Religiosity, Religious Practice, and Antisemitism in Present-Day Hungary

Abstract

Since 1995, Surveys on antisemitism using national representative samples have been regularly carried out in Hungary. In this article, we used data from the 2011 and 2017 surveys to explore the relationship between three types of antisemitism, namely religious, secular, and emotional. Moreover, we scrutinized how different religiosity indicators can be used as explanatory variables for the different types of antisemitism. We found a slight increase in religious and secular antisemitism between 2011 and 2017, while emotional antisemitism remained almost the same. Religious anti-Judaism significantly correlated with both secular and emotional antisemitism, however, its relationship was much stronger with the former. When analyzing the relationship between different types of antisemitism and religiosity indicators, we found that while in 2011, all the indicators were connected to religious, and most of them to secular and emotional antisemitism, in 2017, only the variables measuring subjective self-classification remained significant. The results show that the relationship between religion and antisemitism underwent some substantial changes between 2011 and 2017. While in 2011, personal religiosity was a significant predictor of the strength of antisemitism, in 2017, religion serving as a cultural identity marker took over this function. The hypothetical explanatory factor for the change is the rebirth of the “Christian-national” idea appearing as the foundational element of the new Hungarian constitution, according to which Christian culture is the ultimate unifying force of the nation, giving the inner essence and meaning of the state. In this discourse, being Christian is equated with being Hungarian. Self-declared and self-defined Christian religiosity plays the role of a symbolic marker for accepting the national-conservative identity discourse and belonging to the “Christian-national” cultural-political camp where antisemitic prejudices occur more frequently than in other segments of the society.

1. Introduction

In Hungary, surveys on antisemitism carried out regularly since 1995 show that—often contrary to the perceptions of observers—the share of antisemites among the adult Hungarian population barely changed between 1995 and 2006.1 The percentage of antisemites among the Hungarian population was roughly the same throughout the period 1995–2002 (removed for peer-review). However, after moderate growth in the following years, in 2010, a sudden jump in the proportion of antisemites could be observed (

Table 1. The percentage of antisemites among the Hungarian adult population, 2006–2017.

	Extreme Antisemites	Moderate Antisemites	Non-Antisemites/Unclassifed
2006	18	16	66
2017	26	10	64

).

The results indicate that while the proportion of non-antisemites remained almost constant, the proportion of extreme antisemites significantly grew, which could be the result of the radicalization of the previously moderate antisemitic group.

The surveys mentioned above tried to map the cognitive components of the antisemitic prejudices, among them the presence of traditional religious anti-Jewish contents. In this article, we analyzed two datasets provided by two large-scale surveys (2011 and 2017) on representative national samples of the Hungarian adult population. By using the same questionnaires, both surveys measured not only the strength and tenacity of antisemitic prejudices, but, at the same time, the level of religiosity, forms, and frequency of religious practices, and the denominational affiliation of the subjects. In the analysis, we examined three types of antisemitism: religious, secular, and emotional. The main question the article aimed to answer was whether different religiosity indicators could be used as significant explanatory variables for the different types of antisemitism in the present-day Hungarian society.

2. Results of Previous Research

Empirical sociological studies have demonstrated, on more than one occasion, the continued existence of traditional Christian anti-Judaism and religiously based antisemitism in modern societies. In the United States, several surveys demonstrated that religious attitudes could partly explain the acceptance of secular antisemitic views. The connection between these two sets of attitudes was already explored in the seminal research on authoritarianism of Adorno and his colleagues (Adorno et al. 1969, pp. 208–21). They investigated the correlation between antisemitism/ethnocentrism and religious affiliation, the parents’ religious affiliation, church attendance, the importance of religion and church to the subjects and certain fundamentalist religious beliefs, and they found that “gross, objective factors—denomination and frequency of church attendance—were less significant for prejudice than certain psychological trends reflected in the way the subject accepted or rejected religion and in the content of his religious ideology” (Adorno et al. 1969, p. 221). In the following decades, further research proved the correlation between Christian religious fundamentalism and antisemitism (Glock and Stark 1966; Quinley and Glock 1979; Heinz and Geiser 1988).

Newer investigations on antisemitism in Germany indicated a weak connection between Christian religiosity and anti-Jewish prejudice. According to the antisemitism report of the German Bundestag which summarized the results of several empirical investigations (Bericht des Unabhängigen Expertenkreises Antisemitismus 2017), though 14% of the adult population agreed with the statement that Jews are responsible for Christ’s death, the impact of religiosity on secular antisemitism is low: church membership, denominational attachment and the level of religiosity do not correlate significantly with the variables measuring antisemitism. However, researchers found significant correlations between Christian fundamentalist believes and antisemitism.

Beate Küpper, in her secondary analysis of data stemming from a large-scale survey on group-focused enmity in Germany (2007), found only a weak correlation between the (self-declared) level of religiosity and antisemitism. On a four-point antisemitism scale, the very religious (1.7) and the not-at-all religious (1.8) groups scored somewhat higher than the others (Küpper 2010).

An international comparative analysis of the relationship between religion and antisemitism brought similar results. In his analysis of the World Values Survey data-set2, Arno Tausch (2018) investigated the correlation between the extent of antisemitism and the extent to which respondents attach importance to religion in their life in 28 countries. He found that the importance of religion correlated significantly with the rejection of a Jewish neighbor, however, when he examined in detail the intricate relationship between religion and antisemitism, he found that the relationship between the belief in God and antisemitism is close to zero, while between fundamentalist dogma (belief in heaven, belief in hell, belief in reincarnation) and antisemitism is clearly significant.

The first comparative study on antisemitism in the post-Communist region which contained a question on religious antisemitism was carried out in 1995 by researchers of the University of Vienna (Weiss and Reinprecht 1998, p. 85). According to the results presented in

Table 2. Measurement of religious antisemitism (percentage and average) ¹.

	1	2	3	4	Average
As Christians, we should reject the Jews
Hungary	3	7	21	69	3.6
Czechs	3	6	26	65	3.5
Slovaks	3	6	24	67	3.6
Poland	4	9	29	58	3.4

¹ 1 = strongly agree, 4 = strongly disagree.

, rejection of Jews on religious grounds was the strongest in Poland and equally strong in Hungary and in the successor countries of former Czechoslovakia.

In the last three decades, after the fall of the Communist system, a series of empirical investigations were conducted in Poland in order to explore the structure of contemporary antisemitic beliefs. They discriminated two forms of antisemitism: traditional and modern. Traditional antisemitism was defined as a direct descendant of Christian anti-Judaism expressed by the acceptance of statements about Jewish deicide or blood libel accusation. The results of the first waves of research gave some cause for moderate optimism: the researchers found that the traditional forms of antisemitism occur mainly among older and less educated citizens and residents of rural areas as opposed to city-dwellers (Krzeminski 1993; Krzeminski 2002).

In 2009, the Center for Research on Prejudice of the University of Warsaw conducted a survey on a national representative sample, consisting of measures of both types of antisemitism (Bilewicz et al. 2013). The results concerning traditional antisemitism confirmed former findings. It seemed so that the traditional religious forms of antisemitism are losing ground in Poland. The results showed that traditional forms of antisemitism are shared only by a small percent of the Polish population: 78.5% of participants scored below the mid-point of the religious antisemitism-scale, i.e., disagreed with traditional antisemitic statements. Those who supported statements expressing religious antisemitism were mostly people living outside of big cities, less educated, older, and with lower income.

However, the results of the third wave of the Polish Prejudice Survey conducted on a representative sample of Poles in 2017 show a different picture (Bulska and Winiewski 2018). Forty percent of the respondents agreed (scores 5 and 4 on a five-grade agreement-scale) with the statement that contemporary Jews, too, are responsible for the death of Christ, and that 27% with the statement that Jews had been kidnapping Christian children in the past. The researchers compared these results with data stemming from previous surveys and found that the strength of traditional religious-based antisemitism in Poland has been gradually rising for the last several years. Both items received stronger support than before, for example, in 2009 a little over 13% of the respondents agreed fully (score 5) with the first statement regarding the responsibility of contemporary Jews for the crucifixion of Jesus Christ, whereas in 2017, this number was over 14 percentage points higher. It is worth noting that religious antisemitism significantly correlated with the secular form of antisemitism.

The first empirical studies on antisemitism in post-Communist Hungary already aimed at identifying the religious components of antisemitic prejudices. The researchers repeatedly used two items for this purpose: the subjects were asked about their agreement or disagreement with statements concerning the deicide and its consequences. Although the scale used in the 1995 survey (two-grade agreement scale) was not identical with the ones used in 1994 and 2002 (four-grade agreement scale), the results are somewhat comparable (

Table 3. Anti-Judaism in 1994, 1995, and 2002 (percentage).

	Year	Fully Agree	Partially Agree	Partially Disagree	Fully Disagree	Do not Know	Average (SD)
The crucifixion of Jesus is the unforgivable sin of the Jews.	1994	15	11	20	34	20	2.11 (1.15)
	1995	23		55		22
	2002	8	9	18	35	30	1.87 (1.04)
The suffering of the Jewish people was God’s punishment.	1994	12	12	19	37	20	1.99 (1.10)
	1995	17		58		25
	2002	7	10	18	37	28	1.84 (1.02)

).

The survey conducted by András Kovács in 1995 offered a chance for a detailed analysis of the issue (Kovács 2011, pp. 60–64, 97, 111–13; Hack 2001). As Table 3 indicates, 17%–26% of the sample agreed with the statement about the responsibility for the deicide, and 17%–24% accepted that the historical suffering of Jews was God’s punishment for their sins. The analysis of the results showed that agreement with the two statements was significantly higher among those who were strictly religious and attended church services at least once a week. While church membership and denominational differences did not significantly influence the acceptance of these religious anti-Jewish statements, among the strict believers, a significant difference occurred between the Protestants who accepted significantly more frequently the two statements than the Catholics.

Table 4. Indicators of religiosity by the strength of antisemitism in 1995 (percentage).

	Non-Antisemites	Moderate Antisemites	Extreme Antisemites
Full sample	71	20	9
I am religious
Rather yes	69	22	9
Rather no	75	15	10
Strength of religious convictions
Strictly religious	65	26	9
Religious in my own way	71	20	10
Don’t know whether I am religious	80	17	3
Not religious	76	15	9
Atheist	70	19	11
Do you consider yourself a member of one of the churches?
Yes	71	20	9
No	73	18	9
How often do you attend church?
Several times a week	56	36	8
Once a week	55	30	15
Once a month	80	12	8
Several times a year	76	18	6
Once a year	74	19	9
Never	70	19	11

shows that the relationship between religiosity and secular antisemitism3 is similar to the structure described above: strict religiosity and frequency of church attendance have a significant relationship with secular antisemitism, but church membership and denominational affiliation do not, however, as detailed analysis has proved, strictly religious Protestants are significantly more antisemitic than strictly religious Catholics in this case too.

3. Materials and Methods

3.1. Data

In our present analysis, we used data from 2011 and 2017. Both surveys were designed by (Removed for peer-review) and carried out by Medián Public Opinion and Market Research Institute using PAPI (paper and pen interview) method. The 2011 survey was commissioned by the Ministry of Public Administration and Justice, while the 2017 survey by the Budapest-based Action and Protection Foundation. The sample size of the representative national sample of the Hungarian adult population in 2011 was 1199, while in 2017, 1190. Both datasets were weighted by gender, age, educational level, and the type of settlement where the respondent lives. All the variables we used in our analysis were measured in the same way in both surveys.

3.2. Dependent Variables

As stated earlier, we dealt with three types of antisemitism. Therefore, these are our dependent variables.

3.2.1. Religiously Based Antisemitism

This measure of religiously based antisemitism was composed of two items, including the following statements: (1) “The crucifixion of Jesus is the unforgivable sin of the Jews” and “The sufferings of the Jews were God’s punishment” Responses were measured on five-degree Likert-scales including the following possible responses: “Strongly disagree”, “Disagree”, “Neither agree, nor disagree”, “Agree”, “Strongly agree”. An index was constructed by computing the mean value of the variables mentioned above. According to Eisinga et al. (2013), the Spearman-Brown statistic was used to access the reliability of the two-item scale. The value of the Spearman-Brown coefficient is 0.75 for 2011 and 0.74 for 2017, which is not excellent but acceptable. The minimum value of the created composite measure is 1, meaning that the respondent is religiously not antisemitic, while the maximum value of 5 suggests the opposite. We also categorized the composite measure to differentiate between non-, moderately and extreme antisemites. We categorized those scoring between 1.00 and 2.50 as non-antisemites, those having scores between 2.51 and 3.75 as moderate antisemites, and those over 3.75 as extreme antisemites.

3.2.2. Secular Antisemitism

The measure of secular antisemitism was composed of five items where respondents had to report their agreement on five-degree Likert scales. Values of the Likert-scales had the same meaning as the previous ones. The items included the following statements: (1) “There is a secret Jewish conspiracy that determines political and economic processes”; (2) “Jews are more prone to using unethical means to achieve their goals than others”; (3) “Jews living in Hungary are more loyal to Israel than to this country”; (4) “It would be best if Jews left the country.”, and (5) “In case of certain professions the number of Jews should be limited.” As there were five items, we used Cronbach’s alpha to assess the internal consistency of the items. The values of Cronbach’s alpha were 0.90 for both years, which suggest excellent reliability. The minimum value of the created composite measure was 1, meaning that the respondent can be described by the lack of secular antisemitism, while the maximum value of 5 suggests the opposite. We also categorized the measure into non-, moderately and extreme antisemites using the same thresholds as in the case of religiously based antisemitism.

3.2.3. Emotional Antisemitism

Emotional antisemitism was composed using two variables. The first was a nine-degree Likert scale where one meant that the respondent felt antipathy, while nine that s/he felt sympathy towards the Jews. In the second question, respondents had to choose from two options: whether they felt antipathy toward Jews or not. We combined these two variables. Those who said that they rather did not feel antipathy and scored on the emotional scale four or above were non-antisemites. While we counted as extreme antisemites those who rather felt antipathy toward Jews and had 1 to 5 value on the nine-degree Likert scale. We treated all the others (who had valid values on both variables) as moderate antisemites.

3.3. Independent Variables

In the analysis, we used three independent variables: religiosity, denomination, and strength of religious convictions.

3.3.1. Religious Self-Identity

Respondents were asked whether they felt religious or rather not.

3.3.2. Strength of Religious Convictions

This variable consisted of four categories: (1) Strictly religious, (2) Religious in my own way, (3) Don’t know whether I am religious, (4) Not religious.

3.3.3. Church Membership

Respondents were asked whether they considered themselves or not a member of one of the churches.

3.3.4. Church Attendance

Respondents were asked about the frequency of their church attendance. This variable consisted of five categories: (1) At least once a week, (2) More than once a month, (3) Several times a year, (4) On family occasions and holidays; (5) Never.

4. Results

Table 5. Items of anti-Judaism in 2011 and 2017 (percentage).

	Year	Fully Agree	Partially Agree	Neither Agree nor Disagree	Partially Disagree	Fully Disagree	Do not Know	Average (SD) (Margin of Error4 of the Mean)
The crucifixion of Jesus is the unforgivable sin of the Jews.	2011	9	11	19	10	39	12	2.32 (1.40) (0.08)
	2017	13	13	21	14	23	17	2.74 (1.40) (0.09)
The suffering of the Jewish people was God’s punishment.	2011	5	9	16	12	48	10	2.01 (1.27) (0.08)
	2017	7	10	19	16	32	17	2.33 (1.31) (0.08)

shows the distribution of the items measuring religious anti-Judaism in 2011 and 2017. In 2011, 20% of respondents agreed to some extent with the statement about the responsibility for the deicide, while in 2017 the respective proportion was 26%. In the case of the other statement about taking Jews’ sufferings as God’s punishment for their sins, the proportion of those who agreed was somewhat lower in both years. In 2011, 14% of respondents chose 4 or 5 on the five-degree Likert-scale, while in 2017, 17%. It is important to note that the proportion of those who did not answer the questions increased considerably between 2011 and 2017 for both statements.

Table 6. Items of secular antisemitism in 2011 and 2017 (percentage).

	Year	Fully Agree	Partially Agree	Neither Agree nor Disagree	Partially Disagree	Fully Disagree	Do not Know	Average (SD) (Margin of Error of the Mean)
There is a secret Jewish conspiracy that determines political and economic processes.	2011	14	19	25	7	15	20	3.12 (1.32) (0.08)
	2017	13	16	22	13	15	21	2.99 (1.34) (0.09)
Jews are more prone to using unethical means to achieve their goals than others.	2011	9	16	25	13	22	15	2.73 (1.31) (0.08)
	2017	12	17	20	17	16	18	2.89 (1.35) (0.08)
Jews living in Hungary are more loyal to Israel than to this country.	2011	12	15	29	11	13	20	3.00 (1.26) (0.08)
	2017	12	19	21	14	11	23	3.09 (1.27) (0.08)
It would be best if Jews left the country.	2011	7	12	24	18	29	10	2.47 (1.29) (0.08)
	2017	10	11	21	18	29	11	2.50 (1.35) (0.00)
In the case of certain professions, the number of Jews should be limited.	2011	7	12	25	11	37	8	2.36 (1.32) (0.08)
	2017	12	15	20	13	29	11	2.62 (1.42) (0.09

shows the distribution of the items measuring secular antisemitism in 2011 and 2017. The agreement with these items is much stronger than with those measuring religious anti-Judaism. Approximately one-third of the Hungarian population believed to some extent in a secret Jewish conspiracy in 2011. The proportion of these respondents was slightly lower, 29% in 2017. In 2011, approximately one-fourth of the population agreed with the statements that “Jews are more prone to using unethical means to achieve their goals than others” and “Jews living in Hungary are more loyal to Israel than to this country.” Their proportion slightly increased by 2017. Compared to the previous statements, respondents were less prone to think that the best would be if Jews left the country. However, approximately one-fifth of the respondents agreed with this statement. There is one item where the proportion of those agreeing with it considerably increased. While in 2011, 19% thought that “in case of certain professions the number of Jews should be limited,” their share increased to 27% by 2017. It is also important to note that the proportion of those not answering these questions was much higher than for the items measuring religious anti-Judaism.

Table 7. Items of emotional antisemitism in 2011 and 2017 (percentage).

	2011	2017
Please, indicate on a 1 to 9 scale how sympathetic Jews are were 1 means that they are very antipathetic and 9 that they are very sympathetic.
1—very antipathetic	10	12
2	7	5
3	10	7
4	11	9
5	33	26
6	10	13
7	7	13
8	4	9
9—very sympathetic	4	3
Do not know	4	3
Mean	4.61	4.91
Standard Deviation	2.05	2.18
Margin of Error of the Mean	0.12	0.13
Do you feel antipathy towards Jews?
Rather yes	24	25
Rather no	71	70
Do not know	5	5
Margin of error (percent)	2.52	2.55

shows the distribution of the items measuring emotional antisemitism in 2011 and 2017. On a 1 to 9 scale, respondents found Jews slightly more sympathetic in 2017 (4.91) than in 2011 (4.61); however, the change is only slightly significant. When asked using a dichotomous variable, in both years, approximately one-fourth of the respondents said that they rather felt antipathy towards Jews. It is important to note that the proportion of those rejecting the answers to these questions was much lower than in the cases of the previous items.

Table 8. Different types of antisemitism in 2011 and 2017 (percentage).

	2011	2017
Religious anti-Judaism
Non-antisemites	55	45
Moderate antisemites	19	20
Extreme antisemites	10	14
Unclassifiable	16	21
Secular antisemitism
Non-antisemites	27	25
Moderate antisemites	27	26
Extreme antisemites	16	19
Unclassifiable	30	30
Emotional antisemitism
Non-antisemites	59	63
Moderate antisemites	13	8
Extreme antisemites	20	22
Unclassifiable	8	7

shows the distribution of all three types of antisemitism in 2011 and 2017. Looking at the data of religious anti-Judaism, in 2011, 19% of respondents were moderate, while 10% were extreme antisemites. By 2017, the proportion of the former remained almost the same, while the latter increased slightly by four percentage points. It is important to note that while in 2011, 16% of the respondents were unclassifiable, meaning that did not give a valid answer to at least one of those variables constituting the composite measure of religious anti-Judaism, in 2017, this proportion increased to 21%.

Table 8 shows that secular antisemitism was more widespread both in 2011 and 2017. While in 2011, 27% of the respondents were moderate and 16% extreme antisemites, in 2017, these proportions were 26% and 19%, respectively. The proportion of unclassifiable respondents was much higher than in the case of religious anti-Judaism. This is not only because this composite variable was made of five variables and therefore there was a higher chance that a respondent did not have valid value on one of them but also attributable to the fact that respondents denied answering these variables with a higher chance, as we described above.

When looking at the distribution of emotional antisemitism, the most striking is the low proportion of unclassifiable respondents. It was below 10% in both years. According to emotional antisemitism, the proportion of moderate antisemites were 13% in 2011 and 8% in 2017, which are much lower than in the case of the other two types of antisemitism. The proportion of extreme antisemites were 20% in 2011 and 22% in 2017. These are much higher than in the case of religious anti-Judaism and somewhat higher than in the case of secular antisemitism.

Religious anti-Judaism and secular antisemitism were equally highly correlated in both years. In 2011, their correlation was 0.63, while in 2017, 0.67. In 2011, emotional antisemitism was just slightly more attached to secular antisemitism than to religious anti-Judaism. Its correlation with the former was 0.41, while with the latter, 0.33. However, the difference between the two increased. In 2017, the correlation between emotional antisemitism and secular antisemitism became much stronger (r = 0.58) while with religious anti-Judaism remained almost the same (r = 0.39).

In the next step of our analysis, we analyzed the relationship between the different indicators of religiosity (religious self-identity, the strength of religious convictions, church membership, and church attendance) and the different types of antisemitism, namely religious, secular, and emotional. The distributions of these religiosity indicators remained stable between 2011 and 2017, as

Table 9. Indicators of religiosity in 2011 and 2017 (percentage).

	2011	2017
I am religious
Rather yes	53	55
Rather no	45	43
Missing data	2	3
Strength of religious convictions
Strictly religious	11	12
Religious in my own way	49	49
Don’t know whether I am religious	6	7
Not religious	31	29
Missing data	3	4
Do you consider yourself a member of one of the churches?
Yes	45	43
No	54	54
Missing data	1	4
How often do you attend church?
At least once a week	7	6
More than once a month	7	5
Several times a year	11	15
On family occasions and holidays	40	38
Never	33	33
Missing data	1	3

indicates. A little more than half of the respondents considered themselves as religious in both years. It is clear that this self-identification does not coincide with strict religiosity as the proportion of the later is just approximately 10% of the population. When asked about the strength of their religious convictions, half of the respondents answered that they were religious in their own way, while approximately one-third of respondent said that they were not religious. The proportion of those considering themselves as members of one of the churches little exceeded 40% in both years. However, attending church services regularly was quite rare.

However, it is interesting to analyze how the relationship between religious self-identity and the other types of religious indicators changed from 2011 to 2017. The ones who were strictly religious or religious in their own way considered themselves as religious in both years. However, in 2017, religious self-identification was less typical for the strictly religious group and more for those who were religious in their own way. Besides that, both the relationship between religious self-identification and church membership, as well as between religious self-identification and church attendance weakened considerably from 2011 to 2017.5

In 2011, all indicators of religiosity had a significant relationship with religious anti-Judaism, while in 2017, only religious self-identity and the strength of religious convictions. In both years, those who said that they were religious proved to be much more prone to antisemitism, both in its moderate and extreme forms. In 2011, 31% of the self-identified religious respondents were moderate antisemites and 20% extreme antisemites, the respective numbers for those not feeling themselves religious were only 16% and 6%. In 2017, 30% of those having religious self-identity were moderate antisemites and 26% extreme antisemites, while the respective numbers for those not feeling themselves religious were 20% and 12% (

Table 10. The relationship between religious self-identity and religious anti-Judaism in 2011 and 2017 (percentage).

	Non-Antisemites		Moderate Antisemites		Extreme Antisemites
	2011	2017	2011	2017	2011	2017
Full sample	65	57	23	25	12	18
I am religious
Rather yes	49	44	31	30	20	26
Rather no	78	68	16	20	6	12

).

As

Table 11. The relationship between the strength of religious convictions and religious anti-Judaism in 2011 and 2017 (percentage).

	Non-Antisemites		Moderate Antisemites		Extreme Antisemites
	2011	2017	2011	2017	2011	2017
Full sample	65	57	23	25	12	18
Strength of religious convictions
Strictly religious	46	49	33	29	21	22
Religious in my own way	63	52	25	24	12	24
Don’t know whether I am religious	62	60	24	31	14	9
Not religious	74	66	17	23	9	11

shows, in 2011, those who considered themselves as strictly religious were significantly more religiously antisemitic than the whole sample. In 2017, however, this group proved to be only slightly more antisemitic than the whole sample. Nevertheless, in this year, extreme antisemites are overrepresented among those following their own way in religion. In both years, those who said that they were not religious were the least antisemitic.

As mentioned above, church membership and church attendance only had a significant relationship with religious anti-Judaism in 2011. Those who were members of one of the churches were more antisemitic, either moderately or extremely. Moreover, the more frequently respondents attended religious services, the more prone they were to religious anti-Judaism (

Table 12. The relationship of church membership and church attendance with religious anti-Judaism in 2011 (percentage).

	Non-Antisemites	Moderate Antisemites	Extreme Antisemites
2011
Full sample	65	23	12
Do you consider yourself a member of one of the churches?
Yes	58	26	16
No	71	20	9
How often do you attend church?
At least once a week	46	29	25
More than once a month	43	38	19
Several times a year	53	30	17
On family occasions and holidays	67	21	12
Never	75	19	6

).

Table 13. The relationship of religious self-identity, church membership, and church attendance with secular antisemitism in 2011 and 2017 (percentage)6.

	Non-Antisemites		Moderate Antisemites		Extreme Antisemites
	2011	2017	2011	2017	2011	2017
Full sample	39	36	38	37	23	27
I am religious
Rather yes	29	26	44	35	27	39
Rather no	46	44	34	38	20	18
Do you consider yourself a member of one of the churches?
Yes	33	—	40	—	27	—
No	43	—	37	—	20	—
How often do you attend church?
At least once a week	27	—	40	—	33	—
More than once a month	33	—	54	—	13	—
Several times a year	28	—	51	—	21	—
On family occasions and holidays	43	—	32	—	25	—
Never	41	—	38	—	21	—
Strength of religious convictions
Strictly religious	—	25	—	53	—	22
Religious in my own way	—	35	—	30	—	35
Don’t know whether I am religious	—	29	—	48	—	23
Not religious	—	44	—	38	—	18

shows the relationship of religiosity indicators with secular antisemitism. In 2011, all indicators of religiosity except the strength of religious convictions had a significant relationship with secular antisemitism. In the case of religious self-identity and church membership, the relationship was similar to that with religious anti-Judaism. Those who identified themselves as religious and considered themselves as a member of one of the churches were more antisemitic. In the case of church attendance, the relationship was also similar. However, the pattern is somewhat more clear-cut. Extreme antisemites were overrepresented among those attending religious service at least once a week, while moderate antisemites were among those who did that more than once a month or several times a year. In 2017, in the case of secular antisemitism, similarly to what we observed in the case of religious anti-Judaism, only religious self-identification and the strength of religious convictions remain significant. The relationship with the former was the same: those having religious self-identification were more antisemitic. In the case of the other independent variable, a new type of relationship occurred. Although those strictly religious and those religious in their own way were the most antisemitic, as in the case of religious anti-Judaism, the former group was more moderately antisemitic than the latter. Moreover, moderate antisemites were also overrepresented among the group undecided about their religiosity.

Table 14. The relationship of religious self-identity and the strength of religious convictions, and church membership with emotional antisemitism in 2011 and 2017 (percentage)7.

	Non-Antisemites		Moderate Antisemites		Extreme Antisemites
	2011	2017	2011	2017	2011	2017
Full sample	63	68	15	8	22	24
I am religious
Rather yes	55	61	16	7	29	32
Rather no	71	74	13	9	16	17
Strength of religious convictions
Strictly religious	51	73	18	14	31	13
Religious in my own way	67	66	11	6	22	28
Don’t know whether I am religious	61	67	19	11	20	22
Not religious	62	71	18	9	20	20
Do you consider yourself a member of one of the churches?
Yes	61	—	12	—	27	—
No	65	—	17	—	18	—

shows the relationship of religiosity indicators with emotional antisemitism. In 2011, all indicators of religiosity but church attendance had a significant relationship with emotional antisemitism; in 2017, again, only religious self-identity and the strength of religious convictions. The relationship with the former was the usual one in both years: those who identified themselves as religious were more antisemitic than those who did not. In the case of the strength of religious convictions, in 2011, strictly religious respondents were the most emotionally antisemitic. However, non-antisemites were somewhat overrepresented among those following their own way in religion. In 2017, we observed the same pattern as in the case of secular antisemitism: moderate antisemites are overrepresented among strictly religious respondents, while extreme antisemites among those following their own way. As mentioned above, church membership only had a significant relationship with emotional antisemitism in 2011. The pattern is the usual one: those having church membership were more antisemitic.

5. Discussion

In our analysis, we dealt with three types of antisemitism: religious, secular, and emotional. The first two represented the cognitive component of antisemitism and were composed of variables measuring the agreement with different stereotypical statements. On the other hand, emotional antisemitism grasps the affective component. The result showed only a slight increase in religious and secular antisemitism between 2011 and 2017. While the proportion of moderate antisemites remained the same, that of extreme antisemites increased slightly. In the case of emotional antisemitism, while the proportion of moderate antisemites was a little lower in 2017, that of extreme antisemites remained almost the same. Moreover, it is important to note that the proportion of unclassifiable respondents was much higher in the case of religious, and especially secular antisemitism than in the case of the emotional one. The results show that measuring only the cognitive components of antisemitism is insufficient and should always be complemented with the measurement of its emotional intensity. The analysis of correlation showed, however, that secular antisemitism was more connected to the emotional intensity of antisemitism than religious anti-Judaism. This was especially the case in 2017.

As the next step of our analysis, we scrutinized the relationship between various indicators of religiosity and the different types of antisemitism. We found that in 2011 all indicators, namely religious self-identity, the strength of religious convictions, church membership, and church attendance were connected to religious anti-Judaism and most of them also to secular and emotional antisemitism. However, in 2017, only the indicators based on subjective self-classifications (religious self-identity, the strength of religious convictions) remained in a significant relationship with the different forms of antisemitism. The inclination for accepting antisemitic statements grew above all among those who characterized themselves as being religious in their own way. It is also interesting to note that while in 2011, the above set of variables had a stronger effect on religious anti-Judaism, than on the other types of antisemitism, in 2017, the strengths of effects equalized. Thus, it seems that in 2017, self-defined religiosity is the only indicator which significantly correlated with all forms of antisemitism.

Departing from these results, it is not groundless to presume that the relation of religion and antisemitism has undergone some substantial changes in Hungary. While in Poland, the support of statements expressing religiously based anti-Judaism grew significantly between 2009 and 2017, similar Hungarian data remained basically constant (see Table 5). The changes occurred in another dimension. It seems that while earlier personal religiosity and the level of religious practice played a significant role in influencing the strength of religious and secular antisemitism, nowadays, religion as a constituent of national and cultural identity and as political-cultural identity marker took over its function as a predictor of antisemitism.

The role of religion in the construction of collective identities has been widely analyzed in the literature on nationalism and religion. Rogers Brubaker characterized “religious nationalism as a specific phenomenon, one in which religion provides content for nationalism as a form” (Brubaker 2012, p. 17). According to Roger Friedland, “Religious nationalism … makes religion the basis for the nation’s collective identity and the source of its ultimate values and purpose on this earth.” (Friedland 2001, p. 139). In another article dealing with contemporary anti-Moslem rhetoric of the European far-right, Rogers Brubaker pointed at the phenomenon he called the ‘culturalization of religion’ and the appearance of ‘identitarian Christianism’ whose purpose is not the defense of religious values and religiosity as such but setting boundaries between the Christian “Us” and the Moslem “Them” (Brubaker 2017, p. 1191). In the last decade, since a national-conservative administration governs the country, in Hungary, religion as a cultural marker has been playing a much stronger role in the public discourse on national identity than before. The discourse of the national-conservative right resuscitated the “Christian-national” idea of the pre-WWII decades in a new context. According to this view, Christian culture is the unifying force of the nation, and it gives the inner essence and meaning of the state, and at the same time, the guarantee for the survival of Europe as a collective of Christian nations. “Without Christian culture, there will be no free life in Europe; and that if we fail to defend our Christian culture, we will lose Europe, and Europe will no longer belong to Europeans”, declared Prime Minister Orbán in a speech last year.8 In this discourse, the nation itself appears as a sacred collective entity, and national identification carries religious attributes: being Christian is equated with being Hungarian, belonging to the national community. An extremely powerful expression of this position is the preamble of the new Hungarian constitution called “Fundamental Law” (2011), according to which the constitution is anchored to the Christian roots of Hungary, and the commitment to nation-sustaining power of Christianity. In this context, self-declared and self-defined Christian religiosity can be considered as a symbolic marker for accepting the national-conservative identity discourse and belonging to the “Christian-national” cultural-political camp which represents the national community defined on this basis. Previous research has pointed at the identity creating function of antisemitism for the far-right political scene in Hungary (Kovács 2012). The Christian-national discourse seems to play a similar role. It is not groundless to presume that a substantial group of those for whom religiosity is a marker for political belonging tend to accept other elements of the identity discourses of the same camp, too. However, further research is needed to investigate how far the impact of culturalized and politicized religion on anti-Jewish prejudice can be considered as an independent explanatory factor of the relationship between religiosity and antisemitism.