The Religious Patterns of Young Roma People in Hungary

The Hungarian youth’s religious patterns and beliefs can be analysed with the help of the database of Hungarian Youth Research. According to ROSTA (2013) it seems to be clear that the earlier phenomena and categories (for example “religious in their own way” etc.) do not fit into the current religious transfornations anymore. The main reasons for this change are the development of mixed religious patterns (mixed with esoteric or oriental elements) and the process of secularisation. The aim of this analysis is to map the young roma people’s religious patterns relying on the database Hungarian Youth Research 2012. The roma subsample was separated from the database (N = 625). Quantitative techniques were used (crosstabs analysis, means, ANOVA-test, factor analysis and linear regression). Our first empirical findings refer to the fact the rate of the “traditional religious” people are higher in the roma subsample. The acceptances of religious dogmas are more typical in the roma subsample but these items refer not only to the traditional contents but also to esoteric elements as well (e.g. horoscope). This relationship is verified by the ANOVA test (two factors were identified, the traditional and the esoteric-mixed factors). A linear regression modell was configured in the last step of our analysis and the sociocultural variables (type of settlement, gender etc.), the categories of religious self-identification and the fact of the religious education are involved as well. Our aim was to reveal the effect of the ethnic identity. The effect of the roma ethnic identity is significant in the case of the traditional religiosity but this relationship does not move together with the rejection of the mixed-esoteric patterns. Moreover with the help of the database the effects of the secularization have been revealed as well. DOI 10.14232/belv.2017.4.7 https://doi.org/10.14232/belv.2017.4.7 Cikkre való hivatkozás / How to cite this article: Bocsi, Veronika (2017): The Religious Patterns of Young Roma People in Hungary. Belvedere Meridionale vol. 29. no. 4. 105–121. pp. 1 Thank for the database for Kutatópont (http://kutatopont.hu/). ISSN 1419-0222 (print) ISSN 2064-5929 (online, pdf) (Creative Commons) Nevezd meg! – Így add tovább! 4.0 (CC BY-SA 4.0) (Creative Commons) Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) www.belvedere-meridionale.hu


INTRODUCTION
If we would like to deal with the religious life of roma people a number of research are available abroad and in Hungary as well.Mostly they tend to focus on the local communities, the features of the local networks and practices and it is frequent that they deliver qualitative data (VEKERDI 1982;TESFAY 2009;GOG 2008).We have to lay down that this can be a very important viewpoint from the aspect of sociology or anthropolgy of religion but the macrostatistical data can give us such an empirical findings which reflect the changes of the whole roma society (GYETVAI -DÉSFALVI 2016;BARTL 2014;GYETVAI 2014).With the help of this viewpoint the first signs of social transformations are more detectable than analyzing the practices and life of a closed or local communities are analysed. 2his explains for the fact that sometimes we have the feeling that the quantitative and qualitative research into roma people' religious life deliver us different emphasis and different trends.It is clear-cut that the Hungarian roma population has been transforming -for example the number of the roma people has increased, their educational level as well and the rate of those people which live in towns has also risen.Accordingly we have to focus on the macrostatistical data because the local analysis cannot be cover the whole roma population.If we go through the special literature from these two aspects we have got such a feeling that there are various features in the field of roma people's religious life.Various trends can be found: secularization on the one hand and strong traditional patterns or a religious revival on the other.Moreover if we analyse the religious patterns and practices the wider social contexts and transformations seem to be a very important starting points.In the case of research which focus on the youth this last sentence is more relevant.
In this study this macrostatistical viewpoint will be dominant.The database which were used (Hungarian Youth Research 2012) is representative for gender, age and the type of the settlement so the separated roma subsample does not refer to specific group of the Hungarian roma people or specific geographic area.We know that Hungarian roma population is varied and in these different groups and communities the religion get different features and practices but -as we have mentioned -the changes in the whole roma ethnic group can be followed with this method more succesfully.

Religious patterns in Hungary
In this paper the phenomenon of secularisation was not analyed -several publications in the field of social sciences had done it (e.g.ROSTA 2004;MÁTHÉ-TÓTH 2011).Moreover some parts of this process are evident (the rise of people outside denominations, the fall of the religions's social elements, the individualisation etc.).It is an important fact that the relationship between the modernisation process and the secularisation is detectable.However the secularisation process does not seem to be a linear and irreversible change (ROSTA 2004;TOMKA 1996;HAMILTON 1995).
The socialist system did not prefer the religious practices in Hungary and tried to restrict it but the first elements of secularisation had started in an earlier period.The restrictions of the religious practices were less effective in isolated villages.Besides the religion has got several other functions in the communities which refer to the everyday life, the rhythm of the work, the holidays, the whole cultural system, the normes and the rules of the social behaviour.And this is why the above restrictions did not function effectively in local communities.This duality according to the types of the settlements has still been working and the main features and patterns of religiosity are different in villages and towns.After the regime changed these restrictions have disappeared.But the secularization and the changes in the field of religious life are rooted in not only the political context (FÖLDVÁRI 2009).Moreover the religiosity has not withdrawn only in the small villages -despite the fact it could be expected due to secularisation.For example in the larger towns a part of the youth can be typified with religious mindset.
If we analyse the members of the denominations according to the data of census 2001 and 2010 we can detect a significant transformation.(Moreover it is to be noted that this section of the survey was not obligatory and the phrasing of the question was different in the two census.The phrasing of the survey from 2011 focuses on the feeling that "I am belonging to the denomination" and not on the simple membership.)In 2011 the total of the "no answer" and "without denomination" is about 46%.In the case of the established churches the number of the membership was decreasing -despite the fact that the restrictions of the socialist system have not been in force for one or two decades (Diagram 1).

DIAGRAM 1
The Hungarian society's division according to denominations (Source: KSH, own diagram) But this change does not unequivocally effect the decrease in the religious practices and mindset because this diagram displays only the institutional approach of religiosity.The individual forms of the religiosity have become more widespread -and this process refers to the dissolution of the traditional communities and the preference of indiviual values.The category of "I am religious in my own way" according to TOMKA (1999) links inherently to this transformation.But this type of religiosity can not be measured by means of questions which the survey of the census contains.According to TOMKA (2009) the rate of this category was less than 50% percent of the adult population.The relationship between the religious self identification and the denomination was analysed by ROSTA (2011) in Hungary.
During the socialist system's last decade an other important religious phenomenon was observable.At first we can detect the dynamic progress of the free churches (Methodists, Adventists, Jehowah's witnesses etc.).This change significantly transformed the religious spectrum of the Hungarian denominations.Other important shift was the evolving of the New Age's elements is this term -the eastern religions became more significant and some esoteric notions and phenomena have become popular.To sum up this chapter sweeping changes have formed the religious life in Hungary in the Millenium.

THE RELIGIOUS PATTERNS OF THE YOUTH
Naturally the religious patterns of the youth have got a special features.This notion can be due to causes which refer to the age (e.g.INGLEHART 2008).The age can form the thinking and the values and this change forms the attitude toward the religion as well.Other important factor is the life situation of the younger cohorts.INGLEHART (2008) calls to our attention to the symptom that the thinking and values were formed by the effects of the period in which people grew up.And the childhood of the young cohorts came to an end during the socialist system or after the regime changed -which represents a watershed from the aspect of religion.Moreover if we analyse the religious life of small villages in Hungary we have to be aware of the fact that the young people are underrepresented in this type of settlement.
Another very important external context is the process of individualisation.Naturally this notion refers not only to the religion but rather to every segment of life.This transformation evolved in the Middle Ages according to MACFARLANE (1978).This change has made less the personal networks less tight and the personal religious networks as well.So every fact mentioned earlier tend to shape the religious patterns of young people in less traditional way.
If we analyse the recent features of young people's religious life with the help of the Hungarian Youth Research 2012 interesting patterns can be detected.The survey contains six optional responses (no answer, uncertain, I am not religious, I have got definitely another conviction, I am not religious, I am religious in my own way, I am religious and I am following the rules of my church).
We have to be aware of the fact that religious self identification contains more categories and this attitude can not be described with a simple "no" or "yes" option.In 2012 the most typical category was the "I am not religious" answer and the second was the "I am religious in my own way".The traditional form of religiosity has got only 7 percents (ROSTA 2013).(Table 1).
In 2016 less significant but important transformation is perceptible.The rate of the traditional religiosity has decreased to 6%.The most dominant category was the "I am religious in my own way" (43%) and the second is the "I am not religious" (41%).(SZÉKELY -SZABÓ 2016) TABLE 1 The proportion of the religious self identification according to the Hungarian Youth Research 2012 (N = 8000, per cents) (Source: my own table ) The questionnaire from 2012 contains a very important question which refers to the different religious dogmas.But these items are linked not only to the "traditional" religiosity (Hell, Satan, God, Trinity, Saints etc.) but the items of the "New Age" and other eastern religions (reincarnation, horoscope, talisman, UFO etc.).The main lesson of the analysis of ROSTA ( 2013) is that we have to use this complex approach of the religiosity because the confessed dogmas are rooted not only in traditional ground.Some "esoteric" or "eastern" items were more popular than the traditional items.In this analysis this new approach has been applied to young roma people.

ROMA PEOPLE AND RELIGIOSITY
The religiosity of roma people show us a diverse picture in the different European and non European countries but the most typical notion is that this ethnic group is conformed to the religion of the majority.Moreover this was a long process with full of conflicts and the roma people were frequently viewed as heretic in the late Middle Ages.
Over the past two decades the established denomantions and the free churches have turned to roma people more intensively.Free churces have been more successful according to the data of census.The membership of these denominations has increased intensively in case the adult population (GYETVAI 2014).But the young people's religious patterns are not identical with the adult population.
Moreover the data of census show us a change with opposite direction according to BARTL (2014).She said that the roma people's religious patterns have started to assimilate to the non roma population's features.In 2011 22% of the roma people was not belonging to any denomination and 11% did give not answer this question (Diagram 2 [See next page.]).But these data are linked only to the religious self identification and not to the contents of the religious dogmas.We can suppose that these contents have changed as well.
Nevertheless it would be a mistake if we assume that the contents of the religious dogmas are equivalent in the roma and in the non-roma ethnic group even if they say "I am belonging to Catholic church and I am religious" (or other established denominations) so their religious self identification are similar.The roma people's religiosity was analyzed in the 60's by Mészáros and the empirical findings have showed that for example the religious events (e.g.christianing or funeral) were adapted but the ground of their religious feeling is different from the Christian dogmas.
There is a different relationship between God and the indiviual and some dogmas are not known (e.g.penitence) whilst the elements of the magic are perceptible.(VEKERDI 1982) But these findings come from the Kádár era but we suppose that this dichotomy has been observable yet and these elements are still detectable.At the same time the denominations' recourse to the gypsy population may have transformed the system of religious dogmas.
If we would like to compare the contemporary young roma people's religiosity with the nonroma young population we have to be aware of the fact that the closed comunities and the strong traditions can preserve the "original" forms of the religiosity.Based on the data of census we expect the effects of the secularisation too.(Diagram 3)

METHODOLOGY
As we have mentioned the database of Hungarian Youth Research was used this analysis from 2012 and the roma and non-roma subsample were separated.The question which refers to the ethnic identity was self categorization.This approach is embedded in the external context (politics, prejudices etc.) but the rate of gypsy population in the database (7.8%) was similar to estimates of the demographers.(Moreover we have to be aware of the fact that this method can generate some degree of bias.)The database was representative for gender, age, the type of the settlement between 15 and 29.
The categories of the religiosity (e.g.I am not religious, I am religious and I follow the rules of my church etc.), proportions of denominations and religious dogmas (miracles, heaven, God, horoscope etc) were analysed in this study and the roma and non-roma population's patterns were compared with different statistical methods (per cents, frequencies, ANOVA-test, cross tabulation, factor analysis, linear regression).Our hypotheses are as follows: -We can reveal the effect of the ethnic identity in a linear regression modell according to the different types of the religiosity (factors will be identified with the contents of the religious dogmas).
The variables of the social background will be involved in these models but we suppose that ethnic identity has their own effect.
-According to VEKERDI (1982) we suppose if factors were created from the contents of religious dogmas in the roma and non roma subsample the patterns will be different.

EMPIRICAL FINDINGS
Our first step was to reveal the differences between roma an non-roma young people regarding religious self identification.There were seven optional answers connected with this question.
The differences between the two subsamples are presented in Diagram 4.
The most dominant answer in the non-roma subsample is the "I am not religious" and in the roma subsample is "I am religious in my own way".The traditional religiosity is less typical but the proportion is larger among roma people.About 30% of the roma subsample said that "I am not religious".
The religious thinking is more typical in the roma subsample but the total of the two definitely "religious" answer is about 50%.The survey consists of two questions connected with denomination.The respondents had to name the denomination of the christening and this question focuses rather on the religious tradition of the family.In some free churces (Adventists, Methodists or Jehowas's wittnesses etc.) the christening of the children is not possible so this data refers less to the traditions of the families but rather it can be an effect of a adult conversion.In the second question the respondents had to name the denominition which he/she was belonging to at the moment.If we compare the results of these questions we can identify 54 roma persons (this is about the 9% of the subsample) which named a church in first question but did not in the second.The loss is the more apparent in the case of Roman Catholic church but the Calvinist and Greek Catholic denominations have a less significant decrease.The membership of the free churches is lower in this database -we suppose this is more typical among adult roma citizens (the rate of the membership is about 4.5% according the BARTL (2014) in the roma population).Those shifts which were shown in the earlier chapter according to the non roma population seem to be similar in the roma subsample (Diagram 5).
If we separate those 54 young roma people which left their earlier denomination this subsample is typified with specific social features.(Although the number of the subsample is very low so our findings can not be regarded as valid.)Their paternal education level is higher (Diagram 6), they have finished the vocational school in higher proportion and they are underrepresented in the villages and in the county seats.(Young roma people from the smaller towns tend to leave their denomination.) The next step of our analysis is to reveal religious dogmas.This block of survey contains 16 items and these items were measured with a five grade scale.The means of the items were used and the means of the roma and non-roma subsamples were compared.12 significant differences were identified with ANOVA-test (p < 0.05, Diagram 7 [See next page.]).(In the case of talisman, horoscope, reincarnation and UFO there were no significant relationships.)As we can see the acceptence of the religious dogmas is stronger in the roma subsample but this not a definite traditional pattern (see the means of the horoscope below).The most significant difference refers to God, Heaven, Satan and Hell and the distances of some items are less remarkable (e.g.miracles, immaculate concepcion).This patterns seems to be linked to the empirical findings of VEKERDI (e.g. the varied acceptance of the relgious dogmas -but every traditional religious dogma has a higher mean in the roma subsample).The acceptance of the telepathy is higher in the non-roma subsample.
The framework of the religious dogmas can be rather decoded with the positions of the items (Table 2).According to this pattern the elements of the New Age seem to be more essential part of the non-roma subsample's thinking (e.g.telepathy, horoscope, UFO) and some "traditional" dogmas are less significant (e.g.Hell and Satan).The distance between the item of God and the other "traditional" items alludes to the individual religious interpretations and the lower means in the non-roma subsample refer to the stronger effect of secularisaton.The telephaty and horoscope have a higher position than after-life or Saints.
After this stage of our analysis factors were identified with the help of these variables.Maximum likelihood method and varimax rotation were used (saved information: 66.7%, KMO: 0.956) We could save 15 items (every item except "miracles").The patterns of the religious dogmas seem to be clear-cut because those two types of the religious elements can be revealed which was introduced in the theoretical part of this study (traditional elements and the elements of New Age).The traditional factor includes dogmas which are related to established churches (e.g.Heaven, God etc. -at the same time some variables are accepted not in all established churches, e.g.Saints) and the second factor contains the horoscope, talisman, reincarnation, telepathy, magic and UFO.The name of this factor is "esoteric-mixed" and some parts come from eastern religions (e.g.reincarnation) and other parts are linked to the esotericism (horoscope, UFO etc.).The patterns are shown in Table 3.
If we use an ANOVA test (p < 0.05) signicant difference is detectable between the roma and non-roma subsample in the case of traditional factor (the mean of the factor score is 0.381 in the roma subsample and -0.02 in the non-roma subsample, sig.: 0.000).During the analysis of esoteric-mixed no significant difference can be shown.
As we have seen the religiosity is embedded in social background as well.Thus it is a very important research question that these factors have been shaped by ethnic identity or sociocultural background.A linear regression model was applied in which the dependent variables were the factors identified previously.Beyond the sociocultural variables other variables which refer the religiosity were involved the model.The first was the religious education in the family and the second is the type of the religious self identification.Every model has got four steps. 3(The Table 1 of Appendix contains the types of the variables.)In the case of traditional factor the ethnic identity preserves their positive effect after the involvement of the sociocultural and religious variables (Table 4).Gender, the type of the settlement (the county seat) and religious variables have formed too this factor.Positive effect is linked to the roma ethnic identity and the category of woman.Young people  from the county seats do not tend to assume the traditional religiosity.The religious education refers this to attitude.It is very important fact that the "religious in my own way" identification removes the respondents from the traditional religiosity.But the most important relationship for us in the table is the positive effect of roma ethnic identity.
Table 5 represents the linear regression model of the "esoteric-mixed" factor.It is evident that the ethnic identity does not have any effect in this case -but the other relationships are very interesting because those variables are shown which move people toward the esoteric religious dogmas.The most significant effect refers to the category of woman.In the case of the settlement we can see an opposite influence and the religious education does not have any impact either.The "uncertain" religious self identification strenghtens disposition of the positive attitude toward esoteric elements contrary to the "I am religious in my own way" -whilst reading the special literature we may have the feeling that this type of identification refers to this religious disposition.
Our second hypothesis has focused on the system of the religious dogmas in the roma subsample.We expect a different pattern in which the traditional and esoteric elements show us a special framework.In the case of the whole database the two centers of the religious dogmas (traditional and esoteric-mixed) can be defined.If we conduct a factor analysis in the roma subsample the outcome was similar to the pattern of the whole database, because these two centers were identified (Table 6 [See next page.]).Less significant differences are detectable by the esotericmixed factor -e.g this one does not include the item of reincarnation (This variable refers to every factor with similar weight).The essential frameworks of the religious dogmas are parallel in the roma and non-roma subsample.

SUMMARY
The aim of this study was to compare the roma and non-roma young people's religious patterns with quantitative method and representative database.We suppose that with the help of this approach we may refine the picture which generally has linked with the roma people's religiosity according to the special literature.
Our starting point was the study of BARTL (2014) and ROSTA (2013).Bartl used quantitive data as well and her empirical findings focus on the religious changes inside the roma population.Rosta has drawn attention to the transformations of the religious dogmas inside the young cohort.
Special literature has highlighted that the roma people's religiosity has specific features whose elements have got historical and cultural rootes (e.g.roma cultural and religious traditions).But some elements can be the effects of recent transformations.Due to these facts two hypotheses were formulated.According to the first hypothesis we assume that the roma ethnic identity has its own effect is the linear regression model -this preconception was verified in the case of traditional religiosity.The empirical findings show that the roma ethnic identity is connected with this type of religious thinking.The membership of the denominations and the categories of self-identification allude to fact that the religious thinking is more significant in this ethnic group and this is not the effect of the sociocultural background.This quantitative and nationwide result can validate those qualitative or local findings which highlight the central role of religiosity in the local roma communities' life.But our empirical findings are linked to the outcome of BARTL ( 2014) -as we have revealed the proportion of those young roma people which said "I am not religious" is about 30% and we can separate the group of those young people which left their churches (their rate is about 9%).Moreover the stronger acceptance of the traditional religious elements does not result in the refusal of esoteric items (there was not any significant difference between the roma and the non-roma subsample in the case of this factor).However the traditional items compose the ground of their religious dogmas (as we have seen in Table 2).Approaching the roma people's religious life only with the "traditional" aspect will not result in an elaborate overview.
Our second hypothesis has relied on the study of VEKERDI (1982).To verify our presumption we have used a factor analysis only in the roma subsample and the patterns of the factors were compared.This hypothesis has not been verified because the frameworks of the religious dogmas seem to be similar and we can model them with two centres: first with a traditional and second with an esoteric center.
To sum up our empirical findings we can identify some specific elements of the young roma people's religious life and these results reinforce the fact that this issue needs further reserach with different aspects and methods.That is also evident that we should use the new results of other research in the field of sociology of religion because the system of the accepted dogmas and the features of the religious life have been persistently altering.Furthermore attention has to be paid to the external context (political, institutional etc.) since the religious patterns are embedded in these phenomena as well.

DIAGRAM 2
religious I have got definitely another conviction 7 I am not religious 40 I am religious in my own way 31 I am religious and I am following the rules of my church 7 The non-roma and roma population's religious division according the census 2011 (per cents) (Source: BARTL [2014], my own diagram) DIAGRAM 3 The change of the denominations' proportion according to the roma population from 2001 to 2011 (percents) (Source: BARTL [2014], my own diagram)

DIAGRAM 4
Religious Self identification (N = 625 and N = 7375, cross tabulation, p < 0.05, sig.: 0.000.Hungarian Youth Research 2012) DIAGRAM 5 The church of the christening and the church which I am belonging at the moment (N = 625, per head.Hungarian Youth Research 2012)

TABLE 2
The positions of the religious dogmas in the roma and non-roma subsample (N = 625 and N = 7375.Hungarian Youth Research 2012.means with five grade scale)

TABLE 3
The patterns of the factors according to religious dogmas(N = 8000.Hungarian Youth  Research 2012)

TABLE 4
The linear regression model of the "Traditional factor"(Hungarian Youth Reserach  2012.N = 8000)

TABLE 5 The
linear regression model of the esoteric-mixed factor.(N= 8000.Hungarian Youth  Research 2012)