Ethnic origin and identity in the Jewish population of Israel^*

* First presented at the 3rd International European Social Survey Conference, Lausanne, Switzerland, 13–15 July 2016.

ABSTRACT

The paper addresses the multifaceted quality of ethnicity in the Jewish population of Israel by probing into the ethnic categories and their subjective meaning. The analyses utilise data collected during 2015–2016 on a representative sample of Israelis age 15 and older, as part of the seventh and eighth rounds of the European Social Survey (ESS). Hypotheses are developed concerning the relationship between demographically based ethnic origin and national identity, as well as the effect of ethnically mixed marriages on ethnic and national identities. The analyses reveal a strong preference among Jews in Israel to portray their ancestry in inclusive national categories – Israeli and Jewish – rather than more particularistic, ethno-cultural, categories (e.g. Mizrahim, Moroccan, Ashkenazim, Polish, etc). Yet, whether Israeli or Jewish receives primacy differs by migration generation, socioeconomic standing, religion, and political dispositions. While the findings clearly add to our understanding of Israeli society, they are also telling with regard to immigrant societies more generally. First, they reveal a multi-layered structure of ethnic identification. Second, they suggest that ethnic identities are quite resistant to change. Third, ethnically mixed marriages appear to erode ethnic identities and are likely to replace them with national identities.

KEYWORDS:

• Ancestry
• ethnic identity
• ethnicity
• European Social Survey
• Israel

Acknowledgements

The authors wish to thank Rotem Rabinovitz and Amit Lazarus for their superb assistance in data preparation and analysis. The authors also thank Yossi Harpaz, Nissim Mizrachi and Orna Sasson-Levy for their thoughtful insights and suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

* First presented at the 3rd International European Social Survey Conference, Lausanne, Switzerland, 13–15 July 2016.

1 We used the ICBS algorithm for coding countries into the two categories.

2 For similar identification of the 2.5 and 3rd generations see Cohen, Haberfeld, and Kristal (2007) and Ramakrishnan (2004).

3 The list included, Israeli, Arab, Palestinian, Muslim, Jewish, Ashkenazi, Mizrahi, Sephardi, Bedouin, Druze, Ethiopian, Iraqi, Moroccan, Polish, Romanian, Russian, Ukrainian. Not all categories, of course were relevant for the Jewish population. Furthermore, respondents had the option to add an ancestry that was not included in the list or refuse to choose any ancestry.

4 For first- and second-generation immigrants, official statistics show that immigrants from Asia and Africa (15 years and over) constitute 47% of the Jewish population (ICBS 2017, Table 2.6), whereas the sample figure for Mizrahi and Ethiopian Jews combined (excluding the third generation) is 50%.

5 The dependent variable in the multinomial model has three categories – Israeli, Jewish and other. We set Jewish as the comparison category. Two equations are then estimated simultaneously, one contrasting the choices of Israeli and Jewish, the other contrasting other and Jewish. The coefficient estimates in Table 2 are derived from the first equation.

6 This index is more sensitive to socioeconomic context than a cruder measure of geographic centre and periphery. Yet, there is a high correspondence between the two variables. Consequently, when a variable contrasting the Northern and Southern regions (the best geographic measure we have) with all other regions was added to the analysis it had no statistically significant effect after controlling for socioeconomic standing of the statistical area.

7 One example of this would be Jews from the Balkan. According to their European origin they are listed as Ashkenazim. Yet, most them are Sephardi Jews and may define themselves as Mizrahi.

8 We cannot reject the possibility of an order effect whereby for some, especially older first-generation respondents, this was an easier option to select than going to the last option on the list and choosing one’s own country of origin.