This article was downloaded by: [University of Haifa Library] On: 26 November 2014, At: 10:59 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Ethnic and Migration Studies Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/cjms20

Language Proficiency among Post-1990 Immigrants in Israel Rebeca Raijman, Moshe Semyonov & Rona Geffen Published online: 24 Nov 2014.

To cite this article: Rebeca Raijman, Moshe Semyonov & Rona Geffen (2014): Language Proficiency among Post-1990 Immigrants in Israel, Journal of Ethnic and Migration Studies, DOI: 10.1080/1369183X.2014.982523 To link to this article: http://dx.doi.org/10.1080/1369183X.2014.982523

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

Journal of Ethnic and Migration Studies, 2014 http://dx.doi.org/10.1080/1369183X.2014.982523

Language Proficiency among Post-1990 Immigrants in Israel

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Rebeca Raijman, Moshe Semyonov and Rona Geffen

This paper focuses on the process of linguistic assimilation among post-1990 immigrants arriving in Israel under the Law of Return. We analyse levels of Hebrew proficiency among four immigrant groups (Ethiopia, Former Soviet Union [FSU], Middle Eastern and North African countries [MENA]) and countries in Europe and America [EUAM]) and by language of origin (Amharic, Russian, Spanish, English and French). The analysis is based on a data-set collected by the Central Bureau of Statistics: the post1990 Immigrant Survey. Controlling for all relevant factors, the data reveal substantial differences across origin groups: immigrants of MENA and EUAM countries have higher probabilities of reporting the highest level of command of the language than their FSU and Ethiopian counterparts. Speakers of Spanish and French attain higher levels of Hebrew proficiency than speakers of Russian, English and Amharic. Apparently, the very existence of a large and residentially concentrated ethnic or linguistic community (as is the case for Ethiopian and FSU immigrants) becomes a potential disincentive for improving their Hebrew skills. The fact that English is a lingua franca understood and prized in Israel explains the slow process of English speakers’ language acquisition. The findings are discussed in light of existing theories of linguistic assimilation. Keywords: Language; Israel; Assimilation; Immigrants Introduction Language has been singled out as a crucial domain for assessing immigrants’ economic and social adjustment in their host countries. Consequently, linguistic assimilation has become a major issue of interest for both academics and policymakers (Espenshade and Fu 1997; Espinosa and Massey 1997; Chiswick 1998; Chiswick and Miller 1998; Alba et al. 2002). Proficiency in the local language affects Rebeca Raijman is a Associate Professor at the Department of Sociology and Anthropology, University of Haifa, Haifa, Israel. Correspondence to: Department of Sociology and Anthropology, University of Haifa, 199 AbaHushi, Avenue Mount Carmel, Haifa, Israel. E-mail: [email protected] Moshe Semyonov is a Professor of Sociology at the Department of Sociology and Anthropology, Tel-Aviv University, Tel Aviv, Israel. E-mail: [email protected] Rona Geffen is a Graduate Student at the Department of Sociology and Anthropology, University of Haifa, Haifa, Israel. E-mail: [email protected] © 2014 Taylor & Francis

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

2

R. Raijman et al.

not only labour market outcomes (labour force participation, occupational and earnings attainment) but also efficiency in consumption of goods and services. In addition, language destination skills influence the quality and quantity of social networks enhancing social and civic assimilation in host societies. Indeed, all language-related economic and non-economic benefits provide, in turn, incentives to invest in acquiring a good command of the host country’s language (Espinosa and Massey 1997; Chiswick and Miller 2014). This paper focuses on the process of linguistic assimilation among post-1990 immigrants arriving in Israel under the Law of Return. Most previous studies on linguistic assimilation in Israel focused on immigrants from the Former Soviet Union (FSU; e.g. Mesch 2003; Remennick 2003a, 2004; Ben-Rafael, Olshtain, and Geijst 1998; Ben-Rafael et al. 2006; Tannenbaum 2009); others examined language proficiency among South Africans (Raijman 2013) and French speakers (Ben-Rafael and Ben-Rafael 2013); and several compared language use across multiple groups of immigrants (Beenstock 1996; Chiswick 1998; Beenstock, Chiswick, and Repetto 2001).The studies that compared language use across immigrant groups utilised data obtained from the 1970 and 1983 censuses. The Israeli census provides information on the languages immigrants speak daily (as a primary or secondary language) but do not provide data on degree of fluency in Hebrew.1 Consequently, we know relatively little about linguistic assimilation in different immigrant groups in Israel. We aim to fill this gap by means of the ‘Immigrant Survey’—a unique data-set collected by the Central Bureau of Statistics that provides detailed information on language proficiency among immigrants who arrived in Israel between 1990 and 2007. We analyse levels of Hebrew proficiency among four immigrant groups (Ethiopia, FSU, Middle Eastern and North African countries—MENA, and countries in Europe and America—EUAM) and by language of origin (Amharic, Russian, Spanish, English and French). These five languages comprise 88% of the native languages spoken by post-1990 immigrants in Israel. In what follows, first we set out the theoretical background typically used in the literature to explain language proficiency in foreign-born adults; secondly, we present the case of Israel as an immigration country; thirdly, we introduce the data source and the variables used in the data analysis; fourthly, we present the findings; and finally, we discuss them in light of theories of linguistic assimilation.

Linguistic Assimilation: Theoretical Background Sociologists and economists describe language acquisition as an investment in human capital that improves labour market opportunities for new immigrants. The literature on immigrants’ linguistic assimilation highlights a series of factors associated with language proficiency. The human capital approach tends to stress three main elements: (i) economic incentives for acquiring language skills; (ii) exposure to the new language in the country of origin and the host country and (iii) efficiency in

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 3

converting economic incentives and exposure into language skills (see Chiswick 1998, 255). The logic underlying ‘economic incentives’ is that investment in language skills improves economic assimilation. Since language is also acquired through interaction with co-workers, it is expected that immigrants active in the labour market will acquire more language skills than their counterparts who are not part of the labour force (Beenstock 1996). Although participation in the labour market increases the likelihood of improving language proficiency and use (Espenshade and Fu 1997), the literature suggests that this is particularly relevant for immigrants in highly skilled occupations where the payoff for command of the local language is especially high (Beenstock 1996; Espinosa and Massey 1997; Remennick 2003b; Raijman 2013).2 ‘Exposure’ to the new language both in country of origin and in the host country is expected to enhance linguistic assimilation. Immigrants who have received formal or informal language training in country of origin and/or country of destination tend to be more fluent than those who have not (Beenstock 1996; Beiser and Hou 2000; Raijman 2013). Likewise, language proficiency and use tend to increase with length of stay in the country of destination, as immigrants accumulate experience in their new society (Stevens 1986, 1992; Espenshande and Fu 1997; Van Tubergen and Kalmijn 2009; for the case of Israel, see, e.g. Chiswick [1998]; Beenstock, Chiswick, and Repetto [2001]; Mesch [2003]; Remennick [2004]; Raijman [2013]). Exposure is also related to the degree of ethnic enclosure of specific immigrant groups, associated with residential concentration and ethnic composition of social networks (Stevens 1992; Espenshade and Fu 1997; Mesch 2003). It has been argued that immigrants who tend to bond with co-ethnics and live in neighbourhoods with a larger concentration of immigrants from the same country of origin or in languagespecific communities are less exposed to the local language. Therefore, they are more likely to use their mother tongue in daily interactions and less likely to use the local language (Espinosa and Massey 1997; Chiswick and Miller 2014). In addition to neighbourhood, another relevant arena of linguistic exposure is the family (Stevens 1992; Chiswick and Miller 1998). Individuals married to co-ethnic spouses tend to use their mother tongue more often for daily family interactions, whereas singles and individuals married to non-co-ethnic spouses have more opportunities to become proficient in the local language and to improve their language skills (Van Tubergen and Kalmijn 2009). ‘Efficiency’ refers to the ways in which incentives and exposure affect language proficiency. Age at immigration and education have been singled out as important efficiency variables. Younger immigrants have greater economic incentives to learn the language because socio-economic returns on such an investment are expected to continue for more years. Immigrants arriving at an older age will presumably display lower levels of language acquisition because of the shorter time remaining to enjoy returns on the investment (Stevens 1992; Chiswick 1998; Van Tubergen and Kalmijn 2009; Van Tubergen and Wierenga 2011).3

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

4

R. Raijman et al.

Highly educated immigrants are better able to acquire new language skills than less educated counterparts (Stevens 1999) and are more willing to invest in language acquisition, mostly because language skills improve economic assimilation in the host society (Mesch 2003; Van Tubergen and Kalmijn 2009). The efficiency with which an immigrant can learn the dominant language of the host country may also vary with the immigrant’s language skills. Those who are literate in their native language can use their mother tongue skills to acquire proficiency in the new language. By contrast, illiteracy in mother tongue, namely the absence of reading and writing skills in the native language, negatively influences language acquisition (Esser 2006).4 Gender is likely to affect language acquisition among immigrants, mostly because men and women tend to participate in different social environments; Differences in social environments are expected to have significant implications for language acquisition and language use (Stevens 1999). For example, immigrant men are more likely to join the labour economically active force than immigrant women. Consequently, immigrant men are expected to evince higher levels of local language fluency than immigrant women. However, as immigrant women have been narrowing the employment gender gap, one would expect that gender differences in language proficiency would also narrow (Rebhun 2014). It has also been argued that because women are more likely than men to be responsible for child education and family health care, they might acquire a good command of the destination language in order to perform these activities efficiently (Stevens 1992; Rebhun 2014). In general, however, studies have revealed that immigrant women are less likely than their male counterparts to learn the host country’s language (Jasso and Rosenzweig 1990; Espinosa and Massey 1997; Beiser and Hou 2000). Country of origin is an important factor affecting linguistic assimilation. Country of origin encapsulates pre-migration exposure to the destination language (Beenstock, Chiswick, and Repetto 2001). In the Israeli case, Jewish immigrants arriving from countries in Europe, America and MENA are exposed to Hebrew as a result of early participation in Jewish institutions in the Diaspora. By contrast, exposure to this type of activities was fairly low for immigrants arriving from FSU and Ethiopia. Country of origin also reflects the size of the immigrant group in the destination country. Proficiency in the host country’s language is lower among immigrants who can avoid using it (Chiswick and Miller 2014). Immigrant groups with a high degree of ethnic enclosure, living in places with a high concentration of immigrants, are more likely to speak their mother tongue and have fewer incentives to invest in learning the local language (Beenstock, Chiswick, and Repetto 2001). Beenstock, Chiswick, and Repetto (2001) further suggest that language of origin (mother tongue) may be an important determinant of linguistic assimilation among first-generation immigrants. Linguistic distance as captured by difficulty learning the destination language by speakers of a language of origin has been singled out as an important predictor of linguistic assimilation (Chiswick and Miller 1998). Subsequently, Chiswick and collaborators ‘proposed an index of the linguistic distance of myriad language from English, based on the difficulty of Americans learning these

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 5

languages’ (Beenstock, Chiswick, and Repetto 2001, 45). Although no comparable index has been developed to measure the linguistic distance of languages from Hebrew, Beenstock, Chiswick, and Repetto reported that Hebrew is discerned as linguistically closest to Arabic and Amharic (2001, 46). Therefore, immigrants from MENA countries in which Arabic or Amharic is one of the official languages are expected to have an inherent advantage in learning Hebrew over immigrants speaking other mother tongues linguistically distant from Hebrew. Furthermore, some origin languages like English have an international role as lingua franca and are widely used in many host countries, especially in Israeli society. So immigrants whose mother tongue is English might be under much less social pressure to improve their Hebrew skills than other immigrant groups who speak other languages that are not widely spoken in Israeli society (e.g. Spanish, French or other languages). Unlike English speakers, the latter groups are ‘forced’ to use Hebrew to integrate into society and, hence, are more likely to be willing to invest in acquisition of Hebrew (Beenstock 1996; Raijman 2013). The human capital model has been broadened to include other relevant factors that might affect language proficiency, such as immigrants’ attachment to the host society and commitment to stay in the new country. Specifically, immigrants who are highly attached to the host society may be more willing to invest in language acquisition and, therefore, become more proficient in the local language (Chiswick and Miller 1995; Hochman and Davidov 2014). Likewise, immigrants who feel that they are welcome by the receiving society might have stronger incentives to adapt to the local culture, therefore to invest in language learning. Following this theoretical framework, we examine linguistic assimilation of firstgeneration immigrants who arrived in Israel during the last two decades. When doing so we aim to improve on previous studies conducted in Israel in several ways: first, we examine the linguistic outcomes of several immigrants groups and languages of origin (mother tongue); secondly, we control for specific measures of exposure to the new language, such as Hebrew proficiency before arrival,5 exposure to training in Israel (attendance at ulpan—see below), social networks, the role of social climate embracing immigrant groups (perceived discrimination) and commitment to the new society (degree of Israeli identity). By focusing on the specific case of Israel we also aim to contribute to the existing literature on linguistic outcomes of immigrants in non-English-speaking countries. So far, most research on the topic has concentrated on classic English-speaking countries of destination such as the USA (e.g. Espinosa and Massey 1997; Stevens 1992), Canada (e.g. Chiswick and Miller 1992; Beiser and Hou 2000) and Australia (e.g. Evans 1986; Chiswick and Miller 1996), while the linguistic experiences of immigrants to non-English-speaking societies has been neglected.6 Indeed, the Israeli context provides a broader comparative perspective for the study of linguistic assimilation.

6

R. Raijman et al.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Immigration to Israel Israel is a society of immigrants and their offspring, where at the end of the twentieth century two out of three members of the Jewish majority were foreign born (40%) or of the second generation (30%) (Raijman and Kemp 2010). Unlike most migratory movements, immigration to Israel has been characterised as a returning ethnic migration (Levy and Weiss 2002). The centrality of the idea of immigration as a return from the Diaspora is expressed in the Law of Return of 1950. This law grants Israeli citizenship to Jews and their children immediately upon arrival in the country. Jewish immigrants came to Israel in a series of waves. The first, at the turn of the twentieth century, was mainly from European countries. The second occurred shortly after statehood (1948) as a cumulative immigration of MENA Jews and Holocaust survivors. Immigration in the three decades after the establishment of the state (1960s to 1980s) was more sporadic and less systematic. It was characterised by a slow but constant stream of immigrants from North and South America, as well as from South Africa, Eastern Europe and the FSU, Ethiopia and Iran. By then, a broad infrastructure of public housing and support was available for all new immigrants. Over time, most of the immigrants of the 1970s and the 1980s became fully integrated and achieved higher levels of socio-economic status (Raijman and Semyonov 1995, 1997; Cohen 2002). In the early 1990s, Israel witnessed a renewal of massive immigration flows, mainly immigrants from FSU and their families, under the Law of Return. Between 1989 and 2010 they numbered more than one million people. To these may be added the flow of immigrants from Ethiopia (4.1% of all immigrants arriving in the 1990s and 10.2% of those arriving in the 2000s), a steady influx of Jewish immigrants from Western and Central Europe and North and South America (7.7% of all immigrants arriving in the 1990s and 24.8% of all immigrants arriving in the 2000s) and a smaller flow of immigrants from MENA7 countries such as Iran, India, Morocco, Tunisia, Algeria and Turkey (2% of all immigrants arriving during the 1990s and 5.2% of all arriving during the 2000s) (Raijman and Kemp 2010; Semyonov, Raijman, and Maskileyson 2014).8 Whereas FSU and Ethiopian immigrants display high levels of ethnic enclosure evinced by residential segregation and co-ethnic social networks, migrants from EUAM and MENA countries tend to live in neighbourhoods with low levels of ethnic concentration and enjoy social ties with native-born Israelis (Raijman 2009). Upon arrival, all adult immigrants in Israel are given the opportunity to participate gratis in a 5-month intensive Hebrew course held at ulpanim.9 The primary purpose of the ulpan is to help newcomers integrate into the new social and cultural life through the acquisition of language skills. The newcomer is quickly taught enough basic vocabulary and grammar for simple Hebrew conversation and for reading and writing simple texts. Some studies have shown that on average, 70% of immigrants attended an ulpan during their first year in Israel, and almost a quarter of those who began their studies dropped out before completing them (Adler 2007). As we shall see, although ulpan attendance does not always result in good command of the

Journal of Ethnic and Migration Studies 7

language, upon completion it provides the minimum basis from which immigrants’ linguistic skills progress over time.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Data Source The analysis is based on a unique data-set collected in Israel by the Central Bureau of Statistics: the post-1990 Immigrant Survey. The survey is based on a nationally representative sample of 3952 immigrants aged 27–75 years who arrived in Israel between 1990 and 2007 and was combined with respondents’ socio-economic data obtained from the 2008 national census.10 We focus on those arriving in Israel over the age of 18, hereafter called the first generation. The questionnaire has the advantage (over other data sources) of soliciting from these immigrants detailed information on ‘pre-migration characteristics’ such as Hebrew proficiency before arrival, as well as data on ‘post-migration characteristics’ usually not available in existing data sources such as time invested in Hebrew acquisition, current Hebrew proficiency (reading, writing and speaking), attachment to Israel (Israeli identity), ethno-linguistic environment (social networks) and perception of discrimination. The file also contains standard information such as education, labour market status, age, gender, marital status and tenure in the country. Dependent Variable: Level of Hebrew Proficiency Level of Hebrew proficiency was based on respondents’ self-reports. Although some researchers have queried the validity of self-report of language proficiency, other studies have shown strong correlations between self-reports and objective measures (Stevens 1992; Espenshade and Fu 1997). Following previous studies on the topic (e.g. Remennick 2003b; Van Tubergen and Wierenga 2011), we determined the level of Hebrew proficiency from respondents’ answers to three questions, combined to construct an index: ‘How would you evaluate your current ability to (i) conduct a conversation, (ii) read, and (iii) write in Hebrew?’. Responses were measured on a 5-point scale from 1 = not at all to 5 = very fluent (Cronbach alpha = 0.94). Independent variables A series of variables were selected to test the economic motivation hypothesis, the relevance of exposure for language proficiency and the efficiency hypothesis. To examine the economic motivation hypothesis, we included a series of dummy variables measuring labour market position: Professional, Technical and Managerial jobs (PTM), white-collar jobs and blue-collar jobs (the omitted category is ‘Does not belong to the labour force’). To test the relevance of exposure for language proficiency, we included the following variables: mean level of Hebrew proficiency before arrival,11 number of months attending Hebrew courses in Israel (measured as a series of dummy variables: less than 4 months, 5–6 months, over 7 months; the

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

8

R. Raijman et al.

omitted category was ‘Did not attend at all’), years since immigration and marital status (married = 1). Ethnic enclosure was measured by a set of dummy variables indicating whether the respondent has Israeli friends = 1, no friends at all in Israel and no Israeli friends (omitted category). To test the efficiency hypothesis, we controlled for age at migration (measured in years) and education (measured as total years of formal education). Discrimination was measured by questions asking the extent to which the respondent felt mistreated due to origin at the workplace, government offices, stores and shopping centres, entertainment establishments and other places (3 = often, 2 = seldom, 1 = never). For each respondent we calculated an index based on the average of the answers to these questions. Commitment to the new society was measured by the answer to a question regarding self-definition as Israeli, measured on a 4-point ordinal scale (1 = not at all, 4 = to a large extent). We also controlled for gender (male = 1). Country of origin was measured by a set of dummy variables: immigrants from Ethiopia, immigrants from EUAM and immigrants from MENA12; omitted category: immigrants from the FSU. To test the influence of language of origin, we used the following dummy variables: Amharic speakers, Spanish speakers, French speakers, English speakers and other; the omitted category was Russian speakers.

Data Analysis Data presented in Table 1 reveal that overall immigrants have lived in Israel on average for 13 years, their average age at migration was about 40 years for almost all groups, and most were currently married.13 Despite these general similarities, the four immigrant groups differ significantly in socio-demographic and economic attributes. EUAM and FSU immigrants have most years of formal schooling (15.5 and 14 years, respectively), the highest percentage with an academic degree (51% and 46%, respectively) and the largest proportion of high-status occupations (professional, technical and managerial occupations– PTM). By comparison, MENA immigrants have relatively lower levels of formal education (13 years of schooling and 22% with academic degree) and a lower proportion of high-status occupations. Immigrants from Ethiopia show the lowest level of education (69% did not attend school in their places of origin) and the smallest proportion of professional and managerial jobs. The groups also differ in the degree of ethnic enclosure as measured by the percentage of those having Israeli friends. Whereas a large percentage of EUAM and MENA immigrants reported having Israeli friends (81% and 69%, respectively), Ethiopians and FSU migrants reported relatively lower levels of social ties with Israelis (50% and 35%, respectively). The data also reveal group differences in perceptions of discrimination and attachment to Israel. Perceptions of discrimination are especially high among Ethiopians, followed by FSU immigrants; in the EUAM and MENA groups, they are fairly low. Ethiopians, MENA and EUAM immigrants evince a relatively strong attachment to Israel, reporting a very great feeling of being

Journal of Ethnic and Migration Studies 9

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Table 1. Descriptive statistics of independent variables, by groups, first generation.

Age at migration Labour market position PTM White-collar jobs Blue-collar jobs Does not belong to the labour force Tenure in the country (in years) Marital status % Married Israeli friends % Yes % No % Does not have friends Years of schooling % Did not attend school % Academic degree Discrimination (% yes) Workplace Government offices Stores and shopping Entertainment establishments Attachment to Israel Israeli identity (% to a great extent) Gender % Female N (3373)

Ethiopia

MENA

FSU

EUAM

39.8 (11.8)

43.2 (13.5)

42.2 (13.2)

40.0 (13.4)

0.8 7.5 34.8 56.9 13.2 (5.7)

12.4 17.5 17.1 53 12.7 (5.1)

16.6 19.2 28.2 36.1 13.3 (4.4)

29.6 20.2 11.3 38.9 13 (5.3)

75.4

78.6

70.3

81.3

34.8 49.1 16.1 2 (3.7) 69 0.4

69.4 23.4 7.2 13.2 (4.2) 1.3 22.6

50.4 39.2 10.5 14.1 (3.1) 0.4 45.9

81 17.1 1.9 15.5 (3.6) 0 51.0

42.3 20 9.8 15

15 10.5 13 9.6

36.2 20 13 12

18.5 8.8 7 9.6

52.5

43.2

22.5

36.1

54.4 480

49.1 234

58.1 2155

55.4 504

Israeli (53%, 43% and 36%, respectively); among FSU immigrants, the sense of attachment is much lower (23%). As mentioned, research has shown that higher levels of linguistic assimilation result from exposure to the host society language in the origin and receiving countries; hence, immigrants with Hebrew training in both countries are expected to be more fluent in this language than those without prior training. Table 2 displays self-reported pre-migration levels of Hebrew proficiency in reading, speaking, and writings and levels of immigrants’ investment in language acquisition upon arrival in Israel. Before immigrating to Israel, some respondents had a minimal baseline for further language improvement. Whereas most immigrants from Ethiopia and the FSU had no knowledge of Hebrew before arrival, almost half of the EUAM and MENA respondents reported on some basic knowledge. Many of these immigrants probably acquired some familiarity with Hebrew before immigrating to Israel through their active participation in Jewish communities in the country of origin (attendance at Jewish schools, involvement in synagogue activities and youth movements). As we shall show, although previous training in country of origin does not always promise

10

R. Raijman et al.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Table 2. Descriptive statistics of independent variables, by groups, first generation. Language

Ethiopia

MENA

FSU

EUAM

Level of Hebrew proficiency before arrival (1 = not at all, 5 = very fluent) Speaking Not at all Weak + mediocre Fluent + very fluent Reading Not at all Weak + mediocre Fluent + very fluent Writing Not at all Weak + mediocre Fluent + very fluent Ulpan attendance (%) 1–4 months 5–6 months >7 months Did not attend ulpan Current Hebrew proficiency (1 = not at all, 5 = very fluent). Speaking Not at all Weak + mediocre Fluent + very fluent Reading Not at all Weak + mediocre Fluent + very fluent Writing Not at all Weak + mediocre Fluent + very fluent N (3373)

1.1 (0.3)

1.8 (0.9)

1.2 (0.4)

2 (1.1)

1.1 (0.4) 93 6.6 0.3 1.1 (0.4) 93 6.7 0.3 1.1 (0.3) 93 7 –

1.6 (0.9) 55.4 40.6 4 2.0 (1.1) 43.9 44.4 11.7 1.8 (1.0) 48 44.8 7.2

1.2 (0.4) 84.9 14.9 0.2 1.2 (0.5) 84.8 14.8 0.4 1.2 (0.4) 86.3 13.4 0.4

1.9 (1.1) 46.2 42.1 11.7 2.1 (1.2) 44 40.1 15.9 1.9 (1.1) 48.1 40.6 11.3

8.2 14.9 76.9 15.8 1.9 (1)

29 44.6 26.3 19.7 3.3 (1)

32.3 44.2 23.4 19.2 2.6 (1.1)

29.4 40.9 29.7 23.6 3.4 (1.1)

2.4 (1.1) 26.7 57.7 15.6 1.8 (1.1) 55.7 35.7 8.5 1.7 (1.0) 57.8 35.5 6.7 480

3.5 (1.0) 1.7 52.1 46.1 3.3 (1.2) 6.1 50.2 43.7 3.1 (1.2) 7.9 57.7 34.4 234

2.9 (1.1) 10.7 57.3 31.9 2.5 (1.2) 22.7 55 22.2 2.4 (1.2) 26.4 54.7 18.9 2155

3.7 (1.1) 2.8 38.2 59 3.4 (1.2) 6.2 45.6 48.2 3.2 (1.2) 8 52.2 39.8 504

good command of Hebrew upon arrival, it does serve as a stepping stone for linguistic assimilation in the future. To obtain additional information about investment in acquiring language skills, we asked respondents how many months they studied Hebrew formally in the ulpan. The data show that ulpan attendance was the modal form as overall 80% of all migrants invested in language learning. However, not all ulpan learners completed the course. With the exception of Ethiopians, dropout rate was about 30% (leaving the ulpan after fewer than 5 months). The reasons for dropout were mostly related to the need to find a job to make a living in the new country. Notably, 67% of Ethiopian migrants reported spending over 7 months at the ulpan.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 11

To learn about current levels of Hebrew proficiency, respondents were asked to rate their command of Hebrew on a 5-point scale from 1 (low) to 5 (high) in speaking, reading and writing. Table 2 (bottom panel) presents averages as well as percentages of language proficiency. Overall, all groups display a linguistic hierarchy, with speaking at the top (the easiest skill to learn) and writing at the bottom (the most difficult skill to learn).The reading skill falls in between speaking and writing. Uneven levels of Hebrew fluency are evident across groups. EUAM and MENA immigrants have the highest level, followed by FSU immigrants, and Ethiopians displaying the lowest levels. The relatively low levels of Hebrew proficiency in FSU immigrants might be explained by their higher exposure to a Russian-speaking community, a kind of language ‘ghetto’, which reduces the need and opportunity to use Hebrew. Although the Ethiopian immigrants report the highest levels of ulpan attendance, they display the lowest levels of Hebrew proficiency. This finding is not surprising considering that over 70% of the first-generation immigrants from Ethiopia who arrived after 1990 are in fact illiterate in Amharic. Illiteracy in the mother tongue indeed proved one of the main factors predicting low levels of proficiency in the language of the receiving society (Esser 2006). Although the descriptive data reveal interesting and meaningful differences across groups, it is not clear whether and to what extent the differences in language proficiency that the groups attained in the host country can be attributed to ethnic origin or to differences in socio-demographic attributes. Accordingly, we conducted multivariate analyses, presented below, to examine the extent language outcomes are influenced by ethnic origin net of immigrants’ socio-demographic attributes. Explaining Hebrew Proficiency in the First Generation In this section, we use the theoretical model outlined at the beginning of the paper to identify determinants of Hebrew proficiency in post-1990 immigrants to Israel. Table 3 presents the results of the ordered logit regression model and shows the unstandardised coefficients and standard errors for the variables in the model.14 The proportional odds model specifies a different intercept for each level of the dependent variable (except the highest), but assumes the coefficients for the explanatory variables to be identical at all levels. A measure of explanatory power is provided by means of a pseudo-R2 (Nagelkerke R2). We estimated four models: models 1 and 3 are baseline models including only origin or linguistic groups; in models 2 and 4 we add the remaining explanatory variables. The explanatory variables in the equation are: Immigrant/Linguistic groups; [Incentives] labour market position (PTM jobs, white-collar jobs, blue-collar jobs, unemployed and out of the labour force [reference category]); [Exposure] level of Hebrew proficiency before arrival, number of months attending Hebrew courses in Israel, years since migration, marital status and percent having Israeli friends; [Efficiency] age at migration and education. We also control for gender (male = 1), commitment to the new society (Israeli identity) and perception of social climate towards immigrants (feelings of discrimination).

12

R. Raijman et al. Table 3. Ordinal regression coefficients for language proficiency, first generation.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Model 1

Threshold Don’t know Weak Mediocre Fluent Ethiopian (Ethiopian = 1) Other African (Other African = 1) Europe−America (Europe−America = 1) Language (English speakers = 1) Language (Amharic speakers = 1) Language (Spanish speakers = 1) Language (French speakers = 1) Language (Other speakers = 1) Age at immigration Years of education Marital status (married = 1) Gender (male = 1) Years since immigration PTM White-collar jobs Blue-collar jobs Self-definition as Israeli Hebrew proficiency before arrival Discrimination Friends (Israeli friends = 1) Friends (no friends in Israel = 1) Time studied in ulpan (1–4 months = 1) Time studied in ulpan (5–6 months and up = 1) Time studied in ulpan (7 months and up = 1) Nagelkerke

Model 2

Model 3

Model 4

B

SE

B

SE

B

SE

−1.45*** −0.41*** 0.07 1.3*** −1.23*** 1.02***

0.05 0.04 0.04 0.05 0.1 0.12

0.36 2.07*** 2.9*** 4.98*** −0.5** 0.95***

0.31 0.32 0.32 0.32 0.18 0.16

−1.48*** −0.44*** 0.04 1.26***

0.05 0.04 0.04 0.05

1.25***

0.09

0.38**

0.12 1.25***

0.14

−0.26

0.18

0.1

−0.51*

0.18

1.24***

0.16

0.98*

0.19

1.23***

0.14

0.71***

0.19

0.32**

0.1

0.54***

0.12

−1.29*

0.14

B

SE

0.45 2.17*** 3.0*** 5.09***

0.31 0.32 0.32 0.33

−0.08*** 0.13*** −0.03

0.00 0.01 0.08

−0.08*** 0.13*** 0.00

0.00 0.01 0.08

0.06 0.05*** 1.16*** 0.68*** 0.25* 0.32*** 1.13***

0.08 0.01 0.13 0.11 0.1 0.04 0.08

0.05 0.05*** 1.13*** 0.62*** 0.18* 0.32*** 1.24***

0.08 0.01 0.13 0.11 0.1 0.04 0.08

−0.07 0.81***

0.08 0.09

−0.07 0.84***

0.08 0.09

−0.23

0.13

−0.22

0.14

0.56***

0.11

0.58***

0.11

1.22***

0.11

1.24***

0.11

1.52***

0.12

1.54***

0.12

0.62

Notes: *p < .05; **p < .01; ***p < .001, Very fluent is the omitted category.

0.14

0.62

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 13

Models 1 and 3 reveal that Ethiopians (Amharic speakers) have lower levels of Hebrew proficiency than FSU immigrants (Russian speakers). This gap is likely to decrease after controlling for all relevant variables but still remains significant. EUAM and MENA immigrants have higher levels of Hebrew proficiency than FSU immigrants; the relative advantage remains even after controlling for other language-related attributes. The data in model 3 show that with the exception of Amharic speakers (Ethiopians), all linguistic groups have higher levels of Hebrew proficiency than Russian speakers (the omitted category). After controlling for all relevant explanatory variables (model 4), the effects of some linguistic groups remain significant (Spanish, French, other) while the effect of English is not significant. The data in models 2 and 4 lend support to the ‘economic motivation hypothesis’. The coefficients for labour market position suggest that those employed in the labour market, regardless of occupational position, are more likely than those out of the labour force to acquire high levels of Hebrew proficiency. We also found differences in Hebrew proficiency between those in PTM occupations and those in service occupations, clerical or blue-collar jobs or out of the labour force: the higher the immigrant’s occupational status, the greater the command of Hebrew. In accordance with the ‘exposure hypothesis’, Hebrew proficiency is more pronounced among immigrants who have been in the country longer and tend to have social networks that include Israeli friends. Likewise, language training, in country of origin or in country of destination, affects language proficiency. Specifically, immigrants who arrive in Israel with language skills and those attending ulpan are likely to report higher levels of Hebrew proficiency than those who have not invested in such types of language training. The impact of ulpan attendance on Hebrew acquisition varies. Those who invest more in the ulpan (more than 7 months) are more likely to acquire higher levels of proficiency. The data suggest that attending but dropping out early (after fewer than 5 months) also has a beneficial effect as against non-attendance. In accordance with the ‘efficiency hypothesis’, we found a positive correlation between language acquisition and age at immigration and education. Hebrew proficiency is higher among immigrants who arrived at younger age, as these individuals are better able to acquire language than their older counterparts. Furthermore, the higher the level of formal education, the more likely is the immigrant to master Hebrew. This is either because highly educated immigrants have accumulated more learning skills or because they are inherently more skilful. Finally, those who feel Israeli to a great extent are more likely to report higher levels of Hebrew proficiency. Note that after controlling for all relevant variables, feelings of discrimination, gender15 and marital status16 have no significant effects on Hebrew proficiency. Estimating Predicted Probabilities For a deeper and clearer interpretation of the impact of the model’s variables on language proficiency (as depicted by the coefficients of the ordinal logistic equation), we calculate the predicted probabilities of each immigrant group attaining a high level of

14

R. Raijman et al.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

fluency in Hebrew. The use of predicted probabilities facilitates a clear comparison of the net values for linguistic outcomes and indicates the magnitude of effects of the predictors on language fluency. For each immigrant and linguistic group, we assume baseline profile for people who display the average characteristics.17 We then change the values of the different predictors to evaluate the change in predicted probabilities associated with selected values of the predictors. Predicted probabilities enable us not only to reveal the impact of selected explanatory variables on Hebrew proficiency but also to explore the extent the impact differs across immigrant groups and linguistic groups. Figures 1a and 1b show the predicted probabilities of reporting very fluent levels of Hebrew proficiency for each origin and linguistic group considering the impact of

Figure 1. (a) Predicted probabilities for very fluent Hebrew proficiency by immigrant groups and labour market position. (b). Predicted probabilities for very fluent Hebrew proficiency by linguistic groups and labour market position.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 15

‘economic incentives’ (labour market position). As expected, the occupational position that immigrants occupy in the labour market plays an important role in determining probabilities of acquiring higher levels of Hebrew proficiency. For all groups, those who work in PTM occupations have the highest probabilities of attaining a very fluent level of Hebrew—considerably higher than those working in low white-collar jobs and blue-collar jobs, and those out of the labour market. In the following Figures (2a, 2b, 3a, 3b, 4a and 4b), we show the impact of various indicators of ‘exposure’ on fluency. Figures 2a and 2b display the impact of tenure in the country, Figures 3a and 3b the impact of ulpan attendance and Figures 4a and 4b the impact of having Israeli-born friends on the predicted probabilities of each origin and linguistic group reporting a very fluent level of Hebrew proficiency.

Figure 2. (a) Predicted probabilities for very fluent Hebrew proficiency by immigrant groups and tenure in the country. (b) Predicted probabilities for very fluent Hebrew proficiency by linguistic groups and tenure in the country.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

16

R. Raijman et al.

Figure 3. (a) Predicted probabilities for very fluent Hebrew proficiency by immigrant groups and ulpan attendance. (b) Predicted probabilities for very fluent Hebrew proficiency by linguistic groups and ulpan attendance.

The data reveal that the longer the tenure in the country and the more the months of language training, the higher the probability for speaking Hebrew very fluently. Dropouts—those who did not complete 5–6 months at ulpan—do better than immigrants who did not attend ulpan at all. The data also highlight the role of immigrants’ access to local social capital (Israeli-born friends) in linguistic assimilation. The predicted probabilities for reporting the highest level of Hebrew proficiency are greater especially for immigrants with Israeli-born friends than for those without, or with no friends at all. Figures 5a, 5b, 6a and 6b provide graphic illustration of the impact of indicators of efficiency on language acquisition. Figures 5a and 5b show the predicted probabilities

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 17

Figure 4. (a) Predicted probabilities for very fluent Hebrew proficiency by immigrant groups and having Israeli friends. (b) Predicted probabilities for very fluent Hebrew Proficiency by linguistic groups and having Israeli friends.

of reporting very fluent levels of Hebrew proficiency at different levels of education. The figures reveal that the higher the level of formal education, the greater the probability of attaining a high level of Hebrew language proficiency. This is either because highly educated immigrants have accumulated more learning skills or because they are inherently better able to deal with the complexities of learning a second language. Consistent with our theoretical expectations, the data in Figures 6a and 6b indicate that those who arrive at older ages have lower predicted probabilities of mastering Hebrew than those arriving at younger ages. Jointly, all the graphs presented by this paper suggest that the probability of reaching the highest level of Hebrew proficiency is not uniform across immigrant and

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

18

R. Raijman et al.

Figure 5. (a) Predicted probabilities for very fluent Hebrew proficiency by immigrant groups and education. (b) Predicted probabilities for very fluent Hebrew proficiency by linguistic groups and education.

linguistic groups. Ceteris paribus, Ethiopians always have the lowest probabilities, followed by FSU, EUAM and MENA; the latter two groups always display the highest probabilities of acquiring very fluent Hebrew. In particular, they are twice as likely as Ethiopians to attain the category of very fluent The analysis of predicted probabilities by linguistic group provides some interesting insights that are not evident when the respondents are grouped by origin. The relevant data show that speakers of Spanish, French and other languages tend to have the highest probabilities of attaining a very fluent level of Hebrew, followed by speakers of Russian, English and Amharic. The last-named (the Ethiopians) have the lowest probabilities of belonging to the highest category of Hebrew fluency.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 19

Figure 6. (a) Predicted probabilities for very fluent Hebrew proficiency by immigrant groups and age at immigration. (b) Predicted probabilities for very fluent Hebrew proficiency by linguistic groups and age at immigration.

Discussion and Conclusions In this paper, we aimed to improve our understanding of the process by which firstgeneration immigrants in Israel acquire Hebrew proficiency. To do so, we relied on a unique data-set that collected information on relevant factors explaining Hebrew proficiency, especially those related to unique measures of exposure to Hebrew (at origin and destination), perceived discrimination and identification with the host society. We also estimated the separate effects of country of origin and language of origin on immigrants’ gaining proficiency in the destination language. The findings are consistent with those of most previous studies of linguistic assimilation. Although proficiency levels are conditioned on age at arrival, we found clear evidence that Hebrew proficiency rises with exposure to Israeli society (length of

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

20

R. Raijman et al.

stay in the country and contacts with Israelis) and exposure to training in the Hebrew language (in country of origin or in the host society). Immigrants’ access to local social networks plays an important part in structuring the process of language acquisition. Levels of Hebrew proficiency are much higher for immigrants with Israeli-born friends than for those without them. The immigrants’ human capital also affects language acquisition. Those with higher education levels are more efficient in converting exposure and incentives into higher levels of Hebrew proficiency as they possess the necessary cognitive skills to acquire a second language. There is also a positive synergism between language acquisition and attachment to the host country. Those who firmly define themselves as ‘Israeli’ are more likely to report higher levels of Hebrew proficiency. Our study contributed to the literature by providing further insights on the importance of both country of origin and language of origin for understanding immigrants’ linguistic assimilation. Controlling for all relevant factors, the data reveal substantial differences across origin groups: immigrants of MENA and EUAM countries have higher probabilities of reporting the highest level of command of the language than their FSU and Ethiopian counterparts. The lower levels of proficiency among first-generation Russian-speaking FSU and Amharic-speaking Ethiopian immigrants might reflect the higher level of ethnic enclosure manifested by patterns of residence and the composition of social networks. Ethnic enclosure hinders language acquisition because immigrants have more opportunities to ‘avoid’ the use of the local language. Apparently, the very existence of a large and residentially concentrated ethnic or linguistic community (as is the case for Ethiopian and FSU immigrants) becomes a potential disincentive for improving their Hebrew skills. Analysis of the relations between Hebrew proficiency and language of origin revealed important differences not explained by ethnic origin. Speakers of Spanish and French attain higher levels of Hebrew proficiency than speakers of Russian, English and Amharic. Other things being equal, the higher probabilities of attaining high levels of Hebrew proficiency among Spanish and French speakers can be explained by the relatively small size of their linguistic communities—a circumstance that obliges most of these newcomers to learn and use Hebrew. Furthermore, we suggest that the relative advantage of French speakers is also explained by their exposure to Arabic in their countries of origin. As Arabic is the language with the least linguistic distance from Hebrew (Beenstock, Chiswick, and Repetto 2001, 44), French speakers who are arriving from MENA or from Europe (France or Belgium)18 might enjoy a relative advantage in acquiring Hebrew skills. The lower probabilities of English speakers achieving high levels of Hebrew proficiency can be attributed to lack of social pressure on them (in contrast to other linguistic groups) to invest in learning the local language. Beenstock, Chiswick, and Repetto (2001) found that linguistic disadvantages for English speakers as compared with other groups increased with tenure in the country due to the slow progress in acquiring Hebrew language skills (p. 44).The fact that English is a lingua franca

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 21

understood and prized in Israel explains the slow process of English speakers’ language acquisition (see Beenstock 1996; Spolsky and Shohamy 1999; Raijman 2013). The lower levels of Hebrew proficiency among Ethiopians (Amharic speakers) are not surprising, considering the high levels of illiteracy in this immigrant group. Despite their higher levels of exposure to language training at the ulpan and the smaller linguistic distance from Hebrew, as compared with their counterparts arriving from other regions of the world, first-generation Ethiopians have not managed to convert exposure into better language outcomes. This finding raises questions about the efficiency of current frameworks for Hebrew learning that need to be adapted to a target population with low skill levels even in their native language. Apart from racial discrimination and lack of education, Ethiopian immigrants attempting to adapt to Israeli society encounter multiple disadvantages, largely due to the language handicap. Although the present study focused on Israel, the findings may have important implications for understanding linguistic assimilation trajectories of immigrants in other immigrant societies. Especially interesting is the positive experience of Englishspeaking immigrants, who might enjoy advantages vis-à-vis other linguistic groups in host societies where English is not the official language but is highly prized as lingua franca. This, in turn, may reduce incentives among English-speaking immigrants to invest in the host society language. In addition, our study highlights difficulties among low educated immigrants, especially those who were illiterate upon arrival, to learn the new language, let alone to acquire reading and writing skills of the new language. Indeed, illiteracy exerts a powerful negative impact on language acquisition even in cases where the linguistic distance from the local language is low (e.g. Ethiopians). The differences in linguistic attainment suggest that the process of socio-economic integration for different origin and linguistic groups might display different patterns. For some immigrant groups, lack of language skills might hinder economic opportunity and access to social resources that enhance integration.19 Proficiency in the host country’s language is relevant not only for economic achievements but also for migrants’ identification with their host society’s culture and values (Hochman and Davidov 2014) and their subjective feelings of well-being (Amit and Litwin 2010). Therefore, from a public policy perspective, state efforts should be focused on enhancing immigrants’ interest in learning and improving their Hebrew proficiency and financing advanced language courses of varying intensity and on a variety of topics associated with the specific cultural and social needs of the diverse immigrant groups. Funding This work was supported by the Israeli Science Foundation [grant number 406/12].

22

R. Raijman et al.

Notes [1]

[2]

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

[3]

[4]

[5]

[6]

[7] [8]

[9] [10] [11]

[12]

[13] [14]

[15]

[16]

Paradoxically, since 1983, Israel—a country of mass immigration—has not included in any systematic way questions on Hebrew proficiency in its population censuses and labour force and income surveys, which are the main sources of publicly available data. Espinosa and Massey (1997, 42) suggest that the effect of labour market position might also indicate that those who display some language ability upon arrival are more able to obtain better jobs or are more likely to be employed in the first place (see also Beenstock 1996). In addition, the linguistic literature points out that due to maturational constraints—which are the product of biologically based neurological processes—immigrants arriving at older ages have greater difficulty than younger counterparts in learning a new language, even when controlling for other relevant factors (Stevens 1992; Espenshade and Fu 1997). Research in Germany has shown that immigrants with very good ability in writing the home country language were more likely to be fluent in German, whereas those who were classified as illiterate in the home country language were less likely to be fluent in German (see Dustmann [1994] and Isphording and Otten [2011]). Most available data-sets for the study of language proficiency do not provide information on pre-migration language learning (Chiswick and Miller 2014), so it is hard to examine the competing claims regarding pre- and post-migration exposure (Beiser and Hou 2000). For examples of research conducted in European countries, see, e.g. Van Tubergen and Wierenga (2011) for Belgium, Van Tubergen and Kalmijn (2009) for the Netherlands and Hochman and Davidov (2014) for Germany. Immigrants arriving from the Asian republics of the FSU are included in FSU; Ethiopians are a separate category. Since 1993, we should also add to these numbers a stable flow of non-Jewish labour migrants and of African asylum seekers (arriving since the mid-2000s). Overall, both documented and undocumented labour migrants and asylum seekers comprise about 3% of the total population of Israel. Because official statistics do not include these populations in data collection, we cannot compare their linguistic outcomes with those of immigrants arriving under the Law of Return. An ulpan (plural ulpanim) is an institute or school for the intensive study of Hebrew. Ulpan is a Hebrew word meaning studio or teaching, instruction. The Immigrant Survey provides information for immigrants aged 27 and older. For this reason, we restricted our analysis to this age range. We determined the level of Hebrew proficiency from respondents’ answers to three questions that were combined to construct an index: ‘How would you evaluate your ability to (i) conduct a conversation, (ii) read, and (iii) write in Hebrew before arrival?’ Answers were on a 5-point scale from 1 = not at all to 5 = very fluent. Because data for EUAM and MENA countries were provided only in major categories, we could not distinguish specific countries of origin. Nevertheless, we believe that these two major categories are rather homogeneous and capture two distinct ethnic groups. Unfortunately, the file does not provide information on ethnicity of the spouse. But given the relatively older age on arrival, we can assume that most spouses are co-ethnic. The index measuring general language proficiency clustered around specific points; therefore, we constructed the following ordinal variable: 1–1.5 = 1; 1.5–2 = 2; 2.1–2.7 = 3; 2.8–3.7 = 4; 3.8–5 = 5). Apparently, the lack of a significant gender effect is due to the similarity of post-1990 immigrant men and women of different ethnic groups along human capital characteristics, rates of labour force participation and experiences of integration in Israel (see Remennick 2004). These findings are similar to those of Chiswick and Miller (1998).

Journal of Ethnic and Migration Studies 23 [17]

[18]

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

[19]

Baseline profile: married men, age at migration = 41.6; years of education = 12.5, tenure in the country = 13.2; PTM occupations, level of self-identification as Israeli = 3; perceived discrimination = 1.2; has Israeli friends, ulpan attendance = 5–6 months; language proficiency before arrival = 1.3. According to Amit and Bar-Lev (2014) a third of Jewish immigrants to Israel from France are also of MENA origin and the rest are second-generation immigrants from MENA. It is already well known that language proficiency may not be an important determinant of employment during the initial phases of immigrants’ incorporation into the host society; however, over time, low proficiency may become an obstacle to socio-economic mobility (Raijman and Semyonov 1998; Beiser and Hou 2000). Nevertheless, for transnational and multi-local migrants whose economic activities are strongly connected to specific linguistic communities or located in their countries of origin, the ability to command the local language becomes less relevant as compared to their counterparts who are active in the labour market in the country of destination. It is suggested that at least 20% of the North American immigrants and nearly half of French immigrants (arriving to Israel in the last 10 years) are multi-local. It is also argued that the number of FSU immigrants spending some time in Israel and in their countries of origin is on the increase (Pupko 2007).

References Adler, S. 2007. “Training and Retraining Programs in Israel.” Paper presented at ICSW Conference, Jerusalem, May 1–11. http://www.euro.centre.org/Jerusalem/files/Adler_Presentation.pdf. Alba, R. D., J. Logan, A. Lutz, and B. Stults. 2002. “Only English by the Third Generation? Loss and Preservation of the Mother Tongue among the Grandchildren of Contemporary Immigrants.” Demography 39: 467–484. doi:10.1353/dem.2002.0023. Amit, K., and S. Bar-Lev. 2014. “Neither Here Nor There: Transnational Identity of French Immigrants Employed in Israeli-French Companies.” Final research report, The Institute for Immigration and Social Integration, Ruppin Academic Center. Amit, K., and H. Litwin. 2010. “The Subjective Well-Being of Immigrants Aged 50 and Older in Israel.” Social Indicators Research 98 (1): 89–104. doi:10.1007/s11205-009-9519-5. Beenstock, M. 1996. “The Acquisition of Language Skills by Immigrants: The Case of Hebrew in Israel.” International Migration 34 (1): 3–30. doi:10.1111/j.1468-2435.1996.tb00178.x. Beenstock, M., B. R. Chiswick, and G. L. Repetto. 2001. “The Effect of Linguistic Distance and Country of Origin on Immigrant Language Skills: Application to Israel.” International Migration 39 (3): 33–60. doi:10.1111/1468-2435.00155. Beiser, M., and F. Hou. 2000. “Gender Differences in Language Acquisition and Employment Consequences among Southeast Asian Refugees in Canada.” Canadian Public Policy – Analyse de Politiques 26 (3): 311–330. Ben-Rafael, E., and M. Ben-Rafael. 2013. Sociologieet Sociolinguistique des Francophonies Israéliennes [Sociology and Sociolinguistics of Francophonies in Israel]. Frankfurt: Peter Lang. Ben-Rafael, E., E. Olshtain, and I. Geijst. 1998. “Identity and Language: The Social Insertion of Soviet Jews in Israel.” In Immigration to Israel, edited by E. Leshem and J. Shuval, 333–356. London: Transaction. Ben-Rafael, E., M. Lyubansky, O. Glockner, P. Harris, Y. Israel, W. Jasper, and J. Schoeps. 2006. Building a Diaspora: Russian Jews in Israel, Germany and USA. Leiden: Brill. Chiswick, B. R. 1998. “Hebrew Language Usage: Determinants and Effects on Earnings among Immigrants in Israel.” Journal of Population Economics 11 (2): 253–271. doi:10.1007/ s001480050068. Chiswick, B. R., and P. W. Miller. 1992. “Language in the Labor Market: The Immigrant Experience in Canada and the United States.” In Immigration, Language and Ethnic Issues: Canada and

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

24

R. Raijman et al.

the United States, edited by B. R. Chiswick, 229–296. Washington, DC: American Enterprise Institute. Chiswick, B. R., and P. W. Miller. 1995. “The Endogeneity between Language and Earnings: International Analyses.” Journal of Labor Economics 13 (2): 246–288. doi:10.1086/298374. Chiswick, B. R., and P. W. Miller. 1996. “Ethnic Networks and Language Proficiency among Immigrants.” Journal of Population Economics 9 (1): 1935. doi:10.1007/PL00013277. Chiswick, B., and P. Miller. 1998. “English Language Fluency among Immigrants in the United States.” Research in Labor Economics 17: 151–200. Chiswick, B. R., and P. W. Miller. 2014. International Migration and the Economics of Language (no. 7880). IZA Discussion Paper. Cohen, Y. 2002. “From Haven to Heaven: Changing Patterns of Immigration to Israel.” In Challenging Ethnic Citizenship: German and Israeli Perspectives on Immigration, edited by D. Levy and Y. Weiss, 36–56. New York: Berghahn Books. Dustmann, C. 1994. “Speaking Fluency, Writing Fluency and Earnings of Migrants.” Journal of Population Economics 7 (2): 133–156. doi:10.1007/BF00173616. Espenshade, T. J., and H. Fu. 1997. “An Analysis of English-Language Proficiency among U.S. Immigrants.” American Sociological Review 62 (2): 288–305. doi:10.2307/2657305. Espinosa, K. E., and D. S. Massey. 1997. “Determinants of English Proficiency among Mexican Migrants to the United States.” International Migration Review, 31 (1): 28–50. doi:10.2307/ 2547256. Esser, H. 2006. Migration, Language and Integration. Programme on Intercultural Conflicts and Societal Integration (AKI). Research Review 4. Berlin: Social Science Research Center Berlin (WZB). Accessed November 30. http://193.174.6.11/alt/aki/files/aki_research_review_4.pdf Evans, M. D. 1986. “Sources of Immigrants' Language Proficiency: Australian Results with Comparisons to the Federal Republic of Germany and the United States of America.” European Sociological Review 2 (3): 226–236. Hochman, O., and E. Davidov. 2014. “Relations between Second-Language Proficiency and National Identification: The Case of Immigrants in Germany.” European Sociological Review 30 (3): 344–359. doi:10.1093/esr/jcu043. Isphording, I. E., and S. Otten. 2011. Linguistic Distance and the Language Fluency of Immigrants (no. 274). Ruhr Economic Papers. Jasso, G., and M. R. Rosenzweig. 1990. The New Chosen People: Immigrants in the United States. New York: Russell Sage Foundation. Levy, D., and Y. Weiss, eds. 2002. Challenging Ethnic Citizenship: German and Israeli Perspectives on Immigration. New York: Berghahn. Mesch, G. S. 2003. “Language Proficiency among New Immigrants: The Role of Human Capital and Societal Conditions: The Case of Immigrants from the FSU in Israel.” Sociological Perspectives 46 (1): 41–58. doi:10.1525/sop.2003.46.1.41. Pupko, I. 2007. “New Forms of Aliyah.” Accessed October 17. http://www.ynetnews.com/articles/ 0,7340,L-3421499,00.html. Raijman, R. 2009. “Immigration in Israel: A Map of Trends and Empirical Research, 1990–2007.” Israeli Sociology 10 (2): 339–379. Raijman, R. 2013. “Linguistic Assimilation of First-Generation Jewish South African Immigrants in Israel.” Journal of International Migration and Integration 14: 615–636. doi:10.1007/s12134012-0257-1. Raijman, R., and A. Kemp. 2010. “The New Immigration to Israel: Becoming a de-facto Immigration State in the 1990s.” In Immigration Worldwide, edited by U. Segal, N. Mayadas, and D. Elliot, 227–243. New York: Oxford University Press. Raijman, R., and M. Semyonov. 1995. “Modes of Labor Market Incorporation and Occupational Cost among New Immigrants to Israel.” International Migration Review 29: 375–393. doi:10.2307/2546786.

Downloaded by [University of Haifa Library] at 10:59 26 November 2014

Journal of Ethnic and Migration Studies 25 Raijman, R., and M. Semyonov. 1997. “Gender, Ethnicity, and Immigration: Double Disadvantage and Triple Disadvantage among Recent Immigrant Women in the Israeli Labor Market.” Gender and Society 11 (1): 108–125. doi:10.1177/089124397011001007 Raijman, R., and M. Semyonov. 1998. “Best of Times, Worst of Times, and Occupational Mobility: The Case of Soviet Immigrants in Israel.” International Migration 36 (3): 291–312. doi:10.1111/1468-2435.00048. Rebhun, U. 2014. “English Language Proficiency among Israeli Jews and Palestinian Arabs in the United States, 1980–2000.” International Migration Review. doi:10.1111/imre.12064. Remennick, L. 2003a. “From Russian to Hebrew via HebRush: Intergenerational Patterns of Language Use among Former Soviet Immigrants in Israel.” Journal of Multilingual and Multicultural Development 24: 431–453. doi:10.1080/01434630308666509. Remennick, L. 2003b. “What Does Integration Mean? Social Insertion of Russian Immigrants in Israel.” Journal of International Migration and Integration/Revue de l'integrationet de la migration internationale 4 (1): 23–49. doi:10.1007/s12134-003-1018-y. Remennick, L. 2004. “Language Acquisition, Ethnicity and Social Integration among Former Soviet Immigrants of the 1990 in Israel.” Ethnic and Racial Studies 27: 431–454. Semyonov, M., R. Raijman, and D. Maskileyson. 2014. “Ethnicity and Labor Market Incorporation of Post-1990 Immigrants in Israel.” Population Research and Policy Review. doi:10.1007/ s11113-014-9345-6. Spolsky, B., and E. Shohamy. 1999. The Languages of Israel. Policy, Ideology and Practice. Clevedon: Multilingual Matters. Stevens, G. 1986. “Sex Differences in Language Shift in the United States.” Sociology and Social Research 71 (1): 31–36. Stevens, G. 1992. “The Social and Demographic Context of Language Use in the United States.” American Sociological Review 57 (2): 171–185. doi:10.2307/2096203. Stevens, G. 1999. “Age at Immigration and Second Language Proficiency among Foreign-Born Adults.” Language in Society 28: 555–578. doi:10.1017/S0047404599004030. Tannenbaum, M. 2009. “What’s in a Language? Language as a Core Value of Minorities in Israel.” Journal of Ethnic and Migration Studies 35: 977–995. doi:10.1080/13691830902957742. Van Tubergen, F., and M. Kalmijn. 2009. “Language Proficiency and Usage among Immigrants in the Netherlands: Incentives or Opportunities?” European Sociological Review 25 (2): 169–182. doi:10.1093/esr/jcn043. Van Tubergen, F., and M. Wierenga. 2011. “The Language Acquisition of Male Immigrants in a Multilingual Destination: Turks and Moroccans in Belgium.” Journal of Ethnic and Migration Studies 37: 1039–1057. doi:10.1080/1369183X.2011.572476.