The Legacy of the Linguistic Fence: Linguistic Patterns among ultra-Orthodox Jewish Girls

This study examined linguistic patterns in the Jewish ultra-Orthodox community in Israel, a group that has rarely been studied from a sociolinguistic perspective. Participants were 92 girls, 10-12 years old, who attend a school where Yiddish is the language of instruction and Hebrew, Israel’s official language, is studied only in religious contexts. Results show that the girls use and prefer to use Yiddish in most contexts and rate their fluency level higher in Yiddish than in Hebrew. Their appreciation of Yiddish was significantly correlated with negative attitudes toward Hebrew. Relationships with parents had no linguistic effects. Findings are discussed in light of the role of both languages in their community, the uniqueness of this linguistic minority group, especially in comparison with immigrants, the impact of group ideology, and the relevance of emotional correlates of language usage at both individual and community levels. On Language and Minorities Language is the means through which we are socialized into our culture. Through language, the cultural heritage of the past is received, reshaped, and bequeathed to the following generations (Skutnabb-Kangas, 1981; Sung, 1985). In this context, we can view language use among ethnolinguistic minorities as associated with identity issues. Language therefore is a vehicle of symbolic value, including for the transfer of ideology. As Even-Zohar states: “By adopting a certain language, a certain population or a certain group in society declares what identity it wants to show to itself as well as to the rest of the world” (1986, p. 126). Language has often been used to develop, strengthen, or even create a cultural, national, or other form of group identity. Herder (1986; 1772) argued that language is not only a crucial tool for a nation, but also a legitimate and natural means of justifying its existence and an aesthetic-affective realization of a community’s internal essence. In his view, languages create nations and, following this line of thought, we could view languages as strengthening and legitimizing the existence of smaller groups and cultures within larger societies (see also Fox, 2003). Israel’s multicultural and multilingual setting is a fascinating locale for the exploration of linguistic patterns among minority groups. Since its establishment as an independent state in 1948 and into the present, Israel has received Jews from all over the world. The country also has a large Arab minority, constituting close to twenty percent of the population. Although these circumstances prompted extensive research exploring aspects of acculturation or linguistic patterns among Israel’s ethnolinguistic or cultural minorities, a group rarely studied in these contexts is the ultra-Orthodox Jewish population. The ultra-Orthodox differ from other Orthodox or secular Jews in a number of ways, all characterized by the endorsement of voluntary segregation patterns (living in separate neighborhoods), dress code (extremely modest for women, specific traditional attire for men), 75 Heritage Language Journal, 7(1) Winter, 2010 traditions, and overall lifestyle. This group is heterogeneous regarding its members’ countries of origin, study methods of Holy Scriptures, specific features of clothes, subgroup leadership patterns, and attitudes towards the State of Israel. All these elements entail further repercussions for the daily life of these groups and often act as a visible or invisible barrier between them and their surroundings (Baumel, 2003; Grylak, 2002; Rose, 2006). Nevertheless, their common characteristics have allowed researchers, statisticians, and policymakers to relate to them as members of one, though varied, subgroup. Historical Perspective Most researchers view the communal patterns prevailing in today’s ultra-Orthodox communities as the result of two events, more than a century apart: the Haskalah [Jewish Enlightenment] movement and the Holocaust. The Haskalah was an intellectual movement active during the eighteenth and nineteenth century in European Judaism, which had sought to expand the traditional educational curriculum by adding secular studies, European languages, and Hebrew. The ultra-Orthodox, known as Haredim (literally fearful, referring to fear of the Lord in the verse “Hear the word of the Lord, you who tremble [haredim] at his word” [Isaiah, 66:5]) rejected these attempts, claiming that Haskalah followers had succumbed to the lure of modernity and had abandoned genuine Judaism. The Holocaust is another reason adduced by contemporary ultra-Orthodox groups for their current communal patterns. Not only were six million Jews murdered during the Holocaust, but most centers of Jewish learning were destroyed. Contemporary Haredi society perceives itself as the heir of pre-Enlightenment Judaism and the last remaining ember of traditional Jewish society. Consequently, it feels compelled to endorse segregation to preserve what it considers the most authentic Jewish values and modes of life (Glinert & Shilhav, 1991; Shilhav & Friedman, 1989). Some scholars, however, reject this explanation. Rather than as a continuation of pre-modern Jewish society, they view the social organization of today’s Haredi groups as a “sectarian manifestation” (Katz, 1984, p. 251; see also Liebman, 1964; Katz, 1986), a new, defensive response to the surge of post-traditional society. In Israel, the ultra-Orthodox make up about seven percent of the present Jewish population. Several large ultra-Orthodox communities can also be found in the United States, Belgium, and the United Kingdom. Generally, they choose segregation from the secular majority, creating “a defended territorial enclave within which they can produce and reproduce what they regard as the ‘holy community,’ without being threatened by the behavioral patterns and conduct of the surrounding secular society” (Hason, 2001, p. 314). They have their own religious institutions and educational system, including a special curriculum and separate schools for boys and girls. They also have their own shops and avoid contact with mass culture, including television, theatre, movies, secular books (Baumel, 2003; Fader, 2001; Grylak, 2002; Hason, 2001; Heilman & Friedman, 1991). Resembling the processes that led to the creation of separate communities in the wake of the Haskalah movement, present-day ultra-Orthodox segregation is also widely regarded as a response to the cultural threat posed by the secular majority. Segregation, however, does not entirely prevent interactions with the surrounding secular society. Just as ultra-Orthodox groups overlap geographically with the secular society both in Israel and in the Diaspora, some forms of economic and cultural interchange are carried out between these groups and secular communities. But these interchanges are limited in scope and depth, and are approached with 76 Heritage Language Journal, 7(1) Winter, 2010 distrust and aversion by most Haredi educators (Caplan, 2003; Glinert & Shilhav, 1991; Shilhav & Friedman, 1989). One perspective from which to view these relationships is suggested by the “contact theory” (Allport, 1954). According to Forbes, the “more contact between individuals belonging to antagonistic social groups (defined by custom, language, beliefs, nationality, or identity) tends to undermine negative stereotypes and reduce prejudice, thus improving inter-group relations by making people more willing to deal with each other as equals” (1997, p. ix). The Haredi community invests great effort in avoiding such contacts, presuming it thereby minimizes the potential for evoking positive emotions towards the secular group. Since a more positive perception of the “outside” could induce individuals to leave, these contacts are not in the community’s best interest (see also Yadgar, 2003, on indications of devaluation of the other as a tool for strengthening one’s own collective identity). In Ogbu’s (1992) conceptualization, the ultra-Orthodox constitute an autonomous minority group rather than an ethnolinguistic immigrant minority group. The option of identity change is thus open—some ultra-Orthodox become secular and members of the secular majority join the ultra-Orthodox community. Generally, Haredi society tries to fend off attempts to abandon the group, particularly among the younger generation. For this purpose, it surrounds itself with several fences, as Baumel states: “Traditionally, Haredi society has always tried to segregate itself from surrounding foreign, non-Jewish, influences, whose introduction into the Haredi lifestyle would destroy its moral and cultural underpinnings. Although the extent of this segregation differed from place to place, the form it took was similar: limiting anything but economic contact with non-Jews and secular Jews, and to a certain extent even with Orthodox but non-Haredi Jews” (2003, p. 62). Or, as Fader states: “Both the Jewish dietary laws and the censoring of books are attempts to control what is put into the body and mind” (2001, p. 272). The Linguistic Aspect Language plays an interesting role in this voluntary segregation; most ultra-Orthodox in Israel use Modern Israeli Hebrew as their main language of communication, both inside and outside their community. Their use of Modern Hebrew is equivalent to their mastery of English in the United States or French in Belgium, where they live as a minority group and view these languages as the lingua franca of their external reality, aware of their importance for their physical survival and social mobility (Bogoch, 1999; Fader, 2001; Isaacs, 1999a; Mitchell, 1998). Some of the more extreme subgroups use Yiddish as their primary language of communication and relate to Hebrew, one of Israel’s official languages, as a holy tongue to be reserved for praying and for the study of traditional texts (Baumel, 2003; Fishman, 1991; Hamburger, 2005; Isaacs, 1999a, 1999b; Tannenbaum, Abugov, & Ravid, 2006, 2007). All ultra-Orthodox groups pray and study religious texts in what they call ‘Loshen Koydesh,’ [Holy Tongue], which is essentially Hebrew and Talmudic Aramaic and is basically (though not exclusively) a written language (Baumel, 2003; Ben Rafael, 1994; Shandler, 2000). The wider meaning of this linguistic choice should be placed in historical context. The Zionist endeavor that culminated in the creation of the State of Israel involved a revolutionary linguistic process, the revival of Hebrew as a spoken language. Until the destruction of the First Temple (587 BC), Jews living in the Land of Israel had spoken mainly Hebrew. The impact of Aramaic grew during the Babylonian exile and, following the Jewish exile after the Roman destruction of the Second Temple (70 AD), Hebrew lost its function as a vernacular and was mainly reserved for literary-religious purposes. In most Jewish communities around 77 Heritage Language Journal, 7(1) Winter, 2010 the world, Hebrew was considered ‘the Holy Tongue,’ while the local language or one of several Jewish languages developed in the Diaspora served for day-to-day communication. The revival of spoken Hebrew as a cornerstone of the Zionist enterprise during the late nineteenth and twentieth centuries was accompanied by a determined campaign to eliminate the use of other languages, particularly Yiddish, in daily communication. Hebrew thus became one of the main factors in the efforts to create a melting pot of Jews arriving from many lands. The newly established State of Israel indeed adopted a melting pot policy concerning immigration, and vast efforts were invested in ensuring that everyone learned Hebrew (Ben Rafael, 1994; Ravid, 1995). The Hebrew revival was presented as symbolizing the battle between Zionism, or the ‘new Judaism,’ and the traditional religious way of life in the Diaspora. Hebrew was associated with Israel as an independent state characterized by modernity, while Yiddish was associated with the opposite: “Yiddish ... evoked images of the religious ghetto Jew, odious to the Zionists who promoted the image of a newly ‘liberated,’ secular, Hebrew-speaking homo hebraicus” (Mitchell, 1998, p. 192; see also Glinert, 1993, and Yadgar, 2003). During the struggle waged in the early twentieth century to agree on an official language for the evolving society (known as ‘the linguistic war’), Yiddish was targeted as the bitterest foe, the icon of all that was negative in the Diaspora. Yiddish, then, becomes synonymous with the pre-modern, traditional Diaspora Jew. Yet, whereas Zionists view this simplistic image as negative, the ultra-Orthodox groups embrace it as positive for the same reasons. Members of this community associate mastery of Yiddish with intellectual prowess and greater religious piety, and thus with higher social status. Preserving Yiddish is considered the authentic realization of Jewish culture and identity (Benor, 2004; Davids, 2000; Fishman, 2002; Mitchell, 1998; Rose, 2006), an idea that resonates with Herder’s view (Herder, 1986; Fox, 2003). Mitchell (1998) points to cases of ultra-Orthodox American Jews who learn Yiddish or deliberately speak in accented English, examples that further strengthen this perception. Language and Family Communication The present study focuses on the attitudes of ultra-Orthodox children toward Yiddish and Hebrew, and particularly on their attitudes to language usage in communication with their parents. Several studies of immigrant children report that children’s language choices and attitudes are associated with or affected by parents’ own attitudes toward the majority and minority languages (e.g., Imbens-Bailey, 2000; Oliver & Purdie, 1998), or by the quality of family relations (e.g., Luo & Wiseman, 2000; Tannenbaum & Berkovich, 2005; Tannenbaum & Howie, 2002). The present study examines both ultra-Orthodox children’s language use and attitudes, both in general and in the family. Although we did not explore the family dynamics directly, we did obtain information on family relations as well as on attachment patterns, viewing them as especially significant for parent-child relationships. Attachment Theory Attachment theory, first developed by Bowlby (1969, 1973), attempts to conceptualize and explain the tendency of human beings to form strong affectionate bonds with significant others, and the psychological implications of these bonds. According to attachment theory, the 78 Heritage Language Journal, 7(1) Winter, 2010 infant, and later the child, through continual interaction with the parents, develops internal working models of the attachment figure(s) as one who is or is not accessible for support and protection, especially in circumstances that involve difficulties or stress (Bowlby, 1973). Accordingly, two main types of attachment patterns have been identified, secure and insecure, the latter including avoidant and ambivalent modes of attachment, with some researchers also adding the category of disorganized attachment mode. Individuals tend to interpret their experiences as consistent with their existing internal working models, and continuity in attachment modes is thus expected throughout the life cycle (e.g., Main & Cassidy, 1988; Meyers, 1998). Numerous studies have also shown that early attachment mode is a powerful predictor of various personality characteristics, as well as of social and cognitive skills. For instance, research indicates that attachment mode can predict a child's future ability to cope with frustration and stress, anxiety level, social competency with peers including interpersonal understanding and sensitivity, self-confidence, curiosity, and cognitive performance (e.g., Aspelmeier, 2003; Grossmann & Grossmann, 1991; Jacobsen, Edelstein, & Hofmann, 1994). Research suggests that children across all cultures become attached to their primary caregivers and display a similar range of behaviors in the context of comfort-seeking or stress (Harwood, Miller, & Irizarry, 1995; Sagi, Lamb, Lewkowicz, Shoham, Dvir, & Estes, 1985; van IJzendoorn & Sagi, 1999), although studying this phenomenon requires sensitivity to the ways of displaying attachment in different cultures, and to the meaning of behaviors, emotional reactions, and gestures in relevant contexts (Rothbaum, Weisz, Pott, Miyake, & Morelli, 2000). The Family in the ultra-Orthodox Community Families in ultra-Orthodox communities follow traditional patterns: they are large, with set gender and role divisions, young marriages, and frequent contacts with extended family. Family cohesion is a central value at all stages of the life cycle (Wieselberg, 1992; Woehrer, 1988). Besides being expected to excel as mothers, however, women are also expected to be the main breadwinners, freeing their husbands from financial responsibilities that could take away time from Torah study (Benor, 2004; Blumen, 2002; El-Or, 1992; Fader, 2001; Isaacs, 1999a). The ultra-Orthodox community educates its children in a separate system, which contributes heavily to the continuation of their strict religious way of life and to the solidarity of its society. Boys and girls study separately from an early age; boys start their studies at the heder (traditional Jewish primary school) at three and generally read Loshen Koydesh fluently by the age of five, to begin a lifelong study of traditional texts. Girls usually go to kindergarten from the age of three until the age of six and then to elementary and secondary school. They usually have more limited access to sacred texts, although they learn to read and write in liturgical Hebrew and, in Yiddish-speaking communities, in Yiddish as well (Baumel, 2003; Benor, 2004; Bogoch, 1999; Fader, 2001; Rose, 2006; Shandler, 2000). Since the women work outside the home, they are often exposed to the vernacular more than the men, both in Israel and in other countries (e.g., Fader, 2001; Isaacs, 1999a; on women as change agents in the ultra-Orthodox community, see Blumen, 2002). The Present Study This study had two aims. The first was to explore linguistic patterns among bilingual girls in the ultra-Orthodox community—their attitudes toward Hebrew and Yiddish, their use of these 79 Heritage Language Journal, 7(1) Winter, 2010 languages, and their evaluation of their proficiency in both languages. The second was to assess the relationship between language use patterns and family relations relying on instruments that, in previous studies, had been applied to immigrant groups (Tannenbaum & Howie, 2002). The exploration of linguistic patterns correlated with family relations is based on findings that emerged in studies of immigrant families, and rests on the following rationale: given the central role of language as a symbol of individual, cultural, and group identity, we expect the quality of affective relationships between parents and children in families belonging to ethnolinguistic minorities to be associated with the extent of children’s use of the parents’ language and their attitudes toward it. In the present study, we focused on a population of girls from the ultra-Orthodox community who study at a school, chosen by the girls' parents, where Yiddish is the language of instruction. Yiddish, then, gains a status similar to that of a mother tongue in its explicit characterization by the parents of children in our study as a valuable and significant language, functioning both as a symbol of group identification and of family values. Our aim was to explore whether specific family relations affect attitudes toward the use of Yiddish and Hebrew, two symbolically charged languages. We hypothesized that the girls would report more favorable attitudes to Yiddish than to Hebrew. In addition, we expected a positive correlation between the quality of parent-child attachment relationships and attitudes to Yiddish, with girls showing a pattern of secure attachment expressing more favorable attitudes to Yiddish than those showing insecure attachment patterns, both in terms of usage and of attitudes to the language. Method Participants The sample consisted of 92 ultra-Orthodox bilingual girls, aged 10-12. The girls live in several ultra-Orthodox neighborhoods in the center of Jerusalem and attend a school where Yiddish is the language of instruction, although some of them come from Hebrew-speaking families. At this school, Hebrew is studied only in a religious context, as a holy tongue. Materials Three instruments were used: 1. Language Use and Preference Questionnaire (LUP) (Tannenbaum, 2003). This 30-item instrument is an adaptation of a language maintenance questionnaire used with immigrant children in Australia and Israel. It focuses on children’s language use and preference in a range of interactions with their parents, differentiating between mother and father. Each item describes a specific interaction scenario (e.g., “I don’t feel very well and I am asking my mum for a drink”). In this study, the girls were asked to indicate whether they would usually use Yiddish or Hebrew (termed here ‘Loshen Koydesh’) in this interaction, and whether they would prefer to use Yiddish or Hebrew. Each item could be scored either as 0 (choosing Yiddish), 2 (choosing Hebrew), or 1 (choosing both). That is, the lower the score, the more the girls use and prefer to use Yiddish rather than Hebrew. Test-retest reliability and internal consistency for both summary scores has been shown to be high (Tannenbaum, 2003). For the purpose of the present study, this questionnaire was professionally translated into Yiddish. 80 Heritage Language Journal, 7(1) Winter, 2010 2. Language Attitudes Measure (LAM). This measure was developed for the present study to assess the girls’ perceptions and attitudes toward Hebrew and Yiddish. The girls were asked to express their level of agreement with 12 statements, written in Yiddish, about Yiddish and Hebrew (again, termed here ‘Loshen Koydesh’) on a 5-point Likert scale. In addition, they were asked to rate their level of fluency in speaking, understanding, reading, and writing in both languages. 3. Attachment Styles Classification Questionnaire for Children (ASCQ) (Finzi, Har-Even, Weizman, Tyano, & Shnit, 1996). This questionnaire aims to classify attachment modes among primary school age children. It includes 15 items on a 5-point Likert scale, focusing on social relationships of children with their peer group. These items were developed based on the notion that attachment is a continuous process throughout the lifespan and that, during latency age, attachment style is manifested largely in the social arena. Each attachment mode (secure, insecure/avoidant, insecure/resistant) is measured by 5 items that are supposed to reflect it. Scoring is done by averaging those 5 items for each mode separately. The child is allocated to the mode in which s/he scores highest. The reliability and validity of this instrument are satisfactory. This questionnaire was also professionally translated into Yiddish. Procedure The girls completed all the questionnaires during school hours. Instructions and clarifications, when needed, were given by the experimenter in Hebrew. It bears emphasizing that entering the school to conduct this study required the approval of several individuals, including the rabbi of the school, the school principal, and the curricular inspector. At their request, we adjusted the questionnaires, changing content and terminology, and omitted some items (e.g., mentions of television or small talk between child and parent, changing such formulations as ‘being angry with mother or father’ to ‘trying to negotiate,’ adding polite terms such as ‘please’ and ‘thanks,’ etc.). In addition, we changed the term ‘Hebrew’ to ‘Loshen Koydesh,’ although it was clear to everyone involved, including the girls, that the reference was to Modern Hebrew. Rather than being technicalities, these changes seem to reflect deeper layers of this community’s values in general, and of their values regarding family and language in particular. They also point to the extent and depth of legitimate intervention by the authorities in the community’s perception, an issue discussed below. Results LUP The Language Use and Preference Questionnaire comprise 30 items (15 questions, each including one use and one preference item). A principal component analysis (with direct oblimin oblique rotation) was performed. Two factors with eigenvalues greater than one emerged, explaining 87.9% of the item variance. One factor was composed of all items related to use and language preference with a child's mother (accounting for 74.6% of item variance), and the other consisted of items related to use and language preference with a child's father (accounting for 13.3% of item variance). Two summary scores were calculated, corresponding to the two factors—each was calculated as the average of all items included in it, with all items having equal weight. The questionnaire's internal consistency was very high (α = .99), and so was each separate score (α = .98 for both scores). Both scores indicate that the girls in this study use, and prefer to use, Yiddish more than Hebrew with both parents (both were less than .5; see Table 1). 81 Heritage Language Journal, 7(1) Winter, 2010 The factor structure of this questionnaire indicates no difference between actual reported use of and preference for a language across all listed situations, and the difference between linguistic interactions with mother and with father was not significant. As a result, and because of the questionnaire's high internal consistency, all questionnaire items were averaged for further analysis to create one summary score, called language approach (see Table 1). LAM The Language Attitudes Measure comprises 12 items. A principal component analysis (with direct oblimin oblique rotation) was performed. Two factors with eigenvalues greater than one emerged, explaining 50.1% of the item variance. One factor was made up of items expressing attitudes in favor of Yiddish (accounting for 34.6% of item variance), and the other factor was made up of items expressing attitudes in favor of ‘Loshen Koydesh’ (LK; accounting for 15.5% of item variance). Internal consistency of the whole questionnaire was medium (α = .73). Summary scores were calculated as the average of all items included for each factor, with all items having the same weight. Internal consistency was satisfactory for the score reflecting attitudes towards Yiddish (α = .83), and medium regarding the score for attitudes towards LK (α = .51). It is interesting to note that the correlation between those two scores was negative (r = -.17) though only marginally significant (p = .054): the more favorable the girls’ attitudes toward Yiddish, the more negative their attitudes toward LK. In addition, correlations between scores emerging from this measure and from the LUP questionnaire were significant: the more the girls practice their Yiddish (use and preference), the more favorable their attitudes toward the Yiddish language (as it emerged from the LAM) and the more negative their attitudes toward Hebrew (LK) and its use in everyday life contexts (see Table 1). In addition, the girls were asked to rate their level of fluency on various skills in Hebrew and Yiddish. Their overall proficiency in Yiddish is significantly higher than their overall proficiency in Hebrew or LK (t85 = -2.19, p < .05), though both are high (see Table 1). In specific skills, the girls rated their level of comprehension and writing as significantly higher in Yiddish than Hebrew (t85 = -2.66, p < .01; t85 = -5.60, p < .001, respectively). They rated their speaking as higher in Yiddish than Hebrew, though not significantly higher, and their reading as higher in Hebrew than Yiddish, again not significantly (see Table 1). All intercorrelations among skills within each language were significant (see Table 1). Finally, overall proficiency in Yiddish was significantly correlated with attitudes towards Yiddish (r = .48, p < .001), and overall proficiency in Hebrew was significantly correlated with attitudes towards LK (r = .25, p < .001). Moreover, correlation between overall proficiency in Hebrew was significantly negatively correlated with attitudes towards Yiddish (r = -.41, p < .001). There was no correlation between overall proficiency in Hebrew and Yiddish, or between attitudes toward Hebrew and toward Yiddish. Table 1. Means, Standard Deviations, and Intercorrelations of Scores Emerging from the LUP and LAM (N = 93) LUP Approach LUP Mother LUP Father LAM LK LAM Yiddish YS YC YR YW YP HS HC HR HW HP M .45 .46 .43 3.47 3.18 4.52 4.67 4.06 4.59 4.46 4.26 4.39 4.35 3.83 4.18 SD .71 .76 .73 .98 .91 .90 .68 1.21 .86 .72 1.01 .93 1.07 1.19 .90 82 Heritage Language Journal, 7(1) Winter, 2010