Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=cjms20
Journal of Ethnic and Migration Studies
ISSN: 1369-183X (Print) 1469-9451 (Online) Journal homepage: http://www.tandfonline.com/loi/cjms20
Identity repertoires among Arabs in Israel
Muhammad Amara & Izhak Schnell
To cite this article: Muhammad Amara & Izhak Schnell (2004) Identity repertoires
among Arabs in Israel, Journal of Ethnic and Migration Studies, 30:1, 175-193, DOI:
10.1080/1369183032000170222
To link to this article: https://doi.org/10.1080/1369183032000170222
Published online: 23 Jan 2007.
Submit your article to this journal
Article views: 634
Citing articles: 35 View citing articles
Journal of Ethnic and Migration Studies
Vol. 30, No. 1, January 2004, pp. 175–193
Identity Repertoires Among Arabs in
Israel
M.AmaraDepartment of Political Science and EnglishBar-Ilan UniversityRamat-GanIsraelkhitam@zahev.net.ilMuhammad Amara & Izhak Schnell
This study investigates the structure and components of the Arab identity repertoire in
Israel. We analyse the relative salience of each component and the structural relations
among the components. In addition we interpret the meanings associated with each
component of the Arab identity repertoire in Israel. Our study differs from traditional
investigations of Arab identities in Israel in considering identity as a multi-dimensional
phenomenon, and in interpreting the meanings that the respondents assign to the
relevant identities. Empirically we find that most Arabs in Israel feel strongly attached
to at least three identities, none of them dominating the others. The correlations among
the identities are low, meaning that they are only marginally affected by the feasibility
of alternative identities and more by the type of practices performed in each respective
milieu. In this respect it is interesting to see that the Palestinian identity is not associated
with citizenship beyond national pride, while the Israeli identity is expected to supply
them with a sense of citizenship. Our finding that different groups constitute for
themselves different relations among the components of the identity repertoire suggests
that our multi-dimensional model supplies a better explanation for the structure of Arab
repertoires of identities.
Keywords: Identity Repertoire; Palestinians; Arabs in Israel; Druze; Religion
Introduction
Non-dominant social groups’ repertoires of identities tend to reflect their problematic
status in the wider society. Particularly complex, in this sense, is the position of
Arab citizens in the state of Israel. The unstable political situation imposed on them,
together with globalisation processes and their consequences on Middle Eastern
politics, set challenges for their complex identity repertoire.1
On the one
Muhammad Amara is Senior Lecturer in the Departments of Political Science and English, Bar-Ilan University,
Ramat-Gan, Israel. Izhak Schnell is Associate Professor in and Chairperson of the Department of Geography
and Human Environment, Tel Aviv University, Israel. Correspondence to: E-mail: khitam@zahav.net.il;
schnell@post.tau.ac.il
ISSN 1369–183X print/ISSN 1469-9451 online/04/010175-19  2004 Taylor & Francis Ltd
DOI: 10.1080/1369183032000170222
Carfax Publishing
176 M. Amara & I. Schnell
hand, they are expected to become Israeli citizens despite the fact that they are
excluded from the national collective, because Israel defines and perceives itself as a
Jewish-Zionist state. On the other hand, they remain marginal to the Palestinian
struggle for a national independent state. Furthermore, they are exposed to Pan-Arabism
and to the rising political Islamic world-views. Globalisation seems to have
tremendous impact on individuals’ and groups’ repertoires of identities exposing
individuals to more complex options (Bar-On 1999; Giddens 1991); it may also
threaten the identities of marginalised groups in the world economy, pushing them
to compensate their low statuses by withdrawing into closed reclusive identities
(Castells 1997). In this context, the main discourse regarding Arab identity repertoire
in Israel, focusing on the relative salience and the degree of congruence between
Israeli civil identity and national identities (either Arab or Palestinian), seems
inadequate. There is a need to study Arab identity repertoires in their full com-
plexity.
This paper aims at investigating the structure of the Arab identity repertoire in
Israel at a time when the establishment of a Palestinian state has become a real
option. Several research questions are at the focus of our investigation: What are the
major components of the Arab identity repertoire in Israel? What is the relative
salience of each component of their identity repertoire? What are the structural
relations among the components of the identity repertoire? What are the meanings
associated with each component of the Arab identity repertoire in Israel?
The Study of Social Identities
Social identity is a complex construct, which may be captured satisfactorily only by
a multi-dimensional approach. Philosophically, the question of identity relates to
two questions: how can we understand human vitality, and what makes two or more
individuals become a social group? The two questions of identity should not be
considered separately but as complementing each other, thereby bridging individuals’
sense of ‘self’ and ‘others’, on the one hand, and broader sociological categories
such as class, nationality, ethnicity and gender that help to create the common social
world (Holland et al. 1998). The gap between the micro and the macro level can be
bridged by the structuration theory, which views human agency and social structure
to be dialectically associated (Giddens 1991). Identity may be considered as a set of
abstract schemes of human subjectivity or the ways in which individuals experience
and view themselves, things they know about themselves and ways in which they
evaluate themselves, as well as the ways empathic others view them (Noy 1995).
These schemes are constituted through human reflections on the ways that they
perform their everyday life practices.
Two theories concerning human social identity are offered by social scientists
nowadays. The first, which has been essential to European modernist social psychologists,
assumes that those quintessentially sociological forces like class, gender and
national ideology constitute one unified, cohesive and relatively stable identity which
individuals internalise. This theory, developed by scholars like Tajfel (1970, 1978),
Journal of Ethnic and Migration Studies 177
dominates the study of Arab identity in Israel, with Rouhana (1997) articulating one
of the more detailed arguments in this direction.
The second, developed in the last decade, tends to represent human identity as a
repertoire of unstable, componential and at times even conflicting sub-identities.
These sub-identities may express a multitude of meanings, which develop in respect
to different social, economic, political and psychological contexts which individuals
may be exposed to, due either to the reality of being minorities in nation-states or
to their exposure to globalisation (Sarup 1996). In this complexity individuals are
required to develop more autonomy in constituting their identities. They may
achieve it by a reflexive process of unifying memories of their past, current schemes
of interpreting their reality and interests, and future aspirations, into one coherent
narrative, instead of stressing sameness to others in reference groups or negation of
strangers (Anderson 1997; Weinreich 1986). Therefore, identity may be considered
balanced if each component of a complex repertoire of identities has the freedom to
move to the centre of the identity field as a response to everyday practices. At the
same time, each component may leave room for alternative components to move to
the centre of the field whenever circumstances change (Lewin 1951; Sarup 1996).
Both theories recognise the ability to understand changes in identities as part of
a broader social change (Bhabha 1994). They differ in assigning salience either to the
other as the mirror of the self, or to everyday life events and human temporality in
constituting identities. In addition, they differ in their implications for the formation
of cohesive social groups. Taylor and Moghaddam (1987) argue that members of
social groups adopt either personal or group strategies in constituting their identities,
but we still lack the knowledge to predict under what circumstances each choice
will be made.
Castells’ argument (1997) can be understood as an attempt to characterise the
social circumstances in which groups adopt the option of a cohesive collective
identity. He identifies three typical strategies for the constitution of collective
identities: legitimising, resistance and projective. Social elites and dominant institutions
in society tend to present legitimising identities to the public in order to
achieve hegemony over society at large. In the current historical context, legitimising
identities in liberal Western countries tend to promote national collective identities,
but also leave some personal autonomy by promoting values such as citizenship,
stability and social order. Agents who tend to define new collectivities inside society
in order to improve their position in society at large may constitute projective
identities; thus, they struggle for social change. Agents draw on cultural sources from
the past in their struggle for social change and the redefinition of societal identity.
This process may give meaning to the new groups’ participation in society, enabling
them to identify with their society. Resistance identity most likely emerges from
marginal groups in society who feel threatened by the elite. Certain individuals,
located in crucial positions in society, gain authority over members of such groups
in order to mobilise them to constitute alternative identities (Calhoun 1994).
Attempts to emphasise collective identities and to negate the legitimised identity may
become major motivations in the constitution of resistance identity. Resistance
178 M. Amara & I. Schnell
identities tend to emphasise their members’ differences from the legitimised identity,
and even present themselves in opposition to them, stressing their positive qualities
in comparison to the others’ negative qualities. In this way, members of resistant
groups may apply the first strategy of constituting identities, stressing similarity and
moral privilege to members of the group with whom they highly identify, and
negating the others. This is the case when socio-spatial strategies are used in order
to segregate (Schnell 2002).
According to the aforementioned typology a constant struggle over social identities
is taking place in society. Different social groups may feel at times the need to
redefine their repertoire of identities in a way that social identities will become more
congruent with their personal feeling of uniqueness, belonging and worth. In some
cases they may struggle for recognition of new social categories, empathy toward
their needs and legitimisation of their set of meanings in society. In other cases they
may struggle to constitute an alternative identity, which negates the legitimised one.
Taylor and Moghaddam (1987) present two other options: either assimilation in the
legitimised identity or acceptance of unsatisfied identity because of lack of opportunities.
They follow Lewin (1951), positing that priority given to one identity or
another depends on the everyday life situation individuals are involved in. When
groups fight to change their social status in society at large or when they feel
threatened by alternative identities, they may adopt collective strategies of identity
formation, while in other circumstances they may adopt more individual strategies
of identity formation. In this sense we ask the following question: Are Arabs in Israel
developing a resistance identity that emphasises collective strategies of identity
formation or alternative strategies?
The Investigation
The study is based on a wider survey aimed at understanding Palestinians’ identity
repertoires in Israel, the West Bank and Gaza Strip. The current paper presents the
results for the Arab repertoire of identities in Israel. The interview survey lasted four
months, from June to September 2000, just before the events of the Al-Aqsa Intifada.
The sample consists of 500 Arab men and women aged 20 and over, selected from
five localities (Nazareth in the Galilee; Arara in the Little Triangle; Hora, a Bedouin
semi-urban settlement in the Negev; Daliyyat Al-carmel, a Druze semi-urban
settlement; and Jaffa, an Arab-Jewish mixed city), representing the different Arab
communities in the pre-1967 borders of Israel. In each locality about one hundred
respondents were interviewed as a layered sample.2
We considered mainly three
layers: socio-economic status (educational attainment and occupation), religion and
religiosity, and gender and age groups. Each interviewer received a list that specified
the socio-demographic characteristics of the person who should be chosen in each
randomly chosen household. However, some adjustments were needed when results
for the total population are presented in order to enable a representative sample for
the community as a whole. The calculated adjustments are shown in Table 1.
Women were 3 per cent under-represented in the sample, a value considered
Journal of Ethnic and Migration Studies 179
Table 1 Sampled and real population distributions by religion
Religion % in sample % in population Adjustment factor
Moslems 65 80 1.23
Christians 15 11 0.80
Druze 20 9 0.55
Total 100 100
negligible. One-third of the sample is of the younger age of 20–34 years, 39 per cent
are of age 35–49 and 28 per cent within the range 50–64. In terms of religiosity 59
per cent of the sample presented themselves as secular, 24 per cent presented
themselves as ‘traditional’ and only 17 per cent as religious. In comparing our results
to Smooha and Ghanem’s (2001) findings for Moslems, in their sample 34 per cent
were identified as being religious while in our sample only 22 per cent defined
themselves thus.
Five Arab interviewers were selected, one in each locality, to carry out the
interviews. The principal researchers trained them in the field, while performing ten
pre-test interviews each. The respondents were interviewed, face-to-face, in Arabic
by Arab interviewers, using a partly structured and partly open-ended questionnaire.
In the structured part we asked the respondents to specify their most salient identity
from a list of six possibilities and to evaluate on a scale from 0 to 6 the relative
salience of each of them for their overall repertoire of identities. These answers were
cross-correlated and clustered in order to unravel the structure of the repertoire of
identities. In the open-ended questions we asked the respondents to describe what
makes them proud and/or not proud about each identity in their repertoire. The
descriptions served as a basis for a narrative analysis in which we exposed the
meanings of each identity and the relations among the component of the repertoire
of identities, as the respondents perceive them relevant to their life.
Arab Identity Repertoire
Before we present the results of the study, we offer a description of the development
of Arab identity repertoires since the 1950s. Identities in the Arab world are formed
within a complex socio-political context (Smooha 1992). They include religious,
national, Pan-Arab, ecological-cultural and kinship identities. Some of these identities
incorporate the community as a whole, while others subdivide them into
separate sub-communities. The identity repertoire of the Arabs in Israel is even more
complex because they are caught in the midst of a national and religious conflict.
However, scholars tend to over-simplify the discussion on this Arab identity by
mainly focusing on the tension between their Israeli civil identity and their
Palestinian and/or Arab national identity. Various studies (e.g. Mayer 1988; Rekhess
1998; Smooha and Ghanem 2001) have paid attention to the increasing role of
political Islam in shaping Arab identity repertoire.
180 M. Amara & I. Schnell
The various studies indicate changes in the order and significance of identities
during different periods among the Arab minority in Israel (Amara and Kabaha
1996). This is briefly summarised as follows. During the first period, 1948–67, most
studies establish that there was a delicate balance in the identity of the Arabs in
Israel; this period is called by Amara (1999a) the ‘quest for security and accommodation’.
This found expression in the development of systems for adapting and a
desire to become part of the life of the country (e.g. Bishara 1993; Cohen 1989;
Ozacky-Lazar 1990; Peres and Yuval-Davis 1969; Yisraeli 1981). The Palestinian
element in their individual and collective identity, it was reported, was extremely
weak due to the defeat during the war of 1948, the lack of political and cultural
leadership, and the disconnection of contact with the remnants of the Palestinian
people. The Palestinian leadership abroad did not promote a national Palestinian
identity; rather they served either the goals of their host nations or a Pan-Arabic
worldview. While the Israeli Arabs were deeply engaged in adapting to their new
status as a minority, the process of coming to terms with the contradicting factors
in their identity commenced, hoping the condition was only temporary and fleeting.
They put emphasis on their Israeli identity more than any other identity. The
reawakening of the Palestinian element in their identity began only after the Six-Day
War in 1967.
The second period is 1967–73. As indicated by most studies, the Six-Day War—
with the conquering of the territories and the military defeat of the Arab countries—
amplified Israeli Arabs’ fear and hatred of Israel and strengthened their identification
with the Arab world. Renewed contact with the Palestinians of the Occupied
Territories in the West Bank and Gaza Strip put an end to their isolation and created
direct contact with a population and leadership having a high sense of nationalistic
consciousness. These contacts awakened the Palestinian component in the Arab
identity repertoire in Israel, which connected, anew, with the consciousness of
Palestinian suffering. During this period strong emphasis was placed on PanArabism,
and the first seeds of the development of the Islamic movement were
planted. Identification with the Palestinians’ suffering in the Occupied Territory
increased the salience of the Palestinian identity (Eisenstadt 1990; Gabbai 1984; Peres
1976; Rekhess 1989; Shtendahl 1988).
The third period, from 1973 to the 1990s, was characterised by the strengthening
of the Palestinian component, the reawakening of Islam and the weakening of the
Israeli element in the repertoire of identities (Schnell 1996). The decision of the PLO
(Palestine Liberation Organization) to strive for a Palestinian state instead of a
Pan-Arabic solution since the 1980s and the Palestinian Intifada toward the end of
the decade further awakened the rise of the Palestinian component in their repertoire
of identities. The Oslo agreement with the Palestinians and Rabin’s (Prime Minster
1992–95) positive policies towards the Arabs in Israel raised their expectations to
come to terms with their Israeli identity—expectations that were broken by Barak
(Prime Minister 1999–2000), who neglected them as a politically legitimate power.
Most studies tended to formulate a dichotomous model in which civil and
national identities are usually described as excluding each other, pushing Arabs in
Journal of Ethnic and Migration Studies 181
Israel either to Israelisation or towards nationalisation as Palestinians or Arabs
(Hoffman 1977; Peres and Yuval-Davis 1969). Almost all scholars believe that the
Israeli civil identity remains relevant in one way or another to Arabs in Israel, while
emotionally they increasingly identify with alternative aspects of their identity
repertoire (Bishara 1996; Smooha 1992). Rouhana’s (1993, 1997) more theoreticallyladen
model distinguishes three levels of identity: (1) an instrumental level in the
polity defined by Israeli identity; (2) an intermediate value level which shares both
Israeli and new Palestinian identity; and (3) a deep sentimental and loyal feeling of
common fate, which is monopolised by the Palestinian identity. In his search for a
‘super-organic’ collective identity, Rouhana regards this complexity in terms of an
incomplete identity. As we have mentioned earlier, we argue that such complexities
are structured and restructured in the experiences of a wide range of social groups
in our globalising world. The key questions are, rather, to what extent do individuals
and social groups comprehend all their relevant identities when they wish to do so,
and to what extent do they succeed in perceiving them as sources of richness instead
of emotional burdens?
A model that comes closer to our conceptualisation is offered by Smooha (1988)
who suggests studying civil and national identities within a framework that treats
them independently of each other. We suggest moving one step further by replacing
the dichotomous model of civil and national identities by a multi-dimensional
model, including the wider range of possible identities—each of them only loosely
dependent on the others. Individuals may constitute their identities by categorising
reference groups, feeling different degrees of identification with some of these
reference groups, and evaluating them as worthy or unworthy of trust and appreciation
(Schnell 1996). Each reference group may be relevant to individuals in four
complementary forms: their meanings to the individual’s identity; structural relations
among the different reference groups; the relative salience of each reference
group; and the flexibility of each identity to changes (Rouhana 1997).
The discussion thus far leads to three major possibilities. First, the different
components of the Arabs’ repertoire of identities in Israel are competing with each
other, presenting negative correlations among their relative salience. Second, only
civil and national identities negate each other while other identities remain indifferent
to both of them. Third, the major components of Arabs’ repertoire of identities
in Israel are independent of each other, presenting low correlations among themselves.
In addition we propose that different Arab groups within the Arab society in
Israel may constitute for themselves different repertoires of identities rather than one
cohesive collective identity.
Results
The first step in understanding the Arab identity repertoire is to investigate the
relative salience they assign to each of their relevant identities. Generally speaking,
about half of the respondents (47 per cent) chose Arab identity as the one they prefer
to present when forced to choose one of their identities in their repertoire. In
182 M. Amara & I. Schnell
comparison only 29 per cent chose their religious identities, whether as Moslems,
Christians or Druze. Palestinian and Israeli identities remained far behind with less
than 10 per cent each. When the whole spectrum of the repertoire is considered, they
assign the highest values to their Arab and religious identities. On a scale of salience
that ranges between 0ϭirrelevant and 6ϭvery important, they assign average values
of 4.9 for their Arab identity and 4.6 for their religious identity. The Palestinian and
the Israeli identities remain behind with 3.1 for Palestinian and 2.9 for Israeli Arab,
while the relevance of local identities of the Hamula (1.3) and the home community
(1.7) remain only marginal.
Differences among respondents are quite significant. Standard deviations in the
respondents’ evaluations are generally between 1.3 and 1.8, and in the case of the
Palestinian identity the standard deviation is even higher (2.2), representing deeper
disagreements concerning its salience.
In an attempt to uncover the characteristics that explain variations in evaluating
the relative salience of each identity, we subdivided the results by the following
socio-demographic characteristics of the respondents: age, gender, marital status,
educational attainment, occupational status, ecological-cultural environment, religion
and religious attitudes. From all eight independent variables the last three
present statistically significant explanations of the differences in evaluations, with
religion being significant at the highest level.3
The distribution of the salience assigned to each identity by the respondents’
religion is presented in Table 2. Several conclusions may be drawn. There are
significant differences among the attitudes of the three communities toward major
identities in the repertoire.4
It seems that Moslems and Christians almost unanimously
tend to emphasise the high salience of their Arab identity. Religious identities
are also presented as highly salient, especially among Druze who tend to play down
their sense of belonging to the Arab identity. All three groups, but mainly the Druze,
tend to assign local identities (Hamula and local community) a low salience. The two
other identities—the Palestinian and the Israeli—are more controversial among
Moslems and Christians. About half of them assign their Israeli identity a moderate
level of salience with the rest being distributed between those who consider it
irrelevant and those tending to highly relevant. The salience of the Palestinian
identity is highly emphasised by about 40 per cent of the Moslems and the
Christians, but is irrelevant or only marginally relevant to 17 per cent of the
Moslems and 36 per cent of the Christians. This means that the Palestinian identity,
like the Israeli one, is somewhat controversial in the Israeli Arab minds, with the
Christians particularly divided in their sense of belonging to the Palestinian identity.
Druze have developed extremely different attitudes towards their identity repertoire.
Most of them assign high priority to their religious identity as well as to their
citizenship in Israel. Almost all of them perceive the Palestinian identity to be
irrelevant to their repertoire of identities and about 40 per cent of them feel the same
in respect to Arab identity. It seems that both Moslems and Christians assign the
highest priority to their Arab identity, which has the potential to incorporate all
three communities within one united collective identity.
Journal of Ethnic and Migration Studies 183
Table 2 The relative salience of the identity repertoire by religion
Relative salience (%)
Not
Identity repertoire Sector relevant Low Medium High Total
Moslems 11 6 43 40 100
Palestinian (IP) Christians 28 8 24 40 100
Druze 87 0 2 11 100
Moslems 0 1 11 88 100
Arab (IA) Christians 0 0 11 89 100
Druze 40 0 20 40 100
Moslems 14 27 53 6 100
Israeli (II) Christians 27 4 58 11 100
Druze 17 0 29 54 100
Moslems 5 7 26 62 100
Religious (IR) Christians 1 8 35 56 100
Druze 14 0 14 72 100
Moslems 29 53 17 1 100
Hamula (IH) Christians 71 18 11 0 100
Druze 90 0 3 7 100
Moslems 21 38 40 1 100
Home community (IC) Christians 29 24 43 4 100
Druze 95 0 3 2 100
Two groups, the Christians and the Druze, seem to be internally divided concerning
their repertoire of identities. Christians face a debate concerning their belonging
to the Palestinian identity. While 40 per cent of them perceive their Palestinian
identity to be highly important, almost 30 per cent perceive the Palestinian identity
to be irrelevant. It seems to us that both attitudes may be understood in the context
of the rise of political Islam (Smooha and Ghanem 2001). Some tend to view the
Palestinian identity as the one that can unite Moslems, Christians and Druze,
moderating the tensions in terms of their religious identities. Others tend to
emphasise their differences from the Moslem identity. Such attitudes were strong
among Christians in Nazareth, probably in response to the struggle over the mosque
in the centre of the city. Druze are united in their rejection of the Palestinian identity
but they are divided concerning the relevance of their Arab identity. On the one
hand a group of intellectuals tend to present themselves as Arabs, blaming the Israeli
establishment for inventing a separatist Druze identity. On the other hand 40 per
cent of the Druze stick to their leaders’ choice to join Israeli citizenship as an
ethnic-religious minority.
Probably the most prominent finding that comes out of Table 2 is that most Arabs
in Israel feel attached to more than one identity in their repertoire, with 77 per cent
assigning values of 4 and higher salience to at least three identities. This finding leads
us to raise questions concerning the relations among the components of their
identity repertoire in terms of its structure, meaning and flexibility.
184 M. Amara & I. Schnell
Table 3 Correlation coefficients among the components of the identity repertoire
Identity
repertoire Palestinian Arab Israeli Religious Hamula Local
Correlations Palestinian 1.00 0.35 Ϫ0.52 Ϫ0.23 0.09 0.25
(Pearson) Arab 0.35 1.00 Ϫ0.25 Ϫ0.20 Ϫ0.04 0.22
Israeli Ϫ0.52 Ϫ0.25 1.00 Ϫ0.00 Ϫ0.36 Ϫ0.40
Religious Ϫ0.23 Ϫ0.20 0.00 1.00 0.08 Ϫ0.02
Hamula 0.09 Ϫ0.04 Ϫ0.36 0.08 1.00 0.44
Local 0.25 0.22 Ϫ0.40 Ϫ0.02 0.44 1.00
Significance Palestinian 0.000 0.000 0.000 0.025 0.000
(1-tailed) Arab 0.000 0.000 0.000 0.184 0.000
Israeli 0.000 0.000 0.491 0.000 0.000
Religious 0.000 0.000 0.491 0.034 0.370
Hamula 0.025 0.184 0.000 0.034 0.000
Local 0.000 0.000 0.000 0.370 0.000
The Structure of the Identity Repertoire
Structurally we have defined three possible models. First, the different components
of the identity repertoire are reinforcing each other, defining a comprehensive and
coherent overall identity. Second, the different components of the identity repertoire
are indifferent to each other. In this case low correlation coefficients may be expected
among the salience of the different identities. Third, groups of identities contradict
each other, presenting strong negative correlation coefficients among themselves.
Table 3 summarises the correlation coefficients among the components of the
Arab identity repertoire in Israel. The analysis shows that almost all correlations
among the components of the identity repertoire are relatively low, with most
coefficients lower than 0.4. Only two correlations show higher coefficients—those
between the Palestinian and the Israeli identities (R2
ϭ0.27, meaning that variability
of one component explains 27 per cent of the variability of the second component),
and between the local and the Hamula identities, which partly reinforce each other
(R2
ϭ0.19). At somewhat lower levels the Palestinian and the Arab identities tend to
reinforce each other while local and Hamula identities tend to contradict the Israeli
identity. Part of the correlations present positive values and part of them negative
ones. It seems that Israeli identity exhibits negative values with the rest of the
repertoire of identities, but these coefficients are relatively low. One exception is the
religious identity, which shows a coefficient of zero due to the fact that on the one
hand many of the Druze may perceive these two identities as reinforcing each other,
while part of the Moslems and Christians may view them as negating each other.
The main conclusion from the correlation analysis is that the different components
of the identity repertoire are to a large extent indifferent to each other, with
a low tendency to perceive the Israeli identity as contradicting the Palestinian one.
This result comes close both to Smooha’s model (1992), which defines the repertoire
Journal of Ethnic and Migration Studies 185
Table 4 The structure of the repertoire of identities
Cluster structure % of Adjusted
No. Cluster name [ (IP), (IA), (II), (IR), (IH), (IC) ] sample sample
1 Autonomous Arabs [(4.5),(5.2),(1.8),(5.1),(1.2),(1.9)] 46 54
2 Arab Israelis [(0.9),(5.3),(4.0),(5.1),(0.9),(1.6)] 31 26
3 Arab Nationalists [(5.0),(5.7),(2.7),(1.9),(1.2),(2.4)] 14 15
4 Druze [(0.2),(0.0),(3.9),(5.7),(1.7),(0.5)] 5 3
5 Israeli Druze [(0.0),(0.9),(5.3),(2.6),(0.0),(0.0)] 4 2
Total (%) 100 100
Total (no.) 498
Notes: The numbers refer to the values of the clusters’ centres, varying from 0ϭirrelevant to
6ϭvery important. The symbols refer to the following identities: (IP)ϭPalestinian, (IA)ϭArab,
(II)ϭIsraeli, (IR)ϭReligious, (IH)ϭHamula, (IC)ϭLocal.
of identities along the national and the civil dimensions, and our suggested multidimensional
model, which assumes relatively low correlation coefficients among a
whole cluster of relevant identities.
Clustering the respondents by their repertoire of identities may further clarify the
relations among the components of the repertoire of identities. We apply a mean
cluster analysis twice. A first one is based on the original data and the second on an
adjusted sample in which we randomly duplicated respondents to a degree that is
required in order to adjust the layers of the samples to their real proportion in the
population. In the analysis we find that five clusters best distinguish among different
repertoires of identities (Table 4). Four of the six components of the repertoire [(IP),
(IA), (II), (IR)] are salient to respondents in at least one of the clusters, while two
remain marginal to all of them [(IH), (IC)]. More than three-quarters of the
respondents have identity repertoires that incorporate at least three salient identities
(in clusters 1 and 2 three components have values of 4 and above), and only 9 per
cent of the respondents emphasised only one identity (clusters 4 and 5). Only for
about half of the respondents has the Israeli identity remained low, negating
alternative national and religious identities (cluster 1), while for the rest the Israeli
identity played a significant role, parallel to their national or religious identities.5
The five clusters distinguish different structures of identity repertoires. About half
of the respondents define themselves as an autonomous minority. They assign
extremely high salience to both their Arab and religious identities and a slightly
weaker salience to their Palestinian identity. Their identification with their Israeli
identity remains weak although relevant to their repertoire (Table 5). They differ also
from the Palestinians in the Palestinian Authority who assign higher salience to their
Palestinian and religious identities and obviously regard the Israeli identity as
irrelevant to them. Instead the autonomous Arab cluster assigns higher salience to
their Arab identity, presenting it as a key factor in their repertoire. Close to one-third
of the respondents define themselves as Israeli Arabs, assigning high salience to their
Arab and religious identities and a slightly lower salience to their Israeli identity,
which replaces the Palestinian one. Fourteen per cent of the respondents emphasise
186 M. Amara & I. Schnell
Table 5 The distribution of socio-demographic factors by clusters of
identity repertoire
Clusters of identity repertoire
Independent variable Cramer’s V Significance
Age 0.123 0.058
Gender 0.124 0.103
Marital status 0.126 0.156
Occupational status 0.135 0.060
Ecological origin 0.379 0.003
Religiosity 0.202 0.007
Religion 0.489 0.001
their Arab and Palestinian identities, assigning only low relevance to the religious
identity and a slightly higher salience to the Israeli identity. This means that they
define themselves mainly along national lines, perceiving themselves as a national
minority within Israel. The last two clusters represent two groups of Druze who
express their unique position among the Israeli Arabs, either by emphasising their
religious identity or by attempting to integrate into the Israeli identity. Based on the
adjusted sample (Table 4) we may conclude that the group of autonomous Arabs
increases to slightly above half of the population, while the percentage of Arab
Israelis declines to about a quarter of the population.
In an attempt to typify the socio-demographic characteristics of the respondents
in each cluster, we calculate the Cramer’s V coefficients between our seven independent
variables and the respondents’ participation in the clusters (Table 5). The
results clearly show the clusters vary according to the respondents’ ecological origin,
religiosity and religion. In accord with the ecological factor, Bedouins and Christians
from Jaffa constitute 78 per cent of the participants in cluster 2. Many Bedouins who
serve in the Israeli military forces tend to assign high salience to their Israeli identity
and lower salience to their Palestinian identity, while strongly identifying with their
Arab and Islamic identities and perceiving them complementary to their Israeli
identity. Part of the Christian community in Jaffa succeed in incorporating their
Arabic and religious identities with their Israeli one, expressing their attempt to
integrate into Israeli society while disengaging themselves from the Palestinian
struggle that becomes in their views increasingly dominated by political Islam.
Concerning the factor of religiosity, non-religious respondents show a relatively
higher tendency to participate in the second cluster and less in the first one. They
assign low salience to their religious identity and a relatively high salience to their
Israeli identity. The most significant factor that influences distribution among the
clusters is religion: 58 per cent of the Moslems participated in the first cluster relative
to 40 per cent of the Christians and 10 per cent of the Druze. In contrast, while
about 45 per cent of the Druze and the Christians participated in the second cluster,
only one-quarter of the Moslems participated in the second cluster. At the same
Journal of Ethnic and Migration Studies 187
time it is almost only the Druze who participated in the fourth and the fifth
clusters.
The Meanings Assigned to Identity Repertoire
One problem that was not addressed in most identity research performed on the
Arab community in Israel was the question of how people understand the meaning
of each identity in the repertoire. We asked the respondents to say in a few sentences
what makes them feel either proud or not proud of each of their identities. Some
quotations of their statements help to highlight the meanings that are commonly
assigned to each of the components of the identity repertoire.
The Arab identity as the most salient one was mentioned in three contexts by
most of the respondents: cultural roots; Arab Middle-Eastern solidarity; and to some
extent the common denominator of the Arab community in Israel. Here are some
quotations to demonstrate these conclusions:
Arabism is a major source of cultural heritage, science, literature, poetry, moral values
and family honour … it gives me a sense of worth and makes me proud to belong to
the Arab community.
I am embarrassed by the lack of political unity and solidarity, by political corruption
as well as low cultural achievements in modern time.
I feel ashamed of traditional habits like revenge for abuse of family honour, women’s
status etc.
Arabism represents the tradition that unites all the Arabs in Israel.
It seems that Arabs in Israel draw their cultural roots and heritage from their Arab
identity and they feel they cannot or do not want to escape it, even when they feel
ashamed of some of their traditions.
Unlike the Arab identity, the Palestinian one emphasises much more the sense of
national pride and solidarity on behalf of cultural heritage, as the following quotations
reveal:
The highly determined uprising gives me pride, I feel part of this battle for independence
and respect, and therefore I participate in the ceremonies of the ‘Land Day’ in
which we demonstrate against expropriation of lands by the government.
I feel proud of the Palestinian history and I hope to live in a Palestinian state.
I feel shame in the way the Palestinians murder innocent people in their battle for
independence.
It is interesting that deliberate nationalist attempts to reconstruct a Palestinian
national history were not articulated by the respondents. They rather preferred to
emphasise the pride they gain from the battle for national freedom and self-determination,
which they so strongly lack as a defeated minority of a Jewish state. It seems
from the discussion that many of the respondents perceived their Palestinian identity
as emerging out of the Palestinian battle against Israel. Therefore the Palestinian
188 M. Amara & I. Schnell
identity may at least partly negate their Israeli identity, a result that is further
supported by the negative correlations between the two respective identities.
In reflecting on the meanings that the respondents have assigned to their Israeli
identity, it appears that the negation of national pride represents only part of their
attitudes, as the next group of quotations illustrates:
Proud of our democratic system and our economic achievements relative to the Arab
world.
Feel good about the sense of political and public order and cleanness that exists in
Israel.
In Israel we care for individual human rights, well-being, health and educational care.
I feel bad about lack of equality in Israel, and cases of deprivation and racism against
Arabs.
I am not proud of the Israeli massacres against Arabs in cases like ‘Kefar Qassem’,
‘Sabra and Shatilla’ and ‘Hebron’.
It seems that the Israeli identity is relevant to the Arabs in Israel at least in three
complementary forms—as a democratic system that assigns high priority to individual
rights and well-being, as a wealthy country that secures people a high standard
of living relative to Arab countries, but also as a system that deprives Israeli Arabs
from their rights relative to Jews. Unlike the Arab and the Palestinian national
identities, the Israeli one is frequently defined in negative terms. Furthermore, the
Israeli identity is frequently defined in comparison to the former two identities,
while they were generally defined independent of any other identity.
Religious identities are perceived differently by the different religions. Moslems
emphasised at least four complementary meanings to their Islamic identity, as the
following quotations demonstrate:
Islam is the religion of morals and justice, which distinguishes between right and fault.
Religious Moslems tend to distort the Islamic morals.
I am proud of the new clever commandments introduced into the Islamic law
nowadays.
Belief in Allah supports me with a sense of comfort and security.
Past victories and the heroic resistance of the ‘Hizbu-allah’, made in the name of Islam,
make me proud.
I feel bad that the rich Islamic countries do not help the poor in the Islamic world.
Islam is perceived as a source of morality, commandments and spiritual experience
that so many of the religious and non-religious Moslems feel they lack in the modern
world. A minority of them perceive their sense of belonging to an Islamic identity
in political terms (cf. Amara 1996). Like many nationalist movements, they reconstruct
their ‘golden age’ and call for Islamic solidarity in solving political and
economic problems.
Christians put more emphasis on religious aspects but they also relate to their
Journal of Ethnic and Migration Studies 189
identity as Christians in terms of being an ethnic minority within the Arab
community in Israel. The following quotations demonstrate these interpretations:
Our religion emphasises spirituality, openness and forgiveness. It adopts a liberal
stance and therefore it is the transcendental religion.
Living in the Holy Land—the Land of Historic and Holy places—gives us a sense of
belonging to the country.
Christians are members of the more educated community among the Israeli Arabs.
Conclusions
The study reported in this paper differs from traditional investigations of Arab
identities in Israel in two important ways. First, it adopts a theoretical framework
that considers identity as a multi-dimensional phenomenon, which evidences complex
relations among the different components of individuals’ identity. Traditional
models assumed a dual model in which they focused on the relations between
national and civic identities. Second, unlike former studies, we do not assume that
the meanings of self-declared identities are self-evident. Therefore we apply an
interpretative methodology in order to unravel the meanings that the respondents
assign to each identity that they mentioned to be relevant for them. The results
support the multi-dimensional model proposed by us.
Empirically we present several pieces of evidence for our argument. Most Arabs in
Israel feel strongly attached to at least three identities. Most of them assigned high
salience to several identities in a way that none of them dominates the others. Each
identity relates to a different content of meanings, playing a different role in the
identity repertoire. Accordingly, the correlations among the identities remained
relatively low, emphasising by this their independence of each other. This means that
the relative salience of each identity is more dependent on the type of practices
performed in each respective milieu, than on the attractiveness of alternative
identities. In this respect it is interesting that the Palestinian identity is not associated
with citizenship beyond national pride, while the Israeli identity is expected
to supply Israeli Arabs mainly with a sense of citizenship. The role of the
Israeli identity is even more remarkable when comparing this attitude with
the Palestinians’ one. They did not mention citizenship and individuals’ rights at
all in their identity repertoire, emphasising instead their collective identity as
Palestinians.
One exception is the relatively high negative correlation between the Palestinian
and the Israeli identity. This may be explained by many Israeli Arab feelings of
humiliation due to the fact that their citizenship is not tolerated in Israeli society. In
this sense the Palestinian identity supplies them at least with some sense of pride.
This finding may explain why the dual model of Israeli vs. Palestinian identities may
gain some empirical support, and this is the situation that led Rouhana (1997) to
define Israeli Arabs’ identity to be incomplete. However, our finding that different
groups constitute for themselves different relations among the components of the
190 M. Amara & I. Schnell
identity repertoire hints at the conclusion that our multi-dimensional model offers
a better explanation of the structure of Israeli Arabs’ repertoires of identities. Many
of the Bedouins and some others (represented in cluster 2) have developed an
identity repertoire in which both the Palestinian and the Israeli identity are relatively
highly salient—hence the positive correlation between them. This option fits the
argument of Smooha (1992) who argues that ‘Israelisation’ and ‘Palestinisation’ may
be working in parallel. At the same time Arab nationalists (cluster 3) tend to
emphasise their Palestinian identity but they do not do it wholly at the expense of
their attachment to their Israeli identity. Only the largest group, representing close
to half of the Israeli Arabs, represent negative correlations between the two identities,
as postulated by Rouhana.
The socio-political and historical contexts that lead to the constitution of this
complex repertoire of identities are beyond the scope of this paper. These contexts
have rarely been studied systematically. However, there is evidence that the practices
of everyday life tend to reinforce certain identities in different sets of activity
patterns. Schnell (1996) argues that the daily commute between Arab and Jewish
spaces of more than half of the Arab workforce exposes many Arabs to events that
bring their Palestinian and Arab identities into the centre of their identity field. Their
failure to integrate into the Israeli legitimate collective undermines their sense of
Israeli identity. And Amara (1999b) exemplifies how the daily use of language in
various domains of life (whether on signs or institutions) is closely related to the
identity repertoire.
What we learn from this study is that the use of quantitative methods may bring
some fresh insight into the study of identity. We were able to show that even
members of highly marginalised groups may develop complex repertoires of identities
instead of developing one unified identity. Furthermore, different subgroups
may develop different strategies of identity formation. We also learn, unlike the
existing research on identity in the Arab society in Israel, that each identity is only
marginally influenced by other identities in the repertoire. Consequently, our
argument is different from the current arguments of ‘Israelisation versus Palestinisation’
approaches, which claim dependency between the two.
The discussion cannot be concluded without two more remarks. First, since
identities are continuously socially negotiated, two phenomena must be taken into
consideration in evaluating future trends. First, the Israeli Arab violent demonstrations
in October 2000, which ended in the killing of 13 Arab demonstrators by Israeli
police forces, shocked the Arab community in Israel. Following these events, Arabs
in Israel started questioning their identification with their civil identity as Israelis.
Second, the political discourse is dominated today by the Islamic movement and
the National Democratic Party (a secular-national party, mainly led by the
Knesset member Azmi Bishara). They aspire either to transform Israel into a
democratic non-Zionist state or to recognise the rights of the Israeli Arabs as a
national minority. Both trends threaten to challenge the basic structure of the Israeli
state and the repertoire of identities crystallised among the Arabs over the last few
decades.
Journal of Ethnic and Migration Studies 191
Acknowledgement
The first draft of this paper was presented at the Study Day ‘Jewish–Arab Relations:
October Events’ at The Centre for Peace Studies, Giv’at Haviva, 1 March 2001. This
study is funded by the Israel Science Foundation.
Notes
[1] The term repertoire is borrowed from music, and it is widely employed in sociolinguistics. It
includes all the languages and varieties known and used by a specific community. In a similar
vein, we apply the concept to the study of identity in the sense of a set of multitudes of
identities with complex interrelations among them.
[2] We chose to interview equal numbers of respondents in each town since we intend to
compare communities in our wider study. The equal sample secures representation to small
towns and enables comparison. Since we do not know about internal variability we do not
have any reason to adopt different sample size, assuming that internal variability has a much
larger effect on sample size than population size in a town. In order to overcome distortion
in the total sample, we made the necessary adjustments for each community: Moslems, Druze,
Christians and Bedouins.
[3] Chi-square test equals 0.0001.
[4] Chi-square test equals 0.0001.
[5] The correlations between education and type of locality on the one hand and the type of
identity on the other are low and insignificant, therefore we did not include them in the
analysis.
References
Amara, M. (1996) ‘The nature of Islamic fundamentalism in Israel’, Terrorism and Political
Violence, 8(2): 155–70.
Amara, M. (1999a) ‘Identity dilemma of Palestinians in Israel’, in Hoffman, A., Schnell, I.,
Stephan, G., Bruer, G. and Fenn, P. (eds) Identity and Education in Germany and Israel.
Hamburg: Fachverlag fur wissenschaftliche literatur, 213–34.
Amara, M. (1999b) Politics and Sociolinguistic Reflexes: Palestinian Border Villages. Amsterdam and
Philadelphia: John Benjamins.
Amara, M. and Kabaha, S. (1996) Divided Identity: A Study of Political Division and Social
Reflections in a Split Village. Giv’at Haviva: Peace Research Institute (in Hebrew).
Anderson, J. (1997) ‘Nationalism and geography’, in Anderson, J. (ed.) The Rise of the Modern
State. Brighton: Wheatsheaf, 7–21.
Bar-On, D. (1999) The Others Within Us. Beer-Sheva: Ben-Gurion University Press.
Bhabha, H.K. (1994) The Location of Culture. London and New York: Routledge.
Bishara, A. (1993) ‘On the question of the Arab minority in Israel’, tioria ubikorit, 3: 7–20 (in
Hebrew).
Bishara, A. (1996) ‘The Israeli Arab: scrutinizing a divided political dialogue’, in Ginosar, P. and
Bareli, A. (eds) Zionism: A Modern Polemic. Research and Ideological Approaches. Beer-Sheva:
The Ben-Gurion Legacy Institute, Sde Boker; The Haim Weizmann Zionist Research
Institute, Tel Aviv University; Ben-Gurion University of the Negev Publishing, 312–39 (in
Hebrew).
Calhoun, C. (1994) ‘Social theory and the politics of identity’, in Calhoun, C. (ed.) Social Theory
and the Politics of Identity. Oxford: Blackwell, 9–36.
Castells, M. (1997) The Information Age—Economy, Society and Culture: Volume II—The Power of
Identity. Oxford: Blackwell.
192 M. Amara & I. Schnell
Cohen, R. (1989) In the Thicket of Loyalty: Society and Politics in the Arab Sector. Tel Aviv: Am
Oved (in Hebrew).
Eisenstadt, S. (1990) Israeli Society and its Values. Jerusalem: Magnes Press (in Hebrew).
Gabbai, M. (1984) The Arabs of Israel: A Question of Identity. Giv’at Haviva: Arab Studies Institute
(in Hebrew).
Giddens, A. (1991) Modernity and Self-identity: Self and Society in the Late Modern Age. Stanford,
CA: Stanford University Press.
Hoffman, Y. (1977) Likenesses and Identity of Arab Youth in Israel. Haifa: Haifa University, Arab
Education and Development Research Institute (in Hebrew).
Holland, W., Lachicotte, J., Skinner, D. and Cain, C. (1998) Identity and Agency in Cultural
Worlds. Cambridge, MA: Harvard University Press.
Lewin, K. (1951) Field Theory in Social Science. New York: Harper.
Mayer, T. (1988) The Awakening of Muslims in Israel. Giv’at Haviva: The Institute for Arabic
Studies (in Hebrew).
Noy, P. (1995) ‘What is the self in the psychology of the self?’ Sichot, 9(3): 180–90 (in Hebrew).
Ozacky-Lazar, S. (1990) The Stands of the Arabs of Israel vis-a`-vis the State, 1949–67. Haifa: Haifa
University, unpublished MA thesis, (in Hebrew).
Peres, Y. (1976) Relations of Ethnic Communities in Israel. Ramat-Aviv: Tel Aviv University (in
Hebrew).
Peres, Y. and Yuval-Davis, N. (1969) ‘Some observations on the national identity of the Israeli
Arabs’, Human Relations, 22(3): 219–33.
Rekhess, R. (1989) ‘The Arabs of Israel and the Arabs of the West Bank and the Gaza Strip:
political ties and national identity’, hamizrah hehadash, 32: 165–91 (in Hebrew).
Rekhess, R. (1998) ‘Political Islam in Israel and its relation to the Islamic movement in the
territories’, in Rekhess, E. (ed.) The Arabs in the Israeli Politics: Dilemmas of Identity.
Beer-Sheva: Tel-Aviv University, The Moshe Dayyan Centre for the Study of the Middle
East and Africa, 73–84.
Rouhana, R. (1993) ‘Accentuated identities in protracted conflicts: the collective identity of the
Palestinian citizens in Israel’, Asian and African Studies, 27: 97–127.
Rouhana, R. (1997) Palestinian Citizens in an Ethnic Jewish State: Identities in Conflict. New Haven
and London: Yale University Press.
Sarup, M. (1996) Identity, Culture and the Post-modern World. Athens, Georgia: The University of
Georgia Press.
Schnell, I. (1996) Perceptions of Israeli Arabs: Territoriality and Identity. London: Avebury.
Schnell, I. (2002) ‘Segregation in everyday life spaces: a conceptual model’, in Schnell, I. and
Ostendorf, W. (eds) Studies in Segregation and Desegregation. Aldershot: Avebury, 39–66.
Shtendahl, U. (1988) ‘The Arab population in Judaea and Samaria’, skira hodshit, 8(28): 3–19 (in
Hebrew).
Smooha, S. (1988) ‘The Arab minority in Israel: radicalization or politicization?’, Studies in
Contemporary Jewry, 5: 1–21.
Smooha, S. (1992) Arabs and Jews in Israel: Change and Continuity in Mutual Intolerance. Boulder,
CO: Boulder University Press.
Smooha, S. and Ghanem, A. (2001) ‘Political Islam among the Arabs in Israel’, in Hanf, T. (ed.)
Dealing with Difference: Religion, Ethnicity and Politics: Comparing Cases and Concepts.
Baden-Baden: Nomos Verlagsgesellschaft, 143–73.
Tajfel, H. (1970) ‘Experiments in inter-group discrimination’, Scientific American, 223(5): 96–102.
Tajfel, H. (1978) ‘Social categorization, social identity and social comparison’, in Tajfel, H. (ed.)
Differentiation Between Social Groups: Studies in the Psychology of Inter-Group Relations. New
York and London: Academic Press, 61–76.
Taylor, D.M. and Moghaddam, F.M. (1987) Theories of Intergroup Relations. New York, Westport
and London: Praeger.
Weinreich, P. (1986) ‘The operationalisation of identity theory in racial and ethnic relations’, in
Journal of Ethnic and Migration Studies 193
Rex, J. and Mason, D. (eds) Theories of Race and Ethnic Relations. Cambridge: Cambridge
University Press, 134–58.
Yisraeli, R. (1981) ‘On the problem of identity of the Arabs in Israel’, in Har-Even, A. (ed.) One
Out of Six Israelis. Jerusalem: Van Leer Institute, Mossad Publishing, 179–84 (in Hebrew).