Lettres de Byblos Letters from Byblos No. 14 T HEODOR H ANF E pluribus unum? Lebanese opinions and attitudes on coexistence UNESCO Centre International des Sciences de l'Homme/ International Centre for Human Sciences and Friedrich-Ebert-Stiftung Byblos 2007

Lettres de Byblos/ Letters from Byblos A series of occasional papers published by UNESCO Centre International des Sciences de l’Homme International Centre for Human Sciences The opinions expressed in this monograph are those of the author and should not be construed as representing those of the International Centre for Human Sciences. All rights reserved. Printed in Lebanon. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage or retrieval system, without permission in writing from the publisher. C International Centre for Human Sciences, 2007 Published in 2007 in Lebanon by the International Centre for Human Sciences, B.P. 225 Byblos(Jbeil), Liban. ISBN 978-9953-0-1004-5

Contents Fear of the future, cautiousness, powerlessness: Psycho-social sensitivities 7 The crisis of entrepreneurship: Attitudes to the economy and society 10 A nation of believers 15 Kinship: Attitudes to family and society 17 Strata, communities and identity 20 Snapshots of a political system and political orientations 26 Democrats by conviction and even more so by necessity 37 Perspectives of coexistence 44

Acknowledgements In 1982, the first of hitherto six surveys on attitudes and opinions of economically active Lebanese had gone to the field. The author’s partner at that time was Samir Farah, Director of the Friedrich Ebert Foundation’s Beirut office. A quarter of a century later he was again my partner – and this time my sponsor, too. He deserves twofold thanks. Salim Nasr and the author jointly developed the questionnaire in 1982, and Salim translated it into Arabic. In 2006, he assisted in crafting some additional items addressing the current situation in the country. Salim’s sensitivity for ways of expressing opinions has been essential for creating a survey instrument that has stood the test of time. Boutros Labaki and Kamal Feghali hotly debated and finally drew the sample. They organised the fieldwork – in 2006 under slightly easier circumstances than in 1982. They and their field staff carefully stuck to the same methodological approach as in the previous surveys and thus greatly contributed to assuring the data’s comparability. Rym Momtaz operated the data entry, Petra Bauerle the statistical analysis. John Richardson translated the author’s text into English, and Angela Herrmann aptly took care of its outfit. I am grateful to all of them. All mistakes and shortcomings are my exclusive privilege. Theodor Hanf

E pluribus unum? Lebanese opinions and attitudes on coexistence T HEODOR H ANF Representative surveys during the war and civil war – in 1982, 1984, 1986 and 1987 – showed the clear desire of all Lebanese, regardless of religious community, to live together peacefully in the same country. In the midst of war the most hopeful consequence of precisely this phenomenon was the emergence of a nation. 1 Twelve years after the end of the war, Lebanon was firmly under Syrian control. Despite, or because of, this, a survey in 2002 confirmed the earlier findings. The Lebanese accepted themselves as they are: different from, but dependent on one another. For them coexistence is not only possible, but enriching – a sceptical nation, but a nation. 2 Has this changed since the withdrawal of the Syrian army? An empirical survey of the economically active population conducted in early 2006, a few months before the 1 Theodor HANF, Koexistenz im Krieg. Staatszerfall und Entstehen einer Nation im Libanon, Baden-Baden: Nomos 1990; expanded English edition: Coexistence in War-Time Lebanon. Decline of a State and Rise of a Nation, London: Tauris 1993. 2 Theodor HANF, The Sceptical Nation. Opinions and Attitudes Twelve Years after the End of the War, in: idem& Nawaf SALAM, Lebanon in Limbo. Postwar Society and State in an Uncertain Regional Environment, Baden-Baden: Nomos 2003, pp. 197–228.

Israeli war against Hezbollah, sought to explore this question. The survey was carried out using the proven questionnaire of the earlier investigations and a random sample (N=2016) 3 broken down into regional quotas. Owing to deficiencies in the country’s social and economic statistics, the regional quotas 4 – and consequently the proportions by membership of the different religious communities 5 – cannot be accepted as mirroring reality beyond all doubt. For the purposes of this study, it is not necessary to discuss the different demographic hypotheses, 6 which in Lebanon are the subject of exquisite – politically biased – exegesis. For the purposes of our investigation it is adequate to compare with sufficient statistical accuracy the attitudes and opinions of respondents from various strata, regions and occupations on the one hand and various religious communities on the other. 3 The two postwar surveys differ from earlier ones in the percentage of respondents that refused to answer certain questions. In 2002, this related in particular to questions of political orientation, and in 2006 questions about income. Fear of the tax authorities has replaced fear of political expression. 4 The largest divergences between the surveys during and after the war were recorded in the region comprising Ashrafieh, the East Beirut suburbs, Metn, Kisrawan and Jbeil: in the earlier surveys the economically active population constituted on average 28% of the total, whereas in the later two only 17.1% and 18.7%, respectively, which is problematically low in view of this region’s economic importance. 5 As a consequence of the low quota for the most densely populated Christian region, the proportion of Christians fell to 33%(2002) and 40%(2006) of the total sample compared to 48% in earlier surveys. In 2006, 28.2% of the respondents were Shiite, 25.6% Sunni, 24.3% Maronite, 9.3% Greek-Orthodox, 5.7% Druze, 4.6% Greek-Catholic and 2.1% Armenian and members of other Christian communities. About 6% gave no details of religious affiliation; in the following we have considered only respondents that provided details. 6 Cf., for instance, differences between the data of Kamal FEGHALI, Les communautés du Liban. Approche démographique, Zalka: Makhtariat, July 2002, and Boutros LABAKI, Données démographique communautaires du Liban, 1996, idem, From Regionalization to Globalization. The Lebanese Emigration since the End of the“Wars for the Others”(1990– 2000)(the author thanks the writers for their permission to use the manuscripts). Our sample was constructed on the basis of Feghali’s calculations. 6

E pluribus unum? Fear of the future, cautiousness, powerlessness: Psycho-social sensitivities To all appearances, Lebanon today, as yesterday, is a country full of the joys of life, hospitable and convivial. However, in their attitudes to life and their social environment, the image that individual Lebanese create is one of an almost oppressively palpable threat to this way of life. “When I think of the future, I feel uncertain and afraid.” In the four wartime surveys, about 60% of the respondents answered in the affirmative, in 2002 81% and in 2006 no less than 84%. Fear of the future rises with age 7 and falls with the level of education. 8 It is above average across the entire southern part of the country, in the Shuf and in some districts of Beirut. 9 All in all, some four fifths of the Lebanese feel afraid when they think about the future, and this fear has risen in recent years. Fear usually goes hand in hand with cautiousness and conservatism. “Before you start something, you should know whether it will work or not.” “When you start changing things, they usually get worse.” During the war, between 71% and 74% of respondents were cautious in their dealings with strangers. Twelve years after the end of the war, this figure had risen to 81%; it 2006 it reached 87%. During the war 33% of respondents expressed strong scepticism about change and a fear that things would get worse; in 2002, the corresponding figure was 43%, and in 2006 a full 53%. 10 "You should always be careful. You cannot trust the people you live or work with." In 2002, 78% agreed; in 2006 no less than 84%. Mistrust of one’s social environment is particularly common among farmers and owners of small businesses, in the Christian district of Ashrafieh, in the Shiite Beirut suburbs of Jnah and Uzai, among 7 78%(youngest) – 84% – 87% – 92%(oldest). 8 93%(no formal education) – 88% – 86% – 83% – 77%(university degree). 9 It is above average among the inhabitants of Zokak el Blad, the eastern suburbs, Jnah and Uzai as well as in Nabatieh, Tyre and Zaharani. 10 Scepticism about change rises with age from 48% among the youngest to 59% among the oldest. The figure is over 60% among farmers, civil servants, unemployed persons, small businessmen and white-collar workers. It is lowest among the Sunnites(45%) and above average among the Shiites(60%). 7

almost all respondents in Sidon, Nabatieh, Tyre, Zaharani and Jezzine and throughout the south. There are no statistically significant differences between communities. 11 Whom do people trust? 1987 survey Close relatives 67 Friends 59 Members of my religious community° Superiors 10 Neighbours° Colleagues 20 People from my village* Figures in all tables in%, rounded.* Not asked.° Not asked 2002 survey 76 52 20 24 16 24 17 2006 survey 95 71 41 37 32 31 27 The family is the repository of trust, even more than in previous years. Solidarity among friends appears to have increased. Respondents now trust superiors, neighbours, colleagues and people from the same village more than they did before. The increase in trust in members of the same religious community is nothing less than spectacular. There is no statistical difference between Christians and Muslims in this respect, but there is between individual religious communities: more than half the Greek Catholics and the Druze, but less than one third of the Maronites trust their coreligionists. In short: the intact world is very small. Intact is above all the family; seven in ten respondents include their circle of friends, and no fewer than four in ten their coreligionists. “People like me can’t do anything to improve people’s lives.” Reactions to this statement are a rough measure of how people assess their ability to bring about social change. Between 1982 and 1987 the feeling of powerlessness and helplessness rose from 66% to 80%. In 2002, it was still as high as 78%. In 2006, it had declined again to 70%. That means that somewhat less than one third of the Lebanese believe that they can make a difference. Such people are overrepresented among the young 12 and the better educated. 13 “Even poor people can get ahead if they help one another.” 11 Mistrust is slightly below average among the Sunni respondents(78%). 12 38%(youngest) – 28% – 26% – 24%(oldest). 13 20%(no formal education) – 25% – 28% – 28% – 36%(university degree). 8

E pluribus unum? As in earlier years, nine in ten respondents believe that help for self-help can be successful. Agreement is slightly above average among the Druze and Shiites. In the Shuf and the Beirut districts of Burj Abi Haidar, Tarik al Jadida and Burj al Barajneh as well as in Jbeil every respondent agreed with the statement. In summary: the psychological sensitivities of the Lebanese in 2006 give even fewer grounds for optimism than before. Four in five respondents are fearful of the future, extremely cautious and mistrustful. Half of the respondents fear that change will only worsen the situation. People’s trust is restricted more than ever to their close circle of family and friends, and – with the notable exception of one’s coreligionists, in whom trust is twice as widespread as in previous years – few others. Almost three quarters are still convinced that they can do little to change their lot. 9

The crisis of entrepreneurship: Attitudes to the economy and society The following question measures preferences for investment and consumption: “Imagine you won some money in the national lottery. What is the first thing you would like to do with it?” Buy a house or property Spend on myself and my family Start my own business Put the money in an good bank investment 1981 37 5 42 16 1984 33 3 46 17 1986 26 3 57 14 1987 19 4 51 26 2002 18 11 45 25 2006 14 17 39 30 During the war and in the postwar period, consumption has always been the least popular and the desire for economic independence the favoured option. In 2006, however, the entrepreneurial spirit was noticeably weaker and the preference for bank investments greater in 2006. Moreover, the tendency to spend money also shows a not insubstantial rise. Would-be businessmen are overrepresented among pensioners and housewives, which is indicative of a wish rather than an intention. A disproportionately high number of those who want their own house have no formal education or job. An aboveaverage number of civil servants, small businessmen and farmers would like to spend lottery winnings. The choice for a bank investment correlates almost linearly with the level of education. 14 The country’s economic situation does not encourage entrepreneurships among the Lebanese. Notwithstanding this, fundamental attitudes have changed little, as the responses to the following statement show: “I don’t want to work for someone else all my life; some day I want to be my own boss.” Three quarters of the respondents agree, almost as many as in the war years, though fewer than in 2002. There are no significant differences between Muslims and Christians. 14 34%(no formal education) – 38% – 43% – 44% – 39%(university degree). 10

E pluribus unum? “In your opinion, which of the following factors counts most for success in life?” Religious beliefs Education/Training Achievement Luck Experience Team work Astuteness Connections Inheritance Other * Not asked 1987 ° 16 21 14 21 ° 7 16 5 * 2002 22 19 18 7 7 7 6 4 4 6 2006 23 22 13 8 6 9 6 3 5 5 In 2006, as in 2002,“religious beliefs” was rated highest. As this option was not available in the wartime surveys between 1981 and 1987, it is unfortunately impossible to determine whether this represents a change in mentality and attitudes or not. The answers given during the war show the great majority of people to be performanceoriented. In 2002 and 2006, non-achievement factors such as luck and connections are rate lower than before, but so is experience. More than one fifth of all respondents rate religious beliefs as the most important factor for success. In 2002, the figure for Muslims was more than one in four, but even among Christians 17% shared this view. In 2006, there was no longer any significant difference between Muslims and Christians. A possible interpretation is that this is a reaction to a fairly desperate economic situation, in which neither luck nor connections offer much hope of success, only divine support. This view is supported by the replies to other questions. “One man says: ‘I'd rather work hard, build up my own business and take risks to get ahead and make a lot of money.’ Another says: ‘I'd rather have a safe job with a regular income so that I don't have to think about the future.’ Do you agree with the first or the second man?” Between 1981 and 1987, agreement with the first option rose from two thirds to almost three quarters. In 2002, it got the support of just under half the respondents, and in 2006 of only 38%. The Lebanese have become much less risk tolerant. By education, 11

only respondents at the highest level are notably less risk averse than average. 15 It is much lower in economically backward regions of the south and north than in Beirut and the mountains close to the capital. Christians still exhibit a greater willingness to take risks than Muslims; among the former risk tolerance fell from 56% to 45% and among the latter from 41% to 32%. 16 What about job satisfaction? “ Of course, everyone would like to earn more, but I’m satisfied with my salary.” “On the whole, I’m satisfied with my boss.” “It’s a pity, but I don’t have any prospects in my job.” “If I could, I’d like to have a different occupation.” “If I could, I’d like to work in the Gulf.” 1987 24 47 47 70 52 2002 37 61 62 76 54 2006 52 75 75 72 55 In 2006, more people are satisfied with their salary 17 and their boss than in all earlier surveys. This is only because there is no other alternative, as shown by the fact that people's sense of no career prospects in their present jobs has risen considerably 18 and more people would like to change their occupation. 19 Although earnings prospects in the Gulf are no longer what they were in the 1980s, more than half would still grab the opportunity if offered it. Remarkably, communal or religious affiliation does not significantly affect job satisfaction – with the exception of the desire to work in the Gulf, which is considerably weaker among Christians(48%) than Muslims(59%). 20 The depth of the economic crisis is reflected in opinions about living standards and future prospects from a transgenerational point of view. 15 Own business: 35%(no formal education) – 34% – 35% – 34% – 44%(university degree). 16 In 2006, it was highest among the Maronites at 49%. 17 People satisfied with their salary are overrepresented among the Greek-Orthodox(61%) and Maronites(60%). 18 The perception of having no opportunities declines with rising level of education. There are no significant differences by community affiliation. 19 The desire to change one’s occupation correlates inversely with education and income, but even in the respective highest status groups is still shared by two thirds of respondents. 20 It is strongest among the Druze(73%) and the Sunnites(62%). 12

E pluribus unum? “Compared to how my parents lived, I think I’m much better off.” “I fear my children won’t have it as good as me.” 1987 2002 2006 71 57 59 66 85 87 Substantially more than half the Lebanese still think their generation is better off than their parents'. In the mid-1980s, however, this view was even more widespread. Neither age nor education nor religious affiliation plays a significant role. During the war, fears that the next generation would be worse off were already widespread; and have grown since then. They are shared by the vast majority of respondents across all levels of education and income, and all regions and communities. How do respondents feel about social distinctions in Lebanese society and the distribution of prosperity and economic power? “The differences between the social groups have widened since 1975(2002 and 2006: in the last five years).” “In Lebanon, a rich minority is buying up everything, and the large majority is losing out.” “In Lebanon the majority of people are middle-class, with only a few rich people and not very many poor.” “It doesn’t matter what the workers do, they can never win against the bosses.” “When I see people from the wealthy parts of town, I think to myself that’s how I want to live, and that I have a right to live that way.” Agreement in%, rounded 1981 90 87 68 25 66 1984 91 92 54 37 67 1986 94 94 46 40 63 1987 96 95 17 52 72 2002 87 89 43 54 72 2006 88 88 64 60 71 13

In the view of the Lebanese, neither the long war nor the postwar decade and a half was the great leveller. 21 Then as now the overwhelming majority of people agree with the topos of the"rich minority". The competing view of a middle-class society, which was shattered in 1987, had regained some support by 2002, and in 2006 was the (ideal) view of almost two thirds of the respondents. 22 More than half of all respondents believe that the power of employers is unassailable. In view of the social crisis, it is not surprising that social envy of“people from the better parts of town” has grown – across all education and income groups and all religious communities. To summarise: attitudes and opinions about the economy and society in 2006 were either little changed from those in the war years – or were more pessimistic. The majority of Lebanese would still like to be independent businessmen, but the proportion of people willing to accept the risks associated with entrepreneurial activity has declined. Although most people are convinced that achievement, education and experience are the main factors determining success in life, a substantial minority think religious beliefs are more important. Job and career satisfaction have dropped. People are even more pessimistic about their children's prospects than they were in the 1980s. Social differences have widened and social envy is almost as high. 21 In 2006, 92% of the Christians and 85% of the Muslims are of the opinion that the gap between rich and poor has widened. Occupation and education are not statistically significant factors. Older people – i.e. with greater experience – are more likely to take this view than young people. 22 Whereas 69% of the Christians believe that Lebanon is a middle-class society, only 60% of the Muslims do. Support for this view is above average among respondents with an intermediate education and below average among university graduates. There are substantial differences between the different areas of Beirut and the regions. Not only in Ashrafieh do more than four in five residents believe in a middle-class society; the same is true of respondents in Burj Abi Haidar, Burj al Barajneh, Jnah and Uzai. This perception finds similar support in the Shuf and Aley, Sidon and Jezzine, but is rejected by large majorities in Jbeil, Zahle and the Bekaa. 14

A nation of believers E pluribus unum? Secularising moderation of religious convictions and less observance of religious practice is not part of the Lebanese agenda, not 20 years ago, and today even less so. “I believe in a life after death, in which the righteous will be rewarded and the wicked will be punished.” “I try to live by the teachings of my religion.” “I often visit a place of worship.” “I can be happy and enjoy life even if I don't believe in God.” 1987 2002 2006 71 85 94 75 80 90 38 60 63 11 11 11 The number of Lebanese who state that they are believers, try to live according to the teachings of their religion and visit mosques or churches has continued to rise. The only constant is the proportion of respondents – a good tenth – who state that they have nothing to do with religion. In 2006, as many Muslims as Christians believed in a life after death and divine justice for all. This view is slightly more common among middle-aged people than among the youngest and oldest. 23 The astonishing finding of the 2002 survey that the younger the respondents, the more likely they were to be believers 24 was not confirmed in 2006. The under-24-year-olds and the 25 – 34-year-olds of then are now mostly in the middle age groups, in which, as mentioned, believers are overrepresented. Today’s youth are also generally believers. This is borne out by the responses to a somewhat nebulous statement about believe in an afterlife. “I believe in some form of existence after death.” Seventy-one percent of the respondents agree. Overrepresented are the better educated strata, but also those without any formal schooling, students and housewives, Shiites – and respondents under the age of 24. 25 In short, the Lebanese clergy of all religions have little reason to doubt the religiousness of the youngest generation. 23 92%(youngest) – 97% – 94% – 92%(oldest). 24 From oldest to youngest: 82% – 82% – 82% –88% – 89% – 92%. Cf. Theodor HANF, The Sceptical Nation, op. cit., p. 205. 25 Agreement: 79%(youngest) – 70% – 67% – 67%(oldest). 15

The proportion of respondents who stated that they tried to live according to the teachings of their religion increased from 80% in 2002 to 90% in 2006. The youngest age group is underrepresented, but even among them the figure is four in five. 26 At 92%, Christians agree slightly more frequently than Muslims and women more frequently than men. Agreement declines at the highest educational levels. 27 There are no significant distinctions by religious affiliation. 28 Muslims and Christians visit cult places equally frequently. This practice rises noticeably with age; 29 housewives, pensioners and farmers are particularly zealous visitors. More than half of the respondents pray regularly and another fifth frequently. As to be expected, Muslims are overrepresented among people who pray regularly, whereas Christians are more likely to pray often or sometimes. The good tenth of respondents who are cheerful non-believers do not differ significantly by religious affiliation, age or community membership. One fifth of all respondents without any formal education are part of this group. To summarise: the Lebanese, already pious during the long years of war, have become even more so. But there is little difference by socio-economic markers. There is no significant difference in the degree of religiosity of the different communities and the two major religious groups: the Lebanese are a nation of believers, the young as much as, if not a little more so than, their elders. 26 81%(youngest) – 93% – 92% – 94%(oldest). 27 90%(no formal education) – 92% – 93% – 88% – 86%(university degree). 28 By residence, agreement is highest in Burj Abi Haidar, Jnah, Uzai and Jbeil(100%); it is also very high in Nabatieh(99%), Sidon(97%) and Tripoli(93%). 29 59%(youngest) – 60% – 65% – 74%(oldest). 16

Kinship: Attitudes to family and society E pluribus unum? Although deep family loyalties may compete with loyalty to one's religious community, it is also the social foundation of the latter. According to Fuad Khury, familialism is a more traditional form of kinship loyalty, and loyalty to the community a more modern form. 30 The responses to the following statements served to measure the strength of familialism: “Stand by your brother, be he oppressor or oppressed.” 31 “If my family disagrees with the strongest political tendency in my community, I'll side with my family.” “It is permissible to break the law if it is in the interest of your family.” * Asked for the first time in 2006 1987 45 2002 47 2006 66 50 48 51 ** 43 In the war years, agreement with the saying:"My brother, right or wrong" rose from one quarter in 1981 to almost one half in 1984, where it still stood 15 years later. In 2006, it had risen to two thirds. Now as then, in conflicts between family and community interests respondents are more or less equally divided. No less than 43% of the respondents believe that one may break the law in the family's interest. 32 30 Fuad I. Khury, From Village to Suburb. Order and Change in Greater Beirut, Chicago 1975, pp. 201ff. 31 Lebanese proverb. 32 This opinion is overrepresented in Burj Barajneh, Sidon, the Jezzine and the Akkar, among farmers, small businessmen and construction and transport workers. It declines with rising education, only to rise again among university graduates: 56%(no formal education) – 51% – 45% – 34% – 40%(university graduates). There are no differences between Christians and Muslims. 17

The replies to the following statements indicate the strength of community identity: 1987 "It doesn't matter whether they are rich or poor, I feel close to all members of my community." 58 "It is good if there are no conflicts between sons of the same community." 91 "One man says:'It is not so important if my daughter marries a man with a different religion, as long as she loves him." A second man says:'Marriages between people of different religions are not good and are often unhappy. I don't want my daughter to marry someone with a different religion.' Do you agree with the first or second statement? Agree with the first statement: 40 "I prefer neighbours who have the same background as myself." 55 "If my colleagues are honest and cooperative, I don't mind what group they belong to." 90 "The identity and uniqueness of my community are more important to me than loyalty to my country." 12 Agreement in% 2002 66 78 39 48 92 34 2006 81 91 36 59 96 39 Solidarity with one’s religious community has continued to strengthen in postwar Lebanon: in 2002, two thirds expressed it; in 2006, four fifths. It is somewhat stronger among women than men. There are no differences by education or by community. 33 Concern about conflict within the community initially declined in the postwar period, since conflicts were no longer settled at gunpoint. In 2006, however, such concern is as high as in 1987 and more pronounced among Christians(93%) than Muslims (89%); 94% of Maronites and Greek-Orthodox are concerned about intracommunal conflict. The willingness to let a daughter marry across a religious divide has fallen slightly. It is slightly above average 34 among younger people 35 and the better educated. 36 33 Overall, 79% of Muslims agree, and an even higher 84% of Christians. 34 In the surveys in the 1980s, women were far more likely to approve of mixed marriages than men. In 2002 and 2006, by contrast, gender no longer plays a role. 35 41%(youngest) – 38% – 32% – 30%(oldest). 18

E pluribus unum? There are considerable differences by residential area. There is strong approval of mixed marriages in Burj Barajneh(100%) and Basta(83%), in Jbeil(73%), Ashrafieh (66%) and Chiyah(58%). Approval is well below average in the eastern suburbs of Beirut(25%), in the Akkar(25%), in Jnah and Uzai(17%) and in Nabatieh/ Zaharani (16%). Acceptance is higher among Christians(41%) than Muslims(32%), and highest among the Druze(48%) and Maronites(45%). Compared with the war years, the proportion of those that would prefer neighbours with the same background fell slightly in 2002, but had risen enormously by 2006, and is now higher than in the war years. Whether Christian or Muslim, almost three fifths of the respondents would like to live among people with a similar background. By contrast, in 2006, religion and community affiliation were even less relevant at the work place than in earlier years. The hard core of community loyalists feel that their community is more important than the nation-state. In 2002, their proportion was almost three times that in 1987 37 ; in 2006, it has again risen markedly. There are no significant differences between Christians and Muslims, 38 or by gender, age or education. 39 To summarise: both familialism and community loyalties have strengthened considerably not only by comparison with the war years, but also with 2002. In 2006, two in three Lebanese rank the family above all else, Four in five feel that they have very close ties with their community, three in five prefer people like themselves as neighbours, and two in five feel greater loyalty to the community than the state. But almost no one has problems working with members of other communities. This bears even stronger witness to J.S Furnival's classic observation than even in earlier years:“… two or more elements of social order which live side by side … in one political unit... there is one place in which the various sections of a plural society meet on common ground – the market place”. 40 36 29%(no formal education) – 38% – 31% – 38% – 43%(university degree). 37 Asked for the first time in 1987. 38 Agreement is highest among the Sunnites at 45%. 39 More than half of the small businessmen, skilled manual workers and farmers agreed, but less than one quarter of self–employed persons, white-collar workers and industrial workers. By district, agreement is well above average in Jbeil(71%), Jnah and Uzai(67%), Basta (63%) and Ashrafieh(61%). Agreement is well below average among residents of Tarik al Jadida, the Shuf(none), Burj Barajneh(14%) and Kisrawan(18%). 40 J.S. Furnival, Netherlands India. A Study of Plural Economy, Cambridge: University Press 1939, p. 449. 19

Strata, communities and identity As in all plural societies, three question complexes dominate academic and political debates. The first concerns the connection between economic stratification and group affiliation, the second citizens' perception of this connection(which is not always the same thing), and the third specific group and transcendent identities. The first of these questions, which deals with the"incorporation" of strata and communities, has been coloured for decades by the topos of"rich Christians" and "poor Muslims", exacerbated by the thesis of"communal class". It is not our intention to add to this body of literature, 41 which is largely concerned with interpretations of aggregate data. Sample surveys provide only approximations; they are no substitute for complete socio-economic census data. Our surveys in the 1980s established that, with minor variations, all communities were stratified by income, i.e., each had its wealthy, average and poor groups. Whereas Shiites alone were underrepresented in the highest income group, in the lowest income group – quite correctly termed poor – there were more or less equal percentages of Shiites, Greek Catholics, Armenians and Maronites. In the 2002 study, the connection between communal and religious affiliation on the one hand and income stratification on the other is statistically insignificant. In other words, although the distribution of income as a whole may have become more unequal, there are no longer rich communities and poor communities. In the 2006 survey, interviewees were less hesitant than previously about expressing opinions on religion and politics, but were far more reluctant than in any earlier surveys to provide details about income 42 . While respondents were utterly indifferent to any form of political correctness, they seemed to be very much afraid of the tax authorities. In short: the income data in the 2006 survey provide no new insights into economic status in and between the individual communities. Although the data can serve to illuminate correlations between educational stratification and religious affiliation, the results are not very significant: Sunnites are slightly overrepresented at the lowest educational level and slightly underrepresented among university graduates; the latter also holds for Shiites. Maronites, GreekOrthodox and Greek-Catholics are overrepresented at the two highest levels of education. 43 However, the Christian’s one-time educational advantage 44 has shrunk 41 For an excellent critical survey, see Farid el Khazen, The Breakdown of the State in Lebanon 1967–1967, Cambridge: Harvard University Press 2000, pp. 29ff. 42 Just under half of the respondents provided details about their monthly wage or salary and virtually none about other income or their annual income. 43 Sunnites: 8%(no formal education) – 11% – 16% – 26% – 39%(university); 20

E pluribus unum? considerably. But there is still a difference between Muslims and Christians by occupation: Sunnites are overrepresented in the civil service and in the health and services sectors, Shiites in agriculture, industry, the trades and schooling, the Druze in the professions and the business world, Maronites in the professions, small business, white-collar workers and services, the Greek-Orthodox and Greek Catholics among white-collar workers and the Armenians among the trades. But there are no occupational monopolies: an“ethnically" cleaved labour market such as found in many plural societies does not exist in Lebanon. How do Lebanese classify themselves socially? "Every society has social classes. Which class do you think you belong to?" highest upper middle middle lower middle lowest. The answers to this question correlate with the data on education as follows: Education/ Class perception No formal education Primary school Brevet, occupational Baccalauréat University Overall Estimate Highest 6 4 2 3 4 3 Upper middle 5 8 7 10 10 9 Middle 40 50 54 67 62 60 Lower middle 33 29 31 18 22 24 Lowest 17 9 6 2 2 4 Educational level total 5 9 18 28 40 100 The urge to be middle-class is very apparent. As shown above, two in three respondents believe that Lebanon is still a middle-class society, and as many classify themselves as upper-middle or middle class, and a further quarter as lower-middle class. Not even one in twenty respondents view themselves as lower class. There is a significant correlation between educational status and social self-classification. Shiites: 5%((no formal education) – 11% – 19% – 28% – 37%(university); Maronites: 3%(no formal education) – 7% – 22% – 29% – 39%(university); Greek-Orthodox: 4%(no formal education) – 5% – 13% – 30% – 47%(university); Greek-Catholic:4%(no formal education) – 11% – 19% – 20% – 47%(university); Average: 5%(no formal education) – 10% – 18% – 27% – 40%(university). 44 Cf. Theodor Hanf, Erziehungswesen in Gesellschaft und Politik des Libanon, Düsseldorf: Bertelsmann Universitätsverlag 1969. 21

Respondents with no or only a low level of formal schooling classify themselves disproportionately often as lower or lower-middle class, and secondary-school and university graduates as middle or upper-middle class. The self-assessment as middle class is indubitably a reflection of the generally high standard of education in Lebanon. But education is not the only status factor, as shown by the fact that an above-average number of respondents with a no or little education categorize themselves as upper class. How do the members of different religious communities classify themselves by social status? Upper class Sunnites 3 Shiites 4 Druze Maronites 5 Greek Orthodox 3 Greek Catholics 1 Armenians Average Categorization 3 Upper middle class 9 6 6 8 10 6 25 8 Middle class 58 49 54 68 56 61 46 58 Lower middle class 26 36 36 17 22 30 25 27 Lower class 4 5 4 2 9 3 4 4 Community as a whole 26 28 6 24 9 4 1 Sunnites and Greek Orthodox self-categorization deviates least from the average. Shiites and Druze are more inclined than other groups to classify themselves as lowermiddle class, while Maronites and Greek Catholics have fewest qualms about placing themselves in the middle class. One remarkable result is that only one in twenty Shiites regard themselves as lower class; in 2002 the proportion was twice as high. How do the Lebanese perceive themselves over and above social classification? How do they define their identity? "Each of us belongs to several groups simultaneously: each person is either male or female, belongs to a religion or a community, is rural or urban and has a political viewpoint. And different people attach different importance to each factor. If you were to be asked who you are, how would you describe yourself?" first second, and third?" 22

E pluribus unum? The answers ordered by frequency of the first option are: 1st place 2nd place Lebanese Personal Normal person Political adjective Occupation Arab Muslim Christian Region, village Other 50 28 11 16 7 3 7 5 5 5 5 7 5 9 3 10 2 6 5 11 3rd place 21 25 2 9 7 7 7 7 5 10 Compared to the 2002 survey, in 2006 the choice of“Lebanese” is higher in first, second and third place. Most respondents regard themselves as Lebanese, although priorities differ. 45 The self-description using personal markers –“normal person”, occupation and other – has also risen across all three places compared with 2002. By contrast, the options of“Arab”,“Muslim” 46 and“Christian” 47 in first place are all slightly less popular;“Christian” is more common in second and third place than previously. A novelty in 2006 is the use of political adjectives, a clear reflection of greater polarisation among at least a minority of the respondents. 48 However, the full complexity of the identity patterns only becomes clear when the first, second and third choices are correlated. 49 45 Among those who put“Lebanese” in first place, Greek Catholics(65%), Maronites(60%) and Greek Orthodox(59%) are overrepresented and Sunnites(41%) and Armenians(43%) underrepresented. This choice for first place correlates directly with education: 39%(no formal education) – 45% – 51% – 50%(university degree). By region, it is below average in the Akkar(30%), Sidon( 35%) and Tripoli(38%). 46 Sunnites(14%) are above average and Shiites(6%) just above. 47 Armenians(10%) and Maronites(8%) are above average. 48 To an equal extent among Muslims and Christians. 49 In the interests of clarity, the following table includes only identifications with political and religious connotations; others(occupation, personal markers, regions, etc.) have been omitted. 23

1st place Lebanese Arab Muslim Christian Political adjective Figures in%, rounded 2nd place 23% Lebanese 16% Christian 14% Muslim 9% Arab 73% Lebanese 9% Muslim 5% Arab 36% Lebanese 30% Arab 6% Muslim 47% Lebanese 28% Christian 9% Political 28% Lebanese 20% Political 3rd place 19% Lebanese 9% Political 9% Arab 9% Christian 7% Muslim 40% Muslim 19% Lebanese 34% Lebanese 19% Arab 10% Political 27% Lebanese 15% Political 15% Christians 31% Lebanese 11% Political 9% Muslim Among those that identify themselves in the first place as Lebanese, just under a quarter also do so in second place and just under a fifth in third place. For one quarter of them, being“Christian”, for one fifth being“Muslim” and for just under one fifth being “Arab” is part of their complex understanding of Lebanese identity. Nine tenths of those that identity themselves primarily as“Arab” define themselves as Lebanese in second or third place – and almost half of them also as Muslim. For seven in ten of those that choose“Muslim” first,“Lebanese” is also part of their identity, and“Arab” for almost half. Of those who view themselves as“Christian” first, three quarters put“Lebanese” in the second or third place. Two in five choose“Christian” a second time, and almost one quarter define themselves as“political”. Finally, among those that define themselves primarily by some political affiliation, one third of them repeat this, but three in five also add“Lebanese”. One immediate conclusion stands out: former national ideological cleavages, such as“Lebanese”,“Arab” and“Syrian” nationalism, no longer play a role in the selfperception of the Lebanese. Almost all see themselves as Lebanese, but with varying fervour and considerable nuances. Half of the respondents regard themselves as Lebanese who are simultaneously Muslims or Christians. One tenth view themselves as Muslims or Christians who are also Lebanese. One twentieth are Arabs who are also Lebanese(and for the most part Muslim). 24

E pluribus unum? E pluribus unum ? The findings on identity can be interpreted in this way. Or have the nuances between the different forms of“Lebaneseness” become so strong that the pluribus must now accommodate a plurum rather than an unum ? The first point to make is that the Lebanese society is strongly familialistic, as discussed above. Religiosity has increased markedly and communal bonds have strengthened. On the other hand, regardless of widening economic disparities, we could not identify any income-based economic distinctions between the communities. Communal differences in educational levels are minor. Allowing for nuances, there is an overwhelming tendency across all communities for people to categorise themselves socially as middle-class. The response to the question of an identity that transcends community, in so far as any political identity was articulated, was also, again allowing for nuances, unequivocal. How do respondents view the relationship between economic stratification, social cleavages by community and the role of politics? This question was examined using the following statements: "Of course there are political and religious differences in this country. But the differences between rich and poor are more important." "Those who exploit differences between the communities in politics do this to hide the real differences between rich and poor." 1987 59 52 2002 66 87 2006 62 70 Almost two in three respondents think that the most important cleavages are economic, 50 and seven in ten are convinced that the traditional politics of politicised communities serves economic goals. 51 For a substantial majority, the main problem facing Lebanon is not its identity, but its economy. However, it should not be overlooked that agreement with the thesis that cleavages between the communities are politically manipulated is considerably lower than in the 2002 survey, in particular among younger respondents. 52 Accordingly, it is necessary to look more closely at the specifically political attitudes and opinions of the Lebanese. 50 Agreement with this statement was well above average in Burj al Barajneh, Tarik al Jadida and Jbeil. 51 More Christians(78%) than Muslims(64%). 52 Agreement: 59%(youngest) – 69% – 76% – 82%(oldest). 25

Snapshots of a political system and political orientations "In your opinion, which of the following groups has the greatest influence in Lebanon?" zu’ama’ Religious leaders Party leaders Large landowners Ministers Military officers Bankers Merchants Industrialists Figures in%, rounded before 1975 53 57 11 4 7 9 4 2 3 2 1981 6 7 54 4 1 5 5 13 4 1984 17 10 38 7 6 5 16 1 1986 18 12 41 2 3 6 17 1987 38 3 30 2 1 3 8 14 1 2002 27 7 10 2 20 12 3 3 1 2006 24 23 13 12 11 9 5 2 1 Pre-1975 Lebanon was dominated by the zu'ama, the traditional notables who formed a sort of executive committee in parliament and government that continually negotiated new compromises between family, regional, communal and personal interests in frequently changing coalitions. After the outbreak of war, these notables were displaced by"party leaders", in effect the militia leaders. When it became obvious to all that the latter had plunged the country into chaos, the zu'ama regained some of their influence. The same holds for the large landowners, who are close to the zu'ama. In 2006, just under one quarter of the respondents thought that the zu’ama were the most influential group, and a good tenth the large landowners. This is a significant comeback. That said, another group has enjoyed a spectacular rise in the opinion of the Lebanese: three times as many respondents in the 2006 survey as in the 2002 survey attributed substantial influence to religious leaders. On the other hand, the perception of ministers as important has almost halved and that of the military diminished by a quarter. Economic leaders, whether bankers, merchants or industrialists, have been more or less marginalised. This lack of trust is understandable against the backdrop of the economic crisis. 53 Estimate in the 1981 survey. 26

E pluribus unum? What do respondents think of politicians? "The time of the old politicians is finished. The new leaders represent the gut feeling of the country and function better." "The old politicians are better than the new generation. At least they are tolerant, more moderate and realistic." Agreement in%, rounded 1987 2002 62 32 37 62 2006 55 68 The results show a clear split among the respondents. A good half accept that the new leaders are more efficient, but at the same time two thirds believe that the old politicians are more tolerant, moderate and realistic. How do the Lebanese feel about political involvement? "If you keep out of politics you have peace and quiet and a clear conscience." In 1987, 62% of the respondents agreed with this statement, in 2002 69% and in 2006 not many fewer: 67%. This indicates widespread depoliticisation. But almost a third of the respondents disagree with this statement and clearly articulate their political involvement. Whereas in 2002 only 15% of the respondents said they belonged to a political organisation – overwhelmingly Shiites and Druze – in 2006 no less than 30% did – including an above-average number of Christians. 54 In 2002, only two thirds of party members were prepared to name the organisation, in 2006 almost all did so. In short: in 2006 a strong minority is not at all depoliticised, quite the contrary. Finally, many Christians no longer make any secret about their political convictions. The reawakening of political interest is underscored by the replies to the question of the most and least popular politicians. In 2002, almost a quarter stated that they did not have a favourite politician, 16% gave no answer and a good tenth did not like any of them. 54 36% Christian, 27% Muslim. Party membership is above average among Maronites(42%), Druze(40%) and(34%) and below average among Sunnites(17%). 27

The picture in 2006 is very different: Michel Aoun Hassan Nasrallah Hariri, Siniora Nabih Berri Samir Geagea Sleiman Franjieh Walid Jumblat Other anti-Syrian Other pro-Syrian Gemayel family Emile Lahoud Other None All Figures in%, rounded Favourite politician 17 15 13 8 7 6 5 9 7 1 4 8 Least liked politician 9 7 7 2 9 3 18 8 14 3 4 8 3 4 There is no lack of clear likes and dislikes. Every high-profile Lebanese politician has supporters and opponents, and those with the highest profiles have more of the latter than the former. The state president constitutes an exceptional case in that statistically speaking he has only opponents. Who likes and dislikes whom? Aoun is the favourites of 38% of the Christians and three percent of the Muslims. 55 His support is concentrated in the Shuf and Aley, in Jbeil, the Jezzine and Kisrawan as well as in Beirut in Tarik al Jadida. His supporters are overrepresented among whitecollar workers, housewives, unemployed persons, traders and civil servants – i.e., among more modest people. The preference for Aoun correlates with the level of education. 56 Aoun’s opponents are concentrated among Sunnites(23%) and Druze (18%). Hassan Nasrallah draws the support of 47% of the Shiites, five percent of the Sunnites and two percent of the Christians, who tend to be congregated in Baalbek, the Hermel, Nabatieh and Zaharani, and in Beirut in the districts of Haret Hreik, Burj al Barajneh, Burj Abi Haidar and Zokak al Blad. His supporters are particularly numerous 55 An above-average 49% among Greek Orthodox, 41% among Armenians and 39% among Greek Catholics. 56 7%(no formal education) – 16% – 14% – 19% – 21%(university graduates). 28

E pluribus unum? among pensioners, farmers, traders and industrial and transport workers, and among primary-school and junior secondary-school graduates. Nasrallah’s opponents are overrepresented among Armenians(20%), Greek Orthodox(14%) and Maronites (13%). Nabih Berri has above-average support among the Shiites(26%), but finds favour with just one percent of the Christians. His strongholds are Nabatieh/Zaharani, the northern Bekaa and in Beirut in the southern suburbs of Jnah and Uzai, Chiyah and Burj al Barajneh. His support is overrepresented among farmers and civil servants, people without any formal schooling or with junior secondary school. The assassinated Rafik Hariri, his son, and Prime Minister Fuad Siniora are the most popular politicians among 45% of the Sunnites, three percent of the Shiites and one percent of the Christians. Their strongholds are Sidon(69%), the Akkar(51%) and the Beirut districts of Musseitbe, Basta and Zokak al Blad. They are named particularly often by tradesmen and service workers as well as by students. Dislike of the Hariri group of politicians is mentioned by 14% of the Greek Orthodox. The most controversial politician in 2006 was Walid Jumblat. His admirers are overrepresented among Druze and Sunnites, professionals, small businessmen and students. He is disliked by a good quarter of the Shiites and the Maronites. The supporters of Franjieh and Geagea offer no surprises. Franjieh is the favourite son of the Maronites around Zghorta. Geagea is admired by some 15% of the Maronites, Greek Orthodox and Greek Catholics, and disliked by Muslims. In short: the approval and disapproval of politicians demonstrates the degree to which political Lebanon is a patchwork of confessional monocultures. How strong is respondents' support for their favourite politicians; how do supporters assess politicians' overall support, and do they think they are effective? "Even if the politician that I prefer does something I don't agree with, I will continue to support him." "Most people in my community support this leader." "In the present situation these politicians cannot do much for us." Agreement in%, rounded 2002 21 36 60 2006 44 61 65 29

The willingness to support politicians unconditionally has more than doubled since 2002. 57 The proportion of those who believe that their favoured politician enjoys widespread support has also soared. 58 At the same time, the number of those who think that politicians have little influence in the present situation has also risen. 59 Thus, respondents have few illusions, yet are far more prepared to identify with today’s political elite than they were with the then leaders a few years ago. As shown above, party membership – and the willingness to admit it – has increased dramatically since the 2002 survey. Party membership breaks down as follows: Free Patriotic Movement(Michel Aoun) 25 Hezbollah(Hassan Nasrallah) 18 Lebanese Forces(Samir Geagea) 14 Amal(Nabih Berri) 12 Future Movement(Saad Hariri) 10 Marada(Sleiman Franjieh) 8 Progressive Socialist Party(Walid Jumblat) 6 Syrian National Socialist Party& Baas 3 Communist Party 2 Other 2 Figures in%, rounded Like the support for popular politicians, the overwhelming portion of larger parties’ membership comes from specific communities. Despite some Muslim members, FPM is a predominantly Christian party. 60 The same is true of the FL 61 and the Marada. 62 Hezbollah is exclusively Muslim and pre57 Unconditional allegiance correlates directly with age: 38%(youngest) – 45% – 46% – 48% (oldest), and inversely with education: 55%(no formal education) – 46% – 37% – 36% (university degree). It is above average among Shiites at 56%. 58 At 76% also well above average among Shiites. 59 There are no statistically significant differences by social variables on this statement. 60 45% of the Maronites who belong to a party, 50% of the Greek Orthodox, 52% of the Greek Catholics, 28% of the Druze, 5% of the Sunnites and 1% of the Shiites state that they belong to the FPM. Professionals, white-collar workers and housewives are overrepresented among FPM members. 61 18% of the respondents in the youngest age group stated they were members of the Lebanese Forces. By occupation, office, agricultural and service workers are overrepresented. By community affiliation, 39% of all Greek Catholic, 28% of Maronite, 17% of Greek Orthodox and 1% of Shiite party members are in the FL. 30

E pluribus unum? dominantly Shiite. 63 Amal includes some Druze and Christians, but is also primarily a Shiite party. 64 Future is the main Sunnite organisation 65 and the PSP largely Druze. 66 Only the pro-Syrian and the Communist parties cannot be assigned to a particular community. 67 At the risk of oversimplifying, one could say that Lebanon has a Christian threeparty system consisting of the FPM, FL and Marada, a Shiite two-party system made up of Hezbollah and Amal, and one-party dominance among the Sunnites and Druze: Future and PSP, respectively. One consequence of the geography of the religious communities is that in some regions and Beirut districts two or more parties compete for the same votes, in others one party dominates, and in yet others parties compete fiercely for support across community lines. According to our data on party membership, there is a Christian twoparty situation in the Metn, Kisrawan, Jbeil, the Jezzine and Ashrafieh, where party membership of both FPM and FL is substantial. In Zghorta, both of these and the Marada compete. There are Shiite two-party systems in Nabatieh/Zaharani, where Amal has more members than Hezbollah, and in the Bekaa and Baalbek-Hermel, where Hezbollah activists outnumber those of Amal. In Burj al Barajneh and Haret Hreik Hezbollah predominates, whereas in Jnah and Uzai Amal. One party has the say in Sidon and in Tripoli, two Future strongholds. Parties representing different religious communities compete fiercely in the Shuf/Aley, in the Akkar and above all in the formerly Sunnite, today partially Shiite districts in West Beirut: in Ras Beirut, Musseitbe and Masraa, in Zokak al Blad and in the Basta Future activists have to contend with numerous Hezbollah members. 62 20% of Maronite and 5% of Orthodox party members have joined the Marada, but no Muslim respondents. 63 Membership includes 57% of all Shiites registered with a political party and 7% of Sunnites, but no Christians, are members. Hezbollah’s membership is young: 24% of all respondents under the age of 24. An above-average proportion work in the agricultural and health sectors, and many are still at school or students. 64 34% of all Shiites in a political party are members of Amal, 11% of the politically organised Sunnites, 6% of the Druze and 4% of the Greek Catholics. Middle-aged persons, farmers, civil servants, and industrial and transport workers are overrepresented among Amal’s members. 65 The organisation includes 59% of the Sunnites, 3% of the Shiites, 2% of the Maronites and 3% of the Greek Orthodox who are members of a party. Tradesmen and pensioners are overrepresented in Future’s membership. 66 The PSP has a disproportionately large number of young people, students, farmers and teachers among its members – and 67% of the politically organised Druze and 3% of the Sunnites. 67 Greek Orthodox are overrepresented among PSNS members and Muslims among those of the CP. 31

In summary: in 2006, the Lebanese expressed clear preferences for a limited number of political leaders, and these preferences are based primarily on communal membership. Compared with 2002, the Lebanese are far more politicised, as reflected in a doubling of party membership and the much greater willingness to openly admit it. As previously, if not more so, party membership is drawn from particular communities: the Lebanese respondents feel not only close social and religious, but also political ties with their respective communities. What are their ideals and desires over and above the well-being of their country and community? To ascertain this, respondents were asked the following question: "Which country do you think is the most perfect country?" The list ranked according to the 2006 replies is as follows: 1984 68 1986 1987 France Smaller European countries Arab countries Lebanon Immigration countries USA Iran Other EU countries No country Syria Israel 7 8 9 30 33 47 5 4 15 8 4 1 2 12 4 6 6 7 3 2 10 8 10 10 19 3 --(0.2) 1 1 1 Figures in%, rounded 2002 16 20 2 6 10 3 3 11 10 4 2006 22 18 11 9 9 7 6 5 4 3 In 2006, France’s standing was even clearer than in 2002 – is the secular république une et indivisible the secret ideal of the inhabitants of a country whose own system is the exact opposite? Or does it reflect the sympathy that France shows for Lebanon and gratitude for French support for Lebanon’s independence? What speaks for the latter is France’s particular support among members of communities that took the lead in the new Lebanese independence movement of 2005: 32% of the Druze, 28% of the Greek Orthodox and 26% of the Sunnites. The small European countries – are Switzerland, Austria and the Scandinavian states – are preferred particularly by Shiites and Armenians. The Arab countries – apart from Syria – mentioned in third place are for the most part the Gulf states and chosen by 13% of the Muslims, but also five percent of the Christians; for many Lebanese 68 Not asked in 1981. 32

E pluribus unum? seeking to earn a living, they have become the destination of choice. Almost one tenth, in particular Shiites and Maronites, think Lebanon is an ideal country. An equal percentage, mostly Druze, Greek Orthodox and Armenians, name one of the classic immigration countries, in which many Lebanese have already settled. The USA is more popular than in 2002; 2006 – until the Israeli attack in summer – was the year in which many Lebanese placed their hopes in the USA. Compared with 2002, in 2006 twice as many respondents, in particular Shiites, 69 see Iran as an ideal country. Admiration for Syria has declined by a quarter in the same period. Israel drew no sympathy in 2002, and none in 2006, either. The choice of a country is influenced by many factors: assessment of the social system, practical considerations about possible destinations for migratory labour, the desire to emigrate, and the appreciation of a country’s policies towards Lebanon. How do respondents assess the influence of the international environment on Lebanon? "Here is a list of organisations and countries that may exercise some influence on Lebanon. Please tell us whether you think the influence of each is positive, negative or ineffective?" The replies in 2006 break down as follows: France Vatican UN EU Saudi Arabia Arab League USA Iran Syria Russia PLO Egypt Iraq Jordan Israel Positive 63 63 57 50 39 35 35 29 24 21 19 18 10 9 5 Negative 20 12 19 31 24 13 55 50 64 28 55 34 33 25 87 Ineffective 17 25 24 19 37 52 10 21 12 51 26 47 57 66 8 69 20% of the Shiites and virtually no Christians; a disproportionately large number of unemployed persons and pensioners, but also students. 33

As in the 2002 survey, France tops the list, this time sharing first place with the Vatican. Since 2002, the UN and the European Union have gained ground. The assessment of the positive and negative influence of Saudi Arabia and the Arab League is more or less unchanged. The influence of all others, from the USA to Israel, is held to be more negative than positive. The most significant changes are in the assessment of Iranian and Syrian influence. In 2002, 43% thought Iran’s influence was positive, but only 29% in 2006, while the negative views rose from 32% to 50%. Positive votes for Syria halved and negative votes doubled in the same period. The assessment of foreign influence on Lebanon, particular that of France, Saudi Arabia, Iran and Syria, differs considerably by respondents’ community affiliation. France’s influence is held to be positive by an above-average number of Druze(82%), Greek Catholics(78%), Maronites(76%) and Sunnites(73%), 70 whereas twice as many Shiites as average(42%) regard it as harmful. 71 The proportion of Sunnites (62%) who find Saudi Arabia’s influence helpful is well above average. Almost three quarters of the Shiites view Iranian influence as a blessing, 72 whereas just as many Maronites, two in three Druze, Greek Orthodox and Greek Catholic respondents and 56% of the Sunnites see it harmful. 73 The assessment of Syria marks a parting of communal ways. Sixty percent of the Shiites regard the neighbour’s influence as positive, 74 but not one of the Druze respondents. Syria’s influence is seen as harmful by 81% of the Maronites, 79% of the Sunnites and 78% of the Druze. 75 70 This is reflected in the preferences by place of residence. Positive evaluations of France are above average in the Christian heartland of Jbeil(96%), Kisrawan(84%) and Metn(83%), but also in the Sunnite strongholds of Sidon(81%) and the Akkar(76%). Three in four respondents in Ashrafieh and the Basta share this view. 71 A negative view of French influence is particularly pronounced in the Baalbek/Hermel region (66%) and in the southern suburbs of Beirut(50% and more). 72 Values are above-average in Nabatieh(77%), Baalbek/Hermel(74%) and Haret Hreik (71%). 73 Negative views are especially common in the Shuf/Aley(87%), Jbeil(85%) Metn(79%) and in Sidon(69%). In Beirut this view is shared by 90% of the inhabitants of Ashrafieh and 66% of those of Musseitbe. 74 Syria finds above-average support in Nabatieh(73%) and Baalbek/Hermel(55%) and by occupation among those active in agriculture(44%) and industry(32%). 75 A negative view of Syria’s influence is particularly pronounced in Kisrawan(89%), Jbeil (88%) and the Metn(81%) as well as in Sidon(84%), Tripoli(77%), Zghorta(75%) and the Akkar. In Musseitbe/Mazraa it is shared by 84% of the respondents, the same figure as in Ashrafieh. 34

E pluribus unum? Above we touched on the degree to which political developments in the country reflected communal quasi-monocultures. Hence, it is not surprising that the respective supporters of the most important politicians exhibit even more unequivocal views on the country’s foreign political orientation. France’s influence in Lebanon is viewed as positive by 89% of Jumblat’s followers, 88% of Hariri’s and Siniora’s, 87% of Geagea’s and 84% of those of other opponents of Syria. But no less than 48% of Nasrallah’s supporters view it as negative. 76 Among the Hariri/Siniora group, 77% – well above average – welcome Saudi Arabia’s influence, whereas 36% of Berri’s followers view it as detrimental. 77 The significance of Iran is held to be positive by 80% of Nasrallah’s and 72% of Berri’s admirers, respectively, whereas about eight in ten of the followers of Jumblat and Geagea and more than seven in ten of Aoun’s view it as negative. Syria’s role is viewed favourably by 63% of Nasrallah’s supporters and 65% Berri’s – high, but less than that for Iran. By contrast, it is regarded as negative by just under nine in ten of Geagea’s and Jumblat’s admirers and almost as many Hariri and Siniora supporters. Respondents’ replies to the following statement allowed a more precise view of attitudes on Syria: “Lebanon and Syria are one people in two states.” 78 In 2002, 35% of the respondents agreed with this statement; in 2006, 33% did, virtually unchanged. 79 Thus, two thirds view Syria as a neighbour – and one third as a brother country. Just under one fifth welcome its influence in Lebanon. However, only three percent of the Lebanese view Syria as in ideal country, one in which they would like to live. To summarise: at the beginning of 2006 the only point of foreign policy on which the Lebanese agreed was the rejection of Israel. Apart from that, they were deeply divided in their views of friend and foe. For Druze, Sunnites and Christians, and in particular for the followers of the respective most popular politicians in these communities, the influence of France(and of the Vatican, the UN and the EU) is positive, a view not shared, however, by the Shiites. Saudi Arabia is viewed as a friend by the Sunnites and in particular by the supporters of Hariri and Siniora. Whereas Druze, 76 Another 31% of Nasrallah’s admirers and 34% of Berri’s regard France’s role as ineffective. 77 The view that Saudi Arabia’s influence is ineffective is particularly pronounced among Jumblat’s supporters(50%) and those of Nasrallah(45%). 78 A statement of former Syrian President Hafez al-Assad. 79 Among the under-24-year olds, 40% agreed. This view is heavily overrepresented among the inhabitants of the Baalbek/Hermel region(64%), the Bekaa(61%) and Nabatieh/ Zaharani; agreement is even more widespread in the Beirut suburbs of Burj Abi Haidar (75%) and Jnah/Uzai(67%). Agreement is twice as high among Muslims(43%) as among Christians(20%). 35

Sunnites and Christians regard Iranian and especially Syrian influence in Lebanon as negative, the Shiites take a positive view of Iran and, to a lesser degree, of Syria. The religious communities part ways in particular over their opinions about Syria. In short: in their opinion on foreign preferences and disapproval the cleavage is not between Muslims and Christians, but between Shiites and others, and more specifically between Nasrallah’s and Berri’s followers and all other Lebanese. 80 80 72% of the supporters of Michel Aoun view the influence of Iran and of Syria as negative. It would be interesting to see whether Aoun’s alliance with Hezbollah influences this opinion in future surveys. 36

E pluribus unum? Democrats by conviction and even more so by necessity “Think of a country in which you would like to live. Which do you think is best for that country?” 2002 2006 One party that unites the country or Several parties that citizens are free to choose between. 57 68 Judges that follow directives from the government or Judges that only follow the laws without listening to the government. 80 72 Press censorship to prevent propaganda that causes strife or A free press that can criticize politicians as it likes. Figures in% 70 68 As in 2002, in 2006 more than two thirds of the respondents hold democratic views, but a substantial minority do not. Agreement with a multiparty system has risen sharply, while support for an independent judiciary has decreased. Who are the supporters of a one-party system? They are overrepresented among women and pensioners and more common among older than younger respondents. 81 They constitute 35% of Muslims and 27% of Christians. An obedient judiciary is favoured most strongly by civil servants and the armed forces(44%), small businessmen(42%), farmers(39%) and unemployed persons (38%), older 82 and less educated people. 83 There are no significant differences by religion or community affiliation. Press censorship finds least sympathy among the young and correlates inversely with education. 84 It is supported by 34% of the Muslims and 29% of the Christians – most strongly by Shiites(42%) and least strongly by Druze(15%). 81 28%(youngest) – 30% – 33% – 43%(oldest). 82 19%(youngest) – 27% – 34% – 36%(oldest). 83 31%(no formal education) – 26% – 29% – 23% – 20%(university degree). 84 41%(no formal education) – 35% – 37% – 32% – 28%(university degree). 37

In short: a little less than one third of the respondents lack democratic convictions. One quarter also had no problem with the following statement: “Rigging election results is acceptable for making the best candidate for the country win”. An above-average 46% of farmers and 40% of small businessmen agree, as do 28% of civil servants. Assuming that they are more or less democratic, what economic system and social policies do the Lebanese regard as desirable in 2006? Lebanon has always been a byword for laissez-faire economics. Yet in 2002, only 51% of the working population favoured private ownership of the means of production. This figure had risen to 58% in 2006. Nonetheless, 42% would like the state to take charge of the economy. The fact the latter figure includes an above-average proportion of the youngest age group will hardly please market economists. Education is a statistically insignificant factor. But there is a clear distinction between Muslims and Christians: 62% of the latter favour a market economy as against 53% of the Muslims. A three quarter majority favours economic and social policies that promote equality. “What is best for the country in which you would like to live: A policy that aims to create as much equality as possible between salaries, living conditions and study opportunities, or A policy that allows people who work hard to earn more and live better than others?” In 2006, as already in 2002, 75% chose the first option. Among them, unemployed person(83%) and small businessmen are overrepresented, as are pensioners, farmers, civil servants, even professionals and self-employed persons. By community affiliation, support for equality is strongest among the Druze(89%) and Shiites(84%). The economic crisis has obviously undermined the belief in the superiority of a free-market system and in the wisdom of laissez-faire policies. Naturally, this also influences people's choice of government. “Which would you prefer: An honest and clean government that rules with a firm hand, or A government that is perhaps a bit corrupt but respects civil freedoms?” Already in 2002, two thirds of the respondents had come out in favour of clean, firm government; in 2006, 85% shared this view. Only 15% want more freedom even if a bit 38

E pluribus unum? of corruption is inevitable. It is impossible to avoid the conclusion that indignation about perceived injustice and corruption is challenging the liberal convictions of Lebanese society. This indignation is found without any significant differences across all educational and income levels in all communities; it is significantly higher only among the Shiites. The choice of clean government with an iron fist correlates directly with the level of education. 85 Against this background what are the opinions and attitudes about the current Lebanese system of government and possible alternatives? "There are many countries that, like ours, have a variety of different groups – language groups, religious groups, ethnic groups and others. Together these countries have different systems of government and there are differing opinions about the best system. We would like to present some of these opinions. Please tell us for each opinion whether you think it is acceptable or unacceptable for Lebanon.” "The country must be partitioned and each group should found its own state." "The strongest group should govern. The other groups must accept what this group decides." "The numerically strongest group should govern. The other groups must accept what this group decides." "One group should govern, and groups that don't like it must leave the country." "One single party that everybody can join should govern without an opposition." "Everybody votes for the party of their choice. The party or parties that win form the government; the other parties remain in the opposition." "Everybody votes for the party they want, but the government should be formed on the principle that all groups should have a share of power." 86 Agreement in%, rounded 1987 2002 2006 4 11 9 20 8 6 6 12 10 10 9 6 35 28 33 71 54 51 80 65 68 85 For the first option: 72%(no formal education) – 83% – 86%- 84%- 87%(university degree). 86 In 1987 the item read:"Given the nature of Lebanese society, all important political decisions must have the agreement of all large communities." The new formulation was chosen to facilitate international comparisons. 39

The acceptance of partition was greater after the civil war than during it, but has declined again in recent years. Support for this option is slightly above average among Christians of all communities and much lower among Shiites(5%). The willingness to accept the domination of the strongest group has also declined. There are no significant differences by either religion or community. In 2006, one tenth of the respondents are prepared to accept domination by the largest group. Agreement with this concept is above average among the Sunnites (15%) and below average among the Shiites(9%). Only one percent of the Druze find it acceptable. Support for the extreme form of domination – if you don't like it, get out – is found across all groups without socio-structural distinction, but is now lower than ever. By contrast, a full third of all respondents accept the one-party system. Support is well above average in Burj Barajneh(86%), in Jnah and Uzai(75%), in Tarik al Jadida (50%) and in the eastern suburbs. There are no significant differences by age, education or community. In 2006, a bare majority of respondents approve of a first-past-the-post system of majority democracy, whereas support for the traditional Lebanese system of proportional democracy has risen to more than two thirds. In both cases there are no significant differences between Muslims and Christians, but there are among individual communities. Remarkably, 56% of the Maronites accept a majority democracy, whereas only 46% of the Shiites do. Equally remarkable is the finding that acceptance of the Lebanese system correlates inversely with age. 87 In 2002 and 2006, the survey not only asked about the acceptability of various types of government, but went on to examine which of the alternatives respondents thought was best solution. The replies broke down as follows: Partition Domination by the strongest group Domination by the largest group Extreme domination of one group One-party system Majority democracy Consociational democracy(proportional representation) Majority democracy or proportional representation Other double responses No reply 2002 4 1 2 2 9 2 35 4 4 18 2006 4 1 2 1 17 17 44 14 87 Acceptance of“joint government”: 76%(youngest) – 73% – 61% – 60%(oldest). 40

E pluribus unum? Six in ten respondents in 2002 and 2006 think one of the democratic options is the best solution. Among the non-democratic options, support for a one-party system almost doubled. In Lebanon there is a solid, but by no means overwhelming, majority in favour of democracy. More respondents favour proportional representation than majority democracy as the most suitable solution. Age plays a role in these preferences. The preference for a one-party system correlates directly with age. 88 Majority democracy finds greatest support in the 25 – 34 age group. 89 Agreement with a form of democracy that gives all groups a share of power correlates inversely with age and finds a clear majority among respondents younger than 35. 90 The Lebanese system is the preoccupation not of those existing the stage, but of those entering it. Besides age, community affiliation is also a significant determinant of government preference. Sunnites Shiites Druze Maronites Greek Orthodox Greek Catholic Armenians Average Figures in%, rounded One-party system 18 18 30 17 26 16 33 19 Majority democracy 22 17 24 24 13 18 5 20 Power sharing 50 56 38 51 49 52 62 52 Power sharing finds greatest acceptance in all large communities. Except for the Druze, about half the respondents in all groups find this the best solution, and among the Shiites 56%. Almost a quarter of the Maronites and Druze and one fifth of the Sunnis prefer simple majority democracy. Somewhat less than one third of the Druze and less than a quarter of the Sunnites, the Shiites and the Maronites – i.e., the communities that have parties with claims to leadership – think a one-party system is the best solution. There are interesting differences in the choice of most suitable system between the followers of different politicians. Support for the option of a one-party state is above average among admirers of Jumblat(32%), Berri and Aoun(25% each). Majority 88 11%(youngest) – 15%- 26%- 30%(oldest). 89 21%%(youngest) – 23%- 17%- 17%(oldest). 90 55%%(youngest) – 54%- 50%- 50%(oldest). 41

democracy is particularly favoured by supporters of Franjieh(35%) and Jumblat (26%). Acceptance of power sharing is most pronounced among followers of Berri (59%), Nasrallah(56%) and Hariri/Siniora(55%). The picture among professed party members is different again. Support for a oneparty state is well above average among members of the FPM(30%), the Future Movement(29%) and in particular the SSNP and the Baath(36%). 91 Proponents of competitive democracy are most common among members of the PSP(45%), the Marada(39%) and the Future Movement(29%). 92 Agreement with a system of power sharing is strongest among members of Amal(68%), the Lebanese Forces(55%) – and the CP(78%). Remarkably, only 33% of the Future members favour this option, in contrast to the views of Hariri’s and Siniora’s followers in general. The responses to the following statements indicate that there is a plurality in favour of sharing power for reasons of necessity rather than conviction: 1981 "Because of the nature of Lebanese society, important decisions need the agreement of all large communities." * "Lebanon won't have a strong and united government until authentic representatives of the communities have a share in power." 67 Agreement in%, rounded.* Between 70% and 80% 1984 1986 1987 2002 2006 * * 80 77 90 79 73 66 72 86 Social characteristics are statistically insignificant in the first of these statements. 93 In 2006, no fewer than nine in ten respondents agreed; agreement was equally strong among Muslims and Christians, and an above-average 94% among Shiites. Between 1981 and 1987 at least two thirds of the respondents agreed with the second statement, and in 2006 more than four fifths. 94 This is higher than the level of support for proportional representation – let alone the percentage that regard it as the best solution – and higher than the percentage of respondents who, through their support for multiparty democracy and freedom of the press, showed themselves to be true democrats. 91 However, not one of the Communist Party members in our sample is in favour. 92 Among Amal(7%), FPM and Hezbollah(14% each) members, support for majority democracy is far below average. 93 Agreement correlates only with age: 87%(youngest) – 91%- 91%- 96%(oldest). 94 Agreement also correlates with age: 80%(youngest) – 89%- 87%%- 92%(oldest). 42

E pluribus unum? In short: the Lebanese are in favour of a system of democratic power sharing because they do not see any alternative. A majority are democrats by conviction, and a larger majority by necessity. 43

Perspectives of coexistence “Do you think it is possible to change a system accepted for reasons of necessity?” "Perhaps it doesn't look like it, but it is possible that the political system can be changed in the near future." "The strength of the different communities makes fundamental long-term change in the political system impossible." Agreement in%, rounded 1987 2002 2006 35 50 77 52 62 71 The statements contradict each other to some extent: the first expresses a desire, the second an insight. Agreement with both has increased since 2002, but for the first time the desire is stronger than the insight by a few percent. Respondents are similarly ambivalent in their responses to the following statements, which were presented for the first time in 2006: “The system has been flawed since its inception because it institutionalises the separation of the various communities. For this reason it should be re83 placed by a system more appropriate for a country aspiring to modernity.” “The Lebanese have reached a level of maturity that helps them get past sectarian isolation and we will be able to build a new political system in the 78 near future.” “Come what may, our communities are rooted in society and are not going to disappear. In the long run their power and resilience will stand in the way of important changes.” 76 “Considering the reality of Lebanese society, this system is acceptable, and if correctly applied, it can work.” 75 Agreement in%, rounded The desire for a change of system 95 (laced with positive rationalisations such as modernity and maturity) and the belief in its possibility 96 finds agreement among four fifths of the respondents. But three in four are aware of the social realities of the 95 Agreement is above average among Druze(94%) and Shiites(88%). 96 There are no significant differences between Muslims and Christians overall. Druze(86%) and Maronites(81%) are above average, Shiites(73%) below. 44

E pluribus unum? system 97 – and just as many view the Lebanese system as acceptable, if its rules are respected. 98 Against this background, how do people view possibilities and forms of coexistence? 1981 "During the crisis the regions had to get by on their own. Perhaps economic 99 decentralization is a good thing." 52 "Coexistence between the communities would be easier if each community had its own region." * "Whatever the political solution, Lebanon must remain a single economic entity. * Agreement in%, rounded.* Not asked 1984 61 23 86 1986 46 25 85 1987 31 18 70 2002 50 28 90 2006 78 48 96 In 2006,“economic decentralisation” no longer has the same connotations as“decentralisation” without adjective, which in wartime Lebanon was associated with partition or“cantonalisation” of the country. Today, it is accepted by almost four in five respondents. 100 Support for the idea of a separate region for each community has jumped, from a good quarter of the respondents in 2002 to almost half in 2006. Fifty-two percent of the Christians agree, but so do 45% of the Muslims. Support for an own region is well above average in the Jezzine(92%), but also in Sidon and the Nabatieh/Zaharani region. In Beirut, not only the inhabitants of Ashrafieh(69%) and the eastern suburbs 97 Agreement correlates directly with age: 72%(youngest) – 76% – 78% – 82%(oldest). There are no significant differences between Muslims and Christians. Agreement is above average among Armenians(91%) and Druze(84%). 98 Here, too, agreement correlates with age: 67%(youngest) – 77% – 77% – 81%(oldest). Agreement is above average among Maronites(83%) and Shiites(78%) and below average among Sunnites(68%) and Druze(69%). 99 In 2006, the statement was changed by inserting economic before decentralisation. 100 Slightly more frequently by Christians than Muslims and well above average by the inhabitants of Jbeil(92%), Sidon(91%), Nabatieh(90%) and the Jezzine(100%). In the Beirut district of Ashrafieh it finds 85% agreement and 100% in Burj Barajneh, Jnah and Uzai. 45

(61%) find the concept attractive, but also those of Jnah and Uzai. Respondents in Baalbek/Hermel, Tripoli and the Bekaa tend to disapprove, as do those in most districts of West Beirut and the southern suburbs. Even though one in two Lebanese may find the idea of separate regions for the religious communities attractive, almost none is prepared to give up the country’s economic unity. How do Lebanese view the relationship between religion and politics? 1981 1984 1986 1987 2002 2006 "One should not mix religion and politics." 84 84 87 93 78 79 "The best solution to Lebanon's present dilemma is a completely secular state and society." * 75 52 63 57 65 "Every Lebanese should have the right to join a secularized community that has the same rights as the other communities – personal status law, political representation, etc." * * * 50 54 70 "It doesn't matter what anyone wants, secularization doesn't seem to have a chance in Lebanon. Community membership is a reality you have to accept." * * * 54 64 69 Agreement in%.* Not asked This general formulation on the desirability of separating religion and politics draws strong support. Agreement among the Shiites is a below average 71%, still a majority, but faced with a strong minority that sees no problem in treating religion and politics as a unity. There is less support for precise concepts of full secularisation of state and society that go beyond the general desire to separate religion and politics. But there is greater agreement in 2006 than in 2002: almost two in three respondents are in favour. Support is stronger among Christians(69%) than Muslims and above average among Druze(85%), Greek Orthodox and Greek Catholics(74% each), Shiites(70%) and Maronites(68%), but well below among Sunnites(46%). The attitudes towards the creation of a secular community, with the same rights as the religious communities, which people can opt to join voluntarily, are particularly interesting. This proposal does not go as far as the call for a fully secular state, but makes membership of a religious community optional – freedom of choice for every individual and the opportunity for a gradual transition for society as a whole. Support for this proposal has risen sharply since 2002; for the first time, a clear majority of 70% 46

E pluribus unum? is in favour. Druze(80%) and Greek Orthodox(77%) are strongly and Maronites und Greek Catholics barely overrepresented, Shiites slightly and Sunnites(63%) strongly underrepresented, though this is still a clear majority. Although all religious authorities in Lebanon have always unanimously rejected this proposal for a first concrete step towards secularisation, it appears capable of winning a majority among Lebanese citizens. This also holds for the members of the leading political parties: agreement among Amal and FPM members is 81%, PSP 74%, Future 68%, Marada 63% and FL 61%; among Hezbollah it is 54%, a narrower majority, but still a majority. 101 Notwithstanding, two thirds of the respondents are resigned, regardless of personal choice, to accepting community affiliation with all its consequences as a fact of life. But how do the Lebanese view coexistence between these communities, the reality of whose differences they believe that they have to live with? "In recent years, 102 horrible things have happened. I fear that this has made coexistence between the communities very difficult." "In spite of the recent terrible events, I believe that coexistence between the communities is still possible." Agreement in% 1984 48 82 1986 44 79 1987 21 86 2002 43 80 2006 47 89 Remarkably, in 2006 as many respondents as in the worst years of the war, 1984 and 1986, think coexistence has become more difficult. This view is held by an above average 52% of the Shiites. The question of between which communities relations are perceived as difficult must remain open. 103 Despite this, nine in ten respondents believe that coexistence is possible in the future. What do people think about the balance of power, militancy and peaceful resignation? 101 As many as nine in ten members of the SSNP and the Baath Party agree with this proposal, as do 83% of the Communists. 102 In surveys during the war years:"In recent months,...". 103 It is striking that agreement with this statement is particularly strong in the districts of Beirut in which there were clashes between Shiites and Sunnites later in 2006: Tarik al Jadida(100%) and Chiyah/Ghobeiri(61%). 47

1987 2002 2006 "In the struggle between the different communities in our country we're all losers." 77 83 82 "In the present situation, given the strength of the different communities, it is necessary to search for a compromise and come to some agreement." 92 80 87 Agreement in% The findings of 2006, and of 2002 before that, show that the great majority have learnt from the war. More that four in five respondents agree that everybody loses when the communities fight against one another. 104 The awareness of the need to compromise is not as strong as it was in the war years, but still accepted by somewhat less than nine in ten Lebanese. Opinions differ more widely on the external and internal causes of the conflicts in Lebanon. "Even if foreigners stayed out of Lebanese affairs, it would be hard to reach an understanding; it is possible that the conflicts between hostile Lebanese groups would break out again." "If foreigners stayed out, conflicts in Lebanon would never end in victory or defeat for anybody." "In spite of all that has happened, the Lebanese could reach agreement among themselves if foreign forces would stop interfering in our affairs." Agreement in%, rounded.* Not asked 1984 * * 79 1986 * * 69 1987 50 56 75 2002 66 70 79 2006 63 75 83 In 2006, almost two thirds think that conflict between hostile Lebanese groups is possible, but three quarters are convinced that such conflicts would not end in the victory or defeat for any group, and four in five believe that if foreigners kept out the Lebanese could reach agreement among themselves. 104 Despite the direct correlation with age, 78% of the youngest age group, which did not experience the war, also agree. Agreement is slightly higher among Christians(84%) than Muslims(79%). There are no significant differences by religious community. 48

E pluribus unum? More Christians(66%) than Muslims(60%) believe in the possibility of renewed internal conflict. But more Muslims than Christians are convinced that in such conflict there would be no winners. 105 More Muslims are also convinced that agreement between the Lebanese is possible. 106 Finally, we look at a few opinions and attitudes that are fundamental for coexistence between different communities and groups. "Whether we like it or not: when different language, religious, ethnic or racial groups live in the same country, they must either dominate or be dominated." "Very different religious, ethnic, language or racial groups can live together in the same country, accept one another and respect each other's rights." "A country with groups with different traditions is wealthier for it and its society benefits." "The monotheistic religions like Christianity and Islam believe in the same God and teach similar ethical and social principles." "A good friend is a good friend, whether he is called Georges or Muhammed." Agreement in%.* Not asked 1981 * * * 92 96 1984 1986 1987 2002 2006 * * 53 60 59 * * 70 86 89 ** 64 80 88 82 88 91 81 77 96 96 98 92 96 The conviction that people must either dominate or be dominated has grown in postwar Lebanon and stabilised at around 60% in the past five years. It is particularly strong among Christians(65%), but also shared by a majority of Muslims(55%). 107 Between 2002 and 2006 the view that different groups could come to accept and 105 Muslims 77%, Christians 72%. 106 Muslims 85%, Christians 81%. 107 It correlates directly with age: 55%(youngest) – 61% – 61% – 65%(oldest). Three quarters of the Greek Orthodox agree, but only half the Shiites. 49

respect one another made some headway. 108 The proportion of those who are convinced that cultural diversity enriches a country has risen substantially. 109 In 2006, fewer respondents than in 2002, and fewer than in the war years, agreed that Islam and Christianity had a lot in common. That said, more than three quarters are still convinced of this. 110 More respondents than in 2002 think friendship across religious barriers is possible: 96% of all respondents – and 98% of the Shiites – believe that Muhammad can have a friend called Georges, and vice versa. To summarise: almost four in five Lebanese believe that another political system is in fact possible, but almost as many believe this will never come about. Almost four fifths favour – economic – decentralisation and almost half would like separate regions for each community. Practically all want to maintain the country’s economic unity. Four fifths favour the separation of politics and religion, and two thirds prefer a secular state and society. Seven in ten respondents – a remarkable increase over previous findings – support the creation of an optional secular community. But almost as many are resigned to the inevitability of membership of a religious community. About half of the respondents believe that coexistence between these communities has become more difficult – but nine in ten still think such coexistence is possible. People have learnt from the war: eight in ten respondents are convinced that in the event of conflict between the communities everybody would lose. Hence, nine in ten are convinced that compromise is inevitable. Cultural tolerance is high and diversity is regarded as a treasure. A large majority recognise the common ground between Islam and Christianity, and an even larger majority agree that friendship is possible across religious barriers. The question asked at the beginning can now be answered. All the results of the 2006 survey – like earlier surveys – indicate that the Lebanese want to be a nation. E pluribus unum ? The findings on individuals’ ties with their communities and to the political opinions rooted in these suggest that although the Lebanese nation is an unum – for reasons of political conviction and more so of necessity – at the same time it remains a plurum . 108 Christians below average(87%) and Muslims above(91%). 109 There are no differences between Christians and Muslims, although there are between individual communities: Armenians(96%), Druze(93%) and Shiites(90%) express aboveaverage agreement and Sunnites slightly below-average. 110 This includes a disproportionately large number of Sunnites(87%) and far fewer Maronites (70%), Druze(70%) and Greek Catholics(72%). 50