Gellner and Islam

Gellner and Islam Gellner and Islam
Author: Shankland, David
Journal: Social Evolution & History. Volume 2, Number 2 / September 2003


This paper examines Gellner's theories of Islamic Society, and argues that whilst they are perhaps rather too blunt, they are nevertheless profoundly important. By way of illustration of this contention, material is drawn upon from research in Turkey and in Germany, research mostly conducted amongst Turkish Alevis. Whilst this minority community appears to constitute an exception to Gellner's model, they complement his overall approach rather than refute it. In conclusion it is suggested that despite its flaws, Gellner's work could provide the basis of a wider and extra-ordinarily fruitful investigation into social change and new movements within Islam, and indeed Islam's place in Europe.


Of the diverse fields in which Ernest Gellner made his reputation, his work on Islam remains perhaps the most controversial. It is the one area where he has been dismissed, sometimes out of hand by his fellow academics, and that which is regarded as being most out of kilter with current understanding of ethnographic and anthropological theory2. It has inspired rather little secondary analysis, certainly less than his historical writings, or his theories on nationalism3. Yet, North Africa was the only region where he undertook fieldwork in the conventional sense, and he retained a fascination for the ideas that he developed whilst there even as he concentrated upon his wider analyses of world social history. Certainly, no study of his intellectual contribution could be made without referring to his thoughts on Muslim societies.

We are faced then with a problem. A topic about which Gellner wrote repeatedly, as if it held a significant, even vital place in his overall thought, is regarded subsequently as being amongst his least successful. Why should this be so? Perhaps inevitably with such a varied thinker, the answer is not simple. In part, it might be because he failed to produce one, single work on Islam that might summarise his position in quite the same way as he did with his other theories4. The closest perhaps, is the long first chapter in Muslim Society (1981). However, its allusive style mitigates against straightforward digestion of its argument5. Lacking a straightforward representative text, even those who might regard themselves as being well-versed in Gellner's writings tend to gain an impression of his theories from his summaries of his own position, such as may be found in Postmodernism, Reason and Religion (1992), or Anthropology and Politics (1994). These summaries inevitably render all the more terse the already sometimes rather concentrated prose.

It is also the case that, in his theories of Islam, Gellner had the misfortune to enter unfashionable territory twice over. He relied upon the validity of a famous British anthropological insight, known colloquially as ‘segmentary lineage theory’, for his overall model of Islamic society, so much so that it might be regarded as essential to it. Yet, a whole series of writers over several decades believe that they have refuted lineage theory generally6, and it is exactly on this issue that many of his critics, such as Munson or earlier commentators such as Hammoudi, have felt most confident in attacking him (Munson 1993; Hammoudi 1974). This almost unanimous assault has without a doubt gravely weakened the impact of his ideas within the anthropological fraternity.

Gellner's work on Islamic society also assumes that there is a connection between a person's viewpoint and their place in the social order, a link that the earlier generation of structural-functional anthropologists had taken for granted. Such an assumption underlies indeed not just his work on Islam, but is entirely integral to his wider project, as for example a glance at The Legitimation of Belief (1973), or the later Plough, Sword and Book (1988) illustrates. Yet this approach is today profoundly unpopular, so much so that even those who, upon mature consideration, would not necessarily find such a connection illogical or inappropriate are not given the necessary intellectual background to appreciate that this theme structures much of his writing. Thus, the mere fact of his being part of an earlier generation in a field that has undergone increasingly swift changes has rendered many of those who come later confused by both the immediate complexity of his writings and their underlying presumptions.


There is yet another difficulty, though this time of his own making. Gellner can hardly be blamed for the intemperate rejection of segmentary lineage theory7, or indeed for the success of fashions that he spent considerable effort trying to impede. However, he has been faulted, and I think quite rightly, for working with models that are ultimately too simple. This, for example, has led Alan Macfarlane in a lucid recent monograph to explore inadequacies in his theories of feudalism8. It has also stimulated Roberts to point out that Gellner's approach to the social map of North Africa is rather too blunt (Roberts 2002). Indeed, it often appears that Gellner assumes that tribal formations are the only significant social structures that operate outside the immediate authority of the state, and in Muslim Society, Gellner castigates Montagne quite unmercifully for the seemingly plausible suggestion that ‘checker-board’ moieties may be a significant aspect of the social life of the mountain Berbers.

Most seriously, perhaps, Gellner's theory of Islamic society occasionally appears rather static. It does not consider the dynamic interaction of different religious groups within a similar region, trans-national aspects of religious practice, or indeed how Islam might fare abroad, outside its traditional heartlands (or indeed perhaps even outside the Maghreb). Thus, though there is not the shadow of a doubt that interaction with other faiths, particularly in the context of migration, is a crucial aspect of the Islamic world, it is not immediately clear from his work how Gellner would approach this issue or the extent to which his approach may be generalizable outside its immediate focus.

This list of problems, both deserved and undeserved, is indisputably formidable, and could even be extended, for example, by noting the ways that Gellner failed to take the varied history of Islamic countries into account9.

It might be thought therefore that Muslim Society must remain no more than a limited tour de force, one that will be difficult to take further, or for subsequent generations to build on. Nevertheless, this essay takes up the challenge through an ethnographic presentation of Turkish material, and specifically Turkish migrants in Germany. My study is very narrowly focussed – it would certainly take a much longer account than this to go through the possible ramifications and consequences of each difficulty for his theory – but I think nevertheless that it illustrates certain issues that go to the heart of Gellner's conception of Islamic society.

In spite of the severe qualifications I express, I believe that it may be demonstrated, quite conclusively, that there is much that is valid in Gellner's underlying approach, particularly his assertion of the paramount importance of a group's orientation toward the state, of patrilineal social organisation, of the reciprocal link between hierarchy and ideology, and his emphasis on the relationship between faith, social change and modernisation. Even given its faults, I would hold his essay to be one of the utmost relevance in our study of the Islamic world and indeed its diaspora. The remaining part of this essay is devoted to exploring this point, initially with a brief exposition of Gellner theories, then a consideration of the Turkish ethnography itself.


A useful way to conceptualise Gellner's thought on Islamic societies is to divide it into two: his description of traditional life, and his vision of changes that may be associated with their modernisation or industrialisation. The contrast is sometimes questioned as being wrong-headed, with some of the debate surrounding the use of the word ‘traditional’. This is, I believe, a red herring. The labels are not important. Gellner's point is that society is transformed radically as it modernises, in all sorts of different ways. This was one theme that he did not try to simplify overmuch.

Nevertheless, in Muslim Society Gellner did present a very distinct set of ideas, depending on whether he was discussing the traditional or the modern or modernising Islamic world. We may sum these up thus: drawing upon a number of thinkers, he concurs with them that pre-modern Islamic countries are characteristically divided internally geographically between those who accept central rule, and those who would reject it. He also agrees that a distinctive attribute of those who lead their lives in opposition to central rule is that their societies are largely tribal, and that this collective cohesion provides them with the capability to withstand governmental troops, and on occasion even overcome them. He suggests too, that in those societies opposed to the central state, indigenous mediators may emerge whose right to judge is decided not on the basis of any formal qualifications but rather through birth, and that this birthright is governed by the patrilineage from which they stem. Just which patrilineage turns out to be ‘mediator-producing’ is not inherently predictable, but they are nearly always regarded as in some way appropriate to take on that role because of an auspicious sign from God. This gives rise to a neat circle: because the right to be a mediator is given by religious sanction, and the mediators themselves are also the representatives of religion, there is a mutual reinforcement of temporal and sacred authority that serves to strengthen and protect their position.

Gellner found this picture of traditional social life within Islamic societies entirely persuasive and became markedly irritable when sceptics questioned the lineage-model upon which his description was clearly based. However, whilst empathising with his opinion of the sceptics of lineage theory, when the Turkish material is taken into account, it becomes immediately clear that his analysis may be complemented and refined.

In Turkey, those groups who have in their traditional life expressed opposition to the state (whether Ottoman or Republican) often fall loosely into contrasting positions. There are those who are self-consciously tribal, rather large-scale and frequently even ostentatiously rebellious. These large-scale tribal groups are, just as is often noted in the early literature, prone to a rather Old Testament sense of the right to pursue reciprocal revenge. This results in a rather unstable fission and fusion that may lend itself to mediation by lineages which stand outside the immediate fractious situation. Whilst not entirely so, such groups are typically Sunni Kurds, and occupy the eastern and south-eastern part of Anatolia. They conform closely to the type regarded by Gellner as being the only consistent societal formation within Islamic societies that is founded upon opposition to the influence of the state.

In fact, there are others, still predicated in opposition to the state, still rural but quiescent, sedentary and divided into much smaller groups than the larger tribal formations. What might be called the social tone is also quite distinct. These sedentary groups draw upon a much more quiescent view of Islam than is envisaged in the segementary model. They are usually Turkish, but of a persuasion known as ‘Alevi’, and are profoundly influenced by the teachings of the Bektashi brotherhood. Here, whilst there are still patrilineal mediators, relations between members of the group are much more intimate. Rather than draw upon a philosophy of revenge, there is a strong bias toward religious quiescence. This has sometimes led commentators to assume that there is a direct connection between the Alevis and Christianity10. Whilst intriguing, such a historical connection is not immediately relevant here. What is important to us is that patrilineal dispute mediation is sanctioned by a powerful esoteric philosophy that insists upon peace rather than revenge, and that it is able to work in sedentary rather than nomadic societies that nevertheless predicate much of their social identity through their opposition to central rule. This contrast is summed up in the table below11.

Table illustrating different forms of rural opposition to central rule in Anatolian Islam


Group in Turkey

Dominant religious and cultural philosophy

Transhumant/ nomadic/tribal, in organisation very close to the segmentary lineage model posited by the British school (and Gellner).

Mainly Kurdish tribes in east, but found also in other southern and western regions in lesser numbers.

Strong ethos accepting the idea that violence may lead to reciprocal employment of force, religion frequently expressed through acceptance of sacred hierarchy, knowledge often enthusiastic rather than learned, uneasy relationship with state, barely accepting central authority.

Very-small scale dispersed sedentary communities. Patrilineal links important but localised, no large-scale groups united by kinship.

Turkish Alevis, mainly found in central-eastern areas.

Emphasis on small-group collective rituals accompanied by dance and music, dislike but tend to avoid rather than rebel against central authority, religious teaching is the prerogative of holy lineages who also mediate in quarrels, strong esoteric tendency leading to generalised affirmation of the importance of peaceful, neighbourly relations.


It may be remarked that Gellner made no claim to be making an exhaustive catalogue of human social groups and that, therefore, neglect of this distinction is not significant. In fact it has profound consequences, and most clearly so when the transition to the modern world is considered. In general, Gellner's argument puts him in the position of being an early sceptic of the position that modernity invariably leads to secularism. In Islam, he suggests that the growth of military technology weakens the capability of the tribal groups to compete with the centre, and that the rise of nation-states favours, rather than undermines, the emergence of a simplified, orthodox form of faith.

The explanation that he offers to account for this emergence emphasises the possible fusion of belief in an almost entirely transcendent God with the bureaucratic, individualistic existence that is characteristic of life within a modern nation. This, according to Gellner, permits literal faith and modern citizenship to be reconciled rather than conflict with each other. Whether his account is entirely valid or not, it does contain an important and clear contention: that any esoteric, hierarchical, inward-looking mystical form of faith largely becomes rejected, leaving the field clearer for a much more puritanical, egalitarian expression of belief, one that may express itself in politically active terms.

Both with regard to the Islamic world, and with regard to Turkey, this general contention is borne out. The esoteric, hierarchical aspects of faith were dismissed by the early, secularist Republicans when they sought to implement their modernising reforms. Instead they encouraged a rather sharp Sunni egalitarian puritanism in which little trace of the rich and complex Sufi tradition of the Ottoman Empire remained12. This emphasis is apparent still in Turkish Islamic practice today. There is also no doubt that faith is buoyant, and that despite the secular basis of the nation, religion has become increasingly prominent as each decade since the Second World War has passed13. The more recent, triumphant success of the Islamist Party in Turkey, which came to power with a majority government in late 2002, would seem only to confirm again Gellner's predictions.

However, there are also other currents of thought. The secular Republican model has convinced some people, who though they make up different and perhaps mutually antagonistic sectors of the nation, such as the moderate political parties, the political left, the army, the majority of working women, the dominant chambers of business, nevertheless constitute a highly significant proportion of modern Turkish society. Further, though there is abundant evidence to show that religious faith can be combined with wealth, the established middle classes in general have adopted a form of modestly pious secularism, one that their children seem cheerfully to have converted throughout the eighties and nineties into a consumer enthusiasm for music and enjoyment that leave little space for any but the most vague sense of religiosity. It might be, then, as a recent work by the American anthropologist Jenny White suggests (White 2002), that the Islamic populist resurgence remains linked to poverty rather than economic development, a point that the widespread support that the Islamist parties draw from the poorer migrant shanty-town areas would seem to support.

There is also the case of the Alevis. It would be fair to state, I think, that despite any refinements that certainly need to be articulated, the main thrust of the reformulations that Gellner noted as being typical within Islam do manifest themselves in Turkey, at least amongst the majority Sunni population. Any disagreement would concern the level of shading, the respective proportions to any one factor or faction that one might care to give, arguments that might indeed rest in part upon one's definition of faith or secularism. The Alevis, however, are a quite different matter.

Amongst their community, the changes that have taken place appear to be the very reverse of those predicted by Gellner. As they become part of the nation, far from stressing the orthodox aspects of their religious faith, the Alevis typically affirm the esoteric part of their creed. The ‘five pillars’ of orthodox practice receive less, not more attention. Here, faith becomes redefined not as belief in a supremely transcendent being who may only be appeased through repeated practice but sublimated into a sense of appropriate moral behaviour and justified by frequently offered assertions that a person's worth and the appropriate route to an inner God does not, unlike that of the Sunnis, demand any particular attendance at the mosque.

It would be vastly too confident in our powers of understanding of social change to offer any exact explanation as to why this transformation has taken place. In my published work, I have attempted to link it to the changes that take place as the Alevi communities modernise, suggesting that as they merge with the outside world, the transference of loyalty to the Republican government weakens the patrilineal structures, the holy lineages, and encourages the dispersal of the tightly-linked small communities, and their face-to-face rituals, that they helped to lead (Shankland 2003a). From the point of view of the individual, what appears to remain is a generalised sense of the importance of honest behaviour, conformity to which in itself leads to a sense of fulfilment of religious self-worth. Some individuals may encourage collective religious ceremonies taking place, but the compulsion that is part of village life becomes, in the urban setting, reduced to an optional attendance, even though the individual moral teachings of that esoteric faith remain.

This is inevitably a blunt summary of a very complicated set of issues that are themselves changing very quickly. There can be little doubt, however, that modernisation has led the Sunni and Alevi Turkish populations respectively into quite different dominant ideological directions in their everyday lives. Whereas the political orientation of the Sunni population (whether Kurdish or Turkish) has since the commencement of democratic elections inclined toward requesting the state to take greater responsibility for teaching Islam, the majority of the Alevis have supported the Republican People's Party, Atatürk's vehicle for secular reforms. Whilst wider debates within the Sunni community have typically surrounded the place of faith in the modern world, and the possible reconciliation of science with divine revelation, those within the Alevi community have been markedly more sceptical in tone, questioning instead the worth of belief, or even the existence of God. The Alevis indeed may be said to be an Islamic community questioning literal adherence to any creed - not just at the political level but also often at the level of individual faith. They do so through an almost exact reversal of the cultural mechanism posited by Gellner, whereby it is the orthodox not the esoteric that is dismissed to a secondary role in their life and thoughts.

Of course, no movement is static. Whether this is a permanent change is a matter that only time, and systematic research, will illustrate. However, overall, this development both supports and contradicts Gellner's model. It contradicts it sharply in that the Alevis illustrate a way in which the majority of an Islamic community have embraced secularism. It supports it in that the most important internal divisions within Turkish society turn out to be, just as he assumed, predicated upon whether a group is part of or against the government. Those who accept central rule, the Sunni Turkish majority, have followed a path almost exactly that he regards as most likely for Muslim societies to follow. The two main groups who predicate their existence independently of the state in traditional society have not. The one, the Alevis, has embraced secularism. We have not discussed the other, the Sunni Kurds, here. It may be noted though that as well as being known for their strong piety, they also produced a famous but violent expression of Marxist nationalism channelled through the PKK, a markedly non-religious organisation. Thus Gellner's overall appreciation of the complexity of the transition that these periphery groups undergo needs reconsidering. Nevertheless his underlying assumption that they in some way are radically different is emphatically supported even though the distinction ‘against’ or ‘for’ the state appears to reflect itself in the transition to modernity more strongly than he allowed for.


The study of migratory movements would appear to provide a solid way that this ethnographic contrast may be evaluated and revisited. It might be argued, for example, that the Turkish Alevis' insistence upon supporting the secular basis of the Republic is no more than a logical choice in the light of the persecution that they fear at the hands of the orthodox majority. Accordingly abroad, in the liberal atmosphere of modern Germany, they may feel no such secular compulsion. A parallel case could be asserted concerning the Sunni migrants and the relationship between economic affluence and faith. If religious resurgence is no more than a reflection of perceived economic inequality, as is so often assumed, then one might then expect to find a gradual lessening of religious activism amongst the socially mobile migrant community abroad.

There is an intriguing aspect to Gellner's position that, even though he did not discuss migrants at all, might imply that he would suggest the opposite. Albeit perhaps counter-intuitively, his argument appears to assume that self-identification with any egalitarian bureaucratic state apparatus through a sense of citizenship is enough to provide a framework for the maintenance of Islamic belief14. If this interpretation is valid, it suggests that integration with the host nation, even if nominally Christian, will not necessarily lead to a lessening of faith. This has awkward policy implications because it implies that successful inter-cultural dialogue will need extremely sophisticated articulations of what exactly is being discussed: a blanket appeal to an ill-defined ‘integration’ will be very little likely to be effective because it will not address the relationship between multi-culturalism, secularisation and faith sufficiently directly. Or, to put it another way, if Gellner is right there is no ‘hidden hand’ that will assure secularisation goes hand in hand with absorption into European affluence.

These of course are highly abstract arguments, even if fascinating ones. It is easy to forget how difficult it is to disentangle different causal factors, indeed that several different impulses may be taking place at the same time. With regard to Germany, where this research has taken place, the picture is further complicated because the relationship between religion and the German state (and respective Länder governments) is changing quickly and in part shaped by the complex special rules that govern the practice of religion in Germany as a whole. It is also the case that the vast and growing field that consists of migration studies in Germany possesses a preponderance of sociologists and demographers rather than anthropologists. This leads to a certain emphasis on macro-analysis, on comparative integration studies that deal with very large samples. Often the differentiation between Alevi and Sunni amongst the population selected for analysis is not noted in the initial design, rendering it extremely difficult to disentangle what may be due to differences between the two groups and what is more common to the migrant experience as a whole. Nevertheless, in order at least to begin the discussion, I discuss below the preliminary findings of a research project amongst the Alevi community that we have begun in Germany.


The initial aim of our study is to trace the fortunes of the migrants who have emigrated from one particular Alevi village in Anatolia who now live in Germany15. This gives rise to an apparently initially rather small data set, about 90 households, but nevertheless, as a study, appears to possess certain technical advantages. One of the problems dogging the otherwise resurgent field of Alevi studies is an emphasis on cultural ‘revival’. This has given rise to a stream of publications exploring Alevi traditions and ‘identity’, but led equally to other aspects of the lives of the Alevis being ignored. Everyday questions of social and geographic mobility, economic success, participation in associations, willingness to accept the sobriquet ‘Alevi’ become replaced by a blanket idea of a rather simple ‘Aleviness’ which vastly over-simplifies the complexities of their existence. By concentrating on all the members from one village, we hoped to be able to offer a much rounder picture of social life than is otherwise usual, indicating for example, those inactive, as well as active, within Alevi associations, or those not interested as well as those interested in pursuing their religious philosophy in their new land.

In the event, too, any earlier fears that the sample might be too small were quickly overcome by the fact that, geographically, the villagers households in Germany are vastly spread apart from one-another, from as far as München in the south to Nordhorn on the Dutch-German border. Our problem has become, therefore, how to investigate even this comparatively small sample adequately, as in effect clusters of villagers have to be considered a number of different and disparate social settings rather than just the one that constitutes the village in its traditional Anatolian setting.

Whilst it is premature to make categorical judgements on a piece of research that we hope will last for a sufficient length of time to provide at least cautious longitudinal indications of social trends amongst this community, our preliminary impressions are coloured by one factor in particular: the fact that overall satisfaction levels are extremely high. The answer of any such emotive question depends of course how the query is phrased more precisely. However, it is already clear that work patterns are stable, with unemployment almost zero, and divorce rates are low. Nearly all who can avail themselves of the opportunity to become German citizens as changes in the law facilitate their doing so, and many of the villagers are beginning to buy their own homes in Germany rather than, as is so often noted in the literature, investing in Turkey. Sociologically, there are naturally many fascinating questions that may be investigated, such as links with the village and with Turkey, the maintenance or otherwise of kinship patterns, differential patterns of integration across generations, and so on. There are also, naturally, some persons who have not succeeded in forging a successful life in the foreign setting. These persons perhaps are worn out emotionally or physically, a few have returned to Turkey. Nevertheless, and even including the fact that not everything is straightforward, our enquiry is profoundly coloured by the migrants' overwhelmingly positive expressions of their life overall in Germany16.


In as much as our preoccupation here, the reformulation of traditional religious life in the urban setting, may be summed up simply, it may be stated that once more the migrant Alevis continue to contradict Gellner's model of modernisation within Islamic societies. Broadly, whilst they debate their own traditions strongly, there is no mass movement toward Sunni orthodoxy. Instead, though they remain uncertain and divided as to the most appropriate way to maintain their own distinctive religious tradition, they support secularism firmly. The support for a division between state and personal faith is just as strong as in Turkey, refuting any suggestion that secularism is purely a rhetorical device to avoid repression.

Further, it may be recalled that, in the Turkish setting, the Alevis possess hereditary religious leaders. Known often as dedes (lit. grandfather), they are responsible for teaching Alevi doctrine and take responsibility for religious ceremonies. Gellner assumes that such hereditary privilege is incompatible with modernity, causing such leadership by birth to loose popularity drastically. Amongst the Alevis, such a shift has occurred, but only partially, and then in such a way that merits a brief explanation.

Whilst historical trends can only be summed up with great care, the gradual politicisation of Anatolia within the modern Republic appears to have reflected itself within the Alevi community in certain discernable ways. Those who were most active in the Republican People's Party substantially redefined their traditional religious mores as ‘culture’, kültür, and were opposed to the hereditary religious leaders. However, those villagers less politically active, and indeed the dede lineages themselves, appear to have been less clear-cut in their rejection of Aleviness, Alevilik, as a religious philosophy, and continued to give importance to the concept of hereditary leadership. This gave rise to an internal split within the communities that, whilst potentially serious, was in fact masked by a remarkable degree of ambiguity in the way that Aleviness could be celebrated. For example, even radical villagers would sing songs and dance dances that possessed clear roots in religious tradition, albeit in a secular setting, whilst dedes themselves were not opposed to drinking, music or dancing in wedding celebrations, and joined in willingly17.

Whilst the Alevis in Germany are profoundly influenced by events in Turkey18, it would appear that this underlying division is reforming in a slightly different way amongst them. In as much as an Alevi person seeks religious fulfilment or indeed recognition of their distinctive way of life, they almost invariably join or form civil associations, in Turkish, derneks. Such associations are dotted around Germany, in most towns where Alevi migrant workers may be found, such as Berlin, Essen, Köln, Bamberg or München. Gradually, however, many of these small associations have opted to join an umbrella association known as the ‘Alevi Federation’, based in Köln. All associations who take part in the federation are known by a similar sobriquet, ‘Alevi Cultural Centres’19.

As the name implies, the dominant philosophy of this umbrella association is that ‘Aleviness’ may be understood as a culture as much as a religion. One of their most successful ventures is the organisation of a huge music and folklore festival evening in Köln, an event that was later repeated in Istanbul, which they called ‘The song of a thousand years’, Bin Yılın Türküsü. At an everyday level, they publish a magazine, ‘The Alevis' Voice’ Alevilerin Sesi. In as much as they seek explicit recognition of ‘Aleviness’ from the German state, they do so stressing that it is a separate and distinct form of Islam, one that is not linked to Sunni or orthodoxy in any straightforward way. Indeed, they sometimes add that their ideal is an entirely separate ‘church’ that would be entirely secular, one organised along Christian lines, with a trained clergy rather than one that is decided along hereditary lines. Ironically, even in order to take part in this negotiation for official recognition, they have to redefine themselves as religious rather than cultural community in terms of German law. This has resulted in a leading member of their community, himself a lawyer, concentrating in his doctoral thesis on the necessary legal devices by which such proof may be offered.

The wider Alevi community, however, whilst not opposed per se to the federation in as much as they feel extremely strongly that a failure to organise will lead them to be dominated by an expansive Sunni orthodoxy, are not nearly as strongly opposed to the concept that the hereditary dedes should be responsible for teaching religious mores. When they hold ceremonies around Germany, they still insist that dedes should preside. Those Alevis in our sample who sense a religious calling still only feel able to claim spiritual leadership if they themselves are from a dede lineage. Contrariwise, those who are dedes recall often their duty to the community, and agonise over that point at which they may be regarded, both in terms of their own path in life, and in the eyes of their followers, as ready to take on a moral and spiritual position of responsibility.

Equally, these persons' sense of being part of the Muslim community is often less bounded than those who lead the federation's initiatives. They may, for example, simply regard their form of Islam as being a profounder version of faith, and not regard orthodox Islam as in itself illegitimate or mistaken. Politically, such a position may lead them to sympathise with a movement in Turkey known as the Cem Vakfı, led by İzzetin Doğan. İzzetin Doğan, rather than confront the Turkish state directly, is attempting to seek support from its treasury to maintain Alevilik as a religious path that is not incompatible with Sunni Islam. In Germany, this approach appears to receive its strongest expression through a break-away movement from the federation known as the ‘Alevi Academy’. Led by a dede, and based in part at a large Alevi association centre at Wiesloch, near Heidelberg, the academy seeks to train dedes in various aspects of Islamic and Alevi history so that they will continue to be proficient at leading the community. Whilst courses have newly begun, they have attracted support from known and respected researchers from the wider Alevi and academic body20.

Any attempt to make instant judgments is naturally fraught with difficulty. However, considering the matter throughout the period of fieldwork, it appears to me in part that the partial survival of support for hereditary religious leadership amongst the wider Alevi population in Germany may be explicable as being due to a readiness amongst the population as a whole to secularise at an individual level, or to put it another way, to abdicate religious responsibility to others. It is part of Gellner's thesis, one that he notes throughout his writings and lectures, that in Islam a man is his own constant revolution, that every man is potentially an author as well as actor in the playing out of his religious life. Amongst the Alevis, this is not the case at all. Most men have no desire to see their traditions disappear entirely, nor indeed for religion to vanish, but they view it as the dedes' job to ensure that any decline is halted. Broadly speaking, this leads them to support constitutional representation so long as it has no power to interfere with their own everyday lives. A parallel in our society is not so far away, of course, in that many countries in Europe still possess monarchies that represent a community through hereditary principles without the power to exercise authority that such a position used to entail.

I admit readily that this is no more than speculation. It can only be fleshed out by fieldwork over a much longer time frame. There are also, indisputably, complex undercurrents, such as the interplay between a sense of Turkish and Kurdish ethnicity that sometimes comes to dominate different local associations, that we have not mentioned here. Nevertheless, neither the movement by the federation to model their community on the German established church nor the insistence of the lay Alevis that the dedes should be encouraged and permitted to continue their duties in Germany, would appear to be anticipated by Gellner in his work, making the Alevis an exception to his theories not just in the traditional setting but also as they come into contact with modern Europe.


To return, in conclusion, to the question of the Turkish Sunni community in Germany. Here, it would appear once more that Gellner's approach is more applicable. Recent research by the outstanding European Forum for Migration Studies at the Otto-Friedrich University Bamberg illustrates that satisfaction levels amongst the greater part of the Turkish immigrant community, whether Alevi or Sunni, appear to be very high. Their work, as well as other surveys, suggests that by a variety of measures and in spite of the undoubted problems of youth unemployment, occasional confrontation, and political controversy that may emerge, there is occurring a rapid process of absorption and integration21. Widespread empirical research such as this has provided the background for a fierce debate as to whether Germany is a ‘migrant accepting’ country from the cultural as well as the factual point of view. So successful are the ‘ayes’ in this debate, that they now regard it has having been won: in other words, that it should and will be regarded at highest level that it is an irrevocable part of modern German consciousness that migrants, with their distinctive ways of life, will henceforth be an integral part of the nation-state.

Yet, if faith and belief are looked at more specifically, the evidence as to whether this broadly successful migration experience has led to the secularisation of faith is much less clear. Amongst such a large community there are inevitably enormous and widely differing views, but there would appear to be no doubt that there is a resurgence of belief within the Sunni community, one that centres upon the ‘five pillars’ and emphasises the place of the mosque at the heart of the community. Here again, the pattern that Gellner notes, whereby one single figure notable for his piety and distinct interpretation, leads the community is often valid22. Usually, the centres of religious practice and worship that are so created and led by inspired individuals become embraced by one of the larger streams of thought within Turkish Islam in Germany, such as the Milli Görüş (supporters of Erbakan's political Islamic movement), Süleymancs (a brotherhood-like organisation of that name), Kaplanclars (followers of the Islamic revolutionary army founded by Kaplan), or mosques officially sponsored by the Turkish Directorate of Religious Affairs23. Here, whilst the socio-economic process of integration assuredly continues, these movements could not obviously be described as secular24. The reconciliation between bureaucratic participation in a rule-bound, egalitarian society and activist faith appears to take place far better than the liberal integration model, with its presumption that multi-culturalist toleration is a natural concomitant of participation, would appear to assume.

This rapid resumé is naturally at the expense of a nuanced assessment of the situation. Nevertheless, there would appear be sufficient evidence that Gellner's assumption that integration into a nation-state does not lead inevitably to secularism be further examined and discussed.

It is perhaps interesting, finally, to pause and consider why the Alevis should differ so markedly from these Sunni movements. Perhaps, once more, the clue lies in Gellner, this time on the very first page of Muslim Society, where he notes that Islam has never had to ‘render unto Caesar’ in the way that Christianity has done. This is a little too sweeping a comment. It is certainly not true of the Alevis, who are indisputably Islamic and expressly sought quietism in the face of the orthodox centre. Nevertheless, it might be this, sociologically speaking, that is the key causal factor that can help to trace why they have taken such a different path from the orthodox majority, and through breaking the rule bring it sharply into perspective. The Alevis possess today a desire to seek a modus operandi vis a vis the existing powers, whether they are in Turkey or in Germany, that envisages a separation of their culture and the state's rule. Yet, unlike the Sunni communities (or indeed the tribal communities with their sense of literal independence) the Turkish Alevis have rendered unto Caesar: they do not like central authority, but they have long ago sought to accommodate, ignore and shape authority rather than rebel against its rule. Nevertheless, the contrasting, mutually-reinforcing relationship between orthodox, Sunni Turkish Islam and the state makes it doubly clear how rare the Alevi case may be. Once more, Gellner may have hit the mark in broad even whilst being too simple in the models that he employs. Infuriating though this may be, it strengthens our sense that there is much in his work on Islam that deserves careful and extensive study, both in the Muslim world, and amongst its diaspora populations.


1 Senior Lecturer in Anthropology, University of Bristol. This paper was given initially to the Prague Gellner seminar in November 2002, organised by Dr Peter Skalník, to whom I should like to offer my warmest thanks. The research in Germany described in this paper was supported jointly by the Humboldt Foundation, Germany and by the UK Economic and Social Research Council. Contact:

2 Amongst the many critical of Gellner's Islamic theories are Munson (1993), Geertz (1982), Hammoudi (1974), and Roberts (2002). The late David Hart (2000), in the last set of essays on North Africa published before he died, wrote that he had specifically turned to social history as a rejection of Gellner's position. Not all have been critical, of course, and it should be noted that Wolfgang Kraus has found Gellner's work accurate (1991, 1998), and Şerif Mardin is a consistent supporter (1989).

3 Thus, Macfarlane's recent volume (2001) is in great part a sustained dialogue with Gellner's theories of history. There is nothing comparable discussing his theories of Islamic society in such depth. The excellent general volume on Gellner edited by Hall and Jarvie (1996) for example, nevertheless is substantially reliant upon reprinting the earlier critical articles by Munson (1993) and Hammoudi (1974).

4 In contrast, the path of Gellner's thoughts on nationalism is far easier to trace, from the initial exposition in Thought and Change (1964), through to Nations and Nationalism (1984)and the posthumous Nationalism (1997). We lack such a sustained, separate overview of his thoughts on Islam, and so far as I know no such essay exists in any unpublished form.

5 At risk of introducing a facetious note, one of my abiding memories of being at Cambridge in the 1980s (where Gellner was at that time William Wyse Professor) was an ebullient North American Master's student going around his class-mates challenging them to affirm that they had read the elliptical first chapter of Muslim Society through to its conclusion. I believe that his scepticism turned out to be entirely justified.

6 See, for example, the criticism of Kuper (1982), and the later summary that he made of the debates in his Anthropology and Anthropologists (1996).

7 Fortes noted this in the foreword to a pamphlet aimed against segmentary lineage theory published by Holy (1979) through the Anthropology Department in Queen's University Belfast, when he pointed out that the only way that lineage theory could be dismissed would be to assume that the anthropologists who have reported such phenomena were either incompetent or wilfully misrepresenting the situation on the ground.

8 Macfarlane (2001). Paul Stirling, who both admired Gellner's work, and was briefly his doctoral supervisor at the LSE, was also fond of making this point in conversation.

9 Already when Muslim Society appeared, its lack of historical sensitivity was noted by Vatikiotis in his Encounter review (1982: 68); ‘Nor can it [the book] deal with the core area of Islam, especially during the Abbasid period, when the separation between state and society was complete by 850 AD'.

10 This topic is well-treated by Ayfer Karakaya-Stump (2003).

11 This table contrasts the Turkish Alevis of central eastern Anatolia with the Kurdish Sunnis of the east as described by, for example, Martin van Bruinessen (1992). Though this division is extremely important socially (Shankland 1993, Chapter 1), the overall ethnic composition of Turkey is naturally very much more complex than this (see Andrews 2002). In particular, it should be noted that the Kurdish Alevis may not fit in with my characterisation here. My hypothesis would be that, in such communities, there is a marked clash between the quiescence taught through ‘Aleviness’ and the more aggressive tribal ethos. I believe that there is some support for this idea, though I regret that I have not conducted fieldwork amongst such communities. Recent essays by Van Bruinessen provide a first-class introduction to Kurdish Alevi religious mores (2000). See also the earlier essay by Bumke (1989) which appears to support this contention.

12 See Mardin (1999). Gellner also touches upon this issue in his one essay entirely devoted to Turkey (Gellner 1994).

13 For an account of organised religion's intensifying links with the state bureaucracy and public life since 1945, see Shankland (1999).

14 It is possible that explicit confirmation of this or otherwise may exist in Gellner's various published lectures and articles that appeared after the initial publication of Muslim Society. However, there is a simpler justification for this assumption in that Gellner needed the independence of the nation-state from any one religion for his argument to avoid being circular: if only Muslim nations produce the Islamic faith, then he has proposed no more than an infinite regress.

15 This project is conducted together with Mr. Atila Çetin. A paper given to the Max Planck Institute for Social Anthropology, Halle in November 2002, ultimately for publication, describes the methodological background to this field project in more detail (Shankland and Çetin 2002).

16 It is possible that such high satisfaction is a general characteristic of migrants' life in Germany. This possibility is returned to below. However, looking at the Alevi specifically, this feeling may be linked to the fact that the villages are a minority in Turkey, and therefore are accustomed to leading their lives in an environment where they are not the dominant culture. Occasionally, in conversation with the villagers, they have indeed put forward this explanation of why they should be able to cope with being foreigners within German society. Until survey projects distinguish between Alevi and Sunni in their design, this will remain an important moot point.

17 For a fuller description of this debate, see Shankland (2003a).

18 Cf. a very fine recent article on the Alevi community by Martin Sökefeld (2002), in which he stresses this point. See also Tan (1999).

19 The full title of this federation is Almanya Alevi Birlikerleri Federasyonu. Their journal, Alevilerin Sesi, is not easy to find abroad. However, they have a very informative web-site that may be found at

20 The Wiesloch association has a web-site at the following address,

21 EFMS maintains much of its research results on-line, and may be consulted at . On a positive approach to integration, see also the recent summary published by the Turkish Embassy in Berlin Zur Integration… (2002).

22 See for example, a good description of the establishment of places of Islamic worship in Bamberg by Mhçyazgan (1990).

23 Aspects of these movements have been described in many publications. See Abdullah (1991), Antes (1991), Bielefeld and Heitmeyer (1998), Heine (1997), Jonker (1999), Karakaşoğlu-Aydn (1996), Lemmen (1998), and in particular the works of Schiffauer, eg. (1991, 2000).

24 It is always difficult in an article of restricted length to make general contentions. However, in brief the Turkish Directorate of Religious Affairs (a body with a budget larger than the Ministry of the Interior) even though founded with the express purpose of buttressing secularism within the Republic, now has such a dominant role in the recreation, leading and interpreting of religion that it is highly debatable as to whether it may be regarded as secular in any straightforward way. The other three movements (Milli Görüş, Süleymanclar, Kaplancılar) are founded expressly with a view to strengthen Islam vis-à-vis the state.


Abdullah, M.

1981. Geschichte des Islams in Deutschland. Graz: Verlag Styria.

Andrews, P.

2002. Ethnic Groups in the Republic of Turkey. 2nd edition, with a supplementary volume. Wiesbaden: Dr Ludwig Reichart Verlag.

Antes, P.

1991. Der Islam als politischer Faktor. Bonn: Bundeszentrale für politische Bildung.

Bielefeld, H., and Heitmeyer, W. (eds.)

1998. Politisierte Religion: Ursachen and Erscheinugsformen des modernen Fundamentalismus. Frankfurt: Suhrkamp.

Blaschke, J., and van Bruinessen, M.

1985. Islam und Politik in der Türkei. Jahrbuch zur Geschichte und Gesellschaft des Vorderen und Mittleren Orients 1984. Berlin: EXpress Edition.

Bumke, P.

2002. The Kurdish Alevis: boundaries and perceptions. In Andrews, P., Ethnic Groups in the Republic of Turkey (vol. 1, pp. 510–519). Wiesbaden: Dr Ludwig Reichart Verlag.

Donnan, H. (ed.)

2002. Interpreting Islam. London: Sage.

Fuller, S.

2000. Thomas Kuhn. Chicago: University of Chicago Press.

Geertz, E.

1982. Review of Gellner’s Muslim Society. New York Review of Books, May 27, p. 25.

Gellner, E.

1964. Thought and Change. London: Weidenfeld and Nicholson.

1969. Saints of the Atlas. London: Weidenfeld and Nicholson.

1973. The Legitimation of Belief. Oxford: Basil Blackwell.

1981. Muslim Society. Cambridge: Cambridge University Press.

1984. Nations and Nationalism. Oxford: Basil Blackwell.

1988. Plough, Sword and Book. London: Collins Harvill.

1992. Postmodernism, Reason and Religion. London: Routledge.

1994. The New Circle of Equity. In Hann, C. (ed.), When History Accelerates (pp. 229–237). London: Athlone.

1995. Anthropology and Politics. Oxford: Basil Blackwell.

1997. Nationalism. London: Phoenix.

Gellner, E. (ed.)

1985. Islamic Dilemmas. The Hague: Mouton.

Gellner, E., and Micaud C. (eds.)

1973. Arabs and Berbers, from Tribe to Nation. London: Duckworth.

Hall, J., and Jarvie I. (eds.)

1996. The Social Philosophy of Ernest Gellner. Amsterdam: Rodopi.

Hammoudi, A.

1974. Ségmentarité, stratification sociale, pouvoir politique et sainteté: réflexions sur les thèses de Gellner. Hesperis 15: 147–180.

Hann, C. (ed.)

1994. When History Accelerates. London: Athlone.

Hart, D.

2000. Tribe and Society in Rural Morocco. London: Frank Cass.

Heine, P.

1997. Halbmond über Deutschen Dächern: Muslimisches Leben in unserem Land. München: List Verlag.

Holy, L. (ed.)

1979. Segmentary Lineage Systems Reconsidered. Queen's University Papers in Social Anthropology 4. Belfast: Queen's University Press.

Jonker, G. (ed.)

1999. Kern und Rand. Religiöse Minderheiten aus der Türkei in Deutschland. Berlin: Verlag Das Arabische Buch.

Karakaşoğlu-Aydın, Y.

1996. Zwischen Türkeiorientierung und migrationpolitischem Enga-gement: neure Entwicklungen bei türkisch-islamischen Dachverbänden in Deutschland. Zeitschrift für Türkeistudien 2: 267–282.

Karakaya-Stump, A.

2003. 19th Century Missionary Accounts of the Kızılbaş Religion and a Discussion of their Impact on Later Treatments of the Subject, with Reference to Hasluck's ‘Heterodox Tribes’. In Shankland, D. (ed.), Anthropology, Archaeology and Heritage in the Balkans and Anatolia, or the Life and Times of F. W. Hasluck, 1878–1920 (p. 22). Istanbul: Isis Books.

Kraus, W.

1991. Die Ayt Hididdi; Wirtschaft und Gesellschaft im zentralen Hohen Atlas: ein Beitrag zur Diskussion segmentarer Systeme in Marokko. Wien: Verlag der Ősterreichischen Akademie der Wissen-schaften.

1998. Contestable Identities: Tribal Structures in the Moroccan High Atlas. Journal of the Royal Anthropological Institute 4 (1): 1–22.

Kuper, A.

1982. Lineage Theory, a Critical Retrospect. Annual Review of Anthropology 11: 71–85. Palo Alto: Annual Reviews Press.

1996. Anthropology and Anthropologists. 3rd ed. London: Routledge.

Lemmen, T.

1998. Türkisch-islamische Organisationen in Deutschland. Eine Hand-reichung. Altenberge: CIS Verlag.

Lewis, B., and Schnapper, D. (eds.)

1994. Muslims in Europe. London: Pinter Publishers.

Macfarlane, A.

2001. The Riddle of the Modern World. London: Palgrave Macmillan.

Mardin, Ş.

1989. Religion and Social Change in Modern Turkey: the Case of Bediuzzaman Said Nursi. New York: State University of New York.

Mıhçıyazgan, U.

1990. Moscheen türkischer Muslime in Hamburg: Dokumentation zur Herausbildung religiöser Institutionen türkischer Migranten. Hamburg: Behörde für Arbeit, Gesundheit und Soziales.

Munson, H.

1993. Rethinking Gellner's Segmentary Analysis of Morocco's Ait Atta. Man 28 (2): 267–280.

Roberts, H.

2002. Perspectives on Berber politics: on Gellner and Masqueray, or Durkheim's mistake. Journal of the Royal Anthropological Institute (N.S.) 8 (1): 107–126.

Schiffauer, W.

1991. Die Migranten aus Subay. Stuttgart: Klett-Cotta.

2000. Die Gottes-männer, Türkische Islamisten in Deutschland. Frankfurt: Suhrkamp.

Shankland, D.

1999. Islam and Society in Turkey. Huntington: Eothen.

2003a. The Alevis in Modern Turkey: the Emergence of an Islamic Secular Tradition. London: Routledge.

Shankland, D. (ed.)

2003b. Anthropology, Archaeology and Heritage in the Balkans and Anatolia, or the Life and Times of F. W. Hasluck, 1878–1920. Istanbul: Isis Books.

Shankland, D., and Çetin, A.

2002. An Alevi Village in Germany: Preliminary Thoughts on Attempting to Replicate in a European Setting Stirling's Model of the Social Integration of a Rural Migrant Community. Unpublished manuscript.

Sökefeld, M.

2002. Alevi Dedes in the German Diaspora: The Transformation of a Religious Institution. Zeitschrift für Ethnologie 127: 163–186.

Tan, D.

1999. Aleviten in Deutschland. Zwischen Selbstethnizierung und Emanzipation. In Jonker, G. (ed.), Kern und Rand. Religiőse Minderheiten aus der Tűrkei in Deutschland (pp. 65–90). Berlin: Verlag Das Arabische Buch.

Van Bruinessen, M.

1992. Agha, Shaikh, and State. London: Zed books.

2000. Mullas, Sufis and Heretics: The Role of Religion in Kurdish Society. Istanbul: Isis Press.

Vatikiotis, P.

1982. Review of Gellner’s Muslim Society. Encounter 58, March issue: 68.

White, J.

2002. The Islamist Mobilisation in Turkey, a Study in Vernacular Politics. Washington: University Press.

Zur Integration

2002. Zur Integration der Türken in Deutschland: Allgemeine Behauptungen und Ergebnisse vin Studien, Botschaft der Republik Türkei. Berlin.