IntroductionFootnote ¹

In recent years, both academic and journalist discussion have rendered ‘global Islam’ a prominent focus of discussion. Yet despite the profusion of commentary, the meaning and substance of the term global Islam has become increasingly opaque. Part of the reason for this is that global Islam has rarely been formally demarcated as a field of enquiry. It is far from certain what its study involves, leaving the question of the researcher’s remit vague in the extreme. Indeed, there is room for scepticism as to whether such an entity as global Islam exists outside of the imagination of political scientists and journalists; and as to whether, even if it does exist, such a vast phenomenon can feasibly be studied. Against this background, the aim of this essay is to return to basic principles by offering a working definition of global Islam together with a model how it can be coherently conceived and in turn studied.

The following sections define separately Islam and the global as objects of study, before combining the two to offer an analytical definition of global Islam. In so doing, the approach taken here is firmly rooted in the etic approach of the social sciences rather than the emic approach of theology and some branches of Religious Studies and Islamic Studies. At its core, this essay attempts to respond theoretically to the analytical problem of Muslim diversity, to the empirical fact that Islam is not the same everywhere, and in so doing to develop a methodology for studying global Islam that builds on rather than undermines this empirical recognition of Muslim pluralism. The importance of such a heuristic approach is all the more important in view of the temptation to conceive global Islam as a homogeneous phenomenon, via a singular and monolithic model that appears to be embedded in the very term ‘global Islam’.

Defining Islam: The problem of one and many

When attempting to define Islam for the purposes of social science research, whether historical or anthropological, it is crucial to avoid the interpretive illusions of essentialism. As A. Kevin Reinhart has defined it, ‘phenomenal essentialism is the assumption that there is some intrinsic form of Islam that transcends time and place; an essence of Islam’ (Reinhart Reference Reinhart and Wheeler2003: 24; also Reinhart Reference Reinhart2020). The pursuit of such an essential Islam has its rightful place in human endeavour: it is the goal of theology and of the thoughtful believer in general. But the price of conceiving Islam in such ahistorical, transcendental, theological ways is that it undermines any attempt to analyse Islam as it exists in any concrete context, in any social situation at any point in time or space. For the essential, timeless emic Islam of the theologian is an abstraction, a Geist that cannot be grasped, examined, or dissected. As Reinhart continues, ‘it is crucial to insist that it is Muslims who are agents and to avoid the hypostatization of something called “Islam” into an agent that acts, thinks, or believes. Islam does not think, require, or hold positions; Muslims do’ (Reinhart Reference Reinhart and Wheeler2003: 24).

Yet the essentialist model remains a pervasive one, with deep roots in not only the emic conceptions of believers but also in the etic conceptions of scholars, not to mention the media. During the last century, Muslim reformers and globalizers drew on the theological doctrine of tawhid (divine unity) to stress the singularity of the Muslim faith and the unity (ittihad) of its believers as a single worldwide umma (community). Meanwhile, some ‘Orientalist’ scholars developed their own trans-historical model of an unchanging and essential Islam. In complex ways that have yet to be unravelled, these emic Muslim and etic Orientalist essentialisms cross-fertilized and in turn supported one another. They underlie, and continue to mutually reinforce, both the media and policy conceptions of Islam today. For the idea of the essential unity of Islam – and thence of Muslims, and in turn of a ‘Muslim world’ (Aydin Reference Aydin2017) – has deep intellectual roots, forming habits of thought that are difficult to shake off. Making matters worse is the paucity of descriptive vocabulary for Islamic pluralism, a paucity of terminology that is itself the legacy of these emic and etic essentialisms and unities. The few broadly recognisable descriptive terms that we do have for Muslim pluralism (such as Sunni/Shi‘i or Sufi/Salafi) are themselves emic, generalizing and largely trans-historical labels that unsurprisingly refer to such large groups of people over such long periods of time as to offer only a slightly less generalist picture of a Sunni and Shi‘i bloc. The end result is a linguistic delusion whereby we speak of ‘Islam’ and imagine there is such an entity out there in the world; or if we are a little more sophisticated, we say that there are a handful of ‘varieties’ or ‘versions’ of Islam: Shi‘i, Sunni, and perhaps Salafi or Isma‘ili: again, all emic labels. In either case, the essentialist trap remains the same and this intangible Islam remains as analytically elusive as ever.

To be able to grasp this Islam analytically therefore requires us to make several basic conceptual moves that, however elementary to the social sciences remain remarkably marginal to mainstream scholarly and public discussion. The first move is to shift from abstraction to concretization. This involves bringing to centre stage the aforementioned empirical recognition that there is no single Islam in the observable social world, there are only different Muslims and in turn texts, institutions, and practices that these self-ascribed Muslims actively label as ‘Islamic’ (or alternatively as ‘unislamic’). The second and consequent move is to recognize that because the texts, institutions, and practices that these self-ascribed Muslims label as Islamic show quite marked differences and divergences – to the point that one believer’s Islam is another believer’s heresy – there is not any single entity (or concretely, any set of persons, texts, institutions, and practices) that exclusively deserves the label ‘Islam’.

Via the simple yet radical potential of such empiricism, of forming our categories of description and analysis on the observable evidence of the surrounding world, we are therefore led to the conclusion that there is concretely, tangibly, observably no single version of Islam in practice. Instead, at the observable social level, there are only groups of social actors calling themselves ‘Muslims’ who agree to label their favoured doctrines, texts, institutions, and practices as ‘Islamic’; that is, as expressing and promoting the beliefs and behaviours they consider as constituting ‘Islam’. In this way, we can speak of such a construct as Islam (or rather, a series of such constructs: a multiplicity of ‘Islams’) while retaining the analytically measurable constraints of empiricism. For these different visions of Islam comprise the aggregate of the texts, organizations, and practices that a specific group of people collectively agree or assent to labelling as (their) Islam. Different visions of Islam – sociologically conceived here in the plural – are in this sense ‘packages’ of beliefs and behaviours shared by specific social groups and maintained by particular organizations and their proprietary discourses.Footnote ²

To speak of the latter – of organizations and discourses – is not only to recognize social mechanisms for the maintenance and reproduction of group identity over time. It is also necessarily to introduce the question of authority, of what beyond free choice or conscience induces individuals to commit (at least outwardly) to upholding certain beliefs and behaviours. This issue is often framed in terms of the language of ‘tradition’, but tradition and authority are two ways of describing an interrelated set of processes. After all, ‘tradition’, from the Latin traditio, designates ‘that which is handed down’. And it is authority-building institutions that serve as the mechanisms for ‘handing down’ beliefs and behaviours through time so as to enable these beliefs and behaviours to garner the authority of ancestral practice, to become norms or orthodoxies (the latter itself a variable function of power). And this in turn further legitimizes and empowers the institutional mechanisms of tradition.

As in so much social scientific discussion, we are in danger here of allowing sociological labels that aim at the clarity of the concrete to sound like another set of abstractions. So let us seize the tangible implications of the centrality of institutions to the working of tradition. The implication is this: if it is institutions that select, preserve, and propagate the textual and spoken discourses that persuade a given social group to follow a particular set of beliefs and behaviours, then it is those institutions that stand at the centre of the package of phenomena we are calling religion, or in this case Islam.

Despite arguments about the Eurocentricity of the category of religion (Asad Reference Asad1993), when defined in this sociological manner the category does appear fit for analytical practice. For this package of beliefs and behaviours that a given social group agrees to share through the persuasive power of the institutionalized discourse of tradition does indeed cohere with the Latin designate of religio as ‘that which binds’. And at the core of ‘that which binds’, at the core of Islam as religion, as the distribution centre of the package of belief and practice, stand institutions. Since an institution is ‘an organization or other formal social structure that governs a field of action’ (Rojas Reference Rojas2020), the implication is that in order to investigate Islam in the human world we should focus on these ‘formal social structures’, whether they comprise religious organizations that survive sufficiently long to become more enduring institutions or the more numerous parvenu and short-lived organizations that have emerged in recent decades. There is also a role here for the individual transnational religious activist, many of whom have emerged since late nineteenth century in the wake of the prototypical figure of Jamal al-Din ‘al-Afghani’ (c.1838-97). Because such ‘individuals’ are only able to gain any degree of social impact by either founding their own organizations (however small scale) or attaching themselves to existing organizations (whether religious or secular).

This delimitation of the research focus to religious organizations in turn helps us manage the issue of scale and thereby of exponential datasets, a problem that becomes all the greater when we turn the analytical focus onto global Islam. To understand the social actuality of Islam is therefore not necessarily to investigate the beliefs and behaviours of all Muslims. It is to examine the organizations that define, authorize, and expound Islam; to examine the social mechanisms that not only ‘speak for Islam’, but tangibly embody it and discursively define it in the sociological reality of the world around us. Clearly, there is a danger here that we can precipitously shrink our analytical vista from a pluralistic multitude of believers to a small club of organizations who are not ‘representative’ of the world’s many Muslims. Yet this is a false problem: because ‘represent’ – speak for, platform, propagate – is precisely what religious organizations do. Whether they do so democratically, or in ways that accurately reflect the views of those whom they claim to represent, is another matter entirely, albeit one that does not change the basic social fact that organizations do indeed represent Islam.

What helps us manage the potential divide between organization and individual is the recognition, firstly, that religious organizations come in many different shapes and sizes; and, secondly, that all religious organizations are controlled by individuals or groups thereof. This is particularly the case with hierarchical and non-democratic organizations. In this way, we hang onto the fact that it is human agents qua Muslims – albeit collectively via organizations – who define Islam. The plurality and varied scale of the religious institutions that sociologically comprise Islam therefore admits to the analytical fold a wide variety of what are often loosely called Muslim ‘voices’ and ‘perspectives’. For this reason, a focus on Muslim religious organizations does not reproduce a monolithic or essentialist model of Islam. This becomes clear when we recognize the extraordinary variety of organizational forms through which various visions of Islam – various packages of beliefs and behaviours labelled as ‘Islamic’ – have been propagated over just the past century (Green Reference Green2020; Mandaville Reference Mandaville and Hefner2011). To name but a few, these forms comprise the mosque, the saintly shrine, the Sufi fraternity, the informal study group, the private or state-funded madrasa, the private or state-funded religious university, the student association, the missionary organization, the voluntary association, the online social media group, and the campaign-oriented pressure group.

As should be immediately clear from this brief list, these different types of organization vary in four interrelated ways. Firstly, religious organizations vary in the degree to which they are either based in a particular spatial location (as in the case of a shrine or madrasa) or based around a spatially distributed network (as in the case of the missionary organization or social media group). Secondly, they vary in terms of the amount of material and symbolic capital required to establish and maintain them (as in the contrast between a large state-funded religious university and an internet pressure group). Thirdly, they vary as to whether they are privately- or state-funded (and in turn directed). Fourthly, they vary in the degree to which they are authoritarian or democratic in structure (as in the contrast of the Sufi brotherhood and voluntary association).

What should be noticed about this approach is that it is not based around discursive content, around the substance of the ‘beliefs and practices’ that are often regarded as the beginning and end of religion. Instead, religious discourse is seen here as a function of organizational variation: in an environment where there is a greater plurality of organizations there will correspondingly be greater variation in religious belief and practice. And vice-versa. The implication in turn is that socio-political environments that do allow for a diversity of religious organizations (principally, liberal secular states where the state allows religious freedoms; and failed or weak states that lack the resources to regulate religious affairs) will in principle exhibit a greater variety of religious belief and practice. And, correspondingly, socio-political environments that do not allow for a diversity of religious organizations (principally, religious, Marxist, or ethno-nationalist states where the state attempts to control religious affairs) will in turn exhibit a lesser variety of belief and practice.

While it may be argued that this approach does not take into account what individuals may privately believe in their own consciences, or indeed practice in their homes, the point is that the concern here is with religion as religio, as a collective and public phenomenon, that gains social and sometimes political power (including the authority of tradition) by the very fact of its collective character, whether that collective is a spatially concentrated population or more widely distributed network.

This methodological step from a focus on the model religious institution of sociological theory to the multiplicity of religious organizations of empirical fact in turn allows us to account for the actual mechanisms that drive religious diversity and change. In other words, unlike static essentialist approaches to Islam, this model not only recognizes religious diversity but also explains how it comes about. Here we need to make the shift from the observation of static plurality to the analysis of dynamic competition between different religious organizations and their directing personnel. Because religious change – and the uneven distribution of religious change that adds further to the uneven contours of religious diversity – is driven by the organizational promotion of beliefs and practices that differ from those which already exist among a given population or in a given environment.

Again, the scale of the organization can vary from a state-funded religious police force to a website of an individual religious entrepreneur. From enforcement to persuasion, the techniques that these different religious organizations use to disseminate their proprietary visions of Islam also correspondingly vary. And so, in turn, do their degrees of effectiveness; that is, the degree to which any given organization can generate religious change and in turn increase religious diversity in a given geographical space; or in some cases, reduce religious diversity to generate religious homogeneity if the organization in question gains sufficient persuasive or coercive power.

Once again, the role of the state is essential here: if state resources are deployed to empower particular institutions (and undermine or persecute others), then ceteris paribus the beliefs and practices associated with those organizations will become publically dominant within the spatial boundaries of the state in question. And to the extent that the state does not or cannot regulate religiosity, a plurality of religious organizations will compete with one another for either dominance or, so to speak, market share.

To speak of ‘market share’ is necessarily to invoke the sociology of religious economy. A key contribution of this method has been to point to the importance of not only the ‘supply side’ of religion (here, religious organizations; or in the terms of religious economy, ‘religious firms’), but also to the ‘demand side’ of religion (here, those organizations’ followers; or, in the terms of religious economy, ‘consumers’ of a firm’s religious products and services) (Green Reference Green2014; Obadia Reference Obadia2013). A religious economy-based critique of the model so far presented here might then be that its focus on organizations not only neglects the religious individual qua consumer, but also relegates them into a position of passivity. In other words, the model does not reckon with the forces of religious demand as well as supply. However, this is not necessarily the case. Because religious organizations by their very purpose comprise not only the personnel who control them but also the wider membership comprised of those who, whether on a formal or informal basis, voluntarily ‘join’ or coercively ‘obey’ them. For by their nature as institutions in the sociological sense, religious organizations seek to control a field of social action (in this case, the highly variable obligations of Islam as a set of beliefs and behaviours).

Religious organizations are therefore relational entities that allow us to see how certain Muslims (that is, an organization’s directors or personnel) shape the beliefs and in turn behaviours of other Muslims (an organization’s followers or members). And, to return to the process of religious diversification, the rate and reach of religious change in any given environment is shaped by the extent to which existing organizations do or do not change (tradition); new organizations emerge (diversification); and old or new organizations contend with one other (competition).

What we have, then, is a sociological model that positions ‘Islam’ as the variable discursive and behavioural product of particular and sometimes competing organizations. Moreover, the activities and interactions of those organizations can be studied as a means of tracing the mechanisms of religious stasis or change, of standardization or diversification, in any given environment, whether that environment be the notional religious ‘free market’ of a liberal secular state or the attempted religious ‘monopoly’ of an Islamic state. What is more, the method has no inherent spatial limitations. It is not only suited to examining the ‘local Islam’ of a Muslim shrine in India in the manner of the classic ethnography of Islam. It is equally suited to examining the ‘global Islam’ of the transnational Muslim missionary organization. Because in either case, the primary focus is on organizations, and through them on the potential members (or ‘consumers’) to whom a given religious organization (or ‘firm’) reaches out, whether on a local, national, or international scale. With this more tangible definition of Islam in hand, we can now move on to grappling with the similarly slippery category of the ‘global’.

Defining the global: The problem of scale

This definition of Islam as the variable but nonetheless empirical product of multifarious religious organizations helps us solve the problem of scale in the analysis of global phenomena. For rather than facing the potentially infinite and inaccessible dataset of the private religious reflections of around two billion worldwide Muslims, researchers of global Islam are instead set the far more manageable task of examining the activities of one or more organizations that operate on a transnational rather than local scale. In other words, the study of global Islam becomes the study of the varying methods, success and impact of specific transnational religious organizations. Because for analytical purposes, the ‘global’ is not so much a scale as a process; it is less a question of where than of how. That is, a global analysis does not require the study of the entire planet. Rather, it requires the scrutiny of the means by which a given object of study moves – and is actively transferred and translated – across political or ethnic boundaries from one world region to another. And at a secondary level, a global analysis addresses the question of how the object of study (here, Islam) is transformed through such acts of transference.

This question of transformation is central to the larger field of Global Studies, which has developed its core analytics around issues of linguistic and more broadly cultural translation in the spatialized sense of translatio (from the Latin transfero: ‘to transfer, carry over’). In particular, Global Studies scholars have focussed on the translational process they have termed ‘hybridization’ (Pieterse Reference Pieterse, Featherstone, Lash and Robertson1995). A similarly defining question in the field of Global Studies has been whether globalization is alternatively a homogenizing or hydrogenising force (Hassi and Storti Reference Hassi, Storti and Cuadra-Montiel2012). This has particularly important implications for the study of Islam: namely whether under the impact of globalization, the world’s multitudinous versions of Islam have been further differentiated, or have alternatively experienced homogenization. The question of transformation more generally is similarly fundamental to the study of globalized religion. For while, as we have seen earlier, a few postmodern theologians aside, the emic view of religion is that true religion is unchanging in its essential character, social science approaches by contrast conceive religion as highly malleable and mutable in its forms and expressions. What Global Studies scholars add to the discussion is therefore a recognition of a specific type of change – in a word, hybridization – that frequently characterizes the global transference of cultural beliefs and practices, including religion and in this case Islam.

What this implies for global Islam as a field of enquiry is that its primary focus is not necessarily on theology nor even on scale. Rather, its focus is on two characteristically global processes, namely the means by which (1) particular religious organizations transfer their activities and theologies across geographical, ethnic or political boundaries; and, in turn, (2) the ways in which religion as theology, organization, activity, and community is transformed through its transfer across borders, whether in terms of texts, institutions, practices, or personnel. So, two key processes of globalization: transfer and transformation. In sociological terms, theological content and demographic scale may in this way be secondary, even corollary, to the concrete social entity that define them – namely, the particular religious organization and the particular social actors (whether suppliers or consumers) who control that organization’s decisions and thereby allocate its human, material, discursive, and symbolic resources.

By in turn delineating a set of comparative criteria with which to analyse different forms of religious organization, this approach to the study of global Islam can explain why some Muslim religious organizations (and their attendant packages of belief and behaviour) have increased their transnational impact while others have not. For not all cross-border religious organizations are successful in increasing their numbers of followers. And those that are successful vary enormously in their degree of demographic expansion and social diversification (that is, their spread among different ethno-linguistic groups). The comparative criteria for evaluating such variable rates of success should include, inter alia, strategies of fundraising and communication; mechanisms of transfer and outreach; varieties of doctrinal content; and involvement in high profile public campaigns. To understand the global as a religious process is therefore to explain how and why some religious organizations are successfully able to expand across borders, and thereby gain followers in what are necessarily different ethnic, linguistic, cultural, or political environments, and why other competing organizations fail to do so. We have, then, a model that explains how things happen.

Comparatively manageable as this form of global analysis is, it is clearly a collective research field rather than an individual research project (Clayer and Germain Reference Clayer and Germain2008; Nordbruch and Ryad Reference Nordbruch and Ryad2014). For to move towards a fuller understanding of global Islam requires two levels of research. The first level must comprise a series of specific case studies of the variety of Muslim organizations that operate at a transnational level, ranging from the most to the least successful. As in business studies, case studies of failure are as illustrative as success stories. This we already have to a considerable extent, albeit varying considerably in the degree to which different research projects have pursued the kind of institutional sociology suggested here. The second level must comprise synthetic works that aggregate, and draw larger conclusions from, the findings of these necessarily varied case studies. This we do not yet have, the closest being Matthew J. Kuiper’s conspectus of global Muslim proselytizing movements and Peter Mandaville’s survey of global Muslim political movements (Kuiper Reference Kuiper2021; Mandaville Reference Mandaville2020). At least not in a way that, either methodologically or empirically, grapples with the key problematic that this essay aims to identify and in some measure resolve, namely the empirical fact of global Muslim diversity and the historical process of the diversification (or alternatively homogenization) of multiple visions of Islam in terms of their organizational and discursive variety. Unless the field of global Islam is to be purely descriptive, this processual question must stand at the analytical centre of research. And in placing this question at centre stage, researchers must tackle the processual causes of both diversification and homogenization, namely the process of competition between different religious organizations. For whether through the dominance of one organization and its teachings, or through the co-existence of various organizations, both Muslim religious diversity and uniformity are different outcomes of competition between the purveyors of different ‘packages’ of Islam, that is, between different Muslim religious organizations. It is here that the empirical observation of Muslim diversity shifts from the descriptive to the explanatory level. And in terms of a methodologically global analysis, researchers must ask how these competitive dynamics and their outcomes have been shaped by the processes of trans-border transfer and transformation.

The final issue to be tackled in terms of defining the global is that of periodization. Given the practical importance of charting a field of manageable scale and charging its researchers with manageable datasets, the question of periodization becomes more pressing. For in terms of Islamic history, to adopt the broadest chronological definition of religious globalism as simply the patterns of transference and transformation undergone as religious organizations cross political or ethno-linguistic boundaries would take the field’s chronological remit back to the very beginnings of Muslim history. This is clearly too broad, not only for practical but also for disciplinary reasons: it would render the field of global Islam a mere rebranding of the larger field of Islamic history. Here it is useful to make a conceptual distinction between ‘global’ and ‘world’ Islam that builds on the distinction between global history and world history (Mazlish Reference Mazlish, Mazlish and Buultjens1993; Olstein Reference Olstein2014). For Mazlish and Olstein, global history is the study of a specific period of broader world history and the processes that define it, namely the modern era of globalization. In Olstein’s words, the field of global history ‘adopts the interconnected world created by the process of globalization as its larger unit of analysis, providing the ultimate context for the analysis of any historical entity, phenomenon, or process’ (Olstein Reference Olstein2014: 24). Given that most historians date this process and period of globalization from around 1870 onwards, it makes both practical and conceptual sense for the field of global Islam to adopt and so standardize these chronological parameters.

As James Gelvin and I have argued, this period of global history had a particular impact on Muslims worldwide not only through the political and economic forces of European colonial globalization, but also through the technological forces that saw Muslims exposed for the first time to such communication technologies as printing and steam travel, and in turn electronic and aeronautical communications (Gelvin and Green, Reference Gelvin and Green2014). Moreover, the post-1870 period was also characterized by what scholars have variously referred to as Islamic ‘modernist’, ‘reformist’, and ‘scripturalist’ movements that were active across large and increasingly interconnected regions of the planet. It was, then, a period of intense religious diversification, interaction, and relocation. What the rubric of global Islam can offer is therefore a way of conjoining these various empirical observations – about globalization, communication, diversification, reformation – into a coherent analytical field by asking how Islamic ‘reform’ – or what I prefer to call diversification – was shaped by globalizing processes of transfer and transformation. In this way, researchers can address the larger and shared mechanisms that fostered these multitudinous ‘modernist’, ‘reformist’ and ‘renewal’ organizations; how they competed both with one another and with more customary organizations (including such incumbent religious organizations as Sufi shrines and brotherhoods); and crucially, how these organizations variably succeeded in crossing the borders laid open by the political, economic, and technological forces of globalization.

Defining global Islam: From abstraction to organization

In light of these conceptual discussions of the component terms ‘Islam’ and ‘global’, we are now in a position to delineate a clearer definition of ‘global Islam’ as a field of enquiry. In terms of period, it should focus on a particular era of world and thereby Islamic history that reaches from around 1870 to the present. In terms of content, the field should focus on the range of Muslim religious organizations that moved across geographic, ethnic, or political borders during this period; and by corollary expansion, it focuses on the different discursive packages of beliefs and practices which these organizations promoted with varying degrees of success. And crucially in terms of process, the field should focus on the strategies and mechanisms by which these specific organizations and their proprietary visions of Islam were transferred across borders and transformed by the act of cultural translation.

Here the research field of global Islam has room for methodological flexibility. Because the interconnected environments in which these border-crossing organizations operate can be differently conceived as comprising social constituencies or religious marketplaces. Yet what is indispensable to the field’s etic analytical model is the empirical recognition of Muslim diversity, of the plurality of visions of Islam, however loud the emic claims of, or calls for, Muslim unity made by different Islamic organizations themselves.

As we have seen, this recognition of diversity is predicated both practically and conceptually on the organizational concretization of religion in the shared social world. This sociological and anthropological approach of ‘recognizing Islam’ (in the anthropologist Michael Gilsenan’s classic phrase) as consisting concretely as specific social actors, institutions and their discursive packages first developed in the study of small scale ‘local Islam’, that is, in the classic and manageable field environments of the ethnographic case study (Gilsenan Reference Gilsenan1982). In methodological terms, the question of how this approach can be adapted for the study of the far larger scales involved in the study of global Islam can be answered by keeping the analytical focus on religious organizations as congeries of various social actors, both suppliers and consumers, leaders, and followers. This focus on organizations includes the discursive and communicative strategies that different organizations deploy so as to operate across borders and compete (or cooperate) with other organizations they find there, whether public or private, Muslim or non-Muslim, religious or secular. As a scaling up and scaling out of existing methods, this trans-border approach builds on the well-tested anthropological technique of multi-sited research. For we should not forget that, in terms of both scale and process, to research the global is to study the means by which distant microsites become connected. To study the global is to examine the small scale in the connected aggregate (Green Reference Green2014 and Reference Green2020).

In this way, we can define ‘global Islam’ as an analytical umbrella term that refers to the various visions of Islam propagated across geographical, political, and ethnolinguistic boundaries by Muslim religious activists, organizations, and/or states that emerged in the era of modern globalization from around 1870 (Green Reference Green2020: 8).

While the field of global Islam should therefore be based on the recognition of Muslim diversity, such recognition does not imply that the field takes the study of all those varied historical visions of Islam and their various practitioners as its remit. This larger endeavour – which we might usefully distinguish as the study of ‘world Islam’ as distinct from ‘global Islam’ – would necessarily be unmanageable. Again, the conceptual definitions of the field of global Islam have practical consequences. For if researchers of global Islam are to focus on the strategies and mechanisms by which the proprietary visions of Islam promoted by specific organizations are transferred and transformed across ethno-political frontiers, then this remit necessarily excludes visions of Islam, and their organizations, that have not ventured across such boundaries since around 1870. The study of global Islam is therefore not the study of the ‘local’, ‘regional’ or ‘folk’ Islam of the anthropologist, even as it borrows from the methods developed by such studies. Of course, when organizations promoting such ‘local’ organizations do expand across frontiers (Tschacher Reference Tschacher2006; Werbner Reference Werber2003), they can fall within the remit of global Islam (albeit at the cost of no longer remaining heuristically ‘local Islam’). But, overall, during the twentieth century at least, those ‘local’, ‘regional’ and ‘folk’ iterations of Islam do not appear to have crossed borders as successfully as their ‘reformist’, ‘scripturalist’ and ‘Islamist’ competitors. In this way, by studying the mechanisms and strategies that did enable these other organizations to successfully expand across borders, the field of global Islam has much to offer scholars of ‘local’, ‘regional’ and ‘folk’ Islam by explaining how the latter have been increasingly outcompeted and marginalized by the deepening impact of globalization.

Here it is useful to make a categorical distinction between global Islam and what might be termed ‘world Islam’, whereby ‘world Islam’ refers to the versions of Islam that developed and adapted to local and regional environments during the millennium before the onset of modern globalization (Green Reference Green2020: 8). Thinking in terms of world Islam rather than through such older categories as ‘local Islam’ (or such proprietary counterparts as ‘Indian Islam’ or African ‘Islam noire’) may allow us to acknowledge the premodern interconnections made by Muslim ‘ulama (lawmakers) and ‘urafa (mystics) who crossed continents on foot, camel caravan, and dhow while at the same time recognizing that their period (and religion) had not yet been transformed by the forces of modern globalization.

Moreover, far from being identical, global and world Islam have often been in conflict. Yet in approaching the issue of the marginalization of more regional Muslim religious cultures (i.e., ‘world Islam’), it is crucial that we do not slip into teleological mode, particularly into the familiar teleology of globalization as homogenization. For as we have already seen, the question as to whether global processes of transfer and transformation lend themselves to homogenization or heterogenization remains open (though for my own part, I have argued elsewhere in favor of the heterogenization thesis: Green Reference Green2014: 281-88; Green Reference Green2020). And if there is one meta-question at the centre of the field of global Islam, then it is surely this. The way to answer it, though, remains through a primary focus on organizations as the tangible social instantiation of religion. For all Muslims worldwide to share the ‘homogenous’ doctrinal opinion that they should model their behaviour on the teachings of Muhammad is not the same as agreeing which individuals, and necessarily which organizations, best represent and expound the Prophet’s teachings. As ever in the social sciences, differences disguised by shared idioms and discourses are revealed in their social expression.