CHAPTER FIVE GLOBAL ISMAILISM: RE-THINKING DIASPORA, NATION AND RELIGION

In Chapters Three and Four I analyzed my interviews with twenty Khojah Ismaili informants in Nairobi, and explored their identities and identifications, and the Islamization reforms of the last few decades. These twenty people are part of the local Ismaili community, but also the larger transnational Ismaili population. They all have family in Canada, and many have friends and relatives in UK, France, Tanzania, Pakistan, Uganda, United States, India and Portugal. In this chapter I use the experiences of my informants to understand the relationship of Ismailis to the concepts of diaspora, nation and religion, that relate not only to the dispersed Khojah Ismailis, but also to Ismailis that come from the Middle East and Central Asia. I engage in a dialogue, based on my interviews, on how Ismailis negotiate multiple allegiances to place, people and belief.

Transnational Ismailism

I went to Kenya with the simple notion that the East African Ismailis constituted a diaspora. I was hoping to examine how their lack of a "homeland" could challenge and redefine the dominant definition of that term. After my interviews and through much of my writing, I realized that an engagement solely with "diaspora" is not adequate to fully expound the Ismaili case. Thus, I turn to ideas of the nation and religion as a way to think critically about the transnational Ismaili community. These terms are clumsy and limited, however, they provide modes of analysis which are useful in thinking about my ethnographic material, and more fully understanding the Ismaili case. Ismailis do not comfortably fit into any of the categories that we traditionally use to define migrants or people who share race, culture or belief. I hope the following discussion will be productive in placing the Ismaili case in dialogue with theories of diaspora, nation and religion.

While a strict definition of a term as broad as "diaspora" is not useful, there are several features traditionally considered to be central, especially a myth of return and a sustained interest and connection to the homeland. Although there has been a dispersal from South Asia to East Africa and subsequent migrations from both places to Europe, North America and elsewhere, the Khojah Ismailis who have lived in East Africa for several generations lack a connection to South Asia. Clifford argues that "transnational connections linking diasporas need not be articulated primarily through a real or symbolic homeland" (1994: 306), and my research in Kenya showed this to be the case for Ismailis. Their connections to kin and friends in Canada and other places were more concrete and salient for them, than connections with India or Pakistan.

Clifford also calls for a decentering of diaspora, and Gopinath (as I discuss in Chapter Two) takes this further to write the nation into the diaspora and vice versa. In such a model, the homeland and a teleology of origin/return are less important than lateral connections between dispersed communities. While this is an important intervention, it does not nuance the heterogeneity of differently positioned diasporic communities, and notions of modernity that operate among Ismailis. Living in Canada, for example, is usually a better diasporic location than the United Kingdom, which is seen as superior to East Africa. The quality of a diasporic location depends on safety, economic opportunity, standard of living, quality of education, racial tension, political stability, and so on. In emphasizing lateral links, it is important not to homogenize these diasporic locations, and to understand the political and economic factors which create hierarchies and different experiences within the diaspora itself. In different hostlands, Ismailis occupy distinct class positions - in North America they are a model minority, but in East Africa they form a wealthy merchant class. In emphasizing the transnational, we must be careful not to overlook the national in ways that may be problematic. In blurring the lines between nation and diaspora, we are also limited to thinking about diasporas which have a claim to a nation, or have a bounded territory, something which excludes Ismailis and others such as Parsis and Kurds. In some ways such an interpretation reinscribes the nation/diaspora dichotomy it is working to challenge.

There is a way, however, that Ismailis in India and Pakistan (broadly the "homeland") can be written in as locations in a decentered Khojah Ismaili diaspora, or as nodes in a "postnational network of diasporas" (Appadurai 1996: 171). Many of these South Asians have experienced internal migrations to urban centers and during partition, and physical and emotional displacement is a reality for Ismailis that have never left South Asia. What makes the notion of diaspora, in any of its current definitions, limiting for this project is the way in which it cannot bring Central Asian Ismailis fully into the theory. There have been no substantial physical migrations of these populations, and contact with Aga Khan IV was only re-established in the last decade after years of Communist rule. Unlike the Khojahs whose kin relations are transnational, there are no family links between Ismailis living in Central Asia and elsewhere. Over the last few years, Aga Khan IV has been promoting substantial interaction and exchange between Central Asian and Khojah Ismailis. These economic, social and religious relationships cannot be accounted for within current theories of diaspora.

The reason is that "diaspora" still connotes shared racial heritage. Considering Central Asian and Khojah Ismailis part of the same diaspora is difficult because of their racial differences, even though they are increasingly told to think of themselves as part of a global community. Even Paul Gilroy's theory of the Black Atlantic, which moves away from Afro-centrism, overlooks the interactions of people of African descent with South Asians in Britain and other people of color in the United States. Gilroy falls into "the conflation of...`black' as of African descent" (Gopinath 1995: 305). This reinscription of a racialized concept of diaspora comes from the implicit idea that "diaspora, in its traditional sense, thus refers us to a system of kinship" (Helmreich 1992: 245), which is itself racialized and gendered. While Central Asians cannot be racially tied to Khojahs through kinship, Aga Khan IV metaphorically invokes such associations by positioning himself as a spiritual parent, and offering "paternal and maternal blessings" to his murids, whom he often calls "beloved spiritual children."

At this juncture it is important to reiterate the emphasis that Aga Khan IV is placing on the Central Asians Ismailis, and the role that Khojahs are asked to play in forging connections. In a firman given in March 1997 in Nairobi, the Imam said:

I quote at length here to underscore the rhetoric which imposes a strong sense of duty to assist Central Asian Ismailis, and also the use of "brotherhood" as another invocation of kinship between populations that do not share a place of racial origin. I argue that the connection Aga Khan IV hopes to create between these populations is not purely a religious one, but that he is trying to incorporate Central Asians into the Khojah-dominated and Khojah-centric community which is already linked in terms of intra-diasporic connections. Both communities are adapting to each other by exchanging aspects of their practice, such as the Central Asian qasidas which are now sung in Khojah congregations worldwide. Aga Khan IV's broader reforms are significant in that he is changing both Khojahs and Central Asians to reconfigure the community, and establish a united transnational jamat. This firman excerpt highlights the limitations of the racialized concept of diaspora, in describing the interactions within the Ismaili community.

While a tacit racialized understanding of diaspora is clear, to exhaust the theoretical value the term can offer, it is useful to think about Ismailis in comparison to other communities, especially Jews and Armenians, which have long been held as prototypical diasporas (Tololyan 1996: 3). All three groups define themselves, to some extent, by their unique religious beliefs - Ismaili Islam, Judaism and Armenian Christianity. The group boundaries are not strictly birth-based since all three accept converts and are not racially exclusive. However, there is a strong correlation between race and religious belief which excludes certain people from full access to these communities. Many other diasporas have a religious component including Sikhs. It is unclear whether religious belief is strengthened in diaspora, or religion contributes to migrations - likely the two work together to define dispersals of people in terms of religion.

Formation of a religious diaspora may also be connected to the reason for dispersal. Both Jews and Armenians were forced to leave their homelands because of their religious beliefs. Because coercion was involved, strong feelings of loss and yearning for the homeland support a connection to the place from which people were exiled (Kirshenblatt-Gimblett 1994: 339). These painful histories are associated with physical spaces which diasporics reclaim and re-inhabit, namely Israel/Palestine and Armenia. In comparison, Ismailis left South Asia and East Africa under conditions of stress, usually because of economic or political tension, but they were not forcibly removed (with the exception of Ismailis in Uganda who were expelled during the time of Idi Amin), involved in warfare, or victims of genocide. In fact, their movements were promoted by Aga Khan III and IV, as discussed below.

In speaking of religious diasporas, maybe Ismailis can be grouped with other Muslims. The notion of an inclusive Islamic diaspora is suggested by Barbara Metcalf (1996), who coined the term "diaspora Muslims". While this provides a useful analytical framework based on broadly-defined shared beliefs, the term is too large to adequately link immigrants from the Middle East, South Asia, and elsewhere. Metcalf also includes African American Muslims in this category, emphasizing the similarities between the displacements of cultural migration and conversion to Islam in the West (1996: 4). This analysis does not pay adequate attention to different experiences of racialization or language and religious sects. For many Muslim immigrants, in their homelands these markers are often central to identity, especially differences in specific Muslim practices - Sunni, Shi'a and beyond. The African American experience, in particular, has a historical specificity which is substantially different from immigrant Muslims. It is possible that in future generations, there will be a more significant pan-Muslim identity in diaspora. In informal discussions, I have been told that while many diverse Muslims share mosques in Europe and North America, the social interaction between different communities is minimal. Ismailis, in particular, have jamatkhanas open only to members, a unique relationship to the Aga Khan, and practice a form of the faith that is recognizably different from other forms of Islam. While thinking about the relationships between Ismailis and other Muslims in diaspora is a useful exercise, including them in the category of "diaspora Muslims" is misleading, since based on my fieldwork, their strongest allegiances lie most closely with other Ismailis and not with Muslims in general.

In my discussions with Ismailis in Nairobi and elsewhere, a desire to claim territory has never been raised. This lack of a push for territory may have to do with the stabilizing leadership of the Imam, but may also reflect the comfortable position that most Ismailis occupy as middle class citizens in the countries where they live. As Arjun Appadurai explains, "displacement does not always generate the fantasy of state building" (1996: 166). Even if Ismailis were interested in land, would it be in South Asia, East Africa, Persia or Central Asia? Pre-exile Jews and Armenians had land-based nations from which they were displaced, whereas Persian Ismailis were involved in proselytizing in Central and South Asia and thus created pockets of murids throughout the continent. The issue of origins leads us back to the concept of kinship in diaspora, which operates in Jewish and Armenian communities more than it does with Ismailis who do not share a place of origin. Israelis have recently been grappling with the assimilation of Ethiopian Jews who are racially distinct within Ashkenazi-dominated Israel and the diaspora. To do this, they hark back to the twelve tribes of Israel, one of which is based in Ethiopia - and highlight the original link of this group to the majority of Jews. In this case, history and kinship can be invoked to facilitate interaction across racial difference.

Another critical difference between Ismailis and the prototypical diasporas of Jews and Armenians is language. Hebrew was revived from an unspoken language into the national language of Israel in the latter part of this century. While Hebrew is not spoken by all Jews, it is spoken by all Israelis, and is a symbol exclusive to Jews. Similarly Armenians, many whom have not been to their homeland for generations, continue to learn and speak Armenian. Language is a powerful force in uniting people, especially when it is exclusively used by a single group as a medium to transmit ideas through journalism, media and speech. Ismailis, on the other hand, do not have a single, unique language. There have been reforms to increase the use of Arabic among Ismailis, even though no Khojahs are native speakers, but this is limited to use in religious practices. The secular language of the community, which Aga Khan IV has encouraged all his followers to learn, is English. This is the language in which all his speeches and firmans are read. Neither Arabic nor English are unique to Ismailis. In fact, they seem to be strategic choices to allow Ismailis to form stronger links with other Muslims through Arabic, and to facilitate assimilation into African, South Asian, European and North American contexts through English. In this way, neither language allows for an exclusive community literature, or for an identity based on language, since Ismailis speak a variety of vernaculars.

Discussions about the relationship between territory, language and the formation of identities, are bound to lead to dialogue with Benedict Anderson's notion of an imagined community. He contends that the Western nation was imagined, that is constructed, through print-languages which laid the bases for national consciousness. Print capitalism accomplished this by forming unified fields of exchange, giving a new fixity to language, and creating hegemonic languages which were based closely on vernaculars spoken by people with power (1991: 44-45). The model of the nation which Anderson offers is intimately linked to print - something which is not parallel to the Ismaili case, since Ismailis do not exclusively share a language. He also speaks about the Muslim ummah and Christendom as communities "imaginable largely through the medium of a sacred language and written script" (1991: 13). Again Ismailism is excluded because Ismailis rarely refer directly to the Qu'ran, and do not natively speak Arabic. However, there are other texts which can function in similar ways to holy books.

Ismailis follow the direct advice of Aga Khan IV offered through speeches and firmans. These firmans are compiled into books, and are read during each prayer service. Unlike the Qu'ran, they are not compiled into a single, defined text - each book has different firmans. This allows the Ismaili Institute in London, which produces these books, to remove out-dated, irrelevant or undesirable firmans from circulation. This loose organization allows for substantial control by Aga Khan IV, who can easily remove a firman from circulation by not re-printing it. Thus, the teachings of the Imam provide some common point of reference for dispersed Ismailis, and since they are both read in jamatkhana and available for sale in print form, they are accessible to both illiterate and literate people. While the notion of print capitalism, and subsequent theories of "electronic capitalism" based on film and television (cf. Warner 1992) may not be applicable here, understanding the imagining of a community through texts is an important intervention. How can we fruitfully extend Anderson's definition to think about the imagined community as transnational? Are there other ways for this imagination to occur besides through print capitalism? What are the ways in which Ismailis are a nation or not one?

Ideas about the nation as imagined, like theories of diaspora that I discuss above, often come back to the "trope of the tribe" in that "there is always the suggestion that blood, kinship, race, and soil are... more natural than the imagination of collective interest or solidarity" (Appadurai 1996: 161). It is difficult for many to think outside of a paradigm where race, place and belief coincide. Here is where Ismailism, as a multiply located, yet non-territorially focused nationalism, challenges what constitutes nation. Ismailism is not linked to one space or soil, it encompasses people of several races (if we can still use such the term race as a noun) and it includes relationships not based on kinship or shared biological origin. Appadurai refers to such nationalisms as "trojan", and points out that they "contain transnational, subnational links and, more generally, non-national identities and aspirations" (1996: 165).

Part of the success of the transnational Ismaili community is that it is non-national, in that there is no push for an Ismaili nation. Ismailism does not pose a threat to hostlands or homelands. Ismailis are able to balance their local obligations as citizens of a nation, and their other obligations as members of a transnational community. The 1986 Constitution lays out clear guidelines about the allegiance of a murid to the Imam, but warns that this allegiance "is distinct from the allegiance of the individual murid to his land of abode" (Aga Khan IV 1986: 4). Also the constitution "shall apply to Ismailis worldwide, subject only to the overriding effect of any applicable laws of the land of abode of any Ismaili to the extent of any inconsistency" (Aga Khan IV 1986: 9). These rules result from the historical experiences of persecution and minority status that have forced Ismailis to be invisible at times, but it seems that Aga Khan IV expects his murids to follow Ismaili doctrines as long as they do not conflict with those of the hostland.

Adapting to the local culture has been the recurrent advice of the last two Aga Khans. Current pronouncements to assimilate are part of a history of negotiation with local laws and social systems:

Aga Khan IV has, on many occasions, emphasized the importance of becoming local citizens of the place of residence, which is why Ismailis have the highest rate of East African citizenship among all Asian groups (Salvadori 1989: 232; Kjellberg 1967: 43; Sumra 1990: 5; Kaiser 1996: 46; Walji 1980: 132). In the early 1980s when I was a child, Aga Khan IV came to Canada, and I distinctly remember that he repeatedly urged us to "Make Canada your home." This is common advice given by the Imam, and is very different from advice given by most diaspora leaders who urge sentimental nostalgic attachment to a distant place. It is probably because there is no homeland to return to that Aga Khan IV's advice is given in the first place, and taken to heart by his murids.

For the most part loyalties to the place of residence and to Ismailism do not clash. We can see the types of issues between other religions and the state that have been avoided by Ismailis: prayers in school, abortion, homosexuality, and so on. None of these topics are ever discussed by Aga Khan IV - it is understood that these decisions must be made individually and within the local legal structure. As Nadira, a middle aged woman explains, "in Pakistan [Ismailis] may do namaaz because other Muslims live with them." Others told me that in the past, Ismailis in Zanzibar observed more traditional Muslim practices because of the local Muslim population. Even the Westernization reforms of Aga Khan III, especially the push for Western dress for women in the 1950s, was part of his efforts to make Ismailis more acceptable in East Africa. The Ismaili emphasis on changing with the times lends itself easily to this adaptation, without making Ismailis feel that they are compromising their practices. These changes are easier since according to my informants, Ismailis value the essence of spiritual practice over the ritual surrounding it.

This attitude is largely an adaptation to being a minority group, and trying to be as unobtrusive as possible. The balancing of nation-state and transnational is necessary, and has been relatively easy for the Ismailis because of the mercantile class status that many of them occupy in Kenya and elsewhere. Their wealth, status, education and leadership from Aga Khan IV allow them to exist as non-traditional citizens and Muslims. They have also thrived as a liberal and unorthodox sect in Pakistan, where another Shi'a group, the Ahmadiyya Muslims, were declared un-Islamic by the Pakistani government in 1974 (Haider 1996: 41). The Ismailis, who are as non-traditional as the Ahmaddiyas in their religious beliefs, are unlikely to be treated in the same way. This is because of their wealth, and the status Aga Khan IV has created for Ismailis in Pakistan by building a private medical university and numerous health centers and schools. Philanthropy can be used as a buffer against hostilities towards the minority Ismailis. Their class standing and the stability that Aga Khan IV can provide, are substitutes for what land represents to displaced peoples - safety, comfort and protection. The Ismaili lack of desire for a state makes defining them as a nation less plausible. Furthermore, unlike nation-states which can enforce laws through military, police and legal systems, Ismailis do not have the power structure to coerce and inflict violence and terror - some of the most important tools a nation-state uses to control its citizens.

The impetus to be unobtrusive, to assimilate, to follow the practices of the land of residence, is the way that Ismailis have learned to deal with the state. Despite their allegiances to Aga Khan IV and their religious beliefs, they attempt to be a minimal threat to the nation of residence. As mentioned above, any conflict between national and Ismaili allegiances must be resolved in favor of the place of residence. This is a coping mechanism that some minorities adopt, and is ubiquitous among Ismailis since they are minorities everywhere, and thus occupy a tenuous position within many national structures. Some of the situations that must be negotiated include being a liberal Shi'a Muslim sect in the Islamic state of Pakistan, a merchant-class middle minority in East Africa, and a visible model minority in North America.

Aga Khan IV has also minimized the threat to all nations where Ismailis reside by living in France. In early times, Imams lived physically near their murids in Persia and later in South Asia. By moving to Europe, Aga Khan III was able to gain more personal and community status, and to associate with European aristocracy. This physical distance from a concentration of Ismailis was also effective in diffusing possible allegiance conflicts for murids, and preventing a threat to any nation by his local presence. The possible threat of an organized and wealthy minority in a nation is further dispersed by the significant philanthropy projects of Aga Khan IV in the form of schools, hospitals and clinics in East Africa and South/Central Asia. Philanthropy is a visible way for Ismailis to financially contribute to the nations they inhabit, to negotiate their tenuous status as minorities. The assistance they provide in the form of human services is significant, and is a way for Ismailis to act as good citizens and show their loyalty. Philanthropy is effective in decreasing the racialized financial threat that many minorities, such as diasporic Jews and Chinese, have long been charged with.

While Ismailis are not a nation in the sense that Anderson uses it - a nexus of people, place and language - there are definitely aspects of the transnational community which echo the structure of a nation. The national model had a profound influence on how Aga Khan IV and his grandfather have organized the Ismailis. I use the most recent Ismaili Constitution from 1986 as my primary reference to support this point. A dozen nations where large numbers of Ismailis live have several principle institutions which include a National Council that deals with policy making and planning; a Tariqah and Religious Education Board that deals with religious education and practices; a Grants and Review Board that deals with funding and budgets; a Conciliation and Arbitration Board that deals with legal matters; and central institutions that deal with education, health, social welfare, economic planning, youth and sports, and housing. Each country's Ismaili community has a parliamentary-type structure with each of these institutions headed by a person whose appointment is directly approved by the Imam. There are also regional and local councils and boards which may have committees working under their jurisdiction on specific projects. In many ways this structure resembles a government. Furthermore, the presidents of these national councils constitute the Leaders International Forum which directly advises Aga Khan IV. This internal organization of councils is complemented by four philanthropic Apex Institutions that serve all people regardless of race and religion. Ismailis have been noted and emulated for their massive and thorough bureaucratic organization, and it "both expresses and expedites the modernization of the sect while it also emphasizes and reinforces the singular role of the Imam" (Salvadori 1989: 232).

That there is an Ismaili Constitution, a document usually reserved for a nation, is a telling sign. There have been several local constitutions in East Africa dating back to 1905, with the latest published in 1986 as the first international constitution that applies to Ismailis worldwide. The Imam created and ordained the constitution, and has "full authority of governance over and in respect of all religious and Jamati matters of the Ismaili Muslims" (Aga Khan IV 1986: 5). As the only person who can change the constitution, the Imam is a monarch - His Highness Prince Aga Khan IV. He was given the title "His Highness" by Queen Elizabeth II and the title "Aga Khan" by the Shah of Iran. Aga Khan IV is often accorded head of state status, and is received by Presidents and Prime Ministers when he travels abroad. He has especially close relations with national leaders in Canada, United Kingdom, South/Central Asia and East Africa. Aga Khan IV holds a British passport but lives outside Chantilly, France, in a place called Aiglemont which the seat of the International Ismaili Secretariat that employs 150 people. As a monarch, Aga Khan IV has the authority to give out titles to people who have served the community, similar to the Knighthood which Queen Elizabeth II can bestow. In his grandfather's time, the highest title was Count, something borrowed from the Europeans. Now, in accord with other changes, the honorifics are in Arabic but retain a similar set of meanings. Ismailis also have an official flag, and the Imam has a seal and personal standard.

It would be easy to mistake this transnational structure for that of a nation, though it is likely part of the corporate structure common to many international religious organizations.20 For example, Catholicism also has a prominent leader who interprets the faith, national and international levels of organization, headquarters at the Vatican, and so on. On the other hand, Ismailism is distinctive in several respects. Many Catholics live in places or nations where they are the majority population or part of a dominant Christian majority. Even when they are a minority, they are globally recognized and understood. In contrast, Ismailis are unknown, even within the Islamic world. Largely because of their historical persecution, they are highly private, some would say secretive, and do not make themselves known. Ismaili history and doctrine has been on the periphery of Islam, and in East Africa and the West this relative marginality has been complicated by racial marginalization as well. Like Jews before the creation of Israel, Ismailis are a minority everywhere. They balance continued faith and community spirit, and also push for national integration in terms of physical appearance, language, law and philanthropy.

Most international religious communities have a leader that is chosen by his/her colleagues. The Pope, for example, is chosen by high-ranking cardinals. In Ismailism, the Imamate is passed through male kin, and what allows the Imam to maintain his status, is that he is continually recognized as a direct descendant of Prophet Muhammed. His role is also particular in that he interprets the Qu'ran for his followers. Unlike most faiths and sects of Islam, Ismailis do not rely on a holy book. Of course as Muslims, Ismailis are supposed to follow the Qu'ran. In reality, however, few of them (especially in East Africa) read this holy book, although its popularity is dramatically increasing as Islamization continues. The Ismaili du'a, ginans and other prayers are written down and taught to children, but as history shows, the du'a changes regularly at the discretion of the Imam. The interpretation of the faith is conveyed by the Imam to the jamat, both in official firmans and unofficially through changes or practices enforced through the Tariqah Boards, which oversee religious practice. Absolute control is exercised by Aga Khan IV, and this produces a dynamic and ever-changing set of practices which are not limited by an ancient text. For this reason, many of my informants spoke about the unique, progressive nature of Ismailism, precisely because of the fluidity of practice due to the Imam's power and ever-changing vision.

Transnational Ismailism emerges as a unique community which defies traditional notions of diaspora, nation and religion. Ismailis are a multiracial diaspora that is not limited by the trope of the tribe; a stateless nation without a desire for territory; an imagined community that is transnational; and a sect of Islam that does not rely on the Qu'ran and is headed by a monarch. These unique and seemingly contradictory traits describe a remarkably complex and coherent community, which has been crafted and maintained by the central figure of the Imam, who I turn to in the next section.

The Role of the Imam

I hope it is clear from the discussion above that the Ismaili case does not comfortably fit into any of the current models we use to talk about communities. Each of them - diaspora, nation, religion - have their limitations. But they are also all useful frameworks for thinking about different aspects of the transnational Ismaili community. I want to emphasize the importance of these multiple viewpoints in providing the most comprehensive and deft analysis, which will help to understand the structure and practice of the Ismaili community. My research in Kenya has emphasized one point which I would like to examine in more detail, and that is that Aga Khan IV is, in so many ways, at the center of the Ismaili community. Since the Imam's followers are widely scattered and have little political power where they reside, his power as a leader is immense. He advises his murids not only on religious practice, but also on secular issues including women's rights, education, immigration, finances and so on. The structure of Ismailism in which Qu'ranic scripture is not pivotal, religious leaders are appointed by the Imam, and festivals are all celebrated communally (as opposed to within the family), make Ismailis even more reliant on the leader, since they cannot independently access the religion without going to jamatkhana which itself is highly controlled by Aga Khan IV.21

As I discussed in Chapter Four, the majority of Islamic reforms that have been instituted, have been enacted by Aga Khan IV and his late grandfather. These include all religious and secular reforms including the increased use of Arabic and orientation towards dominant Islam; the push for Westernization, especially in clothing, language and migration; and the removal of icons, ginans, texts and practices which could be construed as un-Islamic. The Imam has also formed various charitable, architectural, and development organizations to strengthen ties with other Muslims, as well as with nations in East Africa and Central/South Asia where many Ismailis live. As sanctioned by the Constitution, the Imam has full control over the Ismaili community, and takes an active role in reviewing all aspects of ritual, practice, texts, appointments, and so on. He has created institutions which will perpetuate these changes, and appoints the people who run them. He has also placed significant emphasis on religious education, as a means to shape how young people think about Ismailism within the broader context of Islam, and in relation to other groups of people. Concentrating on youth is an effective way to change the beliefs of the community. When I look at the standard workbooks produced by the Institute for Ismaili Studies in London, which are used in all religious education worldwide, I am surprised by the strong emphasis on the Qu'ran, Allah, prophets and broader Islam which were not as central to the Ismailism I learned growing up. In this context, the Imams are established as legitimately Islamic by highlighting that they are direct descendants of Prophet Muhammed. All of these changes have been made on the advice of Aga Khan IV, since as many of my informants articulated, the Imam can see into the future and has a vision which must be followed.

In this structure, the Imam is given significant authority over the community and his position is accepted, even required. Aga Khan IV's reforms and advice have substantial ramifications for Ismailis, including where they live, what language they speak, how they dress, and what they believe. As I state above, Aga Khan IV is orienting Ismailis towards dominant notions of Islam, and bringing together the different cultures of Ismailis themselves. These links are not primarily forged through the interaction of people in social contexts, through marriage or in business relationships. These connections have to be actively imagined by Ismailis, on the suggestion and encouragement of Aga Khan IV. To this end, the Imam is responsible for "transnationalizing" the Ismaili community. He encourages changes in the prayer hall which make Ismailis more acceptable to other Muslims, has qasidas taught and sung by Khojahs in Africa and the West, sets up philanthropic organizations in the developing world to which his murids make contributions, sets up institutions that teach a standard Ismaili history at religious education classes worldwide, and oversees the Aga Khan University in Pakistan which is funded by Ismailis from around the world. All of these changes took place under the direction of Aga Khan IV and his staff in France. It is these actions, as well as the rhetoric of brotherhood and family described earlier in this chapter, which causes Ismailis to think of themselves as a single religious community, and also increases their links to other Muslims.

Economics is central to the transnational Ismaili community. Ismailis are supposed to give one eighth of their income in dasond to the Imam. The bulk of this money is used at the local and national levels to run the institutional machinery, though some of it is forwarded to the Imam for global community use. This compulsory giving is complemented by numerous donations for local events, poor Ismailis, and special projects. Many people give money to international development through the Aga Khan Foundation, which funds education and health care projects in East Africa and South/Central Asia, primarily to help non-Ismailis. Philanthropy is a significant component of Ismaili culture, and provides funds for assisting those less fortunate, both Ismailis and non-Ismailis. It is crucial to imagining the transnational Ismaili community, since Aga Khan projects in developing countries are financed by Ismailis from all over the world. People contribute to these charities because they are urged to by the Imam, but also because Ismailis feel a connection to the places and people that will benefit from their assistance. Giving money is a way for them to connect with their spiritual siblings in different parts of the world. Philanthropy is also important for Ismailis because it provides an overt way of showing allegiance to the hostland.22 Several of my informants pointed to the Aga Khan hospitals, clinics, and schools as evidence of Ismaili investment and loyalty to Kenya. One informant pointed out that the hospital catered almost exclusively to Africans.

Transnationalism is often considered to be the creation of social fields across national boundaries through social, economic and political relations (Basch et al. 1994: 22). These spheres are formed through the movement of objects or ideas, which are often tied to the migration of people. Transnational connections are usually based on contact between people, or come from economic or political motivations. In Ismailism, the Imam's work is to transnationalize - that is forge and uphold transnational connections and relationships within the Ismaili community. This is accomplished by establishing institutional ties, similar changes in religious practice, and firmans which continue to link Ismailis across borders. The irony here is that the Ismailis are not a nation, and thus do not have national boundaries - so they are transnationalizing across other peoples' borders. It is these connections, fostered by the Imam, that lead to flows of people, capital and ideas.

Of course immigrant Ismailis have family and personal relationships that exist outside the community structure and are unconnected to the Imam. However, these tend to last only a few generations, as is clear from the lack of engagement East African Ismailis have with South Asia. What keeps these communities connected is the structural relationships they have to each other fostered by Aga Khan IV. For example, when the Aga Khan University hospital was being built in Karachi, large donations were provided by wealthy Ismailis living in East Africa, North America and Europe. Also, many Khojahs have been going to Central Asia over the last few years to work with Ismailis on issues of development. As I explained before, qasidas from Central Asia have made their way to Khojah congregations. Aga Khan IV uses the metaphors of duty and family to keep these distinct communities in contact. His reforms, especially the religious ones, attempt to standardize religious practice. In this way, he acts to transnationalize Ismailis by forging structural and imagined connections between various components of the global Ismaili community.

To actively transnationalize the Ismaili community, Aga Khan IV acts as a facilitator in many respects. He creates institutions, promotes initiatives, and provides funding to increase interactions between his murids, and also form connections between Ismailis and other Muslims and non-Muslims. Furthermore, he acts as both a secular and religious leader. It is on his advice that many Ismailis migrate. As my informants explained, Aga Khan III encouraged his followers to come to East Africa from India in the late nineteenth and early twentieth centuries. After the violence in East Africa in the 1960s and 1970s, the present Aga Khan facilitated another migration. On a recent trip to Kenya in 1997, the Imam advised people to stay in East Africa for economic reasons, and he has also been encouraging East African Ismailis in the West to move back to or reinvest in Africa. Often the Aga Khan will not directly push for migration, but will suggest it in an indirect way by advising Council and community leaders to suggest it to the local community. The role of the Imam in promoting migration to East Africa, Australia, North America and Europe creates what could be termed a "facilitated diaspora" - one whose displacement and movement is fostered and directed by Aga Khan IV.

The Imam's influence is clear from the nature of the East African Khojah dispersal as distinct from the migration patterns of other East African Asians. As I discuss in Chapter Two, Ismailis were among the earliest immigrants to East Africa, and came primarily in intact family units that were intent on permanent settlement. In those early times, most Asian immigrants were single men who came as sojourners. I suspect that early Ismaili immigration and support by Aga Khan III, encouraged many non-Ismaili Asians to migrate to East Africa. Amidst Africanization policies after Independence, while most Asian groups had large migrations to India or England, Ismailis moved primarily to Canada. This was accomplished because Aga Khan IV was a friend of then Canadian Prime Minister Pierre Trudeau, who allowed several thousand Asian refugees and immigrants from East Africa to be quickly absorbed into Canada. My parents and others have explained how lucky they were to have a leader that brought them to Canada, which is less racist than the United Kingdom and more safe and socialist than the United States.

While my informants lacked a connection to the Gujarati homeland their ancestors left several generations ago, they have maintained a strong connection to the Imam, and through him to other Ismaili communities. This relationship is particularly strong because of the continued involvement, advice, and attention given by Aga Khan IV to his followers in all parts of the world. Furthermore, through transnationalization, the Imam has managed to link Ismailis not only through increasingly shared religious practice, but also through shared institutions, philanthropy, and capital flows. The imagined community that has resulted incorporates elements of a diaspora, stateless nation, and religion. While each of these terms has its limits, it is clear from this chapter how they are all useful frameworks for greater understanding of the transnational Ismaili community.