A growingly body of scholarship has examined the impact of religious beliefs and institutions on economic development. Several influential studies have probed the impact of religion on human capital through a focus on the role of Protestantism in Europe, Catholicism in France, and Christian missionaries in former colonies. This expanding literature on religion and development has predominantly focused on the impact of religious beliefs, practices, and norms. Barring a few exceptions, scant attention has been paid to the possible impact of religious elites. This is surprising given the centrality of elites in explaining the persistence of weak institutions. The interests and incentives of elites can matter for development, since they can impede or facilitate prosperity-inducing reforms. For example, elites are shown to oppose technological change if they perceive it as a threat to their future political power. A famous example is that of Ottoman rulers opposing the establishment of printing press as it threatened to undermine their power.
Religious elites are also central to studying the relationship between “religion, politics, and economic processes” which, according to Sriya Iyer, remains a blind spot in the emerging literature on economics of religion. Religious elites play an important role in structuring power relationships in Muslim Societies. In his book, Islam Instrumentalized, Jean-Philippe Platteau has argued that rulers have tended to lean on Islam as a political instrument in periods when states are consolidating their power or when they are exposed to weakness. Similarly, in Rulers, Religion, and Riches, Jared Rubin draws on Middle Eastern history to argue that Muslim religious clerics served as relatively less costly “legitimizing agents” for rulers when compared with commercial and business classes whose interests became marginalized. The political dis-empowerment of economic elites laid the basis for the Middle East’s subsequent economic divergence from Western Europe. The relationship between religious clerics and the state is also central to Ahmet Kuru’s thesis on Islam, Authoritarianism, and Underdevelopment.
There are other more empirically grounded attempts at probing the deep institutional foundations of Islam in the public sphere. This includes work by Eric Chaney who demonstrated that Egypt’s central religious authority assumed greater political power in periods of economic shocks. In a recent paper, Samuel Bazzi and co-authors study the role of Indonesia’s land reforms in 1960 that exempted religious lands supporting Islamic charitable endowments (waqfs) from expropriation. They show that regions that faced a greater threat of land expropriation saw higher transfers of land by rural elites into waqfs and witnessed stronger electoral support for Islamic political parties several decades later. However, the primary focus of this literature is on two aspects of organized religion: religious clergy (Ulema) and Islamic political parties.
In a forthcoming paper, we study a third important yet neglected dimension: the impact of Islamic mystical orders (Sufism) on long-run development. Sufism has long been understood as a central force in the transmission of Islam across vast swathes of Africa, Asia and the Middle East. A large literature in Islamic studies and history has shown that the missionary activities of mystical Islamic brotherhoods not only extended the borders of Islam but left a deep imprint on local social, economic and political structures. It is surprising that the emerging political economy scholarship on religion has completely neglected this central aspect of Islamic civilization. We offer a first systematic empirical enquiry on the longue durée impact of this aspect of Islam in Pakistan, the second most populous Muslim country. Specifically, we study the impact of religious elites who derive their authority and power from distinguished Sufi shrines on contemporary literacy. We show that impact of these elites on literacy is strongly conditioned by history and expressed through an interplay with formal institutional change.
The Context: Holy Muslim Shrines in Punjab
Our empirical evidence is drawn from the country’s largest province, Punjab. Interspersed between Central Asia and the heartlands of India, Punjab is home to some of the oldest Islamic mystical orders of South Asia. Travelling from Central Asia and the Middle East, early Muslim saints settled down in different parts of Punjab and were credited with bringing entire tribes and communities into the fold of Islam. As historian David Gilmartin has shown, the tombs constructed to celebrate the sacred tradition of these saints became the “symbolic cultural outposts” of Islam. Guardians of these shrines acquired immense legitimacy and prominence in the eyes of local populace and became important controllers of economic and political power over time. Shrine caretakers have historically acted as brokers between the centralized state and its peripheral subjects. Whether Mughal rule or Sikh interregnum, British colonial administration or post-independence Pakistan, no ruler has passed without extending favour to powerful shrine families to seek their support.
The power of shrine families is thus historically embedded. No ruler in history has passed without extending a favour to powerful shrine families in lieu of their support. In terms of their power and privilege, shrine guardians thus fit Peter Brown’s description of Christian saints as “patrons par excellence”. While courted by all rulers, the British colonial rule radically transformed their power through the introduction of formal property rights. By barring the sale of land from “agrarian” to “non-agrarian castes”, the Land Alienation Act of 1900 effectively preserved and expanded elite control over land. The British colonial administration treated prominent shrine families as “agrarian castes”, which made them fit to receive landed gentry grants.
Large tracts of land were awarded to prominent families in return for preserving social order and lending support to British rule. An illustrative example is the Kirmani Syeds of Shergarh, caretakers of the shrine of Daud Bandagi, who were recognized by colonial administration and rewarded through a sizeable land grant (1,168 acres). Colonial archives indicate this as a systematic and recurring pattern. The 1904 Gazetteer of the Bahawalpur State, for example, contains several records of landed estates (jagirs) being awarded to shrine leaders. While past rulers similarly granted land to key shrine families, these usually reverted back to the state upon the death of a shrine guardian. However, with the establishment of property rights under British colonial rule, hereditary succession assumed a powerful economic motive and converted these families into both spiritual and feudal masters.
Given their religious and economic power and their ability to act as intermediaries of rulers, shrine guardians were natural contenders in electoral politics when the British selectively opened avenues for it in the early twentieth century. Entering the electoral race from an early period, big shrine families were able to significantly expand their political footprint after Pakistan became an independent country in 1947. Affiliation with a notable shrine is a vital political asset in a milieu where, according to Anatol Lieven, “it is not wealth alone, but wealth plus either kinship or spiritual prestige, or both, that gives political power”. This paves the way for a shrine family’s entry into electoral politics. As historian K. K. Aziz argued: “A leading shrine is a gold mine, which catapults him into the aristocratic category and brings him riches large enough to enter into politics directly at the highest level” (Aziz 2001, p. 109).
No surprise, shrine elites have remained a permanent feature of electoral politics, surviving both periods of military and civilian rule. The above figure shows the political persistence of Syed Yusuf Raza Gilani, an important shrine guardian from south Punjab and former Prime Minister of Pakistan. Hereditary succession of religious authority facilitates the political persistence of shrine elites over time as it spreads religious power across several generations. Due to the dense and inter-locking ties of shrine brotherhood, shrine elites enjoy strong brokerage capacity both in and outside the parliament. These common interests are further solidified through marital ties. The below figure shows the nuptial ties of the Gilani family. The Gilanis are connected through marriage with at least six major shrine families affording them significant influence on the national political stage.
We explore whether shrine-dominated regions witnessed a substantially retarded trajectory of literacy rates over time relative to non-shrine regions. Our focus on literacy is motivated by both historical and contemporary literature that alludes to the aversion of shrine elites to education as it undermines the regime of voluntary obedience upon which the very religious authority of these elites depends. In Religion, Land, and Politics, noted historian K. K. Aziz argues that it is difficult for shrine elites to “countenance any prospect for the education of the masses” as it can unravel the hierarchical structure of power that is based on unquestioned loyalty, obedience, and superstition.
To construct a measure of the power of shrine elites, we compiled a novel dataset on the presence of historically significant shrines from colonial-era district gazetteers. These gazetteers provide a systematic indication of the differential importance of shrines and their guardians across various regions of Punjab (see the above figures). We also constructed a long panel of literacy spanning over a century, 1901-2011. To investigate the impact of shrine elites on literacy, we utilize a universal policy shock to the administration of public goods provision in 1985 that came in the wake of General Zia-ul-Haq’s military coup. The military coup was not driven by shrine elites. But to consolidate his power, General Zia gave elected politicians greater power over spending and allocation of public goods in their respective constituencies. With their greater aversion to mass education, shrine elites leveraged their political power to suppress literacy expansion in shrine-dominated regions after the coup-induced reforms. We show that the impact of shrine elites on literacy only becomes salient after the coup-induced policy shock in 1985 (see the figure below). The shrine effect is economically meaningful. We show that, without the shrine influence, average literacy rate would have been higher by 13 percent.
One empirical concern is that shrines might be situated in regions that were predisposed to lower literacy expansion in the long-run. Prior historical work by Richard Eaton shows that influential shrines were more likely to be established in regions considered peripheral to Delhi-based Mughal Sultanate (1206-1526). This is because frontier regions were typically home to pastoral nomads that were unattached to the Hindu cultural system. These regions also had a stronger agricultural potential either due to their proximity to the river or through other means of access to water. Our empirical analysis systematically accounts for such factors that could have a direct bearing on shrine location and simultaneously influence the evolution of literacy.
Another concern is that the shrine effect can be proxying for the impact of landed elites who might be similarly incentivized to oppose schooling. We empirically demonstrate that our results are robust to including for various measures of land inequality. While our analysis substantiates prior evidence on the opposition of landed elites to schooling, we emphasize religious, as opposed to economic, motives. We argue that, compared with landed elites, shrine elites have both a greater incentive as well as capacity to oppose schooling. Shrine elites have a greater ability to influence cognitive processes, moral perceptions, and beliefs. Deference to these norms creates a domain of voluntary obedience that helps shrine elites to command the undisputed loyalty of their followers and are able to control the social structure at a much deeper level.
This begs another question: What specific strategies could shrine elites have used in retarding education? They could influence the supply and/or demand for schooling. For example, they could restrain public spending, block the construction of schools, make educational provision defective, and/or influence the demand for schooling in their regions of influence. Given that General Zia’s military regime expanded the educational effort and gave control over development spending directly to elected politicians, some of the above strategies became more politically optimal for shrine elites than others. With more state resources now available for development projects (e.g. the construction of schools, hiring of teachers, etc.) and indirectly administered through politicians, Zia’s policies increased the opportunity cost for shrine elites to disengage from this new competition for state patronage. Thus, rather than restraining development spending or outrightly opposing the construction of schools, shrine elites found it politically more optimal to reduce the quality of educational provision. Our evidence is consistent with this explanation. We show that shrine-dense regions have schools that are more distant from target populations and lacked essential facilities, such as electricity and boundary walls. Furthermore, schooling resources were spread more thinly in high shrine regions, which is reflected in sub-optimally lower size of primary schools.
Our work underscores the importance of the relationship between Islam and the underlying distribution of economic and political power. Religion is one variable in a complex configuration of causes, and its effect can be highly non-linear and mediated by other variables in the surrounding institutional environment. As Avner Grief argues, the impact of a society’s informal institutions is conditioned by a historical process where political economy structures routinely interact with cultural factors to shape outcomes. Religion is thus part of a complex institutional matrix, a “system of social factors that conjointly generate a regularity of behaviour.” Grief’s approach to institutional analysis thus emphasized the role of both timing and context.
While modern political economy has highlighted the role of history for explaining contemporary development outcomes the challenge of persistence studies is to “figure out why and how history matters”. The impact of a historical variable can remain latent for a long time until it interacts with other factors to shape contemporary outcomes. Persistence, in other words, could be “time-varying”. Summarizing recent literature on the subject, Vicky Fouka notes that, in many recent studies, “the time-varying persistence of past events is engineered by political elites”. This is a hugely pertinent observation for the study of historic Islamic institutions, such as shrines, on long-run development. We have shown that impact of historically-embedded power of shrine elites on literacy remains latent until the military coup radically shakes up the administrative regime for public goods provision. The idea of historical contingency is thus central to our analysis.
In developing the external validity of our argument, it is important to stress that the impact of religious elites in different Muslim societies will depend on how they are structurally positioned within the prevailing power structure. Fouka’s observation that “the way history’s shadow manifests depends on timing and context” carries particular relevance for examining the relationship between Islam, politics, and development.