1. Long-term effects of governing religious commons

In some of her writings, Nobel Laureate Elinor Ostrom discussed the wider implications for the general prospects for democratic governance of her findings from how local communities could overcome Garrett Hardin's famous ‘tragedy of the commons’ problem. One example is her Presidential Address to the American Political Science Association in 1998, in which she warned that the concentration on top-down rationalistic models of government that did not put forward the importance of social trust and citizens’ participation might produce ‘generations of cynical citizens’ that would undermine the prospects of democratic governance also at the national level (Ostrom, Reference Ostrom1998: 18). Another example is an article titled ‘The Future of Democracy’ in which Ostrom underlined that sustainable democratic governance depended on the fact that people ‘over multiple generations’ had learnt the importance of participation in democratic institutions (Ostrom, Reference Ostrom2000b, cf. Ostrom, Reference Ostrom2000a).

In this article, we will present a case for the existence of such long-term and wider effects of citizens’ participation in local systems for the production and maintenance of public goods. Our case is based on a ‘puzzle’ provided by research on democratization, namely the lack of successful transformation to democracy in what is called the MENA region, which refers to the Arab countries in the Middle East and North Africa. As noted democratization scholar Larry Diamond put it just before the ‘Arab Spring’: ‘The continuing absence of even a single democratic regime in the Arab world is a striking anomaly – the principal exception to the globalization of democracy’. (Diamond, Reference Diamond2010: 93). The most recent question in this debate is if the ‘Arab Spring’ of 2011 is the beginning of a steadfast path to democratization. So far it seems as if the verdict is still out (cf. Fradkin, Reference Fradkin2013, cf. Chaney, Reference Chaney2012). As we will show, this lack of democratization cannot be explained by references to commonly used variables such as economic prosperity, religious beliefs or access to natural resources. As an alternative, we intend to provide an answer to this puzzle by focusing on the long historical traits of the organization of the local financing of religious practices through contrasting the Arab–Muslim region with Northwestern Europe, the region that historically has seen the strongest rooted tradition of representative democracy.

The argument is that in Protestantism, the local costs for religious practices have historically been organized in ways that are strikingly similar to what Elinor Ostrom found when she analyzed successful cases of how local groups had managed to organize the preservation of so-called ‘common-pool resources’ (Ostrom, Reference Ostrom1990). These local organizations had three features that are central to democratic governance, namely institutional arrangements for representation, accountability and transparency. In contrast, in the MENA region, the infrastructure needed for practicing religion has, for the most part, been managed by wealthy patrons in a type of institutional settings that featured no representation, no accountability and no transparency. We argue that the local institutions for organizing the infrastructure needed for practicing religion that we will describe below may have influenced the type of learning over generations about the prospects for democratic governance that is a central message from Ostrom's research.

2. Why has democracy had such a hard time in the MENA region?

A huge debate exists about the relation between Islam and democracy (cf. Chaney, Reference Chaney2012; Hanusch, Reference Hanusch2013; Jamal, Reference Jamal2007; Potrafke, Reference Potrafke2012; Tessler, Reference Tessler2002; Teorell Reference Teorell2010). One argument against Islam as essentially incompatible with democracy can be found by simply looking at ordinary Muslims’ political opinions. In recent years, a number of survey studies have noted an overwhelming support for democracy across the Muslim world, and this is true also for countries where the Arab–Muslim populations are a minority and also for Arab–Muslim countries were people live under autocratic rule (Braizat Reference Braizat2010; Fish, Reference Fish2011; Jamal and Tessler, Reference Jamal and Tessler2008; Teorell, Reference Teorell2010). Furthermore, these studies show that there is no difference in support for democracy between more and less religious individuals belonging to the Muslim culture (Chaney, Reference Chaney2012; Ciftci, Reference Ciftci2010; Jamal and Tessler, Reference Jamal and Tessler2008: 101; Tessler, Reference Tessler2002). For example, Haklai (Reference Haklai2009: 35) notes that the existing body of political culture research shows that

Regarding the actual democratic performance in the Muslim world as a whole, Ibrahim (Reference Ibrahim2007) reminds us that two-thirds of the planet's billion Muslims live under democratic rule, both in minority (most notably in India), majority (Indonesia, Bangladesh, Senegal as well as in Turkey), and in roughly even numbers with Christians (Nigeria and Bosnia-Herzegovina). Furthermore, as Stepan and Robertson (Reference Stepan and Robertson2003) note, the democratic track-record since at least the 1970s of the 31 non-Arab Muslim-majority countries are as good as, and in a number of cases exceed, what can be expected, considering their level of economic prosperity. This leaves us with the Arab countries in the MENA region. While Egypt and Tunisia may now be on their way to democracy, and 86% of the respondents in the states surveyed by the Arab Barometer (see Jamal and Tessler, Reference Jamal and Tessler2008) think democracy is the best form of government, there is still no Arab country in the MENA region that can be considered a stable democracy defined as when more than one election has resulted in a change of government.

In addition to culture and religion, the system for generating wealth forms a second popular explanation for the MENA region's democratic deficit. The so-called ‘oil curse’ is said to have damaged the prospects for democracy (Ross, Reference Ross2001; Ulfelder, Reference Ulfelder2007). The logic behind the argument can be divided into several factors, the most cited being that a government that relies mainly upon ‘windfall’ natural resources for revenue grows less dependent upon taxation and thus the need to create legitimacy through representative democracy (Levi, Reference Levi1988). However, recent research shows that even heavy reliance on natural recourses does not necessarily stand in the way of democratization. (Herb, Reference Herb2005; Oskarsson and Ottosen, Reference Oskarsson and Ottosen2010). Furthermore, several countries in this region such as Jordan and Morocco do not have access to huge resources of oil or any other similar natural resources that can explain their lack of successful democratization.

So the puzzle remains. Teorell (Reference Teorell2010), who has provided what we see as the hitherto most thorough empirical investigation for why some countries have democratized and others not since the 1970s, concludes that the MENA-region puzzle can neither be explained by economic factors, nor by access to oil, nor by an absence of popular mobilization, nor by lack of neighbor diffusion, nor by popular support for democratic values and, lastly, nor by weak links to democratic regional organizations. Teorell (Reference Teorell2010) adds that so far there is no explanation for the lack of democracy in the MENA region that stands up to an empirical scrutiny and ends his discussion by stating the following:

A third explanation has recently been put forward by Chaney (Reference Chaney2012) who argues that the lack of democratization in the MENA region can be explained by the area being conquered by Arab armies following the death of the Prophet Mohammad. This introduced an autocratic institutional equilibrium with rulers and religious leaders working in a close alliance that since then has been inimical to democracy. Chaney argues that the use of slaves in the Arab armies concentrated power in the hands of rulers and prevented the rise of a semi-independent class of landowning nobles as a basis for rulers to acquire military capacity. In Western Europe, the need for rulers to rely on the independent landed nobility contributed to democratization in the West by challenging central authority. While one of the most interesting approaches to the lack of democratization in the Arab countries, we take issue with this analysis on two accounts. First, to our knowledge, the class of landed nobility has generally not been in the forefront for introducing democracy in the West. The typical local feudal institutions in the West, for example the Junker type of rule in Germany, can hardly be seen as ‘schools’ for learning democratic governance. Second, Chaney argues that the system of slave-based armies in the Arab countries created an alliance between rulers and religious leaders that prevented the emergence of an independent civil society in which groups could challenge the central authority. The problem here is that this theory lacks an explanation for why the religious leaders in the Arab counties became so closely connected to the autocratic rulers and why a semi-independent religious sector did not emerge. Below, we will expand on Chaney's analysis by showing that the differences in how religion was financed historically have been substantial between the two regions under comparison. These differences are likely to have had a great impact on the divergence in the religious sector's independence from political authority and, by extension, in the fostering of democratic practices at the communal level.

3. Follow the money: a political–economic explanation

As Teorell (Reference Teorell2010) underlines, blaming specific religions for hindering democratization has proved to be a false road. As he points out, in the 1970s, Catholicism was thought to hinder democratization but since then southern Europe and large parts of Latin America have democratized. In the same manner, Confucianism was thought to be anathema to democracy but this has been disproved by the democratization of South Korea and Taiwan. It should be added that, on the other end of the spectrum, there is not much that speaks in favor of the argument that the inherent religious message in Protestantism has ‘created democracy’ either (Bruce, Reference Bruce2004).

Our central argument is that the lack of democracy in the MENA region should not be explained by Islam as a religion, but by how religion has been financed. A starting point is that religious practices usually come with a quite elaborate and expensive infrastructure. Churches and mosques have to be built and maintained, religious leaders must be hired and get salaries, religious schools and seminars have to be established, burial grounds have to be bought and kept in order, etc. From a strict technical point, these are not necessarily to be seen as public goods since individuals that have not contributed resources can be hindered from using them. However, both Christianity and Islam have strong norms against such exclusion practices, for example hindering poor people from entering churches and mosques or taking part in other religious services. It is more difficult to define if religious service also fulfills the other criteria for what should count as a pure public good, namely non-rivalry. One person attending a religious service does not subtract from another person attending the same service. However, when it comes to education or more personal services, this problem could of course occur. In any case, these religious services do present a collective action problem because ‘excluding non-contributors’ is not a trivial problem and may result in a serious lack of infrastructure needed for religious practice (Ostrom, Reference Ostrom2003: 241). The members of religious communities that fail to solve such a collective action problem do not, as a rule, risk financial ruin as an effect of depleted natural resources, as in the cases analyzed by Ostrom. Nevertheless, the destruction of a community's religious infrastructure brings the equally harrowing risk that its members cannot properly execute one's religious duties and, as such, spiritual ruin. Simply put, for devout Christians and Muslims, living in a local community without a common infrastructure for religious practices, they face a situation in which their prospect for eternal salvation looks grim. This implies that the collective action problem we analyze is a parallel to the ‘common-pool resources’ problem that was the basis of Ostrom's research.

In addition, in both Muslim and Christian societies, the local religious organizations (in the Christian West this has been the parishes, in the MENA region the waqfs, which are religious endowments) have historically been responsible for a whole host of other public services such as social assistance, religious as well as secular education and, in many cases, also important parts of the local infrastructure such as roads and system for water provision. As we will show, the system for financing these local activities, both the religious and the non-religious ones, has historically been very different when we compare the Arab–Muslim and the Northwestern Christian world.

Our central idea for explaining the lack of democratization in the MENA region is thus the one used in political economy, namely to ‘follow the money’, as initially suggested by Palmstierna (Reference Palmstierna2009). Following the approach known as ‘fiscal sociology’, we argue for the existence of a causal mechanism between type of taxation and type of government. According to this approach, we should expect taxation in general – and more direct forms of taxes in particular – to constitute a significant and important part of the social contract between the citizens and the state (Campbell, Reference Campbell1993; Levi, Reference Levi1988; Schumpeter, Reference Schumpeter and Swedberg1991). Scholars adopting this perspective have demonstrated that rulers who depend on direct taxation for their survival are much more likely to care about constructing legitimate forms of government than rulers that do not rely on direct taxation for their survival.

In attempting to account for how the Middle East lagged behind the West in terms of economic development during the last millennium, Kuran (Reference Kuran2011) employs a similar strategy. Rather than focusing on previously offered existing explanations, such as the production of knowledge and the features of state structure, Kuran places his analytical focus on the progress of the corporation, an entity that struggled to develop under the Ottoman legal system. In focusing on the advances of private organizations, our approach is related. The main difference is that we want to explain the lag in democratization. Where Kuran speaks of the importance of the corporation for economic prosperity, what attracts our interest is the democracy-generating significance of its public version, the local municipality.

In this, we build on research that shows that taxation cannot only be seen an administrative system for the authorities to collect revenue. Instead, with taxation usually follows two other things. First, for rulers to extract resources, installing systems for popular consent to ensure legitimacy is likely to increase the amount of revenue they can collect (Bräutigam et al., Reference Bräutigam, Fjeldstad and Moore2008; Levi, Reference Levi1988; Levi and Sacks, Reference Levi and Sacks2009). Second, taxation also increases demands for accountability and transparency in public affairs (Persson, Reference Persson2008). The reason behind this is that people who pay taxes will demand a system in which ‘the books can be checked’. Moreover, people will not be willing to part with their money if the tax administration (and the public administration more generally) is known to be corrupt, discriminatory or engages in other forms of favoritism (Rothstein, Reference Rothstein2011). This has been stated as follows:

In brief, our solution to the MENA-region puzzle is that when financing religion, the Arab–Muslim and the European–Christian parts of the world have historically relied on two very different systems. We want to draw special attention to the local level where religion (and its related activities) has been financed ‘from above’ in the Arab–Muslim world by a special type of foundation that has been insulated from influences from the majority of people in the community. Since people in the Arab–Muslim world as a general rule have not had to pay taxes for the religious services, and to a large extent, not also for other services handled by the local religious organizations, neither demand ‘from below’ nor a need ‘from above’ for representation, accountability and transparency have been established. The system for financing religious practices has been radically different in the North European–Christian world where these public goods have for the most part been financed ‘from below’ through the local congregations or parishes. As we will show, historically this has come with a surprisingly intricate system for semi- (or proto-) democratic representation, transparency and accountability that to a large extent resembles what Ostrom found when studying successful cases of ‘governing the commons’.

In accounting for the local perspective, we have been fortunate to be able to draw on some excellent, but highly specialized, historical research about the organization of parishes, communes and the systems of funding religion and welfare in these two different parts of the world. For Europe, we have concentrated specifically of studies England and from the 16th century and onward, and Sweden from the 17th century. As for their counterpart in the Arab–Muslim world, with a focus on the Ottoman Empire we have relied on similar sources. Analyzing this fascinating historical research, we have found two diametrically different ways of how to ‘finance religion’ which in its turn has led to two different ways of how to organize local political representation, accountability and transparency.

4. The MENA region: the local waqf-fare state

The dominant form in the Arab–Muslim world for financing local religious services and other local public goods has revolved around a specific and unique foundation-type institution known as the waqf (in Arabic [pl. awqaf]; in Turkish vakif). The term loosely translates to ‘forbidding movement, transport or exchange of something’ (Raissouni, Reference Raissouni2001). A waqf is a religious (widely defined) endowment, originally intended for charitable use. The historical origin of this institution is unclear, but already Mohammed encouraged the establishment of such an arrangement. Similar institutions are known to have existed in the Byzantine Empire, as well as ancient Persia (Makdisi, Reference Makdisi1981: 227). While the waqf institutions and its practice exist to this day, albeit on a much-diminished scale, the focus of this study will revolve around its historical significance and the legacy it has left behind. Indeed, the waqf system has existed, and does to this day, across the Muslim world, including contemporary democracies such as India and Indonesia. Still, the further away from the Arabian Peninsula we get, the historical legacy of the ties between the Muslim institution(s) and the formation of states grow weaker, due to the influence of competing powers, such as far Eastern and Western colonizers and traders; although the historical evidence on the matter is relatively scarce, there is some evidence pointing to the fact that out of the Muslim ‘Gunpowder Empires’, waqf was used much more extensively in the Ottoman and Safavid empires, the area we today (broadly) define as the Middle East, whereas the rulers of Mughal India relied more upon other, more direct and less institutionalized, ways of funding welfare projects (Kozlowski, Reference Kozlowski1995). Due to its uniquely long-standing rule and heavy institutional influence over the MENA region, the Ottoman Empire is a particularly suitable empirical focus for this study. In addition, it will occasionally draw examples from outside the region in order to further illustrate the micro-level effects of a waqf-funded public service system.

Due to its many uses, as well as its long history, the waqf institution is certainly a complex one, and only the core of its features and aspects can be dealt with here. What we can conclude is that it seems to have had an immensely important role in shaping the social, political and economic fabric of Middle Eastern life for centuries. Marshall Hodgson (Reference Hodgson1974a: 124) has characterized the waqf system as the ‘vehicle for financing Islam as a society’. As an example, at the fall of the Ottoman Empire in 1923, three quarters of all arable land in Turkey was waqf land (Kuran, Reference Kuran2001: 849). Awqaf were the major producers of many public goods, which could take the shape of mosques, soup kitchens [which in the 18th century fed 30,000 of a total of 700,000 citizens of Constantinople (Kuran, Reference Kuran2001: 850)], schools of higher learning (Medrasahs), roads, and water fountains, among other things (Yüksel, Reference Yüksel1997). It was thereby possible for a person to live in a veritable ‘waqf-fare state’, quite able to accompany an individual from the cradle to the grave. As one observer notes:

As Hodgson (Reference Hodgson1974a: 124) concludes: ‘through the waqfs, the various civic essentials and even amenities were provided for on a private yet dependable basis without need or fear of the intervention of political power’. As for the macroeconomic effects, Çizakça (Reference Çizakça2000: 8) notes that ‘economists looking at the waqf system would be perplexed by the fact that a myriad of essential services such as health and educational services have historically been provided at no cost whatsoever to the government’. We should add that these services were thus provided without the need for taxation. Çizakça continues to note that, for the issue of income distribution, ‘taxation [was] definitely assigned a secondary role’. Furthermore, waqf property itself was generally not subject to taxation (Darling, Reference Darling1996: 47). A waqf would generally consist of two parts: the charitable institution in question, and the revenue-raising side, i.e. investments, which would often be agricultural land or property, generating dues, and as time progressed, cash that could yield interest over time (Faroqhi, Reference Faroqhi1974; Inalcik, Reference Inalcik1969: 133).

Available historical studies show that the waqf system has not been the only means of funding of religious infrastructure and social welfare within Islam. The well-known institution of Zakat, regularized almsgiving of a certain proportion of one's wealth, is considered as one of Islam's five pillars. However, as Hodgson (Reference Hodgson1974b: 214) notes, after the initial expansion of Islam, the giving of Zakat was largely replaced by the waqf, as ‘zakât alms became an incidental personal obligation, on which few or no religious institutions were dependent’.

Founders of Awqaf generally tended to be, but were not exclusively, families of large wealth and influence, most notably members of the religious and military aristocracy (Barnes, Reference Barnes1986: 154). Motives, besides religious piety and a charitable urge, for founding a waqf were plentiful. For one, thanks to weaknesses in Islamic property rights and strictness of Shari'ah inheritance laws, a ‘family waqf’ was one of few means to protect a family's wealth (Kuran, Reference Kuran2001: 853). Another often-cited reason for establishing a waqf was to increase the family's social standing and prestige, which in turn could generate patronage power and political clout (Makdisi, Reference Makdisi1981: 40). By devoting a waqf deed (waqfiyyat) to the educational sector and religious schools (madrasahs), the founder could stipulate what was being thought in the schools: Just as important, the founder could accrue influence and patronage among the ulama, the important class of religious scholars in Muslim societies who were among the prime beneficiaries, and not seldom caretakers, of the waqf system (Makdisi, Reference Makdisi1981: 40; Sedgwick, Reference Sedgwick2003: 40; Zürcher, Reference Zürcher1994: 18).

Although sultans and members of the ruling elite tended to be amongst the prime patrons of waqf endowments, they did so as private persons, rather than as officials of the regime. Makdisi (Reference Makdisi1981: 281) describes how this religious–educational practice was institutionalized in Arab–Muslim countries:

Peri (Reference Peri1992) shows that this limitation of institutional authority was not confined to the educational system. In addition, through the waqf system, the rulers – be they Caliphs or local notables – could provide social welfare, religious services and other public services to their subjects or clients in order to strengthen their power and over the local community. Peri further adds that through this system, the rulers could increase their power ‘far beyond what they could achieve through the formal frameworks of power that were under their hand’ (Peri, Reference Peri1992: 174; cf. Shatzmiller, Reference Shatzmiller2001: 60). Moreover, the founder of a waqf was free to appoint the chief administrator, the mutawalli, who could be a member of the founder's own family, and often an ulama. This position has often been inherited, thus generating a dynastic form of influence over the institution. This made it difficult for beneficiaries, or stakeholders, and of course the local population, to weather complaints regarding maintenance of the waqf or garner insight into its operations. As Baer puts it:

Awqaf were thus not only top-down institutions, but under the power of one trustee who was appointed or inherited the position rather than being elected, highly informal in nature. As such, contrary to Western parishes and other local communes (see below), the waqf administration was not accountable to the people that used their services, which diminished the need for developing accounting rules. Even as mutawallis were conducting their work with the interest of the people in mind, the awqaf inherently lacked the institutional devices ‘from below’ to change and adapt to new conditions, as the endowments were subject to the ‘dead hand’ of the founder's stipulation, whether it be recent or centuries old. Notably, Kuran argues that, ‘[u]nlike western municipalities and other governmental agencies, which were authorized to tax constituents, change their own budgets and impose new ordinances, the waqf system could not make the necessary adaptations’ (Kuran, Reference Kuran2004: 78). Moreover, Yayla (Reference Yayla2007: 14) refers to the bookkeeping system of the waqf as nothing ‘more than an “accounting of conscience”’. In a study of the waqf system in Palestine after the British takeover, Reiter (Reference Reiter1996: xiii) notes how only those directly involved with the workings of the endowment, hence not including its beneficiaries, had insight into its accounts and ‘inner workings’. Although the modern-day waqf institution has changed in many ways, its management is still fraught with problems with ‘no transparency and lack of accountability’ (Ibrahim et al., Reference Ibrahim2006: 27).

During the modernization efforts of the latter 19th and throughout the 20th centuries, which brought forth greater state involvement, the waqf system suffered a decline in use, but it has for the last few decades shown signs of recovery. Centralization processes themselves underwent great difficulties due to the difficulties in establishing the worth of the myriad of awqaf, along with dishonest bureaucrats who themselves managed to embezzle significant portions of the revenue they were set to gather (Barnes, Reference Barnes1986: 155–156; Çizakça, Reference Çizakça1997: 14). Furthermore, the initial shaping of the municipal system occurred in tandem with an authoritative centralization process, which left the municipalities with little autonomous power, and ‘no personality of its own’ (Ortayli, 1978, as cited in Ersoy Reference Ersoy1992: 325). In the 1950s, Charles Issawi (Reference Issawi1956: 39) also noted the weakness of municipalities and cooperative associations in the MENA region, as the citizens now turned directly to the (authoritarian) central governments for social services.

There is certainly some variation in the MENA region that could speak against our argument. Especially, Turkey stands out as a recent democracy. The Ottoman reforms launched during the mid-19th century (Tanzimat) were designed to break with the old, patrimonial systems of authority. However, Anderson (1987, p. 5) argues that ‘[w]ell before these efforts were to bear all their fruit,. . ., Ottoman administrative development was suspended when the Europeans dismantled the empire after World War I’. In particular, Turkey, as the power center of the old Empire, managed to retain most of the existing Western type of bureaucratic achievements. The result in Turkey was not the development of (semi-) independent local government institutions but instead a system wherein the centralized bureaucratic state and military elite came to dominate society (Sosay, Reference Sosay2009; Sozen and Shaw, Reference Sozen and Shaw2002: 480).

Seen in a larger context, the waqf system illustrates a feature that, in different shapes, reproduced itself for at least a thousand years, wherein the state functionally separated its responsibilities from ‘softer’ societal areas, such as religion, education and welfare – a task left to the religious authorities, and funded by the waqf system. Without having to introduce a system of regularized taxation for these types of public goods, the waqf system took care of religious services, schooling and social welfare as well as handling mediation of local disputes (Lapidus, Reference Lapidus1996: 18). As stated by Kuran

As we interpret this, a wealthy family could protect much of their wealth over very long periods of time by creating a waqf. In exchange, the waqf would finance the religious services and also other social services in their local area. An effect of this system was that the local community using the services did not have to pay for them and thereby could not claim any influence in the form of representation or systems for accountability or transparency. To use the terminology from studies of state building, the waqf system led to a lasting legacy of personalism, patronage and lack of accountability. In fact, as Berman (Reference Berman2009: 225) states, the waqf system was effectively ‘crowding out the need to organize municipal provision that would have required tax collection’. In sum, this system of financing local public goods such as religious services solved collective action problems by making both the beneficiaries and the formal state agents ‘freeriders’ by distancing them both from responsibility and influence over its upkeep. A final example from outside the MENA region illustrates the dislodging of control, both from forces above, such as the state, and below, the citizens:

This failure to install the institutional framework of representation, accountability and transparency into the upkeep of the commons serves, as we will demonstrate in the following section, as a stark contrast to medieval and early-modern Northwestern Europe.

5. Northwestern Europe: communalism in the parishes

The central concept for understanding the European equivalent to the waqf-fare state is communalism which has been developed in a number of historical works by Peter Blickle who has investigated the role of local pre-democratic structures in Western Europe from the 15th to the 17th centuries (Aronsson, Reference Aronsson1992a: 25–28 and Reference Aronsson1992b; Blickle, Reference Blickle1989, Reference Blickle1997, Reference Blickle2000). Largely, communalism is defined as the capacity of a town or village to take charge of and promote its interests against the feudal hierarchy. This capacity was based on the commune's power to manage its economic, religious and social affairs by establishing local rules including forms for handling transgression of such rules. Moreover, the enactment of this local regulatory authority did not, according to Blickle, depend on any lordly consent, which would have been antithetical to the commune's claim to assert its own interests. Communalism thus stands as an alternative to feudalism and absolutism that has dominated much historical research of politics and power in Europe during this period.

The most important political structure of these communal arrangements, which according to Blickle had the same basic form in towns and country villages, was the communal or parish assembly, usually an annual meeting with all the householders. This was a proto-democratic structure in which they had to deliberate and decide upon issues that affected the entire commune and, important in this analysis, decide about fees and taxes. Five features of this type of local governance are emphasized by Blickle. One is the idea of egalitarianism in the assembly represented by the practice of one vote per household irrespective of the economic standing. The second is the notion of the existence of ‘a common good’. The third is the idea to elect representatives for managing these ‘common goods’ and the fourth is that these representatives should be held accountable for how they handled economic resources and implemented the locally decided rules and regulations. The fifth principle is that for accountability to be possible, transparency in ‘public affairs’ was needed. Especially after the reformation, it was common that these local institutions came to have considerable political power, not least taking responsibility for the local religious services. As Kümin (Reference Kümin1996: 11) states regarding the situation in the 16th century: ‘Having achieved a strong political and legal position, many rural and urban communities in Central Europe began to supervise the moral life of their clergy, to administer parish funds,. . .political and religious power was anything but a monopoly of traditional feudal elites’. A detailed study of three local parishes in Sweden during the 16th and 17th centuries reveals that the commoners could exercise a considerable amount of self-governance, not least in ecclesiastical affairs. This study points at the connection between control of the economic resources ‘from below’ and power over the church practices:

It should be added that, in general, the parishes were not used as local organization for implementing policies or for the collection of taxes as an agent for aristocracy or the central state. Tax collection for the central state was in general carried out by separate administrative entities (Cederholm, Reference Cederholm2007: 346). Moreover, the elected churchwardens in England could be sworn in when the Bishops made their visitations, but ‘ecclesiastical officials had no authority to refuse or dismiss them. Their principal loyalty was to their community, not the Church’ (Kümin, Reference Kümin1996: 29).

What seems to have characterized the forms of decisions at the parish assembly meetings was a strong emphasis on reaching consensus and to view the goods that the parish had responsibility for as ‘common goods’. Both in the English and the Swedish cases as well as in the areas in Continental Europe covered by Blickle's studies, there seems to have been very little of ‘politics as interest struggle’ within the parishes and other communal local assemblies (Aronsson, Reference Aronsson1992a; Cederholm, Reference Cederholm2007: 546; Kümin Reference Kümin1996; see also Gustafsson, Reference Gustafsson1989). Instead, it seems as if the type of politics resembles what is described by Elinor Ostrom (Reference Ostrom1990) in her analyses of modern-day local communities trying to ‘govern the commons’. However, this local consensus for the common good type of politics did in many instances give rise to a vertical type of conflict between the local communal bodies vis-à-vis the central church as well as the feudal lords (Blickle, Reference Blickle2000; Cederholm, Reference Cederholm2007).

Aronsson (Reference Aronsson1992a) also underlines that during the 18th century, the area of responsibility for the local parishes in Sweden started to grow. A number of secular responsibilities came under the jurisdiction of local communal board such as care for the destitute, the establishment of public schools, local systems for credit and insurance and infrastructure. The same is true for the English parishes, although it seems to have come about earlier, namely that the parishes extended their responsibilities to things such as the ‘maintenance or roads and bridges, the relief of the poor and the preservation of public order’ (see also French, Reference French2001). From the beginning of the 17th century, managing poor relief became a heavy responsibility for the locally elected officeholders in the parishes, who had to collect special taxes for this purpose and, not least, distribute the relief and by this deciding who was eligible and if deemed eligible, how much they should be given according to (their interpretation of) the rules (Hindle, Reference Hindle2004: 296–298; cf. French, Reference French2001).

Regarding the positions as churchwarden and members of the parish council, in Sweden these positions circulated so that over a decade, almost every landholding peasant had been an elected representative (Aronsson, Reference Aronsson1992a: 309). In a similar manner, studies of the structure of the English parishes show that the central position as churchwarden in practice circulated among ‘ordinary people’ and that it was customary that no one could have the position for more than two years. Although the position could carry quite large economic and administrative responsibilities, churchwardens in England represented a surprisingly wide spectrum of the social strata such as bakers, blacksmiths, butchers, shopkeepers, tailors and carpenters (Kümin, Reference Kümin1996: 30). People from the gentry rarely appeared, but it should be noted that on some rare occasions women could be elected a warden.

Thus, the picture that emerges is a system for managing and financing local religious practices and later many secular practices that is a complete opposite of the waqf system. After the reformation in Northwestern Europe, it was local semi-democratic political structures with elected representatives that decided and collected the taxes, decided upon building and maintenance of the religious temples and other constructions and also had a considerable influence over religious practices. There is, however, one more feature of this system of ‘communalism’ in Northwestern Europe that needs to be pointed out. As stated above, present-day Arab–Muslim countries do lack not only democracy but also ‘quality of government’. In a comparative perspective, and especially given their current economic standard, they are to a higher degree than could be expected ridden by problems of corruption, clientelism and patronage. A very likely reason for this is that the economic transactions within the waqf system have lacked institutionalized forms for transparency and accountability. As we have described above, the people who use the religious or social services produced by the waqf system usually have not paid taxes to the waqf system and from which it follows that they have little reason to demand transparency and accountability. From our readings of the history of the parish system in Northwestern Europe, the situation has been the complete opposite. For reason of space, we will concentrate on the situation in the English parishes during the late medieval period. The first thing to note is that the elected churchwardens came to control substantial amounts of money and other resources. As Kümin states

There is no doubt that the economic resources that could be entrusted to a churchwarden were substantial. Contrary to the administrators of the waqf system, churchwardens usually kept a strict separation between their private money and the communal resources they were elected to administer. Several historians have pointed out how remarkably detailed and to the minute the accounts were kept, not least considering that this was before literacy was common among the English commoners (Kümin, Reference Kümin1996: 221; see also French, Reference French2001). This is a stark contrast to the ‘accounting of conscience’ in the waqf system mentioned above. Moreover, in the parishes, audit regulations were strict making the churchwardens accountable to those who had elected them. At the yearly parish assembly, all accounts and receipts were scrutinized, the ‘audit day was an opportunity for the parish assembly to exercise its sovereignty’ (Kümin, Reference Kümin1996: 30).