Thanks to the increasing availability of critically edited classical Islamic texts and to the studies that draw on a wider variety of such texts, we know more about the world-view of the ulama in the classical period than we did half a century ago when many of the now-authoritative accounts of classical Islam were conceived. Like the study of all living traditions, the study of Islam has been a battleground of competing visions and prejudices. Many studied Islamic tradition only to identify a series of deficiencies in comparison to modern developments in the West, while others drew on that heritage for authority and inspiration for their own projects and thinking, deploying it to make claims of the deep compatibility of Islam with democracy, human rights, constitutionalism, the modern state, modern economy, and so on. In both types of interpretation, the relationship between religion and politics has been crucial to the modern readings of classical Islam. However, neither the orientalist nor the Islamist have been sufficiently critical of the terms and judgements they offer for consideration.
In the spirit of this cautiously collated infantry of concepts, what follows is a reading of a remarkable work on the question of the ideal leadership of the Muslim community by a theologian and jurist of fifth/eleventh-century Baghdad. The treatise in question is Abū al-Ma‛ālī al-Juwaynī’s Ghiyāth al-umam fī iltiyāth al-ẓulam (Aid to nations shrouded in darkness). This treatise was presented, addressed and dedicated to the Saljuq vizier of legendary renown, Abū ‛Alī al-Ḥasan al-Ṭūsī Niẓām al-Mulk (d. 1092), Juwaynī’s chief patron, who appointed him at a relatively young age of thirty-odd years to the professorship of the Niẓāmiyya madrasa of Baghdad. While not groundbreaking, this treatise is quite significant. It is more helpful than the popularly referenced works of al-Māwardī (d. 1058) or al-Ghazālī (d. 1111) in shedding light on the scope and rationale of the classical ulama's political ideals, compromises and political imagination. Indeed, had H. A. R. Gibb focused on Juwaynī rather than Māwardī in his assessment of the (Ash‛arī-led) classical Sunnī political discourse, he might not have concluded that it is little more than a series of capitulations to brute reality.Footnote 1 Compared especially to his renowned disciple al-Ghazālī – who seems to have stolen most of the attention of posterity as well as modern scholarship – Juwaynī has received remarkably little attention. None of the standard English works on Islamic political thought, such as those by Watt, Lambton and Crone, affords him a notable mention, let alone detailed treatment. While some important work has been certainly done, and while some recent studies have begun to recognise him, no devoted monograph on his political thought exists.Footnote 2
In order to allow our reading of Juwaynī to shed light on the questions just raised, a bit more must be said about the categories through which medieval Islamic thought is typically received. Much recent scholarship in various disciplines contests the ease with which one might talk about categories such as ‘religion’ and ‘politics’ – pointing out how recent both concepts are, and hence how anachronistic they may be in describing Islamic history. These scholars have argued – indeed shown (I count myself among the converted) – that it is the characteristic activity of the modern secular state to define and confine normative traditions and practices that are now called ‘religion’.Footnote 3 The whole conceptual world that a historian of the non-West had learned to take for granted is, as a result, thrown into disarray; for both the modern state (the active agent) and religion (on the receiving end of definition, legislation, containment) are defined through and against the other.
In the enterprise of writing history informed by these concerns, language, or the task of its translation and interpretation, is both one's best friend and one's worst enemy. Consider the stakes involved in the translation of the following keywords: whether the Qur'anic notion of shūrā is translated as merely ‘consultation’ or ‘democracy’; dawla as ‘government’ or ‘state’ or merely as the ‘state of being in power’; umma as ‘(a modern) nation’ or merely a ‘community of belief’; sharī‛a as ‘law’ or simply a ‘way of life’; siyāsa as ‘politics’ or ‘punishment’; ḥukm as merely ‘authority’ or as ‘sovereignty’; and khalīfa as ‘vicegerent’ or even ‘God's deputy’ or merely ‘successor’. These are choices fraught with consequences. Correspondence between terms and concepts always has a complex history, difficult to detect in part because language – the very tool we must use to preserve, communicate and clarify concepts over time and across space – obscures its own transformations, ambiguities and inadequacies. In order, therefore, to help clarify and sharpen the terms on which the story of the reception, adaptation or rejection of modern political ideals by non-Western traditions may be more plausibly based, an intellectual historian must at least attempt to wrestle with – hoping occasionally to even vanquish – language, all the while relying on little else.
Of the many tricks language plays on us, the most intractable is that it hides how unsettled modern concepts themselves have been and continue to be. In writing the history of the non-West we must simultaneously entertain an historian's scepticism towards the very apparatus the history of the West has handed to us. Let us begin with a proposition that may seem obvious, that the meaning of ‘politics’ in a world without the modern state, its institutions and its attendant concepts such as the public sphere, society, nation, bureaucracy, sovereignty and democracy, to name a few, cannot be taken for granted. But this problem affects the history not only of pre-modern Islam but of the pre-modern West as well. Was it the polis. a territorially bounded community, or the umma, a belief-bound community, that Muslims authors have been concerned with? And if the latter, as we shall show, then might we not more accurately call this discourse ummatics, rather than politics? But then, without suggesting that pre-modern Islam and the pre-modern West have a similar relation to the genesis of the modern state, we note that the modern notion of the state and its politics was absent not only in Islam but also in the pre-modern West.Footnote 4 Orientalism has been blamed too long for simplifying Islam; one implication of my contention is that it was the West itself, rather than Islam, that the orientalists radically simplified. The absence of comparable notions of territory, authority and jurisdiction in our classical Islamic sources, often understood as merely an absence or underdevelopment, may point to differences with respect not only to the modern West but to the pre-modern West as well. These differences will not disappear; yet, once the right questions are asked, they may become comprehensible.
Modern views on the role of Islam in government continue to vary sharply, in part because the flux of ideas and ideologies in the West shapes the lenses through which academic scholars have studied and continue to study the non-West. For instance, moving away from those who insisted on the inseparability of Islam and politics, now limited to a narrowing circle of old-fashioned academics, the new trend seems to emphasise the opposite. Whatever the case may have been in the brief and negligible moment of early Islam, this trend contends, the following millennium of the middle period of Islamic history saw a separation of religion from politics, so much so that some have described this state of affairs as indeed ‘secularisation’ of sorts. This secularisation remained undetected until now because, as one scholar wrote, the ulama continued to pay lip service to the ideals of Islam.Footnote 5 But since we have little more than words – in fact, only those words that happen to have been preserved – to decide between lip service and coherent, serious, sincere devotion, the task, if possible at all, must begin with a careful listening to the ulama's words themselves.
One way to help conceptual translation (not to be confused with linguistic translation, for modern Arabic has coined perfectly good terms for modern institutions) of political ideas across the barrier of time and space is to pay attention to political metaphors. Observers of political life across cultures have long described the more abstract and unintuitive field of politics by analogising it to the more immediately available phenomena. Contemporary American political metaphors such as ‘the nanny state’ and ‘the night watchman state’ do precisely that. For millennia, political metaphors have likened rulers to the shepherd, the weaver, the father, the mind, the military commander, the caravan leader, the gods and the shadow of God, to name but a few. In his pioneering and insightful work The Language of Political Islam, Bernard Lewis has taken stock of the vocabulary employed by various Muslim writers to depict political relationships.Footnote 6 Quite brief and general, however, this work barely scratches the surface, covering all centuries and regions in one sweep and only cursorily attentive to subtle conceptual differences, conflicts and developments within Islam.
One seminal attempt that brings out the possibilities of insights gained through an investigation of political metaphors is that of Michel Foucault, who traces the genealogy of the modern nation state, with its unprecedentedly enormous reach and ambition to shape and govern its subjects, to a confluence of two traditions represented by two opposing metaphors of politics.Footnote 7 On the one hand, there was the pastoral tradition, which Foucault attributes to “the East”, evident in the Egyptian and Assyrian empires, and culminating in the Hebrews, who saw God as their shepherd. On the other, there was the image of the ruler as a weaver, as depicted in Plato's Statesman, which represented the Greco-Roman politician. The Hebraic shepherd God, who cared for his people directly, collectively as well as singly – that is, for the herd as well as every single sheep –represented the apogee of the pastoral ideal of rule. Hebraic prophets were shepherds of their people chosen by God, and so were their kings, but they are more often chastised for failing to uphold the ideal. But crucially, Foucault contends, the pastoral ideal was adapted by the church through the institutionalisation of pastoral care, of a kind and extent unprecedented in human history.Footnote 8 The modern state, in this view, is a secularised version of the Christian church.
How, in light of this, might we see the unfolding of Islamic political imagination? Instead of asking why Islam did not produce the modern state, as many modern thinkers ask, or whether Islam is compatible with it, this interesting genealogy of modern politics invites us to see the various strands both within the Western and Islamic traditions, strands that are in conflict but also, undoubtedly, those that are shared and intertwined.
The present study's goal is modest. It focuses on the first of the three sections in which al-Ghiyāthī (as Ghiyāth al-umam is commonly known) is divided by its author. This section presents the ideal of imāmate, the political leadership of the Muslim community. The second section discusses the rulings pertaining to the absence of an imām for Muslims, and the third section addresses the rulings of a situation when even the expert knowledge of the Sharī‛a ceases to exist. Whereas the last two sections have drawn some attention recently, the first is often glossed over which is lamentable. It helps elucidate the political ideal of the theological Sunnī current better than any other treatise of the period by virtue of its unique realism and concern to account for the prevailing disagreements. The author himself may have helped engender some of this bias. He declares more than once that his real purpose in writing this treatise unfolds in the last two sections, which is what makes his work new and unique, since no one has so systematically and imaginatively addressed such end-time scenarios. However, it is the first section that presents the ideal whose absence is discussed in the last two, and it is significant enough to occupy four-fifths of the entire treatise, leaving only one-fifth for the rest (the author's own comments need not detain us). What further warrants an exclusive focus on the first section is that it is written with a legal theorist's (uṣūlī) attention to evidence and disagreement that surpasses what we find in other comparable works of the period. Juwaynī takes the clarity and rigour of the Ash‛arī-Sunnī doctrine of imāmate to a new level.
The object of this study is to shed light on the metaphors and concepts he employs and offer critical reflection on their meanings and compatibility with modern political concepts, pointing out some of the ways in which these concepts have been translated and mistranslated.
The Ruler
Juwaynī describes those in power through various terms, whose exact nature and range he does not explicitly define but rather assumes, but which can be fairly precisely understood based on his usage. Those in power are referred to as al-wulāt wa'l-a'imma/wa dhawī al-amr min qādat al-umma, which vaguely translates as “those in charge, the leaders and those of authority among the leaders of the Umma” (15).Footnote 9 Given Juwaynī’s proclivity to write in redundant, rhymed prose, it is likely that only two, not three, types of ruler are meant here. The first is wālī (pl. wulāt), applied later in the treatise to governors or rulers of cities and provinces. The second is imām (pl. a'imma; literally “leader”), which is commonly identified as the Shī‛ī term for the supreme leader, but it is also the Sunnī theologians’ preferred term to refer to the holder of the highest office, although its equivalent, khalīfa (caliph), is more widely known in historical and literary works. The second half refers to dhū imra, a less commonly used form of ulū al-amr, encompassing both wālīs and imāms. It is the only explicit Qur'anic term used for those given authority in the Muslim community, the umma (Qur’ān 4:59; 4:83). Another term Juwaynī frequently uses is amīr (literally commanders, those in charge), which is also closely related to the Qur'anic term ulū al-amr but has a more specific usage, as we shall presently see.
Already in this attempt to translate the word for the “ruler”, we can appreciate the challenge of the task at hand. In order to get as close as possible to what the author meant to convey, we must avoid fixing terms and concepts that are not fixed in the author's mind (by assigning them to some standard terms) and, conversely, avoid dissolving fixed terms by translating them literally or unsystematically. For instance, it is misleading to posit a specific correspondence between a contemporary technical term with a highly charged or well-defined meaning (such as democracy) to a pre-modern term without a highly defined meaning (such as shūrā). To this end, I will distinguish two types of term: those with a fixed conceptual range within the larger discourse of which this treatise is a part, such as imām, sharī‛a, umma, bay‛a (pledge of allegiance) etc., on the one hand, and words employed to capture vague or new concepts that have no settled terms yet and hence are variable and substitutable. Some instances of these unfixed terms may be mentioned here. The substitutability of terms for wielders of power other than the imām, such as wālī, amīr, sulṭān, etc., suggest that they are not the object of main theoretical interest – that their authority is derived rather than independent. In the same vein, the terms for the “lands” or territories of Islam, al-khiṭṭa, al-ḥawza, dār al-Islām, diyār al-Islām, are variable, suggesting that the territory is secondary to the community of Islam that inhabits and possesses them. Similarly, the subjects are described by a number of terms, ra‛iyya, nās, al-khāssa wa'l-‛āmma, al-muslimīn, suggesting that they do not represent a formal body that plays any active political role. The only exception are the electors, ahl al-ḥall wa'l-‛aqd (“those who loose and bind”), who do play such a role, however limited, in Juwaynī’s particular interpretation of it. Similarly, terms for the basis of a real or potential imām's influence vary: frequently referred to as shawka (not only the imām but his electors are also expected to possess it because their selection of the imām must be obeyed; soldiers are said to be the shawka of Muslims, 123). It is referred to using equivalent terms such as munna (power, influence) and ‛udda (“resources, resourcefulness”). These, of course, indicate ‘organic’ sources of power rather than specific political institutions. Similarly, one of Juwaynī’s personal desiderata, not necessarily shared by other contributors to the caliphate discourse, is stability, which he identifies variously as al-istiqrār, al-ittisāq and al-intiẓām.
The centrepiece of Juwaynī’s vision and of Sunnī political discourse at large is the imām. While linguistically imām (leader) and amīr (commander) have comparable meanings, and in early Islam amīr al-mu'minīn (“commander of the believers”) is reported to be the title of the supreme leader of Islam, in Juwaynī’s treatise (and indeed in the genre of writings we now call the caliphate discourse in general), imām and amīr have distinct meanings. Consider, for instance, Juwaynī’s insistence that a ruler over merely one of the territories of Muslims “cannot be the imām, for the imām is the one whose rule extends over all Muslims”. The division of Muslims into various territories, while imaginable and even a reality, is unacceptable, he seems to say, and will not be theorised. Necessity, such as an enemy territory or large body of water separating the Muslim lands, might force a commander (amīr) to be appointed over each of the truncated Muslim territories, and his authority would be valid in accordance with the Sharī‛a, but such an authority cannot be called the imām, and “such an era is to be considered devoid of the imām” (177). If the obstacles disappear, the lands would rejoin and the amīrs would submit to the imām (176). It is agreed upon and indisputable, Juwaynī declares, that there be one imām to oversee all the lands of Islam (khiṭṭa al-Islām), and the very purpose of imāmate requires that there be only one (172–173).
So much for the object of the imām's authority. What is its extent and function? The imāmate is defined by Juwaynī as “the supreme authority and general claim to the obedience of the elite as well as the commoners in matters of religion as well as this world” (22). The definition sounds absolutist, and has indeed been read as such, until one considers the possibility that for a jurist like Juwaynī, not unlike modern jurists’ overtaxing of the notion of sovereignty of the state, a clear line of authority may be a device to ground his legal edifice. What kind of authority and obedience is it really?
Juwaynī takes it for granted in everything he writes, in addition to stating explicitly that “the imām, in his adherence to and upholding of the rulings of Islam, is merely one of the people, one who has been deputised to establish these rulings” (277). Indeed, “the matters of this world must run by the course set by the Sharī‛a, for it is the front-runner and leader (imām) in all matters” (85). At the risk of some anachronism, this can be described as a very strong version of ‘rule of law’ or, as many scholars term it, nomocracy.Footnote 10 The rulings that the imām must uphold are for the most part predetermined, and not at all of the imām's own making. In creed, the imām must promulgate the doctrines of the pious predecessors (salaf), whereas in jurisprudence, he must not interfere with the various jurisprudential schools of ulama, letting them do what they do best. Nor must the imām engage in instilling piety, honesty and rectitude in his individual subjects; that is neither feasible nor desirable. His job is to use his resources and authority to root out heretical doctrines, promulgate those of the pious predecessors (salaf), keep peace and order within and spread Islam outside the lands of Islam. Thus, the ideal imām is a glorified executioner of the Sharī‛a, which he does by implementing wise policy (siyāsa), guided by his impeccable knowledge of the Sharī‛a and his ability to harmonise existing opinions, defuse conflagration of disputes and judiciously distribute funds within his authority. Attention to these details should help us appreciate the texture of the tension in Bernard Lewis's recognition, on the one hand, that “Muslim law has never conceded absolute power to the sovereign, nor, with few exceptions, have Muslim sovereigns been able to exercise such power for any length of time”, qualified by his observation, on the other, that “the predominant view of the jurists is authoritarian”.Footnote 11 What does “authoritarian” even mean with such limitations on the authority of the imām? Are the modern democratic states, who own the law, including the powers to make, interpret, implement and suspend it, not incomparably more authoritarian? It seems that the critique implied in this description is really something else: it is not that the government's powers are enormous but that they are vested apparently in an individual rather than the various branches of a state. This brings us to our next issue, the absence of a term that stands for our modern concept of ‘state’.
In contrast with the fixed, key terms such as the ruler (imām), the community (umma) and the law to abide by (the Sharī‛a), there is simply no fixed term for ‘government’, let alone for the modern abstract notion of the ‘state’. Throughout the treatise, the word dawla is used rarely.Footnote 12 Terms employed to refer to the concept of political power, such as dawla, riyāsa, za‛āma, all refer to the authority or state of being in power rather than the modern notion of the ‘state’ that, in modern parlance, refers to the abstract totality of the territory, the population and their history, all represented by a sovereign power. The imām is an individual contracted by, or on behalf of, Muslims at large, whose officials derive their authority from their appointment by him. To render any reference to the lands or government of Islam as the ‘state’ is thus potentially distortive, especially if we accept the contentions of intellectual historians who trace the current use of the concept of state as an abstraction to the rise of the absolutist states in early modern Europe.Footnote 13
The Imām as the Shepherd and the Weaver
As scholars have long noted, the image of the political ruler as the shepherd of his flock is ubiquitous in Islam, and Juwaynī’s treatise is no exception (22). I only seek to complicate our understanding of how this metaphor actually operated.
To understand the metaphors and concepts appropriate to the ideal imām, let us consider Juwaynī’s discussion of the qualifications, responsibilities and conditions of deposition of the imām. It has been frequently noted that unlike Māwardī, Juwaynī is concerned with the power and independence of the imām as being a non-negotiable trait, and does not accept an imprisoned or overpowered imām who loses his loses his power and influence (shawka, ‛udda, munna) (112–113).Footnote 14 Some qualifications are naturally endowed and non-acquirable, such as being a physically fit free male, who is courageous and noble, and of Qarayshī lineage; others are acquirable, the most important of which, besides the obvious one of being a Muslim, are knowledge, piety and political wisdom (79–83). Save for lineage, all these attributes, Juwaynī observes, may be reduced to independence (istiqlāl). In knowledge, the candidate must be a mujtahid, the bar being set rather high here, but the rationale is what should interest us:
The argument for such a requirement is that most foundational matters of religion are attached to the imāms. For whatever falls within the purview of [lesser] authorities [wulāt] and those in charge, doubtless pertains to the imām. In addition to these rulings of the Law, he must also attend to additional virtuous acts such as the commanding of the good and the forbidding of wrong. Were the imām not to be independent in his knowledge of the Sharī‛a, he would need recourse to scholars in deciding particular matters, and that would confuse his opinion and contradict his independence (istiqlāl). . .Since the imāmate is the leadership of religion as well as this world, it is necessary for him to be independent in the matters of . . . religion, for the matters of this world must run by the course set by the Sharī‛a, for it is the front-runner and leader (imām) in all matters” (84–85).
None of this precludes the imām's seeking of advice from others. Piety too, no doubt, is similarly required. However, the skill Juwaynī most emphasises after knowledge may be termed political wisdom, the ability to bring to a halt conflagration of opinions in great matters, to discern hidden aspects of issues and bring together people who are different and who disagree in all manner of bewildering ways. Like all Sunnīs, he argues at length, against the Shī‛a, that the imām need not be infallible or sinless.
In general, there is an unmistakable sense of pragmatism in Juwaynī’s approach, a concern with the realities of compromise that attends all politics, that contrasts in particular with the Shī‛ī concern with the imām's perfection. Yet he is strictly opposed to giving the imām's discretion a free hand to punish or transcend the boundaries of formal jurisprudence, against the Mālikīs and other jurists (219). This suggests that he is not merely being expedient or capitulating to the secular realities of the time. If men were to quibble about the imām's piety, nothing would prevent them from unceasingly delving into all of his acts, he writes, and obedience to the imām would not meaningfully obtain even for an hour. Having established the Sunnī doctrine that the leader of Muslims need not be infallible, it is quite probable that he would fall into sin in all kinds of matters, “those pertaining to his person (yata‛allaqu bi khāṣṣatihi)” as well as concerning “the rights of the Muslims (ḥuqūq kāffat al-muslimīn)”. Thus it would be impossible for any imām's tenure to continue if one accepts the position that his sinfulness makes his deposition an obligation (103). The habit of criticism of government that for moderns comprises the highest ideal of democracy, and which some idealistic Muslims (such as the Khawārij and the Mu‛tazila) no doubt also demanded, would, in Juwaynī’s view, void the pastoral and disciplining functions of governance altogether.
This is not “secularisation”, as some have suggested. Yet there is a recurring sense of both elusiveness and inscrutability of piety and a strong bent toward what is practical, legal, observable, real. The “most beneficial” candidate is one who commands the obedience of the soldiery (al-murtaziqa) and others whose cooperation is necessary for the attainment of the good of the umma. This good includes the religious objectives of upholding the truth and the mission of Islam, the protection and shepherding of the umma, preventing heresy and violence. Nevertheless, this pragmatism also veils something else: the world is not so enchanted that the piety of the ruler would in itself ensure prosperity and piety of the ruled. There is little doubt, as evident in the preceding, that Juwaynī is thoroughly religious in his vision of authority. Yet the piety of the ruler has little supernatural effect on the substance of rule.Footnote 15 The earthly life of the community is split into more than one dimension, and superiority in one dimension (that of piety) does not guarantee superiority in all others. “If we assume that one candidate combines all conditions with exceeding piety and fear of God, and another more competent than the former and better informed in ways of governance and leadership (siyāsa wa riyāsa), even if not like the first one in piety, the more competent one should be put forth” (170). This multi-dimensionality, this complexity, is not best understood as secularity. Yet it resists a simplistic understanding of Islamic political theology.
Respecting the branches of religion, only such acts of worship as constitute a sign and symbol or ritual (shi‛ār) of Islam and require congregative action are within the purview of the imām (198). Remarkably, alerting to what is good and prohibiting evil, although involved in all aspects of the government and authority in Islam, is not the responsibility of the imām qua imām except when it necessitates the use of force. To make individual Muslims pious and honest is not feasible for the imām, nor would it be effective were he to attempt it. Teaching, preaching and training are the job of all Muslims at their respective levels: parents, preachers, scholars and elders – all must be allowed to do their job. The imām's job is “only what pertains to policy-making and enforcement of penalties (siyāsa)” (238–239). The responsibilities of the imām are not entirely comparable to the medieval church's institutionalisation of pastoral care; the imām must only ensure that the right creed is protected and the flock as a whole safe and guided but not to get involved in rearing and guiding each member of the flock.
It is useful here to recall Foucault's attempt to distinguish true politics from pastoral care, based on his reading of Plato:
Making war, giving good judgments in tribunals, as well as persuading assemblies with the art of rhetoric, are not exactly politics but the conditions of its practice. What then is political action in the strict sense, the essence of the political, the politician, or rather the politician's action? It will be to join together, as the weaver joins the warp and the weft. The politician will bind the elements together, the good elements formed by education; he will bind together the virtues in their different forms, which are distinct from and sometimes opposed to each other; he will weave and bind together different contrasting temperaments, such as, for example, spirited and moderate men; and he will weave them together thanks to the shuttle of a shared common opinion. So the royal art is not at all that of the shepherd, but the art of the weaver, which is an art that consists in bringing together these lives “in a community that rests on concord and friendship”.Footnote 16
Most Muslim jurists, Juwaynī included, repeated the well-worn adage that anarchy for a day is worse than sixty years of tyranny; the flock would surely perish without the shepherd. Or would it? This trope has been quite frequently extended beyond its scope, both by Muslims and by modern commentators. Juwaynī, for all his realism and pragmatic license for the sinfulness of the ruler, does not go nearly that far. “To leave men to themselves to clash with one another without a leader to gather them upon truth or falsehood is still better for them than to impose a leader who aids oppressors, supports tyrants, protects aggressors . . . when people are pushed to such a state, opinions become obscure and senses confounded. . . such must be stopped” (106). If it is possible to stop him, one must proceed to do so with great caution, “for none should rebel against [tyrannical] amīrs except one with wisdom and piety”. If feasible, another imām with proper qualifications should be appointed; if his imāmate is established and obedience to him obtained, he should remove the previous one, and if [the disqualified imām] refuses, he is to be treated as a rebel, the rulings for dealing with such a situation are already found in the jurists’ books on rebels (bughāt). When judging whether to rebel, one must weigh one's options, and if the benefits outweigh the potential sacrifice involved in rebellion, one must not shy away from it. In the end, if a tyrannical imām cannot be removed due to his continued hold over coercive power, one must deem this time devoid of an imām (108–109). The flock, after all, is better off without a shepherd who would collude with the wolves.
When it comes to deposing the imām, Juwaynī mentions formal criteria that obligate deposing the imām, and justifications for rebellion if the imām is not moved by persuasion; there is no concept of other peaceful means to depose the imām. Indeed, if the ruler opts to be impious and disregard the criteria formulated by the ulama, the keepers of God's law, what means could there be, other than rebellion? This, perhaps, is the great chasm between the pre-modern Islamic political tradition and the modern one, for the latter sees the ultimate motivation and restraint for humans to be situated here and now, in this world, in interests and desires that can be calculated, structured and checked and balanced against each other. To check the rulers through some contractual agreements with the ruled would have sounded to the pre-moderns an exercise in circular reasoning. How could the ruled ultimately hold the rulers accountable, without destroying the basis of that rule, as in a rebellion? It is God, ultimately, who holds kings or caliphs to account. This is not an indication of absolute power but a comment on the nature of power.
It is difficult to reconcile this description of the nature and role of governance in Islam, which is essentially consistent with all other Sunnī writers on the subject, with some modern claims, including by Islamists in the wake of the Arab uprisings of 2011 for instance, that government in Islam can be best described as dawla madaniyya (a “civil state”). Even if we accept the rendering of the term dawla loosely as “state”, the defining reference to a city or territory implied in the phrase contradicts nearly everything Juwaynī says about it, voicing Sunnī consensus. However, whether Islamic government can be called theocracy can be disputed. Bernard Lewis's conclusion still appears to be correct: there is no theocracy in Islam, as the word is typically used, for there is neither theological nor institutional basis for it. Yet, if by it is meant, literally, “the rule of God”, then “in the juristic conception of the Muslim state, God alone is the supreme sovereign, the ultimate, indeed the sole legitimate source of authority”. He then remarks, “God thus becomes the formal expression of supreme sovereignty, and is often named in much the same way, and in much the same contexts, as the city, the crown or the people in various Western polities”.Footnote 17 If Lewis's remarks are to be lamented, it is not for simplifying Islam, but for simplifying the (modern) West against which Islam is perpetually compared, and through comparison with which all meaning is produced.
Concluding Reflections
The two lines of inquiry followed in this study can now be brought together. The first asks, based on an examination of the ways in which the metaphors of politics are employed, whether and in what sense Juwaynī’s thought is political. The second asks whether it is political in the modern sense.
To answer the first question, we note that the limits on his powers and responsibilities suggest that, in many ways, Juwaynī’s imām, at least in the limited respects noted above, is more of a weaver. This answer is reinforced when we consider the purpose of political rule. The end-goal of divine laws, Juwaynī states, is to draw men nearer to God. God has created men upon a nature given to desire, and provided laws to call men to restrain it, differentiating between permissible and impermissible, establishing justice among people and connecting good deeds with reward in afterlife and bad ones with punishment. But admonition and promise are not sufficient for most people; thus God appointed political authorities (salāṭīn wa ulū al-amr) to protect the weak, strengthen the pious, facilitate means of piety and thwart means of unrighteousness (181). Note that this appointment by God is metaphorical, in the same Ash‛arī theological sense in which all human actions belong to God. In reality, the imāms are neither infallible nor inherently different from the rest of the flock. Politics is needed when men would not submit to divine law on their own. To recall Foucault's reading of Plato again: “Politics begins, therefore, precisely when this first age, during which the world turns in the right direction, comes to an end. [When] the deity withdraws, and difficult times begin . . . The gods have withdrawn and men are obliged to direct each other, that is to say, they need politics and politicians.”Footnote 18 In this respect, as in the limitations on the pastoral role of the imām, Juwaynī’s imām is a weaver, a politician. Or rather, he is a shepherd whose sheep realize that he is one of them, and who must therefore learn to weave.
To answer the second question, we begin by recalling that Juwaynī’s usage of the term imām and its conceptual content is clear, consistent and stable. Even though Muslims have used the word imām in numerous contexts, in a political sense the imām is only the figure that rules over all Muslims and, furthermore, satisfies the agreed-upon qualifications to do so. A territorially limited ruler, either a governor on behalf of the imām or in the absence of an overarching imām, may be called an amīr or a wālī but not an imām. The idea of complete sovereignty of such a territory, of the kind a modern state requires, is not merely absent or unthought but unthinkable, as it would contradict everything, root and branch, that Juwaynī considers to be the obligation of Muslims, established, he insists, definitively by consensus. It is not primarily defined by territorial boundaries that the imām rules but by the community that inhabits these lands. Power over a territory itself does not make one the imām; it must be a territory containing the community of Islam, the umma. The imām, therefore, rules primarily over a community, and only secondarily over its territories. Note that there may be Muslims outside these lands, and there may be non-Muslims within, but these are exceptional conditions that qualify, but do not invalidate, the overall mapping of the lands to the community of Islam. Unlike the post-Westphalian nation-state, the political community in the Islamic imagination that Juwaynī reflects is constituted by the umma, a community of belief, and only secondarily by a territory. The modern Western state, in contrast, begins with a given territory and seeks to forge a community – in Benedict Anderson's terms, an “imagined community” – out of its inhabitants.Footnote 19 The umma, note further, is not an imagined community created only by the state elite in the interest of the nation based on a real or mythical past (nor, incidentally, is it merely a face-to-face, local community as the word sometimes connotes in Western languages). Rather, it is a community, a political but not territorially defined community, created by a shared belief and way of life. Not only is the doctrine that Juwaynī defends as essential to Islam not modern, it cannot be modernised, it would seem, without losing its essence.
