Chapter 4: LAW AND JUSTICE
The sacred law of Islam, the Shari'a, occupies a central place in Muslim society, and its history runs parallel with the history of Islamic civilization. It has often been said that Islamic law represents the core and kernel of Islam itself and, certainly, religious law is incomparably more important in the religion of Islam than theology. As recently as 1959, the then rector of al-Azhar University, Shaykh Mahmud Shaltut, published a book entitled 'Islam, a faith and a law' (al-Islam, 'aqida wa-shari'a), and by far the greater part of its pages is devoted to an expose of the religious law of Islam, down to some technicalities, whereas the statement of the Islamic faith occupies less than one-tenth of the whole. It seems that in the eyes of this high Islamic dignitary the essential bond that unites the Muslims is not so much a common simple creed as a common way of life, a common ideal of society. The development of all religious sciences, and therefore of a considerable part of intellectual life in Islam, takes its rhythm from the development of religious law. Even in modern times, the main intellectual effort of the Muslims as Muslims is aimed not at proving the truth of Islamic dogma but at justifying the validity of Islamic law as they understand it. It will therefore be indicated for us to survey the development of Islamic law within the framework of Islamic society and civilization, tentative as this survey is bound to be. Islamic law itself is one of our most important sources for the investigation of Islamic society, and explaining Islamic law in terms of Islamic society risks using a circular argument. Besides, the scarcity of expert historical and sociological studies of Islamic law has more often been deplored than it has inspired efforts to fill this gap.Islamic law had its roots in pre-Islamic Arab society. This society and its law showed both profane and magical features. The law was magical in so far as the rules of investigation and evidence were dominated by sacral procedures, such as divination, oath, and curse; and it was profane in so far as even penal law was reduced to questions of compensation and payment. There are no indications that a sacred law existed among the pagan Arabs; this was an innovation of Islam. The magical element left only faint traces, but Islamic law preserved the profane character of a considerable portion of penal law. It also preserved the essential features of the law of personal status, family, and inheritance as it existed, no doubt with considerable variations of detail, both in the cities and among the bedouin of Arabia. All these subjects were dominated by the ancient Arabian tribal system, combined with a patriarchal structure of the family. Under this system, the individual lacked legal protection outside his tribe, the concept of criminal justice was absent and crimes were reduced to torts, and the tribal group was responsible for the acts of its members. This led to blood feuds, but blood feuds were not an institution of ancient Arab tribal law, they stood outside the law and came under the purview of the law only when they were mitigated by the payment of blood-money, and at this moment the profane character of ancient Arabian law asserted itself again. There was no organized political authority in pre-Islamic Arab society, and also no organized judicial system. However, if disputes arose concerning rights of property, succession, and torts other than homicide, they were not normally decided by self-help but, if negotiation between the parties was unsuccessful, by recourse to an arbitrator. Because one of the essential qualifications of an arbitrator was that he should possess supernatural powers, arbitrators were most frequently chosen from among soothsayers. The decision of the arbitrator was obviously not an enforceable judgment, but a statement of what the customary law was, or ought to be; the function of the arbitrator merged into that of a lawmaker, an authoritative expounder of the normative legal custom or sunna. Transposed into an Islamic context, this concept of sunna was to become one of the most important agents, if not the most important, in the formation of Islamic law, and the 'ulama', the authoritative expounders of the law, became not in theory but in fact the lawmakers of Islam.
Muhammad began his public activity in Mecca as a religious reformer, and in Medina he became the ruler and lawgiver of a new society on a religious basis, a society which was meant, and at once began, to replace and supersede Arabian tribal society. Already in Mecca, Muhammad had had occasion to protest against being regarded as merely another soothsayer by his pagan countrymen, and this brought about, in the early period of Medina, the rejection of arbitration as practised by the pagan Arabs. But when Muhammad was called upon to decide disputes in his own community, he continued to act as an arbitrator, and the Qur'an, in a roughly contemporaneous passage, prescribed the appointment of an arbitrator each from the families of husband and wife in the case of marital disputes. In a single verse only, which again is roughly contemporaneous with the preceding passage, the ancient Arab term for arbitration appears side by side with, and is in fact superseded by, a new Islamic one for a judicial decision: ' But no, by thy Lord, they will not (really) believe until they make thee an arbitrator of what is in dispute between them and and within themselves no dislike of that which thou decidest, and submit with (full) submission' (Sura 4. 65). Here the first verb refers to the arbitrating aspect of Muhammad's activity, and the second, ' to decide ', from which the Arabic term qadi is derived, emphasizes the authoritative character of his decision. This is the first indication of the emergence of a new, Islamic, concept of the administration of justice. Numerous passages in the Qur'an show that this ideal demand was slow to be fulfilled, but Muhammad's position as a prophet, backed in the later stages of his career in Medina by a considerable political and military power, gave him a much greater authority than could be claimed by an arbitrator; he became a 'Prophet-Lawgiver'. But he wielded his almost absolute power not within but without the existing legal system; his authority was not legal but, for the believers, religious, and, for the lukewarm, political. He was essentially a townsman, and the bitterest tirades in the Qur'an are directed against the bedouin.
Muhammad's legislation, too, was a complete innovation in the law of Arabia. Muh ammad, as a prophet, had little reason to change the existing customary law. His aim was not to establish a new legal order, but to teach men what to do in order to achieve their salvation. This is why Islamic law is a system of duties, of ritual, legal, and moral obligations, all of which are sanctioned by the authority of the same religious command. Thus the Qur'an commands to arbitrate with justice, to give true evidence, to fulfil one's contracts, and, especially, to return a trust or deposit to its owner. As regards the law of family, which is fairly exhaustively treated in the Qur'an, the main emphasis is laid on how one should act towards women and children, orphans and relatives, dependants and slaves. In the field of penal law, it is easy to understand that the Qur'an laid down sanctions for transgressions, but again they are essentially moral and only incidentally penal, so much so that the Qur'an prohibited wine-drinking but did not enact any penalty, and the penalty was determined only at a later stage of Ishmic law. The reasons for Qur'anic legislation on all these matters were, in the firs tplace, the desire to improve the position of women, of orphans and of the weak in general, to restrict the laxity of sexual morals and to strengthen the marriage tie, to restrict private vengeance and retaliation and to eliminate blood feuds altogether; the prohibition of gambling, of drinking wine and of taking interest are directly aimed at ancient Arabian standards of behaviour. The main political aim of the Prophet, the dissolution of the ancient bedouin tribal organization and the creation of an essentially urban community of believers in its stead, gave rise to new problems in family law, in the law of retaliation and in the law of war, and these had to be dealt with. The encouragement of polygamy by the Qur'an is a case in point. A similar need seems to have called for extensive modifications of the ancient law of inheritance, the broad outlines of which were, however, preserved; here, too, the underlying tendency of the Qur'anic legislation was to favour the underprivileged; it started with enunciating ethical principles which the testators ought to follow, and even in its final stage, when fixed shares in the inheritance were allotted to persons previously excluded from succession, the element of moral exhortation had not disappeared. This feature of Qur'anic legislation was preserved by Islamic law, and the purely legal attitude, which attaches legal consequences to relevant acts, is often superseded by the tendency to impose ethical standards on the believer.
Islamic law as we know it today cannot be said to have existed as yet in the time of Muhammad; it came gradually into existence during the first century of Islam. It was during this period that nascent Islamic society created its own legal institutions. The ancient Arab system of arbitration, and Arab customary law in general, continued under the first successors of Muhammad, the caliphs of Medina. In their function as supreme rulers and administrators, the early caliphs acted to a great extent as the lawgivers of the Islamic community; during the whole of this first century the administrative and legislative functions of the Islamic government cannot be separated. But the object of this administrative legislation was not to modify the existing customary law beyond what the Qur'an had done; it was to organize the newly conquered territories for the benefit of the Arabs, and to assure the viability of the enormously expanded Islamic state. The first caliphs did not, for instance, hesitate to repress severely any manifestation of disloyalty, and even to punish with flogging the authors of satirical poems directed against rival tribes, a recognized form of poetic expression which, however, might have threatened the internal security of the state. This particular decision did not become part of Islamic law, but other enactments of the caliphs of Medina gained official recognition, not as decislons of the caliphs, but because they could be subsumed under oneor the other of the official sources of Islamic law which later theory came to recognize. The introduction of stoning to death as a punishment for unchastity under certain conditions is one such enactment. In the theory of Islamic law, its authority derives from alleged commands of the Prophet; there also exists an alleged verse of the Qur'an to this effect which, however, does not form part of the official text and must be considered spurious. Traditions reporting alleged acts and sayings of the Prophet came into use as proof-texts in law not earlier than the end of the first century of Islam, and the spurious verse of the Qur'an represents an earlier effort to establish the validity of the penal enactment in question. That the need of this kind of validation was felt at all, shows how exceptional a phenomenon the legislation of Muhammad had been in the eyes of his contemporaries.
The political schisms which rent the Islamic community when it was still less than forty years old, led to the secession of the two dissident, and later 'heterodox', movements of the Kharijites and of the Shi'a, but they did not lead to significant new developments in Islamic law; the essentials of a system of religious law did not as yet exist and the political theory of the Shi'a, which more than anything else might have been expected to lead to the elaboration of quite a different system of law, was developed only later. In fact, those two groups took over Islamic law from the 'orthodox' or Sunni community as it was being developed there, making only such essentially superficial modifications as were required by their particular political and dogmatic tenets. In one respect, however, the exclusive, and therefore ' sectarian ', character of the two secessionist movements influenced not so much the positive contents as the emphasis and presentation of their doctrines of religious law; the law of the Shi'a is dominated by the concept of taqiyya, 'dissimulation' (a practice which, it is true, was forced upon them by the persecutions which they had to suffer), and by the distinction between esoteric and exoteric doctrines in some of their schools of thought; and that of the Kharijites is dominated by the complementary concepts of walaya, 'solidarity', and bara'a, 'exclusion', 'excommunication'.
At an early period, the ancient Arab idea of sunna, precedent or normative custom, reasserted itself in Islam. Whatever was customary was right and proper, whatever their forefathers had done deserved to be imitated, and in the idea of precedent or sunna the whole conservatism ofArabs found expression. This idea presented a formidable obstacle to every innovation, including Islam itself. But once Islam had prevailed, the old conservatism reasserted itself within the new community, and the idea of sunna became one of the central concepts of Islamic law.
Sunna in its Islamic context originally had a political rather than a legal connotation. The question whether the administrative acts of the First two caliphs, Abu Bakr and 'Umar, should be regarded as binding precedents, arose probably when a successor to 'Umar had to beappointed in 23/644, and the discontent with the policy of the third caliph, 'Uthman, which led to his assassination in 35/655, took the form of a charge that he, in his turn, had diverged from the policy of his predecessors and, implicitly, from the Qur'an. In this connexion, there arose the concept of the 'sunna of the Prophet', not yet identified with any set of positive rules, but providing a doctrinal link between the sunna of Abu Bakr and 'Umar' and the Qur'an. The earliest evidence for this use of the term 'sunna of the Prophet' dates from about 76/695, and we shall see later how it was introduced into the theory of Islamic law.
The thirty years of the caliphs of Medina later appeared, in the picture that the Muslims formed of their own history, as the golden age of Islam. This is far from having been the case. On the contrary, the period of the caliphs of Medina was rather in the nature of a turbulent interval between the first years of Islam under Muh ammad and the Arab kingdom of the Umayyads. Not even the rulings of the Qur'an were applied without restriction. It can be shown from the development of Islamic legal doctrines that any but the most perfunctory attention given to the Qur'anic norms, and any but the most elementary conclusions drawn from them, belong almost invariably to a secondary and therefore later stage. In several cases the early doctrine of Islamic law is in direct conflict with the clear and explicit wording of the Qur'an. Sura 5.6, for instance, says clearly: ' O you who believe, when you rise up for worship, wash your faces and your hands up to the elbows, and wipe over your heads and your feet up to the ankles'; the law nevertheless insists on washing the feet, and this is harmonized with the text by various means. Sura 2.282 endorsed the current practice of putting contracts, particularly those which provided for performance in the future, into writing, and this practice did in fact persist in Islam. Islamic law, however, emptied the Qur'anic command of all binding force, denied validity to written documents, and insisted on the evidence of eye-witnesses,in the Qur'anic passage play only a subsidiary part. It is, of course, true that many rules of Islamic law, particularly in the law of family and in the law of inheritance, not to mention worship and ritual, were, in the nature of things, based on the Qur'an and, we must assume, on the example of Muhammad from the very beginning. But even here we notice (as far as we are able to draw conclusions on this early period from the somewhat later doctrines of Islamic law) a regression, in so far as pagan and tribal Arab ideas and attitudes succeeded in overriding the intention, if not the wording, of the Qur'anic legislation. This went parallel to, and was indeed caused by, the exacerbation of tribal attitudes in the turbulence created by the Arab wars of conquest and their success. The Qur'an, in a particular situation, had encouraged polygamy, and this, from being an exception, now became one of the essential features of the Islamic law of marriage. It led to a definite deterioration in the position of married women in society, compared with that which they had enjoyed in pre-Islamic Arabia, and this was only emphasized by the fact that many perfectly respectable sexual relationships of pre-Islamic Arabia had been outlawed by Islam. As against tribal pride and exclusiveness, the Qur'an had emphasized the fraternity rather than the equality of all Muslims, nevertheless, social discrimination and Arab pride immediately reasserted themselves in Islam. Non-Arab converts to Islam, whatever their previous social standing, were regarded as second-class citizens (mawali) during the first hundred and fifty years of Islam, and all schools of law had to recognize degrees of social rank which did not amount to impediments to marriage but nevertheless, in certain cases, enabled the interested party to demand the dissolution of the marriage by the qadi. The Qur'an had taken concubinage for granted, but in the main passage concerning it (Sura 4.3) concubinage appears as a less expensive alternative to polygamy, a concept far removed from the practice of unlimited concubinage in addition to polygamy which prevailed as early as the first generation after Muhammad and was sanctioned by all schools of law. Also, the Qur'anic rules concerning repudiation, which had been aimed at safeguarding the interests of the wife, lost much of their value by the way in which they were applied in practice. Early Islamic practice, influenced no doubt by the insecurity which prevailed in the recently founded garrison-cities with their mixed population, extended the seclusion and the veiling of women far beyond what had been envisaged in the Qur'an, but in doing this it merely applied the clearly formulated intention of the Qur'an to new conditions. Taking these modifications into account, the pre-Islamic structure of the family survived into Islamic law.
During the greater part of the first/seventh century, Islamic law, in the technical meaning of the term, did not as yet exist. As had been the case in the time of Muhammad, law as such fell outside the sphere of religion; if no religious or moral objections were involved, the technical aspects of law were a matter of indifference to the Muslims. This accounts for the widespread adoption, or rather survival, of certain legal and administrative institutions and practices of the conquered territories, such as the treatment of the tolerated religions which was closely modelled on the treatment of the Jews in the Byzantine empire, methods of taxation, the institution ofemphyteusis, and so forth. The principle of the retention of pre-Islamic legal practices under Islam was sometimes openly acknowledged, e.g. by the historian al-Baladhuri (d. 279/892), but generally speaking fictitious Islamic precedents were later invented as a justification.
The acceptance of foreign legal concepts and maxims, extending to methods of reasoning and even to fundamental ideas of legal science, however, demands a more specific explanation. Here the intermediaries were the cultured converts to Islam. During the first two centuries of the Hijra, these converts belonged mainly to the higher social classes, they were the only ones to whom admission to Islamic society, even as second-class citizens, promised considerable advantages, and they were the people who (or whose fathers) had enjoyed a liberal education, that is to say, an education in Hellenistic rhetoric, which was the normal one in the countries of the Near East which the Arabs had conquered. This education invariably led to some acquaintance with the rudiments of law. The educated converts brought their familiar ideas with them into their new religion. In fact, the concepts and maxims in question were of that general kind which would be familiar not only to lawyers but to all educated persons. In this way, elements originating from Roman and Byzantine law, from the canon law of the Eastern Churches, from Talmudic and rabbinic law, and from Sasanian law, infiltrated into the nascent religious law of Islam during its period of incubation, to appear in the doctrines of the second/eighth century.
The rule of the caliphs of Medina was supplanted by that of the Umayyads in 4I/66I. The Umayyads and their governors were responsible for developing a number of the essential features of Islamic worship and ritual. Their main concern, it is true, was not with religion and religious law, but with political administration, and here they represented the centralizing and increasingly bureaucratic tendency of an orderly adrninistration as against bedouin individualism and the anarchy of the Arab way of life. Both Islamic religious ideals and Umayyad administration co-operated in creating a new framework for Arab Muslim society. In many respects Umayyad rule represents the consummation, after the turbulent interval of the caliphate of Medina, of tendencies which were inherent in the nature of the community of Muslims under Muhammad. It was the period of incubation of Islamic civilization and, within it, of the religious law of Islam.
The administration of the Umayyads concentrated on waging war against the Byzantines and other external enemies, on assuring the internal security of the state, and on collecting revenue from the subject populations and paying subventions in money or in kind to the Arab beneficiaries. We therefore find evidence of Umayyad regulations or administrative law mainly in the helds of the law of war and of fiscal law. All this covered essentially the same ground as the administrative legislation of the caliphs of Medina, but the social background was sensibly different. The Umayyads did not interfere with the working of retaliation as it had been regulated by the Qur'an, but they tried to prevent the recurrence of Arab tribal feuds and assumed the accountancy for payments of blood-money, which were effected in connexion with the payment of subventions. On the other hand, they supervised the application of the purely Islamic penalties, not always in strict conformity with the rules laid down in the Qur'an.
The Umayyads, or rather their governors, also took the important step of appointing Islamic judges or qadis. The offlce of qadi was created in and for the new Islamic society which came into being, under the new conditions resulting from the Arab conquest, in the urban centres of the Arab kingdom. For this new society, the arbitration of pre-Islamic Arabia and of the earliest period of Islam was no longer adequate, and the Arab arbitrator was superseded by the Islamic qadi. It was only natural for the qadi to take over the seat and wand of the arbitrator, but, in contrast with the latter, the qadi was a delegate of the governor. The governor, within the limits set for him by the caliph, had full authority over his province, administrative, legislative, and judicial, without any conscious distinction of functions; and he could, and in fact regularly did, delegate his judicial authority to his 'legal secretary', the qadi. The governor retained, however, the power of reserving for his own decision any lawsuit he wished, and, of course, of dismissing his qadi at will. The contemporary Christian author, John of Damascus, refers to these governors and their delegates, the qadis, as the lawgivers of Islam. By their decisions, the earliest Islamic qadis, did indeed lay the basic foundations of what was to become Islamic law. They gave judgment according to their own discretion or ' sound opinion' (ra'y), basing themselves on customary practice which in the nature of things incorporated administrative regulations, and taking the letter and the spirit of the Qur'anic regulations and other recognized Islamic religious norms into account as much as they thought fit. Whereas the legal subject-matter had not as yet been islamized to any great extent beyond the stage reached in the Qur'an, the office of qadi itself was an Islamic institution typical of the Umayyad period, in which care for elementary administrative efficiency and the tendency to islamize went hand in hand. The subsequent development of Islamic law, however, brought it about that the part played by the earliest qadi in creating it did not achieve recognition in the legal theory which finally prevailed.
A typical example of the way in which the activity of the early qadis influenced Islamic law is provided by the law of procedure. The Qur'an had not only endorsed the use of written documents as evidence; it had also provided for putting the witnesses on oath in certain circumstances (Sura 5.I06-8). Islamic law rejected the first, and neglected the second provision, and had it not been for the early qadis, the hard and fast rule that evidence by witnesses, who are not put on oath, has to be produced by the plaintiff, and if no such evidence is produced, the oath in denial has to be taken by the defendant, would have been applied to the letter. The early qadis, however, constantly tried to impose safeguards on the exclusive use of the testimony of witnesses as evidence, and this tendency has left more or less extensive traces in several schools of Islamic law.
The jurisdiction of the qadi extended to Muslims only; the non-Muslim subject populations retained their own traditional legal institutions, including the ecclesiastical and rabbinical tribunals, which in the last few centuries before the Arab conquest had to a great extent duplicated the judicial organization of the Byzantine state. This is the basis of the factual legal autonomy of the non-Muslims which was extensive in the Middle Ages, and has survived in part down to the present generation. The Byzantine magistrates themselves had left the lost provinces at the time of the conquest, but an office of local administration, the functions of which were partly judicial, was adopted by the Muslims: the office of the 'inspector of the market' or agoranome, of which the Arabic designation 'amil al-suq or sahib al-suq is a literal translation. In the last few centuries before the Muslim conquest this office had lost its originally high status, but had remained a popular institution among the settled populations of the Near East. Later, under the early 'Abbasids, it developed into the Islamic office of the mutasib. Similarly, the Muslims took over the office of the ' clerk of the court' from the Sasanian administration.
The work of the qadis became inevitably more and more specialized, and we may take it for granted that from the turn of the first/seventh century onwards appointments as a rule went to specialists, to persons sufficiently interested in the subject to have given it serious thought in their spare time. Their main concern, in the intellectual climate of the late Umayyad period, was naturally to know whether the customary law which they administered conformed to the Qur'anic and generally Islamic norms; in other words, the specialists would be found normally within that group of pious persons who were at the same time working out an Islamic way of life. Once more, the care for efficient administration of justice and the tendency to islamize went hand in hand. Their interest in religion caused them to survey, either individually or in discussion with like-minded friends, all fields of contemporary activities, including the field of law, from an Islamic angle, and to impregnate the sphere of law with religious and ethical ideas. Their reasoning, which in the nature of things expressed their own individual opinion (ra'y), represents the beginnings of Islamic jurisprudence. In doing this, they achieved on a much wider scale and in a vastly more detailed manner what Muhammad had tried to do for the early Islamic community of Medina. As a result, the popular and administrative practice of the late Umayyad period was transformed into the religious law of Islam. But the close personal connexion between the groups of pious persons andthe qadis notwithstanding, Islamic law did not grow out of the practice, it came into being as the expression of a religious ideal in opposition to it.
The pious specialists on the sacred law were heId in respect both by the public and the rulers, and they owed their authority to their singleminded concern with the ideal of a life according to the tenets of Islam. They stood outside the political structure of the Arab kingdom of the Umayyads, and their main function was to give cautelary advice on the correct way of acting to those of their co-religionists who asked for it; in other words, they were the first muftis in Islam. Islamic law has preserved much of this cautelary character over the centuries; it is dominant in the teaching of Malik in Medina in the second/eighth century, and it recurs in strength in the medieval hiyal, or 'legal devices'. The pious specialists often had occasion to criticize the acts and regulations of the government, just as they had to declare undesirable many popular practices, but they were not in political opposition to the Umayyads and to the established Islamic state; on the contrary, the whole of the Umayyad period was, at a certain distance, viewed as part of the ' good old time'; this idealizing of things past was the first manifestation in Islam of a tendency which, a few decades later, was to lead to one of the most thorough and most successful of literary fictions. The attitude of the pious specialists to the Umayyad government anticipates the attitude of the religious scholars of Islam to any Islamic government.
As the groups of pious specialists grew in numbers and in cohesion, they developed, in the first decades of the second/eighth century, into what may be called the ancient schools of law, a term which implies neither any dehnite organization, nor a strict uniformity of doctrine within each school, nor any formal teaching, nor even the existence of a body of law in the usual meaning of the term. Their members continued to be private individuals, singled out from the great mass of the Muslims by their special interest, the resultant reverence of the people, and the recognition as kindred spirits which they themselves accorded to one another. It can be said that the division of the Muslims into two classes, the elite and the vulgar, dates from the emergence of the ancient schools of law. The more important ancient schools of which we have knowledge are those of Kufa and of Basra in 'Iraq, of Medina and of Mecca in the Hijaz, and of Syria. The differences between them were caused, in the first place, by geographical factors, such as local variations in social conditions, customary law, and practice, but they were not based on any noticeable disagreement on principles and methods. On principle, the ancient schools were inclined to disturb the practice as little as possible; because of the nature of our documentation, this can be particularly clearly observed in the case of the Medinese and of the Syrians.
The doctrines of the several schools enable us to discern the contrast between the social realities in that ancient Arab land that was the Hijaz and the newly conquered territory of old civilization that was 'Iraq, as well as the various reactions of the ancient lawyers of Islam to them. The legal integration of the wife into the family of the husband had begun with the Qur'an, when the wife was guaranteed a share in the inheritance, and the ancient lawyers followed the same tendency by giving the right to inherit to certain female relatives who did not possess it originally. But the school of Kufa alone went so far as to extend the right to inherit, after the agnates, to a group roughly corresponding to the cognates. The school of Medina rejected this absolutely. On the other hand, the tendency expressed by the school of Kufa found its consummation only in the doctrine of the Twelver Shi'is who unite the agnates and the cognates in one single group. The Twelver Shi'is lay emphasis also on the narrowly defined family, consisting of father, mother and their children and grandchildren, against the broader concept of family, merging into the old Arabian tribal system, which forms the background of the Sunni law of inheritance. 'Iraq was indeed the intellectual centre of early Shi'ism, and Shi'i law (and, for that matter, Kharijite law) has occasionally preserved early 'Iraqi doctrines which were later abandoned by the orthodox. The legal position of the unmarried girl and of the wife within the family, and their legal capacity,were decidedly more favourable in 'Iraq than in the Hijaz. On the other hand, the marriage bond was more rigid there, in so far as in 'Iraq the wife was inadequately protected even against grave derelictions of duty by the husband, such as failure to provide maintenance, or grave maltreatment; the school of Medina gave her the possibility of suing for divorce in these two cases, a rule which continued, it seems, a practice of Arab customary law which allowed the abandoned or maltreated wife to recover her freedom. As regards the status of the slave, the doctrines of the school of Medina show a certain paternalism which seems to derive from the social conditions in the cities of the Hijaz not less than from the civilizing influence of Islam. The Muslim slave, within the patriarchal family, enjoys a status similar to that of a free man; he may conclude a valid marriage by himself, without having secured the previous approval of his master (although the master may subsequently dissolve it); he may marry four wives just as a free man may (in contrast with the general rule which reduces all numbers given in the Qur'an by half for the slave); he has (notwithstanding certain restrictions) a real right of ownership; if he is authorized to trade, his transactions engage only his stock in-trade and not his person so that he cannot be sold to pay off a debt; and if he is gravely maltreated he can demand his freedom. None of this is accepted by the school of Kufa; in addition, according to the latter, he cannot act as leader of the ritual prayer if it is performed in common, he is not entitled to the Qur'anic procedure of li'an if he suspects his wife of adultery, his blood-money must always be less than that of a free man, and the master is in no case obliged to acknowledge the paternity of children which his female slave has borne. (The rule that children born by a concubine to her master and acknowledged by him as his own are free and in all respects equal to his children by a marriage with a free wife, goes beyond pre-Islamic practice, and is not explicitly laid down in the Qur'an; it must have asserted itself early in the first century, and it became of great importance in the development of Islamic society.) This hardening shows, no doubt, a more rigid and more differentiated society in which the social classes were more firmly separated, the result, in short, of a certain evolution. On the other hand, the school of Kufa was more ready to set free certain categories of slaves, to reduce the rigours of penal law for the slave, and to protect his life by making a free man who had murdered him, liable to retaliation. The 'adqila, the group of persons called upon to pay the blood-money in a case of unintentional killing or wounding, consisted originally, and still consists according to the doctrine of the school of Medina, of the agnates. According to the doctrine of the school of Kufa, however, it consists of those whose names, as members of the Muslim army, are inscribed in the same army list or payroll, alternatively of the members of the same tribe, or alternatively of the fellow-workers in the same craft. This shows most clearly the result of profound social changes. The qasama, the collective oath in criminal procedure when the person of the murderer is unknown, is, in the doctrine of the school of Medina, an affirmative oath by the members of the tribe of the victim which is sufficient to make the accused liable to retaliation. The Umayyad caliphs tried to mitigate its effect. The doctrine of the Kufans, however, recognized only a contradictory oath, not by the members of a tribe but by the inhabitants of the locality in question. Concerning pre-emption, the school of Medina was satisfied with laying down the rule that the co-owner was entitled to it; this was normally sufficient to ensure that strangers did not infiltrate the property owned by members of the same family or clan. In 'Iraq, however, this formula was not found sufficient, and in order to preclude the intrusion of strangers, it was found necessary to extend the right of pre-emption to neighbours, that is to say, owners of adjoining plots even if they were not technically co-owners of the property in question, provided their respective plots were entered by a common gate from a lane or thoroughfare, a kind of settlement common in the new cities of Islam which nevertheless preserved the identity of tribal associations. This provides a vivid picture of the lay-out of building plots in 'Iraq in the second/eighth century. It was only later that the Hijazi and the 'Iraqi doctrines were crystallized into the propositions that the right of pre-emption belonged to any co-owner or to any neighbour, whoever he might be. It is also not by accident that the degrees of social rank which aimed at perpetuating the social superiority of the Arabs over the mawali were elaborated outside Arabia, in 'Iraq, and that the procedure of becoming a maula by contract was recognized by the school of 'Iraq, where it was of great practical importance, but ignored by that of Medina. The law of property and of obligations, too, as formulated by the schools of Kufa and of Medina respectively, shows society in 'Iraq more differentiated and more closely controlled by the state than in Medina.
Whereas the ancient schools of law reflected different social realities, their general attitude to popular practice and administrative regulations was essentially the same, and it was certainly not the case, as has often, and recently too, been asserted, that the school of Medina was more traditional in its outlook and the school of 'Iraq more given to individual reasoning. It is true that, apart from diflferences in social development which are reflected in the doctrine, the doctrines of the school of Medina represent, generally speaking, an earlier stage of development of legal thought. But this means merely that the doctrinal development of the school of Medina often lagged behind that of the school of Kufa. 'Iraq was the intellectual centre of the flrst theorizing and systematizing efforts which were to transform Umayyad popular and administrative practice into Islamic law, and the ascendancy of 'Iraq in the development of religious law and jurisprudence continued during the whole of the secondleighth century. This is in keeping with intellectual development generally during the period.
The ancient schools shared not only a common attitude to Umayyad practice and, of course, a considerable body of positive religious law but the essentials of a legal theory, the central concept of which was the 'living tradition of the school'. This idea dominated the development of Islamic law and jurisprudence during the whole of the second/eighth century. Retrospectively it appears as the sunna or 'well-established precedent', or 'practice' ('amal), or 'ancient practice' (amr qadim). This 'practice' partly reflected the actual custom of the local community of Muslims, but it also contained a theoretical or ideal element, so that it came to mean normative sunna, the usage as it ought to be. Already at this early stage, a divergence between theory and practice manifested itself. The ideal practice was found in the unanimous doctrine of the representative religious scholars of each centre. This consensus of the scholars, representing the common denominator of doctrine achieved in each generation, expresses the synchronous aspect of the living tradition of each school. It is significant that the real basis of the doctrine of each school is not the consensus of all Muslims (which also exists) but of the scholars; the function of the class of 'ulama' in Islamic society was firmly established in that early period.
The need of creating some kind of theoretical justification for what so far had been an instinctive reliance on the opinions of the majority, led, from the first decades of the second/eighth century onwards, to the living tradition being retrojected, and to its being ascribed to some of the great hgures of the past. This process, too, began in Kufa, where the stage of doctrine achieved in the time of Hammad b. Abi Sulayman (d. I20/738) was attributed to Ibrahim al-Nakha'i (d. 95-6/7I3-I5). The Medinese followed suit and retrojected their own teaching to a number of ancient authorities who had died about the turn of the century, some of whom later became known as the 'seven jurists of Medina'. At the same time as the doctrine of the school of Kufa was retrospectively attributed to Ibrahim al-Nakha'i, a similar body of doctrine was directly connected with the very beginnings of Islam in Kufa by being attributed to Ibn Mas'ud, a Companion of the Prophet who had come to live in that city, and Ibrahim al-Nakha'i became the main transmitter of that body of doctrine, too. In the same way, other Companions of the Prophet became the eponyms of the schools of Medina and of Mecca. One further step in the search for a solid theoretical foundation of the doctrine of the ancient schools was taken in 'Iraq, very early in the second/eighth century, when the term ' Sunna of the Prophet ' was transferred from its political and theological into a legal context, and identified with the sunna, the ideal practice of the local community and the corresponding doctrine of its scholars. This term, which was taken over by the school of Syria, expressed the axiom that the practice of the Muslims derived from the practice of the Prophet, but it did not as yet imply the existence of positive information in the form of ' Traditions ' (Hadith), that the Prophet by his words or acts had in fact originated or approved any particular practice. It was not long before these Traditions, too, came into existence, and the persons who put them into circulation were the Traditionists.
The ancient schools of law themselves represented, in one aspect, an Islamic opposition to popular andadministrativepracticeunderthe later Umayyads, and the opposition group which developed into the Traditionist movement emphasized this tendency. As long as a Companion of the Prophet had been the final authority for the doctrine of a school on a particular point, it was sufficient for a divergent doctrine to be put under the aegis of another Companion of equal or even higher authority, as happened in Kufa where all kinds of minority opinions were attributed to the Caliph 'Ali, who had made Kufa his capital. But after the general authority of the Prophet himself had been invoked by identifying the established doctrine with his sunna, a more specific reference to him was needed, and there appeared detailed statements or 'Traditions' which claimed to be the reports of ear- or eye-witnesses on the words or acts of the Prophet, handed down orally by an uninterrupted chain of trustworthy persons. Very soon the emphasis shifted from proposing certain opinions in opposition to the ancient schools to disseminating Traditions from the Prophet as such, and the movement of the Traditionists, which was to develop into a separate branch of Islamic religious learning, came into being. It was the main thesis of the Traditionists that formal Traditions from the Prophet superseded the living tradition of the school. The Traditionists existed in all great centres of Islam, where they formed groups in opposition to, but nevertheless in contact with, the local schools of law. Initially the ancient schools offered strong resistance to the disturbing element represented by the Traditions, but they had no real defence against their rising tide; they had to express their own doctrines in Traditions which allegedly went back to the Prophet, and to take increasing notice of the Traditions produced by their opponents. Finally the outlines and many details of Islamic law were cast into the form of Traditions from the Prophet. In this way, one of the greatest and most successful literary fictions came into being.
When the Umayyads were overthrown by the 'Abbasids in I32/750, Islamic law, though still in its formative stage, had acquired its essential features; the need of Arab Muslim society for a new legal system had been filled. The early 'Abbasids continued and reinforced the islamizing trend which had become more and more noticeable under the later Umayyads. For reasons of dynastic policy, and in order to differentiate themselves from their predecessors, the 'Abbasids posed as the protagonists of Islam, attracted specialists in religious law to their court, consultet them on problems within their competence, and set out to translate their doctrines into practice. But this effort was shortlived. The early specialists who had formulated their doctrine not on the basis of, but in a certain opposition to, Umayyad popular and administrative practice, had been ahead of realities, and now the early 'Abbasids and their religious advisers were unable to carry the whole of society with them. This double-sided effect of the 'Abbasid revolution shows itself clearly in the development of the office of qadi. The qadi was not any more the legal secretary of the governor; he was normally appointed by the caliph, and until relieved of his office, he must apply nothing but the sacred law, without interference from the government. But theoretically independent though they were, the qddis had to rely on the political authorities for the execution of their judgments, and being bound by the formal rules of the Islamic law of evidence, their inability to deal with criminal cases became apparent. (Under the Umayyads, they or the governors themselves had exercised whatever criminal justice came within their competence.) Therefore the administration of the greater part of criminal justice was taken over by the police, and it remained outside the sphere of practical application of Islamic law. The centralizing tendency of the early 'Abbasids also led, perhaps under the influence of a feature of Sasanian administration, to the creation of the office of chief qadi. It was originally an honoriflc title given to the qadi of the capital, but the chief qadi soon became one of the most important counsellors of the caliph, and the appointment and dismissal of the other qadis, under the authority of the caliph, became the main function of his office.
An institution which the early 'Abbasids, and perhaps already the later Umayyads, borrowed from the administrative tradition of the Sasanian kings was the 'investigation of complaints' concerning miscarriage or denial of justice, or other allegedly unlawful acts of the qadis, difficulties in securing the execution of judgments, wrongs committed by government officials or by powerful individuals, and similar matters. Very soon, formal courts of complaints were set up, and their jurisdiction became to a great extent concurrent with that of the qadis' tribunals. The very existence of these tribunals, which were established ostensibly in order to supplement the deficiencies of the jurisdiction of the qadis, shows that their administration of justice had largely broken down at an early period. since then, there has been a double administration of justice, one religious and the other secular, in practically the whole of the Islamic world.
At the same time, the office of the 'inspector of the market' was islamized. Its holder, in addition to his ancient functions, was now entrusted with discharging the collective obligations of enforcing Islamic morals, and he was given the Islamic title of muhtasib, it was now part of his duties to bring transgressors to justice and to impose summary punishments, which on occasion came to include the flogging of the drunk and the unchaste, and even the amputation of the hands of thieves caught in the act; but the eagerness of the rulers to enforce these provisions commonly made them overlook the fact that the procedure of the muhtasib did not always satisfy the strict demands of the law.
The caliph, too, was given a place in the religious law of Islam. He was endowed with the attributes of a religious scholar and lawyer, bound to the sacred law in the same way as qadis were bound to it, and given the same right to the exercise of personal opinion as was admitted by the schools of law. The caliph retained full judicial power, the qadis were merely his delegates, but he did not have the right to legislate; he could only make administrative regulations within the limits laid down by the sacred law, and the qadis were obliged to follow his instructions within those limits. This doctrine disregarded the fact that what was actually legislation on the part of the caliphs of Medina, and particularly of the Umayyads, had to a great extent entered the fabric of Islamic law. The later caliphs and other secular rulers often enacted new rules; but although this was in fact legislation, the rulers used to call it administration, and they maintained the fiction that their regulations served only to apply, to supplement, and to enforce the sacred law. This ambiguity pervaded the whole of Islamic administration during the Middle Ages and beyond. In practice, the rulers were generally content with making regulations on matters which had escaped the control of the qads~, such as police, taxation and criminal justice. The most important examples of this kind of secular law are the siyasa of the Mamluk sultans of Egypt which applied to the military ruling class, and, later, the qanun-nama of the Ottoman sultans. Only in the present generation has a secular, modernist legislation, directly aimed at modifying Islamic law in its traditional form, come into being; this became possible only through the reception of Western political ideas. But the postulate that law, as well as other human relationships, must be ruled by religion, has become an essential part of the outlook of the Muslim Arabs, including the modernists among them.
Notwithstanding all this, the office of qadt in the form which it essentialiy acquired under the early 'Abbasids, proved to be one of the most vigorous institutions evolved by Islamic society. Qadis were often made military commanders, and examples are particularly numerous in Muslim Spain and in the Maghrib in general. They often played important political parts, although it is not always possible to distinguish the purely personal element from the prestige inherent in the office. Particularly in the Ayyubid and in the Mamluk periods, they were appointed to various administrative offices. They even became heads of principalities and founders of small dynasties from the fifth/eleventh century onwards, when the central power had disintegrated; there are especially numerous examples in Muslim Spain in the time of the Party Kings and others occur in Syria, Anatolia and Central Asia. In the Ottoman system of provincial administration, the qadi was the main authority in the area of his jurisdiction, and elsewhere, as in medieval Persia, he became the main representative of what is called the religious institution. To some extent the qadi (and the other religious scholars, too) were the spokesmen of the people; they played an important part not only in preserving the balance of the state but also in maintaining Islamic civilization, and in times of disorder they constituted an element of stability. Nevertheless, as far as the essence of the qadi's office was concerned, a real independence of the judiciary, though recognized in theory, was hardly ever achieved in practice.
Very soon after the 'Abbasid revolution, Islamic Spain broke away and became, under a surviving member of the Umayyad family and his descendants, an independent amirate and later caliphate. It is therefore not surprising that Islamic law and justice as applied in Spain should have diverged in some respects (not very essential ones, it is true) from their counterparts in the East. Whereas the qadi was always in principle a single judge, it was taken for granted in Spain that he should sit 'in council' (shura). The 'Abbasid institution of the chief qadi took a long time to become acclimatized in Spain. Although the adoption of the Sasanian 'investigation of complaints' by Islamic law probably dated from the end of the Umayyad period in the East, it had no real parallel in Spain. The 'inspector of the market ' retained his ancient title in Spain for centuries, and the theory of his functions was somewhat different there from what it was in the East.
The first half of the second/eighth century was a period of particularly rapid development for Islamic law, and this is well shown by the memorandum which the secretary of state, Ibn al-Muqaffa', presented to the 'Abbasid caliphj al-Mansur, at some time during the last few years of his life (he was cruelly put to death in I39/756). Written by an intelligent and observant outsider, a Persian convert to Islam, it shows us aspects of the stage reached by Islamic law about I40/757-8 which we should not be able to deduce from more conventional sources. Ibn alMuqaffa' deplored the wide divergencies in the administration of justice which existed between the several great cities and even (a completely unexpected piece of information) between their several quarters, and between the main schools of law. He suggested therefore that the caliph should review the different doctrines, codify and enact his own decisions in the interest of uniformity, and make this code binding on the qadis. This code ought to be revised by successive caliphs. The caliph alone had the right to decide at his discretion; he could give binding orders on military and on civil administration, but he must be guided by Qur'an and sunna. This sunna, Ibn al-Muqaffa' realized, was based to a great extent on administrative regulations of the Umayyads. Therefore, he concluded the caliph was free to determine and codify the sunna as he thought fit. The plea of Ibn al-Muqaffa' for state control over law (and, incidentally, over religion, too) was in full accord with the tendencies prevailing at the very beginning of the 'Abbasid era. But this was merely a passing phase, and orthodox Islam refused to be drawn into too close a connexion with the state. The result was that Islamic law grew away from practice, but in the long run gained more in power over the minds than it lost in control over the bodies of the Muslims.
A little later, towards the end of the second/eighth century, al-Shaf'i made the essential thesis of the Traditionists prevail in Islamic law. For him, sunna was not the idealized practice as recognized by the representative scholars; it was identical with the contents of formal 'Traditions' going back to the Prophet, even if such a Tradition was transmitted by only one person in each generation (a fact which, of course, made it very suspect to the ancient schools). This new concept of sunna, the sunna of the Prophet embodied in formal Traditions from him, superseded the concept of the living tradition of the ancient schools. According to al-Shaf'i, even the Qur'an had to be interpreted in the light of these Traditions, and not vice versa. The consensus of the scholars, too, became irrelevant for him; he fell back on the thesis that the community of Muslims at large could never agree on an error, a thesis sufficiently vague for his purpose. All this left no room for the discretionary exercise of personal opinion, and human reasoning was restricted, in al-Shaf'i's thesis, to making correct inferences and drawing systematic conclusions from Traditions. In accepting the thesis of the Traditionists, al-Shafi'i cut himself off from the natural and continuous development of doctrine in the ancient schools, and adopted a principle which, in the long run, could only lead to inflexibility. Also, the positive solutions of problems which he proposed were often, sociologically speaking, less advanced than those advocated by the contemporary 'Iraqis and Medinese; his reasoning, dominated as it was by a retrospective point of view, could hardly be productive of progressive solutions. Al-Shafi'i's was a personal achievement, and his disciples and followers formed from the very beginning the 'personal' school (madhhab) of the Shafi'is. The schools of Kufa and Medina, too, had seen the formation of groups or circles within each school, and early in the third/ninth century the geographical character of the ancient schools gradually disappeared, and personal allegiance to a master and his doctrine became preponderant.
Whereas the Hanafis and the Malikis, who continued the ancient schools of Kufa and of Medina (their names are derived from Abu Hanifa and from Malik, respectively), did not change their positive legal doctrines appreciably from what they had been when al-Shafi'i appeared, they finally adopted in the course of the third/ninth century, together with the Shafi'is, a legal theory of Traditionist inspiration. This theory differed from al-Shafi'i's own thesis in one essential respect, in that it returned to the concept of the consensus of the scholars, which it considered infallible. It endorsed al-Shafi'i's identification of the sunna with the contents of Traditions from the Prophet, but the legal rules which were to be derived from the Traditions were to be determined by the consensus of the scholars, which left the representatives of each school free to determine them for themselves, by interpretation and so forth. The fact that the Shafi'i school itself had to accept this modification of the doctrine of its founder shows the hold which the idea of the consensus of the scholars, embodying the living tradition of the ancient schools, had gained over Islamic law, and, by implication, how strong the position of the class of specialists had become.
Islamic law reached its full development in early 'Abbasid times, and its institutions reflect the social and economic conditions of Islamicsociety in that period more than any other. The various social backgrounds of the doctrines of the Medinese and of the 'Iraqis have already been mentioned. A feature which may, perhaps, reflect conditions proper to the early 'Abbasid period is the detailed treatment of 'usurpation' of the property of another, neither theft nor robbery, but high-handed appropriation. The provisions of Islamic law aim at protecting the rightful owner as much as possible, but at the same time make the frequency of similar acts, and the inability of the qddi to deal with them, painfully clear. The waqf or mortmain, too, found its final regulation at that time. The roots of this institution are various. One, which left only faint traces in Islamic law, and in the Maliki school more than in the Hanafi, can be traced to certain kinds of annuity, to use a modern, roughly approximate term, in use among the ancient Arabs; another, still very important at the beginning of the third/ninth century, though later quite pushed into the background, consisted of contributions to the Holy War, the object of innumerable exhortations in the Qur'an; a third, particular to Egypt during the first few centuries of Islam, seems to derive from the example of the Byzantine piae causae; and a fourth, which expanded enormously, particularly in 'Iraq, in the first half of the third/ninth century, and which was, perhaps, most decisive in shaping the final doctrine of Islamic law concerning waqf, arose from the desire of the Muslim middle classes to exclude the daughters and, even more so, the descendants of daughters from the benefits of the Qur'anic law of succession; in other words, to strengthen the old Arab patriarchal family system, and also to provide for the ma,vdli in order to make them reliable dependants of the family of the founder; both aims being in conflict with the purpose of the Qur'anic legislation. The waqf, and this may be counted its fifth and last root, also enjoyed a degree of security unknown to any other form of tenure, and its use became popular as a guarantee against confiscation. So was another procedure known to Islamic law, the fictitious sale or talji'a. Two things are significant here: confiscation with its concomitant procedure of torture, which had become almost a fixed institution of the Islamic state at the end of the Umayyad and particularly at the beginning of the 'Abbasid period, was not taken into consideration at all by Islamic law; in other words, the pious specialists averted their eyes from procedures which they knew were wrong but which they felt they could not, without material damage to themselves, openly criticize. On the other hand, even the early 'Abbasid caliphs and their highhanded and powerful dignitaries were averse to interfering openly with transactions which on the face of it, were valid under the religious law of Islam. Ibn Qutayba (d. 276/889), Traditionist and man of letters, held that the injustice of rulers and the highhandedness of overweening persons, and even the insistence of a creditor on being paid, justified lies and perjury. At a slightly later period, the poet and philologist Ibn Durayd (d. 32I/933) composed a treatise on equivocal expressions for the benefit of people who were forced to take an oath against their will, so as to enable them 'to mean something different from what they appear to say, and to save them from the injustice of the oppressor'.
Another omission of Islamic law is more difficult to explain, that of practically all reference to wholesale trade. The activities of wholesale merchants covered the whole of the Islamic world and extended beyond it, and they have left permanent traces in the merchant law of the early Middle Ages. Islamic law treats in great detail of many commercial transactions, but they are, as a rule, envisaged exclusively as transactions of retail trade, and the background of wholesale trading can only be inferred from occasional remarks and from isolated chapters such as those on the contract of muddraba or qirad (commenda, which, incidentally, seems to have come to Western Europe from Islamic law). It is true that IsIamic law is in the hrst place concerned with laying down ethical rules for the behaviour of the individual in a society the composition of which is taken for granted, but it is equally true that the wholesale trader, by the nature of his activities, is exposed, from the point of view of Islamic law, to particular moral hazards, which that law might have been expected to point out and safeguard against with the same interest in details as it does with regard to those involved in a householder sending out a minor to buy a loaf of bread. Generally speaking, Islamic law pays particular attention to transactions involving the middle or the lower-middle class; for instance, it appears clearly from legal terminology that the economic reality underlying the contract of salam, the ordering of goods to be delivered later for a price paid in advance, was the financing of the business of a small trader or artisan by his customers. A saying attributed to the Caliph 'Umar is specificaily directed against the activities of the rich speculators, who buy up supplies of food, anticipating a rise in prices, but exempts the small importer, who carries merchandise 'on his back in summer and in winter'. Merchants are also forbidden to meet caravans outside the town and to buy up what they bring, and a sedentary ought not to act as a sales agent of a bedouin. On the other hand, the Hamasa of al-Buhturi (d. z84/897) contains numerous extracts from the poetic effusions of bedouin, who boasted of having cheated the merchants from whom they had bought.
We are particularly well-informed concerning relations between neighbours in Maliki law. As interpreted by this school, Islamic law shows itself more humane than juridical. It puts the interest of certain social groups first; but these groups are, as a rule, neither state nor province nor city; in the last resort it is the family which matters, and this concern is reinforced by an easy-going acceptance of the fait accompli. The society envisaged by Islamic law is mainly urban, just as medieval Islam was essentially an urban civilization, but Islamic law did not recognize the city as such, nor did it admit corporate bodies. The doctrine of Islamic law does not attach great importance to differences of social status between free, male Muslims except, to some extent, in the requirement of the bridegroom's rank being equal to that of the bride, and, more important, the disqualification of members of certain low trades as witnesses. The doctrines of the several schools differ in details, and have undergone certain changes in the course of time. In a society in which the most highly respected economic activity was not that of the producer but of the merchant, the moralists tried to enhance the functions of the farmer and of the artisan, without, however, quite succeeding. Trade in cloth is generally regarded as the most honourable of professions, and sometimes trade in spices is associated with it. The professions of money-changer and of grain merchant are generally discredited, the hrst because it risks transgressing the complicated rules devised against ' usury ', and the second because it leads to speculation on rising prices of food. The two 'low trades ' par excellence were those of cupper and of weaver, and the contempt in which they were held seems to go back, in each case, to pre-Islamic times.
The early 'Abbasid period saw the end of the formative stage of Islamic law, and by the beginning of the fourth/tenth century a point had been reached when the scholars of all schools felt that all essential questions had been thoroughly discussed and finally settled (albeit with a choice of answers provided by the several schools); hence a consensus gradually established itself to the effect that from that time onwards no one could be deemed to have the necessary qualifications for independent rasoning in religious law, and that all future activity would have to be confined to the explanation, application, and, at the most, interpretation of the doctrine as it had been laid down once and for all. This is the 'closing of the gate of ijtihad', of independent reasoning in Islamic law. It is only in the present century that the reopening of this gate has been seriously envisaged by a number of 'ulama' and by Islamic society at large. The doctrine of the 'closing of the gate of ijtihad' was not the cause but a symptom of a state of mind which had been induced by the fear of doctrinal disintegration, a fear which was not far-fetched at a time when orthodox Islam was threatened by the esoteric propaganda of the Batiniyya. When this propaganda had brought the Fatimid caliphs to power, first in Ifriqiya and then in Egypt, they too felt the need of a doctrine of religious law of their own, and their great lawyer, the Qadi Nu 'man, provided it for them. It was a learned production which drew largely on the doctrines of the existing orthodox schools of law, rather than the result of organic growth, and it confirms the absence of a genuine Shi'i (as opposed to the general Islamic) tradition of religious law. Whatever the theory might say on the closing of the gate of ijtihad, the activity of the later scholars was no less creative, within the limits set by the very nature of their work, than that of their predecessors. New sets of facts constantly arose in life, and they had to be mastered and moulded with the traditional tools provided by legal science. This activity was carried out by the muftis, specialists on religious law who were qualified to give authoritative opinions on points of doctrine. The earliest specialists on religious law had been essentially religious advisers, muftis, and the later muftis only continued their advisory and cautelary activity. Their function was essentially private, and although muftis could be, and often were, appointed officially, it did not add to their authority. The most important officially appointed mufti in later times was the Ottoman shaykh al-Islam. The doctrinal development of Islamic law owes much to the activity of the muftis, and their advices, or fatwas, show us the most urgent problems which arose from practice in certain places and at certain times. Their decisions, if found acceptable, were generally incorporated into the later handbooks, and, generally speaking, the accretion of new cases and decisions in the interval between two comparable works of Islamic law represents the outcome of the discussion in the meantime.
Whereas Islamic law had been adaptable and growing until the early 'Abbasid period, from then onwards it became increasingly rigid and set in its final mould. A doctrine which had to be derived exclusively from the Qur'an and, even more important, from a number of detailed Traditions from the Prophet, and became more and more hedged in by the ever growing area of the consensus of the scholars, and by the closing of the gate of ijtihad, was unable to keep pace with the changing demands of society and commerce. This essential rigidity of Islamic law helped it to maintain its stability over the centuries which saw the decay of the political institutions of Islam. From the early 'Abbasid period onwards, we notice an increasing gap between theory and practice. This discordance and mutual interference dominated the history of Islamic law during the whole of the period here under review. This does not mean that Islamic law is entirely utopian. Apart from worship, ritual, and other purely religious duties, where in the nature of things the sacred law was the only possible norm, its hold was strongest on the law of family, of inheritance, and of waqf; it was weakest, and in some respects even non-existent, on penal law, taxation, constitutional law and the law of war; the law of contracts and obligations stands in the middle. The law of family and inheritance has always been, in the conscience of the Muslims, more closely connected with religion than other legal matters because the greater part of Qur'anic legislation is concerned with it. But even here, practice has been strong enough to prevail over the spirit, and in certain cases over the letter, of strict religious law. The legal position of women with respect to marriage and inheritance was occasionally improved in practice, but more often it deteriorated by comparison with Islamic law. AIso, the institution of waqf was used to produce this last result, as has been mentioned above. It is not the most important and essential rules of religious law which are observed most faithfully but rather those which for some reason or other have become part of popular practice, and practice sometimes insists on refinements unknown to Islamic law. The institution of pre-emption inits extended, Hanafi form proved extremely popular among the Muslims who followed that school of law, and in India it became part of the matters sanctioned by religion, concerning which the continued validity of Islamic law for Muslims was guaranteed at the beginning of British rule in 1777; but the Shari'a itself does not attach great importance to it, and the more detailed handbooks describe ways by which it can be avoided. The field of contracts and obligations was ruledby a customary law which respected the main principles and institutions of the Shari'a but showed a greater flexibility and adaptability and supplemented it in many ways, and the same is true of the special rules concerning real estate, of which only a few rudiments exist in the Shari'a. The customary commercial law was brought into agreement with the theory of the Shar'ia by the hiyal of 'legal devices' which were often legal fictions, transactions by which the parties might achieve, through legal means, ends which were made desirable by the economic and social conditions of the time, but which could not be achieved directly with the means provided by the Shari'a. The earliest devices were merely simple evasions of irksome prohibitions by merchants and others, but very soon the specialists in religious law themselves started creating little masterpieces of elaborate juridical constructions and advising interested parties in their use.
Another important area of contact between theory and practice was provided by the continued use of written documents which became the subject of a voluminous and highly technical literature. Islamic jurisprudence ignores custom as an official source of law, however much customs of varied provenance had contributed to forming it. But the Maliki school in Morocco in the later Middle Ages, where it developed in relative isolation from the rest of the Islamic world, took considerable notice of conditions prevailing in fact, not by changing the ideal doctrine of the law in any respect, but by recognizing that actual conditions did not allow the strict theory to be translated into practice, and that it was better to control the practice as much as possible than to abandon it completely. It therefore upheld the principle that 'judicial practice ('amal) prevails over the best attested doctrine', and it allowed a number of institutions unknown to strict theory. This Moroccan Maliki 'amal is not customary law; it is an alternative doctrine valid as long as conditions make it necessary.
We must think of the relationship of theory and practice in Islamic law, not as a clear division of spheres, but as one of intraction and mutual interference. The assimilation of the non-Islamic elements by the Islamic core in the formative period, and the assimilation of the practice by the theory in the Middle Ages, are really stages of one and the same process. This process, seen from outside, appears as a modihcation of the positive contents of Islamic law; whereas, seen from the inside, it appears as an expansion, a conquest of new helds by the ever dominant influence of Islamic law and jurisprudence. The ideal theory showed a great assimilating power, the power of imposing its spiritual ascendancy, even when it could not control the material conditions. Thus an equilibrium established itself between legal theory and legal practice, an equilibrium delicate in fact but seemingly unshakable in a closed society. As long as the sacred law received formal recognition as a reli~ious ideal, it did not insist on being fully applied in practice. But it could not abandon its claim to exclusive theoretical validity, and acknowledge the existence of an autonomous secular law; its representatives, the 'ulama', were the only qualified interpreters of the religious conscience of the Muslims; and the idea that law must be ruled by religion has remained an essential assumption even of modern Muslims. The works of Islamic law, during the whole of the medieval period, properly interpreted in relation to~ their place and time, are one of the most important sources for the investigation of Islamic society. The hold which the religious law of Islam had gained over the minds of the Muslims by the fifth/eleventh century can be gauged from the writings of al-Ghazali (d. 505/IIII), who, whilst deploring the ascendancy of legalism which threatened to extinguish religious life, and firmly restricting the subject-matter of the law to matters of this world, nevertheless protested that this did not imply reducing it to a secular subject of knowledge, and was unable to envisage secular rules for what he had insisted were matters which had nothing to do with religion.
The general and normal conditions described in the preceding paragraphs were occasionally disturbed by violent religious reform movements, such as that of the Almoravids in north-west Africa and Spain in the fifth-sixth/eleventh-twelfth centuries, that of the Fulbe in West Africa in the nineteenth century, and that of the Wahhabis in Arabia in the nineteenth and again in the present century. All these movements made it their aim, in the states which they set up, to enforce Islamic law exclusively, to abolish the double system of administration of justice, and to outlaw administrative and customary law. The effects of these religious reform movements as a rule tended to wear off gradually, until a new equilibrium between theory and practice established itself. Of essentially the same kind, though sensibly different in their effects, were the efforts of established states (later than the early 'Abbasid period) to subject actual practice to the rule of the sacred law. The two most remarkable of these efforts were made in the Ottoman empire and in the Indian empire of the Mughals, whilst the Safavid empire in Persia provides an instructive parallel.
The Ottoman empire in the tenth/sixteenth century is characterized by strenuous efforts on the part of the sultans to translate Islamic law in its Hanafi form into actual practice; this was accompanied by the enactment of qanun-name which, though professing merely to supplement Islamic law, in fact superseded it. On the part of the representatives of law we find, naturally enough, uncompromising rejection of everything that went against the letter of religious law, but at the same time unquestioning acceptance of the directives of the sultans concerning its administration, and, on the part of the chief muftis, a distinctive eagerness to harmonize the rules of the Shari'a with the administrative practice of the Ottoman state. A parallel effort in the Mughal empire in the seventeenth century was part of the orthodox reaction against the emphemeral religious experiment of the emperor Akbar. In the Persia of the Safavids, the religious institution, including the scholars and qadis, was controlled by the sadr, who exercised control over it on behalf of the political institution, thereby reducing the importance of the qadis. The Safavids' supervision of the religious institution was more thorough than had been that of the preceding Sunni rulers, and by the second half of the eleventh/seventeenth century the subordination of the religious institution to the political was officially recognized. This whole development had already begun under the later Timurids.