The first and most important thing to realize about Islamic law is that, seen in its own terms, it is the law of God not of man. No society since the death of Mohammed could enforce Shari’a because, absent divine inspiration, no human had complete and correct knowledge of its content. Strictly speaking, what traditional Islamic courts enforced was not Shari’a, God’s law, but fiqh, jurisprudence, the imperfect human attempt to deduce from religious sources what human law ought to be¹⁹⁶. That fact helps explain how Sunni Islam was able to maintain four different but mutually orthodox schools of law. There could be only one correct answer to what God wanted humans to do but there could be more than one reasonable guess. According to a widely accepted tradition, a Mujtahid, a legal scholar deducing the law from the Koran and the traditions of what Mohammed did and said, got two rewards in heaven if he got it right, one if he got it wrong.

That also explains the nature of the five-fold division of acts in Islamic law. An obligatory act is an act that God will reward you in the afterlife for performing, punish you for not performing. A recommended act is one which God will reward you for performing but will not punish you for not performing. A permissible act is one for which God will neither reward you nor punish you. An offensive act is one which you will be rewarded for refraining from but will not be punished for performing. An unlawful act is one which you will be punished for performing, rewarded for avoiding. From the standpoint of a believing Muslim, while no society’s legal system enforced Shari’a, Shari’a was enforced in all societies at all times and places—by God not by man.

In many cases, taking an unlawful act or refraining from an obligatory act might result in legal punishment as well as divine punishment, but that is not what defines the act as unlawful or obligatory¹⁹⁷. To put the point differently, Islamic law is more nearly a system of morality than a system of law, since its rules primarily describe how one ought to act, only secondarily the legal consequences of action. It is a system of law only from a standpoint that regards God, not the human legal system, as the ultimate judge and enforcer.

To better see the distinction between Shari’a and fiqh, consider the two different meanings of “constitutional” in our legal system. For a law professor teaching constitutional law to his students, what is constitutional is defined by what decisions the Supreme Court has made in the past or what decisions he expects it to make in the future. There is no inconsistency in holding that the New Deal farm program was unconstitutional in 1936, when the Supreme Court ruled against it, but constitutional after 1942, when the Supreme Court approved a revised version—a change due, arguably, not to changes in the legislation but in the position of the Court.

That is a natural view of what “constitutional” means from the standpoint of a law professor but not from the standpoint of a justice of the Supreme Court. The question for him is not how he and his colleagues will vote or have voted but how he should vote. In deciding whether something is constitutional he has to go back to whatever he sees as the sources of constitutional law, his equivalent of Koran and hadith, to engage in what a Muslim scholar would describe as ijtihad, deducing law. The law professor’s constitutional law corresponds to fiqh, the Justice’s to Shari’a.

How was fiqh deduced and applied? The scholar started with the sources of revealed knowledge—the Koran itself and the words and acts of Mohammed and his companions as reported in hadith, traditions. From that information a sufficiently learned religious scholar, a mujtahid, deduced legal rules. Over time, the scholars separated into four schools, each made up of multiple generations of scholars building on the work of their predecessors, each school (madhab) identified with the name of the scholar thought of as its founder. The schools were generally similar but differed in the details of their approaches to interpretation and the rules they deduced; each regarded the others as orthodox.

As scholarship accumulated, more and more questions could be answered by looking at the work of previous generations of scholars. Some modern historians argue that the result, sometime around the tenth century, was the end of the process by which law was deduced, the closing of the gates of ijtihad. In the view of others, while efforts may have gradually shifted from ijtihad to studying the work of previous generations of scholars, there was never a time when original research entirely disappeared, if only because new circumstances sometimes brought up new questions to be answered.¹⁹⁸

Deducing law involved a number of different problems. To begin with, it was not always clear what the Koran commanded. A verse might be taken either as one example of a general principle or as itself the legal principle to be followed. In some cases, such as the rules with regard to wine, later verses were inconsistent with earlier ones. To resolve such questions the scholars developed elaborate rules of interpretation, varying somewhat across the schools.¹⁹⁹

Deducing legal rules from hadith presented another set of problems. The traditions were at first transmitted orally, only later written down. Each came with its isnad, its pedigree, a list of the chain of transmitters starting with the person who first saw or heard what was done or said. It was necessary to decide for each hadith how certain one could be that it was neither invented by someone at some point down the purported chain nor inaccurate due to an error in transmission.

The basic rule accepted by all schools was that if there were a sufficient number of independent chains supporting the same hadith, it could be accepted as genuine with certainty. Short of that, the reliability of each hadith depended on the number of independent chains and how reliable the transmitters in each chain were believed to be. Part of what was necessary to be a mujtahid was extensive knowledge not only of hadith but of the information needed to evaluate them, including the pedigree of each and the reputation for reliability of its transmitters. Eventually several collections of authenticated hadith were produced by scholars who went through a much larger number, eliminating those they thought insufficiently well supported.

In addition to the Koran and hadith, there was one more source of information on divine law—consensus. According to multiple traditions, the Prophet had at some point said that his people would never be all agreed upon an error. While there were not a sufficient number of identical traditions to that effect, it was agreed by scholars of all four schools that there were enough different traditions with the same implication to meet the requirement, hence that it could be taken as certain that the Prophet had made some such statement. It followed that if at any one time all of the scholars were agreed upon a question, that question was permanently settled. Exactly what it meant for all the scholars to be agreed, whether it must be all qualified scholars of all schools or of one, whether it was necessary that all expressed positive agreement with a proposition or only that none expressed disagreement, were questions on which different scholars held different opinions. But all agreed that once consensus had been established it provided an additional source of authority.

The Separation of Law and State

Law, in theory, was not made by the ruler but deduced by legal scholars. In the view of at least some modern scholars, that was largely true in practice as well. After the first few centuries and until the rise of the Ottomans, political authority in the Islamic world was fragmented. The local rulers were frequently foreigners to the populations they ruled, often Turkish princes who had made the transition from mercenaries in service to Arab dynasties to de facto rulers. What they wanted from the legal scholars was support for their legitimacy. While they might occasionally meddle in some legal question of immediate relevance to themselves, they were willing for the most part to leave the legal system in the hands of the scholars. They were even willing to subsidize the scholars by endowing mosques and madrissahs, colleges which provided employment for legal scholars. Think of the resulting system as what Anglo-American common law would be if law professors ran the world, law defined not by the precedents set by judges but by the medieval equivalent of law review articles.²⁰⁰

Between the scholars who deduced the law and the courts that applied it there was an additional layer—the muftis. A mufti was a legal expert who offered advice on legal questions to any who wanted it. It appears to have originally been an unpaid position defined by reputation rather than appointment by the ruler, like the Roman jurisconsult²⁰¹. Someone who wanted an opinion on a legal issue could put the question to a local mufti and be given a fatwa, an advisory legal opinion. Initially the mufti was expected to himself be a mujtahid, a scholar qualified to derive rules from their original sources. But by about the thirteenth century it had become accepted for someone to issue fatwas who was familiar with the doctrine of a legal school but was not himself a mujtahid capable of deriving it.²⁰²

The fatwa might amount to moral advice, an opinion as to what action it was right to take. It might be legal advice in our sense, a statement of the legal implications of the situation described. It was not the mufti’s job to find out what had actually happened, only to report what would be the legal implications of the facts as described to him.

The final actor in the progress from divine revelation to a functioning court system was the qadi, the judge. Unlike everyone above him in the chain, he was appointed and paid by the ruler. While it was desirable that he be an expert in the law it was not essential, since he could rely for the law on fatwas presented to him by the litigants or provided at his request by a mufti in response to questions about the relevant law.

From the perspective of modern American law, the final two stages of the process look like our system turned upside down. In ours, the court of first impression applies the law to the facts and produces a verdict. If the case is appealed, the appeals court takes the facts as already decided and gives a second and authoritative opinion on the law. In their system, the opinion on the law came first, provided by the mufti, followed by the qadi’s application of the law to the facts as he saw them. Under most circumstances there was no way to appeal the qadi’s verdict.

An Alternative View

What I have so far described is the traditional account of how Islamic law, fiqh, was derived. A number of modern scholars, of whom the most influential was Joseph Schact²⁰³, have argued that that account is in large part fictional. He pointed out a large number of cases in which companions of the Prophet, who would surely have been familiar with the legal rules he followed, made decisions that appeared inconsistent with the rules later deduced from hadith. His conclusion was that all or almost all of the hadith, including the authenticated ones, were bogus, invented in the early centuries in the course of conflicts over the law. In his view Muslim law was actually an amalgam of pre-existing Arabic legal rules, administrative regulations created by the first Muslim dynasty, and legal rules taken over from the rules of the conquered provinces. Once disputes arose among different schools of law the authority of the Prophet provided the most decisive evidence available, so both sides to such disputes invented traditions to support their positions. Scholars who argued for derivations of the law not based on divine revelation eventually lost out to those with the opposite position. Later scholars have criticized parts of Schacht’s argument but most, at least among non-Muslim western scholars, appear to accept much of his thesis.²⁰⁴

The Schools of Law

The schools of law into which Sunni legal scholars divided themselves were each associated with the work of a single scholar from whom they took their name: Maliki, Hanbali, Shafi’i, and Hanafi²⁰⁵. They agreed on the broad outline of the law, disagreed on the details. Thus, for example, the punishment for drinking wine was eighty lashes according to three of the schools, forty according to the fourth (Shafi’i). Three schools interpret the rule as forbidding the consumption of any intoxicant, one (Hanafi) as forbidding only wine drinking and intoxication. The Shafi’i school requires zakāt, the religious tax, to be distributed equally among the eight categories of recipients while the Hanafi school permits it to be distributed to all, some, or only one category.²⁰⁶

While the schools differed in detail they regarded each other as mutually orthodox.²⁰⁷ In this respect as in others, the history of Islamic law both resembles and differs from that of Jewish law. The schools of Hillel and Shammai tolerated each other for several generations, but eventually the majority school suppressed the minority. In the parallel Islamic case, the four schools of Sunni law have continued their mutual toleration up to the present day.

The two major branches of Islam are Sunni and Shia, a division that goes back to a dispute over the succession to the caliphate after the death of the Prophet. The Shia believe that Ali, the Prophet’s cousin and son-in-law, ought to have been the first successor and that the succession should properly have run through his descendants. The Sunni support the succession that actually occurred, with Abu Bakr, one of Mohammed’s closest companions, chosen to be his successor, followed by Umar, Othman, and finally Ali. The split became permanent when Muawiya, nephew of the third caliph and governor of Syria, refused to accept Ali’s succession to the caliphate, setting off the first Muslim civil war and eventually establishing the first Islamic dynasty.

The four schools of law are all Sunni; the Shia have their own schools and legal rules, in most respects similar. While different schools were dominant in different areas, a medieval Muslim city could have had separate courts for the four Sunni schools, the Shia, and the other tolerated religions.²⁰⁸ It was a polylegal system; disputes within each community would go to that community’s courts. Non-Muslims had to use Muslim courts for criminal cases but had choice of law for civil matters. In at least some times and places, parties creating a contract, a partnership, a marriage, could choose which school’s legal system they wished to create it under and would be bound to the rules of that legal system in any future dispute. What happened in a dispute between parties adhering to different legal systems is not entirely clear and probably varied across time and space. According to at least one authority, the most common rule was for the dispute to go to the defendant’s court.²⁰⁹

While law was in theory independent of the state, in practice, in most historical Islamic societies, state-created rules played a significant role. It was up to the ruler to appoint the qadi and to enforce his judgments. The ruler could, by his control over jurisdiction, determine what court a case went to. Under the Abbasids, the second Islamic dynasty, the police (shurta) began to investigate, try and punish offenders outside of the fiqh courts. The inspector of the marketplace (muhtasib) created and enforced commercial regulations. The mazalim courts were created to investigate misconduct by officials, including qadis, but over time expanded to take jurisdiction over additional areas. Finally, the ruler was entitled to create administrative regulations to implement fiqh or fill in gaps in the law through the mechanism of siyasa, creating rules additional to and even to some degree contradictory to the rules of fiqh.²¹⁰ In theory all such courts functioned in Siyasa Shari’a, the general spirit of Shari’a, but they were not bound by the specific rules of fiqh, just as the Jewish communal courts were supposed to be faithful to the spirit of Jewish law but were not always consistent with Rabbinic law.

One reason for the development of parallel state courts may have been the desire of the ruler to maintain control. A second was that fiqh had serious limits as a legal system. It provided explicit rules for only a limited set of offenses and its evidentiary standards were in some contexts too hard to meet and in some, perhaps, too easy. For most controversies, proof required the eyewitness testimony of two adult, competent, male Muslims to the same illegal act; once their good character was established the testimony could not be impugned by cross examination or other means.

One further feature of fiqh echoes a pattern we earlier saw in Jewish law, the role of oaths. For some cases, an oath by one party could replace one of the two required witnesses. In addition, after the qadi had ruled in favor of one party, that party still had to take an oath to the truth of his position. If he failed to do so the other party was given the option of swearing to the truth of his account and, if he did so, won the case.

What Happened to Islamic Law?

Wael Hallaq, one of the most prominent of the modern scholars, argues that the Islamic legal system functioned better than most modern systems, providing legal services for free to all, defending the poor and powerless against the rich and powerful. While rulers played some role, biasing the system in their own favor or filling in its gaps, its rules were primarily the creation of scholars making an honest and well-informed attempt to learn and implement the divine will. The scholars maintained their independence in part through the existence of waqfs,²¹¹ the medieval equivalent of foundations or trust funds, providing mosques and madrassahs with a permanent income some of which could be used to support legal scholars. While the original funds frequently originated with a ruler or a ruler’s close kin, those being the people most likely to have large sums available for the purpose, past donations were out of the control of the current ruler. And the rulers, being for the most part more interested in keeping the support of the legal scholars than in the content of the law, were willing to leave the latter mostly in the hands of the scholars.

In Hallaq’s view it was the breakdown of this system in the nineteenth and twentieth centuries, due to the rise of the nation state under western influence, that destroyed the traditional system. In Islamic territories under colonial rule, such as India, Indonesia, and Algeria, the colonial rulers replaced the traditional system of decentralized law independent of the state with a system of statutory law incorporating elements of traditional law, in some cases elements interpreted in ways favorable to the ruling power. After the end of the colonial period, the newly independent states followed the same path. Thus, in his view, modern “Islamists” who view themselves as wishing to reinstitute Shari’a are proposing something quite different and less desirable, a centralized system of state-made law with rules to some degree modeled on traditional fiqh.²¹²

One problem with Hallaq’s thesis that the destruction of the traditional system was due to western colonialism is the case of the Ottoman Empire.²¹³ It was never colonized, yet in it as well the traditional system of independent and decentralized law was replaced by centralized state law.

Hallaq blames that change on indirect western pressure, sees it as a response of the Ottomans in the 19th century to the increasing power of the Christian states of Europe. But the shift began much earlier. The Ottomans gave the Hanafi school of law a legal monopoly within the core areas of the empire and a superior position in areas where other schools had been dominant prior to the Ottoman conquest. Qadis of the other schools were subordinate to the Hanafi qadi, who could reverse judgments that strayed too far from Hanafi doctrine. Over time, more and more of the legal scholars in those areas defected to the Hanafi school, that being where the money, jobs, and power were to be found.

The Ottoman Sultans claimed the right, when Hanafi teaching permitted alternative interpretations of the law, to tell judges which they must follow. And the Sultans proclaimed their own legal rules, kanun, in theory in support of fiqh but in practice sometimes inconsistent with it. Thus fiqh, according to all four schools, forbade loans at interest. Kanun provided for a maximum interest rate.

Interference by the authorities was not limited to the courts. Under the Ottomans there was a Grand Mufti appointed by the Sultan with ultimate authority over the appointment of qadis and the running of mosques and madrassahs. While Hallaq may be correct in his view of the independence of the legists outside the Ottoman Empire, within it that independence had been eliminated long before the beginning of the 19th century, at a time when the Ottoman Empire was a great power with neither reason nor inclination to take lessons from its western neighbors.

The breakdown of the traditional legal system may, as Hallaq argues, be due to the rise of the nation state, but the connection between that and western imperialism is accident not essence. The causes that led to the rise of the nation state in the west,²¹⁴ the replacement of feudalism by absolute monarchy, operated in the Islamic world as well, most notably in the Ottoman Empire. The annexation of the waqfs by the Ottoman authorities parallels the earlier confiscation of the lands of the monasteries by Henry VIII. The result in both cases was to eliminate institutions that competed with the state for power and resources.

The Content of Fiqh

The parts of Islamic religious law that deal with criminal offenses recognize three categories, defined by the nature of the punishment. Hadd offenses, mostly derived from Koranic rules, have fixed punishments. Ta’zir offenses have punishments set at the discretion of the judge. Jinayat offenses, homicide and bodily harm, have outcomes determined in part by law, in part by decisions made by the victim or his kin, and appear to be based on the rules of pre-Islamic Arabic blood feud.

Hadd

There are five hadd offenses: Unlawful intercourse (zina), false accusation of unlawful intercourse (kadhf), wine drinking (shurb), theft (sariqa), and highway robbery (qat’al-tariq), which includes armed robbery of a home. All are considered offenses against God, although in some cases against a human victim as well, and so cannot be pardoned.

Zina is defined as voluntary sexual intercourse with anyone who is not your lawful spouse or concubine. The punishment is either execution by stoning, for an offender who has previously had lawful intercourse, or a hundred lashes (fifty for a slave―half the number of lashes for a slave is a common pattern and I will not bother to note it hereafter) for one who has not. Proof of the offense requires either four eyewitnesses to the same act of intercourse, all four of whom must be competent adult male Muslims of good reputation, or confession; the pregnancy of a never-married woman can also be taken as sufficient proof. The confession is retractable; according to some authorities, an attempt to escape qualifies as a retraction and so voids conviction and punishment. The witnesses must be present at the trial and execution. In the case of stoning, if they do not throw the first stones the punishment is not carried out.²¹⁵

Such severe requirements of proof suggest that fornication and adultery would in practice be almost impossible to prove, but this applies only to the hadd offense.²¹⁶ Unlawful intercourse might also be prosecuted as a Ta’zir offense, with weaker standards of proof and less severe punishment.

Kadhf is defined as a false accusation of zina, where “false” is interpreted as “not proven,” making the role of witness to zina a hazardous one. If, for example, one of the four witnesses turns out to be a minor, the other three can be charged with kadhf. The charge must be brought by the person falsely accused. The penalty is eighty lashes. A false accusation of illegitimacy also counts as kadhf.

The case of a husband accusing his wife of adultery, directly or by challenging his paternity of her child, falls under a special set of rules. The husband who cannot prove the charge can protect himself from being charged with kadhf by swearing four times by Allah that he is speaking the truth and once calling down a curse upon himself if he is lying. The wife can defend herself against the accusation by the same series of oaths. If she is unwilling to do so that is taken as a tacit confession of guilt, subjecting her to the penalty for zina.

Shurb is narrowly defined as wine drinking or intoxication, more broadly as the drinking of any intoxicant, with the detailed definition varying among the different schools of law. The punishment is eighty lashes (forty in one school). Proof is by retractable confession or the testimony of two adult male Muslim witnesses.²¹⁷

The hadd offense of sariqa is defined as theft, but theft that meets a variety of requirements. The thief must be a competent adult. The theft must be intentional, accomplished by stealth, of an item of more than a specified minimum value. The item must be one protected by its owner, so stealing an animal grazing at a distance from its barn does not qualify, nor does stealing from a house where you are an invited guest. Stealing perishable food does not count because it is presumed that the theft is out of hunger and so permitted. If the thief thought the property was at least partly his, whether correctly or not, it did not count as theft. It could be argued that theft of public property never counted, since all Muslims were in some sense joint owners. The victim of the theft must attend both trial and punishment.²¹⁸

Arguably the list of requirements is so extensive because legal scholars, like many non-Muslim commentators, regarded the punishment, amputation of the right hand, as excessive. Since the punishment was Koranic it could not be changed, but it could be hedged around with enough qualifications so that it was unlikely to be applied―the same approach that Jewish legal scholars applied to the rule about stoning a disobedient son. A theft that did not meet the requirements for the hadd offense could still be prosecuted and punished under ta’zir.²¹⁹

Qat’al-tariq is defined as either robbery of travelers far from aid or armed entry into a private home with the intent to rob it. The punishment is amputation of the right hand and left foot. If combined with murder, the punishment was death by the sword²²⁰; if the theft was not only attempted but accomplished and combined with murder, the punishment was crucifixion. Unlike the case of a homicide prosecuted under jinayat, punishment was mandatory; the victim or victim’s kin did not have the option of canceling it or of accepting blood money instead.

Ta’zir

For ta’zir offenses, the punishment was up to the judge, with options ranging from a private admonishment to death but limited, in the view of most of the schools, to less than the punishment that would be imposed (according to some on a slave, to some on a free man) for the corresponding hadd offense. Proof is by the testimony of two witnesses, one of whom can be a woman, or by a non-retractable confession; some legal scholars held that the judge could act on the basis of his own knowledge even without such proof.

The bulk of criminal offenses in fiqh are treated as ta’zir offenses, including both non-hadd offenses and hadd offenses that for one reason or another do not meet the strict hadd standards. Kadhf is only a hadd offense if committed against a Muslim, but ta’zir can be used to protect non-Muslims against unproven accusations of unlawful intercourse.

Jinayat

The part of fiqh that applies to homicide or bodily injury is called jinayat and appears to be based on the pre-Islamic rules of Arab blood feud. The charge is brought by the victim if he is alive or his nearest relative if he is dead. The punishment is either retaliation or blood money (diya). Retaliation occurs only at the request of the victim if alive, his nearest kin if the victim is dead, and is to be inflicted by victim or kin. In the case of homicide, retaliation means death, in the case of injury it means imposing an identical injury. Where retaliation is one of the options, the victim or his closest kinsman may demand blood money instead or negotiate an out-of-court settlement. Jinayat, like modern tort law, is based on private action; there is no official responsible for initiating the case.²²¹ But if the victim has no living relatives, the right of retaliation falls to the state.

The formula for blood money is based on either the heavy diya―a hundred female camels, evenly divided among one-, two-, three-, and four-year-olds―or the lighter diya, eighty female camels again evenly divided in age plus twenty one-year-old male camels; it can also be 1000 dinar or 10,000 dirham.²²² Blood money for homicide consists of the full diya, heavy or light depending on the circumstances. For injury the payment is scaled by a simple, if somewhat arbitrary, formula―half a diya for the loss of something of which the victim has two, such as a hand, arm, foot, or leg, a tenth of a diya for the loss of a finger. This has the odd result that the payment for the loss of a nose, of which the victim has only one, is a full diya, twice the payment for the loss of an arm or leg. Where no such formula is applicable, the payment is that percentage of a full diya corresponding to the percentage by which a slave’s value would be reduced by the same injury.²²³

The law distinguishes among degrees of homicide. In the case of willful homicide―no legal excuse, intentional and committed with a weapon that normally causes death―the victim’s nearest kin is entitled to demand retaliation or the heavier diya. Most schools include as homicide false testimony at trial that results in death.²²⁴ Where the homicide was accidental or the killer believed he was acting legally, retaliation is not an option and the penalty is the lighter diya. In all cases, conviction is by confession or the testimony of two adult male witnesses.²²⁵

In some cases, the payment of blood money is in theory shared between the offender and his ‘Akila, corresponding to the dia-paying group²²⁶ of Somali law described in chapter 11. According to Schact, “the whole institution fell into disuse at an early date.”

That claim is apparently mistaken, a fact I discovered when a group of Saudi LLM students took the course from which this book developed, providing me with primary sources in the classroom. Each was a member of an ‘Akila, which they translated as “clan.” As one of them explained in a paper written for the course, his ‘Akila would be responsible for a third of the diya due for an accidental homicide or injury.²²⁷

Ridda

There is one crime that does not fit clearly into the categories I have listed―ridda, apostasy, the crime of converting away from Islam. Some view it as a hadd offense, others as an offense to be treated under siyasa, administrative regulations created by the state.

The usual view is that apostasy is a capital offense, a position based on the purported practice of Mohammed. Some authorities hold that it may be punished without trial, others that the apostate is entitled to a trial and one school that the apostate must be given three days to repent and return to Islam. Women apostates are to be imprisoned and, according to some schools, beaten until they recant.

The Koran sentences the apostate to hell but prescribes no punishment. Some scholars argue that the death penalty was a punishment not for abandoning Islam but for the (historically closely connected) offense of rebelling against it, and that belief alone was not to be punished. On the other hand, one school holds that not only is apostasy from Islam forbidden, so is apostasy from any of the other religions of the book.²²⁸

Marriage Law

Under Islamic law, marriage is treated as a contract. Husband and wife are legally distinct, each entitled to own property. The wife’s property includes the dowry agreed to as part of the marriage contract. Part of it is paid at the time of the marriage but part, sometimes a large part, may be due only after divorce or the death of the husband. Thus, while a husband is free to divorce his wife, doing so may be costly. A wife does not have the right to divorce her husband but can bargain for a divorce by offering to give up part of what he would owe her as a result. At some times and places, the wife could also apply to have the court, for cause, act in the husband’s stead to divorce her. And the marriage contract could include conditions such as the right of the wife to be divorced if the husband took a second wife.

The wife owes her husband obedience and sexual access, the husband owes the wife adequate intercourse and support at a level suited to her station, defined by the status of her family.

Taxation

The two Koranic taxes are jizya, an annual head tax paid by non-Muslims under Muslim rule, and zakāt, a property tax on productive property owned by Muslims.²²⁹ The latter is supposed to go to a specified set of categories of recipients. The taxpayer may pay it to the ruler to distribute for him, to a middleman who takes a share to pay him for the work of handing out the money, or to the designated categories of recipients himself. In Twelver Shia law the payment goes to the high-level religious scholar whose interpretation of law the individual taxpayer has chosen to follow, to be distributed by him.

In addition to these two, Islamic states have imposed a variety of more religiously dubious taxes to fund their activities.

Jewish and Muslim Law: Similarities

Studying both, I was struck by how much they had in common. For example:

Islamic law permits a man only four wives. According to Torah law “a man may marry several women, even a hundred of them, either at the same time or one after another … provided that he is able to supply each one of them with the food, raiment, and conjugal rights due her.” However Maimonides reports that “The Sages have therefore ordained that a man should not marry more than four wives, even if he has a great deal of money, in order that each wife’s schedule should be at least once a month.”²³⁰

Here the “schedule” is of intercourse with her husband, the “conjugal rights due her.” Maimonides offers a detailed account of how often a man is obliged to sleep with his wife, depending on his circumstances. “For men who are healthy and live in comfortable and pleasurable circumstances, without having to perform work that would weaken their strength, and do nought but eat and drink and sit idly in their houses, the conjugal schedule is every night. … for sailors, once in six months; for disciples of the wise, once a week, because the study of Torah weakens their strength.”

According to fiqh, at least one Shafi’i version, “One should make love to one’s wife every four nights, as is fairest, since the number of wives one may have is four … though one should make love to her more or less than this, according to the amount she needs to remain chaste and free of want for it, since it is obligatory for a husband to enable her to keep chaste.”²³¹

In Islamic law, a man has the right to intercourse with his wife as long as there is no good reason, such as menstruation or a medical problem, not to permit it. In Jewish law a wife who refuses intercourse with her husband is a “rebellious wife.”²³² She cannot be compelled but her husband is to divorce her and she forfeits her ketubah, the money that would normally go to her when divorced or widowed.²³³ A man who is unwilling to have intercourse with his wife is obligated to divorce her and pay her ketubah. The rules around marriage are similar, but not identical, in a variety of other ways between the two systems.²³⁴ Thus “He may not, however, compel them to reside in the same courtyard, but must let each one reside by herself. … .”²³⁵ and “It is unlawful for a man to house two wives in the same lodgings unless they both agree.”²³⁶

Marriage rules are not the only points of similarity. A major source of Islamic law is the Sunna of the Prophet, the practice of Mohammed as reported in hadith. In Jewish law, the act of a person of acknowledged eminence as a halakhic authority can be the basis of a legal norm.²³⁷ Both systems sometimes allow oaths as part of the legal process. One view of the role of custom in Jewish law is that it has no independent force but its existence is evidence of the existence of a legal rule now lost.²³⁸ Islamic legal scholars take a similar approach to acts of Mohammed that do not appear to be based on the text of the Koran—that they imply the existence of lost portions of the text.

Another similarity is in the source of the law. Jewish law is based on the written Torah and the oral Torah, the latter based on rules of law and interpretation believed to have been revealed to Moses and transmitted orally down the generations until finally written down in the Mishnah. Islamic law is based on the written Koran and the hadith, statements and practices of Mohammed believed to have been based on revelation and passed orally down the generations until finally written down. Two differences are that the hadith were transmitted orally for a century or two before being written down, the Oral Torah for more than a millenium, and that each hadith is accompanied by a chain of transmitters back to its origin, whereas rules of the oral Torah are cited on the authority of only the later transmitters.

During the period when there were two schools of Jewish law, some authorities held that one could choose either but should then follow all of its rulings, both lenient and strict, rather than feeling free to pick and choose.²³⁹ A similar position is argued by some Muslim authorities with regard to the four schools of Sunni law. Later Jewish scholars faced the problem of choosing among differing opinions by different eminent scholars of the past. The solution chosen by Joseph Caro for the Bet Yosef was to base his position on the view of the majority of the three most eminent authorities. Muslim scholars came up with similar rules for deciding which among divergent opinions of the scholars of a school should be taken as the school’s doctrine.

Both systems had to come to terms with the limited ability of those applying the law, rabbis, muftis, qadis, to engage in the elaborate scholarship by which the law was derived from religious sources and came up with similar solutions: permitting reliance on conclusions reached earlier by authorities believed competent to produce them.

Appendix: Two Stories

A Convenient Tradition

Ten Traditionists, Ghiyath ibn Ibrahim among the rest, were summoned to an audience. Now Mahdi was very fond of pigeon-racing; and when Ghiyath was presented, and somebody said: Pray recite some tradition to the Prince of the True Believers, Ghiyath recited: So-and-so told us that he had it of So-and-so on the authority of Abu Hurayra that the Apostle of God (God’s Prayer and Peace be on him!) said: There must be no betting save on a hoof or an arrow or a lance-head; and then Ghiyath added: or on a wing.

Mahdi at once ordered him a bounty of ten thousand dirhams; but as Ghiyath rose bowing to thank him, he exclaimed: By God, the nape of your neck looks like the nape of a liar’s neck—I’ll swear you interpolated those last words! And he gave order immediately that all his racing pigeons should be killed.²⁴⁰

Working Around God’s Law

The poet ibn Harma performed for the Prince of the Muslims and so delighted was the Caliph with his performance that he said “name your reward.”

The poet replied, “the reward I wish from the Prince of the Muslims is that he should send instructions to his officials in the city of Medina, commanding that when I am found dead drunk upon the pavement and brought in by the city guard, I be let off from the punishment prescribed for that offense.”

“That is God’s law, not mine; I cannot change it. Name another reward.”

“There is nothing else I desire from the Prince of the Muslims.”

Al-Mansur thought a little, then sent instructions to his officials in Medina commanding that if anyone found the poet ibn Harma dead drunk upon the pavement and brought him in for punishment, ibn Harma should receive eighty strokes of the lash as the law commands. But whoever brought him in should receive a hundred.

And ever after, when someone saw the poet lying dead drunk upon the pavement, he would turn to his companion and say “a hundred for eighty is a bad bargain” and pass on.²⁴¹

---

¹⁹⁶ “Some, in particular younger Muslim scholars, tend to use ‘Shari’a’ as a name for the divine will as only God knows it; an abstract divine law only perceived by Him. … what the jurists have on the basis of this Revelation formulated though their reason is only ‘law’ as science, fiqh.” (Vikør 2005, 2). This is how I use the terms in this chapter, since that seems to me the best way of making sense of them. As Vikør makes clear, not all Muslim scholars agree.

¹⁹⁷ For example, charging interest (riba) is forbidden but scholars have disagreed as to whether it is punishable by law or only by God in the afterlife.

¹⁹⁸

Schacht (1986 Chapter 10, 69-75) argues for the closing of the gates, but his position was challenged by Wael Hallaq in Hallaq 1984 and his view, that complete closure never happened, seems now to be generally accepted. By Hallaq’s account, it was generally agreed by sometime in the fourth/tenth century that no more schools of law would be founded, there being no more mujtahids of the ability of the eponymous founders, but there remained many mujtahids capable of doing ijtihad within the four existing schools. Sometime in the late fifth/eleventh early sixth/twelfth century it began to be debated whether mujtahids could become extinct and ijtihad thus impossible, and in later centuries some scholars argued that it had happened while others denied it.

After the fourth/tenth century, legal doctrine had reached an exquisite level of detail and sophistication, and one would be at pains to find a case entirely without precedent. Yet, the jurists needed legal theory after this time no less than they had before. Just as many elements of it were employed to construct early law, it was summoned in later periods to adjudicate between the many legal opinions that it had itself produced over time.

Hallaq 2009, 76

¹⁹⁹ Al-Nawawi 2009 is a modern translation of a thirteenth century summary of the law with notes aimed at modern Muslim readers. Chapter 8, in the process of explaining why deducing the law from Koran and Hadith is a hard problem that requires scholarly expertise, provides examples of how the source material can be interpreted and misinterpreted.

²⁰⁰ “The fundamental role of the jurisconsult [mufti] in constructing and replenishing the body of substantive and procedural law is underscored by the fact that the legal theorists were insistent throughout the centuries upon equating the mujtahid with the jurisconsult, not with the judge. True, unlike the magistrate’s decision, the jurisoncult’s legal opinion was not binding, but his opinion, by virtue of its having emanated from a highly qualified authority became part of, and indeed defined, the law. It was the common perception in the legal profession that the judge’s decision is particular (juz’i) and that its import does not transcend the interests of the parties to the dispute, whereas the jurisconsults’ fatwā is universal (kulli), and applies to all similar cases that may arise in the future.”

Hallaq 1997, 154

²⁰¹

“The first government appointment of a mufti appears to have been that made in Damascus with the creation of Dar al-‘Adl, House of Justice, in the latter part of the seventh /thirteenth century, or the first part of the eighth /fourteenth century.”

Makdisi 1981, 199

“As the state grows stronger it begins to appoint muftis to particular courts. … We know that the Mamluks, who in many ways were intermediaries between the medieval and the Ottoman periods, did appoint muftis at some of the major mosques, while the rest of the country still had ‘self-appointed’ muftis.”

Vikør 2005, 145

²⁰² Hallaq 1997, 145.

²⁰³ Schacht 1986 Chapters 4-11. The argument was earlier made by Ignatz Goldziher.

²⁰⁴

Thus Wael Hallaq criticizes Schacht while arguing for a position that appears, at least to this non-expert observer, to differ from his only in detail:

However, mounting recent research, concerned with the historical origins of individual Prophetic reports, suggests that Goldziher, Schacht and Juynboll have been excessively skeptical and that a number of reports can be dated earlier than previously thought, even as early as the Prophet. These findings, coupled with other important studies critical of Schacht’s thesis, go to show that while a great bulk of Prophetic reports may have originated many decades after the Hijra, there exists a body of material that can be dated to the Prophet’s time.

Hallaq 1997, 2-3

For context, al-Bukhari, one of the authorities for authenticated hadith, lists 7275 of them, purportedly culled from 600,000 candidates. Another authority, Muslim, included 9200, culled from 300,000.

In a later work, Hallaq writes:

The dramatic increase in Prophetic authority at the turn of the second/eighth century involved projecting on Muhammad post-Prophetic sunan as well. Legal practices and doctrines originating in various towns and cities in the conquered land, and largely based on the Companions’ model, began to find a representational voice in Prophetic Sunna. The projection of the Companions’ model back onto the Prophet was accomplished by a long and complex process of creating the narrative of hadith.”

Hallaq 2009, 45

Which sounds close to Schacht’s position.

²⁰⁵ There were several other Sunni schools that eventually died out.

²⁰⁶ For details see Donaldson 2000, 60-63.

²⁰⁷ The situation with regard to the Shia schools of law is more complicated. Some Sunni legal scholars consider Shia heretics, and similarly in the other direction. There are also substantial divisions among the Shia.

²⁰⁸ The Koran established three tolerated religions in addition to Islam: Judaism, Christianity and a third usually held to be the Sabeans, inhabitants of a Semitic kingdom in what is now Yemen, mentioned in both the Bible and the Koran. After the conquest of Persia, Zoroastrianism was generally held to be an additional tolerated religion.

²⁰⁹

“The most normal situation is that if plaintiff and defendant belong to different madhabs, then they should use that of the defendant. However, a losing party may try to bring his case up at the court of a different madhab, if there is one in the town. But as this might lead to a conflict of competence between the courts of the schools, it would in most cases be difficult to get a judge from one school to accept a case that had already been decided by a colleague from another. …

Instead the losing party could go to a mufti and ask whether the verdict is correct. The mufti cannot then evaluate the evidence of the case … but he can give his view on the way the law was used, and declare this to be faulty. That would not lead to any automatic reversal or reopening of the case, but it would be a good argument for the party to approach another judge for this purpose. Such a new evaluation should only occur when the first judge has died [or been deposed] … but it may also occur that old cases are taken up again in new courts while the first judge is still in office.”

Vikør 2005, 151

“… the doctrine that proclaims judicial review and outright repeal of an existing legal decision to be valid when it can be proven that the decision stands in violation of the dictates of scripture and/or of consensus. … It seems that the only other ground for repealing an earlier decision is discovering that a mistake has occurred in determining the evidence on the basis of which the decision was reached.”

Hallaq 1997, 156

Both are being distinguished from a disagreement about ijtihad, the derivation of legal rules from the textual sources, which is not appealable.

²¹⁰ In theory, Islamic law, like the 18th c. English criminal law described in chapter 15, provides for no public prosecutor; every Muslim has the right to act as a private prosecutor and cases may be settled out of court without ever involving the qadi. But the Siyasa Shari’a courts made possible something more like ordinary criminal law, with cases initiated by state appointed authorities.

²¹¹ The waqfs I discuss here were trust funds dedicated to religious purposes. There were also waqfs dedicated to purposes of secular welfare, such as the maintenance of public fountains, and others that were family foundations, money provided by one generation to be under the control and spent for the benefit of their descendants.

²¹² Modern Saudi Arabia is to some degree an exception to this pattern, with independent legists. There is a dominant (Hanbali) school of law but some options for having controversies adjudicated under the rules of other schools.

²¹³ My account of Ottoman institutions is based mostly on Gerber 1994.

²¹⁴ I am offering no answer to the interesting question of what those causes were.

²¹⁵ Schacht 1986, 176.

²¹⁶ According to Ibn Taymiyah, writing in the late 13th or early 14th century, up to his time Zina, unlawful intercourse, had never been proved by the required four witnesses.

²¹⁷ This restriction applies only to Muslims. Non-Muslims are permitted to make, transport, and consume wine.

²¹⁸ Schacht 1986, 176.

²¹⁹ “There is a strong tendency to restrict the applicability of hadd punishments as much as possible, except the hadd for false accusation of unlawful intercourse, but this in turn serves to restrict the applicability of the hadd for unlawful intercourse itself. The most important means of restricting hadd punishments are narrow definitions. Important, too, is the part assigned to shubha, the ‘resemblance’ of the act which has been committed to another, lawful one, and therefore, subjectively speaking, the presumption of bona fides in the accused.” (Schacht 1986, 176).

²²⁰ According to Vikør, “a highway robbery that did not lead to murder is treated as theft.” (Vikør 2005, 283)

²²¹ Schacht 1986, 177-8. In addition to making a killer subject to diya or retaliation, homicide also creates the obligation to free a believing slave or fast for two months. For a detailed account of the rules of Jinayat, see.

²²² Schacht 1986, 185. Vikør estimates that 1000 dirhem corresponded to five to ten years income at an average level, not far from my estimate for Icelandic wergeld in chapter 10 (Vikør 2005, 289). In modern Saudi Arabia, diya for a deliberate murder is over a hundred thousand dollars.

²²³ Schact 1986, 186. A similar formula appears in Maimonides, although in that case it is the amount by which the victim’s value would be reduced if he were being sold as a slave. (Maimonides 1954, Treatise IV Chapter 1).

²²⁴ Also the Jewish rule, to judge by Maimonides’ analysis of the problem of convicting a murderer who is suffering from a fatal disease.

²²⁵ Non-Muslim witnesses are acceptable against a non-Muslim defendant but not against a Muslim defendant.

²²⁶ I use the romanization “diya” in this chapter, “dia” in the Somali chapter, in each case following the usage of my sources.

²²⁷

“The ‘Akila consists of those who, as members of the Muslim army, have their name inscribed in the list … and receive pay, provided the culprit belongs to them; alternatively, of the male members of his tribe (if their numbers are not sufficient, the nearest related tribes are included); alternatively, of the fellow workers in his craft or his confederates; … . This institution had its roots in the pre-Islamic customary law of the Bedouins, where the culprit could be ransomed from retaliation by his tribe, … . … the whole institution fell into disuse at an early date.” (Schacht, 186) “… the principle is that no individual should be asked to pay more than four dirhem, so if a death has ensued, 2,500 paying colleagues have to be found. If this is not possible the state may intervene or the person who made the accidental killing will have to take on the loss after all.” (Vikør 2005, 289).

According to my Saudi students, members of an ‘Akila share a common last name and, like members of the Irish kin groups discussed in a later chapter, descent from a common ancestor in the male line. Like the Somalis, the Saudis know their male line genealogy for many generations—ten in the case of one of my students to whom I put the question. The size of the clan is typically a few thousand males, similar to the size of a dia-paying group. A student paper on the ‘Akila.

²²⁸ “For the more casuistical Shafi’i school, any act of apostasy was fatal, even from, say, Judaism to Christianity.” (Forte 1999, 161). But that would clearly not apply to conversion to Islam.

²²⁹ The zakāt consists primarily of a wealth tax of 2.5% on gold, silver, and commodities for trading, an income tax of 10% on crops grown in naturally watered land and of 5% on crops grown in irrigated land. (Al-Nawawi 2009 Chapter 4).

²³⁰ Maimonides 1972 Treatise 1 Chapter XIV, 87.

²³¹ Al-Misri 1991 m5.2, 525.

²³² Maimonides IV Treatise 1 Chapter XIV, 89.

²³³ Under Romani customary law (Chapter 2), if a wife refused intercourse with her husband the husband’s kin are entitled to return her to her kin and demand a refund of the bride price paid for her. “… she wouldn’t sleep with him as a wife, only as a sister. … So I got in touch with her father and said I wanted my money back.” (Sutherland 1975, 293).

²³⁴ Both, for instance, take it for granted that marriage and intercourse will sometimes occur prior to puberty, and discuss the legal implications.

²³⁵ Maimonides 1972, 87.

²³⁶ Al-Misri 1991, 538.

²³⁷ Elon 1994, 946.

²³⁸ Elon 1994, 883-5.

²³⁹ Elon 1994, 1064.

²⁴⁰ Schroeder 1995, 281-2.

²⁴¹ Khawam 1980, 153-154, not verbatim.