PROCEDURAL LAW BETWEEN TRADITIONISTS, JURISTS AND JUDGES: THE PROBLEM OF YAMiNMA ' AL-SHÀHW

Modem critical studies of Islamic law have noted the need for the study of the procedural law in Islam and the role of the judicial practice in its formation. Generally, it is believed that qàdi^ had freer hand regarding witness and method of proof in the early period, but gradually it became more restrictive and rigid. Scholars have suggested various dates ranging from the second half of the seventh century (Tyan) to the eighth (Schacht, Dannhauer), to the eleventh century (Coulson) for this development. These scholars have treated the doctrine of yamin ma * al-shàhid (oath with one witness) as an exception to the Shariah procedure which, according to them, strictly and mechanically adhered to the doctrine of two witnesses. According to them, this doctrine emerged under local [Schacht (Egypt) and Dannhauer (Madina)] or Jewish (Patricia Crone) influence. The present paper argues that the doctrine must have grown against the historical setting of the Civil War (Tahkîm period) when the questions relating qisàs, qawad, qasàma and hakam were discussed frequently with reference to the role of the Qur'àn and the common sunna and the need for reforming the institution of hakam was felt. On the basis of scattered statements of scholars close to that period supported by WaM"s remarks and we conclude that Mu'âwiya introduced this method to reform the institution of hakam. The judicial practice of this doctrine also confirms that it was too widespread to call it a Medinese or local practice as claimed by the classical jurists, traditionists and some modem scholars.

as two methods of procedure used by these courts respectively.While Shari 'a court procedure was fixed and restrictive, the siyàsa was free and discretionary.^N. J. Coulson^ places this change in the eleventh century^.He adds that Shari 'a procedure was rigidly formalistic and mechanical, restricted to two witnesses; one witness was accepted only in exceptional cases.^These are quite insightful observations about the changing procedural law of Islam, but they are very general remarks and lack focus on details.As a matter of fact, contrary to Coulson's observation, qàdis accepted one witness quite frequently, and the doctrine of yarrfm ma' al-shàhid (oath with one witness) has been one of the most debated problems among the traditionists, judges and the jurists.
Joseph Schacht,^^ P. G. Dannhauer^^ and Patricia Crone^^ have gone into some details of the doctrine oí yamin ma ' al-shàhid although they remain more concerned with its foreign origins.
Schacht and Dannhauer both find this doctrine growing in Egypt as a local judicial practice that influenced the theory of Islamic law in Medina and Hijaz.They argue that the Egyptian Qâdï Tawba b.  judgement based on the procedure of yamin ma ' al-shàhid, developed into the Medinese tradition against the Iraqi tradition, and was later projected as a Sunna of the Prophet.According to Dannhauer, the doctrine was also attributed to Iraqi authorities to prove its universal practice.^^Although Dannhauer's critical analysis of some of the reports attributing the doctrine to Medina shows the transmitters' connection with Egypt, yet unlike Schacht, he does not dismiss contrary reports.^"^ Schacht does not say clearly that the doctrine originated with Qâdï Tawba.Kîndï, to whom Schacht refers on this point, does not say so either.In fact Kïndfs statement that the qàdi, i.e.Tawba, applied this procedure «even to 6 Tyan, 443 and 572.^ Coulson, N. J., A History of Islamic Law (Edinburgh, 1964), 126ff; and «Doctrine and Practice of Islamic Law», BSOAS, XVffl (1956), 211-226. ^ Coulson, 127. 9 Coulson, 124.^° Schacht, J., Origins of Mohammedan Jurisprudence (Oxford: Clarendon, 1950).'' Dannhauer, P. G., Untersuchugen zur frühen Geschichte des Q-Amies.Inaugural-Dissertation zur Erlangung der Doktorwiirde der Philosophischen.Bonn: Fakultat der Rheinischen Friedrich-Wilhelms-Universitat, 1975.
'^ Crone, P., «Jàhilî and Jewish Law: The qasamah», Jerusalem Studies in Arabic and Islam, 4 (1984),  disagrees with Schacht regarding Darâwardï and Rabi'a being responsible for the circulation of the false reports about the doctrine.smaller things»^^ implies that the practice already existed and the qadi extended it to matters to which the other judges did not.
Comparing the Jahili law of qasàma with Islamic and Jewish law, Patricia Crone suggests that the doctrine in question was a fusion of Jewish and Jahili laws.She disagrees with W. Heffening^^ who suggested that the Muslim jurists turned to Jahili tradition to develop an alternate legal tradition to the Umayyad system.She finds the Islamic law of qasàma directly influenced by Jewish law.
We believe that the quest for foreign origins and extraneous elements in Islamic law does not help to understand the nature of Islamic legal system properly, because such traces cannot characterize Islamic law as a continuity of Roman or of Jewish legal tradition.Second, procedural law, especially relating to oath, as observed by Lawrence Rosen, ^^ has a special social meaning for the people who negotiate their relations, disputes and rights in that framework.The foreign influence is, therefore, immaterial in these cases.Third, Crone has relied on fiqh and hadith literature more than on the judicial practice.In our view, it is the latter that may properly reflect whether a tradition was Iraqi or Medinese.
Besides, none of the above mentioned three scholars explain why the doctrine appeared in Egypt in that particular period, nor do they refer to any specific historical setting that gave rise to this doctrine.In order to explore possible foreign influence on Islamic law, they do not refer to Egypt either.They instead focus on Iraq where they find different possible influences at work.But since none of them describes this doctrine as an Iraqi tradition, the question about its historical background remains unanswered.One cannot assume that the doctrine grew suddenly.Also the serious and widespread debates about the doctrine that continued through the fifteenth century point to its deep social and political implications.^^M. KH.MASUD AQ.XX, 1999 In our view, the problem oíyamín ma ' al-shàhid seems to have arisen in the wake of Civil   between 'All (d. 661) and Mu'awiya (d. 681) on the question of the qisàs for the murder of 'Uthmàn, better known as the issue of tahkim (appointment of hakam, arbitrator).The open-ended and arbitrary hakam procedural law failed quite obviously in this period.Mu'áwiya felt the need for a comparatively strict procedure for the qàdi.
This paper first reviews briefly the events of the Civil War to examine why and how the institution of hakam failed.Next, it studies how Mu'áwiya responded to this issue.For the latter part of our inquiry we rely on Wakî"s (d. 306/918) Akhbàr al-QudàtP We have chosen Akhbàr al-Qudàt primarily because a qàdi wrote it when Islamic procedural law was still in its formative stages.Second, the book was designed to report judicial practice and is certainly preferable to ihtfiqh texts and the Adab al-Qàdi books, which focus on theory rather than on practice.As a matter of fact Tyan had warned against relying on jurists and their Adab al-Qàdi genre of literature for a study of judicial organization in Islam, as they would describe the ideals rather than the actual judicial practice.Furthermore, the Akhbàr al-Qudàt is a source earlier than Kindfs (d.362/972) Kitàb al-wulàt on which Tyan and Schacht had relied.While Kind! limited himself to Egypt, Wakï' deals with the qàdis all over the Muslim world until his period.Wakî' is also not influenced by jurists or their doctrines as he is not known for adherence to any school of law.Nor do the traditionists influence him as he narrates the events on the authority of his own sources.Historians like Ibn Qutayba al-Dïnawan^^ have relied on Waki"s books.-^^The foremost reason for the choice of Akhbàr al-Qudàt is that it refers to actual judgements of the qàds more than any other source in our knowledge.Noting this feature of the book, Schacht characterized Akhbàr al-Qudàt as «a main source for the study of this period».«This book», he added, «tells us about another phenomenon of legal activity, that is the development of law injudicial practice».^^ We have found in Akhbàr al-Qudàt twenty-ñve cases specifically related with the questions of procedural laws, 18 cases are reported with details of facts and ^^ Cf.Muhammad Khalid Masud, «A Study of Wakl"s (d. 306/917)    seven without describing such details.There are also several general statements by the author about certain qàdis who decided cases according to this doctrine.Since these statements do not identify parties or the subject of the dispute, we have not counted them in our data.A summary and analysis of these cases have been appended to the paper as Annexes number 1 and 2. We shall return to it after a brief review of the Civil War and the events related to the issue of tahkimP TAHKIM Literally, tahkim means appointment of a hakam, an arbitrator.In Islamic history, the term refers to the arbitration about the dispute between 'All and Mu'awiya on the issue of qisàs for the murder of 'Uthmàn (d. 656), the third caliph.Caliph 'All succeeded 'Uthmàn in 656, elected by the group who had revolted against 'Uthmàn.Mu'awiya claimed to be 'Uthmán's wan al-dam (next of kin quahfied to demand qisàs), and demanded ' Alï for retaliation according to law.'All refused and asked Mu'awiya first to take oath of allegiance.The dispute finally led to the battle of Siffin (657) between 'All and Mu'awiya.
Modern scholarship has raised several questions about the civil war and the tahkim.Mostly these studies explain the Civil War as a struggle for power between various groups.Caliph 'Alfs alHes were further divided into qurrà\ ridda tribes and groups of people from Kûfa, Basra, Mecca, Medina and Egypt.^"^The divisions ranged from between early and late converts to Islam, between city elite and tribes, between religious and other groups.Hawting considers some of these explanations exaggerated.^^For instance, it is an ^^ For a comprehensive analysis, historical as well historiographical, of the Civil War, see Petersen, Eriing Ladewig, 'Uthman and Mu'awiyah in Early Arabic Tradition (Copenhagen: Munksgaard, 1964), and Hinds, M., «The Siffîn Arbitration Agreement», JSS, 17 (1972)  exaggeration to say that all the Umayyad were late converts, 'Uthmán was not.It is similarly an exaggeration to say that it was a conflict between Syrians and Iraqis, or between Banü Umayya and Banù Hâshim.These studies have sought to simplify a complex phenomenon by analyzing the tribal alliance patterns.These attempts are certainly commendable, but in the process they tend to minimize the significance of the issue of tahkim and qims.Hawting remarks that the arbitration had little significance for the Civil War.^^ We cannot accept this conclusion because, first, qims and the institution of hakam were of central significance in the Arab society.Several scholars have spoken about the centrality of the institution of hakam and tha V (retaliation) in Arab society^^.The Qur'án also confirmed this pre-Islamic institution saying, «There is life for you in qims» (2:179).Second, murder of a caliph was not a small matter.Earlier, caliph 'Umar's murder had been avenged in due process.'Uthman's murder had turned into a blind murder and hence needed to be decided by a hakam.Third, the fact that the tahkim agreement was written down also shows the extraordinary significance of the issue of tahkim and qims.The text of the agreement, with some variations, has been reported in almost all the early accounts of the Civil war.^^Martin Hinds has studied the agreement in detail and found that essentially there are two versions of the text.The later version gives more details and introduces certain anachronistic elements such as the term sunnat al-Nabi.
The document goes into minute details about the time, place, security of the arbiters and the method and sources of the arbitration.We may divide the document into six parts.The first part mentions the two parties, the second mentions the terms of reference for the arbitration and the agreement of the parties to abide by them.The third section names the arbitrators, the sources of AQ.XX, 1999 THE PROBLEM OF YAMD^ MA  arbitration and its method.The fourth spells the safety of the arbitrators.The fifth fixes the place and time for arbitration.The sixth gives names of the persons who witnessed the agreement.Hinds does not find the subject of arbitration mentioned in this document.One finds it hard to believe that a document, which deals with minute details, would not mention the subject of the agreement.In fact, despite variations, all versions of the documents including the texts that Hinds analyzed, mention variably the phrase «to let live whom the Qur'àn allows to live and to cause to die whom the Qur'àn causes to die».^^Obviously, it refers to the term of reference for the arbiters.Certainly, it is not a figurative speech.It refers to the issue of the qisas and that it was to be decided according to the Book.
Earlier, Prophet's wife 'À'isha's demand for the qisas of 'Uthmàn had also led to a war, the battle of the Camel.Caliph 'All defeated her^^ but the question of 'Uthmàn's qisas had remained unsettled.During the battle of Siffin, when Mu'àwiya's forces were on the verge of defeat, his supporters raised the Qur'àn on spears appealing to accept the Book as hakam between the warring factions.Appeal to the Qur'àn as hakam seems to be a familiar method in those days.Resort to this method was also made in the battle of the Camel.^^In the battle of Siffin, after some hesitation on the part of Caliph 'AH, both parties finally agreed to appoint two hakams, one from each side: Abu Mùsà al-Ash'arî, 'All's hakam, and 'Amr b. al-'Às, Mu'àwiya's hakam.
Appointment of two, instead of one hakam, was justified with reference to two Qur'ànic injunctions where two hakams are prescribed in case of the offence of prohibited game in the vicinity of the sanctity of Mecca (haram) during pilgrimage (5: 95), and for reconciliation between a husband and a wife (4:34).32Montgomery Watt notes that the hakams in Siffin had two meetings.^^In the first meeting both hakams agreed that 'Uthmán was killed unjustly and that Mu'àwiya was the rightful claimant for his qisas.Apparently the main task of arbitration was completed in this meeting.The second meeting was to deal with the implementation of qisas itself.It appears that in this discussion, the ^^ Ibn Qutayba,op. cit.,128,Futüh,293, and other texts cited above in note 28.
^* Some reports attribute the raising of the Qur'in in the battle of the Camel to 'All and some to 'À'isha.See these varying versions in Taban, vol. 6, and 3189-3191. ^^ Al-Tabaiî, op. cit., 3351. Al-Kàmil {Al-Mubarrad, Bayrùt: Mu'assasat al-risàla, 1986, vol. 3, 1079-1080)  question turned into the dispute as to who should be the caliph.Probably the issue was raised because a caliph was required to authorize qisas.The details of that aspect are beyond the scope of this paper.What is important is the fact that the hakams could not reach an agreeable judgement.It became immediately controversial.Even one of the hakams, Abu Mùsâ al-Ash'arî, came out disputing the outcome of tahkim.
Caliph 'All and his supporters rejected the judgement.One group among 'All's followers, later to be known as Khawarij, questioned the validity of tahkim, calling it kufr, because, according to them, God alone was the hakam?^ The later Islamic historians have sUghted the centrality of the real issue of qisas in the tahkim event because of their theological and political focus.The document of the agreement does not support their view.The arbitrators were not asked to choose between 'All and Mu'awiya or to appoint a new caliph.In fact, neither 'All nor Mu'awiya abided by the outcome of the tahkim, namely the selection of Mu'awiya as caliph.While 'Alï protested that the arbitrators had violated their terms, Mu'awiya also did not declare himself Caliph in the presence of 'AIL It shows that the real issue for the hakams was qisas of 'Uthmàn, not the selection of a caliph.
Appeal to the Qur'àn as hakam also meant different things to different people.Earlier, Talha^^ recited the following verse of the Qur'àn during the battle of the Camel: «Have you not seen how those who have received the Scripture invoke the Scripture of Allah that it may judge (yahkuma) between them; then a faction of them turn away, being opposed?(3:23)».Mu'áwiya's supporters recited the same verse when 'AH and his party rejected the outcome of arbitration.^^One faction among 'Alfs supporters argued that 'Uthman's murder was destined by God and, therefore, he was justly killed.^^They blamed 'All for appointing human beings to judge what God had decided.^^In this argument, apparently, the hukm of God did not necessarily refer to the Scripture, but rather to an event of history as a decision of God.Another interpretation was to contrast the injunctions of the Qur'àn with personal desires.Condemning both hakams for deciding on the basis of hawà (personal motive), Hasan and 'Abd Allah b. 'Abbas recited the following verse from the Qur'àn: «So judge between them by that which Allah has revealed, and follow not their desires The Qur'an is clear about the laws of qisàs (retaliation) and diya (blood money), but it does not prescribe any clear procedure for decision in such cases.That is probably why the agreement mentions sunna jàmi 'a ghayr mufarriqa, beside the Qur'an, apparently referring to the common custom, and not to the custom of a particular tribe."^^When the hakams failed to settle the dispute of qisàs, Mu'awiya waited until he was in a position to avenge 'Uthman's blood.He pursued all those who had participated in the murder and got them killed one by one."^^It seems that other people also resorted to this method.Life and property were not safe in the period of Civil War and after.Pre-Islamic practices came to gain strength.The period of Civil War ended with Caliph ' Alfs murder by a Khàrijï.Mu'awiya negotiated with 'Alfs successor and declared himself caliph.He also assumed the title khalifat Allah, God's deputy, probably to institutionalize his authority as Allah's hakam and to define the authority of the other offices as delegated.
Patricia Crone and Hinds"^^ explain that the title of khalifat Allah signified Mu'awiya's assumption of the qadà \ as he himself sat in judgement as God's deputy.The point of real significance is not that Mu'awiya acted as qadi, earlier caliphs had also acted as qadi^.The significant perhaps is the fact that he made the qadi^ answerable to himself or to his governors.It was a real departure from the independent and arbitrary institution of hakam.The institution of qadi now became subordinate to the caliph and its jurisdiction as well as powers were fixed.This institutional development is reflected in the debate in the Islamic tradition about whether Mu'awiya was the first Muslim ruler to appoint qadi^.
On the authority of Malik b.Anas and al-Zuhn, Waki'  These remarks may appear to contradict Wakï"s other statements where he mentions the names of the qàdis in the days of the preceding caliphs.It must, however, be noted that Waki"s statement is concerned more with the change in the institution introduced by Mu'àwiya than whether there were qàdis in the earlier period.
Wakî' clarified that qàdis in earlier period were muftis and hakams rather than qàdis.He describes Ibn 'Abbas in 'All's period as mufti and hakam.^^About 'All's qàdi Abù'l-Aswad al-Du'lï, he remarks: «He was in fact a mufti, qàdis in those days were called mufti.This state of affairs continued until 'All was killed in the year 40.»"*^He further describes how the institution of qàdi became totally subordinate to the caliphs and governors, who appointed them as they pleased,^^ dismissed them summarily^ ^ and sometimes overruled their judgements.^^During the 'Abbàsid period, the caliphs used the qàdis to punish their opponents.^^Caliph Mu'awiya not only subordinated the institution of qàdi to caliphal authority, but also introduced some changes in the rules of procedure.In order to ascertain clear and early decisions, he directed qàdis to oblige the plaintiff to swear solemnly that his or her claim was not false.Ibn Wahb reports, on the authority of Ibn AM al-Zanád, that Mu'awiya asked Sa'îd b. al-'As, Ms governor in Medina, to administer required oaths against an accused and deliver the killer to the caliph for punishment.^"^Al-Zulm, as reported by Ibn Hazm,  Why did Mu'awiya modify the procedure?One may argue that the insecurity during and after the Civil War and the increased number of blind murders might have led Mm to modify the procedure.The normal procedure requiring two witnesses and acquittal of the accused or defendant after the oath would only increase unsolved murders.It was necessary to amend the procedure to control the situation.Before we move further to analyze the judicial practice on this point, let us first briefly overview the institution of yamin (oath).

Pre-islamic tradition
Yamin played a very significant role in daily transactions in pre-Islamic Arabia.^^It meant an oath as well as a vow.It was also used to stress the significance of the truth of a statement.It was sometimes invoked as a curse to invite the wrath of God or super-natural calamity if the statement was false.It was, therefore, often used as a form of ordeal by the hakams.^^ The following verse by Zuhayr b.Salma (d.627) reflects the significance of yamin in the pre-Islamic laws: Three things are decisive in [matters of] right (haqq) yamïn, nifàr (award of possession), and jila' (expulsion) 59

Adab
The adab tradition mentions «The burden of proof is on the plaintiff and the oath is for the defendant» as a proverb or maxim and attributes it to Quss b.Sà'ida.^^Quss b.Sá'ida is not a stranger to hadith tradition.He is remembered as the wisest Arab orator whose speeches at 'Ukaz were repeated admiringly by the Prophet.^^It is not, therefore, impossible that this maxim was remembered by the Prophet at some occasion and became a part of hadith.

Qur 'an
The Qur'án recognized the value of yamïn, but it discouraged its excessive and thoughtless use (2: 225, 5: 92).It did not deny the binding nature of yamïn (16: 91-94) but prescribed lesser punishment of expiation in case of breach (5: 92).The Qur'án, however, allowed the testatory use of oath by the witnesses, even a counter oath by the adversary witnesses (5: 111).The use of oath in place of a witness is not found in the Qur'án.
The Qur'án requires normally two witnesses as evidence (2:282).It is usually difficult to find two eyewitnesses for a murder.In case of blind murder the pre-Islamic tribal practice was qasàma, i.e. to ask the inhabitants of the place where the murdered body was found, to take fifty collective oaths to clear themselves, swearing that they did not kill and had no knowledge of the killer.The Qur'án is silent on qasàma.There is, however, a mention of the repetition or shifting of oaths in the Qur'án in the following verse: «O you who believe!Let there be witnesses (shahàda) between you when death draws near to one of you, at the time of bequest -two witnesses, just men from among you, or two others from other tribe, in case you are campaigning in the land and the calamity of death befall you.You shall empanel them both after 59 Ibid, 509.
^° See Abu' 1-Fadl al-Maydànï, Majma' al-amthal (Cairo: Matba ' al-Sa'áda, 1959), vol.1,111.^^ Jalâl al-Dm al-Suyûtî, Al-Lu'àli al-masnü'a fi 'l-ahàdith al-mawdü'a (Cairo: Husayniyya, N.d.), vol. 1, 183-192;and Ibn Qutayba, Al-Ma'àrif {Cmo\ Dar al-kutub, 1960), 61. the prayer, and, if you doubt, they shall be made to swear (yuqsimàn) by Allah, saying: We will not take a bribe, even though it were (on behalf of) a near kinsman nor will we hide the testimony (shahàdà) of Allah, for then indeed we shall be sinful.»But then, if it is afterwards ascertained that both of them merit (the suspicion of) sin, let two others take their place of those nearly concerned, and let them swear (yuqsimàn) by Allah, saying: Verily your testimony is truer than their testimony and we have not transgressed, for then indeed we should be of the evil doers.
»Thus it is more likely that they will bear true witness or fear that after their oath the oath of others will be taken (turadda aymànun)» (Qur'án 5: 106-108).
Although the verse refers to the subject of property, not to murder or qasama, yet the fact that in case of suspicion the oaths are shifted to two other jurors indicates the existence of the practice.We will deal with the question of shifting of oath during our analysis of the judicial practice.

Hadith
The Hanafí jurists^^ refer to "The burden of proof is on the plaintiff and the oath is for the defendant" as a Prophetic hadith.Most hadith collections report it as a saying of the Prophet.^^Ibn Qayyim, however, finds this hadith weaker than the ahàdith supporting oath and witness (yamin ma ' al-shàhid).He adds, «Besides, none of the six collections of hadith report this hadith».^'Àrif al-Nakidî, a modern jurist, has remarked that it was a pre-Islamic Arabian maxim and the hadith only states it as a current practice in those days.^^ The hadith literature reports frequently the statements by the Companions of the Prophet or by the later generation that the Prophet decided cases on the basis of one witness and oath.^^Shâfi'î jurist al-Muzam refers to ten ahàdith, all of them statements by the first or second generation of early Muslims saying that the Prophet decided cases on the basis of one witness and oath.^^Most of these haciïth are reported by Medinese narrators.It is, however, difficult to conclude from this fact alone that it was a Medinese tradition.Malik narrates the hadith «The burden of proof is on the plaintiff and the oath is for the defendant», adding an exception: «except in case of qasàma».^^Malik refers to the Medinese practice, distinguishing qasàma from other cases.It appears that according to the practice in Medina, it was the defendant who was asked to take oath.It did not differ from the so-called Iraqi practice.The practice differed in case of homicide.In qasàma, Malik explains, the accusers were asked to take oath first.He uses phrases like the following to describe this practice: «the agreed view among us», «the view on which the old and new leaders (of opinion) agree», «this is the practice (sunna) on which there is no difference of opinion among us», and « the continuous practice has been to begin oath with the plaintiff (ahl al-dam)».^^ The Sunna of the Prophet in case of qasàma is recorded in the hadith literature with reference to an AnsM found murdered in a Jewish quarter in Khaybar.According to al-BukhM some Ansañs went to Khaybar for business.In Khaybar they disbursed into the town.On their return they found one of them murdered.They came to the Prophet claiming that the Jews had murdered their kin.The Prophet asked if they had any evidence.They replied in negative.The Prophet said that in that case the Jews would be asked to take oath.The AnsMs objected that they could not trust the Jews.The Prophet then asked them if they were ready to take fifty oaths swearing that the Jews had killed their relative and receive the blood money.They declined that they could not do so because they did not witness the murder.The Prophet dismissed the case but paid them the blood money because he did not want the blood go unavenged.^^In another version the Jews were asked but they denied that they had killed the AnsM.

Jurisprudence
Modem Islamic legal scholarship interprets the difference in the hadith literature as different local traditions appearing in the form of hadith.They consider the hadith: «The burden of proof is on the plaintiff and the oath is for the defendant» as an Iraqi tradition, probably because the Iraqi Hanafí jurists rely on it as a hadith.They regard the ahàdith about yamïn ma ' al-shàhid as Medinese tradition.We have referred to Malik's view that refers to both ahàdith as Medinese practice.According to him the practice on the first hadith was limited to cases related to property while the practice in criminal cases allowed the plaintiff to take oath.He particularly refers to qasàma cases where it was allowed to begin oaths with the accusers.
Regarding the Iraqi tradition, it is interesting to note that al-Sarakhsi refers to the hadith of Oath almost as an apology for the Hanafí position.He opens his discussion on evidence with reference to the Qur'ànic verse requiring two witnesses and to a hadith emphasizing that a person must testify only when he had himself witnessed the event as clearly as the sun.^^He finds it difficult to rationalize oath as evidence, but accepts it because the rule is derived from a hadith of the Prophet.This is why he restricts it to only a negative role.He confines its application within its literal meaning, not allowing its use for the plaintiff.The Hanafís are emphatic in their rejection of the doctrine of yamin ma' al-shàhid, which, as we have seen, they consider a bid'a, an addition, introduced by Mu'awiya.
The doctrine of yamïn ma ' al-shàhid has been a subject of debate among the jurists and traditionists.While Hanafí jurists find it contrary to the Prophet's hadith about yamin belonging only to the defendant, the traditionists argue that the doctrine was based on the Sunna of the Prophet.
Al-Shâfi'ï cites a number of hadiths to prove that it was a Sunna of the Prophet.'^^Ibn Hazm is convinced that it was a widely accepted practice based on sunna and athàrP However, reports about this hadith cited in the hadith literature and those mentioned by al-Shàfî'ï, Ibn Hazm and Ibn Qudama^"^ do not refer to concrete cases.Other references that mention concrete examples do not belong to this procedure.For example, in the oft-quoted hadith on this point, dispute on land between a Hadramî and a Kindï was in fact decided on the basis of defendant's oath in the absence of any evidence from the plaintiff.^^It is significant to note that the jurists do not derive their rules from the actual practice of the Prophet, but rather deduce them from the presumptions in the narrative of the ahàdith.For instance with reference to the qasàma case M. KH.MASUD AQ.XX, 1999 mentioned above, the facts are that neither the Ansârïs nor the Jews took oaths, accusative or the purgatory.The diya was not paid by the Jews, but by the Prophet.Yet the traditionists as well as the jurists conclude that the Jews were obliged to pay the diya^^ and the accusers were asked to take oaths.It is in fact casuistry method of deriving rules that the jurists employed.The method does not deduce rules from the actual practice but from the implication or presumption of the rules.For instance the Prophet's offer to the Ansârïs: «You take oaths and you will receive diya» becomes the basis of deduction, even though the Ansàiïs did not take oath and the judgement was not based on their oath.The traditionists and the jurists deduce the rule from the implication that the sentence allowed the accusers to take oath and that diya was to be paid on its basis.Al-Sarakhsî, while refuting the rule, does not question the method.Rather he says that the sentence might have been wrongly reported.^'^

JUDICIAL PRACTICE
We now turn to Waki' for an overview of the judicial practice and procedural laws.As said earlier, we have found 25 cases related to our question in Waki"s Akhbàr aUQudàt.It appears that the qàdis adopted five procedures in these cases (See details in Annex.2).A brief analysis of these cases is given in order to understand the judicial practices.
• Procedure One: The qàdi demands two witnesses from the mudda 1 (plaintiff) and decides the case in his favour.According to this procedure only two witnesses constitute evidence.Waki' mentions three cases (cases 7,8 and 17,Annex. 1) in which the qàdis insisted on the production of two witnesses and refused to accept oath by the plaintiff with no witness (case no.7), or with one witness (cases number 8 and 17).The three cases belong to Iraq, two (numbers 7 and 8) to Kùfa and one (number 17) to Basra.None of these cases belong to caliph Mu'awiya's qadis.
• Procedure Two: The mudda 1 has no witness.The mudda 1 'alayh (defendant) takes oath of denial.The qàdi decides in favour of the defendant.
In two similar cases (numbers 3 and 12) defendants take the oath saying that they had no knowledge of the defect in the merchandise (slave girls in both cases) at the time of the sale.In case number 3, the defendant also names the person from whom he had purchased the slave girl.His statement thus makes him also a plaintiff.The qàdi treats his oath as an accusative oath against the third person to which the slave girl is returned, apparently canceling the two deals of sale.In case number 12, the defendant offered to accept plaintiff's charges if he stated them under oath.The qàdi al-Husayn al-Kindi, in Walïd II's reign, observed that the defendant, apparently a cunning merchant, retumed the oath knowing that the plaintiff would hesitate to do so.The qàdi, therefore, insisted that defendant must take oath.
• Procedure Three: The plaintiff has one witness and offers to take oath in addition to witness.The qàdi decides in favour of the plaintiff.We have discussed this procedure above as the doctrine of yamin ma ' al-shàhid.Most of the cases in our sample were decided according to this procedure.Out of ten, Wakï' reports three cases with details of facts and seven without facts.Among these ten cases, three (numbers 4, 19, 20) belong to Caliph Mu'awiya's period (661-680), three (numbers 21,22, and 23) to CaMph 'Umar two (numbers 13,15) to Caliph al-  and one each to those of Hârùn (786-809) and al- .Mu'àwiya's qàdis who decided according to this procedure were Shurayh (d.699), Zurára b.Awfa and Abu Salma b. 'Abd al-Rahman b. 'Awf, respectively in Küfa, Basra and Medina.
Schacht and Dannhauer regard references of this doctrine to these qadis as attempts by Medinese jurists to show that the practice was widespread even in Iraq.Looking at the cases reported by Wakî', we find that only two cases belong to Medina and one to Mecca.Even if we include Egypt, only four cases belong to the Medinese tradition.Six cases belong to other regions, five of them to Iraq.It is, therefore, difficult to say to which juristic tradition the Iraqi qadis belonged: Iraqi or Medinese?It is perhaps more meaningful to say that qadis were not dependent on either of these traditions.It is also too early to categorize qadis in Mu'awiya's period according to these traditions which developed later.
• Procedure Four: The plaintiff has no witness.He offers to take oath.The qàdi decides in favour of the plaintiff.
Wakî' refers to five cases (numbers 5, 6, 10 and 14 and 16) in which this procedure was employed.Apparently this procedure is not logically justified.We have not found this case reported in any other source.We cannot doubt Wakï"s report as Abu , an expert on Quraysh families, also mentions that Mus'ab was accused of a murder and acquitted.^^ He adds further that the three accused were imprisoned until they agreed to take oath.There are, however, some puzzling questions.First, why did Ibn Zubayr and Ibn Azhar hesitate to take oath?Second, why did the accused hesitate to take oath?In an above mentioned case of a murder in Khaybar in the days of the Prophet, the accusers and the accused both similarly hesitated to take oath.It shows that they did not take oath slightly, even though it weakened their claim.The more crucial question, however, is why Mu'awiya hesitated to ask the accused to swear?Why did he feel frustrated?If qasàma was in practice and if it required the accused to take fifty oath why Mu'awiya found himself in an impasse?His frustration means either that the qasàma was not a regular practice or that the current practice required the accusers to take oath.When the accusers declined to do so, Mu'awiya did not know what to do.He finally did something new, namely, he shifted the responsibility of oath to the accused.That is how Waki' explains Mu'awiya's action.WaM' characterizes CaHph Mu'awiya's method of judgement as the procedure of radd al-ayrmn (shifting of oaths).He remarks that Mu'awiya was first to do that, explaining that «He retumed {radd) them [i.e.50 oaths], one-third each to the three [accused].Mu'awiya was the first to return the oaths (radd al-aymàn).It never happened before.If there was one person less than fifty, the others took additional oaths to complete the number.If there was one person less, diya was imposed...».^^ Waki' seems to be referring to the current Arab tribal procedure of qasàma.
It appears that normally, the defendant took purgatory oath to deny the claim of the plaintiff.If he declined to take oath, the case was generally decided for the plaintiff.Sometimes the plaintiff was asked to take accusative oath when the defendant had declined to do so.This procedure was called radd alaymàn because the oath was shifted to the plaintiff.
In case of murder, the procedure differed.According to Malik, as we have said above, the practice in Medina was to ask the accusers to take oath.That probably explains why Mu'awiya felt frustrated when the accuser declined to take oath.The only way out for him was to shift the oaths to the accused.
According to Waki', Mu'awiya was the first to do that.In the frequently cited precedent^^ of murder in Khaybar in the Prophet's period, when the M. KH.MASUD AQ.XX, 1999 plaintiff AnsMs and the accused Jews both declined to take oaths, the Prophet did not oblige either party to take oath and decided for the payment of diya, which he paid himself.
One may question Wakî"s statement and its implications that Mu'áwiya set aside the Medinese practice or that he introduced a new practice which later became Iraqi tradition.These questions need further investigation.Presently we are concerned with a more significant matter, i.e.Wakï"s method of inference from this precedent.It differs from that of the traditionists and jurists.While he calls it a precedent for radd al-aymàn, the traditionists and jurists describe it a case of qawad fi'1-qasàma.
As we have pointed out above, the jurists and traditionists tend to deduce rules also from the implications of the statements made in the precedent.A critical note by 'Abd al-'Azîz al-Maraghi,^"^ the editor of Akhbàr al-Qudàt, illustrates this point.
In a long footnote, disagreeing with Waki', al-Marâghï contends that this was a case of qawad, not radd al-aymàn.In his support, he refers to al-Sarakhsi, al-Bayhaqî, Ibn Battal and others who state that Mu'áwiya was the first to decide on the basis of qawad in Islam.Al-Marâghî says that it was a common practice to shift the oath to the plaintiff.Only the Hanafís disallowed it.It would be, therefore, meaningless to say that Mu'áwiya was the first to do something, which was a common practice.
It may be observed that al-Marághí's remarks reflect the casuistic method of the jurists because he relies for his support on the arguments developed by some Hanafí jurists.In fact, no qawad was imposed in this case.Apparently this conclusion is drawn from the implication of the application of qawad in Mu'áwiya's offer to Ibn Zubayr that he would be entitled to qawad if he took the oath.Had Ibn Zubayr taken oath the qawad would have been applicable.The fact, nevertheless, is that Mus'ab was acquitted after taking the oaths and no qawad was applied.
In his support, al-Marághí refers to some Hanafí jurists who argued that Mu'áwiya practiced al-qawad fi'1-qasàma and that he was the first to do that.Let us see how the Hanafí jurists come to this conclusion.
Al~Sarakhsî,^^ Ibn Battál^^ and al-Bayhaqi^^ argue that Mu'áwiya introduced qawad in Islam.Al-Sarakhsi makes this statement in a specific context.In order to refute the traditionists' claim that the Prophet and the first two caliphs practiced al-qawad fi'1-qasàma, al-Sarakhsî argues that it was Mu'áwiya who introduced the practice.
Al-Sarakhsî explains that the Umawi caliphs practiced qawad and qasàma.'Umar b. 'Abd al-'Azïz inquired some scholars who told him that it was the practice of the Prophet and the early Caliphs.The Caliph asked Abu Qulaba.He refuted that the Prophet or the Caliphs ever applied qawad on the basis of qasàma.Abu Qulába gave examples from the actual practice of the Prophet.Apparently, Abu Qulaba distinguished the actual practice of the Prophet from the inferences from the sunna to which Caliph's jurists referred.Al-Sarakhsï cites Abu Qulaba's remarks approvingly but he does not cite any case from the actual practice of Mu'áwiya to prove that he introduced qawad and qasàma.He cites al-Zuhri saying that Mu'áwiya was the first to apply qawad on the basis of qasàma.We have already referred to al-Zulm's statement that Mu'áwiya was the first to employ the method of yamin ma' al-shàhid.It appears both statements refer to the cases where qawad was offered to the accusers if they took oaths.Since the Hanafí jurists require the accused to take oath to clear themselves they would not call the procedure of asking the accused to take oath as that of shifting the oath.
It is evident from the special chapters in the hadith books on the question whom should the qàdi ask to begin taking oath that it was a crucial issue of procedural law in those days.Naturally, therefore, shifting of oath to the other party must be an extraordinary practice.Contrary to al-Marághí's argument, therefore, Wakî"s remarks are quite significant in this context.

CONCLUSION
With reference to the doctrine of yamin ma ' al-shàhid, we have seen that its growth against the historical setting of the Civil War period is quite understandable.This was a period when the whole Muslim society seems to be engaged in the questions relating qisàs, qawad, qasàma and hakam.These discussions were taking place with reference to the role of the Qur'án and the common sunna.The rules of procedure were also part of the debate.The need for changes regarding the institution of hakam and modifications in the rules of procedure was quite obvious.The scattered statements of scholars close to that period supported by Waki"s remarks reinforce this view that Mu'áwiya attended to this need.Secondly we also find that this particular doctrine appears to have been introduced to respond to this need.Viewing the judicial practice of this doctrine we find it difficult to describe it as a Medinese practice because qadis used it more in other areas than in Medina.Its characterization as a Medinese tradition is possible only if we confine to the writings of jurists and traditionists.
We therefore find it quite meaningful to study the judicial practice to understand the growth of Islamic law.It gives us a better understanding of the nature of Islamic law.Tyan had already warned against the use of jurists' writings about qàdis to appreciate the role of qadis.In our view there is a greater need to follow Tyan's advice to write a critical history of the Islamic law on the basis of judicial practice.

Table of Cases
[Relating to the procedure of witness and oath as reported by Waki' (1947) Ibn Zubayr declined to take the oaths, as he only knew that Ibn Habbàr's body was found in the place where the three had gathered together.Mu'àwiya asked Ibn Azhar to declare under 50 oaths that the accusations against his client Mus'ab were false.He also declined to swear, as he also did not know definitely.Mu'àwiya faced an impasse and finally decided: «The only way for me is to refer the fifty oaths back to the three accused, and then they pay the blood money».He shifted the oaths to the three.
with a defect.Plaintiff [(A), apparently, the second buyer,] claimed that the slave girl had a physical defect.The defendant [(B), apparently, the first buyer] explained that he had purchased the girl from so and so [(C), the original seller].
The Qàdî asked the defendant to take oath that he did not know the defect at the time of sale, and he did not conceal it.After the oath the slave girl was returned to the original seller [(C)], because he had sold her with the defect.[ Wakï',11/334].Modem critical studies of Islamic law have noted the need for the study of the procedural law in Islam and the role of the judicial practice in its formation.Generally, it is believed that qàdi^ had freer hand regarding witness and method of proof in the early period, but gradually it became more restrictive and rigid.Scholars have suggested various dates ranging from the second half of the seventh century (Tyan) to the eighth (Schacht, Dannhauer), to the eleventh century (Coulson) for this development.These scholars have treated the doctrine of yamin ma * al-shàhid (oath with one witness) as an exception to the Shariah procedure which, according to them, strictly and mechanically adhered to the doctrine of two witnesses.According to them, this doctrine emerged under local [Schacht (Egypt) and Dannhauer (Madina)] or Jewish (Patricia Crone) influence.
The present paper argues that the doctrine must have grown against the historical setting of the Civil War (Tahkîm period) when the questions relating qisàs, qawad, qasàma and hakam were discussed frequently with reference to the role of the Qur'àn and the common sunna and the need for reforming the institution of hakam was felt.On the basis of scattered statements of scholars close to that period supported by WaM"s remarks and we conclude that Mu'âwiya introduced this method to reform the institution of hakam.The judicial practice of this doctrine also confirms that it was too widespread to call it a Medinese or local practice as claimed by the classical jurists, traditionists and some modem scholars.
(c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) http://al-qantara.revistas.csic.esAQ.XX, 1999 THE PROBLEM OF YAMIN MA argues that Mu'awiya b.Abî Sufyán was the first caliph to appoint qàdis.The caliphs before him dealt with public affairs directly themselves."*^Other scholars have generally disagreed with Malik on this point.Ibn 'Abd al-Barr and Ibn Rushd find Malik's statement contrary to the well known facts that qadis were indeed appointed by the Prophet and by the four caliphs.They explain that Malik meant to say that Mu'áwiya was the first caliph to appoint qàdis in the capital (hadra)^^ This does not, however, explain Wakî"s view.Waki' says that Qatada's statement naming six Companions of the Prophet as qàdis was debatable.He refers to al-Zuhri saying: «Abu Bakr and 'Umar had no qàdis until thefitna (Civil war).Thence Mu'áwiya appointed qàdis.»"^^Referring to caliph 'Uthmàn, he observes that it was not definitely known that 'Uthmàn appointed qàdis in Medina till he was killed in Dhü'1-Hijja in the year 35."*^ says that tMs innovation (bid'a)[i.e. the procedure of oath and witness combined in one person] was introduced by Mu'awiya.^^A Hanafí jurist, 'Abd al-'Aziz al-BukhM, also states that Mu'awiya was the first to introduce oath and witness.^^Ibn Hazm and 'Abd al-'Aziz al-Bukhârï have not provided further details.Wakî' names several qàdis of Mu'awiya who used tMs procedure.Waki' reports one such case decided by Mu'awiya to which we shall return shortly.

] 1 .
Ubayy b.appointed by.Medina.Dispute about a wall.The plaintiff had no evidence.TheQàdî asked the defendant to take oath.'Umarswore as follows: «I swear by Allah except whom there is no god that Ubayy has no right in my land»[Waki',.