Navigation – Plan du site

AccueilNuméros140Première partieA l'ombre du cadi : des modes alt...Bringing arbitration (taḥkīm) and...

Première partie
A l'ombre du cadi : des modes alternatifs de résolution des litiges

Bringing arbitration (taḥkīm) and conciliation (ṣulḥ) under the qāḍī’s purview in Mālikī al-Andalus (10th to 12th centuries C.E.)1

Étendre la compétence du cadi sur l’arbitrage (taḥkīm) et la conciliation (ṣulḥ) selon les juristes mālikites d’al-Andalus (xe-xiie siècle)
Delfina Serrano
p. vol 140, 73-100


Cet article examine le discours des mālikites sur l’arbitrage et la conciliation, à une époque où al-Andalus connaissait une série de développements politiques, intellectuels et idéologiques qui affectaient en profondeur les idées reçues, la politique judiciaire, ainsi que la manière dont les juristes percevaient leur propre rôle social de médiateurs et de pacificateurs. Nous nous appuyons sur les œuvres de deux des plus grands fuqahā’ mālikites de la période : al-Bājī (403-474/1013-1081) et Ibn Rushd al-Jadd (450-520/1058-1126). Il apparaît tout d’abord que l’arbitrage était reconnu comme un mode valide de résolution des conflits, bien qu’inférieur et subordonné à la judicature du cadi. On relève par ailleurs que le recours aux pratiques répandues de conciliation fut limité en droit, surtout pour des raisons éthiques.

Haut de page

Texte intégral

  • 1 This paper collects results both from a research carried out at the Institute of Islamic World Stud (...)

1It may be assumed that the resort to arbitration and conciliation as conflict resolution strategies alternative to formal adjudication in a court (qaḍā’; ḥukm) was very frequent, that it used to take place outside the control of religio-legal and governmental institutions, and was informed by customary practice rather than by the rules of fiqh. Attempts at testing the validity of that image in the specific place-time context of al-Andalus, where the Mālikī school of law was predominant, are hindered, however, by the well-known lack of relevant archive materials, as well as by the many limitations facing the reconstruction of everyday life from the sources at our disposal, and by their near silence concerning actual arbitration processes, in contrast with their coverage of ṣulḥ which, as we will see below, is fortunately much better known.

2Research on the historical development of Mālikī doctrine on conciliation and arbitration is not abundant either. Drawing on three outstanding references for the school doctrine, namely, the Mudawwana by Ifrīqian scholar Saḥnūn b. Saʻīd al-Tanūkhī (d. 240/854); the Risāla by Ibn Abī Zayd of Qayrawān (d. 386/996); and the Bidāyat al-mujtahid by the Cordoban Ibn Rushd al-Ḥafīd (or Averroes, d. 595/1198), Peter Scholz (1997: 440-447) examines the issue of ṣulh in the context of Mālikī procedure law. The settlement of marital disputes in al-Andalus and contemporary Libya has received a fair amount of attention (Fierro, 1985 & 2006; Layish, 1991: 181 & 1995). Maribel Fierro has even pointed to the possibility that Mālik b. Anas’ stance on the outcome of the arbitration of Ṣiffīn may be connected with the ultimate Umayyad decision to opt for Mālikīsm in al-Andalus (Fierro, 2006: 329). Other Islamic law scholars, among them Christian Müller and David Powers, have drawn attention to the qāḍī’s frequent involvement in bringing about amicable settlements between parties in order to avoid formal adjudication (Müller, 1999: 212-214; Powers, 2002: 94-140), while Camilo Gómez Rivas has stressed the role of mediation as performed by Mālikī qāḍīs and fuqahā’ in the economic and social development of the Islamic West under Almoravid rule (2015: chap. I). Meanwhile, a recent essay on qāḍīs’ power of decision and execution includes quite a comprehensive approach to Mālikī doctrine on ṣulḥ and taḥkīm, providing a series of interesting cases of actual involvement by qāḍīs in conciliation and mediation processes (Hentati, 2007: 182-193). The sources quoted by the author reflect the relevance of Ibn Abī Zayd al-Qayrawānī’s al-Nawādir wa-l-ziyādāt to trace the emergence, in 4th/10th century Ifrīqiya, of a distinctively Mālikī jurisprudence on arbitration through scattered opinions transmitted in earlier sources, though the relevant conclusion to be drawn from the chronology of this doctrinal development seems to have escaped the author.

  • 2 “Judicial” is used here to refer to the ordinary dispensation of justice as performed by a recogniz (...)

3The pertinence of focusing on a specific historical context to elucidate the relationship between “official”, “extrajudicial”2 and “informal” justice in pre-moder Islamic societies is beyond discussion, but even a compact frame like al-Andalus is too broad for a general overview of Mālikī doctrine and practice on arbitration and conciliation, given the abundance of relevant sources, on the one hand, and, on the other, the lack of a solid corpus of secondary bibliography to rely on in order to study them. For these reasons, we are going to concentrate on one of the most interesting periods in the history of al-Andalus from a political, religious and intellectual point of view. The period running from the end of the 10th to the second half of the 12th centuries C.E. saw the reception and subsequent assimilation of the rational disciplines of uṣūl al-fiqh (legal methodology) and kalām (discursive theology), and the emergence of Ẓāhirism as a serious alternative to the prevailing Mālikī legal school. These developments were, in part, a response to the political and spiritual challenges posed to the local Muslim community by the political disintegration subsequent to the fall of the Umayyad caliphate of Cordoba, and to fear of losing control of the land at the hands either of north Iberian Christians or of north African Muslims, the latter threat having finally materialized in the Almoravid invasion of al-Andalus from 479/1086 onwards.

  • 3 He performed qāḍīship for several ṭā’ifa kings of al-Andalus, though only in minor towns like Orihu (...)
  • 4 His career and his integration into the world of Andalusī men of science are closely connected with (...)

4Our aim is thus not to exhaust the study of Mālikī doctrine on ṣulḥ and taḥkīm but to observe how the above-mentioned political and intellectual developments affected relevant legal discourses, and determine whether the resulting corpus of opinions was effective in shaping legal practice for the promotion of peace, and the prevention of suffering and unnecessary harshness. This will be illustrated through the works of two of the most relevant Mālikī fuqahā’ of the period described: Abū l-Walīd Sulaymān b. Khalaf al-Bājī (403-474/1013-1081)3 and Abū l-Walīd Muḥammad b. Aḥmad Ibn Rushd al-Jadd (450-520/1058-1126)4. The former produced the first comprehensive Andalusī treatment of Mālikī law on arbitration and conciliation –the corresponding legal practice having been duly Islamized and “Malikicized” much earlier on, as will subsequently be shown. Al-Bājī’s lifetime coincided with the political disintegration of al-Andalus into party kingdoms (ṭawā’if) and, consequently, the end of a unified judicial policy and also, probably, that of a legal system prone to fostering or to tolerating modes of arrangement alternative to the qāḍī’s court and authority. Together with Ibn Rushd al-Jadd, al-Bājī represents the assimilation of uṣūl al-fiqh and kalām methodology in Western Mālikīsm which they both strived to apply to the legal traditions in force and to previous authoritative Mālikī scholarship. Most of Ibn Rushd al-Jadd’s public career evolved under the Almoravids, who managed to reunify al-Andalus politically and made a series of interventions in the judiciary which, at certain points, collided with the interests of the legal scholars, triggering a series of reactions on the latters’ part, in which Ibn Rushd played a prominent role (Serrano, 2014: 254-258).

  • 5 On the centrality of the legal commentary in both al-Bājī’s and Ibn Rushd al-Jadd’s intellectual ac (...)

5We will utilize the opinions collected or issued by both jurists as an argumentative thread because of their relevance and the trust these figures placed in legal methodology and kalām – two disciplines they mastered – to bring about a consistent rationally substantiated jurisprudence equipped to meet the criticisms of the Ẓāhirīs and the Shāfiʻīs, evident in their emphasis on commenting and reviewing the extant manuals of the school doctrine (e.g. Mālik’s Muwaṭṭa’, Saḥnūn’s Mudawwana, and al-ʻUtbī’s Mustakhraja, also known as al-ʻUtbiyya)5. However, in order to complete and refine our understanding of 11th and 12th century C.E. Mālikī jurisprudence on arbitration and conciliation, other previous, contemporary and later relevant authors will also be taken into account (e.g. Ibn al-ʻAṭṭār, Ibn Abī Zamanīn, Ibn ʻAbd al-Barr, Abū Bakr Ibn al-ʻArabī, Averroes, and the Ẓāhirī Ibn Ḥazm).

Taḥkīm is legitimate but qaḍā’ ranks higher and is more excellent: Al-Bājī

6The following presentation of al-Bājī’s doctrine on arbitration and conciliation draws on two of his works: a commentary to Mālik’s Muwaṭṭa’ entitled Kitāb al-Muntaqā (“The select”), sharḥ al-Muwaṭṭa’, and a compilation of legal practices prevalent in al-Andalus, meant as a manual for the training of muftīs and qāḍīs. Al-Bājī’s view on conciliation cannot be considered the most elaborate or complete Mālikī treatment of the subject up to his time. His main contribution dwells rather on the unprecedented and carefully detailed attention he pays to taḥkīm, which he approaches keeping a keen eye on the legal practice of his time and place, combining the concerns of the legal interpreter with those of the qāḍī and the social reformer.


7Arbitration is addressed as a subsection of the section on “issuing a judgment at the request of a party” (al-qaḍā’ fī l-daʻwā) included in the chapter on procedure and adjudication (qaḍā’) of the Muntaqā (VII, 215-219). It opens with a general statement outlining the main differences between arbitration and jurisdictional adjudication, namely the parties’ freedom to choose their judge, the latter’s independence and the binding force of his decision:

“If two men appoint an arbiter to settle their dispute, his decision will be legitimate (qaḍā’u-hu jā’iz). This is the view of Mālik as transmitted in the [Kitāb] al-majmūʻa”.

8To further qualify the statement, al-Bājī adds the maxims attributed to other relevant Mālikīs and the divergence of opinions among them. Ibn al-Qāsim (Egypt, d. 191/813) is then reported to have said that this is so, even if the arbiter’s decision raises the disagreement of the jurists or that of the qāḍī, his decision being binding (māḍin) unless its wrongfulness (jawr) becomes clear. This opinion was subscribed by Saḥnūn (Qayrawān, d. 240/854) as reflected in the book by his son (d. 257/871), i.e. the Kitāb Ibn Saḥnūn. According to al-Bājī, the binding force of the arbitrator’s decision relies on his having been selected by the litigants to decide between them on the basis of his discretion (bi-mā yarā-hu), and on the parties’ commitment to abide by it freely, of their own mutual accord and in agreement with him.

9Ibn al-Qāsim’s, Muṭarrif’s (Medina, d. 220/835), Aṣbagh’s (Egypt, d. 220/845), Ibn al-Mājishūn’s (Medina, d. 214/829) and Saḥnūn’s opinions as transmitted in Ibn ʻAbdūs’ (Tunis, d. 261/875) Kitāb al-majmūʻa and Ibn Ḥabīb’s (al-Andalus, d. 238/852) Wāḍiḥa are called to illustrate the question of the compromise struck by the parties:

“If both [litigants] agreed in appointing him (ḥakkamā-hu) [as arbitrator] but, having presented their evidence before him, one of the [litigants] realizes that he would have won [the dispute had he claimed his right before the ordinary courts] (badā li-aḥadi-himā), the arbiter’s authority to settle their dispute [remains intact] and his decision must be considered valid.”

10According to Muṭarrif,

“the party [wishing to withdraw] can claim his right before the ordinary courts (al-nuzūʻ) if the arbitrator (ḥākim) has not yet started to examine the dispute. However, once he has started to examine any aspect of their case, it is no longer possible to withdraw, both [litigants] being obliged to proceed further (al-tamādī) with the arbitration process”.

11In Aṣbagh’s words, this is so even

“if the other litigant does not object to his [opponent’s wish] to withdraw and turn to the qāḍī [instead]; the same applies if he has appointed an agent (wakīl), or dismissed him”.

12In Ibn al-Mājishūn’s opinion, however,

“withdrawal is impossible at any moment, the arbiter’s decision being as binding for them as if it had been taken by the ṣulṭān himself in the exercise of his prerogative to [arbitrate his subjects’ disputes and to] look after the rights of the absentees.”

13Yet Saḥnūn held for allowing withdrawal as long as no decision had been taken by the arbitrator.

14Al-Bājī explains that the opinions of Ibn al-Qāsim, Muṭarrif and Aṣbagh are grounded in the view that arbitration does not bind the parties (lā yulzam bi-l-taḥkīm) since it follows the rules of mandated agency (wakāla) in two respects: (1) the arbiter is a private judge while the official delegation of judicial competences (wilāya) is general; (2) his decision is valid only with the parties’ agreement, both the favoured and the disadvantaged, a fact fitting the definition of mandated agency. Official delegation of judicial competences excludes the possibility that the judge be selected by the parties (taḥkīm al-mutakhāṣimayn) since, according to Ibn al-Mājishūn, the wilāya entails exclusive right to impart justice to the litigants and does not require the parties’ satisfaction with its results, whereas the main object of mandated agency is the fulfilment of the task commanded by the principal (al-muwakkil) at his satisfaction.

15Al-Bājī observes that the point of Ibn al-Qāsim’s view is that the arbiter’s decision acquires binding force from the moment he starts to work in the resolution of the conflict (yalzim bi-shurūʻi-hi fī l-naẓar bayna-humā) and that the obligation to obey his pronouncement is not based on the similarity between arbitration and mandated agency which Aṣbagh had argued to draw a parallelism with the principal who is not allowed to dismiss the agent once the latter has started to litigate [in the name of his principal] before the qāḍī, but is entitled to do so before. Conversely, Ibn al-Mājishūn’s point is that the obligation to accept the arbitrator’s award relies on his appointment and on his acceptance of the assignment, because he imparts justice to two individuals, as in arbitration in cases of marital dispute. The explanation of Saḥnūn’s opinion, then, is that in ordinary litigation (al-khuṣūma ʻinda l-qāḍī) the execution of the judgment does not depend on the parties, but is the qāḍī’s duty (hādhā lāzim la-hu) while the agent cannot operate before a [qāḍī] different from the one [before whom he was authorized to litigate in the name of his principal], the arbiter being in the same position as the agent concerning the mandate to oversee the interests of his principal; the latter can dismiss him at will without the qāḍī’s consent. Al-Bājī does not side with one or the other opinion and submits the right answer to God for “He knows best”.

16Subsequently, he addresses a case when the parties appoint more than one arbitrator to look after a single dispute. The treatment of this contingency serves the purpose of further stressing the similarity between arbitration and mandated agency, and the difference between arbitration and adjudication proper (qaḍā’). In the opinion of Saḥnūn, transmitted in his son’s book, if one of them reaches a conclusion without the other, the latter is not obliged to comply with it. However, the Majmūʻa collects another opinion by Ibn Kināna authorizing the parties to follow the advice of one of the arbitrators, provided they both agree to it. The reason is that, whether they have agreed to appoint two arbiters or only one, they are not obliged to observe the decision taken by just one of them without the other’s agreement. Likewise, if a man appoints two agents to buy a piece of cloth for him or to pronounce repudiation of his wife in his name, but only one of them carries out the task, the principal is not committed to his action. However, had the parties agreed to [resort to several arbiters] and to abide by the decision taken by one of them, to the exclusion of the others, it would be as correct as if they had appointed only one arbiter, or as if only one decision had been taken.

17From this point on, al-Bājī’s discussion mixes elements of arbitration and adjudication (qaḍā’), not so as to blur the limits between the two but rather to present what constitutes the main purpose of this subsection, namely, to invalidate the possibility that in analogy with the admission of two or more arbitrators – as in the case of the two arbiters engaged to settle marital disputes (al-ḥakamayn bayna l-zawjayn) – dual or multiple qāḍīship be admitted:

  • 6 During the ṭā’ifa period, Onda – in present day Spanish province of Castellón – passed from the han (...)

“It is not permitted for two men to be invested (lā yajūz an yuwallā) with the power to issue judgment (qāḍā’) on the condition that they adjudicate every single case together so that the witnesses declare before each of them, and that the judgment is not executed except with their mutual agreement, nor is it possible for two qāḍīs to agree to decide a single case (an yanẓurā fī qaḍiyya wāḥida) on the condition that their judgment is not to be executed until they both agree to it. There is no divergence of opinions among Muslims regarding that; indeed, judicial praxis (al-ʻamal) has followed this [unitary conception of qāḍīship] everywhere from Muḥammad’s time until the present. Nothing of the like is known to have ever been in force until” – and here comes a precious detail about the author’s most immediate entourage – “this blameworthy innovation (bidʻa) appeared in the Andalusī province of Unda (Spanish Onda) where the prerogative to appoint qāḍīs was assumed (tawallā) by someone who, out of sheer ignorance and exceeding the scope of his functions, appointed three [qāḍīs over a single judicial seat] so that no decision could be executed without the agreement of the three of them6. I got to know that the witnesses testified before the first of them; he wrote upon the register of each witness’s testimony, ‘he testified before us’. Subsequently the same process was repeated before the second and the third judge. When I notified them that, in my view, they were acting reprehensibly, one of them withdrew but the other two insisted and stood by their error (fa-aṣarrā wa-tamādayā ʻalā ḍalālati-himā) with the acquiescence of the judges and the fuqahā’ of [the rest of] al-Andalus (jazīrat al-Andalus) due to their lack of interest in this problem.”

18Al-Bājī underpins further the difference between qāḍīship and arbitration and the superiority of the former with respect to the latter by stating that

  • 7 Qur’an, II, 282: The Qur’an, English translation by M.A.S. Abdel Haleem. Parallel Arabic Text, New (...)

“the difference between the qāḍī officially invested to impart justice (al-muwallā li-l-qaḍā’) and two men appointed [ad hoc] by the litigants to settle their dispute is that qāḍīship is an official delegation of competences (wilāya) similar to limited and universal politico-religious authority (al-imāra wa-l-imāma). It cannot be performed by more than one person at the same time. Let it suffice to refer to the action carried out by the Anṣār on the day of al-Saqīfa when they told the emigrants (muhājirūn): ‘there is an amīr of ours and another of yours’. Then ʻUmar told to Sufyān: ‘two cannot be at peace in a single scabbard (ghimd)’ and the people returned to the opinion of Abū Bakr, ʻUmar and the emigrants, and agreed on it. A second aspect to be taken into account is that the authority represented by the caliphate (imāmat al-khilāfa) comprises two meanings; that of [leading] prayer and that of executive and judicial power (al-ṣalāt wa-l-aḥkām) for, if appointing two persons to lead a single prayer is impossible, by the same token it is not permitted to appoint two judges to decide a single legal suit jointly. In third place, only the most satisfying candidates in terms of religiosity, reliability, and knowledge are selected by the imām to impart justice [to the subjects] (li-l-aḥkām), in a similar way to those who are able to impart justice according to the dictates of their capacity [to perform] independent [reasoning and] interpretation [of the sacred sources] (ijtihād). This renders it impossible to compare the [officially invested judge] with someone the validity of whose arbitration award depends upon agreement from his counterpart, for this [shortcoming] is the mark (ṣifa) of one suspected of [easily] going astray given the frequency with which he falls into perdition and his scarce inclination to obey the law. God the Almighty told: ‘If two men are not there, then call one man and two women out of those you approve as witnesses, so that if one of the two women should forget the other can remind her’7”,

whereby the two arbitrators are now equated to the two women whose testimony completed by that of an upright male witness is accepted as full proof in disputes over property rights. Then al-Bājī insists:

  • 8 On Bājī’s rejection of women’s access to qāḍīship, see also Muntaqā: VIII, 184, cited in Carmona, 2 (...)

“No doubt (lā khilāfa… lā yuʻraf ghayru-hu), the exclusive judgment is the majority, the well-known and the obvious opinion (ḥukm al-wāḥid huwa al-mashhūr, al-maʻlūm al-ẓāhir); nothing different has ever been transmitted from any member of the umma… [On the other hand], if [the decision to appoint two judges] were justified under the claim that people are in great need of testimonies [and of settling disputes] and that nobody has been appointed [to perform as witness or as judge], this argument is to be rejected if [the possibility] that two women take the place of one single man [as witnesses is used to justify that two men be appointed over a single judicial seat] given the scarcity of their religiosity, for two male witnesses do not take the place of a single male witness. In consequence, it is not correct that two judges take the place of a single judge. Were this permitted, it would also be legitimate to appoint women as judges and for them to exercise executive power (tawliyati-hinna al-ḥukūma). Moreover we would have two women taking the place of a male, which is unanimously considered null and void8.

Another [negative] development in our times is what is happening in the Andalusī area of al-Riqqa (which may be the present day Ricla in the Spanish province of Zaragoza), where Ibn ʻUmar was appointed qāḍī despite being blind, which all Muslims agree is forbidden, this rule having been [explicitly] subscribed by Abū Ḥanīfa and by al-Shāfiʻī.”

19Al-Bājī’s claim that he “got to know that Mālik prohibited it” and that he personally did not miss the opportunity to criticize the practice every time it happened, expresses his frontal opposition to it, but also highlighs how difficult it was to find a solid doctrinal basis to rely on, in order to present his view as the Mālikī position on the issue. He proceeds further:

20“Two questions remain to be addressed concerning arbitration. The first deals with the conditions necessary to qualify as an arbiter.

The arbiter must be male, free, Muslim, have reached puberty, have the capacity of discernment, be just and honest. The Majmūʻa and the book by his son transmit Saḥnūn’s view that if they appoint someone who is disapproved of (maskhūṭ), or a woman, or a mukātab slave, or an infidel, the [arbitrator’s] decision becomes null and void (bāṭil). Ibn al-Mājishūn, according to the Majmūʻa, also pointed to the freed slave (al-mawlā ʻalay-hi), while in the Wāḍiha he is reported to have added the child, the person disapproved of, and the Christian. Ashhab (Egypt d. 204/820) acknowledged the child and added that if the mentally handicapped and the mentally deranged (al-maʻtūh wa-l-mawsūs) are engaged as arbiters and reach a decision, it will not be valid. This maxim was applied by Muṭarrif to the case of the slave and the woman”.

21Opening a small window to legitimate a practice that must have been more widespread than the jurists were ready to admit, al-Bājī points out, on the authority of the Kitāb Ibn Saḥnūn, that

“in Ashhab’s opinion, if a woman is appointed by the parties, her decision may be endorsed if it concerns a legal matter subject to divergence of opinions, and the same applies to the slave, and the free person who is disapproved of.”

22Stretching this argumentative thread further, al-Bājī mentions, on the authority of Ibn al-Mājishūn that

“if the slave or the woman is far-sighted, wise and trustworthy, appointing them as arbiters is permitted and their decisions are to be considered legitimate, unless they incur in an obvious error, an opinion also subscribed to by Aṣbagh and Ashhab”.

23Ibn Ḥabīb, who is the transmitter of Ibn al-Mājishūn’s aforementioned opinion, declared he had adopted it drawing on the precedent established by ʻUmar

“when he entrusted Shafā’, who was the mother of Sulaymān b. Abī Ḥathama, with [the inspection of] the Medina market, this appointment implying permission to judge among people (lā budda li-wālī al-sūq min al-ḥukm bayna al-nās) even if concerning [only] petty issues (ṣigār al-umūr)”.

24Therefore, Bājī’s rejection of women’s and slaves’ eligibility for qāḍīship does not expel them from the realm of arbitration. The same considerations rendering arbitration licit when performed by a slave, a woman or someone who is disapproved of, are extended by Aṣbagh to someone who has been punished with a statutory sanction (al-maḥdūd), a child with sufficient capacity of discernment, common sense and awareness (idhā kāna qad ʻaqala wa-ʻarifa wa-ʻalima), if the decision they take proves to be wise (aṣāba) “since, normally, the owner of a young slave (ghulām) does not provide him with instruction in the sunna or the art of issuing judgments either”.

25The aforementioned exceptions are not extended to the Christian whom, according to Ibn Ḥabīb’s Wāḍiḥa, Ibn al-Mājishūn had disqualified as a potential arbitrator of disputes held by Muslims. However, the very fact that this contingency is contemplated, if only to be rejected, suggests that it was not totally alien to the array of social relationships and contacts between the Muslim and non-Muslim inhabitants of al-Andalus. In fact, and given Bājī’s obvious preference for the equivalence between arbitration and mandated agency, the following remark may be interpreted in the sense of favouring wisdom and good judgment over gender and religious identity:

“The legal basis [of the opinion] of those who submit [arbitration to the rules of] mandated agency is designed to prevent [the selected arbiter] being someone devoid of mental capacity (dhāhib al-ʻaql), while those who submit it to [the rules of] the official delegation of competences to deal with a specific legal case (al-wilāya fī ḥukm khāṣṣ) do not accept the designation of persons who do not possess all the above described requirements necessary to issue judgment [i.e. being male, free, Muslim, having reached puberty and the capacity of discernment, and being just and honest].”

26The second question deals with legal matters where it is permissible to resort to arbitration:

  • 9 Which obviously includes the qāḍī.

“As to the rules regulating which legal matters can be submitted to arbitration and which cannot, they establish that only disputes arising from properties and from other items subject to the same legal consideration (wa-mā jarā majrā-hā) can be entrusted to the arbitrator. The latter is not authorized to implement a ḥadd punishment or to establish the liʻān oath, as Saḥnūn had it. Aṣbagh specified that the statutory sanction for calumny [was also included in those ḥudūd forbidden for the arbitrator], and added qiṣāṣ, manumission, repudiation, kinship (nasab) and patronage (walā’) to the list of legal matters excluded from his jurisdiction, since these can only be decided by the higher authority (imām)9. However, notwithstanding the prohibition, if the parties seek an arbitrator to settle a dispute concerning any of these matters and he takes a decision, it will be enforced although the ṣulṭān [i.e. the executive authority] will forbid him from doing it again.

  • 10 For a comparison with early Ḥanafī and Shāfiʻī legal doctrine on arbitration, see Tillier, 2009: 30 (...)

The reason is that those legal issues have a special relevance (la-hā qadr), in which case care must be taken that they are decided only by someone enjoying a general delegation of judicial competences (man qāma bi-l-wilāya al-ʻāmma) on the basis that the appointing authority is aware that the nominee meets the conditions required to perform that task and that he is trustworthy, or directly by the imām or the ruler (al-ḥākim) themselves by virtue of their exclusive capacity to intervene when the circumstances require them to do so, but God knows best10.”


  • 11 No reference to arbitration outside the context of the Qur’ānic practice of the ḥakamān is found in (...)

27There is no specific chapter, section or subsection dedicated to ṣulḥ in the Muntaqā, probably because there is no equivalent in its source, the Muwaṭṭa’. This circumstance appears to be confirmed by the occurrence of a similar gap in another authoritative commentary of the Muwaṭṭa’, namely Abū Bakr Ibn al-ʻArabī’s (468/1076-543/1148) Qabas11. Be that as it may, it is noteworthy that in commenting on the foundational text of their legal school, neither al-Bājī nor Ibn al-ʻArabī find it necessary to pay attention to that matter and so broaden the thematic scope of Mālik’s compilation. Instead, the chapter on compensations for blood crimes (al-ʻuqūl) in Bājī’s Muntaqā includes a long section on “granting pardon for intentional killing (al-ʻafw fī qatl al-ʻamd)” (Muntaqā: IX, 112-122), a subject – namely that of negotiating non-application of talio (qiṣaṣ) for a monetary compensation (diya) – with enough weight to be treated independently, and which is beyond the scope of this paper, as it has been fairly well covered by specialists of Islamic penal law.

  • 12 Qur’an, IV, 114 & 128 respectively.

28In contrast, the third chapter of Bājī’s compilation of legal practices (ʻamal) in force among Andalusī jurists and judges, dealing with a variety of procedural matters, does include some brief sections on ṣulḥ showing that his main concern was the protection of minors’ interests, restricting the capacity of their fathers and guardians to conclude agreements in their name (Bājī, Fuṣūl: 284-286). The first section starts with the Qur’anic basis for the recommendation to seek the peaceful settlement of disputes: “There is no good in most of their secret talk, only in commanding charity, or good, or reconciliation between people” … “If they both come to a peaceful settlement, for peace is best”12. Immediately afterwards come the opinions of the ʻulamā’ who, according to al-Bājī, “are unanimous that conciliation is permitted both when there is acknowledgment and denial of the disputed right”. Ṣulḥ, the author explains, to restrict the general application of the statement,

  • 13 That is to say, conciliation here is possible only when the defendant denies the claim (ṣulḥ ʻalā l (...)

“has a permitted and a forbidden aspect. It is permitted that a creditor agrees with the debtor to receive only one part of the debt by the initially fixed deadline. The second case refers to those contingencies in which the agreement becomes forbidden for any given reason, for example, when the debtor agrees to pay part of the debt before the deadline on the condition that he be freed of the rest, since this is not permitted and is forbidden. That a father concludes an agreement in the name of his virgin daughter concerning part of her right to inherit certain properties or another kind of right, is permitted only when that right is subject to dispute or has been challenged in court13. Otherwise he is not allowed to negotiate a decrease in that which corresponds to her since there is no advantage for her in doing so. However, if, notwithstanding the prohibition, a detrimental agreement for her were concluded by her father, she would be entitled to the remaining part of her right at the cost of the debtor, and not at that of her father, unless both the father and the debtor had agreed that the former would place his own patrimony as guarantee of the agreement. In this case, if the girl contests the agreement and the debtor is unable to pay the part he agreed with her father not to pay, liability is transferred to the latter. This was subscribed by Ibn al-Mājishūn and by Muṭarrif, this also being the opinion of Mālik and [the rest of] his followers.”

29The last case contemplated by al-Bājī refers to the testamentary executor (waṣī) who concludes an agreement in the name of his orphan wards.

“This is permitted and effective as long as no fraud is involved; otherwise, it will be forbidden. The executor’s performance is considered legitimate as long as the contrary does not become evident, as Aṣbagh put it on the authority of Ibn al-Qāsim”.

30According to the editor of the Fuṣūl al-aḥkām, al-Bātūl b. ʻAlī, Bājī’s treatment of ṣulḥ is the main source of Ibn Hishām al-Azdī’s (d. 609/1209) adab al-qāḍī treatise known as Mufīd al-ḥukkām, a text which would presumably have provided the basis for Ibn Farḥūn’s (d. Egypt 799/1397) discussion of the subject in his Tabṣirat al-ḥukkām (Fuṣūl: 284 n. 5, 285 n. 1 & 2, and 286 n. 2 & 3).

Excursus: Ibn ʻAbd al-Barr (Cordoba 368/978-Xativa 473/1071)

31The attention received by taḥkīm in Bājī’s Muntaqā is in sharp contrast with the cursory and marginal reference to the practice found in the chapter on adab al-qāḍī in Ibn ʻAbd al-Barr’s compilation of Mālikī legal doctrine (Kāfī: 501). Stressing the need to respect the qāḍī’s judgments by his successor, Ibn ʻAbd al-Barr admits the possibility of resorting to an arbitrator and the obligation to respect his decision by both the parties and the local authorities provided that it is not manifestly erroneous. Conversely, Ibn ʻAbd al-Barr’s treatment of ṣulḥ is quite elaborate (Kāfī: 451-453). Drawing on the authority of Mālik b. Anas he states – like al-Bājī – the permissibility of ṣulḥ ʻalā l-inkār and specifies that it is accepted when both parties agree to it voluntarily and no reprehensible consequence is to be expected from it. Also, he points out to the similarity between sales and agreements in the sense that what is forbidden in the case of sales applies to agreements as well. This parallelism is illustrated with a series of examples on the use of agreements for disguising usury, thereby circumventing the well-known prohibition against illicit gain (ribā) through the exchange of different currencies and of goods of different nature. Interestingly, Ibn ʻAbd al-Barr takes into account the possibility that one of the parties has been pressed to negotiate a deal, a circumstance he considers as invalidating the deal.

It is all very well to avoid conflict but not at any cost: Ibn Ḥazm’s and Ibn Rushd al-Jadd’s doctrines on ṣulḥ

Ibn Ḥazm (Cordoba 384/994-Huelva 456/1064)

  • 14 Which al-Bājī and Ibn ʻAbd al-Barr discuss in the chapter of their aforementioned works dealing wit (...)

32In order to refine our analysis of 11th century C.E. Mālikī jurists’ doctrine on ṣulḥ and taḥkīm, we will take a look from an outside perspective, through the eyes of their Ẓāhirī contemporary Ibn Ḥazm of Cordoba. In his Muḥallā (VIII, 160-168), Ibn Ḥazm tries to restrict the wide scope left for agreement by his Mālikī counterparts, the main traits of his doctrine on ṣulḥ being as follows: like the Shāfiʻīs, he rejects the possibility of reaching an agreement when the disputed right is denied by one of the parties. Also Ibn Ḥazm refuses to understand silence concerning the disputed right as a tacit acknowledgment of the claimant’s right, contrary to the jurists of the other legal schools (Chalmeta, 2000: 647-648). He is also against allowing any agreement that involves giving up an oath which has been previously established as obligatory, or one in which the party who is ready to acknowledge a right negotiates regarding a different right. Conciliation is possible regarding property rights and patrimonial issues, including those derived from unintentional killing and bodily injuries14, as long as the amount or object at stake is known for certain. It is also correct to negotiate the payment of a debt, a part in cash and a part in exchange for a given service, provided that the specific amount of the debt is known since agreements can be reached only regarding specific objects and amounts. An exception to the latter rule forbidding the conclusion of agreements that involve unspecified amounts of money is khulʻ, or divorce at the wife’s initiative, as well as a tooth fractured intentionally, an injury inflicted intentionally, or intentional and unintentional killing, which can be negotiated instead of talio (al-ʻafw ʻan al-qiṣāṣ), for the established blood price (diyya) or for a lower amount.

  • 15 In this early Andalusī compilation of Mālikī opinions, which rely heavily on ʻAbd al-Malik b. Ḥabīb (...)

33Ibn Ḥazm adds a bit more detail regarding the possibilities of withdrawing or retracting from an agreement than does the information offered by al-Bājī and Ibn ʻAbd al-Barr. He specifies that deals concerning killing and bodily injuries inflicted both intentionally and unintentionally can be retracted at the claimant’s initiative, in which case talio or diyya are implemented instead. In this specific respect, his Mālikī predecessor Ibn Abī Zamanīn (Granada, 324/936-399/1008) had observed that, withdrawing the pardon previously granted to the attacker or killer by those entitled to claim the blood of the victim (walī, pl. awliyā’ al-dam), is permitted only with a valid excuse and after swearing an oath that pardon was initially granted without the intention of giving up the rights ensuing from commission of the crime (Ibn Abī Zamanīn, Muntakhab: II, 1021-1023)15.

34Going back to Ibn Ḥazm’s doctrine on ṣulḥ, the extinction or destruction of the object of the agreement is also admitted as a reason entitling the claimant to cancel it, e.g. when someone agrees with someone else in relation to a given item to be awarded, such as the right to live in a certain house or to make use of a certain slave, but the house collapses or the slave dies.

35I have not managed to find any individual section or subsection relevant to taḥkīm in Ibn Ḥazm’s Muḥallā, either in the chapter on qāḍīship and judicial procedure or in those on marriage and divorce, in connection with the Qur’anic practice of the two arbiters.

Ibn Rushd al-Jadd

36In a way, Ibn Rushd al-Jadd’s approach to arbitration and conciliation runs in opposition to that of al-Bājī. Ibn Rushd scarcely appears to be interested in taḥkīm, or I, at least, am unable to find any detailed discussion of this topic, either in his commentary to al-ʻUtbī’s Mustakhraja (i.e. the Kitāb al-bayān wa-l-taḥṣīl) or in that dedicated to Saḥnūn’s Mudawwana (namely al-Muqaddimāt al-mumahhidāt), both providing, together with his fatāwā, the main pool of information for his legal thought. This seems to indicate that for Ibn Rushd, arbitration is devoid of independent entity. It is a default procedure to which the qāḍī can resort to solve disputes among litigants lacking conclusive evidence. For him, the role of the arbitrator is also part of a qāḍī’s prerogative to appoint a delegate, ad hoc or on a permanent basis, an impression reinforced by Ibn ʻAbdūn’s (Seville, d. 530/1135) assertion that non-qāḍī judges should perform as arbitrators rather than as real judges (Ibn ʻAbdūn, Sevilla: 52-53; also see Ibn Abī Zamanīn, Muntakhab: II, 1140-1142 & Muḥammad b. ʻIyāḍ (Ceuta, d. 575/1179), Madhāhib: 35). The only occasion on which engaging an arbitrator is discussed by Ibn Rushd is that of marital disputes, as will be seen below. Further, the second chapter on repudiation in the Bayān refers to the divergence of opinions as to whether the divorce that two arbiters end up pronouncing, when unable to solve the dispute between the spouses, is revocable or irrevocable, a matter Ibn Rushd sums up without indicating his preference for one opinion or the other (Bayān: V, 454-455).

  • 16 The Kitāb al-ṣulḥ in the Bayān, as opposed to what happens in the Muqaddimāt, includes agreements f (...)

37In contrast to taḥkīm, ṣulḥ is treated at length and paid detailed attention in both Ibn Rushd’s Muqaddimāt and Bayān (Bayān: XIV, 135-225; Muqaddimāt: II, 515-524).16 Given its more compact structure where, as happens in Bājī’s Muntaqā and Ibn ʻAbd al-Barr’s Kāfī, matters unfold according to a systematic and thematically coherent order, we present Ibn Rushd’s doctrine on sulḥ on the basis of the Muqaddimāt, instead of the Bayān, which reviews the corresponding casuistry organized according to samāʻs or legal questions orally transmitted from a number of early Mālikī authorities about given legal subjects.

38As is usually the case with his legal methodology, Ibn Rushd’s discussion of ṣulḥ in the Muqaddimāt starts with a thorough substantiation of the relevant Mālikī jurisprudence in the sacred sources. To the verses already quoted by al-Bājī (i.e. Qur’an IV, 114 & 128) – Ibn ʻAbd al-Barr had mentioned none –, Ibn Rushd adds Qur’an XLIX, 8 and Qur’an IV, 35. Ibn Rushd establishes Qur’an IV, 114 as being of general applicability to litigations involving real state property, blood crimes and movables (al-aʻrāḍ) other than gold and silver. With respect to Qur’an XLIX, 8, Ibn Rushd quotes the first part of the verse: “If two groups of believers fight, you [believers] should try to reconcile them” but omits the second part addressing war and peace and the treatment of Muslim rebels: “If one of them oppresses (baghat), fight the oppressors (allatī tabghī) until they submit to God’s command, then make a just and even-handed reconciliation between the two of them: God loves those who are even-handed (al-muqsiṭīn).”

39Subsequently, he introduces Qur’an IV, 128 which, read together with Qur’an IV, 35, provides the basis for the well-known practice of submitting marital disputes to two arbiters, and makes explicit the bond between arbitration and conciliation – a relationship which might also explain why Ibn Rushd does not consider it necessary to deal with taḥkīm independently from ṣulḥ or marriage:

“If a wife fears high-handedness or alienation from her husband, neither of them will be blamed if they come to a peaceful agreement, for peace is best” ... “If you [believers] fear that a couple may break up, appoint one arbiter from his family and one from hers. Then, if the couple wants to put things right, God will bring about reconciliation between them: He is all knowing, all aware.”

40Curiously, Qur’an II, 280 (“If the debtor is in difficulty, then delay things until matters become easier for him; if you were to write it off as an act of charity, that would still be better for you, if only you knew”), which some scholars connect with Islamic legal doctrine on ṣulḥ (e.g., Chalmeta, 2000: 647), is not mentioned by Ibn Rushd.

  • 17 Even in the case of a competent ḥadīth expert such as Ibn ʻAbd al-Barr, in neither of whose al-Isti (...)
  • 18 The traditions quoted by Ibn Rushd are identified by Muḥammad Ḥajjī, editor of the Muqaddimāt (II, (...)

41Ibn Rushd’s treatment of ṣulḥ is also original in that, contrary to his Mālikī predecessors17, the relevant doctrine is grounded in a complete list of ḥadīths18, supporting Ibn Rushd’s commitment to introduce a series of important nuances to the general assumption about the desirability of ṣulḥ, that place his approach in rare harmony with Ibn Ḥazm’s concerns. The list of Prophetic ḥadīths adduced by Ibn Rushd starts with Muḥammad’s words regarding al-Ḥasan:

“‘Verily, this [grand] son of ours is a lord (sayyid). May God grant that an agreement is reached concerning him by two great Muslim sections (fi’atayn ʻaẓīmatayn min al-muslimīn).’ Eventually, things evolved as he had predicted. On a different occasion, he went to see the Banū ʻAmr b. ʻAwf to settle a dispute that had arisen among them (li-yaṣluḥ bayna-hum). Also, it was recounted that the people of Qabā’ quarrelled to the point of throwing stones at each other. When the Prophet got to know about it he said: ‘Let us go to see them and make peace among them.’”

42Turning from physical violence to disputes over financial issues, Ibn Rushd recounts that

“Jābir b. ʻAbd Allāh offered his father’s creditors to collect their due in kind [from his father’s date crop] but they refused on the grounds that this arrangement did not suffice to pay the debt (wafā’). Then he informed the Prophet about it, and the latter told him: ‘When you have picked the [dates] and put them to dry, let me know.’ He followed these instructions and then, Muḥammad, Abū Bakr and ʻUmar appeared and, after blessing the crop, the Prophet told Jābir to call the creditors and pay them their due. [And even after the decrease in the weight of the dried dates with respect to the fresh ones], there were enough of them to pay the creditors and to keep a substantial portion for himself.”

43Having laid the foundations of ṣulḥ in the Qur’an and the prophetic sunna, Ibn Rushd hints that, notwithstanding the desirability of the practice, it is not mandatory either upon Muslim authorities or the parties:

  • 19 Emphasis added.
  • 20 Emphasis added.

“Bringing people to the peaceful resolution of their disputes concerning properties and other legal issues is one of the most praiseworthy and recommendable altruistic supererogatory19 acts that exist (min nawāfil al-khayr). There is no problem20 (lā ba’s) in the supreme authority (imām) engaging in [a practice of] this [kind] by instructing the litigants to [seek a peaceful agreement].”

44This position finds support in its specific ḥadīth source base, the last of which “is the source of a saying which is widespread among the common people and according to which the best agreement is the half and there is no agreement without burden”, whereby the reader is duly reminded of the detrimental consequences of conciliation, especially for the owner of the disputed right. Ibn Rush argues further:

“If the imām instructs, encourages and exhorts the litigants to come to terms with each other (idhā ashāra al-imām ʻalā l-mutakhāṣimayn bi-l-ṣulḥ wa-nadaba-humā ilay-hi wa-ḥaḍḍa-humā ʻalay-hi), but they both refuse, or only one of them refuses, he cannot oblige them nor must he insist on his argument in such a way that they feel pressured by him; rather, he must adjudicate the case according to the relevant legal doctrine and issue the corresponding formal judgment, as God messenger’s did.

According to the Ṣaḥīḥ, al-Zubayr b. al-ʻAwwām litigated with a man from the Anṣār concerning a water pond they both used for irrigation. God’s messenger instructed al-Zubayr to take the water he needed to irrigate [his plot] and to send the rest to his neighbour, but the Anṣārī got angry and said: ‘Oh messenger of God, what [would you say if I were] your cousin (ibn ʻammati-ka)?’ Then God’s Messenger reddened (talawwana lawn wajh rasūl Allāh) and told [to Zubayr: ‘In that case,] take as much water you need and keep the rest for you until it runs out’, whereby he awarded his right to al-Zubayr after having showed him a way out that was advantageous for both but, because the Anṣārī infuriated the Prophet, he issued a formal judgment (ṣarīḥ al-ḥukm) in favour of al-Zubayr, who obtained the [disputed] right. ʻUrwa reported that al-Zubayr exclaimed: ‘For goodness sake! How appropriate [to our case] was the revelation of Qur’an, IV, 65: « But no, by thy Lord! They will not believe till they make thee the judge concerning the disagreement between them, then they shall find in themselves no impediment touching thy verdict, but shall surrender in full submission »’, with just that exception [i.e. the Anṣāri’s refusal to accept the Prophet’s first proposal]’.”

45Though the textual evidence adduced by Ibn Rushd looks a bit forced (e.g. it is not clear whether the Prophet’s reaction was motivated by anger or by the need to respect the Anṣārī’s decision), maybe because he could not find more conclusive arguments, the message he wishes to convey to his audience is clear: nobody must be forced to reach an agreement however foolish and irritating his/her attitude may result.

46Next, a series of conditions rendering ṣulḥ either not recommendable or inadmissible are presented and argued on the grounds of Qur’an and ḥadīth but also on those of the practice of the well-guided caliphs: “It is convenient for the imām to encourage and direct the parties towards reaching an agreement” Ibn Rushd asserts, just to specify that this is so

“only when he is not certain that the right belongs to one of them. Otherwise he should not wait until they reach a settlement to issue his judgment in favour of that party according to the precedent established by ʻUmar b. al-Khaṭṭāb in his letter to Abū Mūsā al-Ashʻarī: ‘Seek (iḥriṣ) agreement as long as you are certain that the issue does not need to be decided by means of a formal judgment (mā lam yatabayyan la-ka faṣl al-qaḍā’)’.”

47Though referring to ṣulḥ and to the judge rather than the litigants, this point seems to run counter to the opinion expressed by Ibn al-Qāsim, Muṭarrif and Aṣbagh, as mentioned above, and quoted without further nuance or qualification by al-Bājī, stating the need to comply with the arbitrator's decision even if, having examined the evidence produced by the parties, one of them realizes that his/her right would have prevailed if s/he had resorted to ordinary justice.

“Conciliation is not possible concerning a forbidden matter for, as stated by God’s messenger, conciliation among Muslims is permitted except when it aims at allowing something forbidden or at forbidding something that is permitted. However, opinions vary when, [notwithstanding the restriction], an agreement of this kind is concluded. According to some, it must be cancelled (yufsakh), rejected and redressed, this being the opinion of most scholars (jumhūr ahl al-ʻilm), whereas Aṣbagh considered it permitted from a strictly legal point of view (fī wajh al-ḥukm), although ethically (fī mā bayna-hu wa-bayna Llāh) unacceptable”,

48whereby permissiveness towards forbidden agreements is undermined without having to rebut Aṣbagh more openly. Ibn Rushd proceeds further:

“Aṣbagh’s opinion was used to argue in favour of ʻAlī b. Abī Ṭālib who, according to the tradition, once a document containing the terms of a certain agreement was submitted to his approval, commented: ‘This is forbidden; were it not because it is something upon which an agreement has already been reached, I would declare it null and void’”.

49The link between Aṣbagh and ʻAlī is neutralized by Rushd, declaring the first opinion as the most correct one, and asserting the Prophet’s superiority over ʻAlī and the soundness of the ḥadīth collections and authorities transmitting the relevant evidence:

  • 21 This ḥadīth was also quoted by Ibn Ḥazm to argue against being permissive with erroneous agreements (...)

“It [i.e. the first opinion] is supported by the Prophet’s words and deeds (qawlan wa-ʻamalan). Moreover, the [following] ḥadīth qualifying the Prophet’s [refusal of agreements forbidding something permitted or permitting something forbidden] was transmitted in the Ṣaḥīḥ: ‘He who introduced in this rule of ours something that does not correspond to it, reject him’. Another ḥadīth establishing what the Prophet put into practice and the judicial decisions he took (qaḍā’u-hu) [to realize the refusal] is attributed to Abū Hurayra and was transmitted by Mālik in the Muwaṭṭa’ on the authority of Abū Hurayra and Zayd b. Khālid al-Juhanī. According to it, two men litigated before God’s Messenger and one of them told: ‘Oh, Messenger of God, decide (iqḍi) our case on the basis of the Qur’an’, while the other recited the aforementioned ḥadīth in its entirety until the Prophet asserted: ‘Regarding the cattle and the slave woman that belong to you, they must be returned to you’21. And [the Prophet] declared null and void those conditions of the agreement which did not appear in the sacred book, for that [lack of Qur’anic reference] renders them null and void, even if they (i.e. the conditions) amounted to one hundred.

Conciliation consists just in taking something in exchange for something else. It resembles the commercial deal [al-mubāyaʻa] being, in most circumstances, ruled by the same principles. It is not, therefore, appropriate that the agreement helps confirm what is forbidden in the case of sales. No divergence of opinions affects the need to cancel any settlement involving something that is forbidden for one of the parties, e.g. when someone from whom another claims a debt of 10 dinars acknowledges only 5, but finally agrees to delay the payment (ilā ajl) of the total amount initially claimed by the creditor, but does so in dirhams, and other such deals. The only question raising a certain divergence of opinions is the fact that the forbidden aspect of the agreement affects only one of the parties since the creditor is not allowed to receive his ten dinars in dirhams at a delayed deadline, whereas the debtor is allowed to assume that very same compromise under oath. This view was endorsed by Aṣbagh, who did not consider [that kind of agreement] null and void from a strictly legal point of view drawing on ʻAlī b. Abī Ṭālib’s [aforementioned] tradition, though the dominant opinion (al-mashhūr) holds for nullifying it. Regarding the reprehensible agreement (al-ṣulḥ al-makrūh), some were in favour of tolerating it once concluded but Ibn al-Mājishūn held for its revocation if the agreement was disclosed shortly after its conclusion. [A reprehensible agreement] is one in which the parties agree on something obviously corrupt and which cannot be realized by one of the parties, for example when each of them claims a number of dinars or dirhams from the other and they both refuse to acknowledge the debt to their opponent; then, they agree to postpone the payment of what they claimed to each other to a given term, and both argue that there is no prohibition in what they did since what they gave was by way of free provision (wāhib). It thus becomes evident that the reason why they both postponed [the payment of] their [mutual] debt was to move the other to do the same and, then, to agree with him to pay their mutual due in advance (aslif-nī wa-uslifu-ka). Also, it is feared that each of them is entitled to part of what they claim from one another and that postponement of the payment aims at obtaining a surplus (ʻalā an yazīda-hu fī-hi), that is to say, the usury that is forbidden by the Qur’an. [In sum,] be aware that [Mālikīs approach problematic agreements] in three ways: (1) unanimity on the need to cancel the agreement; (2) divergence of opinions on the need to cancel it; (3) unanimity on the option to maintain it unless the agreement is disclosed [shortly] after its conclusion, with divergence of opinions on the need to cancel it”.

50Commenting on reprehensible arrangements, Ibn Abī Zamanīn had quoted Muṭarrif’s and Ibn al-Mājishūn’s opinions according to which they can be respectively validated in all cases or only when they have been put in practice for long after their conclusion. Mixing both opinions, Ibn Abī Zamanīn identifies Muṭarrif as the authority on which ʻAbd al-Malik b. Ḥabīb relied, to hold in favour of validating reprehensible agreements if they had already been put in practice. A look at Ibn Abī Zamanīn’s treatment of this specific case (Ibn Abī Zamanīn, Muntakhab: II, 1109-1111) makes clear that Ibn Rushd differed from Ibn Ḥabīb, and hence from Ibn Abī Zamanīn himself, in their readiness to accept reprehensible and forbidden agreements discovered once they had been put into practice, and that he rejected the charitable donation of something equivalent to the value of the object agreed upon as a valid way to atone for the ethical dilemma involved.

  • 22 This expression can also be understood to mean it is impossible to claim when the slave is no longe (...)

51Subsequently, Ibn Rushd applies the above-mentioned requirements to a practical example, namely that of the sale of a slave in whom the buyer finds a defect regarding which he is ready to reach a settlement with the seller. This contingency, to which Ibn ʻAbd al-Barr had also referred though briefly, articulates the rest of Ibn Rushd’s particular approach to ṣulḥ. After establishing a primary distinction between the expiration of the time limit to claim anything on account of the slave’s defect (an yakūn al-ʻabd fā’itan)22, and the lapse during which claiming [anything concerning the slave] is still possible (al-ʻabd qā’im), Ibn Rushd addresses with utmost care and detail a vast array of instances that must have been common in daily commercial transactions and the lucrative slave trade. The accuracy with which Ibn Rushd scrutinizes every possible relevant aspect is meant – as was the case with Ibn ʻAbd al-Barr’s treatment of ṣulḥ – to undermine legal devices helping to disguise usury, but also to refine, clarify and complete the relevant sections in Saḥnūn’s Mudawwana, which Ibn Rushd found to be deficient in those respects.

  • 23 On which see Ibn Rushd al-Jadd, Fatāwā: I, 631-649. The full contents of the fatwā as well as the h (...)

52Ibn ʻAbd al-Barr admits the possibility that the usurper of a given object (e.g. a slave, a piece of cloth or something similar) and its legitimate owner can reach an agreement concerning it (also see Ibn al-ʻAṭṭār, Wathā’iq: 445-446). However, in line with his uncompromising doctrine on usurpation, requiring either the return of the usurped object or reimbursement of its value23, that case is not discussed by Ibn Rushd in connection with ṣulḥ.

Retrospective supplement to Andalusī doctrine on arbitration: withdrawal, and the “managers” of conciliation

  • 24 On the istirʻā’ typology and its use in legal contexts different from ṣulḥ see Müller, 1999: 180-18 (...)

53Both Ibn ʻAbd al-Barr and Ibn Rushd point to external pressure as an element going against the spirit of ṣulḥ and thus invalidating any decision taken under its influence. We have also seen that Ibn Rushd was opposed to validating agreements when, after their conclusion, one of the parties manages to present upright witnesses demonstrating the veracity of his/her initial claim and confirming his/her full right to the disputed object both at the interested party’s initiative or at that of the judge. Be that as it may, none of the authors whose doctrines we have explored so far is particularly eloquent regarding the possibility of withdrawing from an agreement. The reason is certainly not rooted in irrelevance or infrequency, for 10th century Cordoban legal practice was already fairly well prepared in that respect, as Ibn al-ʻAṭṭār’s (d. Cordoba 399/1009) above-mentioned model shurūṭ collection shows. Apart from a series of model forms appropriate for recording agreements on inheritances, bequests, the payment of the kāli’ and other debts, as well as homicide and bodily injuries, Ibn al-ʻAṭṭār includes a model deed known as ʻiqd al-istirʻā’ (“deed of caution” or “guardedness”). This kind of deed serves the purpose of having a series of facts recorded in the presence of witnesses, facts which the owner of the document reserves the right to wield in court should certain circumstances converge in the future, in order to claim the cancellation of an agreement and, hence, exemption from the obligation to fulfill it (Wathā’iq: 419-456 & 438-440)24. Indeed, Ibn al-ʻAṭṭār’s contemporary, Ibn Abī Zamanīn, observes, drawing on the authority of Saḥnūn, that

“a document of istirʻā’ – which he conceives as a written record of an assessment made in the presence of witnesses in secret (idhā ashhada ʻalā hādhā fī l-sirr) – specifying that someone was forced to reach an agreement with someone else disputing his/her right, or refusing to acknowledge it (because there was no other choice and on the assumption that it was impossible for him/her to demonstrate his/her right), entitles the claimant to contest the agreement should s/he manage to find witnesses in favour of his/her claim after the conclusion of the agreement, provided that the istirʻā’ document was drafted before the latter was reached, all the more so when those witnesses are also aware of the defendant’s refusal (jaḥd) to acknowledge the initial claim”. (Muntakhab: II, 1062 & I, 297-299)

  • 25 See Ibn Abī Zamanīn, Muntakhab: II, 752, though it might be a confusion with walī or matrimonial gu (...)
  • 26 For an illustration of this see al-Qurṭubī, Jāmiʻ: VI, 290 & 297.

54As a final note, we would like to point that in the above narratives, the trustworthy, virtuous and learned people engaged in the conflict resolution process are usually encouraged to become involved by the qāḍī or the provincial governor (wālī)25, consolidating the idea that arbitration and peaceful settlement might evolve outside the courts but under the authorities’ close scrutiny and control. As P. Chalmeta observes (2000: 651), three of the model legal forms included in the chapter on ṣulḥ of Ibn al-ʻAṭṭār’s al-Wathā’iq wa-l-sijjillāt, refer to the intervention of a “good man” offering his help to solve the dispute held by the parties as a common scenario (Wathā’iq: 434-435, 444-446 & 452-453). Given that these well-intentioned people are described as intervening after a claim has been filed before the judge, it may be assumed that they used to step in at the latter’s recommendation or that, especially in large cities like Cordoba, they were close enough to the activities of the courts to be acquainted with ongoing disputes. The same impression is conveyed by Mālikī jurists’ approach to the role of the ḥakamān or the amīn, the latter usually being portrayed as a figure summoned by someone other than the spouses themselves, such as the authorities, the qāḍī, the matrimonial guardian or the community26. It is thus difficult to use the above-mentioned testimonies to document the handlings of “free-lance” peacemakers operating independently from the established courts, which does not mean we have to rule out their having actually existed.


  • 27 See, for example, Ibn Qudāma, Mughnī: IX, 107, quoted by El-Ahdab & El-Ahdab, 2011: 14.
  • 28 We refer to the opinions selected by the author of the entry, not the sections on arbitration inclu (...)

55The terseness of expression regarding taḥkīm from a school generally characterized by an unusual openness towards arbitration27, and in a region whose adherence to Mālikism might have been connected with that very same attitude, is noteworthy. This might be a false impression resulting from our inability to locate relevant discussions in the sources consulted. Yet, resort to a useful reference tool such as the al-Mawsūʻa al-fiqhiyya has not been of much help either. The Mālikī opinions mentioned in the section on “Taḥkīm” (X: 233-247) are not particularly abundant and are extracted from late sources including Qurṭubī’s (d. Egypt, 671/1272) al-Jāmiʻ li-aḥkām al-Qur’ān, commentaries and supercommentaries of earlier sources, Khalīl b. Isḥāq’s (d. Egypt 767/1365) Mukhtaṣar, Ibn ʻĀṣim’s (Granada, d. 829/1426) Tuḥfat al-ḥukkām and Ibn Farḥūn’s Tabṣirat al-ḥukkām, without adding anything substantial to the points already raised28.

  • 29 Secondary because they are later and less authoritative than Mālik’s Muwaṭṭa’ or Saḥnūn’s Mudawwana (...)
  • 30 As per Bājī’s explicit identification of his sources. None of the edited fragments of the Wāḍiḥa(...)

56This enhances the novelty of al-Bājī’s doctrine on arbitration and makes his selection of the commentary, rather than the adab al-qāḍī genre, to discuss an issue which he had to reconstruct from “secondary” sources29 (e.g. Ibn Abdūs’ Majmūʻa, Muḥammad b. Saḥnun’s Kitāb, and Ibn Ḥabīb’s Wāḍiḥa)30, all the more intriguing and interesting. His aim was to set clear limits on the scope and legitimacy of arbitration with the ultimate intention of arguing in favour of the superiority of qāḍīship and of protecting it from contamination and instability in a context of political fragmentation and religious anxiety. Yet instead of obliterating arbitration, as his peers did, he seems to have considered it sensible to avoid placing the prerogatives of qāḍīship and its holders at even greater risk by acknowledging people’s right to choose their own judge – be this a woman, a slave, a disapproved Muslim, or even a dhimmī – and by preserving the Islamic aura of a conflict resolution strategy that could afford to operate without documents and upright witnesses: A version of Islamic justice adapted to the possibilities of common, non-influential people?

57Doctrinal and political specificities notwithstanding, al-Andalus does not seem to have escaped a phenomenon that has been detected in other similar contexts, as observed for example by Uriel Simonsohn in the sense that Islamic legal treatises “make reference to arbitration as a legitimate form of Islamic adjudication”; yet “Muslim jurists, irrespective of their legal affiliation, tended to allow this form of legal resolution in very few instances, rendering it significantly inferior to the qāḍī’s authority” (Simonsohn, 2011: 69).

  • 31 Extra-judicial in the sense of taking placing outside the courts and other spaces associated with t (...)
  • 32 Talking about Saudi qāḍīs, Frank Vogel (2000: 153) observes that “Cases that proceed to final judge (...)
  • 33 See Muḥammad b. ʻIyāḍ, Madhāhib: 36-37.

58We can thus assume that arbitration was common and successful enough to be resented by the jurists as a threat to qāḍīship and to their status as privileged dispensers of Islamic justice. As an alternative to avoiding the disadvantages of formal adjudication, they offered mediation directly exercised by themselves or by a third party appointed by them and acting under their supervision. From this perspective ṣulḥ may be seen as a controlled, “domesticated” alternative to arbitration. The frequency with which both pre-modern legal sources and present day judicial practice portray qāḍīs in action while trying to convince the parties to come to an agreement and so avoid having to decide their dispute by means of a judgment, may be the result of a bias in the written sources – authors having intentionally neglected the extra-judicial settlement of disputes31 and daily life in the countryside – but is at least telling of the success of conciliation with respect to formal adjudication32. In the case of rural or tribal areas escaping the control of the ruler and lacking officially invested judges during the Almoravid period, we know of the jurists’ insistence that conflict resolution be entrusted to the most virtuous Muslims, to those in possession of some religious training, or to the community33, whereas the Almohads used their resources to fill those gaps by creating new judicial seats. The case of Qāḍī ʻIyāḍ (d. 544/1149) is paradigmatic in that regard. After being banished from Ceuta for his famous attempt to rid the city of the Almohad grip, he was sent to the area of Dāy, close to Marrakech, to impart justice to the local tribes, in a joint move to punish him for his rebellion and replace the tribal chiefs with a charismatic and cosmopolitan religious scholar such as himself (Serrano, 2009: 413).

  • 34 On actual legal cases reporting attempts to use conciliation to circumvent the law (including a ref (...)
  • 35 To judge by Ibn Farḥūn’s doctrine on arbitration (Tabṣira: I, 50-51), we should rule that possibili (...)

59Ibn Rushd’s attempt to transform Mālikī doctrine on ṣulḥ by raising awareness about its negative consequences and the risk that amicable settlement be abused in order to circumvent well established principles like the prohibition of ribā or the need to pay one’s debts and fulfil one’s commitments34, only confirms the above impression. Whether he managed to shape later legal doctrine and practice cannot be ascertained at this stage in our research35. What is clear is that his treatment of received Mālikī doctrine on ṣulḥ provides another example of the way in which his concerns with the ethical dimension of Islamic law and with the evil consequences to which its thoughtless application might lead, eventually drew him remarkably close to such sharp critics of his school as Ibn Ḥazm, and al-Shāfiʻī, the latter having rejected ṣulḥ ʻalā l-inkār on the grounds that it was liable to end up permitting alien property to be illicitly used and disposed of without compensation (Ibn Rushd al-Ḥafīd, Bidāya: II, 293-294).

60Agreements are supposed not to violate Islamic principles or go against what is explicitly stated in the sacred sources, but we can safely assume a strict following of fiqh was not required either, especially when qāḍīs and other legal professionals were not involved in the decision-making process. In this context, non-fiqhī – though still perfectly Islamic – valid considerations providing the basis for taḥkīm are described in terms of their compliance with prudence, trustworthiness, moral virtue, experience, know-how, common-sense, general interest, etc.

61Conciliation as dealt with in the relevant sections and subsections of the works we have consulted refers to peaceful settlement of disputes concerning private rights, especially property rights. It did not govern the relationship with rebels and non-Muslim enemies. Yet as a special legal procedure less strict than qāḍā’, arbitration and conciliation are quite relevant to facilitating dhimmīs’ resorting to Islamic justice whenever this was liable to turn out to their advantage, thereby allowing them to overcome the limitation represented by the impossibility of them qualifying as upright witnesses, or the difficulties encountered in securing upright Muslim witnesses in their favour.

Haut de page



Al-Bājī, Sulaymān b. Khalaf, Fuṣūl al-aḥkām wa-bayān ʻalā mā maḍā ʻalay-hi l-ʻamal ʻinda l-fuqahā’ wa-l-ḥukkām, Rabat, Wizārat al-awqāf wa-l-shu’ūn al-islāmiyya, 1990.

Kitāb al-Muntaqā (“The select”), sharḥ al-Muwaṭṭa’, ed. M. ʻAbd al-Qādir Aḥmad ʻAṭā, Beirut, Dār al-kutub al-ʻilmiyya, 1999, 9 vols.

Ibn ʻAbd al-Barr, Abū ʻUmar Yūsuf b. ʻAbd Allāh b. Muḥammad al-Namarī al-Qurṭubī, al-Istidhkār al-jāmiʻ li-madhāhib al-amṣār wa-ʻulamā’ al-aqṭār, ed. Beirut, Dār al-kutub al-ʻilmiyya, 9 vols.

al-Kāfī fī fiqh ahl al-Madīna al-Mālikī, Beirut, Dār al-kutub al-ʻilmiyya, 1327/2006.

Ibn ʻAbdūn, Muḥammad b. Aḥmad, Sevilla a comienzos del siglo XII. El tratado de Muhammad b. Ahmad Ibn Abdun, Spanish trans. Évariste Lévi-Provençal & Emilio García Gómez, Madrid, Moneda y Crédito, 1948.

Ibn Abī Zamanīn, Muntakhab al-aḥkām, ed. M. Ḥammād, Rabat, Markaz al-dirāsāt wa-l-abḥāth wa-iḥyā’ al-turāth, 1439/2009, 2 vols.

Ibn al-ʻArabī, Abū Bakr Muḥammad, Kitāb al-Qabas fī sharḥ Muwaṭṭa’ Mālik b. Anas, ed. M. ʻAbd Allāh Uld Karīm, Beirut, Dār al-Gharb al-Islāmī, 1992, 2 vols.

Ibn al-ʻAṭṭār, Kitāb al-wathā’iq wa-l-sijillāt, ed. Pedro Chalmeta & Federico Corriente, Madrid, Academia Matritense del Notariado-Instituto Hispano-Árabe de Cultura, 1983; Spanish study and translation by Pedro Chalmeta & Marina Marugán, Formulario notarial y judicial andalusí, Madrid, Fundación Matritense del Notariado, 2000.

Ibn Farḥūn, Tabṣirat al-ḥukkām fī uṣūl al-aqḍiya wa-manāhij al-aḥkām, Beirut, ed. Jamāl Marʻashlī, Dār al-kutub al-ʻilmiyya, 1316/1995, 2 vols.

Ibn Ḥabīb, ʻAbd al-Malik, Kitāb al-wāḍiḥa, edition & Spanish translation of a number of fragments of the work preserved in Ibn Abī Zamanīn's Muntakhab al-aḥkām by María Arcas Campoy, Kitāb al-wāḍiḥa (Tratado jurídico). Fragmentos extraídos del Muntajab al-aḥkām de Ibn Abī Zamanīn (m. 399/1008), Fuentes Arábico-Hispanas number, 27, Madrid, Consejo Superior de Investigaciones Científicas, 2002.

Ibn Ḥazm, Abū Muḥammad ʻAlī b. Aḥmad b. Saʻīd, al-Muḥallā, ed. M. Munīr al-Dimashqī, Cairo, Idārat al-ṭabāʻa al-munīriyya, 1347/1931.

Ibn Qudāma, Mughnī, ed. R. Riḍā, Cairo, 1367/1947, 9 vols.

Ibn Rushd al-Ḥafīd, Abū l-Walīd Muḥammad b. Aḥmad, Bidāyat al-mujtahid wa-nihāyat al-muqtaṣid, Beirut, Dār al-maʻrifa, 1982, 6th edition, 2 vols.

Ibn Rushd al-Jadd, Abū l-Walīd Muḥammad b. Aḥmad, al-Bayān wa-l-taḥṣīl wa-l-sharḥ wa-l-tawjīh wa-l-ta’līl fī masā’il al-Mustakhraja, ed. M. Ḥajjī, Beirut, Dār al-Gharb al-Islāmī, 1984, 20 vols.

Fatāwā Ibn Rushd al-Jadd, ed. al-M. b. al-Ṭāhir al-Talīlī, Beirut, Dār al-Gharb al-Islāmī, 1987, 3 vols.

al-Muqaddimāt al-mumahhidāt, ed. M. Ḥajjī, Beirut, Dār al-Gharb al-Islāmī, 1988, 2 vols.

Muḥammad b. ʻIyāḍ, Maḏhāhib al-ḥukkām fī nawāzil al-aḥkām, ed. M. Bencherifa, Beirut, Dār al-ġarb al-islāmī, 1990.

al-Qurṭubī, al-Jāmiʻ li-aḥkām al-Qur’ān wa-l-mubayyin li-mā taḍammana min al-sunna wa-āy al-furqān, ed. ʻAbd Allāh b. ʻAbd al-Muḥsin al-Turkī, Beirut, Mu’assasat al-Risāla li-l-ṭibāʻa wa-l-nashr wa-l-tawzīʻ, 2006, 24 vols.


El-Ahdab Abdul Hamid & El-Ahdab, Jalal, 2011, Arbitration with the Arab Countries, Third Revised and Expanded Edition, Bedfordshire, U.K., Kluwer Law International.

Arcas Campoy María, 2002, See above, Ibn Ḥabīb.

Carmona Alfonso, 2000, “Le Malékisme et les conditions requises pour l’exercise de la judicature”, Islamic Law and Society, 7/2 (2000), p. 123-158.

Chalmeta Pedro, 2000: see above, Ibn al-ʻAṭṭār.

Fernández Félix Ana, 2003, Cuestiones legales del Islam temprano: la “ʻUtbiyya” y el proceso de formación de la sociedad islámica andalusí, Madrid, Consejo Superior de Investigaciones Científicas.

Fierro Maribel, 1985, “Los malikíes de al-Andalus y los dos árbitros (al-ḥakamān)”, Al-Qanṭara, 6, p. 79-102.

– 2005, “Proto-Malikis, Malikis, and Reformed Malikis in al-Andalus” in Bearman, Peri, Peters, Rudolph & Vogel Frank (eds.), The Islamic School of Law: Evolution, Devolution and Progress, Cambridge, Massachussets, Harvard University Press, p. 57-76.

– 2006, “Ill Treated Women Seeking Divorce: The Qur’ānic Two Arbiters and Judicial Practice among the Malikis in al-Andalus and North Africa” in Masud, Muhammad Khalid, Peters, Rudolph & Powers, David (eds.), Dispensing Justice in Islam. Qadis and their judgments, Leiden, Brill, p. 323-347.

– 2012, “Al-Bāŷī, Abū l-Walīd” in Lirola, Jorge & Puerta, José Miguel (dirs. & eds.), Biblioteca de al-Andalus. De al-ʻAbbādiyya a Ibn Abyaḍ, Enciclopedia de la Cultura Andalusí, vol. 1, Almería, Fundación Ibn Tufayl de Estudios Árabes, 75, p. 233-243.

Gómez Rivas Camilo, 2015, Law and the islamization of Morocco under the Almoravids, Leiden, Brill.

Hentati Nedjmeddine, 2007, “Mais le cadi tranche-t-il ?”, Islamic Law and Society, 14, p. 180-203.

Layish Aaron, 1991, Divorce in the Lybian Family: A study based on the sijills of the shariʻa courts of Ajdabiyya and Kufra, New York, New York University Press.

– 1995, “Dar ʻadl-Symbiosis of Custom and Shariʻa in a Tribal Society in Process of Sedentarization”, Jerusalem Studies in Arabic and Islam, 19, p. 198-213.

Al-Mawsūʻa al-fiqhiyya, Kuwayt, Wizārat al-awqāf wa-l-shu’ūn al-islāmiyya, 1414/1993, 4th edition.

Müller Christian, 1999, Gerichtspraxis im Stadtstaat Cordoba: Zum Recht der Gesellschaft in einer malikitisch-islamischen Rechtstradition des 5.11. Jahrhunderts, Leiden, Brill.

Powers David, 2002, Law, Society and Culture in the Maghrib, 1300-1500, Cambridge, U.K., Cambridge University Press.

– 2006, “Four Cases Relating to Women and Divorce in al-Andalus and the Maghrib, 1100-1500” in Masud Peters and Powers (eds.), Dispensing Justice in Islam, p. 383-409.

Scholz Peter, 1997, Malikitisches Verfahrensrecht. Eine Studie zu Inhalt und Methodik der Scharia mit rechtshistorischen und rechtsvergleichenden Anmerkungen am Beispil des malikitischen Verfahrensrecht bis zum 12. Jahrhundert, Frankfurt am Main etc., Peter Lang.

Serrano Ruano Delfina, 2009, “ʻIyāḍ b. Mūsā”, in Lirola Jorge (dir. & ed.), Biblioteca de al-Andalus, vol. VI, Almería, Fundación Ibn Tufayl de Estudios Árabes, number 1479, p. 404-434.

–, 2013, “Ibn Rushd al-Jadd” in Arabi, Oussama, Powers, David & Spectorsky, Susan (eds.), Islamic Legal Thought. A Compendium of Muslim Jurists, Studies in Islamic Law and Society, number 36, Leiden, Brill, p. 295-322.

–, 2014, “Judicial pluralism under the “Berber empires” (last quarter of the 11th century C.E./first half of the 13th century C.E.)”, Bulletin d’Études Orientales, 63, p. 243-274.

–, 2016, “Forum shopping in al-Andalus (II): Discussing Coran V, 42 and 49 (Ibn Ḥazm, Ibn Rushd al-Jadd, Abū Bakr Ibn al-ʽArabī and al-Qurṭubī)” in Echevarría, Ana, Monferrer, Juan Pedro & Tolan, John (eds.) Law and Religious Minorities in Medieval Societies: between Theory and Praxis. De la teoría a la práctica en el derecho de las minorías religiosas en la Edad (RELMIN 9), Turnhout, Brepols Publishers, p.173-193.

Simonsohn Uriel, 2011, A Common Justice: the Legal Allegiances of Christians and Jews under Early Islam, Philadelphia, University of Pennsylvania Press.

Tillier Mathieu, 2009, Les cadis d’Iraq et l’État abbasside (132/750-334/945), Damascus, Institut français du Proche-Orient.

Vogelw Frank, 2000, Islamic Law and Legal System: Studies of Saudi Arabia, Studies in Islamic Law and Society, number 8, Leiden, Brill.

Haut de page


1 This paper collects results both from a research carried out at the Institute of Islamic World Studies of Zayed University (Abu Dhabi, UAE) funded by the Spanish Ministry of Education, Culture and Sports, ref. PR-2011-0068, and the in-progress R+D project “In the footstepts of Abū ʻAlī al-Ṣadafī: Tradition and devotion in al-Andalus (XIth to XIIIth centuries C.E.)”, funded by the Spanish Ministry of Economy and Competitiveness, ref. FFI2013-43172-P.

2 “Judicial” is used here to refer to the ordinary dispensation of justice as performed by a recognized authority (ḥukm, qaḍā’), and not to taḥkīm or to the ad hoc – mostly dependent on the will of the parties concerned – justice exerted by the arbiter (ḥakam). We are aware, nevertheless, that an arbiter may be and usually is on the same footing as a judge and that conciliation may, and does in fact tend to, take place at the initiative of officially invested authorities.

3 He performed qāḍīship for several ṭā’ifa kings of al-Andalus, though only in minor towns like Orihuela, and developed a close relationship with the ruler of Zaragoza, for whom he carried out a series of diplomatic missions that reached the Christian kingdoms (Fierro, 2012).

4 His career and his integration into the world of Andalusī men of science are closely connected with the rise of the Almoravids, for whom he performed as chief qāḍī of Cordoba and as counsellor (Serrano, 2013).

5 On the centrality of the legal commentary in both al-Bājī’s and Ibn Rushd al-Jadd’s intellectual activity and their classification as renewers of Mālikī legal doctrine, see Fierro, 2005 & 2012: 233-234; Fernández Félix, 2003: 263-274 & 368-373.

6 During the ṭā’ifa period, Onda – in present day Spanish province of Castellón – passed from the hands of the ṭā’ifa rulers of Valencia (the ʻĀmirids – i.e. descendans of Almanzor or al-Manṣūr Ibn Abī ʻĀmir, the powerful ḥājib of the Umayyad caliph Hishām II), to those of the Banū Hūd, rulers of Lerida and Zaragoza and to those of the Banū Dhī l-Nūn of Toledo. In all likelihood, al-Bājī’s testimony refers to a period subsequent to his return from the East around 1047 C.E. which renders it difficult to be precise as to what ruler in particular, or lieutenant of this ruler, he refers. Given his good relationship with the Hūdids of Zaragoza, it might be assumed that he was not pointing at them when describing a practice he disliked so much.

7 Qur’an, II, 282: The Qur’an, English translation by M.A.S. Abdel Haleem. Parallel Arabic Text, New York, Oxford University Press, 20102.

8 On Bājī’s rejection of women’s access to qāḍīship, see also Muntaqā: VIII, 184, cited in Carmona, 2000: 145. According to Carmona, Bājī is the first Mālikī scholar to have addressed the issue.

9 Which obviously includes the qāḍī.

10 For a comparison with early Ḥanafī and Shāfiʻī legal doctrine on arbitration, see Tillier, 2009: 301-318.

11 No reference to arbitration outside the context of the Qur’ānic practice of the ḥakamān is found in Abū Bakr Ibn al-ʻArabī’s Qabas either, qāḍīship being defined there as “the pillar of the sharīʻa, the axis of legal norms and God’s succession to humanity (al-qaḍā’ bayna l-nās aṣl al-sharīʻa wa-madār al-aḥkām wa-khilāfat Allāh fī l-khalq)” (Qabas: II, 758-759 & 869). Arbitration is discussed in his al-Jāmiʻ li-aḥkām al-Qur’ān but the focus there is dhimmīs’ right to resort to Muslim judicial authorities rather than Muslims’ capacity to opt for alternative adjudication forms (Serrano, forthcoming).

12 Qur’an, IV, 114 & 128 respectively.

13 That is to say, conciliation here is possible only when the defendant denies the claim (ṣulḥ ʻalā l-inkār).

14 Which al-Bājī and Ibn ʻAbd al-Barr discuss in the chapter of their aforementioned works dealing with blood crimes, rather than in those corresponding to ṣulḥ like Ibn Ḥazm.

15 In this early Andalusī compilation of Mālikī opinions, which rely heavily on ʻAbd al-Malik b. Ḥabīb’s authority, the information relevant to ṣulḥ is presented in a dispersed manner without apparent organizational logic, contrary to the later legal works we have examined.

16 The Kitāb al-ṣulḥ in the Bayān, as opposed to what happens in the Muqaddimāt, includes agreements for intentional killing (qatl ʻamdan) (see Bayān: XIV, 209).

17 Even in the case of a competent ḥadīth expert such as Ibn ʻAbd al-Barr, in neither of whose al-Istidhkār al-jāmiʻ li-madhāhib al-amṣār wa-ʻulamā’ al-aqṭār; al-Istiʻāb fī maʻrifat al-aṣḥāb and al-Tamhīd li-mā fī l-Muwaṭṭa’ min al-maʻānī wa-l-asānīd have I managed to find any specific treatment of Prophetic traditions relevant to ṣulh, other than the section on the two arbitrators (Mā jā’a fī l-ḥakamayn) in Istidhkār: VIII, 182-184.

18 The traditions quoted by Ibn Rushd are identified by Muḥammad Ḥajjī, editor of the Muqaddimāt (II, 515-518, n. 7-11 & 18-20).

19 Emphasis added.

20 Emphasis added.

21 This ḥadīth was also quoted by Ibn Ḥazm to argue against being permissive with erroneous agreements, due to the risk they involve that someone’s legitimate assets might be unjustly appropriated by someone else.

22 This expression can also be understood to mean it is impossible to claim when the slave is no longer extant, due to death, or vanishing because of abduction, etc...

23 On which see Ibn Rushd al-Jadd, Fatāwā: I, 631-649. The full contents of the fatwā as well as the historical circumstances that surround it were addressed in my paper “Sharia and rule under the Almoravids”, presented at the International Society for Islamic Legal Studies VII Conference. Islamic Law and State: Doctrine and History held in Ankara (Turkey) 30 May-1 June 2012.

24 On the istirʻā’ typology and its use in legal contexts different from ṣulḥ see Müller, 1999: 180-181 & 183-188 (“Memorialurkunde”); Powers, 2002: 31 n. 35; 2006: 402-403 & n. 9. Powers points out that the acceptance of a rasm istirʻā’ (“memorial document”) by the judge is “conditioned by his assessment of the integrity of the professional witnesses who recorded the testimony”.

25 See Ibn Abī Zamanīn, Muntakhab: II, 752, though it might be a confusion with walī or matrimonial guardian

26 For an illustration of this see al-Qurṭubī, Jāmiʻ: VI, 290 & 297.

27 See, for example, Ibn Qudāma, Mughnī: IX, 107, quoted by El-Ahdab & El-Ahdab, 2011: 14.

28 We refer to the opinions selected by the author of the entry, not the sections on arbitration included in the Mālikī manuals consulted by him.

29 Secondary because they are later and less authoritative than Mālik’s Muwaṭṭa’ or Saḥnūn’s Mudawwana, which do not appear to have catered for Bājī’s interest, though they do certainly address the issue of arbitration of marital disputes as Fierro’s work shows.

30 As per Bājī’s explicit identification of his sources. None of the edited fragments of the Wāḍiḥa – which has not been preserved in its entirety – appears to include any section on taḥkīm. On the existing manuscripts and fragmentary editions of the Wāḍiḥa and their contents, see M. Arcas Campoy 2002: 28. Ibn Abī Zayd al-Qayrawānī's al-Nawādir wa-l-ziyādāt is not mentioned either by al-Bājī or by Ibn ʻAbd al-Barr in that connection, which appears to indicate that Andalusī doctrine on taḥkīm emerges in parallel with but independently from its Ifrīqiyan Mālikī counterpart.

31 Extra-judicial in the sense of taking placing outside the courts and other spaces associated with the activity of the qāḍīs and the fuqahā’.

32 Talking about Saudi qāḍīs, Frank Vogel (2000: 153) observes that “Cases that proceed to final judgement are rare: most are settled by agreement of the parties usually with the qāḍīs’ assistance”.

33 See Muḥammad b. ʻIyāḍ, Madhāhib: 36-37.

34 On actual legal cases reporting attempts to use conciliation to circumvent the law (including a reference to Ibn Rushd’s reaction towards this possibility), see Hentati, 2007: 189-192.

35 To judge by Ibn Farḥūn’s doctrine on arbitration (Tabṣira: I, 50-51), we should rule that possibility out, at least with regard to 14th century Egyptian Mālikīsm.

Haut de page

Pour citer cet article

Référence papier

Delfina Serrano, « Bringing arbitration (taḥkīm) and conciliation (ṣulḥ) under the qāḍī’s purview in Mālikī al-Andalus (10th to 12th centuries C.E.) »Revue des mondes musulmans et de la Méditerranée, 140 | 2016, vol 140, 73-100.

Référence électronique

Delfina Serrano, « Bringing arbitration (taḥkīm) and conciliation (ṣulḥ) under the qāḍī’s purview in Mālikī al-Andalus (10th to 12th centuries C.E.) »Revue des mondes musulmans et de la Méditerranée [En ligne], 140 | 2016, mis en ligne le 30 janvier 2017, consulté le 02 octobre 2023. URL : ; DOI :

Haut de page


Delfina Serrano

Institute for Languages and Cultures of the Mediterranean and the Near East-CSIC, Madrid

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC-BY-NC-SA-4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Search OpenEdition Search

You will be redirected to OpenEdition Search