Al-Shtifi and the Exclusionary Project

“O Prince of Believers, what would you say of two men, one sees me as a brother and the other sees me as a slave. Whom would I prefer?” He [Han] said, “The one who sees you as a brother.”3?؛ Al-Shfi'i then re­sponded, ‘That is you [your clan], Prince of Believers, you are descen­dants of al-Abbas, they are descendants of 'Alr, and we are Banb l-Muttalib. You, descendants of al-Abbas, see us as brothers and they [de­scendants of Alr] see us as slaves.”4?؛

If true, this response was probably intended to achieve two important objec­tives: first, to refute logically the accusation, and second, to emphasize al- Shafrs noble lineage and his blood relation to the caliph.5?؛ Impressed by al-Shahr knowledge and ability, al-Rashrd is reported to have asked al-Shafr'i for advice on managing the caliphate.

ensure that the haram (sacred house) of god and the haram of his proph­et are renovated, secure the roads, manage affairs of the nation and give the descendants of the muhajirun [those who left Mecca to Medina with the Prophet] and the ansar [the inhabitants of Medina who welcomed the Muslim immigrants] their share of the booty.6?؛

Despite credibility doubts shed on a!-Razr's narrative,؛?? it is noteworthy that the above quote specifically mentions division of the booty as a matter of prime importance. Singling out the booty system as one of the “salvation” mea­sures recommended to al-Rashrd must be read in light of the Abbasid revolu­tion's appeal to those who were deprived of their booty rights by the Umayyad dynasty.8?؛ Moreover, this focus proves more interesting when it is recalled that

173 AbtrZahra, O-Shafff, 23.

-η⅛ AbbZahra, dtshafff, 23.

175 Fakhr al-Drn al-Razr, Manaqib, 73.

176 Fakhr al-Drn al-Razr, Manaqib, 79.

177 It should be noted that Fakhr al-Drn al-Razr's account of the mihna is subject to serious challenges. For example, he claims that Abu Yusuf and al-Shaybanr attempted to antagonize the caliph against al-Shafi'r and to assert his affiliation with 'Alrd insurgencies. See Fakhr al-Drn al-Razr, Manaqib, 71. But according to al-Razr, the mihna happened in 184/800, whereas Abu Yusuf died in 182/788. Moreover, other accounts such as Abu Hatim al-Razr's (see Adab, 32) state that al-Shaybanr defended al-Shafi'r against the accusation and that al-Shafi'r became his student afterwards.

178 Faruq 'Umar, al-Abbasiyyun al-awaHl (Beirut: Dar al-Irshad, 1970), 1:36, 73-75. al-Shafi'i was a member of al al-bayt and was accordingly entitled to a booty share that he had not received prior to this incident.

Al-Bayhaqi claims that al-Rashid offered al-Shafi'i money after that debate and requested that he hold a judicial position, but al-Shafi'i refused to be ap­pointed as a judge, distributing the money among the poor when he left the caliph's palace.

Sources disagree over the reasons behind a!-Shafi'i's choosing to reside in Egypt. Abfi Zahra argues that he left Baghdad, around 198-9/813-14, because he did not wish to live in the capital during al-Ma'mfin reign as caliph, partly because “Persians had the upper hand.... It was not possible for this Qurayshi to accept living under a Persian government,” and partly because of the rising influence of the Mutazila, whom al-Shafi'i thought “repulsive” and whose ap­proaches he rejected.184 Al-Bayhaqi, on the other hand, seems to argue that al-Shafi'i went to Egypt to spread his jurisprudential views in a territory other than the two established regional camps in Iraq and the Hijaz.185

Regardless of the reasons behind a!-Shafi'i's move to Egypt, one can witness how his strong Arab lineage and the powerful Islamic empire contributed to his comfort and his revered intellectual independence. Had it not been for his blood relationship with the Prophet (and thus the caliph), he would not have been able to establish his noble status to al-Rashid, and he would not have re­ceived a share of the booty. Similarly, as the 'Abbasid empire expanded and

179 al-Bayhaqi, Manaqib, 152.

180 AbfiZahra, CttShafH, 25-27.

181 Amin, Duha, 2:220.

182 Khadduri, introduction to Risala, 17.

183 Kecia Ali, Imam Shafih: Scholar and Saint, Kindle ed. (Oxford: One World Publications), loc. 807.

18٩ AbfiZahra, a--Shaff't,28.

185 al-Bayhaqi, Manaqib, 238.

consolidated its power, it received generous revenue, which reflected on the wealth of those entitled to shares in caliphal revenue, including al-Shh'i. Both factors, al-Shafir Arab/Islamic lineage and the strength of the Islamic empire, seem to have influenced al-Shahis approach to the Islamic regulation of armed conflict, as established later in this chapter.

2.2 AL-Shafi^cVs Theory ofUsul al-Fiqh

Al-Shafir is argued by many to have been the earliest jurist to have dedicated a book, al-Risala, to the rationalization of jurisprudence and the theorization of the hierarchy of the sources, a held of study later referred to as usul al-fiqh. Hal- laq, however, argues that al-Risala was written to establish the authority and supremacy of the Qur'an and sunna rather than to set a theory for legal reason- ing.186 Al-Shafirs prominent role in early accounts of Islamicjurisprudence was thus attributable more to his synthesis of a reconciliatory approach to jurisprudence, an approach that affirms supremacy of traditions yet asserts the need for q،yas.187 Hallaq concedes, however, that the reader can detect some hypotheses about legal reasoning in the book.188 This section provides a brief summary of these hypotheses, referred to loosely in the rest of the chapter as al-Shafr^tr^ls theory of usul.

Lowry, who agrees with Hallaq that defending the sunna as an authoritative source of law was a primary objective of al-Risala,189 argues that the text as­pires to “provide a general account of the fundamental structure of Islamic law, in which prophetic authority is only one element.”19٥ He argues that the com­mon understanding of al-Risala as a treatise focused on the “four” sources of law is erroneous because al-Shafr^tr does not consistently refer to four sources; sometimes, he refers to two, three, or more sources.191 Rather, Lowry posits that al-Risala is an attempt to understand bayan (expression of God's will to humans) - a term frequently used in the text - primarily through the Qur'an and sunna. In other words, al-Risala is concerned primarily with how we un­derstand bayan in a particular issue given the different channels ofrevelation.192

In al-Risala, al-Shafr argues that God illustrated Man's obligations through four channels: through the Qur'an (as in the prohibition on eating pork and

186 Hallaq, “Master Architect?” 592.

187 Hallaq, “Master Architect?” 600.

188 Hallaq, “Master Architect?” 592.

189 Lowry Raa,8.

1-90 Lowry Raar

191 Lowry Raa,i2.

192 Lowry Rtsala,24.. consuming wine); through the Quin with the specific method of upholding the obligation elaborated by the Prophet (as with prayer); through the Prophet, without any Quranic text indicating such an obligation; and through Ijtihad (the human obligation to exert one's best effort in seeking God's satisfaction).^

2.2.1 The Qur'an: General and Specific Verses and Mastery of Aiabic Al-ShafiT contends that verses of the Qur'an exhibit the unique feaPires of the Arabic language. They are either general with the intent of generality, general with the intent of specificity, specific with the intent of generality, or specific with the intent of specifici(y.!94 General verses can sometimes simply indicate general application. For example, “There is not a creaPne that moves on earth whose provision is not His concern''195 indicates that God is responsible for all living creatares.i90 Other verses could appear general, but in fact indicate both general and specific meanings. For example, “People, We created you all from a single man and a single woman, and made you into nations and tribes, so that you should get to know each other. In God's eyes, the most honoured of you are the ones most aware of him''197 is general and specific. With regards to the creation of “mankind” into nations and tribes, it is general, but it is also spe­cific in stating that the righteous is the most honoured because the classifica­tion upon the basis of righteousness is limited to those who are capable of it as indicated in other instances in the Qur'an and sunna.198 Finally, he argues that some verses appear general, but their intention is acPrally specific, such as, “Those whose faith increased when people said, ‘Fear your enemy, they have amassed a great army against you"'199 because “people” excludes the Prophets CompanionsOO Finally, God commands pilgrims to “surge down where the rest of the people do”20i is also specific because “it is certain that not all the people had been present at Arafa during the time of the Apostle, and that it was [only] the Apostle and those who had been with him were those to whom this communication was addressed.''202

Al-Shafi'i argues that examples of general verses intended for a specific meaning prove that there is a need for mastering the Arabic language in order to understand the intention of the verses in terms of the specific and the gen­eral. He says that while a non-Arab would find the first verse the easiest to un­derstand, the second slightly more difficult, and the third much more difficult, an Arab would find them equally clear.203 As pointed out by Lowry, al-Shafirs analysis “feels much more like a theological claim about the special qualities of Arabic than an assessment of Arabic as a linguistic phenomenon.'^ In the above-mentioned verses, al-Shafi'i does not clarify why the ability to compre­hend them is subject to proficiency in Arabic rather than to awareness of the general context of the Qur'an and sunna. In the second verse, one would as­sume that a non-Arabic speaker would understand that believers are excluded from their enemies mentioned in the verse. The third verse refers to pilgrims following the footsteps of others. Naturally, that would mean that the followed cannot be “all people.”

Considering al-Shafirs emphasizing that command of the Arabic language should be a basis for jurisprudential reasoning on linguistic grounds, Abfi Zayd argues that al-Shafirs defence of the Arabic language and his claim in al-Risala that the language, and subsequently the Qur'an, were “pure” of foreign influ­ence and infiltration indicates an “ideological bias” for Arabic and, more spe­cifically, for the Qurayshi dialect. According to Abfi Zayd, this bias reflects Arab favouritism and aresentment ofthe rising Persianinfluence inAbbasid times.205 Of course, undeniably, al-Risala was intended primarily at asserting the su­premacy of Islamic sources and providing “a universal basis for Islamic judg­ment... [to] supersede the various legal traditions.”2٥6 But one cannot disregard the fact that emphasizing the Arabic language as a key tool for comprehending sources inevitably sets the Arab local tradition as a benchmark to interpreta­tion of the universal one.

2.2.2 The Sunna and Its Role in the Interpretation of the Qur'an As Hallaq argues, one of the primary objectives of al-Risala is “to define the role of the Prophetic Sunna in the law.”2٥7 Al-Shafi'i does not recognize the

203 al-Shaβ'i, Risala, 101.

204 Lowry, “Hermeneutics,” 10.

205 Nar HSmid Abfi Zayd, al-ShafiTwa tasis al-idiyulujiyya al-wasatiya (Cairo: Sina li-l-Nashr, 1992), 15-16.

206 John Kelsay, “Divine Command Ethics in Early Islam: al-Shaβ'i and the Problem of Guidance,” TheJournal OfReligious Ethics 22, no. 1 (1994), 103.

207 Wael Hallaq, A History OfIslamic Legal Theories (Cambridge: Cambridge University Press, ¹997), ²9.

possibility of contradiction between the two main sources of law, the Quin and the sunna. Accordingly, the two sources do not and cannot abrogate each other.208 However, the sunna could specify and explain the Qurtin. He argues that the sunna is a binding legal source because God has indicated in several incidents that all Muslims should obey the Prophet.2٥9 The sunna, according to him, can confirm rules in the Qurn elaborate on obligations referred to gen­erally in the Qurtin, introduce a new rule not referred to in the Qur'tin,2!9 specify general verses,211 and provide evidence for abrogation of older verses.212 Interestingly, the issue of booty comes to prominence again when he provides rules of inheritance as an example of sunna specifying general verses. He ar­gues that verses addressing inheritance were specified by the sunna that re­quired unity of religion as a condition for inheritance2i3 and specified Banti !-Mutalib and Banti Htishim as the sole blood relatives of the Prophet entitled to a share of the booty.214

Finally, al-Shtifir accepts al-khabar al-wahid, but the person “must merit confidence in his religion, and be known as reliable in his transmitting, com­prehending what he transmits, aware of any pronunciation that might change the meaning of the tradition, [and] capable of transmitting the tradition word for word.''215 Again, the criteria for transmission assert mastery of Arabic as a condition for transmission of hadlth.

2.2.3 The relationship between the Qurtin and the Sunna

The sunna cannot abrogate the Qurtin, because “God has declared that He ab­rogated [communications] of the book only by means of other communica­tions in it; that Sunna cannot abrogate the book, but that it should only follow what is laid down in the Book.”2i6 To support this argument, al-Shtifi'i relied on a verse where the Prophet rejected others' demands for another Qurtin, saying, “It is not for me to change it of my own accord; I only follow what is revealed to me.''2i7 He equally refirsed to allow the Qurtin to abrogate sunna in order to

prevent abandonment of sunna,2i8 and he argued that if a sunna were abro­gated by the Qur'an, a new sunna abiding by the new obligation or rule would exist, and in that case the sunna would be abrogated by the later sunna rather than by the Qur’an.219 Because he was reluctant to allow either of the two sources to abrogate the other, al-Shafi'i reftrsed to acknowledge the possibility of conflict between the Qur'an and the sunna.220

Although al-Shafi'i ruled out the possibility of conflict between the Qur'an and the sunna, he acknowledged contradiction among KadIths. He initially called upon jurists to attempt to reconcile the contradicting hadlths with each other. If reconciliation attempts failed, “one should not be abandoned for the other except for a reason that shows that the one adopted is stronger.''221 When asked what that reason would be, al-Shafi'i stated that the KadIth more consis-

tent with the Qur'an was the one that should be adopted.222

2.2.4

Other Sources

Al-Shafi'i also recognized ijmd^c, because “we know that the public can neither agree on anything contrary to the sunna of the Prophet nor on an error.''223 Al- Shafi'i acknowledged qiyds, more a tool than a source in itself, as one of the approaches to formulating legal rules. To him, qiyds was the extension of rul­ings revealed in the textual sources to analogous situations on which said sources were silent. “If the case in question is similar to the original meaning [of the precedent], no disagreement on this kind [is permitted]. The second, is applying qiyds on the basis of precedence if the examined situation resembles an established precedent.''224 Finally, he rejected istihsdn, which he defines as stating an opinion without relying on a textual source or qiyds.225 To sum up, al-Shafir project was aimed at emphasizing the role of the KadIth in juristic interpretation, which led him to be “the first jurist consciously to articulate the notion that Islamic revelation provides a full and comprehensive evaluation of human acts.''226 This focus on textual sources will be the primary benchmark

for assessing consistency in al-Shafir work on jihdd.

218

218 al-Shaβ'i, Risdla, 126.

219 al-Shaβ'I, Risdla, 126.

220 al-Shaβ'I, Risdla, 184.

221 al-Shaβ'I, Risdla, 204.

222 al-Shaβ'I, Risdla, 213.

223 al-Shaβ'I, Risdla, 286.

224 al-Shaβ'I, Risdla, 290.

225 al-Shaβ'I, Risdla, 305.

219

221

222

223

226 HaIIaq, Origins, 117.

2.3 Al-S٠I CLultheRgationofAimedConflict

2.3.1 Islamic Hegemony and Revenue 2ا.اه DeJlnitionof Peopieof heBook

Most jurists agree that the Qur'an indicates that the People of the Book are granted the choice between payingjizya and converting to Islam, but the treat­ment of other non-Muslims spurred significant debate. The primary trigger for this debate was a hadlth in which the Prophet is reported to have ordered that the magi be treated like the People of the Book when he was asked about their status. Bonner argues that most jurists eventually agreed that the magi were monotheists of some kind,227 but his argument disregards the nuances of the legal arguments on the matter. Although somejurists like alShafi'i adopted this position, al-Shaybanι228 and the Malik⅛229 argued that the magi were not People of the Book and that acceptingjizya from them was proof that the re­gime applied to non-scriptuaries (whose religion was not based on a “holy” book).

Al-ShafiT argued that the objective ofjihdd is to convert non-Muslims to Is­lam, citing numerous Quianic verses to that effect. In addition, he cited this hadlth to indicate that infidels, with the exception of the People of the Book, must convert or face death:

I have been commanded that I should fight against people till they de­clare that there is no god but Allah, and when they profess it that there is no god but Allah, their blood and riches are guaranteed protection on my behalf exceptwhere it is justified by law, and their affairs rest withAlah.230

Thus, identification of the People of the Book has serious implications for tar­geting, since a!-Shafi'i's hypothesis leads to the conclusion that any adult male23! who is not a follower of a scripted religion may, and should, be targeted unless he converts to Islam. In order to support his argument that the magi are People of the Book, al-Shafi'i offered textual evidence from a prophetic hadlth, a Com-

227 Michael Bonner,y،hadin Islamic History (Princeton, NJ: Princeton University Press, 2006), 89.

228 al-Shaybani, Siyar, 224.

229 Mξ⅛k b. Anas, al-N^ndawc^na CtLkubra i----ImUm Mdiikb. AnUs ctl-^?^؛^؛^c^L^l fBe⅛ut. Dtff al- Kutub alllmiyya, 1994), 1:529.

230 Sahih Muslim, 1:32, trans. Abdul Hamid Siddiqui, Center for Muslim Jewish Engagement, University of South California, online:, accessed 30 December 2010.

231 With the exception of monks, who are dealt with later in the chapter. panion's statement ('Aii b. Abi Tib), and interpretations based on qiyas. The first evidence, the prophet's hadlth, is one that is used by both juristic camps equally. According to al-Shafi'i, 'Umar only tookJizya from the magi after a Companion testified that the Prophet had taken it from the Hajar232 magi and that he had heard the Prophet say, “Follow the same sunna with them that you follow with the People of the Book.”233 Al-Shafir says,

If he [the Prophet] meant all infidels other than People of the Book, he would have said, and God knows better, “Follow with all infidels the same sunna that you follow with the People of the Book,” but since he said, “Follow with them,” then he specified them, and if he specified them, then others ae different andthose others are Oscriptuaries [do not-fodowa holy book].234

He additionally relied on the Qur'an's reference to the existence of holy books other than the Torah and the Bible to prove that the magi might be People of the Book.235 He also relied on a statement that 'Ali is supposed to have made regarding the status of the magi. According to that tradition, two men, arguing over the reason for taking Jizya from the magi, went to ask 'Ali. 'Ali said the magi had a book, but their king violated it, and to avoid the punishment stipu­lated therein, forged the religion.236 Al-Shafi'i used this incident to prove that the magi were People of the Book, arguing that this incident was evidence that “'Ali... did not know that the Prophet (pbuh)... tookjizya from them except because they are People of the Book. If Jizya were to be taken from anyone other than People of the Book, 'Ali would have said” so.237 He also argued that no Companion cited the Prophet as having takenjizya from non-scriptuaries.238

Additionally, al-Shafi'i argued that Jizya is not dependent on ethnicity or lineage, which precludes differential treatment for non-Arab non-scriptuaries. To prove thatJizya is not dependent on nasab (lineage), al-Shafi'i stated that the Prophet had takenjizya from a man from Ghassan, from Arab People of the

232 In modern-day Bahrain.

233 al-Shafi'i, al-Umm, 4:245-46. Translation provided in Malik, Muwatta^y, 17.24.43, online:, accessed 29 December 2010.

234 al-Shaβ'i, al-Umm, 4:246 (my italics).

235 al-Shaβ'i, al-Umm, 4:245.

236 al-Shaβ'i, al-Umm, 4:245.

237 al-Shaβ'i, al-Umm, 4:245.

238 al-Shaβ'i, al-Umm, 4:245.

Book in Yemen and Najran.239 According to him, the Prophet takingJizya from Arab People of the Book proves that ethnicity does not play a role in determin­ing status. In other words, he is arguing that there are possibly two criteria for determining whether or not a person is allowed to payJizya, rather than being forced to convert. The first could possibly be ethnicity and in that case, Arabs would be denied the right to payJizya, while non-Arabs would be allowed to pay it regardless of religion. To refute this argument, he states that the Prophet tookJizya from Aiab People of the Book. By virtue of qiyas, he deduces that if Arab People of the Book are allowed to pay the Jizya, then the determinant fac­tor for eligibility forJizya payment is not ethnicity. Accordingly, it is not possi­ble to claim that non-Arabs are allowed to pay Jizya on the basis of ethnicity and regardless of their religion. Since ethnicity is not the determinant factor of the right to payJizya, then we are left with religion as the sole determinant of who may or may not pay Jizya, with People of the Book entitled to claim the right to pay Jizya and maintain their faith, and other non-Muslims left choos­ing between converting to Islam and death. In addition to qiyas, al-Shafi'i de­clares that allowing non-scriptuary non-Arab people to pay the Jizya, while denying non-scriptuary Arabs the right to pay it, violates the principle of equal­ity, because God never differentiated between Arabs and non-Arabs, neither in belief nor in disbelief 240

Despite his extensive arguments for including the magi among the People of the Book, one can detect several attempts to interpret the texts in a!-Shafi'i's arguments that are not in congruence with his al-Risala work on legal theory. First, he relies on the Prophet's hadtth, “Follow the same sunna with them that you follow with the People of the Book,” to prove his argument. However, this reliance in itself is rather problematic for a formalist who claims that under­standing the details of linguistic expression is necessary to understand the text. Even a person who is not a native speaker would understand from the language of the hadtth that the Prophet instructed similar treatment but did not state that the magi were People of the Book. In fact, a strictly literalist in­terpretation would argue that the analogy drawn between the magi and the People of the Book is specific to the sunna to be followed with them (in other words, payment of Jizya). Following the approach adopted by al-Shafi'i in his linguistic interpretation of general and specific Quranic verses, the reference to similar treatment of “them” and “the People of the Book” indicates the exis­tence of two distinct groups (People of the Book as belonging to one group and the magi to the second). Moreover, al-Shafi'i commented on the hadtth and

239 al-Shafi'i, al-Umm, 4:244-45.

240 al-Shafi'i, al-Umm, 4:342. said that if the Prophet had been referring to all non-Muslims, he would have said so.241 In this specific case, he allows his literalist interpretation to read the hadlth in a narrow and specific manner that prevents the inclusion of other People of the Book in the ruling regarding the magi. However, he disregards the fact that his own argument may be used against his inclusion of the magi as People of the Book. Similarly, arguably if the Prophet had wanted to say that the magi are, in fact, People of the Book, he would have said so, rather than instructing that they be treated like People of the Book.242

Although al-Shahr reliance on Alfs statement arguably is a possibly strong proof for his argument, it is still inconsistent when read against his approach to the positions of the Companions in dealing with armed conflict. For exam­ple, a later section of this chapter shows that al-Shafi'i deviates significantly from Abfi Bakr's views on targeting old men and monks. Moreover, his claim that Companions did not report that the Prophet ever accepted from fol­lowers of any religion other than the People of the Book is self-inferential since he uses his own arguments that the magi are People of the Book to neutralize the potential interpretation of this tradition to indicate that the Prophet may have accepted Jizya from non-scriptuaries. In other words, he acknowledges that the Prophet tookjizya from the magi, but he relies on his own understand­ing of the magi as People of the Book to prove that Companions never reported that the Prophet accepted Jizya from non-scriptuaries, which in turn is relied on to prove that the magi are People of the Book.

Finally, his argument on the basis of the principle of equality is equally problematic, because the Prophet is reported by many jurists, including al- Shafi'i, to have granted all Meccans mercy without any condition, regardless of religion.

He prohibited their killing, except for some individuals he named unless one of them fought and was killed [on the battlefield]. When gathered in the mosque, he told them, “What do you think I will do to you?” They said, “Good [treatment]. [You are] a generous brother and the son of a generous brother.” He said, “Go, you are freed,” and he didn't take as booty anything, whether big or small, from their property.243

True, al-Shafi'i argues that Mecca was “conquered” on the basis of an aman safety pledge, thereby protecting the property of its residents, but many other

241 al-Shaβ'i, al-Umm, 4:246.

242 Badawi, “Sunni Islam,” 310.

243 al-Shaβ'i, al-Umm, 7:593-594. jurists argue that Mecca was conquered by force.244 Moreover, al-Shafi'i pro­hibits imams from concluding long-term agreements with unbelievers con­taining conditions that contradict objectives of jihad. Since jihads primary objective according to him is to convert unbelievers, with the exception of People of the Book, or otherwise they must be killed, then the Mecca case would still be a special case of preferential treatment, even if it were conquered on the basis of a safety pledge.

As mentioned earlier, the suggestion that the magi are People of the Book limits the potential for followers of any other religion to “benefit" from the pro­tection guaranteed under this system, thereby expanding the destructive pow­er of the army in the conduct of armed conflict with them. Yet, the suggestion is particularly expansive of military might with regards to the acquisition of booty. Although People of the Book may avail themselves of the jizya regime before the commencement of battle - thereby protecting their property, in­cluding their women and children, who may otherwise be taken as captives - others are denied this privilege. Thus, if they refuse to convert, they may be fought with the objective of eliminating them and acquiring all their property, rather than imposing a simple fixed tax.

This approach undeniably asserts the supremacy of the Muslim caliphate in its relationship with its others by emphasizing their complete and full assimi­lation through conversion to Islam, or in other words, by emphasizing the elimination of the enemy as a political entity.245 It also guarantees the Muslim caliphate a more abundant source of revenue through the potential wholesale acquisition of enemy property. As stated previously in this chapter, the Islam- ic-Byzantine border had reached a stalemate in terms of potential for expan­sion of the Muslim state, and the wars conducted between the two empires had reached their climax in terms of Islamic expansion, and “Muslim fighters no longer returned loaded with booty and slaves."246 This change is likely to have guided al-Shafi'i to direct his attention to another potential venue for ex­pansion, the east, where most Turks, Indians, and Chinese were not followers of any of the religions of the People of the Book according to his criteria, and whose property would all be subject to confiscation in the case of victory. One can even imagine how al-Shafi'i may have been responding to what he is likely to have perceived as illegal conduct of war in India, where jizya was accepted from Buddhists and Hindus for pragmatic purposes.247

244 Ibn Rushd (alHafid), Bidayat, 2:163.

245 Badawi, “Sunni Islam," 310.

246 AJmar, al-KhtlaJa al'Abbasiyya, 335.

247 Hugh Kennedy, The GreatArab Conquests (London: Wiedenfield and Nicolson, 2007), 306.

2..3,.1-.2. DeliberateTargeting

Like other Muslim jurists, al-Shafi'i argued that Muslims and non-Muslim women and children may not be targeted. But much like his rule formulation with regards toJizya, there are some gaps in the internal coherence of his argu­ments on targeting. Similarly, the overall approach to targeting offers strict limitations on the protection of the other and expands the military's powers to ensure that maximum harm is inflicted on the enemy.

Unlike his mentors, Malik and al-Shaybanr, al-Shafi'i argued that all adult men, with the exception of monks, should be targeted. As detailed below, the primary proof he relied on was the alleged lack of textual sources prohibiting the targeting of anyone other than women or children. It should be noted that al-Shaybanr relied on a prophetic tradition to prove that the Prophet specifi­cally instructed refraining from targeting older men. However, al-Shafi'i made no reference to this tradition, despite his alleged familiarity with al-Shaybanr's works. Malik, on the other hand, primarily relied on a widely cited instruction given by Abh Bakr to the commander of the army:

You will find a people who claim to have totally given themselves to Al­lah. Leave them to what they claim to have given themselves. You will find a people who have shaved the middle of their heads. Strike what they have shaved with the sword. I advise you ten things: Do not kill women or children or an aged, infirm person.248

Although al-Shafi'i states that he follows Abh Bakr's speech, except with re­gards to a later part prohibiting the burning of trees,249 he seems to ignore that the speech ordered soldiers not to kill old men.250 It is hard to assume that al- Shafi'i was unaware of the speech, because Malik's al-Muwatta^, is confirmed to have been memorized by al-Shafi'i.251 But his al-Umm does not refer to that element of the speech, acknowledging only the prohibition on targeting of monks.252 In fact, in several other instances, he argues that monks may be tar­geted as well. It is unclear why al-Shafi'i would omit such reference, despite his declared knowledge of Malik's book, especially given that he could have re­jected the rule, as he did with the prohibition on the burning of trees. However,

248 Malik, Muwattaد, online:, accessed 18 January 2011.

249 al-Shaβ'i, al-Umm, 4:369.

250 Malik, Muwatta^y, 297.

251 AbhZahra, al-SaJi't,19.

252 al-Shaβ'i, al-Umm, 4:340.

this rejection obviously gave him more grounds to build a case for targeting older men and any man regardless of whether or not he is fighting.

If someone said, “What is the proof that infidels who are not fighting may be killed?” We say, “Companions of the Prophet (pbuh) killed Durayd b. alSumma.... [H]e was around one hundred and fifty years old, and the Prophet did not object to his killing, and I do not know any of the Mus­lims objecting to the killing of unbeliever males except for monks.253

Of course, one could argue that the lack of reference to those texts is possible proof for Calder's thesis that al-Shaybnis al-Asl and Malik's al-Muwatta’ were falsely attributed to the two masters. But as previously mentioned, Calder's hypothesis has been widely challenged by several scholars.254 Moreover, al- Shfiis refutation of Mhliki theory in al-Umm makes numerous references to traditions in al-MIwa#a,,255 suggesting prior circulation of the text, as well as his own familiarity with it.

Al-Shfii approach to monks is indicative of a similar inclination to assert permissibility of targeting them, thereby expanding the power of the army in its conduct of military operations.256 Although al-Shafi'i admits the authorita­tiveness of Abfi Bakr's tradition with regards to monks, he argues in another part of the book that Abfi Bakr never prohibited targeting monks and that if it were proven that he did, then he must have prohibited their targeting to avoid the army's distraction with those who are not likely to fight them and to en­courage them to engage in military combat that is likely to inflict more damage on the enemy.257 Al-Shfi'i was thus clearly aware of the reference to the pro­hibition of targeting monks in the Abfi Bakr speech, even when he instructed that monks be targeted. It is unclear whether the difference in opinions on the matters in the same book were outcomes of his or his disciples' revisions of the book or the views reflected a change of position on al-Shfi'!'s part, since it is well established that al-Sh fii extensively revised hisjurisprudence. But we have no tools to assist us in speculation over which view predated the other. Whether contradictions over the positions of monks reflected a change of view, what is most important is that al-Shfi'1 was keen to mitigate Abfi Bakr's

253 al-Shafi'i, al- Umm, 4:340.

254 See e.g. Wael Hallaq, “On Dating Malik's Muwatta,” UCLAJournal of Islamic and Near Eastern Studies, 1 (2001-2), 47-65, and Lowry, “Hermeneutics,” 1-42.

255 al-Shafi'i, al-Umm, 7:307-24.

256 Badawi, “Sunni Islam,” 331.

257 al-Shafi'i, al-Umm, 4:406.

instructions in a manner that expanded the permissibility of targeting, even with categories he concedes were addressed therein.

2,3>A3> Idscbrnate Targeting

Although al-Shafi'i acknowledges the prohibition on killing non-Muslim wom­en and children, he argues that the prohibition applies to deliberate targeting rather than infliction of death on them during the course of action. However, if one traces the language used in his approach to targeting, one finds that he regards the targeting of women and children permissible in almost all situa­tions except in cases of individual targeting. His interest in expanding the free­dom of targeting by Muslim combatants is perhaps best understood if one contrasts the language used in the directives regarding the killing of non-Mus- lim women and children, on the one hand, with the prohibition on targeting Muslims, on the other hand, when launching an offensive against a non-Mus- lim army. Most jurists accept the rule that children and women may not be targeted. However, there seems to have been an extensive debate and a strong intellectual interest in the limitations on such prohibition, particularly in the use of “indiscriminate weapons.” Although one might argue that using arms against a diverse group of people, knowing that the attack is likely to inflict harm on all of them, is an act of collective targeting, al-Shafi'i argues that the Prophet only prohibited the deliberate singling out of women and children for targeting.258 To prove this assertion, he relies on several prophetic traditions that include the use of hurling machines against the people of Ta'if and the Banti l-Nad,259 as well as on a hadlth attributed to the Prophet, in which the Prophet is reported to have been asked about the death of women and children during night raids and responded by stating that “they [women and children] are from them.”26٥ This hadlth is understood by al-Shafi'i and others to indicate that these women and children may be indiscriminately targeted during con­flict.

And if the enemy shields itself..., there is nothing wrong with attacking them with mangonels, catapults, fire, scorpions, snakes, or anything they hate, or open up water on them so that they drown or get buried in mud, regardless of whether they have children, women, or monks with them.261

258 al-Shafi'i, al-Umm, 4:347-48.

259 al-Shafi'i, al-Umm, 4:347.

260 al-Shafi'i, al-Umm, 4:337. The editor of al-Umm, Malmud Mutraji, provides evidence that this prophetic tradition was abrogated by other traditions. See al-Umm, 4:337-338. Al- Shafi'i refutes this claim in al-Risala. See Lowry, Risala, 132-34.

261 al-Shafi'i, al-Umm, 4:347.

Lowry argues, in his analysis of al-Shdfir al-Risala, that al-Shfir claim that women and children may be killed in the course of fighting, but may not be individually targeted, in fact reflects an attempt to balance two seemingly conflicting hadlths, the one generally prohibiting the killing of women and children, and the other declaring that women and children killed during night raids are “from them.”262

Although an aversion to directly targeting women and children may have led al-Shafi'i to the said conclusion, or perhaps the influence of other scholars such as al-Shaybn as previously discussed, al-Shafi'i adopts a different tone when addressing the issue of Muslim presence in non-Muslim lands.263 Despite a prohibition on killing non-Muslim women and children (or, at least, a direc­tive not to do so), a!-Shfi'i does not attempt to limit the use of “indiscriminate weapons” if they are present in targeted areas. But, when he anticipates the presence of Muslims in the land, he shows more reluctance to give soldiers the same freedom in attack.

If they [the enemy] shield themselves with Muslim or non-Muslim chil­dren and the Muslims have already engaged in combat, then there is no problem in targeting the warriors and avoiding the Muslims and chil­dren. If they were not engaged, I prefer that they refrain from attacking them until they can fight them unshie!ded.264

If there are Muslim captives in the territory or traders given aman, I hate attacking them with methods that are not discriminate such as burning or drowning or anything similar, but I do not blatantly prohibit it, for if the land is permissible [for attack], nothing indicates that the presence of a Muslim in it makes it prohibited. But I say it is hated [to resort to those methods] out of cautiousness, because we are allowed to avoid attacking it [even] if Muslims are not present in it, so if we fight this territory, we should attack it without using indiscriminate weapons such as burning or drowning.265

Al-Shfi'i thus portrays a bias similar to al-Shaybnr for protecting Muslim life, despite the existence of textual sources equally prohibiting targeting both Muslims and non-Muslim women and children. Nevertheless, he goes a step further than al-Shaybn, in that he states that causing the death of a Muslim

262 Lowry Rssala,132-3¾.

263 Badawi, “Sunni Islam,” 335.

264 al-Shafi'i, al-Umm, 4:409.

265 al-Shafi'i, al-Umm, 4:348. makes the warrior liable for kaffara (repenting through commission of a par­ticular act) via the manumission of a slave.266 But contrasting the two posi­tions shows a more genuine interest in sparing Muslim lives. This interest allows al-Shafi'i to resort to an approach that he harshly criticizes, namely Istihsan, through instructing the army to avoid territories where damage is likely to be inflicted on Muslims. Yet, he is reluctant to employ the same rea­soning when referring to the exclusive presence of non-Muslim women and children and the permissibility of an indiscriminate attack. Of course, one might argue that the prophetic hadlth allowing for the death of women and children during the course of fighting provides al-Shafi'i with some basis for making such a distinction between the presence of Muslims and the presence of non-Muslim women and children. But drawing an analogy between the un­intentional deaths of women and children during a night raid and the deliber­ate targeting of an army, despite the use of women and children as a shield, is not accurate, especially if we see that in other instances, al-Shafi'i envisions the potential performance of this attack with a mitigated attempt to avoid groups reprehensible to inflict loss of life on such as Muslim captives.

Both al-Shafi'i's and al-Shaybani's distinct approaches reflect the inconsis­tency of the expansion of warriors' resort to indiscriminate targeting when only non-Muslims are involved, an expansion aimed at widening legitimate force in armed conflict. Landau-Tasseron argues that the inclination to protect Muslim life and the willingness to dispose of non-Muslim combatants ema­nated from the distinct legal treatment of both, where the taking of Muslim life was muharrama (strictly prohibited), whereas taking non-Muslim life was manhΓanha (forbidden).267 But the jurisprudence of al-Shaybani and al-Shafi'i does not indicate such a conscious distinction between instilling harm upon Muslims and non-Muslims. Neither jurist perceives targeting in the presence of non-Muslims legal and in that of Muslims illegal, and with the exception of the demand for kaffara in the case of al-Shafi'i, neither jurist prescribes strict rules that change the course of action by the military. Thus, it is true that the bias is still evident in their treatment of the two groups, but it reflected itself in a subtler manner than the sharp legal/illegal distinction suggested by Landau- Tasseron.

266 al-Shafi'i, al-Umm, 4:349.

267 Ella Landau-Tasseron, “Non-Combatants in Muslim Legal Thought,” Center on Islam,

CindtheFuUrre of theMuslim WoridMonographs, Se eno.t, Paper ¾(,2.006)".

6-7, online:, accessed 9 March 2007.

2,¾.r,4 Desttof Property

Al-Shafi'i also expands the permissibility of destruction of property if it is like­ly to weaken the enemy's strength. In the statement below, al-Shafi'i encour­ages Muslims to destroy as much property as possible in an undefeated territory, again through the employment of the term uhibb (prefer) and with­out reliance on any textual sources to support the call for destruction. It is true that al-Shafi'i relies on Q 59:5, which is arguably interpreted to permit destruc- tion.268 But as argued in the previous section, nothing in the verse indicates that it is preferable to destroy property.

If Muslims attack a territory of war... and they cannot win them and con­vert the territory to a territory of Islam or a land of truce, I prefer that they cut, burn, and vandalize whatever they can from their fruits and trees and take their property, especially light items, as booty. Whatever they can't take, they should burn and flood.269

In the above case, al-Shafi'i seems more interested in the power of the Islamic state and its revenue than in his stated objective for jihad. He starts off his chapter on jizya with a highly indicative verse from the Qur'an, “I created jinn and mankind only to worship me,''270 declaring that the objective ofjihad is to get human beings to Ifflfil their raison d'etre and worship God. This is contrary to Bonner's claim that al-Shaβ'i's jihad regime focused on defensive warfare.271 But his interest in the relentless destruction of the enemy's might shows that al-Shafi'i does not always appear more interested in converting non-Muslims to Islam than in consolidating Muslim power. After all, if the objective is to at­tract inhabitants of a city to a new religion, then destroying property to guar­antee military advantage is unlikely to attract those inhabitants to the religion. This finding is corroborated by Griffels analysis of a!-Shafi'i's approach to apostasy. Griffel points out that al-Shafi'i is more interested in an alleged apos­tate's public pronouncement than in exploring his belief system. If a person declares himself to be a Muslim, then he may not be treated as an apostate.272 Arguably in both cases the interest is hegemony, where undeclared belief is

268 al-Shafi'i, al-Umm, 4:368.

269 al-Shafi'i, al-Umm, 4:367.

270 Q 51:56.

271 Bonner, Jihad, 107, 160.

272 Frank Griffel, “Toleration and Exclusion: al-Shafi'i and al-Ghazali on the Treatment of Apostates” Bulletinof,^heS(^hoolof OrieiaandAfiicanSudies 64, n0.¾f200ff,3A5,. irrelevant, and the simple act of submission through artificial espousal of the Islamic faith suffices.

2.3.2 Arab/Muslim Supremacy Within the Empire

After providing proof of the magi being People of the Book, which in turn means that the expansion of the right to pay the Jizya is granted only to them, excluding all non-Arab non-scriptuaries, al-Shafi'r argues that the magi belong to a special category of People of the Book.273 Generally, Muslims are allowed to marry and eat the food of the People of the Book. However, al-Shafi'i argues that the magi are allowed to payJizya as People of the Book, but their status is different from Christians and Jews. Accordingly, Muslims may not eat their food, and Muslim men may not marry their women.

Evidence in sources citing the Prophet (pbuh), and I do not know of any­one contesting that, indicates that he meant the people of the Torah and the Bible from the Israelites and not the magi.... And I do not know of anyone contesting that magi women may not be married and that their slaughtered animals may not be eaten. So consensus indicated that there are two judgements for the People of the Book, the women of some may be married and slaughtered animals eaten, and others whose women may not be married and slaughtered animals may not be eaten.274

In order to prove his thesis that not all the women of the People of the Book may be married and their slaughtered animals eaten, he argues that the per­mission only applies to Israelites and relies on 'Umar saying, “Christian Arabs are not People of the Book. We are not allowed to eat their slaughtered animals or marry their women. I am not leaving them until they convert to Islam or I smite their necks.''275 However, in his attempt to prove the special status of the magi, al-Shafi'i contradicts himself in more than one way. First, he earlier argued that the Prophet had taken Jizya from Arab Christians to prove that Jizya is not dependent on ethnicity. Relying on this tradition presumably can­cels out the 'Umar tradition cited above. Second, nothing in 'Umar's statement calls for distinguishing between those who may be married, and whose food may be eaten, and those who may only payJizya. In fact, 'Umar's statement on Arab Christians denies the permissibility of their food and their “right” to pay Jizya on the basis of them not being People of the Book. Thus, the statement

273 Badawi, “Sunni Islam,” 321.

274 al-Shaβ'i, al-Umm, 4:258-59.

275 al-Shaβ'i, al- Umm, 4:259. arguably contradicts al-Shafirs distinction with its assertion of the link be­tween payment of Jizya and those two practices, since 'Umar obviously links the two to each other, whereas al-Shafi'i argues that not all People of the Book are treated equally in terms of marriage and food sharing.

Al-Shafirs positioning of the magi as People of the Book might appear to be an expansive approach to the definition of the People of the Book through the inclusion of a new category among them. But practical implications prove that this definition leads to a limitation on the “enjoyment of the right' to pay the Jizya by non-Muslims, which prevents them from living in Muslim lands and from continuing to enjoy their property. What is clear is that al-Shafi'i was faced, on the one hand, with the indisputable fact that the Rightly Guided Ca­liphs allowed the magi to live in the Muslim state and enjoy the “privileges' enjoyed by the People of the Book and, on the one hand, with the Quranic reference to payment ofJizya by People of the Book. In his attempt to mitigate sources and to resolve the seeming conflict between the Quranic tradition and the prophetic hadlth on the treatment of the magi, he was unwilling to follow the approach taken by the Hanafis, whereby all non-Arabs were allowed to pay theJizya. Hence he devised a new category for People of the Book, ones who may pay the Jizya, but whose food may not be consumed and women not be married, despite the lack of textual basis for the existence of such a category. Alternatively, he could have argued that the Prophet's hadlth is specific in na­ture and that such a privilege is granted only to the magi. Although this argu­ment conforms more closely to his theory of usul al-fiqh, al-Shafi'i does not follow that line of argument and prefers to pursue the other option, position­ing the magi as People of the Book. It might be that he was guided in this choice by the fear of expansion of the category of non-Muslims granted the status of dhimmls through qiyas. This choice is particularly convenient, considering that he ends up creating a special category for the magi as People of the Book. This category is inferior to other categories of the People of the Book, since interac­tion with Muslims through marriage and food sharing is curtailed. Thus, the regime advocated by al-Shafi'i effectively leads to a scenario identical to an argument of specificity of the hadlth to the magi andJizya payment.

There is some indication that a!-Shafi'i's distinction between People of the Book reflects an interest in Arab/Muslim supremacy. As mentioned in the In­troduction, superiority debates existed between different ethnic groups. Al­though al-Shafi'i is said never to have engaged in similar debates, evidence from his biographies indicates a strong sense of pride in his Arab identity. This pride leads Abfi Zahra to believe that al-Shafi'i left Iraq because of increasing Persian influence, and Abfi Zayd to argue that in al-Risala, al-Shafi'i was inter­ested in asserting Arab supremacy. Although Persian domination is a highly contested issue, with some historians arguing Arabs always had the upper hand and others denying it, it is hard to contest that they enjoyed certain priv­ileges and were close to decision-making circles. Al-Rashid, for example, re­ferred to his Persian minister as his father (before turning against the Baramika).276

This increased power might have agitated al-Shahi, who seems to have shown suspicion of other ethnicities. For example, al-Razi documents an inci­dent where one of al-Shafir disciples bought him perhime, and when asked from whom he had bought it, he said a blonde man. Al-Shahi responded by ordering him to return it, because “he has never seen anything good from a blonde person.''277 In the context of the early stages of the Islamic state, it is fair to assume that most fair-coloured people did not come from Arab origins, and that the suspicion of blonde people is in fact revealing in terms of al- Shafrrs ethnic perceptions. He equally showed allegiance to his Arab descent. For example, when given money by al-Rashid, he is said to have distributed it among Qurayshi and Meccan men.278 Meccans were not the only ones fa­voured by al-Shafi'i, according to al-Bayhaqi, who recounts a prophetic hadlth recited by al-Shafi'i indicating the purity of Medina and its protection by an- gels.279 Perhaps the strongest indication of a!-Shafi'i's favouring Arabs is his treatment of the issue of enslaving non-Muslim Aiab men. Although he denies the claim that Arab non-Muslims may not be enslaved, he shows his prefer­ence for such an approach, saying, “If it were not a sin to wish for that, we would have wished for it.''280

But this statement portrays that al-Shafi'i revered Islam as a source of iden­tity and refused to let Arab identity trespass upon it. Accordingly, it would be shallow to assume that he allowed that bias to seep into his analysis to the extent of introducing the contradictions highlighted earlier. However, because of his strong reverence for Islam as an identity, he is likely to have been agi­tated by the increasing influence of the Barmakids and other Persian elites. Those elites were accused of heresy and of encouraging older Persian religions, to the extent that several poets wrote about their heresy and enthusiasm for non-belief.281 In fact, the important 'Abbasid administrator al-Fadl b. Sahl (d. 202/817-8), a late convert to Islam after the fall of the Barmakids, played

276 Hugh Kennedy, The Early Abbasid Caliphate (London: Croom Helm, 1981), 116.

277 Ibn Abi Hatim al-Razi, Adab al-Shafi’l, 130--1.

278 al-Bayhaqi, Manaqib, 116.

279 al-Bayhaqi, Manaqib, 121.

280 al-Shaβ'i, al-Umm, 4:387.

281 Abu Khalil, Amlr al-khulafa^,, 146.

a crucial role in the civil war between his patron al-Ma'mfin (d. 218/833), and his brother and predecessor al-Amin (d. 198/813).282 The suspicion of Persian subjects may have led al-Shafi'i to limit the inclusion of non-Muslims in the Jizya scheme and to adopt a contradictory approach to the magi/Persians, an approach that acknowledged their established status as Jizya payers but that aimed to limit their further integration into society.

In a further attempt to limit the expansion of the “right” to pay Jizya, al- Shafi'i argues that People of the Book are only those who, or whose parents, adopted their religion before the transmission of the Qur'an to the Prophet, and that anyone who newly adopts one of the “book” religions later than that date is treated like a pagan. If a Christian or a Jewish person converts to one of the other two religions, s/he is denied dhimml status. Seemingly, al-Shafi'i was unable to find a textual source to back up this argument.

The source we rely on [to prove that conversion of one of the people of the Book to another religion of the book is unacceptable] is that Jizya is not accepted from anyone who adopts a book religion except if his par­ents or he himself adopted that religion before the transmission of Qur'an and is accepted from anyone who adheres to his religion and the religion of his parents before the transmission of the Qur'an.283

This statement gives the reader the impression that a solid rule for accepting Jizya only from those who adopt one of the religions of the People of the Book before the advent of Islam was earlier established by al-Shafi'i. Nevertheless, the matter is very briefly dealt with in the section focusing on eligibility for Ji- zya payment, and no proof is given for his statement that only those who had adopted a “book” religion before Islam were entitled to payJizya. He simply treats his hypothesis that time of the adoption of the religion affects an indi­vidual's dhimml status as an uncontested given and proceeds to address how one can establish time of adoption of a “book” religion:

As for anyone, Arab or non-Arab, who encounters the advent of Islam before adhering to [one of] the religions of the People of the Book and wishes to payJizya and be affirmed in his religion or to convert to [one of the] religions of the People of the Book, the imam may not acceptJizya

282 Kennedy, Early Abbasid Caliphate, 117.

283 al-Shafi'i, al-Umm, 4:260.

from him and should fight him until he converts to Islam, just as pagans are fought until they convert to Islam.284

It should be noted that this limitation again meets the objective of asserting Muslim hegemony during armed conflict, as well as within the boundaries of the state. In armed conflict, it expands the power of the army by limiting pro­tection to those who converted before the advent of Islam, leading to conse­quences similar to those elaborated upon when addressing his definition of the People of the Book. But at the same time, it ensures stability and limited freedom of practice within the state itself. If conversion to one religion is trans­lated in triumph of the said religion, then any attempt to convert from one scripted religion to the other is somehow an affirmation of a relative triumph of the religion converted to. If one were interested in limiting any legitimation of other religions and reluctantly accepted their habitation in Muslim lands, then any conversion would likely be faced with scrutiny.

2.3.3 Limiting the Caliph's Authority

233.1 FreeingFemaleandMinorCaptives

In the previous section, al-Shaybanr was shown to deviate from the interest in asserting caliphal authority in favour of expanding the potential military might of the army. Interestingly, al-Shafi'i addresses the same issue dealt with by al- Shaybanr, the issue of grace and ransom of captives, to reach the opposite con­clusion that male captives may be freed or ransomed, whereas women and children may not. In the case of victory, if men do not convert, they may be killed, graced upon, or ransomed in return for money or Muslim captives, which effectively gives the imam complete discretion in dealing with male captives.285 However, women and children may only be taken as booty and divided as slaves according to booty rules.286 They may not be released even if the captured people demand that they pay Jizya for their women and chil- dren.287 In order to justify this argument, al-Shafi'i relies on incidents in which the Prophet captured women and children and divided them like property, but he does not offer any source that prohibits freeing those two categories.288 Again, the Quranic verse addressing ransom is sidelined in the reasoning. Al­though the verse does not specify gender and age as conditions for ransom or

284 al-Shafi'i, al-Umm, 4:247.

285 al-Shaβ'i, al-Umm, 4:335.

286 al-Shaβ'i, al-Umm, 4:335.

287 al-Shaβ'i, al-Umm, 4:250.

288 al-Shaβ'i, al-Umm, 4:335. grace, al-Shafi'i does not offer any insight into his reasoning on the verse. One can speculate that he believes the verse was specified by the Prophet to adult men. Yet, as mentioned above, prophetic traditions relied on by al-Shafi'i do not indicate a reluctance to free women and children.

Moreover, al-Shafi'i, who treats women and children as property in this case and thereby denies them the opportunity for freedom from captivity and slavery,289 treats women as independent individuals enjoying full capacity in other instances. In fact, he argues that if male non-Muslims convert to Islam before capture, their souls and properties are untouchable. Nevertheless, their women's legal position depends not on their men but on their own legal status, and they may thus be either enslaved or killed.290 It is noticeable that the con­tradictory reasoning regarding the treatment of women leads in both cases to their enslavement, which leads the reader to question al-Shafir approach and, perhaps, to assume that his main objective is to maximize revenue through enslaving women and children.

'2,.¾.¾.'2, TippingthePowerBcLmce InFcorof Rebels

Just like al-Shaybani, al-Shafi'i was interested in formulating a legal regime that limited the privileges of the political authority vis-a-vis rebels. Again, the com­peting interests of promoting stability and limiting the caliph's power in deal­ing with rebellion are weighed against each other, through asserting the necessity for taWll and significant numbers, on the one hand, and denying rebel liability for damages caused to life and property and instructing that flee­ing and wounded rebels may not be killed, on the other hand.291 Interestingly, unlike his approach to collateral damage to Muslim life in conflicts with non­Muslims, al-Shafi'i does not demand kaffara from rebels, in case of similar damages to Muslim life, hinting perhaps at an interest in suppressing potential routes for holding rebels legally accountable for rebellion.

Al-Shafirs approach to rebellion is indeed favourable to rebels in a context where rebels were often crucified and muti!ated.292 He argues that rebel weap­ons may not be used against them293 and that indiscriminate weapons such as mangonels may only be resorted to in reciprocity or to avert serious damage to the imam's people.294 But what is of prime importance to us here is his ap-

289 al-Shafi'i, al-Umm, 4:335.

290 al-Shafi'i, al-Umm, 4:343-44.

291 al-Shafi'i, al- Umm, 4:303--.

292 Abou El Fadl, Rebellion, 157.

293 al-Shafi'i, al-Umm, 4:318.

294 al-Shafi'i, al-Umm, 4:311. proach to the issue of pursuing fleeing rebels. He argues that the existence of a group that they may reunite with does not affect the treatment of rebels and that captives are still not to be killed, the wounded not to be dispatched, and fugitives not to be pursued.295 In doing so, al-Shafi'i relies more consistently on Als conduct with rebels than his Hanafi contemporaries. However, his ap­proach to a Quranic verse that he holds primarily relevant to rebellion is less consistent. The verse states that “if two parties among the believers fight each other, then make peace between them. But if one of them transgresses (baghat) against the other, then fight, all of you, against the one that transgresses until it complies with the command of God.''296 The Quranic verse uses the term taff ila amr allah, translated above as, “complies to God's command.” However, the term tafi carries several meanings and can also suggest physical retreat. Thus, arguably, the term could support al-Shafi'i's contention that retreating rebels may not be fought. He argues that the verse refers to retreat and does not ad­dress regrouping, thereby prohibiting pursuing fugitives even if they are flee­ing to reunite with their army.297 Nevertheless, al-Shafi'i does not address the fact that the verse conditioned retreat upon submission to God's command, possibly supporting the Hanafi argument that physical retreat does not suffice on its own to require the pursuit of rebels to end, since a retreat for regrouping and remobilization purposes is not a submission to God's command.

According to Abou El Fadl, al-Shafi'i seems interested in restricting the bru­tality with which rebels were treated by Islamic dynasties through setting up a pragmatic legal system that holds the ruler accountable regardless of rebel conduct and regardless of the justice of the ruler himself.298 More important, unlike al-Shaybani, al-Shafi'i elaborates on the qualifying nature of ta’wil, argu­ing that an erroneous ta’wil does not mean that legal regime of baghy (rebel­lion) does not apply to rebels.299 Thus, al-Shafi'i's proposed legal system aimed at shaming caliphs into a more restrictive conduct of armed conflict with reb­els, or as Abou El Fadl puts it, “It is as if al-Shafi'i is arguing to the 'Abbasids of his age, ‘If you claim to be the rightful and just rulers, then this is how just rul­ers treat rebels.'”3٥٥

295 al-Shaβ'i, al-Umm, 4:311.

296 Q 49:9. Translation provided in Abou El Fadl, Rebellion, 37.

297 al-Shaβ'i, al-Umm, 4:311.

298 Abou El Fadl, Rebellion, 156.

299 Abou El Fadl, Rebellion, 156.

300 Abou El Fadl, Rebellion, 149.

3