Previous · All chapters · Next

B8 · باب بيان الحجة الشرعية وأحكامها · The binding proofs: Kitāb, mutawātir Sunna, mashhūr, and ijmāʿ (its cause, pillar, qualified members, conditions and grades) (v1/277–321)

1. B8.01 · الحجة والبينة والبرهان والآية والدليل والشاهد · The names of proof: some bind with certainty, some bind to act, and dalīl is the proof that reveals (v1/277–279)
2. B8.02 · أصول الحجج · Three roots (Kitāb, Sunna, ijmāʿ), with qiyās drawn from them; all certain proofs go back to hearing from the Prophet (v1/279)
3. B8.03 · الكتاب · The Qurʾān is known only by mass transmission, not by its inimitability; ⚑ minimum recitation; the basmala is a revealed verse between sūras; the Persian-recitation question (v1/279–282)
4. B8.04 · المتواتر · Mass-transmitted reports yield certain knowledge, and that knowledge is immediate (ḍarūrī), not reasoned (v1/282–291)
5. B8.05 · المشهور · ⚑ A report single at its root, that scholars received with acceptance and practice: al-Rāzī counts it as tawātur; ʿĪsā b. Abān, "which is sound in our view", as mashhūr giving reassurance; that reception grounds its adding to the Qurʾān (v1/291–295)
6. B8.06 · حجية الإجماع · The community's consensus yields certainty as an honour given to it for the religion, not because many cannot err; but the certainty of its own form and transmission is a separate question (→ B8.07, B8.11) (v1/295–300)
7. B8.07 · سبب الإجماع · Consensus may rest on revelation, on inference from a text, or on opinion; against Ibn Jarīr, a single report or qiyās can ground it; consensus itself may be reported by single reports (v1/300–303)
8. B8.08 · ركن الإجماع · ⚑ The pillar is agreement in word or deed, or its concession: one voice made known and the rest silent, after the question reached them and time to reflect passed (v1/303–310)
9. B8.09 · أهلية الإجماع · Mujtahid scholars clear of heresy or open sin count, but mujtahid status binds only in rulings that need specialist knowledge; ⚑ Sarakhsī: concealed hawā counts outside the issue he errs in, concealed fisq counts, open display of either does not; the number need not reach tawātur; not the Companions, Medina or the Prophet's family alone (v1/310–315)
10. B8.10 · شرط الإجماع · The end of the generation is not a condition; ⚑ a lone dissenter blocks consensus only if the others accepted his ijtihād (v1/315–317)
11. B8.11 · حكم الإجماع · Consensus has grades: the Companions' explicit consensus makes its denier an unbeliever; lower grades do not; ⚑ a later consensus after an earlier dispute (v1/318–321)

B8.01 · الحجة والبينة والبرهان والآية والدليل والشاهد · The names of proof: some bind with certainty, some bind to act, and dalīl is the proof that reveals (v1/277–279)

• Terms: ḥujja, from "ḥajja" (he overcame), or from returning to it (as to the House), since one must return to it in practice. Bayyina is from bayān: what makes the binding force plain to the heart, whether it yields certain knowledge or less. Burhān is used like ḥujja. Āya is a sign; used absolutely in the law it means what yields certainty (the prophets' miracles). Dalīl is "one who guides"; in the law, a proof that brings out what was hidden. Shāhid is like dalīl, certain or not (witnesses in court). (v1/277–279)
• Objection → Reply: some denied the prophets after seeing miracles. Reply: miracles yield knowledge through fair reflection, not by compulsion, or reward and punishment would lose their point; deniers turned away from reflecting (Q 41:5, 41:26) or denied in obstinacy (Q 27:14). (v1/278)

Back to the list

B8.02 · أصول الحجج · Three roots (Kitāb, Sunna, ijmāʿ), with qiyās drawn from them; all certain proofs go back to hearing from the Prophet (v1/279)

• Position: «ثمَّ اعْلَم بِأَن الْأُصُول فِي الْحجَج الشَّرْعِيَّة ثَلَاثَة الْكتاب وَالسّنة وَالْإِجْمَاع وَالْأَصْل الرَّابِع وَهُوَ الْقيَاس هُوَ الْمَعْنى المستنبط من هَذِه الْأُصُول الثَّلَاثَة» — "Know that the roots of the legal proofs are three: the Kitāb, the Sunna and ijmāʿ; the fourth root, qiyās, is the meaning drawn out of these three." (v1/279)
• They are of two kinds: those that yield certain knowledge, and those that only "make permissible" (mujawwiz): acting on them is required though they do not yield certainty (Q 17:36 forbids following what one does not know). (v1/279)
• The certain ones are four: the Kitāb, the Sunna heard from the Prophet, the Sunna transmitted from him by tawātur, and ijmāʿ. Their root is hearing from the Prophet, who is protected from falsehood; ijmāʿ binds because he said God would not unite his community on error. (v1/279)

Back to the list

B8.03 · الكتاب · The Qurʾān is known only by mass transmission, not by its inimitability; ⚑ minimum recitation; the basmala is a revealed verse between sūras; the Persian-recitation question (v1/279–282)

• Definition: the Qurʾān sent down to the Prophet, «الْمَكْتُوب فِي دفات الْمَصَاحِف الْمَنْقُول إِلَيْنَا على الأحرف السَّبْعَة الْمَشْهُورَة نقلا متواترا» — "written between the covers of the muṣḥafs, transmitted to us in the seven well-known modes, by mass transmission". (v1/279) «وَبَاب الْقُرْآن بَاب يَقِين وإحاطة فَلَا يثبت بِدُونِ النَّقْل الْمُتَوَاتر كَونه قُرْآنًا» (v1/280) — "The Qurʾān is a matter of certainty and full grasp, so nothing is established as Qurʾān without mass transmission." So prayer with words that only Ibn Masʿūd read is invalid: they are then like reciting a report. (v1/279–280)
• Objection → Reply: "inimitability proves it is Qurʾān." Reply: less than a verse, or a short verse, is not inimitable, yet it is Qurʾān. And a miracle proves the Prophet truthful, not that the words are God's speech (compare ʿĪsā reviving the dead). (v1/280)
  • ⚑ So Abū Yūsuf and Muḥammad require three short verses or one long one in prayer (the shortest sūra, al-Kawthar, has three). Abū Ḥanīfa: "what is easy of the Qurʾān" is met by a short verse, though limiting oneself to it is disliked. (v1/280)
• The Companions wrote in the muṣḥafs only what they agreed on, stripped of additions (they disliked the ten-verse marks), to secure mass transmission. (v1/280)
• Objection → Reply: the basmala is mass-transmitted in the muṣḥafs, yet you do not make it part of al-Fātiḥa or of each sūra. Reply: Abū Bakr al-Rāzī reported «أَن الصَّحِيح من الْمَذْهَب عندنَا أَن التَّسْمِيَة آيَة منزلَة من الْقُرْآن لَا من أول السُّورَة وَلَا من آخرهَا» (v1/280) — "that the sound view in our school is that the basmala is a revealed verse of the Qurʾān, belonging neither to the start of a sūra nor to its end". It is written in a separate line, to separate sūras. (v1/280–281)
  • Hence it is recited quietly (as part of the recitation, not of al-Fātiḥa), it is disliked for the junub to recite it as Qurʾān, but on Abū Ḥanīfa's view it does not fulfil the obligation of recitation, since the dispute over whether it is a complete verse creates a doubt. (v1/281)
  • Ibn Masʿūd's "three consecutive days" was acted on as a report he heard from the Prophet, not as Qurʾān. (v1/281)
• Who: ⚑ (Persian recitation) many of our shaykhs say the Qurʾān's inimitability lies in wording and meaning together, especially for Abū Yūsuf and Muḥammad, who hold that Persian recitation does not fulfil the obligation. Sarakhsī: they did not mean the meaning alone is not inimitable: God's speech is uncreated while all languages are created, and the Prophet was sent to all people, so the challenge must reach a non-Arab in his own language. So Abū Ḥanīfa allowed Persian recitation in prayer; the Ṣāḥibān allow it for one who cannot recite Arabic, and for one who can, they require following the Prophet and the salaf. Abū Ḥanīfa counts that following only as making Persian recitation disliked. (v1/281–282)

Back to the list

B8.04 · المتواتر · Mass-transmitted reports yield certain knowledge, and that knowledge is immediate (ḍarūrī), not reasoned (v1/282–291)

• Definition: what reaches us from the Prophet through a group so large and so widely spread that their agreeing on a lie cannot be imagined, and so on at each stage, "its first like its last, its middle like its ends": the number of rakʿas and prayers, zakat and blood-money amounts. People's aims differ, so only a shared hearing could unite them. «وَلِهَذَا كَانَ مُوجبا علم الْيَقِين عِنْد جُمْهُور الْفُقَهَاء» (v1/283) — "Hence it yields certain knowledge in the view of most jurists." (v1/282–283)
• Who: (1) Some, a party of those who deny the prophets' messages, say reports never yield knowledge, since reporters could invent together. Sarakhsī: this person is a fool, like the sophists who deny what they see; one does not argue with him, one only shows he denies what he knows (his own birth, his parents, that Mecca exists). (v1/283–284) The knowledge comes from God's making, not from the reporters' act: people of differing natures agree only when something unites them; this keeps the law alive after prophecy ended. (v1/283–284)
• (2) Some say tawātur yields only reassurance (ṭumaʾnīna), since many single reports still each carry doubt, and Christians and Jews mass-transmitted the killing of ʿĪsā, and Zoroastrians Zoroaster's miracles. (v1/284) Reply: that view ruins its holder's faith (he cannot truly know the prophets) and denies what every sane person knows (the way to Mecca, Bukhārā to the west, Kāshghar to the east). (v1/285)
  • The crucifixion report goes back to four in a house with ʿĪsā, and to seven who entered, guided by a man paid thirty dirhams, so it was never tawātur; the crowds saw only from afar. Or the mass report was true of the man they killed, but he was made to resemble ʿĪsā ("it was made to appear so to them", Q 4:157). (v1/285–286) Objection: that makes all sense-knowledge unreliable. Reply: casting a likeness was a wise defence of ʿĪsā against those God knew would not believe; it is not imaginable toward those who come to prophets to believe (Iblīs appeared as a shaykh of Najd and as Surāqa; Jibrīl as Diḥya). (v1/286–287)
  • Zoroaster's wonders were tricks done before the king's inner circle, who colluded to secure the king's rule; collusion in a small circle stays hidden. (v1/287–288)
  • The Companions were of many tribes, about thirty thousand, with hypocrites and spies among them; collusion among them would have leaked, as a Qurʾān rival would have. (v1/288–289)
  • "It is contrary to custom, so we grant reassurance but not certainty." Reply: reassurance counts as knowledge unless something higher lies beyond it (as waking lies beyond a dream); nothing lies beyond tawātur, which equals seeing. (v1/289–290)
• Objection → Reply: a group is only individuals added together, no stronger than any one of them. Reply: this is false. One of us can pronounce every letter of the alphabet, yet no one concludes from that he could produce something like the Qurʾān, though the Qurʾān is made of those same letters; a dull man can recite every wise line of Imruʾ al-Qays's poetry without being able to compose its like; a man may hit the truth once by conjecture without being right whenever he speaks that way. A group's agreement on truth comes from what unites them, the claim of religion and honour on truthfulness, not from adding up fallible individuals. (v1/290)
• Objection → Reply: a qāḍī who hears by tawātur that Zayd's property belongs to ʿAmr does not rule without evidence. Reply: two answers are possible. He may rule, since he is commanded by the law to rule on knowledge; or he may not, as where he learned the cause by direct observation before his appointment (in matters that stand despite doubts); and in God's ḥudūd, which doubts avert, even knowledge gained after appointment does not let him rule until witnesses testify. On the second answer, judging carries conditions beyond knowledge (he cannot rule on a witness who says "I inform" instead of "I testify"). (v1/290–291)
• Who: «ثمَّ الْمَذْهَب عِنْد عُلَمَائِنَا أَن الثَّابِت بالمتواتر من الْأَخْبَار علم ضَرُورِيّ كَالثَّابِتِ بالمعاينة» — "The doctrine of our scholars is that what mass-transmitted reports establish is immediate knowledge, like what is seen." The followers of al-Shāfiʿī say it is certain but reasoned, since people differ over it. Reply: a child knows his parents by report; one cannot avoid this knowledge by not reasoning; differences come from defective minds, as with the senses. (v1/291)

Back to the list

B8.05 · المشهور · ⚑ A report single at its root, that scholars received with acceptance and practice: al-Rāzī counts it as tawātur; ʿĪsā b. Abān, "which is sound in our view", as mashhūr giving reassurance; that reception grounds its adding to the Qurʾān (v1/291–295)

• Who: ⚑ the report that is "single at the root, mass-transmitted at the branch" — one the scholars received with acceptance and practice (khuffs, the ban on mutʿa, marrying a woman with her aunt, the six ribā items). (v1/291–292)
  • Abū Bakr al-Rāzī: one of the two kinds of tawātur; it yields certainty, but reasoned; so it can abrogate, and adding to the text is abrogation. (v1/292)
  • ʿĪsā b. Abān: only what yields immediate knowledge is mutawātir, and this kind «فَأَما النَّوْع الثَّانِي فَهُوَ مَشْهُور وَلَيْسَ بمتواتر وَهُوَ الصَّحِيح عندنَا» (v1/292) — ʿĪsā: "as for the second kind, it is mashhūr and not mutawātir"; Sarakhsī adds: "and this is the sound view in our school". Its denier is not declared an unbeliever, and on reflection a doubt remains at its root, so it yields reassurance. (v1/292–293)
• Position: «وَلَكِن مَعَ هَذَا تجوز الزِّيَادَة على النَّص بِهَذَا النَّوْع من الْأَخْبَار» — "But even so, adding to the text is allowed by this kind of report", because the scholars' acceptance of it is a binding proof (the consensus of the second and third generations). (v1/293) The crucifixion report is also single-root, and a stronger report (God's) overrides it. (v1/293)
• ʿĪsā's three grades of mashhūr (Sarakhsī: sound, since it rests on the scholars' acceptance): (1) its denier is misguided but not an unbeliever: stoning (only the Khawārij dissented); (2) its denier is wrong and feared for in sin: khuffs (ʿĀʾisha's and Ibn ʿAbbās's early doubts, then their return) and ribā al-faḍl (Ibn ʿAbbās's "no ribā except in delay", then his return); (3) its denier is wrong but not in sin: reports the jurists dispute in every generation. (v1/293–294)
• The rare, disowned report (gharīb mustankar): «وَأما الْغَرِيب المستنكر فَإِنَّهُ يخْشَى المأثم على الْعَامِل بِهِ وَذَلِكَ نَحْو خبر الْقَتْل فِي الْقسَامَة وَخبر الْقَضَاء بِالشَّاهِدِ وَالْيَمِين لِأَنَّهُ مُخَالف لظَاهِر الْقُرْآن» — "As for the rare, disowned report, one fears sin for whoever acts on it, such as the report of execution in qasāma and the report of judging by one witness and an oath, because it contradicts the apparent sense of the Qurʾān." Scholars of the first and second generations left it, so it nears falsehood, just as the mashhūr nears truth. (v1/294)
• No fixed number for tawātur (ʿĪsā; sound): a small number does not suffice, a great number does; the line is clear at its ends and unclear in the middle, like miles walked or pounds eaten. (v1/294–295)
• [Editorial] Debate note: the gharīb mustankar paragraph names two reports (execution in qasāma; one witness plus an oath). Before using this line, check who acts on them and on what grounds. The filter itself ("contradicting the apparent Qurʾān") is developed in B9d.

Back to the list

B8.06 · حجية الإجماع · The community's consensus yields certainty as an honour given to it for the religion, not because many cannot err; but the certainty of its own form and transmission is a separate question (→ B8.07, B8.11) (v1/295–300)

• Position: «اعْلَم أَن إِجْمَاع هَذِه الْأمة مُوجب للْعلم قطعا كَرَامَة لَهُم على الدّين» — "Know that the consensus of this community yields certain knowledge, as an honour to them for the religion", not because a large number cannot err: Jews, Christians and Zoroastrians outnumber us and agreed on error. This is the view of the jurists and most theologians. (v1/295)
• Who: al-Naẓẓām and some Imāmīs: consensus is only individuals added up, each fallible. Reply: this contradicts itself; joined together, people carry a beam none can carry alone, and morsels satisfy. «وَمن أنكر كَون الْإِجْمَاع حجَّة مُوجبَة للْعلم فقد أبطل أصل الدّين» (v1/296) — "Whoever denies that consensus is a proof yielding knowledge has destroyed the root of the religion." (v1/295–296)
• Proofs from the Kitāb: "the best community… enjoining the right" (Q 3:110): the absolutely right is what they agree on; a single mujtahid's view is right only for him. "Whoever opposes the Messenger… and follows other than the way of the believers" (Q 4:115). "No intimate besides God, His Messenger and the believers" (Q 9:16). "Their religion which He has approved for them" (Q 24:55). "A middle community, that you be witnesses over the people" (Q 2:143): middle means just and approved, and their witness is set beside the Messenger's. (v1/296–299)
  • Objection → Reply: "the plural attributed to a group covers each one, yet no single mujtahid is certain." Reply: «قُلْنَا لَا بل المُرَاد هُنَا أَن جَمِيع الْأمة أَو أَكْثَرهم بِهَذِهِ الصّفة» — "No: what is meant here is that the whole community, or most of it, has this quality", as in "when you said, O Moses…" (Q 2:55) and "when you killed a soul" (Q 2:72), said of some of them, or "the Banū Hāshim are wise". So when most of this community agree, that is the right in the absolute sense, and when they differ, the right does not lie outside their views. (v1/296) → B8.10
  • The Messenger's witness binds with certainty, so theirs does; this is about this world, since the next-world witness is named in other verses (Q 4:41; 16:84). The People of the Book were also called witnesses, but for the covenant to make plain the Prophet's description, which they broke. Single reports and qiyās bind without certainty out of necessity; here the witness is absolute. (v1/297–298) Objection from "I created jinn and men only to worship Me" (Q 51:56), which does not stop them uniting in not worshipping. Reply: the lām of "that you be witnesses" marks an honour given for that purpose; in Q 51:56 the apparent sense is set aside for a proof (worship must be chosen to earn reward), which does not affect the apparent sense elsewhere. (v1/298–299)
• Proofs from the Sunna, widespread and mass-transmitted in sense: "whoever wants the heart of Paradise, let him hold to the jamāʿa"; "God's hand is with the jamāʿa"; "My community will not unite on error"; "what the Muslims see as good is good with God". (v1/299)
• Reason: the Prophet is the last, his law lasts to the Hour and revelation has ended, so the law survives only if the community is protected from agreeing on error. Like a qāḍī's ruling by ijtihād, which becomes binding and cannot be undone. (v1/300)
• Objection → Reply: "the Hour will come only upon the worst of people". Reply: «فِي صِحَة هَذَا الحَدِيث نظر» — "the soundness of this ḥadīth is open to question"; it seems to contradict Q 2:257. If sound, it means the wicked will prevail while the righteous remain, and the second report ("until 'Allāh' is not said on earth") refers to the time between the two trumpet blasts. (v1/300)
• [Editorial] Debate note: Sarakhsī questions that report's soundness on the ground of its apparent clash with the Qurʾān, not on its chain. Anyone citing this passage should first check where the report is collected and with what chain.

Back to the list

B8.07 · سبب الإجماع · Consensus may rest on revelation, on inference from a text, or on opinion; against Ibn Jarīr, a single report or qiyās can ground it; consensus itself may be reported by single reports (v1/300–303)

• Causes: revelation (the mothers and daughters; blood-money for hands; not reselling food before taking possession); inference from a text (ʿUmar's kharāj on Sawād from Q 59:10; Abū Bakr's caliphate from the prayer); opinion (the impotent man's term; the wine punishment). (v1/300–301)
  • The wine punishment: the whipping itself came from the Sunna (palm stalks and sandals, about forty men); the number (forty men, each striking with his sandals: eighty) was fixed from the text; only the move from sandals to lashes was by opinion, supported by the qadhf punishment ("when he drinks he raves… eighty lashes"). ʿAlī: "no one I punish who then dies troubles me, except the wine punishment, which we set by our opinion." (v1/301)
• Who: Ibn Jarīr: consensus that yields certainty cannot come from a single report or qiyās. Sarakhsī: «وَهَذَا غلط بَين فقد بَينا أَن إِجْمَاع هَذِه الْأمة حجَّة شرعا بِاعْتِبَار عينه لَا بِاعْتِبَار دَلِيله» — "This is a clear error: we have shown that the community's consensus is a legal proof in itself, not by virtue of its basis." Demanding a certain basis makes consensus pointless. The early community agreed on using qiyās; the dissenters were theologians without fiqh and later people "whose dissent is not counted and whose agreement brings no comfort". (v1/302)
• Transmission: consensus reaches us by tawātur, by fame or by single reports (ʿAbīda al-Salmānī: four before ẓuhr, dawn prayer in the light, the sister's ʿidda; Ibn Masʿūd: four takbīrs at funerals). Those who deny transmission of consensus by single reports are wrong, since the Prophet's own word is transmitted that way, then binding for action only. (v1/302–303)

Back to the list

B8.08 · ركن الإجماع · ⚑ The pillar is agreement in word or deed, or its concession: one voice made known and the rest silent, after the question reached them and time to reflect passed (v1/303–310)

• Position: «ركن الْإِجْمَاع نَوْعَانِ الْعَزِيمَة والرخصة» — "The pillar of consensus is of two kinds: the strict and the concession." The strict: all agree in word or deed, all the people in matters everyone needs (zinā, ribā), all scholars of the age in matters of specialists (aunt and niece, zakat details). The concession: a view spreads from some scholars of the age and the rest stay silent, neither voicing dissent nor refuting it, after the fatwā was put to them or became known, «فالإجماع يثبت بِهِ عندنَا» — "and consensus is established by it, in our view." (v1/303)
• Who: (1) some scholars: silence does not make consensus. (2) Reported from al-Shāfiʿī: yes if most spoke and a few were silent, not if one or two spoke. Their proofs: silence may be awe or caution. ʿAlī stayed silent when ʿUmar consulted on surplus funds and on the miscarriage of the frightened woman, then spoke only when asked. Ibn ʿAbbās on ʿawl: "he was a man of awe, so I held back". (v1/303–304) (3) ⚑ ʿĪsā b. Abān: Dhū al-Yadayn (the Prophet asked Abū Bakr and ʿUmar rather than relying on their silence). (4) ⚑ al-Karkhī: in matters of ijtihād one mujtahid may not censure another, so silence is no agreement. (v1/304–305)
• Proof: requiring each to speak would mean consensus never forms, and binding a thing to an impossible condition is denying it. A mujtahid may not stay silent if he holds otherwise, so his silence is read in the lawful way. The majority-speaks view is weaker, since silence is more telling when only one or two spoke. (v1/305–306)
• Replies: in the funds and miscarriage cases both options were acceptable and ʿAlī gave the better one when asked; silence counts as agreement only if the ruling had then been settled by their view, or the asker had shown hesitation in answering, with the would-be dissenter still silent after that; neither is reported to have happened here. The Ibn ʿAbbās report is said to be unsound, since ʿUmar sought his counsel; if sound, it is a young man's awe of a senior. Dhū al-Yadayn: speech in prayer was then allowed and the waiting period had not passed. al-Karkhī's point is met by our condition: «فَهُوَ خَارج على هَذَا الْحَرْف» (v1/308) — "it comes out on the same principle", since silence counts only as leaving unsaid a contrary view one is bound to state. (v1/306–308)
• Conditions: «وَلِهَذَا اعْتبرنَا فِي ثُبُوت الْإِجْمَاع بِهَذَا الطَّرِيق أَن يسكت بعد عرض الْفَتْوَى عَلَيْهِ» — "Hence for consensus by this route we require that he stays silent after the fatwā is put to him", and after a period for reflection; after that, no one may retract by new opinion. Wide spread of the fatwā counts as putting it to them. (v1/308) This is also how the Qurʾān's inimitability is known: the Arabs never rivalled it. (v1/308–309)
• Objection → Reply: muzāraʿa spread after Abū Ḥanīfa while his companions stayed silent. Reply: their dissent was argued openly; they only stopped censuring those who practise it, as with a qāḍī's ruling on a disputed question. Custom shows people voice objections to what they reject (the preacher shot on the pulpit). (v1/309–310)
• Limited positions: «فَإِن الْمَذْهَب عندنَا أَن هَذَا يكون دَلِيل الْإِجْمَاع مِنْهُم على أَنه لَا قَول فِي هَذِه الْحَادِثَة سوى هَذِه الْأَقَاوِيل» — "our doctrine is that this [differing on a set of positions] shows consensus among them that there is no view on the case besides those positions", so no one may invent a new one. ⚑ Some of our shaykhs restrict this to the Companions; the reason covers every age. A sermon of one of the caliphs from among the Companions, left unopposed, is consensus by this route. (v1/310)
• Who: some, "not worth heeding", say certain consensus exists only in things like the Kaʿba's location. Reply: how did you learn that consensus, if not by some speaking and the rest silent? (v1/310)

Back to the list

B8.09 · أهلية الإجماع · Mujtahid scholars clear of heresy or open sin count, but mujtahid status binds only in rulings that need specialist knowledge; ⚑ Sarakhsī: concealed hawā counts outside the issue he errs in, concealed fisq counts, open display of either does not; the number need not reach tawātur; not the Companions, Medina or the Prophet's family alone (v1/310–315)

• Who: some say consensus needs all the sects, true and misguided. (v1/310)
• Position: «فَأَما الْمَذْهَب عندنَا أَن الْحجَّة اتِّفَاق كل عَالم مُجْتَهد مِمَّن هُوَ غير مَنْسُوب إِلَى هوى وَلَا معلن بفسق فِي كل عصر» — "Our doctrine is that the proof is the agreement of every scholar capable of ijtihād who is not associated with heresy nor openly sinful, in every age", since the qualities in the verses (middle, witnesses, enjoining the right, honoured) require it. (v1/311)
• Heresy (hawā): one whose heresy is unbelief is not of the umma; one who calls to his misguided heresy is suspect (so the Rāfiḍa on Abū Bakr and the Khawārij on ʿAlī are not counted). ⚑ For a known but non-proselytising heretic, some of our shaykhs: not counted where he is misguided, counted elsewhere. Sarakhsī: «وَالأَصَح عِنْدِي أَنه إِن كَانَ مُتَّهمًا بالهوى وَلكنه غير مظهر لَهُ فَالْجَوَاب هَكَذَا» — "the sounder view, to me: if he is suspected of heresy but does not display it, the answer is so" (i.e. as those shaykhs said); if he displays it, he is not counted at all. The Khawārij, who held most Companions to be unbelievers, cannot be trusted in the law they transmitted. (v1/311)
• Sin (fisq): ⚑ the Iraqis: the sinner of sound belief is not counted. Sarakhsī: «وَالأَصَح عِنْدِي أَنه إِذا كَانَ مُعْلنا لفسقه فَكَذَلِك الْجَواب» — "the sounder view, to me: if he is open about his sin, the answer is the same"; if not open, he is counted though his testimony is rejected, since he remains worthy of the honour of the religion. (v1/312)
• Expertise: ijtihād matters in rulings known only to scholars. A theologian without uṣūl al-fiqh is not counted (from al-Karkhī), «وَكَذَلِكَ من يكون مُحدثا لَا بصر لَهُ فِي وُجُوه الرَّأْي وطرق المقاييس الشَّرْعِيَّة لَا يعْتد بقوله فِي الْإِجْمَاع» — "and likewise a ḥadīth scholar with no insight into the ways of juristic reasoning and legal analogy is not counted in consensus", being like a layman on these rulings. (v1/312)
• Number: some require the tawātur number. «وَالأَصَح عندنَا» — the sounder in our view: a group suffices, since consensus raises no suspicion of lying, only of error, from which a group is protected by honour. (v1/312) Objection: a group might later turn sinful or apostate. Two answers from our shaykhs: God protects them after their consensus binds, or He raises others in their place. (v1/313)
• Companions only: «وَهَذَا ضَعِيف عندنَا» (v1/313) — "this is weak in our view": the Prophet praised the following generations too, and the qualities are not tied to a time. Abū Ḥanīfa's "we follow the Companions; the Followers we compete with" was because he was himself a Follower (he saw four Companions and debated al-Shaʿbī). (v1/313–314)
• Medina only: true of Medina in the Prophet's time; in Sarakhsī's day, he says, no people anywhere in the abode of Islam were less learned or more plainly ignorant than those in Medina. The ḥadīths describe Medina during the hijra. (v1/314) The Prophet's family only (the "two weighty things" ḥadīth; Q 33:33): their honours are agreed, but consensus rests on texts and meanings not tied to lineage (Q 31:15). (v1/314–315)

Back to the list

B8.10 · شرط الإجماع · The end of the generation is not a condition; ⚑ a lone dissenter blocks consensus only if the others accepted his ijtihād (v1/315–317)

• Who: some, including al-Shāfiʿī, make the death of the generation a condition, citing Abū Bakr's equal stipends (ʿAlī later preferred some) and ʿAlī: "my view and ʿUmar's agreed that umm walads are not sold… then I thought to keep them slaves". (v1/315)
• Position: «وَأما عندنَا انْقِرَاض الْعَصْر لَيْسَ بِشَرْط» (v1/315) — "In our view the end of the generation is not a condition": what consensus establishes is like a text, and the condition would mean consensus never forms, since Followers competed with Companions and so on. Once formed, no one may dissent by opinion. The stipends question was disputed from the start (ʿUmar to Abū Bakr). ʿAlī's "keep them slaves" means they are not freed at death until the heir frees them, not that they may be sold. (v1/315–316)
• Who: ⚑ al-Karkhī (and al-Shāfiʿī): one or two dissenters block consensus ("my Companions are like stars"). Sarakhsī: «وَالأَصَح عِنْدِي مَا أَشَارَ إِلَيْهِ أَبُو بكر الرَّازِيّ» (v1/316) — "the sounder view, to me, is what Abū Bakr al-Rāzī indicated": if the others accepted the dissenter's ijtihād as legitimate (Ibn ʿAbbās on the mother's share with husband and parents, or wife and parents), no consensus without him; if they rejected it (Ibn ʿAbbās on ribā al-faḍl, which he retracted), consensus stands without him. So Muḥammad in al-Imlāʾ: a qāḍī's ruling allowing one dirham for two is void. (v1/316–317) Proofs: "God's hand is with the jamāʿa"; "hold to the great mass". (v1/317)
• Who: Abū Ḥāzim al-Qāḍī: when the rightly guided caliphs agree, that is consensus and dissent is not counted ("my Sunna and the Sunna of the rightly guided caliphs"). So he ordered the caliph to return estate funds taken from dhawū al-arḥām cases. When Abū Saʿīd al-Bardaʿī objected that the matter had been settled on Zayd's view, he answered that he did not count Zayd's dissent against the rightly guided caliphs, and that he had ruled, so no one could undo it after him. (v1/317)

Back to the list

B8.11 · حكم الإجماع · Consensus has grades: the Companions' explicit consensus makes its denier an unbeliever; lower grades do not; ⚑ a later consensus after an earlier dispute (v1/318–321)

• Muḥammad's four sources (from Hishām): what is in the Qurʾān, the Sunna, what comes from the Companions, and what the Muslims see as good, and what resembles each. (v1/318)
• Grades:
  1. «فَفِي هَذَا بَيَان أَن مَا أجمع عَلَيْهِ الصَّحَابَة فَهُوَ بِمَنْزِلَة الثَّابِت بِالْكتاب وَالسّنة فِي كَونه مَقْطُوعًا بِهِ حَتَّى يكفر جاحده» — "This shows that what the Companions agreed on is like what the Kitāb and Sunna establish, certain, so that its denier is an unbeliever." It is the strongest, as it includes the people of Medina and the Prophet's family. (v1/318) Objection: the Prophet himself erred by opinion (Q 9:43; 8:67). Reply: he was protected from being left in error. (v1/318)
  2. The Companions' dispute on set positions (the reward for returning a runaway slave): no new view may be invented, but it is lower, being inferred, so its denier is not an unbeliever. The "choose" (ikhtārī) divorce case is not a third view: al-Karkhī traced it to Muʿādh. (v1/318–319)
  3. Consensus of a later age: like a mashhūr report; the denier is not an unbeliever, but it can abrogate. (v1/319)
  4. ⚑ Consensus after a dispute in an earlier age: some say it is not consensus. «وَعِنْدنَا هُوَ إِجْمَاع وَلكنه بِمَنْزِلَة خبر الْوَاحِد فِي كَونه مُوجبا للْعَمَل غير مُوجب للْعلم» — "In our view it is consensus, but like a single report: it binds action, not knowledge." (v1/319)
• Who: ⚑ Sarakhsī's teacher (al-Ḥalwānī): consensus on Muḥammad's view only; not on Abū Ḥanīfa's and Abū Yūsuf's, since Muḥammad voids a qāḍī's ruling allowing sale of an umm walad, while they let it stand. Sarakhsī: «وَالْأَوْجه عِنْدِي أَن هَذَا إِجْمَاع عِنْد أَصْحَابنَا جَمِيعًا» (v1/320) — "the more cogent view, to me, is that this is consensus according to all our colleagues"; Abū Ḥanīfa lets the ruling stand only because whether this counts as consensus is itself disputed. (v1/319–320)

• Objection → Reply: the dead dissenter's view does not die with him; otherwise Ibn ʿAbbās and Ibn Masʿūd would be misguided; and you drop the ḥadd in the "khalīya" case because ʿUmar held it revocable. Reply: consensus belongs to the living, who enjoin the right; it is like the Prophet's confirmation of a fatwā, which does not make earlier holders misguided (Qubāʾ praying toward Jerusalem; Ibn ʿAbbās's return on mutʿa). The khalīya ḥadd is dropped for the doubt that the scholars' dispute about this kind of consensus creates. (v1/320–321)

• [Editorial] Debate note (bāb): for debate, prepare these three questions. (1) Silent consensus (B8.08) holds only after the fatwā reaches the silent scholars and time for reflection has passed. (2) The lone dissenter (B8.10) blocks consensus only if the others accepted his ijtihād. (3) The ḥadīth scholar without fiqh is not counted in consensus (B8.09). Quote Sarakhsī's conditions with each; they are narrower than the slogans.

Back to the list

Previous · All chapters · Next