How rulings are derived

Fiqh is the body of rulings. Uṣūl al-fiqh is the method by which those rulings are reached — which texts count as evidence, in what order, how a text is read, and what a scholar may and may not do when the texts run out. It exists for one reason above all others: so that a human conclusion is never mistaken for the word of Allah. Learning even its outline changes how you hear religious speech, because you stop asking only what is the ruling and start asking where did this come from.

Three different sentences, three different weights

Most confusion in religious argument comes from three sentences that sound alike and are not alike at all. Keep them separate and most of the noise falls away.

Read that carefully, because it settles the whole question of tone. The Prophet ﷺ describes a qualified man reaching a wrong ruling and still being rewarded. That is only coherent if scholarly conclusions are understood as effort, not as revelation. Where two schools disagree, at least one of them is wrong — and the one that is wrong is not thereby disgraced. What is a fault is presenting that effort to ordinary Muslims as though Allah had said it.

The four sources everyone agrees on

Sunni scholarship agrees on four sources, in a fixed order of recourse. The order is not arbitrary: each later one only operates where the earlier is silent, and none of them may contradict the earlier. The Qur'an itself names the first two and tells you where to go when people disagree:

1 · The Qur'an — the text itself

The Qur'an is the first reference and the one that overrules everything else. What it contributes is not a law code — only a minority of its verses are legal at all. It gives the obligations, the prohibitions, the limits, and above all the reasons: it tells you why a thing is commanded far more often than it tells you exactly how. Some of its rulings are decisive and admit no reading around them — "Allah has permitted trade and forbidden ribā" (2:275) is one. Others are stated in words that genuinely bear more than one sense, and those are where honest scholars part company.

2 · The Sunnah — what the Prophet ﷺ said, did and approved

The Sunnah is not a supplement to the Qur'an; the Qur'an assigns it its job. Allah says the Messenger was sent to make the revelation clear to people (16:44), and commands obedience to what he brings:

What the Sunnah actually contributes is best seen in the prayer. The Qur'an commands the prayer dozens of times and never once states its number of units, its timings in the clock sense, or its words. All of that is Sunnah — and he ﷺ handed it over as something to be copied, not reasoned out:

Because the Sunnah reaches us by transmission, an extra step stands in front of it that never applies to the Qur'an: is this report actually established? A ruling built on a weak report is a ruling built on nothing. This is why the site names a report's grade wherever it can, and says plainly when a report is weak.

3 · Ijmāʿ — consensus

Ijmāʿ means the agreement of the qualified scholars of a generation on a ruling. Where it is genuinely established it is treated as decisive, on the reasoning that the whole community of scholars could not settle together on an error. The report usually quoted for this — that this Ummah will not agree upon misguidance — is narrated in the Sunan collections and elsewhere; its chains have been criticised by the hadith scholars, and those who accept its meaning generally do so from the weight of many weak routes together rather than from any single sound chain. It is honest to say that much rather than to quote it as though it were in the Ṣaḥīḥs.

Two cautions follow, and they matter more than the definition. First, most reliably attested consensus concerns matters already obvious from the texts — that the five prayers are obligatory, that pork is forbidden. Second, claims of consensus are made far more often than consensus can be shown. Imām Aḥmad is reported to have warned against exactly this: a man who says "the scholars have agreed" is claiming knowledge of what every scholar of an age held, which he cannot have. When you hear "there is ijmāʿ on this", it is fair to ask who verified it and when.

4 · Qiyās — reasoned analogy

Qiyās extends a ruling from a case the texts settle to a case they do not, when both share the operative reason (علة, ʿillah) behind the ruling. The standard illustration: the Qur'an forbids khamr (5:90), and the operative reason identified in the texts is intoxication — so a substance unknown in that era which intoxicates falls under the same prohibition, not because it is named, but because the reason that got wine named applies to it.

Be clear about what qiyās is. It is not a fifth voice alongside revelation; it is an attempt to apply revelation to a case revelation did not mention. Its result is only as strong as the identification of the ʿillah, which is a human judgement — and where scholars identify the ʿillah differently, they reach different rulings from the same verse. Usulis point to the Qur'an's call to reason from what you observe — "so take a lesson, O you of vision" (59:2) — as an indication that this kind of reasoning is legitimate; that is an inference drawn by scholars, not a verse instructing anyone to perform qiyās, and it should not be quoted as if it were.

The order above is often taught through a narration about Muʿādh ibn Jabal, sent to Yemen and asked how he would judge: by the Book of Allah; then, if he found nothing, by the Sunnah of His Messenger; then by his own considered effort. The chain of that narration is graded weak (Sunan Abī Dāwūd 3592), and it is fair to say so even though the jurists accepted its substance — the order it describes is established by the verses above without needing it. A weak report is a poor foundation even for a conclusion that happens to be sound.

The disputed sources — named as disputed

Beyond the four, later scholars proposed further considerations. These are not agreed, and no school treats them as revelation. They are tools for the space the texts leave open, and the schools accept, restrict or reject them differently. Anyone presenting one of these as "the ruling" without saying which tool produced it is skipping a step you are entitled to see.

• DisputedIstiḥsān — departing from the answer strict analogy would give, in favour of a stronger consideration in the texts. Its critics argued it opens a door to ruling by preference; its defenders replied that it is itself governed by evidence, not taste.
• DisputedAl-maṣāliḥ al-mursalah — weighing a public benefit the texts neither endorse nor forbid. Powerful, and for that reason the most argued-over: everything can be dressed as a benefit, which is precisely why its critics restricted it.
• DisputedSadd adh-dharāʾiʿ — blocking a permitted act because it reliably leads to a forbidden one. The disagreement is rarely about the principle and almost always about how reliably "reliably" must hold.
• DisputedʿUrf — established local custom, used to settle what an undefined term in a contract or an oath means. Custom never makes the forbidden lawful; at most it fills in a blank the texts left to people.
• DisputedIstiṣḥāb — presuming a known state continues until evidence shows it changed. This is the reasoning behind the familiar rule that certainty is not removed by doubt.
• DisputedQawl aṣ-ṣaḥābī — the ruling of an individual Companion where no text settles the matter. Some treated it as evidence, some as strong guidance, some as one opinion among others.
• DisputedSharʿ man qablanā — law revealed to earlier prophets, where our own revelation reports it without cancelling it. Accepted narrowly by some, declined by others.
• DisputedʿAmal ahl al-Madīnah — the settled practice of the people of Madinah, argued by some to preserve the Prophet's ﷺ practice more reliably than a lone chain. Others did not grant it that weight.

The five rulings — where every action lands

Every human act is sorted into one of five categories. Two of them bind, two of them lean, and one leaves you free.

Two working defaults sit underneath the five, and they too are scholarly principles rather than verses — though each is argued from the texts. In ordinary worldly matters, things are permitted unless evidence forbids them; in acts of worship, the reverse holds — nothing is legislated except by evidence, which is why an added devotion needs a proof and a new kind of food does not. That second default is the one the Prophet ﷺ stated directly:

Note too that the middle of the scale is where scholars are genuinely needed, and the Prophet ﷺ said as much — the extremes are plain, and the ground between them is not:

How the texts are actually read

A great deal of what looks like scholars disagreeing about the religion is scholars disagreeing about language. Three habits of reading account for most of it.

General and specific — ʿāmm and khāṣṣ

A text stated generally may be narrowed by another text. The Qur'an gives divorced women a waiting period of three qurūʾ:

Sūrat aṭ-Ṭalāq then carves cases out of that generality: a woman past menstruation or who does not menstruate waits three months, and a pregnant woman's period runs until she delivers (65:4). The general verse is not contradicted; it is specified. That is a settled operation, and both verses are revelation.

Now notice the other half. The word qurūʾ in 2:228 carries two meanings in Arabic — a menstrual period, and the interval of purity between them — and the Companions themselves are reported to have differed over which is meant. The schools split accordingly, which is why the waiting period is counted differently in different madhhabs. That difference is not one school following revelation and another abandoning it. It is one Arabic word bearing two senses, and later scholars weighing which the verse intends. Saying so plainly is exactly the distinction this page exists to draw.

Absolute and qualified — muṭlaq and muqayyad

A text may state a duty without conditions while another states it with one. Scholars then ask whether the condition governs both places or only the one it appears in — and their answers differ. The same applies to commands and prohibitions: a command in Arabic can express an obligation, a recommendation, or permission after a restriction, and a prohibition can express prohibition or mere dislike. Which one is meant is read from context and from other texts. That reading is ijtihād, however confidently it is delivered.

Abrogation — naskh

Revelation came down over twenty-three years into a community being changed, and in a few places a later ruling replaced an earlier one. Allah states the principle:

The clearest illustration is the prohibition of wine, delivered in stages: first a statement that in wine and gambling there is great sin and some benefit, and the sin is greater (2:219); then a prohibition on approaching prayer while intoxicated (4:43); then the decisive command to shun it altogether (5:90). A ruling that stood at one point no longer stands, and the Qur'an preserves the record of both. The same happens in the Sunnah — visiting graves was forbidden and then permitted, and the later permission is what stands (see Visiting Graves).

Three limits keep naskh honest, and they are worth memorising:

• Only revelation abrogates revelation. Nothing produced after the Prophet's ﷺ death — no consensus, no analogy, no scholar, no ruler — can abrogate anything. Revelation ended; the door closed with it.
• It requires proof of sequence. To claim abrogation you must establish which text came later. Without that, two texts that appear to conflict are simply two texts a scholar has not yet reconciled.
• It is over-claimed. Long lists of "abrogated verses" circulate, and later scholars examining them found most cases to be specification, or qualification, or a reconciliation the claimant had missed — and cut the genuine instances down to a small number. Treat any confident claim of abrogation as something to ask the evidence for.

What a fatwa is — and what it is not

A fatwā is an answer: a qualified scholar's considered reply to a question, telling the questioner what he understands Allah's ruling to be in that situation. The Qur'an directs the non-specialist to ask:

Now the negatives, which are where the misunderstandings live.

• A fatwā is not legislation. It does not create a rule. It reports what the scholar believes the existing evidence requires — the way a doctor's diagnosis reports a condition rather than causing it. Legislation belongs to Allah; a mufti's task is to convey, not to enact.
• A fatwā is not binding on anyone but the person who chooses to act on it. It does not bind other scholars, and it does not bind a questioner who follows a different qualified scholar. This is distinct from a judge's ruling (qaḍāʾ), which does bind the parties to the case it decides — and binds nobody else.
• A fatwā is answer-shaped by its question. Change the facts and the answer can change: the questioner's circumstances, capacity, hardship and situation are part of the input. This is why a fatwā given to one person is often useless when forwarded to another, and why every page on this site ends by telling you to take your own case to a scholar.
• A fatwā can be wrong. The hadith at the top of this page says so outright. A mufti is a fallible reader of infallible texts.
• A fatwā is not the Qur'an, and does not inherit its authority. Two things follow: you may ask what the evidence is, and a scholar of integrity will tell you — and where a fatwā is shown to contradict an established text, it is the fatwā that gives way.

Telling revelation from conclusion

You do not need to be a jurist to hear the difference. You need a few habits.

And the two questions that do most of the work: where is this from, and who disagrees, and on what basis. A ruling with an honest answer to both is one you can act on knowing what you are acting on. A ruling with no answer to either may still be correct — but you are being asked to take a human conclusion on the authority of revelation, and that is the one trade this discipline exists to prevent.

Common questions

Related: Rulings from Quran & Hadith · Comparing the Four Madhhabs · How Hadith Are Graded · Occasions of Revelation · The Scholars of Islam · Glossary