al-Ḥujurāt 49:9

Wa in ṭāʾifatāni mina'l-muʾminīna'qtatalū fa-aṣliḥū baynahumā, fa-in baghat iḥdāhumā ʿala'l-ukhrā fa-qātilu'llatī tabghī ḥattā tafīʾa ilā amri'llāh, fa-in fāʾat fa-aṣliḥū baynahumā bi'l-ʿadli wa aqsiṭū, inna'llāha yuḥibbu'l-muqsiṭīn

"If two groups of the believers fight one another, set things right between them. If one of them attacks the other, then fight the one that attacks until it returns to God's command. If it returns, set things right between them with justice, and act equitably. God loves those who act equitably."

In the first clause of the verse stands the single most striking expression in the sūra: ṭāʾifatāni mina'l-muʾminīn — "two groups of the believers."

Two sides that have come to armed conflict are still called "believers".

To see how unusual this is, one must look at what the verse does not do. The verse does not say:

• "One of the two groups is not to be counted as having believed."
• "Those who come to conflict cease to be believers."
• "Of the two, one is a believer and the other is not."

And, more importantly: the attacking side is not named either. The verse does not call it kāfir, does not call it munāfiq, does not even call it fāsiq. The only word it uses is a verb: allatī tabghī — "the one that attacks". Not an epithet but a verb. What is described, that is, is not the identity of the persons but what they are doing at that moment.

And the next verse will carry this further still: innama'l-muʾminūna ikhwa — "the believers are but brothers." The two sides in conflict are not merely counted as believers; they are counted as brothers.

The consequence of this detail of word order is heavy and must be written plainly: the text does not put the sides of a conflict outside the religion. What the Qurʾān sets up here is not a ruling about the identity of the sides but a ruling about their conduct. This is the sharpest instance of the measure recorded again and again in this commentary: what the verse describes are qualities, not identities.

What the neglect of this measure has produced in Islamic history is well known: in internal conflicts, the sides declaring one another outside the religion. The wording of the verse is closed to this, and the great majority of the classical scholars have recorded, on the basis of this verse, that the sides of an internal conflict cannot be declared unbelievers.

I write this as a note of history and of legal theory; no side is taken in this text about any current conflict.

Root: ط-و-ف. This root was treated in the sections on al-Raḥmān and al-Wāqiʿa; its root sense is to turn, to go round about. Ṭawāf, ṭāʾif and ṭāʾifa are from the same root.

Ṭāʾifa, according to what the lexicographers record, comes from the idea of "a company that turns about a thing"; in usage it means a group of people, and its limit in number is indefinite — definitions ranging from "at least one person" to "a company" are transmitted in the sources.

The indefiniteness serves a purpose here: the verse does not fix the scale of the conflict. It may be two people, or it may be two tribes.

This is the most discussed grammatical point of the verse.

The subject is dual: ṭāʾifatāni — "two groups". But the verb comes in the plural: iqtatalū — "they fought" (for three and more).

The expected form would have been iqtatalā (dual). Why the plural?

The explanation the grammarians record is this: ṭāʾifa is singular in wording but a company in sense. When "two groups" is said, those actually fighting are not two persons but two crowds. The verb has followed not the wording but the sense. In Arabic this is called ḥaml ʿala'l-maʿnā (construction according to sense).

And the real beauty of the word order is here: after the verb has come in the plural, the command that follows immediately returns to the dual: fa-aṣliḥū baynahumā — "set things right between the two of them."

The sequence is: two → many → two.

I record this as an observation about the word order, and I add the following reading as my own: when a conflict begins the sides scatter, grow into crowds, and who is with whom becomes unclear — the verb goes into the plural. The work of reconciliation, on the other hand, again presupposes two sides; for an agreement can be made only between two sides with definite parties. The verse makes a scattered conflict solvable by reducing it once more to two sides.

Form: Form VIII (iftiʿāl) — iqtatala. This form here states reciprocity: they fought one another. The verse, that is, describes not a one-sided attack but a mutual conflict — the determining of the attacking side will come at a second stage.

Root: ص-ل-ح. This root was treated in the sections on al-Qalam and al-Munāfiqūn; the sense recorded: a thing's being in a serviceable, sound, fit condition. Ṣalāḥ — soundness. Ṣulḥ — peace: the mending of a relation that has broken down. Its opposite is fasād.

Form: Form IV (ifʿāl) — aṣlaḥa: to set right, to repair. The verb, that is, is transitive: someone else sets a thing right.

And the picture the root carries is important: iṣlāḥ is not making from nothing but repairing what has broken. When a vessel is broken, the work done is iṣlāḥ; not the making of a new vessel. The word presupposes that what is repaired was sound before.

This carries the whole logic of the verse: between the two sides in conflict there is a relation to be repaired. The tenth verse will give it its name: brotherhood.

Who is the one addressed by the command? The verb is plural: aṣliḥū — "set right". And the one addressed is neither the first nor the second side; it is the third party. The verse, that is, lays an obligation on those who stand outside the conflict.

This is the most practical side of the verse and is worth recording: to stand outside a conflict, on the verse's terms, produces not a right to neutrality but an obligation to intervene. Standing aside is not counted among the options.

Root: ب-غ-ي. Its root sense is to want, to seek, to demand. Ibtighāʾ — seeking; bughya — the thing sought. In the Qurʾān: "Seek of God's bounty" (al-Jumuʿa 62/10) — wa'btaghū min faḍli'llāh.

How, then, does one pass from "to want" to "to attack"?

The transition the lexicographers record is this: the demand's overstepping the bound. A person wants what is his; when he begins to want what is not his, the act turns into baghy. Baghy, that is, is wanting in its boundary-overstepping form.

The derivatives of the root show this line:

• بَغْي (baghy) — wrongdoing, injustice, attack.
• بَاغٍ (bāghī) — one who oversteps, who attacks. Its plural is bughāt.
• بَغِيّ (baghiyy) — used in one place in the Qurʾān of unchastity: "Nor have I been unchaste" (Maryam 19/20).

The common idea: a wanting that oversteps a boundary.

And this agrees with the whole of the sūra. Fisq was coming out of the husk; baghy is a demand that oversteps a boundary; ḥujra is the boundary itself. The sūra is woven with words of boundary.

The striking side of the usage in the verse: the verse begins with baghat (perfect: it attacked) and continues with allatī tabghī (imperfect: the one that is attacking). The object, that is, is the side whose attack continues. When the attack ends the description ends too — and the close of the verse will say this openly.

This is the most beautiful word in the verse.

Root: ف-ي-ء. The root sense the lexicographers record: to turn, to come back.

And the concrete picture of the root is this: الفَيْء (fayʾ) is the shadow that comes back after the sun has passed its zenith.

Arabic distinguishes two words for shadow, and the difference lies exactly here:

The picture is this: in the morning the shadow is long; as the sun rises it shortens; at noon it comes to its shortest — it almost disappears; then it comes back and lengthens until evening.

That is fayʾ: the return of the shadow. The coming back of what was thought to be lost.

And that this word was chosen in the verse is worth noticing. The verse could have said "ḥattā tarjiʿa (until it returns)" — that is the common verb for returning in Arabic. It chooses the verb tafīʾa.

What the difference brings I record as my own reading: rujūʿ is a neutral returning — in any direction. Fayʾ, however, is a returning to one's own former place, and it is the name of a natural event: the shadow turns back not under compulsion but of itself, by the course of the sun. The word pictures the return not as a surrender but as a coming back to a natural state.

The attacking side, that is, is called not to "be defeated and surrender" but to return to its place.

The same root occurs in the Qurʾān as a legal term too: fayʾ is property obtained without fighting and "returning" to the community (al-Ḥashr 59/6-7). In the section on al-Ḥashr this verse (49/9) had already been named. What the two uses share is the same: a thing's returning to where it belongs.

The most critical word in the command is this: "until."

Fa-qātilu'llatī tabghī — fight the side that attacks. But the sentence does not end there: ḥattā tafīʾa ilā amri'llāh — until it returns to God's command.

The fighting, that is, has a purpose and an end point, and neither of them is the destruction of the attacking side. The purpose: return. The end point: the return's taking place.

And the verse says as well what is to be done after the return: fa-in fāʾat fa-aṣliḥū baynahumā bi'l-ʿadl — "if it returns, set things right between them with justice."

Note: after the return, no punishment of the attacking side is commanded. What is commanded is a repetition of the command given at the head of the verse: set things right between them. But this time with a qualification: bi'l-ʿadl.

There is a ring in the word order:

fa-aṣliḥū baynahumā (beginning) → fighting → fa-aṣliḥū baynahumā bi'l-ʿadl (end)

The same command, twice. The fighting in the middle does not take the place of the command — it is a parenthesis opened so that the command can be carried out. Iṣlāḥ stands both before and after the conflict.

I record this as my own reading, but the ring stands in the text and the words are repeated exactly.

At the end of the verse two different roots come one after the other, and both are translated "justice":

عَدْل — root ع-د-ل. Its root sense is to make equal, to level, to make straight. The two pans of a balance coming even. Taʿdīl — setting straight, evening out.

قِسْط — root ق-س-ط. Its root sense is a share, a portion; and from this to give a share as it is due. Qisṭ — a portion. Muqsiṭ — one who gives the portion rightly.

The interesting side of the root: ق-س-ط, when it goes into Form IV (aqsaṭa), states acting justly, whereas in its bare triliteral form (qasaṭa) it states acting wrongfully. The same root takes opposite senses according to the form — it is among the instances the lexicographers name under the heading aḍdād. In the Qurʾān: "As for those who act wrongfully, they have become fuel for Hell" (al-Jinn 72/15) — wa ammā'l-qāsiṭūna.

This distinction was treated in the section on al-Jinn.

The difference between the two words:

In the classical sources this distinction is recorded in different ways and there is no complete agreement. I impose no preference.

But that the two come together in the verse says something: the work of reconciliation must have regard both to the balance of the whole (ʿadl) and to the right of each side (qisṭ). A conflict can be ended "by silencing both sides" — that gives balance but does not give what is due. Or it can be ended by giving one side everything it claims — that gives what is due but does not establish balance. The verse asks for both at once.

إِنَّ ٱللَّهَ يُحِبُّ ٱلْمُقْسِطِينَ — This expression occurs in several places in the Qurʾān: al-Māʾida 5/42, this verse, and al-Mumtaḥina 60/8. It was named in the sections on al-Jinn and al-Mumtaḥina.

This verse is the principal ground of a discussion known in classical jurisprudence as aḥkām al-bughāt (the law concerning the attacking group). The subject is what is to be done when armed division arises within a Muslim community.

I issue no legal ruling here. I relate the headings debated in the classical sources together with their disagreements. Not one of the lines below is a preference of this text and none of them is binding.

The last line is particularly important and requires a note. The command of the verse is plural (fa-qātilū) and its addressee is indefinite. Some of the classical scholars have not read this as the authority to stop a conflict being handed out to everyone at large; on the contrary, they have tied it to the office that bears responsibility for order. The ground is practical: reading the command as open to all could produce the very thing the verse means to prevent — the spreading of the conflict.

This text makes no choice. But this must be recorded: the order the verse itself sets up is plain and beyond dispute — first iṣlāḥ, then (if necessary) stopping, then iṣlāḥ again. To read the verse beginning from the middle item is to break the text's word order.

According to what is transmitted in the sources, the verse came down over a quarrel that broke out between two groups in Medina and went on with slaps or light blows. Some reports state that the sides struck one another with palm branches and sandals — that is, the scale is small.

I relate this account with the note "it is transmitted"; there are differences in the details and I give no names.

But the scale of the account is worth recording: if the report is sound, the verse came down not over a great civil war but over a small quarrel. And even so its language is extremely heavy: iqtatalū (they fought), fa-qātilū (fight).

One explanation of this is the following, and it is pointed to in the classical sources as well: the way to prevent small conflicts from growing is not to treat them as small. By describing in the language of war an incident still at the level of a quarrel, the verse records the weight of the matter from the start.

I relate this as an explanation, not as a settled ruling.

In writing what this verse says to the present I enter into no current conflict; this is the plain rule of the method (Method). What can be recorded is the structure of the verse, and the structure is general:

One. The responsibility laid on the third party. The verse addresses not the two sides in conflict but those outside. To stand aside in a dispute does not figure as an option in the verse's sequence.

Two. The fixing of the purpose. The qualification ḥattā ("until") placed at the end of the command limits the purpose of the intervention. The purpose is not to finish off one side but to stop the conflict. Intervention without a purpose produces nothing but the spreading of the conflict to the third party.

Three. The preservation of identity. This is the verse's firmest note: the sides do not lose their identity even while they are fighting. To count the other side in a dispute as "no longer one of us" in effect destroys the ground of any agreement. In order to preserve that ground the verse goes on calling both sides in conflict by the same name.

Four. That the ending has two stages. The stopping of the conflict and the repairing of the relation are separate works, and the verse commands each separately. A quarrel that has merely been silenced has not ended.