24. Sūrat al-Nūr

Sixty-four verses. Medinan. It takes its name from the expression ٱللَّهُ نُورُ ٱلسَّمَٰوَٰتِ وَٱلْأَرْضِ in the thirty-fifth verse and from the parable of light built in that verse.

This sūra sets side by side two things whose standing together in the Qurʾān looks surprising at first: the rulings that govern the gravest accusations a society can make and the densest image of light in the Qurʾān. The first thirty-four verses of the sūra speak of things that must be covered, of things that must not be made public, of doors that are not entered without leave, of glances that are to be lowered. Then the thirty-fifth verse comes and lights a lamp.

This is the first thing that can be said about the sūra as a whole, and I record it as my own reading: Sūrat al-Nūr takes as its subject the difference between seeing and looking. The glance is lowered (30-31), the house is covered (27-29), an accusation is bound to four witnesses (4, 13), a thing heard is not carried in the mouth (15-16) — and in the middle of all these restrictions a light is lit. The sūra, that is, narrows man's power of sight while binding his capacity to see to something else.

The sūra is also unique in this commentary in another respect: it is the one place in the Qurʾān where a sūra names itself "sūra" (the first verse). I take this up in detail in the section below.

نُور — root ن-و-ر: light.

The analysis of the word was made at Nūḥ, 71/16; I rely on that treatment. The note there was this: nūr is light itself — what illuminates, what removes darkness; sirāj is the body that produces the light. The Qurʾān distributes these two words consistently and keeps the distinction.

From the same root:

And there is a pair inside the root: nūr (light) and nār (fire) come from the same root. The difference between them is that fire both burns and gives light; light only gives light. The thirty-fifth verse will dwell on this distinction: there it will be said of the olive oil that "it would almost give light though no fire had touched it."

It should not be forgotten that the name was given later. The names of sūras are not the text's claim but the tradition's choice. But I can say that this choice is apt, and its ground stands in the text: the root nūr occurs in the sūra not only in the thirty-fifth verse but in the thirty-ninth and fortieth as well, and there it is named together with its opposite (wa man lam yajʿali'llāhu lahū nūran fa-mā lahū min nūr). The sūra's central word is thus built once within the sūra and once negated.

I record this as my own reading; the places of the words can be counted in the text.

There is no disagreement that the sūra was revealed in the Medinan period, and the text itself supports this:

• There is an order of rulings and an authority to apply them: penalties are commanded to a community (2, 4), the procedure of witness is regulated (4, 6-9, 13).
• An accusation spreading within a community is on a scale that would shake the whole of it (11-20).
• There are house doors, rooms, the hour of the midday rest, the coming and going of servants and of children (27-29, 58-59): a settled city life.
• Around the Prophet there is an assembly where business is done in common, and an order of general mobilisation (62-63).
• There is an anxiety about security and a promise of "safety after fear" (55).

I say nothing definite about which year the sūra belongs to. In the sources it is widely transmitted that it falls around the fifth or sixth year after the emigration; there is disagreement over the detail, and this commentary does not give datings it cannot establish. What can be said is that the sūra presupposes a settled Medinan society.

The sūra consists of seven blocks. The boundaries of the blocks are marked by changes of address and by the calls yā ayyuhā'lladhīna āmanū.

The most striking thing about the structure is that the fourth block (27-34) stands immediately before the central verse (35), and the seventh block (58-64) at the end of the sūra. The matter of asking leave is taken up twice: once on entering a house from outside (27-29), once on passing from room to room within the house (58-59). Between them lies an entire section of light.

I record this as my own reading; the places of the verses stand in the text.

The weave of Sūrat al-Nūr is built upon a contrast, and this contrast can be verified from the text:

The left column restricts man's appetite for showing; the right column shows what is to be shown. The sūra does not only say "do not look" — it says "look here."

I record this reading as my own reading. Its ground is the places of the verses in the two columns.

Sūrat al-Nūr is one of the sūras where this commentary's rules of style most need to be observed. I therefore write at the outset what is not done:

One. No legal ruling is issued. From the second, fourth, sixth and thirty-first verses of the sūra, broad discussions have arisen in classical jurisprudence. In this section the disagreements of those discussions are transmitted in tables, none is preferred, and none is binding. To decide about the practical application of a ruling is not the business of this text.

Two. No debate of criminal law is entered. Neither how the penalties are to be applied today, nor whether they are to be applied, nor the current controversies around them are the subject of this text. What is treated is the language of the verses: the roots of the words, the word order, the sequence.

Three. The reports about the ifk affair are not narrated. That the verses from eleven to twenty point to an event is plain, and in the sources that event is transmitted in detail. In this section no personal name is given, no discussion is entered about who was accused, and no details of the reports are transmitted. The reason is the verse itself: the text has not given the names of persons. In the twenty-sixth verse it says ulāʾika mubarraʾūna mimmā yaqūlūn — "they are free of what is said" — and the most correct thing this commentary can do is to stop where the verse stops. What will be treated is the frame the verses themselves give: how a report spreads, how language works, and how "thinking it a small thing" is diagnosed.

Four. No current politics and no current controversy is entered. How the fifty-fifth verse (the promise of istikhlāf) and the thirty-first (covering) are used in the controversies of today is not the subject of this text.

Five. The reader is not scolded. The subjects of this sūra — slander, gossip, the glance, privacy — are the subjects that have produced the most scolding in the language of preaching. This section does not do that.

The word-by-word analysis of the basmala was made in the al-Fātiḥa section; I do not repeat it here. The numbering of the verses counts without the basmala.

One point particular to this sūra may be recorded: in the first verse of a text opened with the names al-Raḥmān and al-Raḥīm there stands the verb faraḍnāhā — "we cut and determined". This nearness between mercy and the drawing of a limit was recorded at the head of al-Ḥujurāt as well. In al-Nūr the relation is a step plainer: most of the limits laid down through the sūra are drawn around something that needs protecting — a person's standing, the inside of a house, the intimacy of a family.

Sūratun anzalnāhā wa faraḍnāhā wa anzalnā fīhā āyātin bayyinātin laʿallakum tadhakkarūn

"This is a sūra We have sent down and whose rulings We have determined precisely. In it We have sent down clear verses, that you may take heed."

This verse is one of the most striking openings in the Qurʾān in point of syntax, and the reason is this:

The Qurʾān uses the word sūra in many places. The addressees are asked to "bring a sūra like it" (al-Baqara 2/23, Yūnus 10/38); it is told what men did when a sūra was sent down (al-Tawba 9/64, 9/86, 9/124; Muḥammad 47/20). But in all these places the word points from outside at a text being spoken of.

Here, however, the sūra names itself. The subject of the sentence is the text itself: sūratun — "this is a sūra". And immediately after, two verbs come about that sūra: anzalnāhā (we sent it down) and faraḍnāhā (we determined it).

The explanations the lexicographers record about the root of the word diverge, and none is certain:

I prefer none of them, and none is binding. But what the three explanations have in common may be recorded: all three arrive at the idea of a "separated portion" — whether walled around, or set apart as a step, or broken off from a whole.

And this is directly connected with the verse's second verb.

Root: ف-ر-ض. And the concrete sense of the root carries the whole weight of the verse.

The primary sense of the root: to notch, to cut a nick, to set a mark by cutting into an object. To cut the notch in which a bowstring will sit, to nick a measure into a rod — that is the verb.

This root was analysed at al-Baqara, 2/68 (lā fāriḍun wa lā bikr — in the description of the cow, in the sense "advanced in age"); the note there was this: fāriḍ, of an animal, means "past bearing, of a determined age", and from the same root come farḍ and farīḍa: an obligation determined precisely, with its limit drawn.

The derivatives of the root show how the concrete sense spread:

And the Qurʾān uses this root in the division of inheritance and in the fixing of a dower: naṣīban mafrūḍan — "a determined share" (al-Nisāʾ 4/7); fa-niṣfu mā faraḍtum — "half of what you determined" (al-Baqara 2/237). In both places the word comes together with a number, a fraction, an amount.

So faraḍnāhā means this: what is inside this sūra was not left indefinite; its bounds were drawn, its shares were set apart, its measure was notched.

And this agrees exactly with the content of the sūra. In Sūrat al-Nūr there are rulings given with numbers: a hundred (2), eighty (4), four (4, 6, 8, 13), five (7, 9), three (58). Few other sūras in the Qurʾān carry so many numbers.

I record this as my own reading, and its ground can be counted: the root of the verb faraḍnāhā is "to notch and set a measure", and the sūra does in fact set measures.

A variant reading is transmitted in the sources: it is recorded that beside wa faraḍnāhā (undoubled) the word is also read with the consonant doubled. The doubled reading is Form II and conveys multiplicity, going into detail: "we determined it in detail", "we determined many rulings one by one". Since I cannot state with certainty which reader read which, I write no names; this is the practice adopted in this commentary. The two readings do not exclude one another: the one conveys the precision of the determining, the other its detail.

In the verse the verb anzalnā occurs twice, and its two objects differ:

The verse thus names the whole and the part separately: first the sūra as a unit, then what is in it, item by item.

This is the finest point of syntax in the verse, and it is worth recording. The text claims its own wholeness: this is not a set of scattered rulings brought together — it is a text sent down as a unit. And this claim is a ground for those who hold that the sūra really has a single weave.

I record this as my own reading; that the verb occurs twice and that its objects differ is the wording of the text.

بَيِّنَات — root ب-ي-ن. The root was treated in detail at al-Ḥujurāt, 49/6 (fa-tabayyanū); the note there was this: the concrete sense of the root is separation, the space between two things (bayn), and bayān is an operation of separating — separating what is mixed, making it distinguishable.

So bayyināt does not mean "easy verses" — it means "verses that distinguish". Verses that separate one thing from another, that make a boundary visible.

And the word will return three more times in the sūra: āyātin mubayyināt (34), āyātin mubayyināt (46), yubayyinu'llāhu lakumu'l-āyāt (58, 59, 61). The distribution of the root through the sūra shows how the sūra names its own work: separating.

That it comes indefinite is worth noting too: āyātin bayyinātin — "certain clear verses". There is no article. The rule the classical lexicographers record is that the indefinite here conveys magnification; I impose no preference.

تَذَكَّرُونَ — root ذ-ك-ر, Form V: to try to remember, to call a thing back to one's own mind.

The choice of the form matters. The verse does not say taʿlamūn (that you may know), it says tadhakkarūn. Dhikr is the bringing back of what has been forgotten; not learning but remembering.

This says something, when the content of the sūra is considered. The subjects of Sūrat al-Nūr — guarding a person's standing, not spreading what one has heard, not entering without leave, lowering the glance — are none of them things man has never known. Instead of giving new knowledge, the verse speaks of calling back something known.

And the word will return in the twenty-seventh verse: dhālikum khayrun lakum laʿallakum tadhakkarūn — at the end of the manners of asking leave, the same verse-ending.

It is rare for a sūra in the Qurʾān to open in this way. Sūras generally begin with a group of letters, with an oath, with praise, with an address, or directly with the subject. Here there is a sentence of presentation: "this is such and such, it was done thus, and this is what for."

And the presentation prepares what comes immediately after. In the second verse a very heavy ruling will come. The first verse says in advance where that ruling comes from and what kind of thing it is: sent down and determined. Neither a custom men produced, then, nor a piece of advice open to interpretation.

I record this as an observation about the word order. Setting a frame before a heavy ruling is a structure seen also in the long legislative sections of al-Baqara: before the ruling, the nature of the ruling is stated.

What this verse says to the present stands in the root of the second verb.

Faraḍnāhā — to notch, to set a measure. A thing's having a measure means that it is not without limit. And the whole subject of the sūra is the domains in which man of himself recognises no limit: speech, curiosity, the glance, accusation.

All of these, left to themselves, tend to grow. A word grows, a curiosity grows, an accusation grows. The sūra's first sentence cuts a notch into the growing.

And the last word of the verse says how this will work: tadhakkarūn — that you may remember. The limit, then, is laid down not as a barrier coming from outside but as a measure remembered from within.

Al-zāniyatu wa'l-zānī fa'jlidū kulla wāḥidin minhumā miʾata jalda, wa lā taʾkhudhkum bihimā raʾfatun fī dīni'llāhi in kuntum tuʾminūna bi'llāhi wa'l-yawmi'l-ākhir, wa'l-yashhad ʿadhābahumā ṭāʾifatun mina'l-muʾminīn

"The woman who commits adultery and the man who commits adultery — flog each of them a hundred strokes. And let no pity for them take hold of you in the religion of God, if you believe in God and in the Last Day. And let a group of the believers witness their punishment."

This verse and those that follow gave rise to one of the broadest discussions in classical jurisprudence. In this section the disagreements of that discussion are transmitted, none is preferred, and none is binding. How the penalty is to be applied today, whether it is to be applied, and the current controversies around it are not the subject of this text.

What will be treated is the language of the verse: the roots of the words, their forms, their order, and the verse-ending.

The verse begins with al-zāniyatu — the feminine first.

This is not the usual order in the Qurʾān's verses of ruling. In the fourth verse the ruling on slander comes with alladhīna yarmūn (masculine plural); the ruling on theft at al-Māʾida 5/38 comes with al-sāriqu wa'l-sāriqa (the masculine first).

In the classical commentaries several explanations are given for this order:

I prefer none of them, and none is binding. The one certain thing is that the order stands so in the text.

But an observation about the word order can be made, and I record it as my own reading: the verse names the two parties separately and equally, and then says the equality once more with kulla wāḥidin minhumā — "each of them". That the ruling is bound to both parties in the same measure is thus said three times: the two nouns, the dual pronoun (minhumā), and kulla wāḥidin.

Root: ج-ل-د. And the concrete sense of the root is directly the name of an organ.

جِلْد (jild) — skin. In Turkish the borrowed word cilt means both skin and the leather binding of a book, and from there a volume of a book as well — each part of a book bound in leather.

And the verb is derived from the noun: jalada — to strike the skin. The literal sense of the verb, then, is not "to punish" but "to strike his skin".

The derivatives of the root:

The choice of the word is worth recording, and this is one of the most important points in the language of the verse.

In Arabic the general verb for striking is ḍaraba. The verse does not use it. The verb it uses carries in its own name where the stroke will reach: the skin.

In the classical juristic literature a conclusion has been drawn from this word, and I transmit it as it is transmitted: because the verb points to the skin, it has been said that the penalty must not go past the skin — that is, that the stroke must not maim or do lasting harm. Over the detail of the application — the force of the stroke, the kind of instrument, which parts of the body are to be avoided, the case of the sick and of the pregnant — there is broad disagreement among the schools, and none of it is given here as a ruling.

And the root's second cluster of senses stands in an interesting place: jalad — endurance; tajallud — fortitude. From the same root come both "to strike the skin" and "to bear up". The connection the lexicographers draw is this: the skin is the body's enduring layer — the first to meet what comes from outside, and the thing that protects what is within.

I record this as an observation about the root; I build no interpretation of the verse upon it.

The verse gives a number: a hundred.

This is the direct counterpart of the verb faraḍnāhā in the first verse. The root ف-ر-ض — to notch, to set a measure. And here the notch is cut: a hundred.

The function of the number is a limit. The penalty is not left indefinite, not left to the discretion of the one who applies it. Not an upper limit but a fixed number is given.

And that the word jaldatan comes in the singular is worth noting: miʾata jalda — "a hundred one stroke". In Arabic the numbers from three to ten take a plural specifier; a hundred takes a singular one. This is a rule of grammar, not a subtlety; but its outcome is worth recording: in saying "a hundred strokes" the sentence leaves the word in the form of a single stroke.

رَأْفَة — root ر-أ-ف. This root was analysed at al-Ḥashr, 59/10, and at al-Ḥadīd, 57/9; I rely on those treatments.

The note there was this, and it sits exactly here:

The lexicographers draw a distinction between raḥma and raʾfa: mercy is general and may sometimes contain severity (as a treatment that gives pain in order to heal a patient); raʾfa is the refusal to give pain at all, the refusal to hurt. The definition recorded in al-Ḥadīd is sharper still: raʾfa is above all the feeling of not being able to bear that another should be harmed.

And the verse's choice of word is built exactly upon this distinction.

What the verse forbids is not mercy. What is forbidden is that raʾfa — the inability to bear witnessing pain — should stop the ruling.

And this distinction agrees with the language of the word elsewhere: the Qurʾān uses raʾfa positively of God and of the Prophet (al-Ḥadīd 57/9, al-Ḥashr 59/10, al-Tawba 9/128). The word, then, is not bad in itself. What is bad is its shifting a measure that has been laid down.

The form of the verb says this too: lā taʾkhudhkum — "let it not take hold of you, let it not seize you". Root أ-خ-ذ: to take, to hold, to seize. Raʾfa is thus pictured here as an agent: something that takes hold of a man and binds his hand.

The sentence does not say "do not pity". It says "let your pity not take hold of you". The feeling is not forbidden; the feeling's stopping the act is forbidden.

I record this distinction as my own reading; its ground is the form of the verb: akhadha — to take hold.

This is the part of the sentence most easily passed over: fī dīni'llāh — "in the religion of God".

The prohibition is not absolute but qualified. It is not said "show no pity anywhere": it is said where the application of a determined ruling is at issue.

And this qualification joins with faraḍnāhā in the first verse. What has been determined stays as the one who determined it determined it. Raʾfa is free in the domains not determined; in what has been determined it cannot alter the measure.

The conditional clause is placed at a noteworthy point. Not after the command of a hundred strokes, but after the qualification that pity must not stop the ruling.

The condition, that is, is tied not to the carrying out of the penalty but to the feeling's not disturbing the measure.

And the particular naming of the Last Day carries a logic: the inability to bear that a ruling should give pain arises from a gaze that stays within this world. The mention of the hereafter lengthens the range of the gaze.

I record this as my own reading; the position of the condition and its content are the wording of the text.

طَآئِفَة — root ط-و-ف: to turn, to go round about a thing. Ṭawāf is from the same root. A ṭāʾifa is the group that goes round a place, that is present there. The word was analysed at al-Ḥujurāt, 49/9 (ṭāʾifatāni mina'l-muʾminīn); I rely on that treatment. The note there: the word states no number, it conveys an indefinite group; in the classical sources broad disagreement is transmitted about its minimum.

And the function of the command here needs dwelling on, because it appears to stand in tension with the sūra as a whole.

From beginning to end the sūra commands covering: do not enter without leave (27), lower your glance (30-31), do not spread what you have heard (15), bind an accusation to four witnesses (4). But here the very opposite is asked: that the penalty be seen.

The explanations widely transmitted in the classical commentaries about the purpose of this command are these:

I impose no preference. But the second explanation's link with what precedes the verse may be recorded: the preceding clause regulated the feeling of the one who applies the penalty (raʾfa). The verse thus limits the applier both from softening and from exceeding the measure: the one by a condition, the other by a witness.

I record this as my own reading.

And there is one further detail in the wording: in saying "let them witness their punishment" the verse says ʿadhābahumā — "their punishment". The word jazāʾ or ḥadd is not used. The weight of the word is not softened.

The juristic discussions arising from this verse are broad. The table below is meant to show the main divisions transmitted in the classical sources. None of them is this text's preference and none is binding. In the detail there are differences of view within each school as well.

This table is not a legal opinion and cannot be read as a guide to practice. The reason it is placed here is the method's rule about disagreement: the disagreement is not concealed, but no preference is imposed.

And one further thing must be recorded: the fourth verse comes immediately after this one and binds the accusation itself to a sanction. The sūra, that is, in laying down the penalty also closes the road that leads to it. This is the most important move in the sūra's arrangement, and it will be taken up in detail at the fourth verse.

What this verse says to the present can be written without entering the debate about penalties and without scolding the reader; what is treated are the words.

One. The general logic of the raʾfa qualification.

The distinction the verse draws — the separating of feeling from measure — is a distinction that works outside the application of a ruling too. When a measure is laid down, what breaks it is most often not ill will but a momentary softening. The verse does not name this a bad thing; its name is raʾfa, and the word is used in the Qurʾān of God as well. All that is said is this: a measure that has been determined is not re-determined according to feeling.

Two. The witness qualification works in two directions.

A sanction applied in the open makes the one who applies it visible too. A penalty applied in secret is a penalty whose measure cannot be checked. The verse's ṭāʾifa qualification can be read for the side of it that secures this — and this reading agrees with what the sūra's later verses repeat: this sūra is from beginning to end about the checking of claims and of power.

Three. And a limit.

From this verse no conclusion follows that "nothing ought to remain hidden"; the rest of the sūra says exactly the reverse. The verse regulates what comes after a ruling has been given. The stage before the ruling — suspicion, accusation, gossip — will be closed off in the sternest terms in the sūra's later verses.

Al-zānī lā yankiḥu illā zāniyatan aw mushrikatan wa'l-zāniyatu lā yankiḥuhā illā zānin aw mushrik, wa ḥurrima dhālika ʿala'l-muʾminīn

"The man who commits adultery marries none but a woman who commits adultery or a woman who associates; and the woman who commits adultery is married by none but a man who commits adultery or a man who associates. This has been forbidden to the believers."

In the second verse the order began with the woman; here it begins with the man. Two verses in succession, in two different orders.

This sets up a balance in the wording, and I record it as an observation: the pair of verses makes it hard to draw any superiority or precedence from which party is named first. The order changes between the two verses.

This verse is one of the places in the Qurʾān most disputed as to the kind of sentence it is, and the source of the disagreement is this: the sentence is built in the indicative, but at its end there stands a prohibiting clause (wa ḥurrima dhālika).

This disagreement takes up wide space in the classical works of commentary and of jurisprudence. Among the schools too there is division over the marriage ruling drawn from this verse: some jurists have drawn a binding prohibition of marriage from it, while the majority have not taken it so and have explained the verse in another way.

I make no preference, I issue no legal ruling, and none of these is binding. I content myself with recording that the disagreement exists.

The sense the lexicographers record: one thing's joining with another, interlocking with it. Tanākaḥati'l-ashjār — the branches of the trees interlaced. The word is used both of the contract of marriage and of union, and occurs in the Qurʾān in both senses.

Which sense is meant in the verse is part of the disagreement above. That the word carries two senses is one of the reasons the sentence can be read in two ways.

What the sentence does is to put both parties in the same scale.

The verse consists of two clauses, and the two are like each other's reflection in a mirror:

The symmetry is exact. And the work of the symmetry becomes visible when it is read together with the sūra's later verses: in this sūra there is no burden laid on one side alone. In the second verse the penalty was bound equally to both parties; here the description is bound equally to both; in the sixth verse the oath of liʿān will be left equal in number for both.

I record this as my own reading; the symmetry itself is the wording of the text and can be counted.

To use this verse as an instrument for ranking people runs against the sūra's own logic.

The reason lies within the sūra: the fourth verse is the verse that binds the naming of someone with this attribute — unless four witnesses are brought — to a penalty. The sūra, that is, takes as its addressee, one verse later, the man who picks up the language of the third verse and fastens it to a person.

This is the same measure recorded in al-Ḥujurāt: the text is an instrument of diagnosis, and the diagnosis is applied first to oneself.

Wa'lladhīna yarmūna'l-muḥṣanāti thumma lam yaʾtū bi-arbaʿati shuhadāʾa fa'jlidūhum thamānīna jaldatan wa lā taqbalū lahum shahādatan abadā, wa ulāʾika humu'l-fāsiqūn

"And those who cast a charge at chaste women and then do not bring four witnesses — flog them eighty strokes, and never again accept any testimony from them. It is they who are the transgressors."

The place of this verse explains the whole weave of Sūrat al-Nūr, and this is the observation about arrangement most dwelt on in this section.

The second verse laid down a penalty for an offence. The third framed a sentence about the parties to that offence. And the fourth bound the ascribing of that offence to someone to a penalty.

The order, then, is this:

The meaning of this order is this: immediately after forbidding an offence, the text forbids the use of that offence's name as a weapon.

And to see how unusual this is, it is enough to look at the general structure of legal texts. A penal norm generally defines the act and lays down its sanction. The binding of the accusation itself to a sanction is a further and a distinct step — because it takes as its subject the misuse of the penal order itself.

The balance the verse sets up is visible in the numbers too:

The penalty for slander, then, is four-fifths of the penalty for the act charged. The number is smaller — but it is on the same scale and of the same kind, and to it is added a permanent loss of a right (the rejection of testimony).

I record this as my own reading, and its ground is the numbers themselves: the verse places an unproved accusation not as a light fault but as something comparable in weight with the offence itself.

And one further observation: in the second verse raʾfa (pity) was forbidden. In the fourth there is no such qualification. What comes in its place is the exception of repentance (5). The qualifications around the two rulings look in different directions: the one prevents the application from slackening, the other opens what comes after it has been applied.

Root: ر-م-ي. Its concrete sense is to throw, to hurl — a stone, an arrow, an object.

And this is the most concrete picture in the language of the verse. The verse does not say "they slander", "they lie", "they accuse". It says "they throw".

The derivatives of the root:

The Qurʾān uses the same verb of an arrow and of a stone: wa mā ramayta idh ramayta wa lākinna'llāha ramā (al-Anfāl 8/17) — there too what is thrown is something concrete.

The picture carries three things, and all three are a description of slander:

1. What is thrown goes from a distance. The thrower does not touch the target; he strikes with a word. 2. What is thrown cannot be recalled. Once out of the hand, the thrower's control is over. 3. What is thrown leaves a mark. A sign stays where it struck.

And this picture goes on living in language: in English too one speaks of hurling an accusation, of slinging mud, of throwing dirt at a man. The same verb, the same picture. I record this as an observation about the language.

The verb's being used without a stated object is also worth noting: yarmūna'l-muḥṣanāt — "they throw at chaste women". What they throw is not said. In Arabic, when this verb is left absolute in such a context, the content of the charge is understood from the context. The text does not repeat the charge itself.

I record this as an observation about the word order, and I add this: the verse does not itself utter the words it forbids.

Root: ح-ص-ن. This root was analysed at al-Ḥashr, 59/2 (ḥuṣūn — fortresses); I rely on that treatment. The definition there: to make firm, to take into protection, to make unreachable.

The choice of the word describes what slander does. A muḥṣana is, in the root of the word, one who is protected like a fortress. And slander is the throwing of something over that wall (yarmūna).

The picture the two words build together is this: the piercing of an area protected from within by something thrown from outside.

I record this as my own reading; the concrete senses of the two roots stand in the dictionary.

And this picture is the same as the picture recorded in al-Ḥujurāt. There, at 49/4, there were those who called out from behind the wall: the wall stood in its place but was crossed by a voice. Here too the fortress stands in its place but is crossed by a word thrown.

There is disagreement over the scope of the word, and it needs to be transmitted:

I transmit these as views that are transmitted; I make no preference, and none is binding.

Here is the heart of the verse, and its place in the arrangement must be recorded.

The condition is not to prove the slander — it is to prove the charge. The verse is built thus: yarmūna … thumma lam yaʾtū bi-arbaʿati shuhadāʾ. The penalty, that is, comes not because the charge has been proved false, but because the charge has not been proved true.

This shows where the burden of proof is placed: on the one who makes the claim.

And the particle thumma is worth noting. Not wa but thumma — "then". In Arabic this particle conveys that time has come in between. The one who throws the charge is thus given a span in which to bring his witnesses; the ruling is framed not immediately after the charge but after the chance of proof has gone unused.

I record this as an observation about the word order.

That the number is four: in no other matter does the Qurʾān require four witnesses. The general ruling on witness at al-Baqara, 2/282, is two men. This, then, is the heaviest condition of proof in the Qurʾān.

And the consequence of this weight is this, and it must be said openly: the condition is so heavy that in practice it makes proof all but impossible. In the classical works of jurisprudence this weight has been dwelt on separately, and the conditions of the witnessing (seeing the same event, at the same moment, in the same manner) have been kept extremely narrow.

In the classical sources a widespread conviction is transmitted that this is the result of an intention: the purpose of the ruling is that the act, so long as it stays hidden, should not be pursued and should not be brought before the community. I transmit this conviction as a view that is transmitted.

And the verses' own arrangement agrees with that conviction, and I record this as my own reading: the nineteenth verse will announce a heavy recompense for "those who love that indecency should spread among those who believe." The sūra, that is, holds a line that aims not at the act but at the publishing of the act, and the condition of four witnesses is the first link in that line.

The same root: ج-ل-د. The same verb as in the second verse, the same form, the same specifier.

The sūra's two penalties are expressed with the same word. This is a fact of the arrangement, and it shows this: the text places slander not as a separate category but as a violation of the same kind.

This is the second and more lasting half of the penalty, and its logic stands within itself.

The man had come forward with a claim to have witnessed: he had said "I saw". And he could not support his claim with four witnesses. The result is the loss of the competence to bear witness.

The sanction, then, is chosen from the kind of the violation itself. The weight of the word of the man who has misused his word is taken away.

I record this as my own reading; its ground is that the word in the two clauses is from the same root: shuhadāʾ (witnesses) and shahādatan (testimony). The verse uses the same root in both the condition and the penalty.

أَبَدًا — "never, for ever". The word conveys that the sanction is without term. And the fifth verse will come directly upon this word.

فَاسِق — root ف-س-ق. This root was analysed in detail at al-Ḥujurāt, 49/6; I rely on that treatment. The note there was this: the concrete sense of the root is a date's coming out of its skin; fisq is not the name of a sin but the name of a position — a thing's being outside where it ought to be inside.

And the place of the word here forms a ring with the one in al-Ḥujurāt:

The two verses look at the same matter with the same word: the carrier of an unverified word. al-Ḥujurāt regulated him from the side of the receiver (fa-tabayyanū — verify); al-Nūr regulates him from the side of the source.

I record this complementarity as my own reading, and at the end of the sūra I shall give in a separate table how the two sūras complete each other.

And the work the three sanctions do together:

All three together reduce the weight in every domain of the man who uses his word without measure.

One. That the accusation is itself an act.

The ground the verse lays down is this: to charge a man with something is an act. Not the expression of an opinion, not a guess, not a question. And because it is an act it has consequences.

This is the distinction most often lost in everyday speech. The man who puts a claim about most often sees himself only as "passing it on", "saying it", "conveying what he heard". The verse's verb does not recognise that position: yarmūn — they throw.

Two. Where the burden of proof lies.

The verse does not ask the accused to prove the charge false. It asks the accuser to prove the charge true.

This distinction points in the same direction as the logic of what is called in legal thought the presumption of innocence. This is not a claim of identity — what the verse builds is not a procedure of trial but a regulation of accusation. But what the two structures have in common is this: a claim that cannot be proved is not counted proved; and the maker of an unproved claim pays a price.

Three. The meaning of one qualification: abadā.

The permanent rejection of testimony looks very heavy at first. And it is heavy. But the fifth verse comes directly after and opens a door upon that weight. To read the two verses apart is to read both of them wrongly.

Illa'lladhīna tābū min baʿdi dhālika wa aṣlaḥū fa-inna'llāha ghafūrun raḥīm

"Except those who repent after that and set things right. God is truly forgiving, merciful."

This short verse is the most cited example of a debate that has a name of its own in the Islamic tradition of legal theory: when an exception follows several clauses, does it apply to all of them or only to the last?

In the fourth verse there were three rulings:

1. Eighty strokes 2. That testimony is never to be accepted 3. The name fāsiq

And the fifth verse begins with illā. The question is: which of these does the exception apply to?

The first view is grounded on the rule that in Arabic an exception following clauses that come one after another applies to the nearest of them. The second is grounded on the rule that the exception applies to all the clauses, and appeal is also made to the general ruling on repentance.

This division is a question of legal theory over which the schools differ, and it is debated in the classical works of jurisprudence by way of this verse. In this section no preference is made, no legal ruling is issued, and none of these is binding.

But what the two views have in common may be recorded: both agree that the penalty itself (the eighty strokes) falls outside the scope of the exception. Repentance, that is, does not retroactively lift a sanction that has been laid down. Where they part is over whether the lasting part of the sanction is lifted.

The condition of the verse is not one verb but two, and both are worth noting.

تَابَ — root ت-و-ب. Its concrete sense is to turn, to turn back. The word was taken up in al-Taḥrīm and at al-Ḥujurāt, 49/12. Tawba is to turn from one direction and pass into another; it is the name not of a feeling but of a movement.

أَصْلَحَ — root ص-ل-ح. The root was analysed at al-Ḥujurāt, 49/9-10; I rely on that treatment. The note there: iṣlāḥ is to repair what has been spoiled, to make it workable again. The word conveys not an intention but a work.

The coming of the two verbs together fixes the content of the condition:

And without the second verb the first is not counted sufficient. This is the most concrete note in the verse's arrangement: slander has left a damage outside, and turning alone does not remove that damage.

In the classical works of jurisprudence it has been debated what this second verb means — whether it is retracting his word, or setting his own condition right, or remaining upon the straight road for a long time. I content myself with transmitting this disagreement; I make no preference.

The same pair occurred in al-Ḥujurāt too, in the discussion of repentance for backbiting, and a similar division was recorded there as well: in repenting for the violation of another's right, what is to become of the party violated.

The demonstrative is worth noting: dhālika — "that". What it points to is not said openly: the slander, the penalty, or both at once?

In the classical commentaries all three possibilities are mentioned. The leaving of the reference indefinite leaves indefinite — that is, wide — the stage at which the door of repentance stands open.

I record this as an observation about the word order.

The sūra's first verse-ending comes here, and it is a verse-ending of forgiveness.

The sūra had opened with two heavy penal rulings. The first pair of divine names comes after those rulings, and inside a note about repentance.

غَفُور — root غ-ف-ر: to cover. Mighfar is from the same root: what covers the head and protects it. The root's sense of "covering" was recorded in Muḥammad, where it was compared in a table with the other roots of covering.

And the word's place here is worth noting: immediately after penalising a man's laying bare another's fault, the sūra closes with the name of God that covers.

I record this as my own reading; its ground is that the root of the word ghafūr is "to cover" and that the subject of the verse is "laying bare". The two words stand face to face in the same pair of verses.

One. A door set beside a permanent sanction.

The fourth verse said abadā — "never". The fifth says illā — "except". One verse later, a sanction without term is bound to a condition.

This has a general logic, and it may be recorded thus: the permanent identification of a man with a single fault is not absolute even within the order the text itself establishes. There is a sanction, it is heavy, and it has a door.

Two. And the condition of the door: not remorse alone, but repair.

The verse asks for two verbs. In today's terms this corresponds to something like this: to retract a claim is not the same thing as never to have put it about. The word has gone into circulation. The verse's second verb (aṣlaḥū) means the work to be done where the word left off.

Three. A limit: this verse is not an instrument for settling accounts.

To use the verse in order to measure whether another man's repentance is valid runs against the whole logic of the sūra. The verse binds the condition of the one who repents to God: fa-inna'llāha ghafūrun raḥīm.

Wa'lladhīna yarmūna azwājahum wa lam yakun lahum shuhadāʾu illā anfusuhum fa-shahādatu aḥadihim arbaʿu shahādātin bi'llāhi innahū la-mina'l-ṣādiqīn · Wa'l-khāmisatu anna laʿnata'llāhi ʿalayhi in kāna mina'l-kādhibīn

"And those who cast a charge at their wives and have no witnesses but themselves — the testimony of such a man is four testimonies sworn by God that he is of the truthful. And the fifth, that the curse of God be upon him if he is of the liars."

Liʿān (لِعَان) is a subject of its own in classical jurisprudence, and there is broad disagreement among the schools about its consequences. In this section no juristic detail is entered into and no ruling is issued. What will be treated is the structure and the language of the verses.

It opens with the same verb: yarmūn — they throw. The very verb of the fourth verse.

But the condition changes: in the fourth verse four witnesses were required; here there is the qualification wa lam yakun lahum shuhadāʾu illā anfusuhum — "if they have no witnesses but themselves".

The verse thus sets apart the one case in which the heavy condition of proof laid down in the fourth verse cannot be applied, and there it puts an oath in the place of a witness.

That the number is kept is worth noting: four. The number of witnesses was four; the number of oaths is four as well. The text does not change the weight required for proof — it changes only its kind.

I record this as an observation about the word order.

The structure of liʿān is the same for both parties, and this symmetry is the most striking side of the verses:

The symmetry is all but exact, and the only difference lies in the word of the fifth oath: laʿna and ghaḍab.

لَعْنَة — root ل-ع-ن: to drive away, to expel. A laʿna is a being driven far from mercy — at the centre of the word stands distance.

غَضَب — root غ-ض-ب: anger. At the centre of the word stands a turning towards.

The classical commentaries have met the difference between the two words with several explanations, and I prefer none of them:

These are transmitted explanations; I impose no preference, and none is binding.

The sentence does not say "let him swear four times". It says "his testimony is four testimonies".

The oath, that is, is named with the word shahāda. What was required in the fourth verse was shuhadāʾ (witnesses); what is required here is shahādāt (testimonies).

The same root, in two different forms: the one a person (four separate human beings), the other an act (four separate declarations).

I record this as my own reading, and its ground is the form of the words: the text puts a man's own declaration in the place of four separate testimonies, and does so without changing the word. The oath is placed here not as an expression of feeling but as a procedural act.

What this regulation does is to rescue both parties from a deadlock.

The condition of the fourth verse — four witnesses — is in practice all but impossible for a husband. Had the condition been applied as it stands, two results would follow: either the claimant would receive eighty strokes for failing to prove his claim, or the claim could never be voiced at all.

The verse opens a third way: the claim is stated within a procedural frame, the other party answers within a procedural frame of the same weight, and the matter is closed by leaving the truth of the oaths to God.

Liʿān, then, is not a procedure of proof but a procedure of exit. Which of the two is telling the truth is not settled by a ruling.

I record this as my own reading; its ground is that the oaths of both parties are qualified by a conditional clause (in kāna) — that is, the text does not itself announce which of the two is telling the truth.

Over the juristic consequences — the state of the marriage, paternity, the possibility of coming together again, the ruling where one refuses to swear — there is broad disagreement among the schools. None of it is transmitted in this section and no ruling is issued.

Wa yadraʾu ʿanha'l-ʿadhāba an tashhada arbaʿa shahādātin bi'llāhi innahū la-mina'l-kādhibīn · Wa'l-khāmisata anna ghaḍaba'llāhi ʿalayhā in kāna mina'l-ṣādiqīn

"And it averts the punishment from her that she testify four testimonies sworn by God that he is of the liars. And the fifth, that the anger of God be upon her if he is of the truthful."

Root: د-ر-أ. Its concrete sense is to push, to thrust back, to repel. Daraʾa — to push a thing away. From the same root come dirʾ (in the sense of pushing) and tadāruʾ.

The choice of the verb is worth recording. The verse does not say "the woman too swears"; it says "it pushes the punishment away from her".

The work of the woman's oath, that is, is not to prove a claim — it is to thrust back something that is coming. The verb is chosen from the field of defence.

And this fixes the position the verse gives to the two parties:

The two verbs run in opposite directions, and both are physical. I record this as an observation about the language; the concrete senses of the two roots stand in the dictionary.

The verse does not say which punishment it means in saying "the punishment", and this is a point debated in classical jurisprudence. That debate is not entered in this section.

What can be recorded is a linguistic observation: the word is the same word as ʿadhābahumā in the second verse. The sūra names the punishment always with the same word.

The readings al-khāmisatu (nominative) in the seventh verse and al-khāmisata (accusative) in the ninth are recorded in the sources, and this difference between the two verses is taken up in the works on the variant readings.

That the difference does not change the meaning, and consists only in a choice of case within the construction of the sentence, is recorded in the classical sources. Since I cannot state with certainty which reader read which, I write no names.

The eighth and ninth verses repeat, word for word, the structure of the sixth and seventh. Only three things change: the subject, the word of the fifth oath (laʿna / ghaḍab), and the direction of the conditional clause.

This repetition is a wholeness rarely seen in the Qurʾān's verses of ruling, and by itself it says something: the procedural position of the two parties is one and the same.

I record this as an observation about the word order; the repetition itself is the wording of the text and can be compared.

Wa law lā faḍlu'llāhi ʿalaykum wa raḥmatuhū wa anna'llāha tawwābun ḥakīm

"And were it not for God's bounty towards you and His mercy, and that God is ever-relenting, wise…"

The most striking side of this verse is that the sentence is left unfinished.

Law lā — "were it not for" — is a particle of condition and calls for an answer: "were it not for it, such and such would be". But the answer is not spoken.

In Arabic this is called ḥadhf jawāb al-sharṭ, and the rule the lexicographers record is this: the omission of the answer conveys the answer's greatness and its unspeakableness. What is not said is heavier than what is said; the reader is asked to complete it himself.

And this pattern will be repeated four times in the sūra:

Four repetitions, balanced by two spoken and two unspoken answers. I record this as an observation about the word order; the places of the repetitions can be counted in the text.

This verse is the close of the second block (2-10) and serves as a seal.

The block had opened with heavy rulings. The close names the quality standing behind those rulings: faḍl (bounty), raḥma, tawwāb, ḥakīm.

تَوَّاب — root ت-و-ب, an intensive form: He who accepts turnings again and again. That in Arabic the word can be used both of the servant and of God is worth noting: the servant "turns", God "accepts the turning" — the same verb, in two directions.

حَكِيم — root ح-ك-م: to judge; and the concrete sense of the root is the bit set in an animal's mouth (ḥakama), that is, to rein in. The root was named at al-Ḥujurāt, 49/8.

And the name's place here is worth noting: at the end of a block of rulings there comes a name from the root of ruling. The sūra seals the measures it has laid down with the name of the one who laid them.

The logic of the block is this, and I record it as my own reading: the text forbids an act, then forbids the use of that act's name as a weapon, then opens a way out for the one case in which proof is impossible, and at the end names repentance and mercy.

The block, then, is not a series of penalties — it is a series of limitations. What is limited, in order: the act, the charge, and the ruling itself.

And what comes after the block will show why this logic is built so tightly: in the eleventh verse it will be told how an accusation spreads within a community.

Inna'lladhīna jāʾū bi'l-ifki ʿuṣbatun minkum, lā taḥsabūhu sharran lakum, bal huwa khayrun lakum, li-kulli'mriʾin minhum ma'ktasaba mina'l-ithm, wa'lladhī tawallā kibrahū minhum lahū ʿadhābun ʿaẓīm

"Those who brought the ifk are a band from among you. Do not think it an evil for you; rather it is a good for you. For each man among them is what he earned of the sin. And for the one among them who took upon himself the greater part of it there is a great punishment."

The verses from eleven to twenty point to an event that is transmitted in detail in the sources.

In this section the details of the reports about that event are not transmitted, no personal name is given, and no discussion is entered about who was accused.

The reason has two layers, and it needs to be written openly:

One. The verses themselves give no name. Alladhīna jāʾū — "those who brought". ʿUṣbatun minkum — "a band from among you". Alladhī tawallā kibrahū — "the one who took upon himself the greater part of it". The text is nameless from beginning to end. And the namelessness stands here not as a lack but as a choice: the verses close the matter in the twenty-sixth by saying ulāʾika mubarraʾūna mimmā yaqūlūn — "they are free of what is said".

Two. This commentary's general practice. The reports around an event diverge from one another at the level of detail; and, more importantly, this sūra itself takes as its subject what it is to speak about named persons. For a section to be written by doing the very thing the verse it treats forbids would be an inconsistency.

What will be treated is the frame the verses themselves give: how a report spreads, how language works, and how "thinking it a small thing" is diagnosed.

The root was analysed at al-Aḥqāf, 46/11 and 46/28, and also at al-ʿAnkabūt, 29/17 and 29/61; I rely on those treatments.

The note there was this: the concrete sense of the root is to turn a thing from its direction, to turn it inside out. As was recorded in al-Aḥqāf: "the word used for 'fabrication' means 'turned about'." In al-ʿAnkabūt the root's use in two directions was given in a table: takhluqūna ifkā (active — they fabricate) and fa-annā yuʾfakūn (passive — they are turned about). The sentence there was this: "what they fabricate becomes in the end the thing that turns them about."

And the choice of the word here uses exactly the picture this root carries.

The verse does not say kadhib (a lie). The word buhtān will come in the sixteenth verse, but not here. The word used here is ifk: the thing turned about.

The difference is this:

Ifk is not the fabricating of a thing out of nothing — it is the turning inside out of something that exists. That is the picture in the root: the bottom goes on top, the front to the back.

And this picture describes how slander works. A slander is generally not built from nothing; it is built by turning the direction of real material. An event happened, a person was seen somewhere, a word was said — and their direction is reversed.

I record this as my own reading; its ground is the concrete sense of the root, and that sense was recorded independently in al-Aḥqāf.

Root: ع-ص-ب. Its concrete sense is to bind tightly, to wrap.

The choice of the word is one of the verse's most important diagnoses.

The verse does not say qawm (a people), farīq (a party), nās (people) or ṭāʾifa (a group). It says ʿuṣba: those bound to one another.

The spreading of a report, that is, is described not as the independent movement of scattered individuals but as the movement of a bound band.

The idea of "binding" in the root of the word tells the social mechanics of slander, and I record this as my own reading: an accusation does not spread on its own. Among the people who carry it there is already a bond — a common circle, a common enmity, a common expectation. And the report flows along that bond.

Moreover the same root gives the word ʿaṣabiyya, and this word conveys the taking as one's measure not a man's truthfulness but whether he is on one's own side. Since the two words are from the same root, I record this as an observation about the root and draw no ruling from it.

This is the heaviest word in the verse, and it consists of two letters.

Those who brought the slander are not from outside but from within. And this is exactly the measure recorded in al-Ḥujurāt: there too the behaviours corrected were the behaviours of those who believe, and the address was turned inward upon them.

This word settles from the outset how the verses are to be read. To take up the section and use it to describe an outside group is to turn the direction of the address on its head.

Here is the most unlooked-for place in the verse's arrangement.

A slander that shakes a community is spoken of, and the text says this: "do not think it an evil for you; rather it is a good for you."

Three notes must be entered so that the sentence is not misunderstood:

One. The sentence does not declare the slander a good thing. At the end of the same verse it says ʿadhābun ʿaẓīm of those who spread it; in the nineteenth verse ʿadhābun alīm, in the twenty-third luʿinū fi'l-dunyā wa'l-ākhira. The ruling about the act is of the utmost severity.

Two. To whom does the pronoun lakum refer? The one addressed is not those who brought the slander — it is the whole of the community. By saying lā taḥsabūhu (a plural imperative) the verse opens the address to the community.

Three. What is the "good"? The verse does not say; but the following verses give an answer: the whole of the verses from twelve to twenty is a lesson in method drawn from this event. What is "good", then, is not the event itself — it is the measure that comes after the event.

I record this as my own reading, and its ground is the arrangement of the sūra: had the verses only meant to record an event, the verses from twelve to eighteen would not have been needed. The whole of those verses builds a method.

And one further observation: a similar sentence structure occurs at al-Baqara, 2/216 — wa ʿasā an takrahū shayʾan wa huwa khayrun lakum. The two verses do the same work: the ruling upon an event is separated from the outcome of that event.

This clause divides the responsibility within a group among individuals, and it is one of the places in the verse that must be read most carefully.

At the head of the eleventh verse a group had been described (ʿuṣbatun). And now the verse binds each person within that group separately to account: li-kulli'mriʾin minhum — "for every man among them".

The group, that is, does not dissolve responsibility. A thing's being done in company does not free a man of his own share.

ٱكْتَسَبَ — root ك-س-ب, Form VIII. The root was recorded in al-Shūrā. And the difference of form is a point the classical lexicographers dwell on:

I give this distinction as a tendency the lexicographers record; that it is not an absolute rule is also stated in the sources. But that the form in the verse is Form VIII is worth recording: the carrying of a report is placed as something a man does by his own effort.

The verse sets up a difference of degree within the group.

تَوَلَّىٰ — root و-ل-ي: to be near, to take upon oneself, to take a matter in hand. A walī is one who takes a matter in hand; a wālī is one who governs. Here the sense is: the one who takes upon himself the largest part of a business.

كِبْر — root ك-ب-ر: greatness. Two readings of the word are recorded in the sources: kibrahū and kubrahū. Both mean "the greatest share of it"; the difference is in the form. Since I cannot state with certainty which reader read which, I write no names.

And the correspondence of this difference of degree in the arrangement is this:

The verse thus does two things at once: it exempts no one from responsibility, but neither does it put everyone in the same place.

I record this as my own reading, and its ground is that the two clauses stand side by side. And I add this: this is the distinction most often lost when a report spreads. The one who first says a word, the one who carries it insistently, the one who repeats it once and the one who listens in silence do not stand in the same place — but none of them is entirely outside it either.

Law lā idh samiʿtumūhu ẓanna'l-muʾminūna wa'l-muʾminātu bi-anfusihim khayran wa qālū hādhā ifkun mubīn

"When you heard it, ought not the believing men and the believing women to have thought well of their own, and to have said: 'This is a plain slander'?"

In Arabic law lā is used in two distinct functions, and the two must be kept apart:

And the second function, used of a past act, conveys a reproach: you did not do what you ought to have done.

The particle occurs seven times in the sūra and divides in two:

Three reproaches and four conditions, interlocked. And the two series run in and out of each other: 10 (condition) → 12, 13 (reproach) → 14 (condition) → 16 (reproach) → 20, 21 (condition).

I record this as an observation about the word order; the places of the particle can be counted in the text. And what the structure does is this: beside every reproach a note of bounty is set. The text reproaches, but does not leave the reproach alone.

The note of time is worth remarking: idh samiʿtumūh — "the moment you heard it".

The verse binds what ought to have been done to a moment: the first instant the report reaches the ear.

And this is exactly where the variant reading at al-Ḥujurāt, 49/6, looks. There the two readings were given in a table: fa-tabayyanū (verify) and fa-tathabbatū (hold your ground). And the note there was this: tathabbut is the precondition of tabayyun — verifying takes time, and to gain time one must halt.

al-Nūr 24/12 says what is to be done in that moment of halting.

This is the expression in the verse that most needs dwelling on, and it carries a difficulty of translation.

Literally: "let the believing men and the believing women think good about their own selves".

And there are two readings of what bi-anfusihim means:

The two readings do not exclude one another, and both are transmitted in the classical commentaries. I impose no preference.

But what the two readings have in common is what gives the expression its force: the word anfus puts one believer's judgement about another in the same place as his judgement about himself.

And this pattern occurred exactly so at al-Ḥujurāt, 49/11: wa lā talmizū anfusakum — "do not find fault with yourselves". The note there was this, and it sits here exactly: that the object chosen is anfusakum means that the one found fault with is named not "the others" but "yourselves".

The two sūras use the same word for the same work:

The same word in three verses, with three different verbs. This is a fact of the arrangement and can be counted.

The most striking side of the verse is how it frames its command.

The verse does not say:

• "Verify the report you have heard"
• "Do not spread it"
• "Be silent"

It says: ought you not to have thought well and said, "this is a plain slander"?

What is asked, then, is not a passive silence but a judgement. There are two verbs: ẓanna (to think) and qālū (to say). The one within, the other without.

And the order is worth noting: first the inside, then the outside. The word comes as the uttering of an opinion already framed.

I record this as my own reading, and I ground it on the chain at al-Ḥujurāt, 49/12. There the chain was built thus: suspicion → prying → backbiting, three steps growing from inside outward. And what was recorded there was this: "The prohibition does not begin with backbiting. It begins with suspicion — that is, where no one can see, and where no harm has yet come to be."

al-Nūr 24/12 builds the positive counterpart of the same chain:

The two verses show two directions of the same mechanism. al-Ḥujurāt cuts the chain at its very beginning; al-Nūr builds a counter-chain in the same place.

I record this comparison as my own reading; the order of the steps in the two verses stands in the text and can be compared.

مُبِين — root ب-ي-ن, the active participle of Form IV: what is clear; and what makes clear. In Arabic the word can be understood both intransitively and transitively.

And this is the same root as āyātin bayyināt in the first verse. The sūra had presented its own verses as bayyināt (distinguishing); here the believers are asked to say of a report that it is mubīn (plain).

The text, that is, hands the addressee the very word of measure it uses of itself. I record this as an observation about the root.

And the definiteness of the sentence is worth noting. The verse does not say "perhaps it is not true", "let us not be sure", "let us wait". It asks for a sentence of judgement: hādhā ifkun mubīn.

At first sight this may look to stand in tension with the command of tabayyun (verify) at al-Ḥujurāt, 49/6. But the two commands look at different situations, and the difference must be recorded:

The two verses do not contradict one another; what fixes the difference between them is the condition of proof that will be stated in the thirteenth verse. If a report, by definition, is not held to exist in law unless it is proved by four witnesses, then there is no "verifying" to be done about it either.

I record this distinction as my own reading; its ground is the wording of 24/13.

One. The question of the default position.

What the verse builds is not a counsel of "thinking positively". What it builds is a default position: when a report arrives, the mind of the one who receives it is not a blank page — it starts from somewhere. The verse fixes that starting point.

And the practical consequence is this: to say "who knows?" in the face of a claim is in fact not a neutral position. It is to let the claim take room in the mind without carrying its burden of proof.

Two. That silence is not enough.

The verse does not command silence; it commands the saying of a sentence. This is a concrete distinction that comes out of the arrangement.

When a report is spreading, those who do not carry it but do not oppose it either do not slow the speed of its spread. What the verse asks for is the breaking of the chain — and a chain is broken not by the links staying still but by one of them changing direction.

Three. And a limit.

This verse does not mean that no claim is ever to be taken seriously. The sūra itself has built a procedure of proof (4, 6-9, 13). What the verse closes off is the claim that circulates outside the procedure of proof — the one that goes about not in court but in the mouth.

Law lā jāʾū ʿalayhi bi-arbaʿati shuhadāʾ, fa-idh lam yaʾtū bi'l-shuhadāʾi fa-ulāʾika ʿinda'llāhi humu'l-kādhibūn

"Ought they not to have brought four witnesses to it? Since they did not bring the witnesses, it is they who, in God's sight, are the liars."

The verse repeats the condition of the fourth verse exactly as it stood: arbaʿati shuhadāʾ.

And the place of the repetition is worth noting: the fourth verse had laid down a ruling; the thirteenth applies that ruling to an event. The sūra, that is, works the rule it has laid down upon the case it has itself recounted.

I record this as an observation about the word order; the identity of the expression in the two verses is the wording of the text.

Here is the qualification in the sentence that must be read most carefully: ʿinda'llāh — "in God's sight".

The verse identifies the failure to prove with lying. But it does so with the qualification ʿinda'llāh.

In the classical commentaries the function of this qualification has been dwelt on, and the explanations transmitted are these:

I impose no preference, and none is binding.

But what lies beneath all three explanations is the same, and I record it as my own reading: an unproved claim, even if it is true, has the standing of a lie from the moment it goes into circulation. For no one who hears it has anything in hand with which to verify it.

This is the sharpest statement of the matter of the burden of proof recorded at the fourth verse. The verse does not say "they lied" — it says "they are liars"; that is, it states a position. And what fixes the position is not the content of the claim but the absence of its proof.

The particle is worth noting: fa-idh — "since, given that".

The sentence draws a conclusion. And the manner of the drawing is logical: a condition had been laid down (four witnesses), the condition was not met, the ruling followed.

This is the working of the verb faraḍnāhā in the sūra's first verse. A measure had been notched; the measure is applied.

Wa law lā faḍlu'llāhi ʿalaykum wa raḥmatuhū fi'l-dunyā wa'l-ākhirati la-massakum fī mā afaḍtum fīhi ʿadhābun ʿaẓīm

"Were it not for God's bounty towards you and His mercy in this world and in the hereafter, a great punishment would have touched you for what you plunged into."

Root: ف-ي-ض. And the concrete sense of the root is one of the most striking word-pictures in this section.

Fāḍa — a vessel's overflowing, water's going over the brim and spreading. Fayḍ — an overflowing, abundance. Ifāḍa — to make overflow, to pour out.

And the Qurʾān uses this verb in the setting of the pilgrimage: fa-idhā afaḍtum min ʿArafāt (al-Baqara 2/198) — "when you pour out from ʿArafāt". What is described there is a great crowd's pouring from one place to another.

The use of the word here is a verb of water chosen for a word:

The verse, that is, describes the spreading of a report as an overflowing.

And this picture carries three things, which I record as my own reading:

1. What overflows has passed the vessel's bound. A word has gone outside the place it belonged to. 2. An overflow cannot be steered. Water that has spilled does not go by a road the spiller has fixed; it finds its own way. 3. An overflow is the act of a crowd. A drop does not overflow; a mass does.

And that the verb comes in Form IV rather than Form V is worth noting: afāḍa — transitive. The verse names its addressees not as the water that overflows but as those who made it overflow. The word did not spread of itself; it is not told as "it spread" — it is said, you made it overflow.

This agrees with the whole logic of the sūra: talaqqawnahū bi-alsinatikum in the fifteenth verse, yuḥibbūna an tashīʿa in the nineteenth. In all three verses the spreading is set down not as a passive occurrence but as an act.

There is a grammatical point in the verse, debated in the classical commentaries: what does the phrase fi'l-dunyā wa'l-ākhira attach to?

Both readings are transmitted in the classical sources; I impose no preference, and neither is binding.

But the first reading's agreement with the position in the sentence may be recorded: the phrase comes immediately after faḍl and raḥma. I give this as an observation about the word order, not as a preference.

Root: م-س-س. Its concrete sense is to touch, to make contact — the lightest contact.

And the choice of the verb is not a softening but a sharpening. The verse does not say la-nazala ʿalaykum (it would have come down upon you) or la-akhadhakum (it would have seized you). It says "it would have touched you" — and what touches is described as ʿaẓīm.

Only the grazing of the thing described as "great", then, is at issue.

I record this as an observation about the style. The same verb will return in the thirty-fifth verse: wa law lam tamsashu nār — "though no fire had touched it". In both places the same verb, in the same sense of the lightest contact.

Idh talaqqawnahū bi-alsinatikum wa taqūlūna bi-afwāhikum mā laysa lakum bihī ʿilmun wa taḥsabūnahū hayyinan wa huwa ʿinda'llāhi ʿaẓīm

"When you were taking it up on your tongues, and saying with your mouths what you had no knowledge of, and thinking it a small thing — while in God's sight it is great."

This verse describes the mechanics by which a report spreads in four verbs, and each of the four does a distinct work in the arrangement:

The four steps build a chain, and the last link of the chain is the cause of the others. I take this up at the end of the verse.

Root: ل-ق-ي. Its concrete sense is to meet, to encounter, to come upon a thing. Liqāʾ — a meeting. Talaqqī (Form V) — to receive, to take up and make one's own, to take over.

The Qurʾān uses the same form elsewhere: fa-talaqqā Ādamu min rabbihī kalimāt (al-Baqara 2/37) — "Adam received words from his Lord". There too the verb conveys the receiving of a word, its being met and made one's own.

And the most striking side of the usage here is the object of the preposition: bi-alsinatikum — "you were taking it up with your tongues".

This is an unlooked-for combination in the language. A thing is normally received with the ear. The verse binds the receiving to the tongue.

And this combination describes how a report spreads. Thus:

A report is not received by hearing and left there. The act of receiving is completed only when it is repeated. A man has not really "received" a report — he has received it the moment he says it to another. Saying is part of receiving.

The verse, that is, removes the distance between hearing and saying and joins the two in a single verb: talaqqī bi'l-lisān.

I record this as my own reading, and its ground is the expression itself: that the instrument of the verb "to receive" is given as the tongue stands in the wording of the text, and it is out of the ordinary.

The explanations given for this expression in the classical commentaries look in a similar direction: one man's taking the report over from another, its going from mouth to mouth, everyone taking it from the one before and giving it to the one after. I transmit these explanations as views that are transmitted.

A variant reading is transmitted in the sources: it is recorded that the word is also read تَلِقُونَهُ (talikūnahū). This reading comes from the root و-ل-ق, and the sense the lexicographers record for it is to say a thing quickly and without stopping, to keep telling it.

Both readings are transmitted in the sources; since I cannot state with certainty which reader read which, I write no names.

And the two readings complete each other: the one describes how the report comes (by being taken over), the other how it goes (by being said without ceasing). I record this complementarity as my own reading.

The verse names two organs, and both are worth noting: alsina (tongues) and afwāh (mouths).

The expression "you say with your mouths" looks at first like a needless repetition. Saying is done with the mouth in any case.

But the Qurʾān uses this expression for a definite work, and this was recorded earlier in this commentary. At al-Aḥzāb, 33/4, the same pattern occurs: dhālikum qawlukum bi-afwāhikum — "that is your word with your mouths". The context there was an ascription that had no reality.

The work of the pattern, therefore, is this, and it is explained in this direction in the classical commentaries too: the qualification bi'l-fam conveys that the word stays only in the mouth, that there is no reality behind it. It comes out of the mouth but does not rest on any knowledge.

And the verse's next clause says this outright: mā laysa lakum bihī ʿilm — "what you have no knowledge of".

The work of the two organs in the arrangement:

The verse, that is, pictures man as a pipe through which the report passes: it goes in at one end, comes out at the other, and nothing is done to it in between.

The name of what is not done is given in the verse: ʿilm.

I record this as my own reading; its ground is that the two organs are named separately and that the note of the absence of knowledge is set between them. The order stands in the wording of the text.

The expression is a pattern repeated in the Qurʾān and occurred earlier in this commentary: wa lā taqfu mā laysa laka bihī ʿilm (al-Isrāʾ 17/36) — "do not pursue what you have no knowledge of".

And the pattern here uses a language of possession: laysa lakum bihī ʿilm — "you have no knowledge of it".

Knowledge, that is, is placed as something held in the hand. For a man to be able to speak on a matter he must have something in hand. And in slander there is nothing in the hand at all — the fourth verse had already bound this to a condition: four witnesses.

And the measure recorded at al-Ḥujurāt, 49/12, works here exactly: the note there was this — "The Qurʾān does not censure supposition in itself. What it censures is the putting of an unproved assumption in the place of certain knowledge."

al-Nūr 24/15 shows how that "putting in the place of certain knowledge" comes about: as a word passes from tongue to tongue it gains no knowledge — but it gains the look of knowledge. For what is said by many people appears, by the very frequency of its being said, to have been verified.

I record this as my own reading.

This is one of the strongest sentences in the sūra and deserves separate treatment.

هَيِّن — root ه-و-ن. Its concrete sense is lightness, ease, smallness of weight. Hawn — softness, not being heavy. Hīn / hāʾin — counted worthless. Ihāna — to make light of a man, to devalue him. The Turkish hain comes from this root, though its sense has drifted.

The word is used in the Qurʾān of God's power as well: huwa ʿalayya hayyin (Maryam 19/9) — "that is easy for Me". The word, then, is not negative in itself; it conveys a measure of weight.

And what the verse builds is a contrast of measures:

The relation of the root ع-ظ-م with ʿaẓm (bone) was recorded in al-Ḥāqqa. The picture that appears when the two roots are set side by side is this: lightness on the one hand, a skeleton on the other.

And what the sentence does is to make a diagnosis.

The verse does not say that those who spread the slander were of ill intent. What it says is finer and far more common: they held the measure wrongly.

The importance of this diagnosis lies here, and I record it as my own reading:

A man doing something he knows to be wrong feels he must defend himself. But doing something he supposes to be of no importance he offers no resistance at all — for in that case there is not even a decision.

And this is exactly what makes the mechanics of a report's spread possible. The crowd that spread the slander was not, taken one by one, doing a great evil: each of them was only saying a sentence. And each one's own share was, in his own eyes, hayyin.

But the verse had said in the eleventh that the shares are divided: li-kulli'mriʾin minhum ma'ktasaba mina'l-ithm. The text, that is, establishes this: the share may be small; but to have a share is not to be exempt from a share.

And it is said who holds the measure: ʿinda'llāh. The qualification is the same as in the thirteenth verse: there too the ruling was given with the qualification ʿinda'llāh.

The contrast of measures the verse builds has one further side, and I record it as my own reading.

There is a reason for supposing it hayyin: in a report that spreads, no one sees the whole of it.

The one who first says it does not see where it will go. The link in the middle does not know through how many people it has already passed. The last link does not know the first source. Every link sees only its own share, and its own share really is small.

But the harm is the sum of the shares, and no one sees the sum.

And the verse's qualification ʿinda'llāh sits exactly at this point: the party that sees the sum is the party that holds the measure.

I record this reading as my own reading; its ground is that the verse has set the qualification ʿinda'llāh between hayyin and ʿaẓīm.

These two sūras are the texts in the Qurʾān that treat the ethics of report and of speech in the greatest detail, and they complete each other. Relying on the discussions of al-Ḥujurāt 49/6 and 49/12, I record the comparison:

What the two sūras build together is this, and I record it as my own reading:

al-Ḥujurāt regulates what is to be done with a report; al-Nūr fixes what an accusation is.

In al-Ḥujurāt the report may turn out to be true, and if it does one acts. In al-Nūr the matter is different: a claim about the privacy of a person is not even a thing whose truth is to be looked into, so long as it is not proved by four witnesses.

The shared backbone of the two sūras, however, is the same: the word is stopped before it goes into circulation. al-Ḥujurāt does this by waiting, al-Nūr by closing the door.

One. There is no such thing as the innocence of the middleman.

The verse's four verbs — to receive, to say, not to know, to count light — describe not a source but a carrier. And the text takes the carrier as its addressee.

This is the distinction most often lost in today's terms. There is a difference of degree between the man who first puts a claim about and the man who passes it on — the eleventh verse already recorded this (alladhī tawallā kibrahū). But a difference of degree is not an exemption.

Two. The diagnosing of "thinking it a small thing".

The most useful side of the verse is that it diagnoses not a behaviour but an error of appraisal. And the symptom of that error can be recognised: not having stopped to think at all before saying a word.

A man hesitates when he says something he thinks weighty. If there is no hesitation, that word is for that man hayyin. The verse gives a measure, and the measure can be read from within.

Three. And a limit: this verse does not forbid speaking.

What the verse forbids is the saying of what one has no knowledge of; not the saying of words. In the twelfth verse the saying of a sentence was commanded. The sūra asks not for silence but for the saying of what one has in hand.

Wa law lā idh samiʿtumūhu qultum mā yakūnu lanā an natakallama bi-hādhā subḥānaka hādhā buhtānun ʿaẓīm

"When you heard it, ought you not to have said: 'It is not for us to speak of this. Glory be to You — this is a great slander'?"

The verse opens with the same note of time as the twelfth: idh samiʿtumūh — "when you heard it".

The sūra, that is, returns twice to the same moment and twice asks for a different sentence:

The two sentences look in two different directions, and this is the verse's most important point of arrangement.

The sentence in the twelfth verse rejects the truth of the report. The sentence in the sixteenth, without entering the question of truth at all, rejects the speaker's standing to speak.

I record this as my own reading, and its ground is the content of the two sentences.

In Arabic the pattern conveys a lack of standing. Literally: "it is not for us", "this does not belong to us". The Qurʾān uses the same pattern elsewhere: mā yakūnu lī an aqūla mā laysa lī bi-ḥaqq (al-Māʾida 5/116) — "it is not for me to say what is not my right".

And the work of the pattern here is worth recording.

The verse does not say "this is not true" or "we do not know this". It says "this is not our business".

Two distinct positions, then, can be taken in the face of a report, and the sūra asks for both:

Recording the second position is important, because it is wider than the first.

We cannot always know whether a report is true. But we can always know whether a report belongs to us. The second question requires no knowledge.

And this is the same as the logic of the prohibition of tajassus (prying) at al-Ḥujurāt, 49/12. The measure recorded there was this: "The difference lies in whose the knowledge is: where a lost brother is is no one's privacy; but a man's hidden condition belongs to him."

al-Nūr 24/16 builds the same measure from the side of speech: if a piece of knowledge is not mine, then speaking about it is not mine either.

I record this connection as my own reading.

Root: ك-ل-م. And the root has an interesting side: from the same root come both kalima (a word) and kalm (a wound).

Kalm — a wound opened upon a body, a cut. The connection the lexicographers draw is this: a word leaves a mark upon the one who hears it — just as a wound does.

I transmit this connection as an explanation the lexicographers record, not as a claim of derivation. That the two words are from the same root stands in the dictionary.

And the context here shows the reasonable side of the connection: the verse is speaking of what a word does in a human being.

Here is the most striking place in the verse, and it has been dwelt on in the classical commentaries.

Subḥānaka — "glory be to You". Root س-ب-ح: to swim, to go flowing; and from there tasbīḥ: to keep a thing clear of what does not befit it. The root has been taken up at many places in this commentary; it will return in the forty-first verse of the sūra.

And the difficulty is this: why is a sentence of tanzīh said in the face of a report of slander?

The explanations transmitted in the classical commentaries are these:

I impose no preference, and none is binding.

But an observation about the word order may be recorded, and I give it as my own reading: the sentence the verse asks for has three parts, and each of the three looks in a different direction.

In a single sentence, then, the verse asks for three distinct judgements: of one's standing, of the measure, and of the report.

Root: ب-ه-ت. This root was analysed in al-Mumtaḥana; I rely on that treatment. The note there:

Root ب-ه-ت: to strike dumb, to leave stunned. Buhtān is not an ordinary lie: it is a slander so unlooked for that it stuns the one it is aimed at, one before which he does not know what to say. The word itself describes the effect of the slander.

And the usage at al-Baqara 2/258 was named there: fa-buhita'lladhī kafar — "the one who disbelieved was struck dumb".

Now the three words in the sūra can be set side by side, and it is seen that each does a distinct work:

Three words in three different places, and none of the three is a needless repetition. The sūra names the same event from three different sides: what it is, what it does, and how it spreads.

I record this table as my own reading; the places and roots of the words stand in the text.

And one further observation: the sense of buhtān as "leaving without an answer" describes the condition of the one slandered.

The heaviest side of a slander is that it cannot be answered. That a thing not done was not done cannot be proved. There is no evidence in hand, because a thing that did not happen leaves no trace either. The one slandered is, for precisely this reason, buhita — struck dumb.

I record this as my own reading; its ground is the sense in the root of the word, and that sense was recorded independently in al-Mumtaḥana.

And why the fourth verse puts the burden of proof upon the claimant is understood exactly here: because the other side cannot prove.

Yaʿiẓukumu'llāhu an taʿūdū li-mithlihī abadan in kuntum muʾminīn · Wa yubayyinu'llāhu lakumu'l-āyāt, wa'llāhu ʿalīmun ḥakīm

"God admonishes you never to return to the like of it, if you are believers. And God makes the verses clear to you. God is knowing, wise."

Root: و-ع-ظ. The sense the lexicographers record: to remind a man in a way that softens his heart and holds him back from a deed. Mawʿiẓa — an admonition.

In the definition of the word there are two elements at once: reminding and holding back. Waʿẓ, that is, is not the giving of knowledge; it is the bringing back of something known, in a way that changes behaviour.

And this is on the same line as tadhakkarūn in the first verse. In both places the sūra speaks not of giving new knowledge but of calling back something known.

Two words in the expression are worth noting.

The first: li-mithlihī — "to the like of it". The verse does not say "do not return to this"; it says "to anything like this".

This carries the scope of the ruling out of the event and over to the kind. What is forbidden is not a particular report but everything of that kind.

I record this as an observation about the word order and find it in agreement with the measure recorded in al-Ḥujurāt: there too it was stated that the verses regulate not single events but kinds of event.

The second: abadā — "never". And this word occurred in the fourth verse too: wa lā taqbalū lahum shahādatan abadā.

The two usages answer each other, and I record this as my own reading: the one closes a door, the other a road. And between the two stands the note of repentance in the fifth verse.

This conditional clause occurs for the second time in the sūra. The first was in the second verse: in kuntum tuʾminūna bi'llāhi wa'l-yawmi'l-ākhir.

In both places the condition is set against a feeling or a habit. In the one, pity; in the other, the wish to carry a report.

I record this as an observation about the word order.

The root ب-ي-ن returns here after the first verse, and it will return again and again in the sūra: 34, 46, 58, 59, 61.

The root's distribution through the sūra builds a weave:

The root occurs ten times in the sūra, and at each occurrence a work of separating is done. This shows how the sūra names its own work.

I record this as my own reading; the places of the words can be counted in the text.

Inna'lladhīna yuḥibbūna an tashīʿa'l-fāḥishatu fi'lladhīna āmanū lahum ʿadhābun alīmun fi'l-dunyā wa'l-ākhira, wa'llāhu yaʿlamu wa antum lā taʿlamūn

"For those who love that indecency should spread among those who believe there is a painful punishment in this world and in the hereafter. God knows, and you do not know."

This verse draws one of the finest distinctions in the sūra — and perhaps in the Qurʾān — and it will be taken up here in detail.

The verse does not censure an act. It censures a wish. And what is wished is not an act either, but the spreading of an act.

Let us open out the structure of the sentence:

What is censured is the third layer.

The verse, that is, speaks not of the one who commits the act but of the one who wants it to be heard that the act was committed. And this man may not have committed the act himself — indeed, in the picture the verse builds, it is likely that he did not.

Why this distinction matters so much I record in three items, and this is my own reading:

One. The verse names a type of harm. The harm of an act is confined to the one who commits it. The harm of an act's spreading, on the other hand, spreads over the whole community — for what spreads is no longer the act but an image.

Two. The verse points to an evil that goes about in a moral dress. The man who spreads an indecency generally spreads it by condemning it. The sentence "have you heard what happened" is, in form, a sentence of condemnation. And precisely for that reason the spreader thinks himself set against the act.

Three. And this is why the verse's word is yuḥibbūn: "they love". The verb conveys not an act but a leaning.

Root: ش-ي-ع. This root was analysed at al-Qamar, 54/51, at al-Ṣāffāt, 37/83, and at al-Qaṣaṣ, 28/4; I rely on those treatments.

The note there was this: the concrete sense of the root is to spread, to scatter, to go in every direction. Shāʿa'l-khabar — the report spread. Ashāʿa — he spread it. The Turkish expression şuyu bulmak ("to become widely known") is from the same root.

And the root's second sense: a body of people who follow someone, a party of partisans (shīʿa, plural shiyaʿ and ashyāʿ).

The sentence recorded in the al-Ṣāffāt section works exactly here: "at the centre of the word there are two things: to follow a person, and to spread that person's cause."

And the connection between the two senses of the root explains the mechanics of the verse, and I record this as my own reading:

The spreading of a thing and the forming of a group are named from the same root. For when a report spreads, a bond forms of itself among those who carry it — a shared piece of knowledge, a shared stance, a shared subject of talk.

And this joins with the word ʿuṣba in the eleventh verse: there those who brought the slander were named "a band bound to one another". In the nineteenth verse it is seen how that bond is formed: by spreading.

The two words name the same fact from two sides. I record this table as my own reading; the roots of the words stand in the dictionary.

Root: ف-ح-ش. The sense the lexicographers record: a thing's being so ugly as to exceed its measure, its excess. Fāḥish — excessive, beyond measure. Fuḥsh — ugliness.

At the centre of the word stands "excess", and the word is not used only of an act. In Arabic fāḥish is used of a price as well: a price that exceeds the measure.

And the word will occur twice more in the sūra: al-faḥshāʾ in the twenty-first verse (as what Satan commands), and the root in the thirty-third, in the domain named together with bighāʾ.

The usage in the verse is worth recording: the verse does not say al-zinā, it says al-fāḥisha. What is named, that is, is not a particular act but a kind. This widens the scope of the prohibition: what is condemned in its spreading is not one kind of act alone.

And that the word al-fāḥisha comes definite draws a limit: the verse takes as its subject not "the spreading of every kind of report" but the spreading of what is of this kind.

The preposition is worth noting: fī — "within". Not bayna (between) or ʿalā (upon).

The spreading, that is, is described as something happening within a community. And this looks in the same direction as the qualification minkum — "from among you" — in the eleventh verse.

From beginning to end the sūra keeps the matter inside a community. Not as an attack coming from outside but as a mechanism working within.

Root: ح-ب-ب. Its concrete sense is a seed, a grain (ḥabb). The connection the lexicographers draw: love is what settles in the heart like a seed and strikes root.

And the use of the verb in this verse is the place that must be read most carefully.

The verse does not censure an act: it censures a leaning. And a leaning lies further within than an act — it is seen by no one, it cannot be measured, it cannot be proved.

This is the same logic of "a circle narrowing from outside inward" recorded in al-Ḥujurāt. It was recorded there that the sūra puts the least visible thing last. al-Nūr does the same: it begins from an act that could be proved by four witnesses and comes down to a leaning that could never be proved at all.

And precisely for this reason the verse lays down no worldly sanction. The end of the sentence is this: lahum ʿadhābun alīmun fi'l-dunyā wa'l-ākhira — in this world and in the hereafter. This speaks not in the language of the court but in the language of the reckoning.

I record this as my own reading; its ground is that the verse gives neither a procedure of proof nor a measure of penalty. In the fourth verse there were both; here there is neither.

The verse-ending is like the seal of the whole block, and it is extremely short: "God knows, and you do not know."

And the arrangement of the sentence is worth noting: the two parties are set side by side, the same verb is used (yaʿlam / taʿlamūn), the one affirmative, the other negative.

This is the summary of the whole section. In the fifteenth verse it had said mā laysa lakum bihī ʿilm — "what you have no knowledge of". The nineteenth repeats the same word in its last clause: wa antum lā taʿlamūn.

The sūra, that is, first identifies the ignorance as confined to one event, then records it as a general position.

I record this as an observation about the word order.

One. Spreading in the form of condemnation.

This is the most familiar thing the verse points to, and it can be said in language that does not wound: the commonest way of telling an indecency is to tell it by condemning it.

And this form clears the spreader in his own eyes. For the grammatical structure of the sentence he says is a sentence of condemnation. But where the verse looks is not the structure of the sentence but its outcome: has a thing spread, or has it not?

The verse does not declare the man who does this to be evil. The verb it uses is yuḥibbūn — it describes a leaning. And leanings go on working so long as the man himself does not notice them.

Two. Where the distinction is placed.

A thing's happening and that thing's being heard of are separate events, and their outcomes differ.

An act has been committed. If three people know of it, the harm is at one place. If three hundred know of it, the harm is at another — and the increase of harm is the product not of the one who committed the act but of those who spread it.

The verse names this increase and binds a separate ruling to it.

Three. And the link with the sūra as a whole.

This verse explains why the condition of four witnesses in the fourth verse is so heavy. The condition is so heavy that bringing an act before the community is all but impossible. And the nineteenth verse says the reason: what is asked to be brought is most often in the service not of justice but of the spreading.

Four. A limit: this verse is not a command to cover things up.

To read the verse as "let no evil be spoken of" runs against the sūra itself. The sūra has built a procedure of proof; the existence of a procedure presupposes that matters can be brought somewhere. And as was recorded at al-Ḥujurāt, 49/9, as the counterpart of the ninth verse, the Qurʾān commands that action be taken against wrongdoing.

What the verse closes is not the road of justice — it is the road of gossip. The difference between the two stands in the verse's own word: tashīʿa — to spread. A matter taken to an authority does not spread; a matter recounted does.

I record this distinction as my own reading.

Wa law lā faḍlu'llāhi ʿalaykum wa raḥmatuhū wa anna'llāha raʾūfun raḥīm

"And were it not for God's bounty towards you and His mercy, and that God is most kind, most merciful…"

The pattern of the tenth verse is repeated here, and again its answer is not spoken.

But there is a difference, and it is worth noting: in the tenth verse the close was tawwābun ḥakīm; here it is raʾūfun raḥīm.

In the second verse raʾfa had been forbidden — on the condition that it should not stop the ruling. And now the same root returns as a name of God: raʾūf.

This is one of the finest rings in the sūra's arrangement, and I record it as my own reading:

The sūra thus separates this: a man's raʾfa cannot alter a measure that has been set; but the raʾfa of the one who sets the measure stands behind the measure itself.

The definition recorded at al-Ḥashr, 59/10, shows its use here: raʾfa contains the refusal to give pain at all, the refusal to hurt. And this is exactly the work done between the eleventh and the twentieth verses — a community has made a heavy mistake, and in correcting it the text has entered a note of bounty and mercy four times in ten verses.

I record this as an observation of count: 10, 14, 20, 21.

Yā ayyuhā'lladhīna āmanū lā tattabiʿū khuṭuwāti'l-shayṭān, wa man yattabiʿ khuṭuwāti'l-shayṭāni fa-innahū yaʾmuru bi'l-faḥshāʾi wa'l-munkar, wa law lā faḍlu'llāhi ʿalaykum wa raḥmatuhū mā zakā minkum min aḥadin abadan wa lākinna'llāha yuzakkī man yashāʾ, wa'llāhu samīʿun ʿalīm

"O you who believe, do not follow the footsteps of Satan. Whoever follows the footsteps of Satan — he commands indecency and what is wrong. Were it not for God's bounty towards you and His mercy, not one of you would ever have been purified; but God purifies whom He wills. God is hearing, knowing."

Root: خ-ط-و. Its concrete sense is to take a step. Khuṭwa — a step, the distance between two feet.

That the word comes in the plural, and that it is not a figure, is worth noting. The verse does not say "do not follow Satan's road" — it says "his footsteps".

And the difference between the two words describes a mechanism, which I record as my own reading:

A road is seen from the start; one decides whether to enter it or not. A step, on the other hand, is not a road by itself. As one steps, where one is going is not visible; only the place of the next step is.

And what is told between the eleventh and the twentieth verses is exactly a process that works step by step: a report is heard (12), repeated (15), thought a small thing (15), spread (19). At no stage does anyone say "I am now doing a great evil".

This is why the verse chooses the word khuṭuwāt, and the connection can be read from the arrangement: the verse comes immediately after the diagnosis of hayyinan (thinking it a small thing) in the fifteenth.

The form is worth noting too: tattabiʿū — Form VIII, ittibāʿ. This form conveys here a going after, a following of tracks. The man is not looking ahead; he is looking at someone's tracks.

And the expression is a pattern repeated in the Qurʾān: the same expression occurs at al-Baqara, 2/168 and 2/208, and also at al-Anʿām 6/142. The same structure in all four places: lā tattabiʿū khuṭuwāti'l-shayṭān.

فَحْشَاء — root ف-ح-ش: analysed at the nineteenth verse. An ugliness that exceeds the measure.

مُنكَر — root ن-ك-ر: not to be recognised, to be found strange. The root was analysed at Luqmān, 31/19 (ankara'l-aṣwāt — the most jarring of voices) and occurred at 31/17 as wa'nha ʿani'l-munkar; I rely on those treatments.

The coming of the two words together sets up a scale:

The one, that is, looks at quantity, the other at recognition. The opposite of munkar is maʿrūf: what is known, recognised, accepted by everyone as right.

زَكَىٰ — root ز-ك-و. This root has been taken up at many places in this commentary — al-Aʿlā, al-Shams, al-Layl, ʿAbasa. Its concrete sense: to increase, to grow, to be fruitful; and to be cleansed.

The lexicographers draw the connection between the two senses of the root thus: a thing grows only when it has been cleared of foreign elements. A crop yields nothing until it is cleared of weeds.

And the word will return again and again in the sūra:

The same root in four places, in three different tasks. And in all three the same thing is said: to hold back from a thing is the condition of increase.

I record this as my own reading; the places of the words can be counted in the text and their repetitions seen.

And the qualification in the verse's arrangement is important: mā zakā minkum min aḥadin abadā — "not one of you would ever have been purified". Purification, that is, is not set down as an outcome man reaches on his own. The sentence goes on with an exception: wa lākinna'llāha yuzakkī man yashāʾ.

This enters a note upon the whole of the section. The verses from eleven to twenty-one were correcting a community's fault. And the section closes by saying that those who were corrected did not achieve it themselves.

This note also settles how the sūra is to be read: to take up the text and use it to sort the purified from the unpurified runs against the verse's last sentence.

Wa lā yaʾtali ulu'l-faḍli minkum wa'l-saʿati an yuʾtū uli'l-qurbā wa'l-masākīna wa'l-muhājirīna fī sabīli'llāh, wa'l-yaʿfū wa'l-yaṣfaḥū, a lā tuḥibbūna an yaghfira'llāhu lakum, wa'llāhu ghafūrun raḥīm

"And let not those of you who have bounty and means swear not to give to their kinsfolk, to the needy, and to those who emigrate in the way of God; let them pardon and let them overlook. Do you not wish that God should forgive you? God is forgiving, merciful."

The verse's place in the sūra is as important as the verse itself.

The verses from eleven to twenty-one took up a slander, described those who brought it in the heaviest terms, announced a penalty and built a procedure. And the twenty-second verse comes directly after that section and says go on giving.

The one addressed by the verse is the party harmed by the slander. This is understood from the subject of the sentence: ulu'l-faḍli minkum wa'l-saʿa — "those of you who have bounty and means". That is, the giving party, the one who has the means in hand.

And this arrangement is one of the most unlooked-for moves in the sūra:

The text, that is, sets two rulings one after the other: the act is grave, and let the one harmed by the act pardon.

The logic of this is given in the verse's last sentence: a lā tuḥibbūna an yaghfira'llāhu lakum — "do you not wish that God should forgive you?"

This is the most disputed word in the verse in point of language, and the root's two distinct senses lead to two distinct readings.

Root: أ-ل-و. And the two clusters of sense the lexicographers record are these:

And the verb yaʾtali in the verse (Form VIII) is understood in two ways, according to which of the two senses it carries:

This disagreement is recorded openly in the classical commentaries and both readings are transmitted. I impose no preference, and neither is binding.

Moreover it is transmitted in the sources that the word is read in a different form as well — in Form V (yataʾalla), again with the sense of swearing. Since I cannot state with certainty which reader read which, I write no names.

What the two readings have in common may be recorded: both forbid the cutting off of giving. Where they part is over whether the cutting off is done by an oath or in silence.

And an observation: the first reading gives the cutting off a form. An oath is an attempt to make a decision irreversible. The first reading, that is, describes the tendency of a grievance to make itself permanent.

I record this as my own reading.

A note on the root: the derivative آلَاء (ālāʾ — favours), formed of the same letters, was treated at al-Najm, 53/55, and in the al-Raḥmān section. The lexicographers are not agreed as to whether these derivatives come from a single root; I draw no connection from it here.

The verse names its addressee with two attributes, and both are worth noting:

فَضْل — root ف-ض-ل: to increase, to be over and above; and superiority. The concrete sense of the word is "the part left over"; English covers the two ends of this root with two unconnected words — surplus and excellence.

سَعَة — root و-س-ع: breadth. Wāsiʿ — wide; saʿa — breadth of means, material ease.

The two words return again a few verses later in the sūra:

And this distribution says something, which I record as my own reading: the same word is used in one verse of what God gives to His servant, and in the next of a servant's giving to a servant. The faḍlu'llāhi ʿalaykum of the twenty-first verse and the ulu'l-faḍli minkum of the twenty-second stand back to back.

This transition in the arrangement builds in advance the logic of the verse's last sentence: you resist over giving what you were given; and you go on expecting to receive.

The verse uses two distinct verbs of pardon, and the difference between them is recorded by the classical lexicographers.

عَفَا — root ع-ف-و. The root was analysed in al-Mujādala and at al-Shūrā, 42/40; I rely on those treatments. The note there: the concrete sense of the root is to wipe out, to erase the trace — as the wind erases footprints on sand. And its difference from ghufrān: what is covered is still there but not seen; what is erased is no longer there.

صَفَحَ — root ص-ف-ح. The root was analysed at al-Zukhruf, 43/5 and 43/89; I rely on those treatments. The note there: the concrete sense of the root is the name of a surface — ṣafḥa: the side of a thing, its flat face.

And the concrete senses of the two words describe two distinct acts:

The classical lexicographers count ṣafḥ as going further than ʿafw, and the ground stands in the picture of the root: to pardon is to drop a claim; ṣafḥ is, after that, not to turn one's face away, to bring the relation back to what it was.

I transmit this distinction as an explanation the lexicographers record; I do not claim that it is an absolute rule.

And that the verse asks for both verbs together is worth noting. A man may drop a claim of his and still turn his face from the person. The verse asks for both at once.

And one further detail in the arrangement: the command to pardon comes after the command to go on giving. The order, that is, is this:

What belongs to the deed first, what belongs to the heart after. The verse asks first that the hand should not stop.

I record this as my own reading, and its ground is the order of the commands. And I add this: had the order been reversed — first pardon, then go on giving — pardon would have been the condition of the act of giving. In this order, giving is placed before pardon.

This is the strongest sentence in the verse, and it is framed as a question.

And the verb draws attention: tuḥibbūn — "do you not wish?". The same verb occurred three verses earlier (19) in a context of censure: yuḥibbūna an tashīʿa'l-fāḥisha — "they love that it spread".

The same verb, three verses apart, with two opposite objects. This is a repetition that can be verified from the text, and I record it as an observation about the word order.

And the logic the question builds rests not upon reciprocity but upon consistency.

The verse does not say "pardon so that you may be pardoned"; it sets up no exchange. It says: you wish to be forgiven. Then you know what forgiving is.

I record this as my own reading; its ground is the mood of the sentence: not a conditional but a question.

One. A grievance making itself permanent.

The verb in the verse's first reading — to swear — describes a behaviour: the turning of a grievance into a decision.

And the logic of this can be understood. A grievance grows less of itself with time. A decision does not. An oath is a grievance's attempt to protect itself against time.

The verse stands against this, and it does something striking: it does not find the grievance unjust. The verses from eleven to twenty-one condemned in the heaviest terms the act that caused the grievance. The verse asks that a just grievance should not turn into a severance.

Two. On what plane the pardon is asked for.

The verse does not lift the penalty. The sanction of the fourth verse stands where it is; the ruling that will come in the twenty-third stands where it is. What is lifted is the requital that lies in a man's own hand.

This distinction matters: pardon does not take the place of justice. It stands beside justice, in a place of its own.

Three. And a note: this verse is not an instrument of pressure.

To use the verse in order to tell another man "you must pardon" is to change the verse's addressee. The verse's address is directed at the one who is to pardon — not at the one who expects something from him.

And this agrees with the sūra's general measure: as was recorded in al-Ḥujurāt, the text is an instrument of diagnosis, and the diagnosis is applied first to oneself.

Inna'lladhīna yarmūna'l-muḥṣanāti'l-ghāfilāti'l-muʾmināti luʿinū fi'l-dunyā wa'l-ākhirati wa lahum ʿadhābun ʿaẓīm

"Those who cast a charge at chaste, unsuspecting, believing women are cursed in this world and in the hereafter; and for them there is a great punishment."

The verse names the party slandered with three attributes, and all three are worth recording.

The second attribute is one of the most affecting words in the sūra and deserves separate treatment.

Root: غ-ف-ل. Its concrete sense is to be unaware of a thing, not to hold it in mind, to have it escape one's notice. Ghafla — inattention; ghufl — bearing no mark upon it, unbranded (used of a piece of land or of a camel).

The second sense of the root is interesting: ghufl is what bears no trace or mark upon it. An unbranded camel, a stretch of land with no track across it. At the centre of the word, then, there is also the idea of "bearing no mark".

And in the verse's context this builds a picture: the party slandered is the party upon whom there is no mark at all. And slander is an attempt to set a mark where there is none.

I record this as my own reading; the second sense of the root is a usage the lexicographers record.

And the attribute has another side: ghāfilāt means "those to whom such a thing never even occurred". The attribute conveys, that is, not only that the act was not done but that there was not even a thought of it in the mind.

This sense of the word is widely recorded in the classical commentaries.

The verb is passive: luʿinū — "they are cursed". Its agent is not named.

And the tense is perfect. The ruling is announced not as something to be given in the future but as already given.

The root ل-ع-ن occurred in the sixth and seventh verses in the discussion of liʿān: laʿnatu'llāhi ʿalayh. The same word, in two different places:

What a man calls down upon himself conditionally in the seventh verse is announced as a ruling in the twenty-third. This is an observation about the word order, and the places of the words stand in the text.

The verse repeats the verb of the fourth (yarmūn) but announces a different outcome:

The two verses give the counterpart of the same act on two different planes. And the increase in the attributes (three of them) shows that the second verse takes as its subject a narrower case — the case in which innocence is certain.

I record this as an observation about the word order.

Yawma tashhadu ʿalayhim alsinatuhum wa aydīhim wa arjuluhum bimā kānū yaʿmalūn

"On the day when their tongues, their hands and their feet will bear witness against them concerning what they used to do."

Up to now the root ش-ه-د has occurred four times in the sūra: at 2 (wa'l-yashhad), at 4 (shuhadāʾ, shahādatan), at 6 and 8 (shuhadāʾ, shahādāt), at 13 (shuhadāʾ).

And now it occurs for a fifth time — but the witness has changed.

The sūra builds a chain of witnessing, and at the end of the chain the witness and the accused meet in the same body.

I record this as my own reading; the distribution of the root can be counted in the text.

The order of the limbs is worth noting: tongue, then hand, then foot.

In similar verses of the Qurʾān the order may differ (at Fuṣṣilat 41/20-21, hearing, eyes and skins; at Yā Sīn 36/65, mouths, hands and feet). That the tongue is put first here has directly to do with the subject of the section.

For the act treated in the verses from eleven to twenty-three was an act done with the tongue:

The same organ, nine verses apart, in two opposite tasks: first it carries the report, then it bears witness against the one who carried it.

I record this ring as my own reading; the places of the words and their sameness are the wording of the text.

The preposition is worth noting: ʿalā — "upon, against". In Arabic shahida li (he bore witness in someone's favour) is distinguished from shahida ʿalā (he bore witness against him). Here it is the second.

And the picture the sentence builds is this: a man is given up by his own body. There is no possibility of defence, because the other party is no stranger.

This is the reverse of the word buhtān in the sixteenth verse, and I record it as my own reading: there the one slandered could not answer, because he had no evidence in hand. Here the one who slandered cannot answer, because the evidence is his own body.

Yawmaʾidhin yuwaffīhimu'llāhu dīnahumu'l-ḥaqqa wa yaʿlamūna anna'llāha huwa'l-Ḥaqqu'l-Mubīn

"On that day God will pay them their due in full, and they will know that God is the plain Truth."

Root: و-ف-ي. Its concrete sense is to fulfil a thing completely, without any lack. Wafāʾ — to keep one's word in full. Awfā — he measured in full. Tawfiya — payment without deficiency.

And the word is the opposite of the root treated in al-Muṭaffifīn: taṭfīf is short measure; tawfiya is full measure.

دِين — root د-ي-ن. In Arabic the word has three fields of sense, and all three issue from this root:

The sense here is the second: the recompense that is due. The word's link with "debt" joins with the verb of the verse: yuwaffī — to pay in full. The verse, that is, builds a scene of payment.

And the attribute al-ḥaqq is worth noting: the recompense is not only "in full" but also "just". The two words reinforce one another: complete, and in its right place.

This is the only place in the sūra where the name al-Ḥaqq occurs, and the only place where the root ب-ي-ن comes together with a divine name.

And an observation about the word order may be recorded: the section between the eleventh and the twenty-fifth verses opened with a slander and closes with the name al-Ḥaqq al-Mubīn.

The two pairs of words stand facing each other: in the twelfth verse the believers were asked to say hādhā ifkun mubīn; in the twenty-fifth God is named al-Ḥaqq al-Mubīn. The same adjective, upon two opposite objects.

I record this as my own reading; the wording of the two expressions stands in the text and can be compared.

Al-khabīthātu li'l-khabīthīna wa'l-khabīthūna li'l-khabīthāt, wa'l-ṭayyibātu li'l-ṭayyibīna wa'l-ṭayyibūna li'l-ṭayyibāt, ulāʾika mubarraʾūna mimmā yaqūlūn, lahum maghfiratun wa rizqun karīm

"The corrupt are for the corrupt, and the corrupt for the corrupt; the good are for the good, and the good for the good. Those are free of what is said. For them there is forgiveness and a generous provision."

This verse is one of the verses in the Qurʾān most disputed as to how it is to be understood, and the cause of the disagreement is that what the subjects of the sentence are is not stated.

Al-khabīthāt and al-ṭayyibāt — feminine plural adjectives. But what they are adjectives of is not said. In Arabic this pattern can answer to persons, to words, or to acts.

Three main readings are transmitted in the classical commentaries:

No preference is imposed in this section, and none is binding.

But a note of method needs to be entered, and it is a direct requirement of the method (Method): no ruling about persons can be drawn from this verse.

The reason lies within the verse and can be shown from the arrangement:

• The second half of the verse itself states the conclusion to be drawn from the first: ulāʾika mubarraʾūna mimmā yaqūlūn — "they are free of what is said." The point the sentence arrives at, that is, is not the classifying of anyone; it is the announcing of the invalidity of something that has been said.
• And the verse comes at the end of a whole section (11-26). The subject of that section, from beginning to end, is what it is to lay down a ruling about people without evidence.

To use the verse in order to divide people into two groups, then, is the very thing the section it stands in forbids.

I record this as my own reading, and its ground is the second half of the verse.

And there is one further practical consequence: to establish which group a person falls into, the condition of proof the sūra itself has laid down (four witnesses, 24/4 and 24/13) is required. The verse, that is, cannot be read apart from the sūra's own procedure.

This root is analysed here for the first time in this commentary.

The sense the lexicographers record: a thing's being corrupt, foul, useless in its substance. Khubth — badness, filth. Khabath — the dross that rises to the top when a metal is melted, the part that is thrown away.

And this last usage gives the concrete picture of the root: the scum that rises to the surface when gold or silver is melted, and is discarded, is called khabath.

At the centre of the word, then, there is an idea of separating out: the part that separates under fire and is thrown away.

I record this as an observation about the root and add this: this picture in the word agrees with the structure of the verse's sentence. The verse describes a separation — but does not say how the separation comes about.

This root was analysed at al-Baqara, 2/168; I rely on that treatment. The note there: what is pleasant, clean, good; the word answers both to what is pleasant to the senses and to what is clean in its substance.

The opposition of the two roots:

That both roots are used in the same fields is one of the reasons for the indefiniteness in the verse. The words do not, by themselves, settle what they are adjectives of.

Root: ب-ر-أ. Its concrete sense is to part from a thing, to slip free of it, to be delivered. Barāʾa — a slipping free of an obligation. Bāriʾ — the Maker (who separates a thing out of nothing); this name was taken up at al-Ḥashr, 59/24. Bariʾa mina'l-maraḍ — he recovered from the illness.

The form is worth noting: mubarraʾūn — the passive participle of Form II.

It is not said "they cleared themselves". It is said "they have been held clear". The agent is not named, but it is understood from the sentence's context.

And this is the answer given to the matter of buhtān in the sixteenth verse, and I record it as my own reading:

It was recorded there: the one slandered cannot defend himself, because a thing that did not happen leaves no trace. The twenty-sixth verse takes that defence out of the man's hands and puts it elsewhere: the passive voice.

Acquittal, that is, comes not by being proved but by being announced.

The verse names in a single word the thing they are held clear of: mā yaqūlūn — "what they say".

At the end of the section, that is, the whole matter is reduced to one word: speech.

And this forms a ring with the section's opening:

The same root (ق-و-ل) three times in the section, and the section closes with it.

I record this as an observation about the word order.

The section closes with two promises: maghfira and rizqun karīm.

كَرِيم — root ك-ر-م: precious, generous, noble. The word was analysed at al-Ḥujurāt, 49/13 (akramakum).

And the placing of the word rizq here is worth noting. The subject of the section was a matter of standing; at its close a word of material favour comes. In the classical commentaries it is recorded that the attribute karīm qualifies the kind of the provision: provision given without being kept waiting for, without reproach, without deficiency.

Yā ayyuhā'lladhīna āmanū lā tadkhulū buyūtan ghayra buyūtikum ḥattā tastaʾnisū wa tusallimū ʿalā ahlihā, dhālikum khayrun lakum laʿallakum tadhakkarūn

"O you who believe, do not enter houses other than your own until you have gained leave by establishing familiarity and by greeting their people. That is better for you, that you may take heed."

With the twenty-seventh verse the sūra passes into a new block, and the passage is marked by an address: yā ayyuhā'lladhīna āmanū.

And the block is the exact counterpart of the block before it.

The verses from eleven to twenty-six took as their subject a word's entering into a person's privacy. The twenty-seventh verse takes up the same matter on the physical plane: a body's entering into a house.

Three blocks, three distinct boundaries and three distinct instruments of violation. The sūra repeats the same idea on three planes.

I record this as my own reading; the subjects of the blocks and their order stand in the text.

And this is the direct continuation of the matter of privacy recorded in the section on the sūra in al-Ḥujurāt. Two notes were entered there:

In the discussion of 49/4: "a physical boundary can be crossed by a voice. A door may be shut and a man may still disturb the one inside. Privacy is not kept by a wall alone; it is kept by the wall's being respected." In the discussion of 49/12: "The sūra takes up privacy twice: in the fourth verse as a physical boundary, in the twelfth as a boundary of knowledge."

al-Nūr takes up the same two layers — but reverses the order: first the boundary of knowledge (11-26), then the physical boundary (27-29).

And al-Nūr adds a third layer: the glance, which is not in al-Ḥujurāt (30-31).

The two sūras complete each other, and I record this table as my own reading; the places and subjects of the verses stand in the text.

بَيْت — root ب-ي-ت. Its concrete sense is to pass the night. Bāta — he passed the night; bayāt — a passing of the night. And bayt — the place where the night is passed.

That the word comes from this root is worth noting, and I record it as an observation about the root: in Arabic a house is named not "the place one dwells in" or "the place one is protected in" but "the place where one passes the night".

And the night is the time when a man is least protected. The sūra's fifty-eighth verse will regulate this separately: three times will be enumerated at which leave must be asked, and two of them will have to do with the night.

I record this connection as my own reading; the sense of the root stands in the dictionary.

And the wording of the expression is worth noting: buyūtan ghayra buyūtikum — "houses other than your houses".

The verse does not say "do not enter another's house"; it says "houses other than your own". The boundary, that is, is drawn not by ownership but by a person's own domain. And buyūtikum is plural: a man may have more than one house, or the address is directed at the community.

This is the most striking word in the verse and will be taken up here in detail.

Root: أ-ن-س. The root was analysed at al-Qaṣaṣ, 28/29; I rely on that treatment. The note there:

ءَانَسَ — root أ-ن-س: to catch sight of, to grow warm. From the same root come ins (man) and uns (familiarity). The lexicographers record the word's sense of "to sense a light from afar".

The derivatives of the root:

And now the verse's most important question: why is the verb of asking leave chosen from this root?

In Arabic there is a direct verb for "to ask leave": istaʾdhana (root: أ-ذ-ن). And the sūra uses this verb in many places: in the twenty-eighth verse (yuʾdhana lakum), in the fifty-eighth (li-yastaʾdhinkum), in the fifty-ninth (fa'l-yastaʾdhinū), in the sixty-second (ḥattā yastaʾdhinūh).

But in the twenty-seventh verse — that is, in the first and founding command — this verb is not used. Tastaʾnisū is used.

This is the most important point in the verse in point of language, and it has been dwelt on in the classical commentaries too.

What the roots of the two verbs say:

And the difference is this, which I record as my own reading:

Istiʾdhān runs one way: one party asks, the other gives or does not give. It is a transaction.

Istiʾnās runs both ways: in its root there is a mutual growing warm. To stand at the door and ask leave is not enough; the one behind the door must also not be discomposed.

The verse, that is, frames the act of asking leave not as a procedure but as a relation.

And its concrete counterpart is recorded thus in the classical commentaries — I transmit them as explanations that are transmitted:

I impose no preference. But what the three explanations have in common is worth noting: all of them include the visitor's making himself known beforehand.

And the measure recorded at al-Ḥujurāt, 49/4-5, is completed here:

There those who called out were censured — because there was no waiting. And what was commanded there was patience: wa law annahum ṣabarū ḥattā takhruja ilayhim.

al-Nūr 24/27 regulates the other side of the same matter: there what was censured was calling out from behind the wall; here what is commanded is establishing familiarity before the door.

The difference between the two verses lies in the direction of the voice, and I record this as my own reading:

The same act (making a sound), with two different intentions, receives two opposite rulings. The difference lies in whom the sound is made for.

سَلَّمَ — root س-ل-م. Its concrete sense is to be sound, entire, without flaw. Salāma — being clear of harm. Salām — a noun from this root.

And the function of the greeting here is worth recording: a greeting is an address, and it is the acknowledgement of a person's existence.

The two verbs the verse asks for, then, do two distinct works:

The one looks towards oneself, the other towards the one opposite.

And the verb from the root س-ل-م will return twice more in the sūra: in the sixty-first verse, fa-sallimū ʿalā anfusikum — "greet yourselves". There is a ring between the two verses, and it will be taken up at the sixty-first.

The particle is worth noting: ḥattā — "until". In Arabic this particle conveys a limit (ghāya).

The prohibition, that is, lasts up to a certain point, and when that point comes it is lifted. The prohibition is not absolute but conditional.

And the same particle is repeated through the sūra:

In all four usages the particle conveys a waiting. I record this as an observation about the word order.

The verse-ending is the same as that of the sūra's first verse: laʿallakum tadhakkarūn.

This verse-ending occurs twice in the sūra: at 1 and at 27. And both occurrences stand in verses that open a distinct block of the sūra.

I record this as an observation about the word order; the places of the verse-endings can be counted in the text.

And the expression khayrun lakum is worth noting too. The same expression occurred in the eleventh verse: bal huwa khayrun lakum. In both places it is said that something hard for the addressee is for his own good.

One. The root of the verb of asking leave.

The root the verse chooses — uns, to grow warm — settles what asking leave is: not a transaction but a relation.

The practical counterpart of this is that leave may be asked and given in form, and yet what the verse asks for may not have come about. To knock at a door and walk in is different from knocking at a door and waiting for the one inside to be ready.

Two. Where the boundary lies.

The verse sets the door of the house as a boundary. And as was recorded at al-Ḥujurāt, 49/4, a man's door today is not only the door of his house:

"His telephone, his message box, the end of his working hours, his day off — each of these is a chamber: an enclosed, separated space."

What al-Nūr 24/27 adds is what is to be done in entering that space: first make yourself known, then greet the other, then wait for leave. Three steps, and all three before entering.

Three. And the addressee of the command.

The verse does not tell the householder "lock your door". It tells the one entering "ask leave". The obligation lies not with the one who needs protecting but with the one who approaches the boundary.

This is the same logic recorded at al-Ḥujurāt, 49/6: there too the obligation lay not with the one who spread the report but with the one who received it. Both sūras place the obligation upon the party whose behaviour is in his own hands.

Fa-in lam tajidū fīhā aḥadan fa-lā tadkhulūhā ḥattā yuʾdhana lakum, wa in qīla lakumu'rjiʿū fa'rjiʿū, huwa azkā lakum, wa'llāhu bimā taʿmalūna ʿalīm

"If you find no one in them, do not enter until leave is given you. And if you are told 'go back', then go back; that is more purifying for you. God knows what you do."

The verse fills the two gaps the twenty-seventh left:

And what the two rulings have in common is worth recording: in both, the prohibition rests upon an absence of answer or upon a refusal.

The first possibility looks unnecessary at first: if there is no one, there is no one to give leave either; so what is there to wait for?

But the verse's logic is this, and I record it as my own reading: the twenty-seventh verse had commanded the asking of leave. And a man who gets no answer is inclined to count leave as given — "there is no one, so I may go in". The verse closes off that inference.

Silence, that is, is not leave. This is a concrete note that can be read from the arrangement.

The second half of the verse does something finer than the first, and its arrangement is worth noting.

The same verb occurs twice: irjiʿū (go back) and fa'rjiʿū (then go back). The one is the word spoken, the other the command given.

The verse, that is, puts the householder's word directly in the place of a ruling. No ground is asked for, no reason is required, no insistence is made.

And this is a thing worth recording: the verse lays no obligation upon the householder to "say why". The refusal is valid without being explained.

I record this as my own reading; its ground is that the verse lays down no condition of a reason.

The root ز-ك-و was analysed at the twenty-first verse. Its concrete sense: to increase, to grow; and to be cleansed.

The form is an elative: azkā — "more purifying, cleaner". And the other side of the comparison is not stated: more purifying than what?

The rule the classical lexicographers record is this: the omission of the compared term after an elative leaves the comparison wide. "Than insisting", "than being hurt", "than resenting" — all fall within its scope.

And the placing of the word azkā here says something worth recording, which I give as my own reading:

To be turned back from a door touches a man's pride. And the verse gives this situation the name of a purification. Going back, that is, is placed not as something lost but as something gained.

And the same word will come again two verses later: dhālika azkā lahum (30) — the outcome of lowering the glance. Both commands are commands to hold back, and both are rewarded with the same word.

I record this parallel as my own reading; the sameness of the words is the wording of the text.

The verse-ending is almost the same as that of al-Ḥujurāt, 49/18: wa'llāhu baṣīrun bimā taʿmalūn.

The difference is a single word: al-Ḥujurāt has baṣīr (seeing), al-Nūr ʿalīm (knowing).

And the difference agrees with the subject of the two sūras, and I record this as my own reading:

The last verse of al-Ḥujurāt took as its subject a man's claim about himself — something that cannot be told apart from outside. There baṣīr is fitting: the one who sees what is not seen.

al-Nūr 24/28 takes as its subject something that happens before a door — a moment no one sees: the door is knocked at, no answer comes, and the man decides alone what to do. There ʿalīm is fitting: the one who knows what is not known.

This is not a claim of preference but an observation of reading.

Laysa ʿalaykum junāḥun an tadkhulū buyūtan ghayra maskūnatin fīhā matāʿun lakum, wa'llāhu yaʿlamu mā tubdūna wa mā taktumūn

"There is no fault in your entering houses that are not lived in, in which there is something of use to you. God knows what you make public and what you conceal."

The verse brings an exception to the prohibition laid down two verses before, and the exception has two conditions:

And that the two conditions are required together is worth noting. The first alone is not enough: it is not said that every uninhabited house may be entered. The second condition asks for a reason.

I record this as an observation about the word order.

Root: س-ك-ن. Its concrete sense is the stopping of motion, coming to rest. Sakana — it stopped, it grew still. Sakīna — tranquillity. Maskūn — dwelt in, inhabited. Sakīna and sukūn are from the same root.

And the choice of the word gives a definition: what makes a place a "house" is someone's coming to rest within it.

The verse's exception, that is, looks not at a building but at a life. Not every structure with walls is a protected space; what is protected is someone living there.

I record this as my own reading; the sense of the root stands in the dictionary.

Root: ج-ن-ح. Its concrete sense is to lean, to incline to one side. Janāḥ — a wing (the limb that folds to the side of the body). Janaḥa — he inclined.

And junāḥ: a straying from the right road by leaning aside; whence "fault, objection, blame".

The picture in the root of the word is worth noting: junāḥ is a swerving aside from the straight road. Its first sense, that is, is an error of direction, not a weight.

And the word occurs four times in the sūra:

All four usages are the lifting of an obligation. And all four stand in the sūra's blocks of manners.

This is a characteristic of the sūra, and I record it as my own reading: in laying down rules of manners the sūra writes beside every rule the place where that rule does not apply. A rule is set, and then the rule's limit is set.

Two verbs stand face to face: