al-Taḥrīm 66:4

In tatūbā ila'llāhi fa-qad ṣaghat qulūbukumā, wa in taẓāharā ʿalayhi fa-inna'llāha huwa mawlāhu wa Jibrīlu wa ṣāliḥu'l-muʾminīn, wa'l-malāʾikatu baʿda dhālika ẓahīr

"If you two repent to God — your hearts have inclined. But if you back one another against him, then God is his mawlā; and Jibrīl, and the righteous among the believers. And after that the angels too are a backer to him."

The first word of the verse gives a piece of grammatical information, and that information is the key to reading the verse.

تَتُوبَا — the dual form. In Arabic the singular, the dual and the plural are separate conjugations; the dual is used only for two persons. The same form continues throughout the verse: qulūbukumā (the hearts of you two), taẓāharā (if you two back one another).

That is, the verse is said to exactly two persons.

This is information lost in translation. English "you" serves for two people and for twenty alike; Arabic makes the distinction. That the verse's addressee is two is written in the verse itself.

And this is the narrowest point in the sūra's map of addressees. In the previous verse one person was spoken of (baʿḍi azwājih); here two persons are addressed; two verses later it will pass to all the believers. The text first narrows as far as it can, then widens as far as it can.

Names are given in the reports as to who is meant. For the grounds I recorded in the introductory section I write no name here: there are differences among the reports, and the verse itself gives no name.

This is the most disputed grammatical point in the verse.

The sentence is built thus: "If you two repent — your hearts have inclined."

This is an incomplete sentence in Arabic as it is in English. The condition "if you repent" must have an answer: if you repent, what happens? The verse does not say. What it says stands rather like the ground for repenting.

The explanations given for this gap in the classical sources on grammar:

All three are defensible and in practice they come out at the same place. I do not impose a preference.

But there is something that follows from the form of the sentence and it must be recorded: the verse does not command repentance. It does not say tūbū (repent); it says in tatūbā (if you two repent). The conditional form does not compel the one addressed — it shows the door.

And this becomes meaningful when compared with the eighth verse. There the same root will come in the imperative: tūbū ila'llāhi tawbatan naṣūḥā. That is:

Where it is particular, an offer; where it is general, a command. I record this ordering as an observation: the text does not give a command to two persons whose identity is known, it shows the possibility; when it says the same thing to everyone it uses the imperative.

Root ت و ب (t-w-b). This root was taken up at al-Baqara 2/37 and the findings there are not repeated. I recall it in one sentence: the sense of the root is to turn, and the same root is used both of the servant and of God — the servant tāba ila'llāh (turns towards God), God tāba ʿalayh (turns towards the servant). The main point recorded there was this: God's turning comes before the servant's turning; the servant can turn because he has already been turned towards.

What is new here: the preposition. In the verse tatūbā ila'llāh — "towards God". That is, the servant's direction is what is recounted, not God's turning. In the eighth verse the same preposition will come: tūbū ila'llāh.

This consistency is meaningful: in recounting the side that falls to the servant, the sūra always uses ilā. The servant's work is to change direction.

The most pictorial expression in the verse.

Root ص غ و (ṣ-gh-w). The concrete sense of the root: to bend, to lean to one side, to slip.

The liveliest use of the root has to do with the ear: aṣghā ilayhi — he gave ear to him. And this describes exactly a bodily movement: bending the head to that side in order to hear a sound better. From the same root comes iṣghāʾ (giving ear). The Qurʾān uses it in this sense too: "so that the hearts may incline towards it" (al-Anʿām 6/113) — there too the same root recounts the heart's bending to one side.

The other concrete usages strengthen the same picture: a vessel tilting to one side, the sun leaning to the west, a wall slipping to one side.

The function of the word in the verse. The verse does not say "your hearts have gone bad", it does not say "your hearts have darkened", it does not say "your hearts have become unbelieving". It says "they have inclined".

This is a thoroughly measured word and its choice is meaningful. To bend is not to break. A bent thing can be straightened; a broken thing cannot. In making the diagnosis the word leaves the door open.

And one further nicety: bending is a movement that is hard to notice. If a vessel overturns you see it; if it tilts slightly you do not. The word says that a heart can change direction by a small slippage — and that this slippage becomes visible only through an intervention from outside. This is what the verse does: it states the inclination.

An observation, as my own reading: the word's link with the movement of the ear falls into the context of the sūra. The third verse was about the hearing and the relaying of a word. The fourth verse speaks of the heart's inclining — and the same root is the root that recounts bending the head to hear a sound. I do not present this as a claim of wordplay; I only record that the picture the root carries falls into the context.

There is a mismatch in the verse that goes unnoticed: the subject is dual, but qulūb (hearts) is plural.

The expected form would be qalbākumā (the heart of you two — dual). Two persons, two hearts. But the verse says qulūbukumā: "the hearts of you two", plural.

The explanation the classical grammatical sources give is this: in Arabic, for organs whose possessor is dual and of which there is one apiece in each person, the plural is preferred instead of the dual. The reason the lexicographers give is the heaviness that arises from duals coming one after another — in the form qalbākumā two dual markers come one upon the other and this falls heavily on the Arab ear; the plural is lighter.

The same usage is found elsewhere in the Qurʾān as well; for example for the hand in the verse on theft (al-Māʾida 5/38) and in similar structures. This is recorded as a rule in the books of grammar.

I do not turn this into a point of meaning. That is, I make no interpretation such as "the plural points to the dividedness of the heart" — such an interpretation has been made in the classical sources too, but since the rule of grammar already requires this form, drawing a meaning from it remains without support. I record the rule and add nothing to it.

Root ظ ه ر — treated at the third verse: back, outer face, backing. I do not repeat it.

The pattern: تَفَاعُل (tafāʿul). And here we come again to a pattern that has met us many times in this commentary.

In al-Takāthur the function of this form was recorded in detail: the tafāʿul form states reciprocity (mushāraka) — it shows the act going back and forth between two or more parties. Fakhara (he boasted) → tafākhara (they boasted against one another); kaththara (he multiplied) → takāthara (they vied with one another in multitude). I do not repeat that analysis.

What the pattern adds here: ẓahara is to back someone; taẓāhara is to back one another. That is, the verse speaks not of two persons each taking a position, but of their forming an alliance.

And the difference is large. That two persons hold the same view is one thing; that they stand in that view by leaning on one another is another. In the second, each draws strength from the existence of the other. The correctness of the view is not separately tested — because the other party already thinks the same thing, and this works like a confirmation.

This is a mechanism observed in social psychology: that a conviction is shared by someone else increases the felt confidence in the correctness of that conviction — even if nothing at all has changed in the conviction itself. Even a group of two is enough to produce this effect.

I record this not as a claim of a "scientific miracle" but as the coinciding of the mechanism the pattern of the word describes with a modern observation. The pattern already says it: the verb goes back and forth between the parties.

And the place of the tafāʿul form in this sūra takes on meaning together with the other examples in the corpus:

All four are on the same pattern. And the pattern itself is neutral: reciprocity is neither good nor bad, it takes its value from its subject. The finding recorded at al-Baqara 2/148 holds good here too: the Qurʾān does not forbid competition or reciprocity; it changes their subject.

The second half of the verse gives a list and the form of the list is striking.

"God is his mawlā; and Jibrīl, and the righteous among the believers. And after that the angels too are a backer to him."

The structure of the list sets up an ascent:

1. الله — and first huwa mawlāh is said: He is the one who is mawlā. This is a repetition of the word that occurred in the second verse; there "your mawlā", here "his mawlā". 2. جبريل — the bearer of the revelation. 3. صالح المؤمنين — the righteous among the believers. 4. الملائكة — and after these, the angels.

And the effect of the ascent is this: on one side two persons, on the other side this list. The verse states the meaninglessness of the alliance by showing the disproportion.

The expression صَالِحُ الْمُؤْمِنِينَ is disputed and here I do not speak with certainty. The construct is built with a singular word (ṣāliḥ); against that, its sense may be plural. The classical explanations:

The first explanation requires fewer assumptions linguistically. But I do not impose a preference and it is not binding.

And the real point of grammar is at the end: ظَهِير is singular.

The sentence says "wa'l-malāʾikatu baʿda dhālika ẓahīr". The subject is plural (al-malāʾika — the angels), the predicate singular (ẓahīr — a backer). The expected form would have been ẓuharāʾ or some similar plural.

The explanations given in the classical sources:

All three have been defended in the classical sources. I adopt no preference.

But the third explanation has a beautiful side in point of meaning and I content myself with recording it: if the company has been taken as a single whole, the sentence says this — on one side there are two persons and the two back one another; on the other side there are countless beings and all of them are a single backing. Multitude of number does not guarantee being one; the side that is one is named with a singular word. I write this as my own reading; it is not transmitted.

One. That an alliance does not produce truth.

The mechanism the verse records with the tafāʿul pattern touches one of the commonest errors of daily life: taking the finding of someone who shares a view to be the confirming of that view.

That two persons think the same thing gives no new information about the truth of that thing — both may be looking from the same place, with the same incomplete information, with the same interest. But the felt certainty rises. And because it rises, the matter is not thought through again.

This is a relative of the finding recorded in al-Bayyina: there it was said that knowledge produces division, and that the reason for this is not a lack of knowledge but the using of knowledge to enlarge one's own position. Here the mechanism is on a smaller scale: a ring of confirmation made of two people.

The measure the verse gives is simple and comes from outside: how large is the list facing you? That is, a conviction is weighed not by the number of those who share it but by the weight of what stands against it.

Two. The diagnosis "it has inclined".

The verse's choice of word — not broken but inclined — shows a method.

The word used in describing a fault also determines the fault's correctability. A diagnosis that says "you are like this" leaves the other party no possibility of change; because being is not a thing that can be changed. A diagnosis that says "you have inclined in this direction" describes a movement, and every movement can be undone.

This is what the verse does: it says the situation exactly, but it gives no adjective, it describes a movement.

Its practical counterpart holds both in warning someone else and in assessing one's own state. A person's saying about himself "I am just that sort of person" is a sentence that looks like a diagnosis but in fact makes change impossible. To say "I have inclined in that direction" says the same truth without covering it, and makes something doable.

Three. That repentance is not commanded.

The verse does not command repentance to the most particular addressee; it offers it in the conditional. And this is of the same family as the attitude seen in al-Muzzammil and ʿAbasa: a correction is made, then the one addressed is set free.

What this says to the present has to do with how a warning is made. To turn a warning into a command places the one warning in the position of a supervisor — and from that moment the matter is no longer the wrong that was done but the relation of authority between two people. The conditional form prevents this transformation: the door is shown, no one is pushed.