113. Sūrat al-Falaq

Five verses, twenty-three words. The second-to-last sūra of the Qurʾān and the first of the two sūras that form the close of the codex.

The sūra builds a single sentence and completes that sentence four times: "I take refuge — from this, from this, from this, from this." There is no ruling in it, no narrative, no description. From beginning to end there is a single act: taking refuge. And the side with whom refuge is taken is named once, the things refuge is taken from four times.

This is the plainest thing a text can do: to count the danger and to point to the refuge.

In the tradition Sūrat al-Falaq and Sūrat al-Nās are almost never treated separately. The two together are called al-Muʿawwidhatān: "the two sūras that give refuge".

The word comes from the root ع-و-ذ — the root of "taking refuge", which we shall see in detail shortly. Muʿawwidha, in the pattern of the active participle, means "that which gives refuge"; with the dual ending -atān it becomes "the two that give refuge". It is also transmitted that the word is read as muʿawwadhatān (passive: "the two by means of which refuge is taken"); both readings are meaningful and the difference between them is unimportant — one looks to the function of the sūras, the other to their use.

When Sūrat al-Ikhlāṣ is added to these, the three together are called al-Muʿawwidhāt. Both this threefold usage and the twofold usage occur in the sources; the boundary of the terms is not always sharp.

To read the two sūras together is not merely a habit but an invitation issued by the structure of the text itself. Both open with the same words (qul aʿūdhu bi-Rabbi), both continue with the same pattern (min sharri), and each completes the other. Where al-Falaq closes, al-Nās opens. The detail of this relation will be taken up after both sūras have been treated, at the end of the section on al-Nās; but the basic difference to be kept in mind throughout the reading is this:

al-Falaq is a taking of refuge against the evils that come from outside, al-Nās against the evil that comes from inside.

The name "al-Falaq" is taken from the word in the first verse of the sūra. This is the commonest way of naming in the Qurʾān.

In some sources the sūra is also named by its full name, "Qul aʿūdhu bi-Rabbi'l-falaq" — naming short sūras by their first sentence is a widespread practice.

The majority say it is Meccan. There is also a view that it is Medinan, and that view rests on some of the reports about the occasion of revelation that we shall take up separately below — in particular the report that the Prophet was bewitched in Medina.

The content and style of the sūra (short verse-endings, a dense weave of sound, direct address) fit the general character of the Meccan sūras; but this alone is not decisive, because there are short Medinan sūras as well. I do not issue a definite ruling. This distinction does not change how the sūra is understood.

Qul aʿūdhu bi-Rabbi'l-falaq "Say: I take refuge with the Lord of the falaq (of the morning that is split open and lets the light out / of everything that is split open)."

The meaning of this imperative's having been left inside the text was treated in detail in the section on al-Ikhlāṣ; I shall not repeat it here. But two points need to be added as belonging particularly to this context.

First: the words of taking refuge are themselves taught words. When a person is afraid he says something of his own accord — a prayer, a cry, a word. The sūra meets this spontaneity with a formula: I am telling you how to take refuge as well. The moment of fear is the moment when a person is most defenceless and most easily directed; if in that moment the source of the words that come out of his mouth is left indeterminate, the taking of refuge may turn towards other doors. Indeed the Qurʾān records a practice in which people took refuge with the jinn, and states its outcome: "And there were men among human beings who took refuge with men among the jinn, and so they increased them in insolence" (al-Jinn 72/6). The same root is used in that verse too (yaʿūdhūna). The verb "to take refuge", that is, is not a neutral verb; made to the wrong address, it produces harm.

Second: the one addressed by the imperative is the Prophet. He is the first to learn the taking of refuge. This shows that taking refuge cannot be counted a mark of weakness — taking refuge is not the helplessness of one who has no protection but the right conduct of one who knows where protection lies.

The five sūras that open with "Qul". al-Jinn (72), al-Kāfirūn (109), al-Ikhlāṣ (112), al-Falaq (113), al-Nās (114). The last three are consecutive. In the section on al-Ikhlāṣ we touched on the point that these three are a sequence that states a position: in al-Kāfirūn the boundary of belief is drawn, in al-Ikhlāṣ the one believed in is described, in al-Falaq and al-Nās the taking of refuge is declared. The order is meaningful: first you say in whom you believe, then you take refuge with Him. That al-Falaq comes immediately after al-Ikhlāṣ does not look accidental in this respect — 112 answers the question "who is He", 113-114 the question "what then is to be done".

The verb of the sūra and the shared backbone of the two sūras.

Root: ع-و-ذ. The original sense of the root is to take hold of a thing, to cling to it, to get behind it and be protected. ʿĀdha bihi — he took refuge with him, he clung to him, he was protected by him. In taking refuge there are two movements at once: moving away from danger and drawing near to something. The English expression "to take refuge" carries the same doubling: one takes refuge from a thing and in a thing — going behind it, under it, inside it.

Derivatives of the same root:

• istiʿādha (استعاذة): the seeking of refuge. The isti- pattern (istifʿāl) conveys in Arabic "to want, to seek"; istiʿādha is to seek, to ask for refuge. This pattern occurs in the Qurʾān as an imperative: "When you are about to recite the Qurʾān, take refuge with God from the accursed devil" (al-Naḥl 16/98).
• maʿādh (معاذ): the place refuge is taken in, a shelter. Maʿādha'llāh — "I take refuge with God", the English "God forbid". It occurs in the Qurʾān on the lips of Joseph (Yūsuf): "Maʿādha'llāh!" (Yūsuf 12/23).
• ʿiyādh (عياذ): taking refuge. It is used in the expression ʿiyādhan bi'llāh.
• taʿwīdh / muʿawwidha (تعويذ / معوذة): the words recited in order to take refuge; and the text that contains those words. The name al-Muʿawwidhatān comes from here.

The conjugation of the verb says something. The verse has aʿūdhu — imperfect (muḍāriʿ, the present and general tense) and first person singular.

Its being imperfect: taking refuge is not a work done once and finished but an ongoing state. Not "I took refuge" but "I take refuge / I am taking refuge". In Arabic the imperfect verb conveys repeated and continuing action. Because the danger is continuous, the taking of refuge is continuous too.

Its being first person singular: it does not say "we take refuge". Istiʿādha, even when it is done in a crowd, is a work done one by one. No one can take refuge on another's behalf. This is to be thought together with another point that will meet us towards the end of the sūra: in Sūrat al-Nās the threat is directed at a plural field (the breasts of the people), but the one taking refuge is singular. The danger is shared, the taking of refuge is personal.

The verb aʿūdhu does not take a direct object; it becomes transitive by means of the letter bāʾ: aʿūdhu bi- — "I am protected by…", "I cling to…".

In Arabic grammar this bāʾ is called bāʾ al-ilṣāq (the bāʾ of adhesion); it is the letter that brings two things into contact and removes the distance between them. The root's sense of "taking hold" is thus written into the grammar of the sentence as well: to take refuge is not to call for help from a distance but to make contact with something.

This also shows the fine difference between taking refuge and prayer. In prayer something is asked for; in istiʿādha one moves into a place.

The sūra does not say "aʿūdhu bi'llāh" (I take refuge with God); it says "aʿūdhu bi-Rabbi'l-falaq". Yet the best-known formula of istiʿādha is "aʿūdhu bi'llāh". This choice calls for attention.

The word Rabb. Root r-b-b: to raise a thing step by step, to look after and bring up, to nurture; and at the same time to own, to hold the authority of disposal over it. The detailed analysis of the word was made in the section on al-Fātiḥa. Three of its aspects come to the fore here:

1. Rabb is a name that conveys a relation. "Allāh" is a proper name and denotes the Essence; "Rabb", on the other hand, is always the Lord of something — Rabb al-ʿālamīn, Rabb al-falaq, Rabb al-nās. In the moment of fear a person turns to an authority with which he has a bond. Refuge is taken not with a power you do not know but with one who knows you.

2. Rabb means the one who looks after and watches over. Protection is contained within the concept of nurture: one who raises protects what he raises. This is exactly the name to be turned to for protection from evil. The choosing of an attribute to fit what is asked for is a widespread manner of the Qurʾān; an accord is established between the thing asked for and the name it is asked by.

3. Rabb means the one who has disposal. And this is the most critical point: the one with whom refuge is taken is also the owner of the thing refuge is taken from. The expression "the Lord of the falaq" states in whose hand the feared domain lies. The darkness is His and the morning is His; the creation that harbours the evil is His too.

This third point is the tawḥīd backbone of the sūra. None of the evils that will be counted through the sūra is an independent power. Evil is not outside the dominion but inside it. The one who takes refuge takes refuge with his enemy's master.

This is why it is striking that the overwhelming majority of the expressions of istiʿādha in the Qurʾān are built with the name "Rabb":

• "My Lord! I take refuge with You from the promptings of the devils; and I take refuge with You, my Lord, from their being present with me" (al-Muʾminūn 23/97-98).
• "I take refuge with You from asking of You that of which I have no knowledge" — the words of Noah (Nūḥ) (Hūd 11/47).
• "I have taken refuge with my Lord and your Lord from every arrogant one who does not believe in the Day of Reckoning" — the words of Moses (Mūsā) (Ghāfir 40/27; similarly al-Dukhān 44/20).
• In the words of Mary (Maryam), on the other hand, another name is chosen: "I take refuge with al-Raḥmān from you" (Maryam 19/18) — because what is asked for there is mercy, protection from a threat.

The choice of name, that is, varies with the context and is always meaningful.

The most debated word of the sūra, and its name.

Root: ف-ل-ق. Its concrete sense is to split, to crack, to divide in two. But falq in Arabic is not any kind of cutting; it is the opening of something closed from within or by a force, the parting of a whole and the coming out of something from between. Filqa is the name of the split-off piece — the half of a rock, of a stone, of a plank.

The derivatives of the root and their uses in the Qurʾān show the heart of the matter:

فَالِق — the one who splits. The Qurʾān uses this active participle of God in two consecutive verses:

"God is indeed the one who splits the grain and the date-stone. He brings the living out of the dead, and He is the one who brings the dead out of the living. That is God; how then are you turned away?" (al-Anʿām 6/95)

"He is the one who splits the dawn. He has made the night a rest, and the sun and the moon a measure of reckoning. That is the ordaining of the Mighty, the Knowing." (al-Anʿām 6/96)

The two verses follow one another and use the same word on two different scales: fāliqu'l-ḥabbi wa'n-nawā (the one who splits the grain and the date-stone) and fāliqu'l-iṣbāḥ (the one who splits the dawn).

These two images say the same thing. The seed under the soil is closed, is in darkness, looks dead; it cracks at one point and a green shoot comes out of it. The night is closed, is dark, is motionless; it cracks at one point on the horizon and light comes out of it. In both, something closed and dark is split, and life comes out of it.

ٱنْفَلَقَ — it was split. The quasi-passive (muṭāwaʿa) pattern of the same root occurs in the splitting of the sea: "Thereupon the sea was split, and each part was like a great mountain" (al-Shuʿarāʾ 26/63). Here too a mass that is closed and impassable is split, and a road is opened through it.

What, then, does "al-falaq" in the verse mean? Is the word here a verbal noun or a noun, and what concrete thing does it point to? The views transmitted in the classical commentaries can be gathered as follows:

The majority prefer the first view. But the second does not exclude the first; on the contrary, it widens it. The splitting of the morning is the most visible and most regular instance of the event of "splitting" — a splitting repeated every day and seen by everyone. The word's opening out from this concrete instance to the general sense is the familiar working of Arabic.

Why this name, in this context? The sūra is a sūra of taking refuge from evil. The densest representation of evil is darkness. And the one with whom refuge is taken is presented as the Lord of that very splitting which cleaves the darkness and brings the light out of it.

In the very first verse, that is, the sūra's whole answer has been given: the darkness itself looks closed and absolute; but there is a power that splits the darkness, and it is with Him that refuge is taken. The morning is the proof that darkness is not the last word — and it is repeated every day.

We can take this one step further. The image of splitting has three scales, and all three are images the Qurʾān uses: the splitting of the seed (the plant), the opening of the womb, of the egg (the birth of a living creature — the Qurʾān covers this too with the language "He brings the living out of the dead"), the splitting of the morning (the cosmos). At all three scales the same law works: life comes out by the splitting of a closed and dark place. This is not a claim of a "scientific miracle"; it is the finding that the image the Qurʾān has chosen catches a pattern that really does recur in nature.

There is a confusion into which readers in English, and sometimes even those who know Arabic, fall. فلق (f-l-q) and فلح (f-l-ḥ) are not the same root. Their last letters differ: one is qāf, the other ḥāʾ. In Arabic the root is determined by the identity of its three consonants; the change of one letter changes the root.

Even so, their fields of meaning overlap strikingly:

• ف-ل-ح: to split, to crack (especially the soil). From here fallāḥ — the one who splits the soil, the farmer. And from here falāḥ / aflaḥa / mufliḥ — to attain salvation, to attain what one hoped for, to succeed. "Qad aflaḥa'l-muʾminūn" — "The believers have indeed attained salvation" (al-Muʾminūn 23/1). The "ḥayya ʿala'l-falāḥ" of the call to prayer is from this root too.
• ف-ل-ق: to split, to crack. From here falaq — the splitting of the morning, the thing that is split.

The Arabic lexicographers noticed and debated such nearnesses of meaning among roots that share their first two letters; in the tradition of the language this topic is treated under the heading of "greater derivation". In the series of roots beginning with ف-ل (فلق، فلح، فلذ، فلج، فلي) a recurring sense of "separating, splitting" is observed.

But this is not a claim that they are one root, and it should not be presented as one. How much of it is a real historical kinship and how much is coincidence or a pattern constructed after the fact is disputed in linguistics. I make no claim about origin here.

A conceptual link, on the other hand, can be made, and this link is a reading of my own, not a transmitted report: Falāḥ is the splitting of the obstacle in front of one and the opening of the road. Without the soil being split the seed does not come out; the farmer splits the soil, and salvation too is the splitting of a closedness. Falaq, for its part, is the splitting of the night and the coming out of the morning. The two words are the counterparts of the same image in two separate fields — one a road opened by human effort, the other a road opened not of itself but by God's splitting.

I do not present this reading as "the Qurʾān says this". I record it only as a scene that appears when we set the two words side by side without confusing them.

Al-falaq has the article (with al-). Two of the three items of evil that come later in the sūra, on the other hand, are without the article (ghāsiqin, ḥāsidin). This opposition is meaningful: the side with whom refuge is taken is definite, the evils refuge is taken from are indefinite. The refuge is known, is one, has an address. The threats are countless and shapeless.

We shall see this again at the end of the sūra.

Min sharri mā khalaq "From the evil of the things He has created."

Root: ش-ر-ر. One of the basic words the Qurʾān uses for evil, and the backbone word of this sūra, repeated four times.

Sharr, in its most general sense, means what is not wanted, what does harm, what is avoided. Its opposite is khayr. In Arabic these two words form a pair, and the Qurʾān often uses them side by side: "Whoever does an atom's weight of good will see it; and whoever does an atom's weight of evil will see it" (al-Zilzāl 99/7-8).

A point of grammar: sharr and khayr are used in Arabic both as nouns ("evil", "good") and as the comparative pattern (ism al-tafḍīl: "worse", "better"). Normally the afʿal pattern is required for comparison (akbar, aḥsan); these two words, however, convey comparison without changing pattern: "You are worse in position" (Yūsuf 12/77). Their being this firmly settled in the language shows how basic the concepts are.

From the same root come sharār / sharar — spark. In the Qurʾān it is used of the fire of Hell: "It throws up sparks like a castle" (al-Mursalāt 77/32). Some lexicographers make a link of meaning between "sharr" and "spark": that which leaps, spreads, sets alight. This association is interesting but is not certain in point of language; I record it as an association.

The state of the word in the verse: min sharri — sharr stands here in the position of muḍāf (the first term of a construct), that is, it is always the evil of something. The sūra does not say abstractly "I take refuge from evil"; each time it binds the evil to a source: the evil of what He created, the evil of the ghāsiq, the evil of those who blow, the evil of the envier.

This is one of the most important features of the sūra's word order, and it has a theological consequence: evil is not a being that stands on its own. It is not an opposing god, an independent power, a principle of evil existing in its own right. It is always the quality of something, an effect issuing from something. In the world into which the Qurʾān came down the duality of a god of good and a god of evil was a living form of belief; the linguistic structure of the sūra leaves no room for that duality.

Two readings of "mā". In Arabic mā can perform more than one function, and two readings have been made of the mā here:

1. Mā mawṣūla (the relative pronoun): "From the evil of the things He created." The evil is ascribed to created beings. This is the preference of the majority.

2. Mā maṣdariyya (the particle of the verbal noun): "From the evil of His creating." On this reading the evil is bound to the act of creating. This reading is mentioned in the classical sources but is not preferred, because it is theologically problematic and is forced in point of language as well.

I take the first reading as my basis; I merely record the second.

There is a debate about scope as well. It is transmitted from some early exegetes that the expression "mā khalaq" means Iblīs and his offspring in particular. This is a reading that narrows the expression, and it has remained a minority one. The wording of the word is general; to narrow it requires evidence.

At first sight this verse seems to say something disturbing: evil is inside what has been created. And the one with whom refuge is taken is the one who created what has been created.

It would not be right to pass this matter over. But to settle it is not within the competence of this text either — it is one of the longest debates in Islamic thought, and it will not be closed by a few paragraphs given here. What I can do is to show what the language of the verse says and does not say.

First: the verse does not say "from the evil He created". The construct is min sharri mā khalaqa: "from the evil of the thing He created". The evil is named not directly as the object of the act of creating but as a state, a work of what has been created. This may look like a fine difference, but in Arabic it has been read as a deliberate choice.

Second: the Qurʾān observes a consistent manner in this matter. Evil is not ascribed to God directly and in an affirmative sentence. A few examples:

• On the lips of the jinn: "We do not know whether evil is intended for those on earth, or whether their Lord intends right guidance for them" (al-Jinn 72/10). Note: for the evil the passive ("is intended") is used, for the guidance the active ("their Lord intends"). Two different voices in the same sentence.
• On the lips of Abraham (Ibrāhīm): "And when I fall ill, it is He who heals me" (al-Shuʿarāʾ 26/80). The illness to the first person, the healing to God.
• "Whatever good comes to you is from God; and whatever evil comes to you is from yourself" (al-Nisāʾ 4/79). When this verse is read together with the one before it (4/78: "all is from God"), it is seen that the matter is taken up on two separate planes: on the plane of creation everything is from Him, on the plane of responsibility evil is from the human being.

Third: the main positions in the tradition of kalām. Without going into detail, in order to show where the debate is knotted:

I give this table not in order to settle the debate but in order not to present a single view as though it were the only view. I state no preference.

Fourth — and this is what the sūra really says: the verse does not offer a theodicy (a defence in the face of evil). It does not explain why evil exists. It accepts that evil exists and says what is to be done in the face of it: take refuge.

This is an attitude not very widespread among religious texts in general. The sūra does not deny evil, does not make it small, does not pass it off by saying "there is really a good in it", does not say "your point of view is wrong". Evil is real, it has names, it is counted one by one. And a single thing is proposed: to go to the owner of the source.

The second verse covers all evils. There is no evil that falls outside the expression "the evil of the things He created". Why, then, are verses 3, 4 and 5 needed?

In rhetoric this manner is called "the mention of the particular after the general": after a general statement has been made, particular instances falling within that statement are named separately. This is not repetition; it is the drawing of attention. What is named separately has an importance, a danger, or a likelihood of being forgotten.

The Qurʾān uses this manner in other places as well — as in speaking of the angels and then naming Gabriel (Jibrīl) separately.

The function here is this: a general expression can equally relieve a person and paralyse him. The sentence "I take refuge from the evil of everything" is true but abstract; it points to nothing. After it the sūra counts three concrete evils and so makes the abstract fear recognisable. A fear that has no name cannot be managed.

And the choice of the three instances is not random. We shall take this up as a whole after the fifth verse.

Wa min sharri ghāsiqin idhā waqab "And from the evil of the darkness when it settles."

Root: غ-س-ق. The root has two wings of meaning, and both occur in the Qurʾān:

1. Dense darkness, the pressing down of the night's darkness. Ghasaqa'l-layl — the night grew dark, the darkness settled. In the Qurʾān: "Perform the prayer from the declining of the sun to the west until the darkness of the night (ghasaqi'l-layl)" (al-Isrāʾ 17/78). Here ghasaq denotes the stretch of time running towards the darkest point of the day.

2. A thick, heavy, flowing thing. From the same root, ghassāq occurs in the Qurʾān as the drink of the people of Hell (Ṣād 38/57; al-Nabaʾ 78/25) and is explained in the commentaries with senses such as pus, or an intensely cold and foul-smelling liquid.

The place where the two senses meet is important: a density that flows, that seeps, that spreads. In the conception carried by this Arabic root, darkness is not something that "is" but something that flows. Not the cutting off of light but the pouring out of darkness. Like a liquid it fills the hollows, the empty spaces, the gaps.

Ghāsiq is an active participle: "that which darkens, whose darkness settles". No single English word carries it; renderings such as "the darkening one" or "when the darkness falls" are used.

Root: و-ق-ب. The concrete sense of this root explains the whole of the matter.

Waqb — a pit, a hollow, a recess. A hollow in a rock, the mouth of a well. And in Arabic the socket of the eye is called waqb al-ʿayn. The word, then, first names an emptiness.

The verbal form waqaba, for its part: to enter that emptiness, to sink into it, to seep in, to fill it. Waqabati'sh-shams — the sun set, that is, it withdrew into, entered, its place below the horizon.

The verb occurs in the Qurʾān only in this verse. (I am not 100% certain of this finding; it does not rest on a search but on my memory — it should not be taken as a definite claim.)

What does the choice of the word do? It could have said "when the night comes" (idhā jāʾa), it could have said "when the darkness presses down" (idhā ghashiya). Instead, the verb of seeping into a hollow has been chosen.

This describes how the darkness comes: it does not enter by the door, it seeps in through the gaps. Not all at once, but little by little. By the time you notice it, it is already inside. The moment at which the darkness comes cannot be fixed; one cannot say at a particular minute "there, now it has become dark". Seeping is exactly such a movement.

And one step further: what seeps in seeps into an emptiness. Waqb is a hollow. Darkness fills not a full place but an empty one.

I do not present this as a deliberate implication of the verse — it is an association I have made myself. But because the concrete sense of the word stands here, I record it.

Because the word comes indefinite, its scope is wide, and more than one rendering has been given in the classical commentaries:

On the view "when the moon is eclipsed". This view is related often and is sometimes presented as though it were certain. The honest position is this: such a report has been transmitted, it is found in the sources, but by itself it does not determine the lexical meaning of the word. Moreover there is no tension between what is said in the report and the wording of the sūra: the moon too is a part of the night, and when it darkens it leaves the night wholly dark. The report, that is, can also be read as showing an instance of the sense "night" — and some of the exegetes have read it so.

That the eclipse or the darkening of the moon was counted an ill omen in that period is a known thing. The general attitude of the Qurʾān and the sunna is to reject the loading of ill omen onto events in the sky. To read the report as meaning "the moon is an ill omen" therefore strains both the wording of the report and this general attitude.

I state a preference, and I record that it is not binding: the word's coming indefinite shows that it should not be nailed to a single object. "The darkness of night" is the original sense; renderings such as the moon, the star, the snake are instances of that original sense, not alternatives to it.

The verse does not say "from the evil of the darkness"; it says "from the evil of the darkness when it settles". This restriction occurs twice in the sūra (here and in verse 5), and it is important.

Darkness is not in itself evil. The Qurʾān names the night as a favour: "He made the night a rest" (al-Anʿām 6/96 — in the same verse, immediately after the expression "the one who splits the dawn"); "We made your sleep a rest, and We made the night a covering" (al-Nabaʾ 78/9-10; the same pair occurs in al-Furqān 25/47 as well).

The night, that is, is an ordered and good part of creation. The evil lies not in the night itself but in a particular state of the night and in the possibilities that state brings.

This has a concrete counterpart. Darkness is a covering that makes harm possible: it gives the possibility of hiding, it cuts off sight, it makes defence harder, it enlarges fear. Throughout history the night has been the time of theft, of raids, of attack, of getting lost. It is also the time when illnesses grow worse, when pain increases, when loneliness is felt most — this is a universal part of human experience, and the verse rests on that experience.

I draw no scientific claim from this. I state only that the image the verse builds rests on real human experience.

Wa min sharri'n-naffāthāti fi'l-ʿuqad "And from the evil of those who blow on the knots."

Root: ن-ف-ث. To send air out of the mouth; but in a particular manner.

In Arabic three degrees of blowing are distinguished, and the books of the language record this distinction regularly:

In the verse the middle one has been chosen. This shows the physical character of the work described: a work done with voice and breath, but one that does not remain entirely "in the air". An operation in which a recited word is thought to be transferred onto an object by being blown.

The word has other uses as well. It is used of a snake's spraying venom. And of poetry: the Arabs would describe the poet's inspiration with nafth — something being blown inward. It is transmitted that in a saying ascribed to the Prophet in the ḥadīth sources the expression "the Holy Spirit blew into my heart (nafatha fī rawʿī)" occurs. The root, then, can be used of a good transmission too; the negative charge in the verse comes not from the word itself but from the context.

The pattern: نَفَّاثَات — it is the plural of the faʿʿāla pattern. In Arabic this pattern conveys intensification: one who does a thing much, who has become identified with that work. Like ʿallāma (one who knows much), raḥḥāla (one who travels much). The verse, that is, describes not one who blows once but one who has made this his occupation.

Root: ع-ق-د. Its concrete sense is to tie a knot, to bind, to tie two ends to one another. ʿUqda — knot; its plural is ʿuqad.

The derivatives of the root show how central a concept the verb "to bind" is in Arabic:

• ʿaqd (عقد): a contract. The binding of two parties to one another. In the Qurʾān: "O you who believe! Fulfil the contracts" (al-Māʾida 5/1). It is used of the "binding" of an oath as well (al-Māʾida 5/89).
• ʿaqīda (عقيدة): belief. The thing to which the heart binds itself.
• iʿtiqād (اعتقاد): to believe; to be knotted to something.
• ʿuqda used figuratively: a knot, a matter that cannot be untied. It occurs in Moses's prayer: "Loosen the knot in my tongue" (Ṭāhā 20/27).

In Arabic, then, "being bound" is at once a physical, a legal and a doctrinal concept. The word the verse chooses stands inside this network.

The verse points to a concrete practice: tying knots in a cord, and reciting something over each knot and blowing on it. This was a form of magic known in pre-Islamic Arabia and over a wider geography. The knot represents the symbolic bond established with the person targeted; the blowing is the loading of the word onto that bond. The untying of the knot was counted as the lifting of the effect.

There is historical information about the practice itself; its details and its regional forms are the subject of the history of religions and of ethnography.

The word has come as a feminine plural. This is one of the points most dwelt on, and the explanations transmitted are these:

A caution is needed. The second, third and fourth explanations show that the word's coming feminine may arise from a grammatical necessity or from intensification. These explanations are found in the classical commentaries; they are not excuses invented afterwards.

And as a matter of principle this needs to be stated: the verse issues no ruling about a sex; it issues one about an act. This is the general method of the Qurʾān — the attribute is described, and whoever bears that attribute falls within the ruling. The other three items of evil in the sūra (the evil of created things, the darkness, the envier) are defined by way of an act and a state as well; here too what is defined is the work.

There are also widening readings made in later periods: "blowing on knots" has been understood as corrupting ideas, decisions and relationships by secretly blowing words onto them. I do not present this reading as the literal meaning of the verse — the verse speaks of a concrete practice. But the image the word carries (working with words, in secret, upon a bond) lends itself to this widening, and I shall return to it in the section "What this says to the present".

Three particular evils are counted in the sūra, and two of them are indefinite while one is definite:

• ghāsiqin — indefinite
• ٱلنَّفَّٰثَٰتِ — definite
• ḥāsidin — indefinite

Why is this so? Two grounds stand out in the classical explanations:

1. A pointer to a known practice (ʿahd). In Arabic the article al- is used to point to a particular thing the addressee knows. The first addressees knew this practice; what is said is "those blowers".

2. Darkness and envy are general states any person may meet with; this, on the other hand, is an act belonging to a particular group. The general ones have come indefinite, the particular one definite.

This second ground is my own reading; the first occurs in the classical sources.

This verse is one of the central texts of the debate about magic in Islamic thought. Rather than issuing a sweeping ruling on the subject, it is more correct to show the ground of the debate.

What does the Qurʾān say about magic?

The Qurʾān speaks of magic in several places, and what it says can be gathered under these three points:

1. Magic is named as a fact. al-Baqara 2/102 is the most detailed verse on the subject: it speaks of what was sent down to Hārūt and Mārūt in Babylon, and says that people learned "things by which they might part a man from his wife".

2. But the same verse draws the limit of the effect. In the same verse there is this restriction: "Yet they could harm no one by it except by God's leave." This sentence coincides exactly with the whole logic of the sūra: evil is not an independent power. Magic constitutes no exception.

3. Part of magic is the deceiving of the eye. In the narrative of Moses: "Their ropes and their staffs, because of their magic, appeared to him to be moving" (Ṭāhā 20/66) — the key word here is yukhayyalu, "it was made to appear". And: "They bewitched the eyes of the people" (al-Aʿrāf 7/116). And for the end of magic: "God will surely bring it to nothing" (Yūnus 10/81).

The disagreement: the reality of magic.

It is said that the general tendency of the Muʿtazila is close to the second view, and that the majority of the Sunnī theologians incline to the first. I state no preference.

As for its legal side: the ruling on working magic is a separate topic in the books of fiqh, and there are differences of weight among the schools of law. No legal ruling is issued in this text; it can only be recorded that the tradition unanimously condemns the working of magic, and that it is counted neither an act of service nor a legitimate occupation.

What the verse itself says, on the other hand, is narrow and clear: refuge is taken with God from this. The sūra does not describe magic, does not classify it, does not teach a counter-technique. It points in a single direction.

Wa min sharri ḥāsidin idhā ḥasad "And from the evil of an envier when he envies."

Root: ح-س-د. The definition of the word is quite precise in the classical sources: to want a favour that someone has to go from him.

This definition has two parts, and both are necessary: 1. Someone else has a favour. 2. The envier wants that favour to be taken from him.

Without the second condition there is no envy. This separates the concept from a neighbouring one:

This distinction is drawn regularly in the classical books of ethics, and it is important: to covet what another possesses is not envy. Envy is the name not of coveting but of grudging.

It is transmitted that in some dictionaries the root is linked with the name of a parasite that clings to the skin and sucks blood (the tick). This linkage offers an attractive image — envy's clinging silently and consuming from within — but it is not certain in point of language; I record it as a transmitted report, not as a ground.

The first victim of envy is the envier. This is not something the Qurʾān says directly; it is a finding often voiced in the tradition of ethics, and it is reasonable: because the envier constantly measures his own lack against what another has, he wounds himself without cease. The favour is with the other, the pain is with the envier.

This restriction is one of the most important fine points of the sūra, and it is often passed over.

The verse does not say "from the evil of an envier"; it says "from the evil of an envier when he envies". To say "when he envies" of someone who is an envier already looks at first sight like an unnecessary repetition. It is not.

Two main explanations, transmitted and defensible:

1. As long as envy stays within, its harm is to the one who holds it; when it moves into action it reaches another. A feeling that passes through a person does nothing to another by itself. The evil appears at the moment that feeling breaks out, when it turns into a word, an obstruction, a trap, a look. The restriction marks exactly this moment of transition.

The moral consequence of this explanation is important: you are responsible not for what passes through your heart but for what you do. This is the general line of the Qurʾān and of the tradition — a person is called to account not for inner states that come of themselves and over which he has no mastery, but for what he does by his will.

2. The restriction states the case in which envy actually occurs. That is, not the tendency to jealousy present potentially in every person, but the case in which that tendency actually turns into envy. There is a difference between something stirring inside a person and that something being envy.

A third explanation is transmitted as well: the restriction points to envy's coming out by way of the eye (naẓar, ʿayn). In the tradition a link is made between envy and "the striking of the evil eye"; reports are transmitted in the ḥadīth sources to the effect that "the eye is real" (al-ʿaynu ḥaqqun). The views that make this link are found in the classical commentaries. I do not go into the detail of the subject; what must be recorded is that the wording of the verse says nothing of an "eye", and that this is an interpretation.

The Qurʾān uses the word "envy" in several places:

• "Many of the People of the Book would like to turn you back into unbelief after your faith, out of the envy in their souls (ḥasadan min ʿindi anfusihim), after the truth has become clear to them" (al-Baqara 2/109). The expression min ʿindi anfusihim here — "arising from within themselves" — is important: envy issues not from an outward cause but out of the person's own inside. And note: the verse says not "the People of the Book" but "many of the People of the Book" — here too the Qurʾān lays no wholesale ruling upon a group but describes an attitude.
• "Or do they envy people for what God has given them of His bounty?" (al-Nisāʾ 4/54). The object of the envy here is explicitly "what God has given". This brings out the hidden logic of envy: envy is to appearance directed at a person, but in reality it is an objection to the distribution itself. Without realising it, the envier calls the apportionment into question.

And two narratives the Qurʾān tells show what envy can do:

The two sons of Adam (Ādam). "Recite to them the account of the two sons of Adam, in truth: when the two of them offered a sacrifice, and it was accepted from one of them and not accepted from the other. He said: I will surely kill you" (al-Māʾida 5/27 and what follows).

Note: the Qurʾān does not give the names of these two; it says "the two sons of Adam". The names Cain and Abel do not occur in the Qurʾān; they come from other sources. This needs to be stated.

The structure of the narrative is the exact description of envy: a favour (the acceptance of the sacrifice) has been given to the other; the one who could not receive the favour turns to doing away with the one who has it. The point to notice: the one who did not receive the favour chooses not to put his own sacrifice right but to destroy the other. Envy prefers lowering the other to raising itself; because the matter is not the favour but the comparison.

The brothers of Joseph. "When they said: Joseph and his brother are dearer to our father than we are, though we are a company" (Yūsuf 12/8). Here too the word "envy" does not occur, but the structure is the same: a favour (the father's love) is with the other, and this becomes unbearable through the comparison "we are more and we are stronger". The outcome is the well.

What the two narratives have in common is striking: in both cases the envy comes from a brother. From the nearest one. This is important for the ordering of the sūra, and we shall return to it in the next section.

In the tradition Iblīs's not prostrating is often bound to envy as well. The ground the Qurʾān records is pride — "I am better than he; You created me from fire and created him from clay" (al-Aʿrāf 7/12) — but the interpretation that behind this ground there lies a "grudging" is widespread. It needs to be stated that this is an interpretation; the wording of the verse does not say "envy".

The most striking side of the sūra is the order of the three instances. This order is not random, and it advances on several axes at once.

At every step the danger comes nearer. The darkness is outside; magic is worked in an unknown place by an unknown person; envy, on the other hand, is at the other end of the table.

The darkness has no intention towards you; it merely settles. In magic there is an intention, but the target is sometimes indeterminate. In envy, on the other hand, the intention is direct, addressed and personal — the envier knows you, looks at you, looks at the particular thing you possess.

You see the darkness (or rather, you know that it is dark). You do not notice magic, but it may leave a trace. Envy you neither notice nor does it leave a trace — most often even the one who envies is not aware of his own envy. A person generally finds other names for his own envy: a sense of justice, being in the right, criticism, concern.

These three axes look in the same direction: evil is dangerous not as it grows more distant but as it comes nearer. Harm that comes from a stranger can be defended against; harm that comes from someone familiar cannot, because no defensive position is taken up.

And here the sūra stops. It counts no fourth evil. It is as though one had come as near as possible and had finished there.

But it has not finished. The next sūra will go to a place nearer still than the envier: inside the person's own breast.

Envy, the last item of al-Falaq, is the nearest evil that can come from outside. The whispering, the single item of al-Nās, does not come from outside — it speaks within. The transition between the two sūras takes place at exactly this point.

Let us look at the last letters of the five verses of the sūra: ق ق ب د د (falaq — khalaq — waqab — ʿuqad — ḥasad).

In Arabic these five letters have a feature in common: all of them are among the letters of qalqala. Qalqala (ق ط ب ج د) means that, when one stops on them in the state of sukūn, the sound comes out with a bounce, with a knock. When you stop on a word ending in these letters, the sound is not cut off flatly; it ends with a small stroke, with a skip.

All five verses of the sūra, that is, close with a stroke. A cut-off, hard, striking sound. The recitation of the text carries the sound of the thing it describes — of the blows that come from outside.

I record this finding not as a claim of a "miracle" but as an observation about the sound weave of the text. But it will become far more meaningful when it is compared with the sound texture of the next sūra: all the verses of Sūrat al-Nās end in س — a fricative, lengthening, hissing sound. The two sūras describe two different kinds of threat with two different sounds.

The report most often named as the occasion of revelation of these two sūras is this: a man of the Jews called Labīd b. al-Aʿṣam worked magic on the Prophet; knots were tied in hairs caught in a comb and it was left in a well; the Prophet was affected for a time, then the state of affairs was made known to him, the knots were brought out, and as al-Muʿawwidhatān were recited (or as they were sent down) the knots were untied one by one.

One must be honest about the standing of this report; and the matter must be divided into two separate parts.

The first part: the account that magic was worked on the Prophet. This account is found in reports transmitted from ʿĀʾisha in those ḥadīth sources counted the most authoritative — in al-Bukhārī and Muslim. It has therefore been accepted as strong in point of its chain.

The second part: that these two sūras came down upon this event and that each verse untied a knot. This detail is not a part of the reports of the two Ṣaḥīḥs; it comes by other routes counted weaker. Details such as "eleven knots" and "with each verse a knot was untied" are found in transmissions that are disputed in point of their chains.

The importance of this distinction is this: that a part of a report is counted sound does not mean that the whole of it is sound. What is presented as the occasion of revelation is precisely the second part, and that part is the weakest. Indeed the majority's counting these sūras Meccan shows that they do not accept this report as the occasion of revelation.

The first part, that is the account of the magic itself, has not been accepted without dispute either. The objection comes from this point:

The Qurʾān records the expression "a bewitched man" as a slander of the mushrikūn. It occurs in two places:

"The wrongdoers said: You are following none but a bewitched man." (al-Furqān 25/8)

"…The wrongdoers said: You are following none but a bewitched man." (al-Isrāʾ 17/47)

The Qurʾān transmits this saying and rejects it. According to those who raise the objection, to accept that the Prophet was affected by magic is to confirm this slander of the mushrikūn.

Among those who voice this objection are the theologians of the Muʿtazila and, in later periods, some scholars. The objection is generally rested on the principle of ʿiṣma (the protectedness of the prophets).

The answer of those who defend the report, for its part, has run along this line: the effect of the magic concerns the Prophet's bodily and worldly states — such as supposing he had done a thing when he had not. It had no effect whatever on the revelation, on its delivery, or on the office of prophethood. The claim of the mushrikūn, on the other hand, means "his mind has gone; his speech is not his own, he speaks under the effect of magic"; what is related in the report is not the same thing. Prophets too are human beings; they fall ill, they are wounded, they suffer pain — this does no harm to ʿiṣma.

The commentators on ḥadīth have elaborated this defence. I ascribe no particular sentences to particular names, because it would not be right for me to speak with certainty about exactly what which scholar said without consulting the source.

The conclusion I record is this, and it is not binding:

1. The account of the magic itself is found in strong sources; it cannot easily be rejected. 2. The claim that the sūras came down upon this event, on the other hand, is weak; it does not suffice for saying "the sūra came down upon this". 3. When the debate is carried into the field not of the methodology of ḥadīth but of kalām, it has not been settled, and it continues today as well. 4. And most important of all: the understanding of the sūra does not depend on this report. The text is entirely intelligible even if no occasion of revelation is known.

There is another historical matter that honesty requires be recorded. In some sources it is transmitted that Sūrat al-Falaq and Sūrat al-Nās were not present in the private codex of ʿAbdallāh b. Masʿūd, one of the Companions.

What has been said about this transmission can be summarised as follows:

• There are sources that relate the transmission; the account itself is on record.
• Some of the scholars have rejected the soundness of the account.
• Some of those who accept it have interpreted it thus: Ibn Masʿūd did not deny that these sūras are Qurʾān; he may not have written them into his own copy because they were used as prayers of refuge and were in any case memorised by everyone.
• The consensus of the community on the text of the Qurʾān, and these two sūras' having been transmitted by mass transmission (tawātur, uninterrupted and widespread transmission), have been counted the real proof that closes the matter.

I write this not in order to turn the subject into a problem but in order not to conceal it. The Islamic scholarly tradition recorded this account in its own sources and debated it; to ignore it would run counter to the tradition's own attitude.

First: naming evil. The first thing the sūra does is to count the evil. This stands over against two attitudes that are widespread today.

The first attitude is to deny evil: "everything is really good", "do not think negatively", "change your point of view". This isolates the person who is living through the pain once more — because he is being told that what he sees is not real.

The second attitude is to freeze in the face of evil: the world is full of dangers, there is nothing to be done.

The sūra does neither. It counts the evil — four times, one by one, by name. And precisely because it counts it, it can point to a place over against it. A fear that has not been named cannot be managed; a fear that has been named can be directed.

Second: to take refuge is not to take no precaution. The sūra says "I take refuge"; it does not say "I do nothing". The general line of the Islamic tradition is that trust in God does not take the place of precaution. One who takes refuge from the evil of the darkness locks his door as well. These two things are not alternatives to one another; taking refuge begins where precaution cannot reach — and there is no place precaution can reach such that everything up to it is secured.

Third: the present-day counterpart of "naffāthāt" — and two errors.

Tying knots in a cord and blowing on them is a practice that has no place in most people's lives today. But the structure the verse describes — doing harm in secret, with words, by working upon a bond — has not aged at all.

To blow words onto a relationship: to leave words, one by one, between two people without either of them seeing. To blow words onto a reputation: to form a shadow that no one asserts directly but everyone has heard of. To blow words onto a decision: to plant in the person who is to decide, right up to the moment of decision, suggestions that are indirect rather than direct. What all of these have in common is the two elements in the verse: the word and secrecy.

This is not an exegesis but an application; I do not claim that the wording of the verse means this.

And here two errors are to be avoided. The first is to reduce the verse entirely to metaphor and to disregard its concrete sense. The second — and by far the more widespread today — is to attach every setback in one's life to magic, to the evil eye, to hidden enemies. This second looks at first sight like piety, but its function is exactly the opposite: it produces an escape from responsibility. The person whose work is not going well does not look for his own mistake; the person who built the failed relationship does not ask about his own share. The sūra teaches taking refuge from evil; it does not teach handing off one's own share.

Fourth: envy and comparison.

Envy is born not from the other's having something but from measuring. You and him. And in order to measure, the two of you must be laid against the same rule.

This existed in the world into which the sūra came down as well — the tribe, the neighbour, the brother. But the possibility of measuring was limited: with how many people could you compare yourself? Today the possibility of comparison is practically unlimited. A person can see the best moments of thousands of people he does not know side by side with his own ordinary moment. This did not create envy, but it opened the way before it.

The restriction in the verse is of use here: idhā ḥasad — when he envies. A pang passing through you does not make you an envier. The difference lies in what you do with that pang: do you turn it into a failure to rejoice, a belittling, an obstruction, a bad word; or do you notice it and let it go.

And the finding in al-Nisāʾ 4/54 is to be remembered: envy is not directed at a person, as it appears to be. It is directed at the apportionment. The real question facing a person who notices his own envy is this: am I angry at this person, or at this share having been given to me?

Fifth: being the one who is envied. The sūra teaches taking refuge from the one who envies. There is something it does not teach: diagnosing, marking out, announcing the one who envies. The verse describes an attribute; it names no one. The sentence "so-and-so envies me" is easily constructed, and the moment it is constructed it begins to produce an enemy in the person's own mind. The movement the sūra teaches is not outward but upward.

Sixth: the knowledge that the darkness will be split. The sūra takes refuge with the Lord of "al-falaq" — that is, of the one who splits the darkness and brings the morning out of it. The choosing of the name in this way places a piece of knowledge inside the taking of refuge: the darkness is not permanent.

This is not a consolation of the "everything will turn out fine" kind. It is an observation: splitting is the way creation works. The seed is split, the night is split, the sea is split. The opening of what looks closed is the rule, not the exception. The sūra gives this not as a promise but as the presentation of the one with whom refuge is taken — which is stronger: no word is given to you; you are told in whose hand you are.

• Whether the sūra is Meccan or Medinan is disputed; no preference was stated.
• The views about which concrete sense the word al-falaq has were given in a table; none was presented as certainly correct. It was stated that the sense "morning" is the strongest in point of language, but this is not a binding preference.
• The relation between فلق and فلح is not a claim that they are one root. The two roots are separate. The link that is made is conceptual and is my own reading; it has not been presented as a transmitted report.
• It was said that the verb waqaba occurs in the Qurʾān only in this verse; this finding rests on memory, not on a full search.
• The report stating that ghāsiq means "when the moon is eclipsed" was related, but its source was not named with certainty and no ruling was issued about the chain of the report.
• On the question of the creation of evil the positions of the schools of kalām were given in very general outline; no detailed view was ascribed to any of them and the debate was not settled.
• The disagreement about the reality of magic was given in a table; no preference was stated. No legal ruling was issued.
• All the explanations about al-naffāthāt's coming as a feminine plural occur in the classical sources; no preference was stated as to which of them is correct.
• The linking of the root of envy with the "tick" was related as a transmitted report; it was stated that it is not certain in point of language.
• In the report of Labīd b. al-Aʿṣam it was said that the account of the magic is found in the two Ṣaḥīḥs; it was stated that the detail "with each verse a knot was untied" comes by weak routes. No sentences were ascribed by name to the scholars who defend or reject the report.
• The account concerning Ibn Masʿūd's codex was related; no ruling was issued about its soundness.
• The finding about qalqala is an observation about sound; it has not been presented as a theological claim.