Abrogation (Naskh): When the Eternal Word Revises Itself

For some sixteen or seventeen months after the migration to Medina, Muhammad and his community prayed facing Jerusalem. Then, according to Islamic tradition, in the middle of a prayer, the revelation came that changed the direction: “We have certainly seen the turning of your face, [O Muhammad], toward the heaven, and We will surely turn you to a qiblah with which you will be pleased. So turn your face toward al-Masjid al-Haram”—the Sacred Mosque in Mecca.¹ A mosque in Medina is still called Masjid al-Qiblatayn, “the mosque of the two qiblahs,” because the worshippers there are said to have pivoted, mid-prayer, from one sacred direction to the other.

It is a vivid scene, and it sits at the heart of one of Islam’s most fascinating and least-understood doctrines: naskh, abrogation—the principle that a later word from God can cancel an earlier one. The qibla change is the gentle, almost beautiful version of the doctrine. But naskh reaches further than the direction of prayer. On the classical theory, it governs how a community grew from sobriety-at-prayer to total prohibition of wine; how it moved from patience under persecution to permission for war; and—most startlingly—why certain verses that Muslims believe were once recited as Quran are not in the Quran today. The question this essay asks, as fairly as it can, is whether a book that presents itself as the eternal, uncreated, perfectly consistent speech of an unchanging God can also be a book that revises its own rulings—and what it means that Islam’s own scholars never agreed how often it does.

Stating the Doctrine Carefully

First, what naskh is and is not. It is not a Christian invention, nor a polemical charge smuggled in from outside. It is a mainstream feature of Islamic jurisprudence, grounded in the Quran itself, and developed into a formal science—‘ilm al-nasikh wa al-mansukh, the science of the abrogating and the abrogated. To discuss it is not to attack Islam; it is to take seriously a doctrine Muslims themselves built and have debated for fourteen centuries.

The textual root is two verses. The first is Surah 2:106: “We do not abrogate a verse or cause it to be forgotten except that We bring forth [one] better than it or similar to it. Do you not know that Allah is over all things competent?”² The second is Surah 16:101: “And when We substitute a verse in place of a verse—and Allah is most knowing of what He sends down—they say, ‘You, [O Muhammad], are but an inventor [of lies].’”³ Notice that the second verse already anticipates the objection: critics in Muhammad’s own day apparently said that a God who replaces his own words looks like a man making things up as he goes. The Quran’s answer is that the substitution is itself an act of divine knowledge and competence—God knows what he sends down.

From these verses the classical jurists derived three recognized types of abrogation, and the distinctions matter a great deal for what follows.⁴ First, naskh al-hukm wa al-tilawa: abrogation of both the ruling and the recitation—a verse whose wording and legal force are both withdrawn, so it survives in neither the text nor the law. Second, naskh al-hukm duna al-tilawa: abrogation of the ruling but not the recitation—the words remain in the Quran you can hold in your hand, but the command they express is no longer in force. Third, and most striking, naskh al-tilawa duna al-hukm: abrogation of the recitation but not the ruling—a verse whose wording has vanished from the text, yet whose legal obligation is held to remain binding. The mere existence of these last two categories is what makes naskh more than an academic curiosity. The second means the written Quran contains commands Muslims believe God has retired; the third means the law contains obligations that, on Islam’s own account, were once Quran and no longer appear in it.

How the Doctrine Developed

The science of naskh did not spring fully formed from the text. Modern scholarship—notably the careful work of John Burton, whose The Sources of Islamic Law: Islamic Theories of Abrogation (1990) remains the standard study—argues that abrogation theory was elaborated by later jurists wrestling with two practical problems.⁵ The first was internal: certain Quranic passages appear, on the surface, to pull against one another, and a tool was needed to say which ruling governs. The second was external: Islamic law as practiced contained rules—the stoning penalty for adultery is the classic example—that have no clear basis in the Quranic text as we have it. Naskh, especially the category of “abrogated recitation, surviving ruling,” could explain how a law could be Quranic in origin even when its words were no longer on the page.

This is an important and contested historical judgment, and it should be labeled as such. Burton’s thesis—that naskh functioned partly as a back-formation to reconcile law and text—is a Western academic reconstruction, and traditional Muslim scholars would say instead that the categories were transmitted from the Prophet and his Companions and merely systematized later. Both sides agree on the data; they disagree on the direction of explanation. What is not in dispute is that by the classical period an enormous literature on al-nasikh wa al-mansukh existed, and that its leading authorities could not agree on its scope.

The Famous Cases

The doctrine is best understood through its standard examples, told fairly. The qibla change (2:142–144) is the textbook instance of abrogated ruling: the obligation to face Jerusalem gave way to the obligation to face Mecca, and the tradition treats this as God’s deliberate redirection, not his reversal of an error.

The prohibition of wine is the favorite illustration of naskh as gradual legislation. The Quran addresses intoxicants in stages: 2:219 concedes that wine and gambling hold “great sin and [yet] some benefit for people, but their sin is greater than their benefit”; 4:43 forbids approaching prayer “while you are intoxicated”; and 5:90 finally declares intoxicants, gambling, and divining arrows “defilement from the work of Satan, so avoid it.”⁶ Many Muslims point to this sequence as a model of divine pedagogy: a community is weaned off an entrenched practice by degrees rather than by an impossible overnight command.

The hardest case is the “verse of the sword,” Surah 9:5: “And when the sacred months have passed, then kill the polytheists wherever you find them and capture them and besiege them and sit in wait for them at every place of ambush.”⁷ Some classical exegetes held that this single verse abrogated a great many earlier passages counseling patience, forbearance, and even tolerance—by one widely repeated (and much-disputed) reckoning, more than a hundred peaceable verses.⁸ It must be said immediately, in fairness, that this maximalist reading is itself contested within the tradition: the immediate context exempts polytheists who keep their treaties (9:7) and commands safe-conduct for those who seek it (9:6), and many scholars—especially in the Hanafi tradition and among modern Muslims—tie the verse to its specific occasion, the Arab idolaters who had broken their pacts, rather than reading it as an open-ended cancellation of the peace verses.⁹ The point for our purposes is not to settle Islamic ethics but to notice that abrogation could be invoked to let a later, harsher command override earlier, gentler ones—and that even the application of the doctrine here is internally disputed.

Finally, the cases of naskh al-tilawa—abrogated recitation—are the most arresting, because they concern verses no longer in the Quran. The best-attested is the so-called “verse of stoning.” In a sermon preserved in Sahih al-Bukhari and Sahih Muslim, the caliph ‘Umar declared that God had sent down a verse prescribing stoning for adultery, that the early community recited and acted on it, and that he feared people would one day deny it was ever revealed.¹⁰ On the classical account, its wording was abrogated while its ruling remained—the textbook example of the third category. (These “lost verse” reports are treated in depth in our companion essay on the Quran’s preservation; here they matter only as instances of naskh.)

The Internal Tension

Now the difficulty can be stated—and it should be stated as a real tension, not a knockout. The Quran makes two strong claims about itself that abrogation strains. It insists on its own perfect consistency: “Then do they not reflect upon the Qur’an? If it had been from [any] other than Allah, they would have found within it much contradiction” (4:82). And it insists on the fixity of God’s words: “None can alter His words” (6:115; cf. 10:64).¹¹

Abrogation does not produce a flat logical contradiction with these verses—the doctrine is precisely the tool designed to prevent that. But it does exert pressure on the natural sense of the words. A book that contains commands God no longer wills (category two) is, in a plain sense, a book some of whose contents have been superseded. And a doctrine whose whole job is to manage passages that would otherwise read as conflicting is, at minimum, an admission that the surface of the text contains tensions requiring resolution. The deeper theological knot lies in Sunni orthodoxy’s doctrine of the Quran as the eternal, uncreated speech of God. If the speech is eternal and the speaker unchanging, what does it mean for one eternal word to cancel another? The orthodox answer is subtle: God did not change his mind; he knew from eternity that one ruling would apply for a season and another thereafter. The cancellation is in the ruling’s temporal application, not in God’s eternal knowledge. Whether that fully dissolves the tension or merely relocates it is exactly the live question.

The Mainstream Muslim Account, at Full Strength

Before assessing, the strongest Muslim defense deserves to be laid out in its best form—because at its best it is carefully constructed, and a great many Muslims hold it sincerely.

The case runs like this. Abrogation is not God revising a mistake; it is God legislating progressively for a community that is itself changing. An unchanging, all-knowing God can will different obligations for different circumstances without any change in himself—just as a wise physician prescribes one regimen for the acute phase of an illness and another for recovery, without ever being inconsistent or uncertain. The wine sequence is the perfect illustration: a sudden, total prohibition of a deeply rooted social practice would have been unbearable and unheeded, so the law descended in stages, each fitted to the community’s growing capacity to obey. On this view naskh is not a bug in the revelation but a feature of it—evidence of divine wisdom and mercy, of a God who meets people where they are and leads them forward. Far from undermining the Quran’s perfection, abrogation displays it: the perfection of a law perfectly suited to its moment.

And the “contradiction” objection, the defender continues, misunderstands the categories. Two commands that apply at different times do not contradict each other any more than “fast today” and “do not fast tomorrow” contradict each other. The no-contradiction verse (4:82) concerns genuine inconsistency—a book at war with itself—not the orderly succession of rulings under a single divine purpose. The unchangeable-word verses (6:115, 10:64) concern the security of revelation against human tampering and falsification, not God’s own sovereign freedom to legislate in stages. To read them as forbidding naskh, the defender says, is to read them against their context and against the plain testimony of 2:106, where God explicitly reserves the right to bring a better verse in place of another. Christians, moreover, should feel the force of this: the New Testament itself teaches that the ceremonial law of Moses gave way to a new covenant (Hebrews 8–10). Progressive revelation is not foreign to biblical faith; the Muslim is simply claiming for the Quran what the Christian claims across the two Testaments.

That last point is fair and worth conceding without reserve. The bare idea that God may give different commands in different eras is not absurd, and a Christian who finds naskh self-evidently incoherent should remember Hebrews. The objection, if there is one, must be sharper than “God changed the rules.”

Weighing It Honestly

So where does the defense hold, and where does it strain? Three things should be said.

First, the progressive-legislation account works best for legal abrogation—category two, abrogated ruling. The wine example really is a reasonable pedagogy, and the physician analogy genuinely fits it. On this narrow point — gradual legal pedagogy — the Muslim case is strong, and the Christian should grant it. But the account fits far less comfortably the case of abrogated recitation—the category in which the very wording of a revealed verse is withdrawn. Progressive legislation explains why a command might change; it does not naturally explain why the eternal, uncreated text of a verse should be removed from the Book while its ruling stays behind. The stoning verse is not a softer law replaced by a sterner one for pedagogical reasons; it is a piece of claimed revelation that, on the tradition’s own telling, was once recited and is now absent. The physician analogy has no obvious purchase there.

Second, and most tellingly, the experts could not agree how much of the Quran is abrogated—and the range is staggering. Early authorities counted abrogated verses by the hundreds: Ibn Salama’s well-known list runs to over two hundred, and figures in the same neighborhood circulate in the classical literature.¹² The great polymath Jalal al-Din al-Suyuti (d. 1505), in his standard manual of Quranic sciences al-Itqan, pared the genuine cases down to around twenty.¹³ The eighteenth-century Indian reformer Shah Wali Allah of Delhi (d. 1762), in al-Fawz al-Kabir, argued the number could be securely established at only five, explaining away the rest.¹⁴ The modern Egyptian scholar Mustafa Zayd, in his detailed 1963 study al-Naskh fi al-Qur’an al-Karim, concluded after sifting every claimed instance that the real cases number a mere six.¹⁵ And a minority of modern Muslims—notably Ahmadiyya scholarship and various reformers—argue that the true number is zero: that intra-Quranic abrogation does not occur at all, and that 2:106 refers to the abrogation of previous scriptures rather than the Quran cancelling itself.¹⁶

Some of this spread is definitional—earlier scholars used “abrogation” loosely, lumping in specification, qualification, and the mere lapsing of temporary commands, which later scholars distinguished out. That concession is real and important; the gap from 200 to 20 is partly a tightening of terms, not pure chaos. But not all of it is definitional. Scholars genuinely disagreed about which verses are cancelled, and even about whether the doctrine applies to the Quran at all. That is substantive disagreement about the operative content of a book the Quran calls “a clear Book” (5:15) whose verses are “clarification for all things” (16:89). A doctrine on which sincere, learned Muslims range from “over two hundred” to “none” is not a settled feature of a self-evidently clear text; it is a centuries-long argument about which of the Book’s own commands remain in force.

Third, the honest critic must let the Muslim have the last word before the reply. The strongest rejoinder is this: every revealed tradition that spans time must reconcile earlier and later teaching, and Christianity is no exception—the relation of the Testaments is precisely an exercise in working out what abrogates, fulfills, or transforms what, and Christians disagree among themselves about it no less than Muslims disagree about naskh. The diversity of scholarly opinion, on this view, is not a defect peculiar to Islam but the normal cost of a long revelation interpreted by fallible humans. That is a serious and fair point.

The reply is that the parallel is real but not symmetrical, and the asymmetry is the whole point. Christianity locates the transition between covenants in history—in the coming of Christ, a public event around which the church openly organizes its reading of the older Scriptures, which it preserves entire and unaltered. The naskh tension is different in kind: it is internal to a single book delivered through a single man within roughly two decades, a book that simultaneously claims to contain no inconsistency and to consist of unchangeable words. The difficulty is not that God legislated progressively—Christians agree he may—but that this particular text makes maximal claims about its own seamless perfection while its own tradition documents rulings retired and verses withdrawn, and cannot agree on how many. The Christian claim is more modest at exactly the point where the Quranic claim is most absolute.

The Honest State of the Question

It is important not to overstate the case, and several things must be granted plainly. Naskh is a coherent, internally consistent doctrine; on its own terms the “lost” rulings and verses were withdrawn by God’s will, not lost by human failure, and it is unfair to describe this as the Quran being “corrupted.” Many Muslims minimize abrogation to a handful of cases or deny it outright, and that range is itself part of the honest picture: there is no single “Islamic position” on naskh to refute. The maximalist sword-verse reading is a minority view even within the classical tradition. And the bare principle of progressive revelation is one Christians share. These concessions are not throat-clearing; they are the difference between a fair argument and a caricature.

What survives the concessions is narrower and sturdier. A book advertised as the eternal, uncreated, perfectly consistent, unchangeable speech of God turns out, on the testimony of its own most revered scholars, to contain rulings that have been superseded and—on the classical account—verses whose wording has been withdrawn, with no agreement among the experts on how often either occurred. That is not proof of anything by itself. It is a genuine internal tension between the doctrine of abrogation and the strongest forms of the Quran’s self-description—a tension the tradition manages, but does not entirely dissolve.

Conclusion: What This Does and Does Not Establish

Naskh does not show that the Quran is incoherent, or that abrogation is irrational, or that Muslims are unaware of the issue—they have written libraries on it. It does not, by itself, disprove Islam. What it does is far more modest and, precisely because modest, harder to dismiss: it shows that the Quran’s most exalted claims about itself—flawless consistency, unchangeable words, perfect clarity—sit in real tension with a doctrine the tradition needed, built, and never settled. The eternal Word, on Islam’s own account, revises its rulings; sometimes it withdraws its very wording; and the people best equipped to count the changes could not agree whether there were two hundred or none. Put plainly: the maximal claim—a perfectly consistent, unchangeable, eternal Word—does not survive the tradition’s own doctrine of abrogation.

For the Christian, this is not an occasion for triumph but for an invitation, offered with gentleness and respect (1 Peter 3:15). The deepest Christian conviction is not that we possess a book miraculously immune to all development—our Scriptures span covenants and centuries openly. It is that the Word of God is finally not a text at all but a person: “the Word became flesh and dwelt among us” (John 1:14). A book’s rulings may be abrogated; a living person, raised from the dead, is not superseded. That is the conversation toward which a fair reckoning with naskh quietly points—and where, in love, it should lead.