The Sira: How Much Do We Really Know About Muhammad?

Ask a believing Muslim how she knows that the angel Gabriel first appeared to Muhammad in a cave on Mount Hira, that he migrated from Mecca to Medina in the year we now call 622, that he won at Badr and stumbled at Uhud, that he was gentle with children and stern with hypocrites—and she can answer in vivid detail. The striking thing is where almost none of that detail comes from. It does not come from the Qur’an, which says remarkably little about the events of its prophet’s life, naming him only a handful of times and narrating almost no biography.¹ It comes instead from the sira, the prophetic biography—a body of tradition assembled by later Muslims who told the story of the man around whom their scripture had formed.

This essay asks a single, careful question: how solid is that biographical foundation? It is worth saying at the outset, and saying plainly, what is not being argued. This is not the claim that Muhammad did not exist. It is not the claim that nothing can be known about him. Both of those would overreach the evidence badly. The argument is the more modest and, I think, more honest one: the detailed life of Muhammad that most people—Muslim and non-Muslim alike—take for granted rests on sources that are late, that survive only in edited form, and that were shaped by the theological needs of the communities that preserved them. The right posture toward the specifics, therefore, is one of held-loose confidence rather than certainty.

The Sources, and the Gap

The earliest connected biography of Muhammad we can name is the Sirat Rasul Allah (“Life of the Messenger of God”) by Ibn Ishaq, who died around 767. Muhammad died in 632. That is already a gap of roughly 120 to 135 years between the events and their first sustained narration—the distance, in our own terms, between the present and the 1890s, recounted by someone working only from what had been passed down.²

But the situation is more delicate still, because Ibn Ishaq’s original work does not survive. Not a single complete copy has come down to us. What we possess instead is a later recension—an edited redaction—produced by Ibn Hisham, who died around 833, fully two centuries after Muhammad.³ Alongside Ibn Hisham, large stretches of Ibn Ishaq’s material are preserved as quotations in the massive universal chronicle of al-Tabari (839–923), whose History of the Prophets and Kings reaches us from the early tenth century and in fact contains some of the fullest surviving extracts of Ibn Ishaq.⁴ The standard English translation, Alfred Guillaume’s The Life of Muhammad (1955), is itself a scholarly reconstruction: Guillaume stitched Ibn Hisham together with Ibn Ishaq material drawn from al-Tabari and others, attempting to recover a lost original he could only approach at second and third hand.⁵

Two things follow. First, the chain from event to surviving text is long: oral and early written reports across roughly a century, then Ibn Ishaq, then a redactor or chronicler another forty to ninety years on. Second—and this is the part that gives even sympathetic readers pause—the redaction we depend on was openly selective. In his preface, Ibn Hisham tells the reader that he has left things out: matters he judged would “distress certain people,” reports his own teacher could not vouch for, poems regarded as spurious, and, in Guillaume’s rendering, “things which it is disgraceful to discuss.”⁶ We are not dealing with a raw archive but with a curated one, and the curator told us so.

A Disputed Author, an Impressive Method

Ibn Ishaq was also a contested figure within his own tradition. Several major early scholars expressed reservations about him as a transmitter; the celebrated Medinan jurist Malik ibn Anas—founder of the Maliki school of law—is reported to have clashed with him and to have called him a “liar” (kadhdhab) and an “imposter” (dajjal, here in the sense of charlatan rather than the eschatological Antichrist), while others suspected him of relying too readily on Jewish converts and of weak chains of transmission.⁷ Fairness requires immediately adding the other side of the ledger: many equally weighty authorities defended him, and later hadith scholars often graded his reports as acceptable in biography even while hesitating over them in law. The point is not that Ibn Ishaq was a fraud—that would be its own overreach—but that the very tradition that preserved him did not regard him as beyond question.

And here charity demands real respect for what the Muslim historians achieved. The transmission of the sira was bound up with the sophisticated science of the isnad, the chain of named transmitters appended to a report, which classical scholars scrutinized with extraordinary care. This was not credulous storytelling; it was an early and serious attempt at source criticism, and it has no real parallel in the ancient world. To wave it away would be both unjust and unwise. The difficulty, which Muslim scholars themselves acknowledged, is structural: isnad criticism interrogates the chain of names more searchingly than the content, and a chain can be counterfeited—a problem we treat at length in the companion essay on the hadith.

The Seventh-Century Silence—and Its Few Voices

If the Muslim narrative sources are late, what about contemporary, seventh-century evidence—the kind a historian most wants? Here the record is thin, and honesty requires saying so in both directions. There is no contemporary documentary life of Muhammad: no inscription he commissioned, no letter in his hand, no dated chronicle written within his lifetime that narrates his career. The earliest Arabic-Islamic documentary traces—coins, an inscription or two, administrative papyri—begin to mention him by name only in the latter decades of the seventh century.⁸

Yet the silence is not total, and the most important counterweight comes from outside the Muslim community. A Greek Christian tract called the Doctrina Jacobi nuper baptizati (“Teaching of the Newly Baptized Jacob”), whose setting is a debate of 634 and whose core is usually dated to around 634–640—within a decade or so of Muhammad’s death—refers to “a prophet” who has appeared among the Saracens (the Arabs), proclaiming the coming of an anointed one and holding, it says, “the keys of Paradise.”⁹ The text is hostile and confused on details, and it does not name Muhammad. But it is powerful evidence that, within a few years of 632, non-Muslims in the eastern Mediterranean already knew that an Arabian prophet had arisen and that his followers were on the move. Other early non-Muslim notices—Syriac, Armenian, and Greek—corroborate the same broad picture across the seventh century.¹⁰ These sources do not vindicate the detailed traditional biography; they were not in a position to. What they do, decisively, is anchor the bare existence and rough dating of Muhammad and his movement in testimony nearly contemporary with the events, and independent of the Muslim tradition.

The Revisionist Challenge—Stated at Full Strength

In the 1970s a group of scholars pressed the source problem far harder than the mainstream had, and their challenge has to be stated at its strongest before it is weighed. John Wansbrough, in Quranic Studies (1977) and The Sectarian Milieu (1978), argued that the Qur’an and the sira alike were not transparent records of seventh-century Arabia but the literary deposit of a sectarian community that took shape considerably later, perhaps in the eighth or ninth century—the sira, on this reading, a body of pious storytelling (he used the term salvation history) generated to anchor scripture in a prophetic life, not a sober chronicle of one.¹¹

His students Patricia Crone and Michael Cook went further. In Hagarism: The Making of the Islamic World (1977), they proposed setting the Muslim sources almost entirely aside as too late and too tendentious to trust, and reconstructing Islam’s origins from contemporary non-Muslim sources instead—arriving at a startlingly different picture of a messianic Arab-Jewish movement.¹² Whatever one makes of the conclusion, the underlying methodological worry is serious and cannot be dismissed: if our narrative sources postdate the events by a century or two and were written to serve later doctrinal needs, on what basis do we trust their particulars? That is a fair question, and it is the engine of this whole essay.

The Mainstream Reply—and What the Revisionists Conceded

The reply came swiftly and, in the end, largely prevailed—but it prevailed by conceding the moderate point while rejecting the radical one. Critics noted that the non-Muslim sources Crone and Cook leaned on are themselves fragmentary, hostile, and no more obviously reliable than the Muslim ones; building a confident alternative history on them simply trades one set of problematic sources for another.¹³ Most strikingly, the authors themselves stepped back. Cook stated flatly, “The central thesis of that book was, I now think, mistaken”; Crone, while never abandoning a critical stance toward the sources, likewise distanced herself from the book’s specific reconstruction.¹⁴ Wansbrough’s very late dating of the Qur’an, meanwhile, sits poorly with the early Qur’anic manuscript evidence—most notably the Birmingham folios, radiocarbon-dated with high probability to before 645, and other early Hijazi codices—which point to a seventh-century text far earlier than his model allowed.¹⁵

Now the steelman of the position this essay ultimately commends—the mainstream historical-critical view—deserves to be put as strongly as it can be, because it is the strongest position in the room. A real Muhammad, an Arabian religious reformer active in the early seventh century who founded the community that erupted out of Arabia within a few years of his death, is about as well established as anything in late-antique history. The convergence is the argument: the Qur’an itself as a primary seventh-century text; the near-contemporary non-Muslim notices; the explosive, datable Arab conquests that demand a galvanizing founder; the early documentary traces; and a Muslim tradition that, however late its written form, preserves a recognizable core. Scholars such as Fred Donner have argued powerfully that the tradition, critically sifted, yields a credible early-seventh-century “Believers’ movement” around Muhammad, and that wholesale skepticism throws out far more than the evidence warrants.¹⁶ Oral cultures, moreover, can transmit a stable core of material across generations even where exact wording and chronology drift. The skeleton—Mecca, Medina, preaching, community, conquest—is firm.

I find that steelman convincing, and this essay does not contest it. The error to avoid is sliding from “the sources are late” to “therefore we know almost nothing,” which is precisely the revisionist overreach the field corrected.

Where That Leaves the Details

But notice exactly what the mainstream view secures, and what it does not. It secures the skeleton. It does not, by the same evidence, secure the flesh—the specific scenes, sayings, miracles, motives, and tight chronology that make up the traditional biography and that believers often hold with great confidence. Those particulars rest overwhelmingly on the late, redacted sira-and-hadith material, not on the early non-Muslim notices or the Qur’an. The cave of Hira, the wording of particular sermons, the dramatized councils before battles, the miracle reports—these come to us through Ibn Hisham’s admittedly pruned recension and the hadith corpus, with all the distance and shaping that implies.

Here the critic should get the last word before the reply. A defender of the tradition will say: granting the late written form, the isnad system and the sheer mass of cross-referenced reports let us recover far more detail than your “skeleton only” verdict allows; you are smuggling in a double standard, demanding of the sira a documentary contemporaneity you would never demand of, say, Alexander the Great, whose main biographers also wrote centuries after him. That is a genuinely strong objection and deserves a genuine answer rather than a dismissal.

The answer is twofold. First, the Alexander comparison cuts both ways: precisely because Arrian and Plutarch wrote so long after Alexander, careful historians do hold the colorful particulars of his life loosely, distinguishing the secure outline from the legendary embroidery—which is exactly the discipline being urged here for the sira. The plea is not for a double standard but for a single one, applied evenly. Second, the isnad method, impressive as it is, scrutinizes chains more than content and can be defeated by a skilled forger; and the tradition’s own admission—that the great bulk of circulating reports were judged unsound, and that the surviving biography was deliberately edited—is itself reason to treat the rich detail as the most vulnerable layer, not the most secure.¹⁷ Conceding the secure core is not skepticism; it is the same critical caution every ancient historian practices.

The Honest State of the Scholarship

Where does the field actually stand in 2026? The radical revisionism of the 1970s has been substantially rolled back—by the early Qur’anic manuscripts, by its own authors’ retractions, and by the implausibility of dissolving Muhammad into a later fiction. There is broad agreement that a historical Muhammad existed and was active in early-seventh-century Arabia. At the same time, the source-critical worry that motivated the revisionists has been absorbed into the mainstream rather than refuted: serious historians now routinely distinguish the well-grounded outline of Muhammad’s career from the layered, tendentious, late-attested detail of the traditional biography, and treat the latter with calibrated caution. Chase Robinson’s standard survey of Islamic historiography stresses just this—that early Islamic historical writing served social and theological functions and must be read with its purposes in view, not as neutral reportage.¹⁸ The live debates today are about how much of the detail can be recovered and by what method (the refinement called “isnad-cum-matn” analysis is one promising tool), not about whether Muhammad existed.

Conclusion: What This Does and Does Not Establish

The source situation establishes, with as much confidence as history allows, that the traditional biography of Muhammad rests on late and theologically shaped foundations: a first life written more than a century after the events, surviving only through a redactor who told us he edited out what might embarrass, and a documentary seventh-century record that is sparse. It establishes that confident, granular claims about the specifics—particular scenes, words, motives, miracles—should be held loosely.

It does not establish that Muhammad is a myth; he is not, and to say so would be to repeat the revisionists’ corrected mistake. It does not establish that nothing can be known; the convergence of Qur’an, near-contemporary non-Muslim notices, and the conquests secures a real man and a real movement. What it does, for the Christian engaging a Muslim friend, is invite something fair rather than triumphal: a single, consistent historical standard applied to both faiths’ founding sources. Apply it honestly, and the New Testament documents—written within living memory, resting partly on eyewitness testimony, and behind them the creed of 1 Corinthians 15:3–7 datable to within a few years of the crucifixion—turn out to stand closer to their events than the sira stands to Muhammad. That is not a weapon to wield over anyone made, as we all are, in the image of God. It is an invitation to follow the evidence together, with gentleness and respect, wherever it honestly leads.¹⁹