If one scholar reports a consensus and another reports disagreement—whether naming the scholar who held the dissenting view or merely reporting disagreement without naming anyone—it is not sufficient to say, “The report of disagreement is unsubstantiated.” For the reply would be, “The report of consensus is likewise unsubstantiated.” Indeed, the person reporting consensus denies the existence of disagreement, while the other affirms it, and affirmation takes precedence over denial.

If someone says, “The person reporting disagreement may have erred in affirming it, either because the chain of transmission is weak or because the statements cited do not indicate disagreement,” the reply is: “The person denying disagreement is even more liable to error. There may be opinions on the issue that have not reached him. Or they may have reached him, but he considered their chains of transmission weak, while others found them authentic. Or he may have thought the statements did not indicate disagreement when they did.” Every error possible for the person affirming disagreement is also possible for the person denying it, with the additional possibility of being unaware of the disagreement altogether. This applies to disagreements generally: lack of knowledge is not knowledge of nonexistence, especially regarding the opinions of the scholars of Muhammad’s community—may Allah bless him and grant him peace—which none but the Lord of all worlds can fully encompass.

For this reason, Ahmad and other scholars said: “Whoever claims consensus has lied. This is the claim of al-Marīsī and al-Asamm. Rather, one should say, ‘I know of no disagreement.’” Those who used to cite consensus, such as al-Shāfiʿī, Abū Thawr, and others, explained that they meant, “We know of no disagreement,” and would say, “This is the consensus we claim.”

It is thus clear that a consensus of this kind, countered by a report of disagreement, when neither has been established