CONSENSUS AND DISAGREEMENT  93

the Iraqian concept of consensus of scholars in each generation.¹ Basing himself on the traditions expressing 'unsuccessful' Medinese opinions and on recent (mostly spurious) information regarding old Medinese authorities, he denies the existence of real consensus in Medina and charges the contemporary Medinese with diverging from the consensus of their old authorities.² He tends to replace the old concept of consensus, on which the Medinese rely, by his idea of sunna (§ 71). Against the provincialism of the Medinese in their concept of consensus, he points out that the Medinese are only a minority and claims that, if a consensus exists in Medina, it exists also in the other countries, and if there is disagreement in Medina, the other countries also disagree.³

Ikhtilaf al-Hadith, itself the latest of the treatises, contains early passages, and we find both the old concept of consensus and Shāfi'ī's new one. Some typical examples of the former (which are, however, not all necessarily early) occur on pp. 5, 37, 73, 170, 176, 207, 246, 262. For the latter see, for instance, pp. 141 ff., which is directed against the assumption of a silent consensus of the Companions, and of a consensus of Companions in general: 'the alleged consensus [of Companions and later authorities] on many points of detail cannot be properly claimed'; Shāfi'ī considers the opinion of his opponents to the contrary as ill-advised, ignorant, and pretentious. 'The forebears never, if I am right,⁴ held that all details of law are based on consensus in the same way in which there is consensus on the Koran, the sunna, and the essentials.' Apart from the essential duties which the public at large are obliged to observe, no consensus has been claimed by any of the Companions or of the Successors or of the following generation or of those after them, or by any scholar on earth whom Shāfi'ī has known, or by anyone who was regarded as a scholar by the public, except occasionally when someone claimed it after a fashion approved by no scholar Shāfi'ī can think of, and to his personal knowledge rejected by many.⁵

But Shāfi'ī was unable to dispense completely with the idea of consensus of the scholars; he tried to reconcile it with his concept of the consensus of the community at large by opposing the opinion of the generality of scholars ('awāmm ahl al-'ilm) to that of the specialists (khāṣṣa) among them (pp. 56 f.).⁶ The unanimous opinion of the