158 THE CONFLICT OF DOCTRINES AS REFLECTED

The main tradition in Tr. III, 5, represents 'Urwa b. Zubair as being converted to a certain doctrine by a tradition from the Prophet which he came to know (this is obviously already a counter-tradition). Muw. i. 79 has a statement, through 'Urwa's son Hishām, on 'Urwa's doctrine which he had heard from his father Zubair, to the same effect as 'Urwa's revised opinion in the first tradition. This obviates the claim of a change in 'Urwa's doctrine. The first tradition occurs in a more elaborate form, designed to give it greater authority, in Ṭaḥāwī, i. 43.

The essential features of the common ancient doctrine on slaves captured by the enemy and recaptured by the Muslims, a doctrine for which Auzā'ī and Abū Ḥanīfa did not yet know a tradition, are expressed in an Iraqian tradition from the Prophet which appears for the first time in Abū Yūsuf in Tr. IX, 18. The ruling is given in general terms which do not well agree with the circumstantial story which has been added in order to provide an authentic touch. This form is improved and a further personal touch is added in the versions in Dāraqutnī and Baihaqī respectively (see Comm. ed. Cairo, loc. cit.). Ḥasan b. 'Umāra, in the generation preceding Abū Yūsuf, is the lowest common link in the three isnāds, and he or a person using his name must be responsible for the creation of this tradition and the fictitious higher part of the isnād. But Ibn 'Umāra was impugned, and the tradition is therefore related alternatively, on hearsay authority, through 'Abdalmalik b. Maisara who is, however, also considered weak.

The same doctrine is expressed in two Medinese traditions with the first-class isnād Abū Yūsuf—'Ubaidallāh b. 'Umar-Nāfi'— Ibn 'Umar, both quoted for the first time by Abū Yūsuf in Tr. IX, 18, and in Kharāj, 123,¹ respectively. The first gives it as a general ruling of Ibn 'Umar, the second purports to describe the loss by Ibn 'Umar of a slave and a horse to the enemy, and the subsequent restitution of the one during the lifetime of the Prophet and of the other after his death, by Khālid b. Walid who had recaptured them. In its older forms, which are preserved, without an isnād, in Muw. ii. 299 and in Siyar, iii. 107, this anecdote lacks the indirect reference to the Prophet² or is even explicitly dated to the time of 'Umar³. None of this is genuine, and the fact that Mālik, who relates many traditions from Nāfi'—Ibn 'Umar, does not yet know it as a formal tradition from Ibn 'Umar, makes it likely that the isnād with Nāfi'