Josef Horovitz
[264] In his new edition of Noldeke's Geschichte des Qorân,1 Schwally expresses criticism of my showing the isnād to have had its origin in Jewish literature. I wish to answer this criticism in brief.
1. "The chain of transmitters was never as important a part of Jewish literature as it was of Arab ḥadīth, even by the end of the first century AH." It is correct that in the end Jewish literature does not show the same consistency in tracing chains of transmitters as do ḥadīth and sīra Schwally himself points out that the unconditional acceptance of the isnād only developed gradually.2 The question, however, is not where the consistency comes from, as this is without difficulty explained as an internal Islamic development, but where the first instance of the use of a chain of witnesses comes from.
2. " This Jewish usage does not have a history either irt Judaism or in Israelite culture. Consequently, a foreign origin has to be assumed." The topic under discussion was not the origin of Jewish but of Islamic usage. Islam can only have learned it from a quarter where it was either still in use in the seventh and eighth centuries, or at least known from older national or religious literature. Theoretically it is perfectly possible that Judaism originally came to know the system of the chain of witnesses from an older body of literature that has so far remained unknown to us. As a source for the Islamic use of the chain of transmitters, this could only be considered if Islamic scholars of the seventh and eighth centuries could have had immediate access to it, without Jewish mediation. We do not know of any literature from either those centuries or Antiquity in the Middle East to have used the system of chains of witnesses, apart from Jewish literature. Besides, as can be seen from the Book of Jubilees (VII.38, 39), the chain of witnesses was in use as early as the second century BC, albeit in a slightly different form.3
3. "The question about the isnād cannot be separated from that about the origin of other peculiarities of the older historical literature of the Arabs." A petitio principii of this kind can easily lead to closing the paths to the right solution, especially in a field where syncretism is as triumphant as it is in Islam. If we consider that ideas, traditions, and laws of diverse origin influenced ḥadīth and sīra and, before them, the Qur'an, we should be sufficiently warned off tracing back all the peculiarities to one single source in this case.
[265] Schwally's attempt at explaining the isnād from the traditions of the ancient Arab reciters of poetry seems to me just as futile as his criticism. If there were such a connection, it would have to be expected that the isnād, having its origin in the poetry rather than in the prose, would have been used for poetical texts before being transferred to prose texts as well. However, Ibn Isḥāq, who does use the isnād freely, never uses it for songs. The passages at the end of each of the chapters about the campaigns consist exclusively of songs, and there is not a single isnād to be found in them. In the few instances in which lines of poetry do actually appear with an isnād, it does not refer to them at all but only to the prose passage in which they are embedded. Ibn Sa'd still only very rarely gives more authorities for poetry than his immediate source, as does Ibn Hishām.
After all this, it seems impossible to me that the isnād should have its origin in the transmission of poetry. I consider it most probable that at first it was only applied to the Prophet's own sayings, and that it was first used in ḥadīth and only taken into the sīra when a "scientific" approach to the Prophet's biography became more relevant. Where the emphasis was on continuous narrative, the isnād frequently turned out to be in the way. This problem was solved quite early by merging several reports and by prefacing all their isnāds to the unified version of the story.4 Al-Wāqidī and Ibn Sa'd, on the other hand, would begin by composing a unified narrative based on the information collected by their predecessors. Then they have the references collected by themselves follow the main part of the story.5 Transferring the system of isnād to the information about the poets and their songs was part of this later development.
1 Theodor Nöldeke, Geschichte des Qorâns, 2nd ed. by Friedrich Schwally, Gotthelf Bergsträsser, and Otto Pretzl (Leipzig, 1909-38), II, 128-29.
2 Ibid., II, 131, 133.
3According to Hippolytus, Pkilosophumena V.vii.l [ed. Miroslav Marcovich, Refuta-tio omnium haeresium (Berlin, 1986), 142:18-19] (and similarly X.ix.3 [384:18-20]) the Naassenes said that their teachings had been taught to Marianne by Jacob. Similar references to authorities are frequent, but unlike the passage from the Book of Jubilees quoted above they do not actually constitute a chain of witnesses. The examples quoted in Dele-haye, Les légendes hagiographiques, 2nd ed. (Brussels, 1906), 80-81, where the legend of the saint is told by one of his pupils, do not belong to this category either.
4Cf. above, 42.
5Cf. Ibn Sa'd, Kitāb al-ṭabaqāt ai-kabīr, ed. Eduard Sachau et al. (Leiden, 1904-40), II.1, Preface, v.