The argument we are advancing in these blog posts is that when al-Tabari (d. 310/923) created his Taʾrikh al-rusul wa-l-muluk, Jamiʿ al-bayan ʿan taʾwil ay al-Qurʾan (Tafsir) and Tahdhib al-athar, he relied on notes that were bound into notebooks of some kind, and we can reconstruct something of their contents.
We base our argument of the notebooks’ existence on practical evidence: the long interval between the time of writing and the time of his contact with the people he cites, the extensiveness of the citations (which would represent an astonishing feat of memory without written records) and the quick pace at which al-Tabari appears to have worked. There is also more direct evidence. This evidence consists of the citations themselves, including the ways in which they map onto probable notebooks, and of the alignments between all three books revealed by text reuse data.
A further point that emerges from our data set is the extent to al-Tabari appears to have gone back to the same sets of notes when composing each of his works. Let us look more closely at our data and numbers on this point.
Viewing all 849 names on our list of direct informants provides only a weak picture of sources common to the works. This list, as noted in an earlier post, contains many names that occur only once or twice in al-Tabari’s works, and the identities of many of them are uncertain. The list shows that al-Tabari relied on 347 direct informants for the Taʾrikh, 543 for the Tafsir and 342 for the Tahdhib. Cross-checking pairs of these sub-lists shows that the Taʾrikh and the Tafsir have 139 direct informants in common, the Taʾrikh and the Tahdhib have 110 in common and the Tahdhib and the Tafsir have 229 in common. All three works share 104 direct informants. These figures seem to suggest that he drew on many different sources and mostly used different sources for each work.
But the picture becomes more interesting when we focus on the smaller number of people on whom al-Tabari relied intensively. Recall that 51% of all of al-Tabari’s citations in our data set came from just eight people; twenty people accounted for 79% and fifty people for 89% of citations. It is these relatively few direct informants who mattered most to al-Tabari.
By focusing on this small subset we get a clearer picture of al-Tabari’s working habits. Of the fifty direct informants cited most frequently, thirty-eight are cited in all three of his works. Of the top twenty, eighteen appear in all three, and each of the top ten is cited in each work. In other words, al-Tabari’s most important direct informants constituted his core source base for all three of his works.
Among these direct informants, there is a further subset consisting of people cited numerous times in each of the three works. The Taʾrikh and the Tafsir took years to complete, and in the drawn-out writing process, the notes from these informants were al-Tabari’s go-tos and trusted favourites. In their case it seems especially probable that he had at his disposal stable notes bound together for easy reference and citation.
| Rank | Name | Died | Tabari met in | Total count | Count in Taʾrikh | Count in Tafsir | *Count in *Tahdhib ** |
| 1 | Muhammad b. Humayd al-Razi | 248/862 | Rayy | 2879 | 457 | 2126 | 296 |
| 6 | Muhammad b. Bashshar | 252/866 | Basra | 1644 | 34 | 1240 | 370 |
| 7 | Abu Kurayb | 248/862 | Kufa | 1609 | 69 | 1033 | 507 |
| 11 | Muhammad b. al-Muthanna | 250/864 | Basra | 1081 | 47 | 649 | 385 |
Table 1: Four direct informants on whom al-Tabari relied heavily and whose material was likely preserved in well-organised notebooks.
Among al-Tabari’s informants, Muhammad b. Humayd al-Razi (d. 248/862), Muhammad b. Bashshar (d. 252/866), Ab Kurayb (d. 248/862) and Muhammad b. al-Muthanna (d. 250/864) stand out. It is not hard to see why al-Tabari saw these individuals’ works as particularly worthy of careful collation and frequent citation. Muhammad b. Humayd transmitted Ibn Ishaq’s Sira, an important source both for history (Taʾrikh) and for interpretation of the Qur’an (Tafsir; he is also cited 296 times in the Tahdhib within our data set).
The other figures are well-known Hadith scholars. Scott Lucas’s work on al-Tabari’s ‘teachers’ has shown that Hadiths from each of these three are cited in all six of the canonical Sunni Hadith collections (i.e. those of al-Bukhari, Muslim, Abu Dawud, al-Tirmidhi, al-Nasaʾi and Ibn Maja).1 These numbers are of interest for our understanding of knowledge transmission in the third/ninth and fourth/tenth centuries and beyond. How did an author whose works mattered so much to posterity put together his very large works? Reliable materials mattered. And those that al-Tabari deemed reliable he returned to again and again in his works.
In our next post, we examine text reuse alignments between al-Tabari’s works, which also suggest a common source base.
-
Scott C. Lucas, Selections from the Comprehensive Exposition of the Interpretations of the Verses of the Qurʾān (Cambridge: Royal Aal Al-Bayt Institute for Islamic Thought and Islamic Texts Society, 2017), esp. ii, 467–82 (‘Appendix B: Ṭabarī’s Teachers’). ↩