Post 3: Ibn ‘Asakir’s Direct Informants

Ibn ʿAsākir names many persons from whom he acquired information for the TMD. What can our data tell us about them?

For what follows, we also consulted Ibn ʿAsākir’s Muʿjam al-shuyūkh, as well as two major biographical works containing entries for the period: Yāqūt al-Ḥamawī’s biographical dictionary Muʿjam al-udabāʾ and al-Dhahabī’s Siyar aʿlām al-nubalāʾ.

The Pareto distribution

Ibn ʿAsākir cites many people as his direct informants – hundreds or possibly thousands. The number of Ibn ʿAsākir’s direct informants far exceeds that of al-Ṭabarī’s. However, like al-Ṭabarī, he relies extensively on a much smaller number of people, with twenty or fewer persons responsible for over 60% of all information and 100 people responsible for 88%. He cites these persons repeatedly across the TMD, and he also provides biographies for many of them in the TMD and includes entries for them in his Muʿjam al-shuyūkh.

Number of names	% of all citations
1	11
2	18
3	22
4	27
5	31
6	34
7	38
8	41
9	43
10	46
20	61
30	70
40	76
50	79
60	82
70	84
80	85
90	87
100	88

Table 3.1: The percentage of Ibn ʿAsākir’s citations accounted for by his most commonly cited direct informants, at various sample sizes.

These figures are estimates because the many unnormalised names complicate our counting. But if anything, they overstate the number of people upon whom Ibn ʿAsākir relied. That is because we have done considerable work to disambiguate the twenty most commonly cited names, but we have done much less for the next eighty names, which often appear in multiple lists (in different combinations and orders). Many of the latter names have not been normalised, and they may thus in reality refer to a smaller number of people than it appears.

In any case, it is clear that a small number of people are responsible for most of Ibn ʿAsākir’s citations. By the numbers, this pattern seems to resemble the so-called Pareto distribution (or ‘power law’), which is more typically associated with the areas of science, engineering and finance.¹ However, the percentages in our data are not actually Pareto-like. In fact, the distribution is far more unequal. Whether we consider the top twenty, eighty or even one hundred most frequently cited individuals, these people represent less than 5% of Ibn ʿAsākir’s informants yet contribute the vast majority of his quotations. The figure of 5% is based on a conservative estimate that would put the total number of people he cites in the 2,000–3,000 range. That exceeds the number of shaykhs he lists in Muʿjam al-shuyūkh and a separate work on his female teachers (according to al-Dhahabī, Muʿjam al-shuyūkh mentions 1,636 teachers and the work on women adds another eighty-odd names). But it is still less than the 7,637 names and combinations of names that appear in our data set as direct informants, most occurring only once or twice.²

Although Ibn ʿAsākir does cite many people in the TMD, the sheer number of variant surface forms for direct informants’ names vastly inflates our sense of the size of his informant pool. Typically, he is not trying to communicate his efficiency in gathering information; he is simply referring to persons he knew well using terms that came immediately to mind. One of his uncles (d. 537/1142), for example, is mentioned in the ways listed in Table 3.2, all of which are normalised in our data to Abu Ma'ali Muhammad b. Yahya.

Normalisation	Arabic Surface Form
Abu al-Ma'ali Muhammad b. Yahya	القاضي أبو المعالي محمد بن يحيى
Abu al-Ma'ali Muhammad b. Yahya	خالي أبو المعالي القاضي
Abu al-Ma'ali Muhammad b. Yahya	خالي أبو المعالي محمد بن يحيى
Abu al-Ma'ali Muhammad b. Yahya	خالي أبو المعالي محمد بن يحيى القاضي
Abu al-Ma'ali Muhammad b. Yahya	خالي أبو المعالي محمد بن يحيى بن علي القاضي
Abu al-Ma'ali Muhammad b. Yahya	خالي القاضي أبو المعالي محمد بن يحيى
Abu al-Ma'ali Muhammad b. Yahya	خالي القاضي أبو المعالي محمد بن يحيى القرشي
Abu al-Ma'ali Muhammad b. Yahya	خالي القاضي أبو المعالي محمد بن يحيى بن علي
Abu al-Ma'ali Muhammad b. Yahya	خالي القاضي أبو المعالي محمد بن يحيى بن علي القرشي

Table 3.2: The various ways in which Ibn ʿAsākir refers to Abu Ma'ali Muhammad b. Yahya in the TMD.

Many relatives are identified in similarly varying ways, with the result that until the names have been normalised, the pool of direct informants appears artificially large. Ibn ʿAsākir’s most frequently quoted direct informant, Abu al-Qasim al-Samarqandi, is described in dozens of different ways, including as the son of another man’s daughter. There are also many references to ‘the qāḍī so-and-so’, and to individuals identified as ‘the jurist’ (faqīh), the Qurashī, the Anṣārī, the Dimashqī or the like alongside a proper name. Similarly, that a person was the companion or student (ṣāḥib) of another person is occasionally mentioned. Ironically, such efforts at disambiguation by Ibn ʿAsākir himself are the cause of difficulties in pinning down his informants’ identities.³

In addition, Ibn ʿAsākir sometimes mentions where he heard a piece of information, and the inclusion of place names with personal names requires some work to separate the two. The examples in Tables 3.3 and 3.4 illustrate another difficulty – that caused by Ibn ʿAsākir’s use of very different-looking names for one and the same individual.

Abu Muhammad al-Sulami	عبد الكريم بن حمزة
Abu Muhammad al-Sulami	أبي محمد بن السلمي
Abu Muhammad al-Sulami	أبي محمد بن حمزة السلمي
Abu Muhammad al-Sulami	أبي محمد عبد الكريم السلمي
Abu Muhammad al-Sulami	أبي محمد عبد الكريم بن حمزة السلمي

Table 3.3: Variant surface forms referring to Abu Muhammad al-Sulami.

Abu al-Qasim al-Mustamli	أبو القاسم الشحامي
Abu al-Qasim al-Mustamli	أبو القاسم زاهر
Abu al-Qasim al-Mustamli	أبو القاسم زاهر بن طاهر الشحامي
Abu al-Qasim al-Mustamli	أبو القاسم زاهر بن طاهر ا المستملي

Table 3.4: Variant surface forms referring to Abu al-Qasim al-Mustamli. Since Ibn ʿAsākir cites many Abu al-Qasims, this table probably does not capture all the ways in which this individual is cited, and more work could be done to disambiguate the Abu al-Qasims in general.

All these features of Ibn ʿAsākir’s citations challenged our identification of distinct informants and our ability to trace the routes through which information was passed down.

Regardless of the variations and consequent uncertainty, the big picture is clear: Ibn ʿAsākir cites a relatively small number of people very often, and a large number of people infrequently.

If we think in human terms, this makes utter sense. There was inevitably a limit to the number of persons whom Ibn ʿAsākir or any author could consult to produce a work as long as the TMD. He was efficient and got much out of a small number of people. Working primarily with quotations from his core pool of direct informants gave him time to collect also the rare isnāds and the material from his many only occasionally cited informants. This approach allowed him to compile what is the largest pre-1500 work in our corpus.

To understand Ibn ʿAsākir’s overall transmission methods, we need to focus on the people he cites most often. This is because in the transfer of information, the frequently cited direct informants are the highways, whereas those rarely cited are the country lanes.

Ibn ʿAsākir’s youthful information gathering

When we examine the biographies of the people Ibn ʿAsākir cites most often, one detail jumps out. Given their death dates, he would have to have been very young to have acquired information personally from them.

Ibn ʿAsākir was born in the month of Muharram in the year 499/September or October 1105 and died on the 11^th^ day of the month of Rajab in the year 571/31 January 1176, at the age of 72 (Hijri) or 70 (Gregorian).⁴ Many of his top informants died before he began writing the TMD in 529/1134–5, and all of them were long dead by the time he completed it in 559/1164. At least three died when he was still very young. As we argued in the parallel case of al-Ṭabarī, this situation raises questions of access and storage. How did Ibn ʿAsākir gain access to the words of these informants, and in what form did he have them to hand when he composed his book?

Table 3.5 lists the twenty individuals whom Ibn ʿAsākir cites most often in the TMD. Quotations from them account for more than 60% of all isnāds in our data set.⁵

Name	Died	Number of citations
Abu al-Qasim al-Samarqandi	536/1142	8,858
Abu Ghalib	527/1133	4,871
Abu al-Barakat	538/1143	3,449
Abu Muhammad al-Akfani	524/1129	3,380
Abu Muhammad al-Sulami	526/1132	3,224
Abu 'Abd Allah	531/1136	2,835
Abu al-Qasim al-Mustamli	533/1138	2,687
Abu al-Qasim al-'Alawi	508/1114	2,069
Abu 'Ali al-Haddad	515/1121	1,950
Abu Bakr Muhammad b. 'Abd al-Baqi	535/1140	1,899
Abu al-Qasim b. al-Husayn	525/1131	1,577
Abu 'Abd Allah al-Furawi	530/1136	1,341
Abu al-Ghana'im	510/1116	1,259
Abu 'Abd Allah al-Khallal	532/1137	1,245
Abu al-Muzaffar	532/1138	1,209
Abu Ghalib al-Mawardi	525/1130	1,120
Abu Bakr Muhammad b. Shuja'	533/1138	1,077
Abu al-Fadl b. Nasir	550/1155	1,011
Abu al-Hasan al-Faradi	533/1139	949
Abu al-Qasim	?	866

Table 3.5: The top twenty direct informants in the TMD, each with death date and number of isnāds started.

Ibn ʿAsākir’s biographers make much of his youthful learning, but how did the material transmitted by these persons really wind up on his pages? What does it mean when he quotes Abu al-Qasim al-'Alawi, who died when Ibn ʿAsākir was 9 years old, or Abu al-Ghana'im, at whose death Ibn ʿAsākir was 11 years old? He mentions dates in isnāds on only a few occasions, but although these dates all relate to his early life, none we have seen record the date on which he obtained information from his top informants. Nor does he explain how he stored the information he cites.⁶ His older brother Hibat Allāh reportedly transmitted material to him through audition (sammaʿahu) from the year 505/1111–12 onwards; perhaps he was an intermediary between Ibn ʿAsākir and those of Ibn ʿAsākir’s sources who greatly predeceased him?⁷ Possible intermediaries such as Ibn ʿAsākir’s brother are not indicated in individual citations within the TMD.

Another possibility is that Ibn ʿAsākir obtained the written lectures of these informants and felt able to quote them because he had been present when the informants had lectured on the relevant topics, even if the notes were not his own or he acquired them later. Given how young he must have been on many of these occasions, this theory seems plausible in at least some cases.

These and similar scenarios were discussed in detail less than a century after Ibn ʿAsākir’s lifetime by Ibn al-Ṣalāḥ al-Shahrazūrī (d. 643/1245) in his Kitāb Maʿrifat anwāʿ ʿilm al-ḥadīth. According to Ibn al-Ṣalāḥ, scholars disagreed over the age at which a child could reasonably be assumed to benefit from hearing Hadith. He reports that a Hadith expert was asked, ‘When should a boy (ṣabī) hear Hadith?’ The expert replied, ‘When he becomes able to distinguish between a cow and a riding animal’ or, according to another version of this answer, ‘between a cow and a donkey’. After summarising various views on the issue, Ibn al-Ṣalāḥ notes that five years is the age ‘on which the practice of the modern scholars of the Hadith has settled’.⁸

The scenarios that Ibn al-Ṣalāḥ presents in his book illustrate the variety of transmission methods in his day – and in that of Ibn ʿAsākir. They show that one could gain access to and transmit the words of direct informants in many different ways, and that some of these ways enabled knowledge transfer even to children such as Ibn ʿAsākir would have been at the time of some of his transmissions.

For example, Ibn al-Ṣalāḥ says that the terms anbaʾanā and nabbaʾanā (‘he communicated to us’) are rarely used to introduce material heard in spoken form from a teacher.⁹ Readers today might assume that the phrases ‘to us’ and ‘to me’ indicate in-person communication (unless they are followed by a phrase such as fī kitābihi, ‘in his writing’), but according to Ibn al-Ṣalāḥ that was not the case in his day.

When Ibn ʿAsākir cites Abu 'Ali al-Haddad, he uses anbaʾanā 75% of the time (in 1,458 out of 1,950 isnāds). Abu 'Ali al-Haddad died when Ibn ʿAsākir was about 16 years old. Similarly, when Ibn ʿAsākir cites Abu al-Ghana'im, whom he met when very young, he uses anbaʾanā 80% of the time (in 1,011 out of 1,259 isnāds).

Perhaps anbāʿnā, for Ibn ʿAsākir, marks a mode of transmission suited to the circumstances of a young person acquiring information from a direct informant via intermediaries and their notes. This possibility is supported by the description of this practice given by al-Ḥākim al-Naysābūrī (d. 405/1014) before Ibn ʿAsākir’s lifetime. He writes in his Maʿrifat ʿulūm al-ḥadīth (‘Knowing the disciplines of Hadith’) that ‘the phrase “he communicated to me” (anbaʾanī) is employed when someone shows their written material to a transmitter of Hadith (ʿaraḍa ʿalā al-muḥaddith) and is given an ijāza (i.e. a licence to transmit) orally (shifāhan).’¹⁰ Perhaps the young Ibn ʿAsākir obtained such authorisations from his much older informants after presenting transcripts of their material written down by and acquired from other scholars serving as intermediaries.¹¹

Ibn al-Ṣalāḥ also outlines other ways in which a very young person could acquire reports. For example, in the method of transmission that he calls ‘recitation to the teacher’ (al-qirāʾa ʿalā al-shaykh), the reciter recites a Hadith to the teacher, allowing the latter to correct any errors. But it ‘makes no difference whether you are the reciter or someone else recites while you listen’.¹² Therefore, a young child present on such an occasion could subsequently transmit the report in question. In his Muʿjam al-shuyūkh, Ibn ʿAsākir frequently uses phrases that seem to suggest that he himself recited materials back to a shaykh (bi-qirāʾatī ʿalayhi). But perhaps this formula also accommodates situations in which he was simply present at a recitation.

Elsewhere, Ibn al-Ṣalāḥ deals with licences issued by a shaykh for transmitting a written text (ijāzas). He quotes al-Khatib (al-Khaṭīb al-Baghdādī), who held the opinion that it is permissible to give an ijāza to a young child – even one who was not actually present at an audition session. According to al-Khatib, ‘we saw all of our teachers licensing children who were absent, without inquiring about their age and level of discrimination. We never saw them under any circumstance license anyone who was not yet born.’

Explaining al-Khatib’s stance, Ibn al-Ṣalāḥ writes:

It seems that they regarded a child fit for this way of receiving Hadith so that he could validly transmit on the basis of it after he had attained the qualifications necessary for transmitting. This arose from their desire to expand the means of perpetuating the isnād – which has been granted to this Community alone – and to bring the child closer to the Messenger of God (Peace be upon him). God knows best.¹³

These ideas, which continued to circulate in Ibn ʿAsākir’s day and beyond, suggest that we need to consider critically the terms Ibn ʿAsākir uses to describe transmission – including the phrases that supplement opening expressions in isnāds, such as akhbaranā and ḥaddathanā.

We might expect to find clarification on Ibn ʿAsākir’s methods of accessing and storing information in his Muʿjam al-shuyūkh. But the terminology in this book does not generally match his citations in the TMD. A search for the term anbaʾanā, for example, turns up zero matches.

Still, the reports for both Abu 'Ali al-Haddad and Abu al-Ghana'im in the Muʿjam al-shuyūkh indicate that he acquired the information he quotes from them via a written source. For Abu 'Ali al-Haddad, the sample isnād refers specifically to a written work: Ibn ʿAsākir writes, ‘Al-Ḥasan b. Aḥmad b. al-Hasan b. Mihra Abū ʿAlī al-Ḥaddād al-Muqriʾ al-Iṣbahānī [Abu 'Ali al-Haddad] informed us (akhbaranā) more than once in his writing (kitābihi) …’¹⁴ For Abu al-Ghana'im, the sample isnād reads, ‘Muḥammad b. ʿAlī b. Maymūn Abū al-Ghanāʾim al-Kūfī al-Narsī, known as Abū al-Ḥāfiẓ [Abu al-Ghana'im], informed us via an ijāza …’¹⁵

Top five transmitters

What do we learn about the people from whom Ibn ʿAsākir quotes most often and Ibn ʿAsākir’s transmission generally, when we delve more deeply into our data?

1. Abu al-Qasim al-Samarqandi

Born	444/1052–3
Died	536/1142
Where he met Ibn ʿAsākir per Muʿjam al-shuyūkh	Baghdad
Total # of his isnāds in TMD	8,858
Percentage of all isnāds in data set	11%
Transmission term used in Muʿjam al-shuyūkh’s example	akhbaranā/bi-qirāʾatī ʿalayhi
Most common transmission term in TMD	akhbaranā
# of his isnāds using this term in TMD	7,609
Percentage of his isnāds using this term in TMD	86%
Name as given in TMD bio	إسماعيل بن أحمد بن عمر بن أبي الأشعث أبو القاسم بن أبي بكر السمرقندي
Name as given in Muʿjam al-shuyūkh	إسماعيل بن أحمد بن عمر بن أبي الأشعث أبو القاسم بن أبي بكر السمرقندي الحافظ
Name as given in al-Dhahabī, Siyar aʿlām al-nubalāʾ	ابن السمرقندي إسماعيل بن أحمد بن عمر الشيخ الإمام المحدث المفيد المسند أبو القاسم إسماعيل بن أحمد بن عمر بن أبي الأشعث السمرقندي الدمشقي المولد البغدادي الوطن صاحب المجالس الكثيرة
Minimum # of surface forms of his name in TMD	36
Sample surface forms	أبو القاسم بن السمرقندي أبو القاسم اسماعيل بن أحمد أبو القاسم إسماعيل بن أحمد بن عمر بن السمرقندي أبو القاسم الدلال
Most common surface form	أبو القاسم بن السمرقندي
# of his isnāds using most common surface form	6,955
Percentage of his isnāds using most common surface form	79%
Minimum # of times named alongside other direct informants in TMD by most common surface form	454
Minimum # of times cited as direct informant along with others	352
# of isnāds citing him as direct informant that contain an author’s name	2,831
Percentage of isnāds citing him as direct informant that contain an author’s name	32%
# of distinct author names in isnāds that cite him	74
Author names in these isnāds (normalised)	Abd Allah b. Ahmad, 'Abd Allah b. Wahb, 'Abd al-Rahman b. Ibrahim al-Dimashqi, 'Abd al-Razzaq, Abu 'Abd Allah Muhammad b. Ahmad b. Muhammad, Abu 'Abd Allah al-Balkhi, Abu 'Aruba al-Harrani, Abu 'Awwana, Abu 'Ubayd al-Qasim b. Sallam, Abu Bakr b. Abi Dawud, Abu Bakr b. Abi Dunya, Abu Bakr b. Ziyad, Abu Bishr al-Dulabi, Abu Dawud, Abu Hudhayfa, Abu Ishaq al-Fazari, Abu Khaythama, Abu Muhammad al-Akfani, Abu Muhammad b. Abi Nasr, Abu Nu'aym, Abu Ya'la, Abu Zur'a, Abu al-Aswad, Abu al-Hasan 'Ali b. Muhammad b. Ahmad, Abu al-Qasim al-Sahmi, Ahmad b. Hanbal, Hisham b. 'Ammar, Ibn 'Adi, Ibn Abi Dawud, Ibn Abi Shayba, Ibn Ishaq, Ibn Ma'in, Ibn Makula, Ibn Mujahid, Ibn Sa'd, Ibn Sallam, Ibn al-Mubarak, Ibn al-Munadi, Ishaq b. Bishr al-Bukhari, Khalifa, Muhammad b. Harun, Muhammad b. Jarir, Muhammad b. Muhammad, Musa b. 'Uqba, Muslim, Sayf, Sufyan b. 'Uyayna, Surayj b. Yunus, Waki', Ya'qub b. Sufyan, Yahya b. Muhammad b. Sa'id, Yunus b. Bukayr, al-Asma'i, al-Baghawi, al-Bayhaqi, al-Bukhari, al-Daraqutni, al-Faryabi, al-Harith b. Abi Usama, al-Hasan b. Sufyan, al-Humaydi, al-Khara'iti, al-Khatib, al-Kilabi, al-Mada'ini, al-Nasa'i, al-Qati'i, al-Shafi'i, al-Shirazi, al-Suli, al-Tabarani, al-Walid b. Muslim, al-Zubayr b. Bakkar, al-Zuhri

Table 3.6: The informant profile of Abu al-Qasim al-Samarqandi, the person whom Ibn ʿAsākir cites most often as a direct informant.

Ibn ʿAsākir cites Abu al-Qasim al-Samarqandi 8,858 times in the TMD, and 11% of all isnāds name him as the direct informant (see Table 3.6). Overwhelmingly, Ibn ʿAsākir uses the term akhbaranā, ‘he informed us’, to describe how he obtained information from this informant. In the Muʿjam shuyūkh, he uses the same phrase in his example citation of a Hadith from Abu al-Qasim al-Samarqandi, although he also notes personal presence on the occasion of transmission: ‘via my reading back to him’ (bi-qirāʾatī ʿalayhi).¹⁶ The Muʿjam al-shuyūkh also records the place where he received the sample Hadith: Baghdad. In our data set, Ibn ʿAsākir’s isnāds sometimes likewise note that Abu al-Qasim al-Samarqandi passed on information to him in Baghdad.¹⁷ One isnād starts thus: ‘Abū al-Qāsim al-Samarqandī [Abu al-Qasim al-Samarqandi] informed us (akhbaranā) in Baghdad, saying: Abū Bakr al-Khaṭīb [al-Khatib] told us (ḥaddathanā) in Damascus in the month of Rabīʿ al-Ākhar in the year 458 [March or April 1066] that Abū al-Hasan Aḥmad b. Muḥammad b. Mūsā b. Hārūn b. Ṣalt …’¹⁸ Interestingly, Abu al-Qasim al-Samarqandi had been born in Damascus and started his learning there, studying, among others, with al-Khatib, but according to Ibn ʿAsākir’s biography of him in the TMD, he subsequently moved to Baghdad. Perhaps Ibn ʿAsākir relished the geography of this report, as it involves him, a Damascene, learning in Baghdad and al-Khatib, a Baghdadi, transmitting in Damascus.

Two centuries later, al-Dhahabī tells us that Abu al-Qasim al-Samarqandi hosted many study sessions (ṣāḥib al-majālis al-kathīra), and when he lists the scholars from whom Abu al-Qasim received material through audition, he names al-Khatib first.¹⁹

The details about Abu al-Qasim al-Samarqandi and other top informants also help us understand how and when the names of authors appear within Ibn ʿAsākir’s isnāds. We identified names as authorial on the basis of the attributions of Scheiner, Mourad and al-Daʿjānī (see Post 2 under ‘Author names and titles of works’).

About a third of the isnāds passing through Abu al-Qasim al-Samarqandi as the direct informant contain author names. What is remarkable is that Ibn ʿAsākir seems to cite such a large number of different authors – seventy-four of them – on this informant’s authority. Many of their names occur only rarely, and it is possible our search has misidentified some of them. But in many cases the names are so specific that it is highly likely our identification is correct.

Some of the authors in question are cited hundreds of times; see Table 3.7.

Author	Died	# of times cited via Abu al-Qasim al-Samarqandi
Abu al-Qasim al-Sahmi	428/1036	773
Ya'qub b. Sufyan	n.d.	448
al-Baghawi	317/929	193
Abu Bishr al-Dulabi	310/923	178
Abu Bakr b. Abi Dunya	281/894	174
Yunus b. Bukayr	199/815	159
Sayf	180/796	154

Table 3.7: The authors whose names appear most frequently in isnāds in which Abu al-Qasim al-Samarqandi is the direct informant.

When scholars have considered Ibn ʿAsākir’s working methods, these sorts of citations have posed puzzles. How are we to imagine he acquired the text passages he seems to be quoting in these instances? Were they derived from well-known books, or perhaps obtained in reading sessions?

With our isnād data set, we tested statements made by modern scholars about premodern book transmission as exemplified in the TMD. As discussed in Post 5, these scholars do not typically spell out the bases of their conclusions, so their statements may reflect data other than isnāds, such as other citations within the TMD (including serendipitous discoveries); inferences of various sorts, including from the mere mention of a name in connection with a topic on which the named person wrote a book; and statements in other, later sources.

For example, Mourad says that Abu al-Qasim al-Samarqandi gave Ibn ʿAsākir a certificate in Baghdad to transmit four works: the Kitāb Futūḥ al-Shām of Isḥāq b. Bishr al-Bukhārī (d. 206/821, normalised as Ishaq b. Bishr al-Bukhari); the Kitāb Ṭabaqāt al-shuʿarāʾ al-jāhiliyyīn of Muḥammad b. Sallām al-Jumaḥī (d. c. 232/846, normalised as Ibn Sallam); the Kitāb Tārīkh Jurjān of al-Jurjānī (d. 427/1036, normalised as Abu al-Qasim al-Sahmi); and the Kitāb Ṭabaqāt al-fuqahāʾ of Abū Isḥāq al-Shīrāzī (d. 476/1083, normalised as al-Shirazi).²⁰ In our data set, isnāds with Abu al-Qasim al-Samarqandi as the direct informant cite Ishaq b. Bishr al-Bukhari three times. They also cite Ibn Sallam (fifteen times), Abu al-Qasim al-Sahmi (773 times) and al-Shirazi (six times). Within these citations, none referring to Ishaq b. Bishr al-Bukhari mentions the Kitāb Futūḥ al-Shām and none referring to Ibn Sallam mentions his Kitāb Ṭabaqāt al-shuʿarāʾ al-jāhiliyyīn. Fifteen of Abu al-Qasim al-Sahmi’s 773 citations refer explicitly to his Tārīkh Jurjān, and all six citations of al-Shirazi refer to his Kitāb Ṭabaqāt al-fuqahāʾ.²¹

We propose as a general explanation that Abu al-Qasim al-Samarqandi shared with Ibn ʿAsākir whatever pieces of existing written works he had to hand in what form he had them (perhaps at Ibn ʿAsākir’s request). In some cases, this may have involved the granting of rights to transmit entire works, as Mourad indicates. The repeated quotations from al-Shirazi’s book point in that direction. In each instance, Ibn ʿAsākir appears to specify the part of the book that he is citing, variously appending to the book’s title, Kitāb Ṭabaqāt al-fuqahāʾ, the phrases min aṣḥāb Abī Ḥanīfa (once), min al-Shāfiʿiyyīn (three times), min aṣḥāb al-Shāfiʿī (once) and min aṣḥāb Mālik (once).²² For example, Ibn ʿAsākir writes: ‘Abū al-Qāsim b. al-Samarqandī [Abu al-Qasim al-Samarqandi] informed us, saying that Abū Isḥāq Ibrāhīm b. ʿAlī al-Shīrāzī [al-Shirazi] in Kitāb Ṭabaqāt al-fuqahāʾ min al-Shāfiʿiyyīn said: Among them is Abū al-Faraj Muḥammad b. ʿAbd al-Wāḥid b. Muḥammad b. ʿUmar …’²³ This passage can be traced directly to al-Shirazi’s Ṭabaqāt al-fuqahāʾ as preserved today in the OpenITI corpus.²⁴

That said, it is noteworthy that Ibn ʿAsākir generally quotes Abu al-Qasim al-Samarqandi using terms that do not point to books or even written transmission (we discuss such terms in later posts). Moreover, the quotations can only occasionally be traced to passages in extant written works. In other words, the evidence indicates that more often than not, Abu al-Qasim al-Samarqandi passed on information in a piecemeal fashion, sharing individual reports or sections of text (including passages from books) with Ibn ʿAsākir. Like Ibn ʿAsākir himself, he seems to have been a collector of bits of material that Ibn ʿAsākir subsequently drew on, just as later scholars who gained transmitted material from the TMD might have been chiefly interested in particular parts treating the history or topography of Damascus or the biographies of specific individuals. Ibn ʿAsākir also cites Abu al-Qasim al-Samarqandi in his other works. He presumably possessed a collection of quotations from Abu al-Qasim, and when he began to write the TMD, he tapped into that collection.

A major question in interpreting isnāds is what Ibn ʿAsākir means when he cites several people together as direct informants for a single report – something he does frequently, including with Abu al-Qasim al-Samarqandi. Since not all names in the TMD have been normalised, we cannot know precisely how common such citations are. We employed two ways of estimating the number of joint citations of Abu al-Qasim al-Samarqandi, and although both are prone to undercounting, they can yield a rough idea. First, we selected the most common surface form of his name and counted how many times it appeared with other names (478 times). And second, we counted instances in which a verified reference to Abu al-Qasim al-Samarqandi accompanied a reference to a second direct informant (at least 352 cases). One possible explanation for these citations is that Ibn ʿAsākir possessed multiple parallel isnāds for the quoted material, each through a different direct informant – but see below for further discussion of how we might interpret these isnāds.

2. Abu Ghalib

Born	?
Died	527/1133
Where he met Ibn ʿAsākir per Muʿjam al-shuyūkh	Baghdad
Total # of his isnāds in TMD	4,871
Percentage of all isnāds in data set	6%
Transmission term used in Muʿjam al-shuyūkh’s example	akhbaranā/bi-qirāʾatī ʿalayhi
Most common transmission term in TMD	akhbaranā
# of his isnāds using this term in TMD	3,312
Percentage of his isnāds using this term in TMD	68%
Name as given in TMD bio	N/A
Name as given in Muʿjam al-shuyūkh	أحمد بن الحسن بن أحمد بن عبد الله بن البناء أبو غالب بن أبي علي الحريري
Name as given in al-Dhahabī, Siyar aʿlām al-nubalāʾ	أبو غالب بن البناء أحمد بن الحسن بن أحمد الشيخ الصالح الثقة مسند بغداد أبو غالب أحمد ابن الإمام أبي علي الحسن بن أحمد بن عبد الله بن البناء البغدادي الحنبلي
Minimum # of surface forms of his name in TMD	58
Sample surface forms	أبو غالب أبو غالب أحمد بن الحسن أبو غالب أحمد بن الحسن بن البنا أبو غالب بن البنا
Most common surface form	أبو غالب بن البنا
# of his isnāds using most common surface form	1,218
Percentage of his isnāds using most common surface form	25%
Minimum # of times named alongside other direct informants in TMD by most common surface form	30
Minimum # of times cited as direct informant along with others	1,915
# of isnāds citing him as direct informant that also contain an author’s name	2,635
Percentage of isnāds citing him as direct informant that contain an author’s name	54%
# of distinct author names in isnāds that cite him	46
Author names in these isnāds (normalised)	Abd Allah b. Ahmad, 'Abd al-Razzaq, Abu 'Abd Allah al-Balkhi, Abu 'Aruba al-Harrani, Abu Bakr b. Abi Dawud, Abu Bakr b. Abi Dunya, Abu Dawud, Abu Muhammad b. Abi Nasr, Abu Nu'aym, Abu al-Ghana'im, Abu al-Hasan b. Sumay', Abu al-Qasim al-Sahmi, Hisham b. 'Ammar, Ibn Abi Dawud, Ibn Abi Khaythama, Ibn Abi Shayba, Ibn Ishaq, Ibn Jusa, Ibn Ma'in, Ibn Makula, Ibn Sa'd, Ibn Sallam, Ibn al-Kalbi, Ibn al-Mubarak, Khalifa, Muhammad b. Jarir, Muhammad b. Muhammad, Sufyan b. 'Uyayna, Surayj b. Yunus, Waki', Yahya b. Muhammad b. Sa'id, al-Baghawi, al-Daraqutni, al-Faryabi, al-Harith b. Abi Usama, al-Humaydi, al-Khatib, al-Khutabi, al-Kilabi, al-Mada'ini, al-Qati'i, al-Suli, al-Tusi, al-Walid b. Muslim, al-Zubayr b. Bakkar, al-Zuhri

Table 3.8: The informant profile of Abu Ghalib.

Ibn ʿAsākir himself followed the Shāfiʿī legal tradition, but his informants included adherents of other rites. Abu Ghalib followed the Ḥanbalī madhhab.

Ibn ʿAsākir places Abu Ghalib in Baghdad when he says in his Muʿjam al-shuyūkh that he recited material back to him there. But he provides no entry for Abu Ghalib in the TMD, suggesting that the latter had never travelled to Damascus to justify his inclusion.

Most often (68% of the time), Ibn ʿAsākir says in his reporting that Abu Ghalib ‘informed us’ (akhbaranā), but he also uses other phrases, most notably variations of ‘I read back to’ (qaraʾtu ʿalā). He uses qaraʾtu ʿalā 1,231 times, or 25% of the time.

Al-Dhahabī describes Abu Ghalib as musnid Baghdād, an honorific that suggests his role as something of an isnād nexus. As explained by Devin Stewart, the term denotes someone ‘who provides a sanad or isnād’; the person is an ‘“anchor” or pivotal “support” who provides an important link for an entire generation of scholars in the chain back to earlier Muslim authorities’. The title indicates that the person ‘plays a crucial role in the transmission of authoritative religious knowledge in general, probably as one of the last living direct links to the leading scholars of a bygone generation’.²⁵ Whereas Ibn ʿAsākir uses this term only rarely in the TMD for anyone,²⁶ al-Dhahabī identifies several people as musnids, as we discuss below (he does not use the term for Ibn ʿAsākir himself).²⁷

When Ibn ʿAsākir cites Abu Ghalib jointly with another person, the other informant is most often Abu Ghalib’s brother, Abu 'Abd Allah.

Ibn ʿAsākir quotes at least forty-six authors on Abu Ghalib’s authority. The most frequently mentioned of these authors is al-Zubayr b. Bakkar (d. 256/870, cited 653 times), followed by Ibn Sa'd (d. 230/845, 627 times). The other forty-four authors he cites appear less frequently, and together, citations of these authors make up only about half of all of his author references.

3. Abu al-Barakat

Born	462/1069–70
Died	538/1143
Where he met Ibn ʿAsākir per Muʿjam al-shuyūkh	Baghdad
Total # if his isnāds in TMD	3,449
Percentage of all isnāds in data set	4%
Transmission term used in Muʿjam al-shuyūkh’s example	akhbaranā/bi-qirāʾatī ʿalayhi
Most common transmission term in TMD	akhbaranā
# of his isnāds using this term in TMD	3,312
Percentage of his isnāds using this term in TMD	96%
Name as given in TMD bio	NA
Name as given in Muʿjam al-shuyūkh	عبد الوهاب بن المبارك بن أحمد بن الحسن بن بندار أبو البركات الأنماطي الحافظ
Name as given in al-Dhahabī, Siyar aʿlām al-nubalāʾ	الأنماطي عبد الوهاب بن المبارك بن أحمد الشيخ الإمام الحافظ المفيد الثقة المسند بقية السلف أبو البركات عبد الوهاب بن المبارك بن أحمد بن الحسن بن بندار البغدادي الأنماطي
Minimum # of surface forms of his name in TMD	43
Sample surface forms	أبو البركات أبو البركات أبو البركات ابن المبارك أبو البركات الأنماطي ابن المبارك
Most common surface form	أبو البركات الأنماطي
# of his isnāds using most common surface form	2,108
Percentage of his isnāds using most common surface form	61%
Minimum # of times named alongside other direct informants in TMD by most common surface form	619
Minimum # of times cited as direct informant along with others	395
# of isnāds citing him as direct informant that also contain an author’s name	1,211
Percentage of isnāds citing him as direct informant that contain an author’s name	35%
# of distinct author names in isnāds that cite him	35
Author names in these isnāds (normalised)	Abd Allah b. Ahmad, Abu 'Abd Allah al-Balkhi, Abu 'Ubayd al-Qasim b. Sallam, Abu Bakr al-Nisaburi, Abu Bakr al-Wasiti, Abu Bishr al-Dulabi, Abu Dawud, Abu Muhammad b. Abi Nasr, Abu Nu'aym, Abu al-Ghana'im, Ahmad b. Hanbal, al-Haytham b. 'Adi, Harun b. Hatam, Ibn Abi Shayba, Ibn Ishaq, Ibn Ma'in, Ibn Manda, Ibn Sa'd, Khalifa, Ma'mar b. Rashid, Muhammad b. Harun, Muhammad b. Jarir, Muhammad b. Muhammad, Waki', Ya'qub b. Sufyan, Yahya b. Muhammad b. Sa'id, al-'Uqayli, al-Baghawi, al-Bayhaqi, al-Bukhari, al-Faryabi, al-Harith b. Abi Usama, al-Tabarani, al-Waqidi, al-Zuhri

Table 3.9: The informant profile of Abu al-Barakat.

Ibn ʿAsākir cites Abu al-Barakat consistently in about 3,312 isnāds with the phrase ‘he informed us’ (akhbaranā). In the Muʿjam al-shuyūkh, Ibn ʿAsākir places him in Baghdad and reports having recited material back to Abu al-Barakat.

Al-Dhahabī refers also to Abu al-Barakat as a musnid. He quotes Ibn ʿAsākir’s contemporary al-Samʿānī (d. 562/1166), the author of Kitāb al-ansāb, as saying of Abu al-Barakat:

He was a ḥāfiẓ, trustworthy, precise and with a vast number of authorised transmissions (wāsiʿ al-riwāya). He was constant, quick to weep and good company. He produced histories and gathered together narratives on uncovered topics. He was occupied with disseminating Hadith. I read a great deal back to him.²⁸

The term ḥāfiẓ here does not mean someone who has memorised the Quran but rather is used as an honorific for a Hadith expert who has memorised a large amount of material.²⁹ A riwāya is a citation chain that grants those included in it authority to transmit a work onward (see Post 5 for further discussion of this term and its interpretation).

Al-Samʿānī goes on to say, as quoted by al-Dhahabī:

It might even be said that there was no volume that he had not read and obtained a copy of. He copied the great works, such as the Ṭabaqāt of Ibn Saʿd [Ibn Sa'd] and the Tārīkh of al-Khaṭīb [al-Khatib]. He was devoted to riwāyas and would not permit an ijāza to follow after an ijāza. He composed something on that. I read to him al-Jaʿdiyyāt, the Taʿrīkh al-Fasawī and the Selections by al-Baqqāl from al-Mukhalliṣ (Intiqāʾ al-Baqqāl ʿalā al-Mukhalliṣ).³⁰

Ibn ʿAsākir mentions several authors in isnāds for which Abu al-Barakat is the direct informant; they include Ibn Sa'd, but not al-Khatib. The author whose name appears most often Abu al-Barakat’s isnāds (388 times) is Ibn Abi Shayba. But none of these citations refers to a work by him.

4. Abu Muhammad al-Akfani

Born	444
Died	524/1129
Where he met Ibn ʿAsākir per Muʿjam al-shuyūkh	Damascus
Total # of his isnāds in TMD	3,380
Percentage of all isnāds in data set	4%
Transmission term used in Muʿjam al-shuyūkh’s example	akhbaranā/bi-qirāʾatī ʿalayhi
Most common transmission term in TMD	akhbaranā
# of his isnāds using this term in TMD	2,686
Percentage of his isnāds using this term in TMD	79%
Name as given in TMD bio	هبة الله بن أحمد بن محمد ابن هبة الله بن علي بن فارس أبو محمد بن أبي الحسين ابن أبي الفضل الأنصاري المعروف بابن الأكفاني
Name as given in Muʿjam al-shuyūkh	هبة الله بن أحمد بن محمد بن هبة الله بن علي بن فارس أبو محمد بن أبي الحسين بن أبي الفضل بن الأكفاني الأنصاري المزكي
Name as given in al-Dhahabī, Siyar aʿlām al-nubalāʾ	ابن الأكفاني أبو محمد هبة الله بن أحمد بن محمد الدمشقي المعدل المعروف بابن الأكفاني
Minimum # of surface forms of his name in TMD	38
Sample surface forms	أبو محمد ألاكفاني أبا محمد بن الأكفاني أبو محمد ابن الأكفاني أبو محمد هبة الله بن أحمد بن الأكفاني
Most common surface form	أبو محمد بن الأكفاني
# of his isnāds using most common surface form	2,581
Percentage of his isnāds using most common surface form	76%
Minimum # of times named alongside other direct informants in TMD by most common surface form	93
Minimum # of times cited as direct informant along with others	162
# of isnāds citing him as direct informant that also contain an author’s name	N/A
Percentage of isnāds citing him as direct informant that contain an author’s name	N/A
# of distinct author names in isnāds that cite him	N/A
Author names in these isnāds (normalised)	N/A

Table 3.10: The informant profile of Abu Muhammad al-Akfani.

In his biographical entry on Abu Muhammad al-Akfani, al-Dhahabī says that he began participating in audition sessions (samiʿa) at the age of 9. Al-Dhahabī lists eleven people from whom Abu Muhammad al-Akfani received material through audition, beginning with Abu Muhammad’s father, and then adds, ‘a large number of people’ (wa-khalq kathīr). The list also includes al-Khatib. However, it contains no mention of what Abu Muhammad al-Akfani heard from these individuals.³¹

We have already mentioned that Ibn ʿAsākir’s older brother Hibat Allāh reportedly transmitted material to him via audition (sammaʿahu) from the year 505/1111–12 onwards. Yāqūt adds that Ibn ʿAsākir received independent auditions from his father and from Abu Muhammad al-Akfani (samiʿa huwa bi-nafsihi min wālidihi wa-Abī Muḥammad al-Akfānī).³² This seems to suggest that Abu Muhammad al-Akfani held pride of place as a direct informant for Ibn ʿAsākir in the latter’s youth.

Our isnād data set also contains a specific and immediate reference by Ibn ʿAsākir to Abu Muhammad al-Akfani:

Abū Muḥammad Hibat Allāh b. al-Akfānī [Abu Muhammad al-Akfani] informed me orally (shifāhan) and I transmit it from his handwriting that he said that Abū Muḥammad ʿAbd Allāh b. al-Ḥasan b. Ṭalḥa b. al-Naḥḥās al-Tinnīsī (may God be pleased with him) told him via an ijāza that Abū al-Ḥasan ʿAbd al-Salām …³³

Indeed, Ibn ʿAsākir emphasises having listened to Abu Muhammad al-Akfani personally on other occasions in the TMD, where he says repeatedly, using variations of the term shifāhan, that Abu Muhammad al-Akfani informed him of things orally.³⁴

Our data model assumes that whichever author mentioned in an isnād as the closest in time to Ibn ʿAsākir is the author on whose work he draws for the material transmitted through that isnād. A subset of our isnād data set collects the chains that run back to individuals identified as authors. We have not recorded isnāds going through him to other authors in this subset.

This case illustrates the complexities of authorship in the period in question and the challenge these pose for our data collection. Scheiner classifies Abu Muhammad al-Akfani as an author on the basis of a work ascribed to him entitled Tasmiyat quḍāt Dimashq. However, as we discuss in Post 6, the term tasmiya is used widely in the TMD for written collations that often do not fit the model of a composed book (separate pieces but also sections of larger works). If authorship of a tasmiya makes an individual an author, the TMD contains far more authors than hitherto reckoned. Looking closely at the data set, we find that Abu Muhammad al-Akfani passed on quotations from several (other) authors on our list, including al-Khatib (in 309 isnāds). Ibn ʿAsākir’s frequent use of tasmiyas suggests that these works mediated his access to the materials conveyed in them.

5. Abu Muhammad al-Sulami

Born	?
Died	526/1132
Where he met Ibn ʿAsākir per Muʿjam al-shuyūkh	Damascus
Total # of his isnāds in TMD	3,224
Percentage of all isnāds in data set	4%
Transmission term used in Muʿjam al-shuyūkhs example	akhbaranā/bi-qirāʾatī ʿalayhi
Most common transmission term in TMD	qaraʾtu ʿalā
# of his isnāds using this term in TMD	1,913
Percentage of his isnāds using this term in TMD	59%
Name as given in TMD bio	عبد الكريم بن حمزة بن الخضر بن العباس أبو محمد السلمي الحداد أخو سليمان وكيل المقربين
Name as given in Muʿjam al-shuyūkh	عبد الكريم بن حمزة بن الخضر بن العباس أبو محمد السلمي الحداد الوكيل المعروف بأخي سلمان
Name as given in al-Dhahabī, Siyar aʿlām al-nubalāʾ	عبد الكريم بن حمزة بن الخضر بن العباس الشيخ الثقة المسند أبو محمد السلمي الدمشقي الحداد وكيل المقرئين
Minimum # of surface forms of his name in TMD	54
Sample surface forms	عبد الكريم بن حمزة أبي محمد بن السلمي أبي محمد بن حمزة السلمي أبي محمد عبد الكريم السلمي أبي محمد عبد الكريم بن حمزة السلمي
Most common surface form	أبي محمد السلمي
# of his isnāds using most common surface form	1,341
Percentage of his isnāds using most common surface form	42%
Minimum # of times named alongside other direct informants in TMD by most common surface form	0
Minimum # of times cited as direct informant along with others	96
# of isnāds citing him as direct informant that also contain an author’s name	1,466
Percentage of isnāds citing him as direct informant that contain an author’s name	45%
# of distinct author names in isnāds that cite him	36
Author names in these isnāds (normalised)	Abd al-Razzaq, Abu 'Aruba al-Harrani, Abu 'Ubayd al-Qasim b. Sallam, Abu Hudhayfa, Abu Muhammad al-Akfani, Abu Muhammad b. Abi Nasr, Abu Nu'aym, Abu Zur'a, Ahmad b. al-Ma'ali, Al-Haytham b. 'Adi, Hisham b. 'Ammar, Ibn Abi Khaythama, Ibn Ma'in, Ibn Makula, Ibn Manda, Ibn Sa'd, Ibn al-A'rabi, Ibn al-Mubarak, Muhammad b. 'A'idh, Muhammad b. Harun, Muhammad b. Jarir, Muhammad b. Muhammad, Waki', Ya'qub b. Sufyan, al-Baghawi, al-Bukhari, al-Daraqutni, al-Kalbi, al-Khara'iti, al-Khatib, al-Kilabi, al-Mada'ini, al-Tahawi, al-Walid b. Muslim, al-Waqidi, al-Zuhri

Table 3.11: The informant profile of Abu Muhammad al-Sulami.

Ibn ʿAsākir cites Abu Muhammad al-Sulami in a variety of ways, but most often he uses the phrase qaraʿtu ʿalā, indicating that he read the transmitted material back to his informant. This mode of transmission is not common among Ibn ʿAsākir’s top five informants, but it constitutes a dominant pattern in the TMD generally, as we discuss in Post 4.

Al-Dhahabī calls Abu Muhammad al-Sulami a musnid, like Abu Ghalib and Abu al-Barakat, and he quotes Ibn ʿAsākir as saying that he read many items back to him (qaraʾtu ʿalayhi al-kathīr).³⁵

Forty-five percent of Abu Muhammad al-Sulami’s citations in the data set contain the name of an author – a relatively high percentage. One might hypothesise a connection with his mode of citation: perhaps the transmission term qaraʾtu ʿalā goes back to an author-informant more often than do other modes of citation. When we surveyed all isnāds that cite authors, we did not find this to be generally the case. Isnāds that include authors do not begin with qaraʾtu ʿalā more often than other isnāds do. However, in the specific case of Abu Muhammad al-Sulami, there does seem to be a correlation between the term qaraʿtu ʿalā and material obtained from an author. Of Abu Muhammad’s isnāds that feature authors, 57% begin with qaraʾtu ʿalā (831 out of 1,466 isnāds). In 539 isnāds, Abu Muhammad al-Sulami is preceded in the chain by Ibn Makula (d. 485/1092), who is credited with authoring books, and all but nine of these start with the formula qaraʾtu ʿalā.

A few more observations about the top five informants

Each of the five persons whom Ibn ʿAsākir cites most frequently as his direct sources represents a unique case, but they enable some general observations.

In terms of geography, Ibn ʿAsākir occasionally gives geographical information about his direct informants in isnāds. He mentions receiving information in Damascus but more often refers to locations beyond Damascus. It is hard to know how much of the TMD comes from sources outside of Damascus. The TMD connects to Damascus many people who merely passed through it or lived there briefly. His direct informants likewise often hail from elsewhere, which makes sense if we assume that he relied on notes from his youthful study trips when composing the TMD. Moreover, given Baghdad’s important role both historically and in his own day, it might be natural that he would seek to cite Baghdadi sources in particular.

Al-Dhahabī labels four of Ibn ʿAsākir’s top five transmitters musnids. Ibn ʿAsākir might well have valued these direct informants especially highly because of the access they brought to important isnāds. Ibn ʿAsākir’s reliance on these direct informants in particular supports the possibility that transmission itself was a key focus of the TMD.³⁶

Although the term musnid does appear at least once in the TMD, it is not one that Ibn ʿAsākir generally employs. (Yāqūt does not use the term either, at least as far as we have been able to establish through searching.) Al-Dhahabī uses the label musnid throughout the Siyar aʿlām al-nubalāʾ. For example, he says of one ʿAbd al-Malik b. ʿUmayr b. Suwayd that he ‘became the musnid for the people of Kufa’.³⁷ A search of the Siyar aʿlām al-nubalāʾ yields references to such isnād specialists in many (or even most) centres of Islamic learning in the centuries up to al-Dhahabī’s own day. He names musnids for al-Andalus, the Maghrib, Cairo/Egypt (Miṣr and al-diyār al-miṣriyya), the Ḥijāz, Palestine (ahl Filistīn), Syria (al-Shām and ahl al-Shām fī zamānihi), Iraq, Jurjān, Khurāsān and Transoxania as well as various cities and other locations, including Baghdad, Basra, Cordoba, Marrakech, Alexandria, Mecca, the Ḥaram (i.e. the Temple Mount/Jerusalem), Damascus, Homs, Aleppo, Kufa, Wāsiṭ, Dīnawar, Hamadhān, Isfahan, Nishapur, Samarqand, Herat, Bukhara and Merv. There are also many vague references to individuals as the musnids of their times (musnid waqtihi), eras (musnid ʿaṣrihi), regions (musnid tilka al-diyār fī zamānihi) and the world generally (musnid al-āfāq and musnid al-dunyā). Al-Dhahabī also mentions a specialist in isnāds as the leader of Quran reciters and Hadith specialists (musnid al-qurrāʾ wa-l-muḥaddithīn).³⁸ And there are references to especially long-lived (al-muʿammar) isnād specialists, a feature that is significant if one is seeks transmission chains that are as short as possible. Al-Dhahabī appears to have appreciated the special role of these specialists in ‘carrying on the tradition’.³⁹

In his isnāds, Ibn ʿAsākir’s most commonly uses the expression akhbaranā, ‘he informed us’, to explain how he acquired information. With four of his five top direct informants, Ibn ʿAsākir typically says that he received the material as part of a group. This wording is common throughout the TMD, as we discuss in Post 4.

Each of the five informants transmits reports that contain authors’ names, but only in the cases of Abu Ghalib and Abu Muhammad al-Sulami do such author-containing isnāds represent a significant percentage of the informant’s total isnāds. It may be significant that with Abu Muhammad al-Sulami specifically Ibn ʿAsākir favours a term indicating that he read the material back to him.

In general, the data set yields a complex picture of citation. We accept Ibn ʿAsākir’s implicit insistence that the isnāds are signifiers, but decoding them is not straightforward.

Citing people jointly

Ibn ʿAsākir repeatedly asserts that he acquired material while in a group. He employs a variety of transmission terms to indicate that an informant relayed information to ‘us’. In our data, he rarely says that a direct informant provided information to him alone (‘to me’), unless an explicitly written method of transmission is involved.

This tendency to refer to collectivities also occurs in another, distinct sense within the TMD. Ibn ʿAsākir frequently credits two, three or more people together, at the start of an isnād, for the same information. For example, consider the following isnād:

وأخبرنا أبو غالب وأبو عبد الله ابنا البنا قالا أنا أبو الحسين بن الآبنوسي أنا أبو الحسن الدارقطني إجازة

Abū Ghālib [Abu Ghalib] and Abū ʿAbd Allāh [Abu 'Abd Allah], the two sons of al-Bannāʾ, both informed us, saying that Abū al-Ḥusayn b. al-Abnūsī told them that Abū al-Ḥasan al-Dāraquṭnī told them via an ijāza (ijāzatan) that …

By our very rough estimates – each calculated in a different way – more than 12% of the isnāds in our data set feature more than one direct informant.⁴⁰ Abu Ghalib and Abu 'Abd Allah, the Bannāʾ brothers, appear together in various ways at least 1,132 times.⁴¹

Names	Count
Abu Ghalib, Abu 'Abd Allah	1,132
Abu al-Husayn, Abu 'Abd Allah al-Khallal	487
Abu al-Husayn b. al-Farra', Abu Ghalib, Abu 'Abd Allah	384
Abu al-Husayn, Abu 'Abd Allah	246
Abu Sa'd al-Mutriz, Abu 'Ali al-Haddad	233
Abu 'Abd Allah al-Furawi, Abu al-Muzaffar	225
Ibn al-Mubarak, Abu al-'Izz Thabit b. Mansur	219
Abu al-Barakat, Abu 'Abd Allah al-Balkhi	181
Abu al-Husayn al-Abraquhi, Abu 'Abd Allah al-Khallal	171
Abu 'Abd Allah, Abu Ghalib	155
Abu al-Qasim al-'Alawi, Abu al-Wahsh	153
Abu al-Barakat, Abu al-'Izz Thabit b. Mansur	126
Abu Talib, Abu Nasr	121
Abu al-Hasan b. Qubays, Abu Mansur b. Khayrun	120
Abu 'Ali, Jama'a	98
Abu al-Qasim al-'Alawi, Abu al-Hasan 'Ali b. Ahmad	73
Abu 'Abd Allah, Abu al-Qasim al-Samarqandi	62
Abu al-Husayn Hibat Allah b. al-Hasan, Abu 'Abd Allah al-Husayn b. 'Abd al-Malik	61
Abu al-Hasan al-Faradi, Abu al-Qasim al-Samarqandi	53
Abu al-Muzaffar, Abu al-Qasim al-Mustamli	53

Table 3.12: Sets of informants who are often cited together.

Table 3.12 offers only a sample of informants who are frequently cited together, and the counts are inevitably incomplete. They reflect a search for surface forms referring to the twenty most-cited informants amidst isnāds that cite more than one person in the first position in the isnād. Careful readers will note that the order in which names appear, within this position, matters, so that the same set may be counted several times if the individuals in it are listed in a different order. This situation is not ideal for getting a real handle on citations, and tackling it will require further work on the names and their normalisation.

What exactly does it mean when Ibn ʿAsākir cites people together?

The Bannāʾ brothers could plausibly have transmitted information to their audience together, but in many cases of joint citation there is no reason to think that they delivered the information in the same place and at the same moment in time. The combinations of names are too numerous and varied, and the known geography of many informants makes joint delivery even more unlikely. Furthermore, the Bannāʾ brothers are often listed alongside other people.

Bearing in mind that 39% of the TMD is made up of isnāds, we should expect that several different models of collation are at play. For example, the repeated citation of the same individuals in the same sequence within the first position within an isnād, using the same surface forms for their names, suggests that all the information may have come from the same notebook. In other cases, Ibn ʿAsākir may have started with a single-informant isnād for a particular report, but when he received corroboration for it from another direct informant, he may have added the additional source or sources to his isnād in the first position.

So far, we have been unable to directly link the contents and organisation of the TMD to Ibn ʿAsākir’s sources. It does not seem to be the case, for example, that he relied on specific notebooks containing information from certain sets of informants for particular sections of the work. Another project, distinct from ours, will consider the co-occurrence of certain names (including their various surface forms and orderings) alongside the reports that they accompany to determine whether the informant sets are defined by thematic or other principles. This project will also take into account the informants’ geography and the placement of the citations within the TMD. It will thus involve much more advanced work on pattern recognition, and it can illuminate how authors such as Ibn ʿAsākir collated their materials to produce their books.

1. The distribution is named after the Italian polymath Vilfredo Federico Damaso Pareto (d. 1923). Pareto’s original point pertained to income distribution – that a substantial portion of Italy’s wealth belonged to a relatively small percentage of the population (Pareto’s idea gained traction with the Fascists, as a picture of stable inequality suggested the pointlessness of efforts to change it). Pareto’s findings were based on tax tables from 1880 to 1890 from Prussia and Saxony, as well as Swiss and Italian cities; as Thomas Picketty notes, the information was ‘scanty [and] covered a decade at most’. Thomas Piketty, Capital in the Twenty-First Century (Cambridge, MA: Belknap Press, 2014), 462–5. ↩

2. As noted in post 1, our edition of the Muʿjam al-shuyūkh contains 1,621 shaykhs, slightly fewer than in al-Dhahabī’s count. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:556, ms. 6648. ↩

3. As nicely summarized by Lorenz Nigst in a personal communication on 10 May 2023. ↩

4. Yāqūt, Muʿjam al-udabāʾ, 4:1698, entry 743, ms. 1199. ↩

5. We exclude Abu Ghalib and Abu 'Abd Allah’s shared isnāds from this figure, since the two transmitters’ individual counts include cases in which they are listed with other people. ↩

6. For a report according to which Ibn ʿAsākir was present at the recitation of one Abu 'Abd Allah Muhammad b. 'Ali, see Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 62784, ms. 21493. ↩

7. Yāqūt, Muʿjam al-udabāʾ, 4:1698, entry 743, ms. 1199; al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:554, ms. 6647. ↩

8. Ibn al-Ṣalāḥ al-Shahrazūrī, Muqaddimat ʿulūm al-ḥadīth, ed. Nūr al-Dīn ʿItr (Beirut: Dār al-Fikr al-Maʿāṣir, 1397/1977), 129, 06430643IbnSalahShahrazuri.MuqaddimatCulumHadith.JK000537, ms. 054–5; Ibn al-Ṣalāḥ al-Shahrazūrī, An Introduction to the Science of the Ḥadīth (Kitāb Maʿrifat anwāʿ ʿilm al-ḥadīth), trans. Eerik Dickinson (Reading: Garnet, 2005), 96–7. We mostly use Dickinson’s translation for our quotations, but with slight modifications. For his translation, Dickinson relied on an earlier version (Damascus 1387/1966) of the edition cited here and secondarily on the edition of ʿĀʾisha ʿAbd al-Raḥmān (2^nd^ ed., Cairo: Dār al-Maʿārif, 1989); ʿAbd al-Raḥmān’s edition in the OpenITI corpus is 0643IbnSalahShahrazuri.MuqaddimatCulumHadith.Shamela0029878BK1. Dickinson explains his edition in Ibn al-Ṣalāḥ, Introduction, xxv. In these blog posts we cite ʿItr’s 1397/1977 edition. See also Jonathan Berkey, The Transmission of Knowledge in Medieval Cairo: A Social History of Islamic Education (Princeton: Princeton University Press, 1992), 32–3. ↩

9. Ibn al-Ṣalāḥ, Introduction, 99; Ibn al-Ṣalāḥ, Muqaddima, 134, ms. 056. ↩

10. Abū ʿAbd Allāh Muḥammad b. ʿAbd Allāh al-Ḥākim al-Naysābūrī, Maʿrifat ʿulūm al-ḥadīth (Beirut: Dār al-Kutub al-ʿIlmiyya, 1977), 261, 0405HakimNaysaburi.MacrifatCulumHadith.Shia002093, ms. 240. See also J. Robson, ‘Ḥadīth’, in Encyclopaedia of Islam, 2^nd^ ed., http://dx.doi.org/10.1163/1573-3912_islam_COM_0248. ↩

11. It may be relevant that he cites Abu al-Ghana'im alone, not alongside other persons (as is his frequent practice with other direct informants). He does cite others alongside Abu 'Ali al-Haddad, however. ↩

12. Ibn al-Ṣalāḥ, Introduction, 100; Ibn al-Ṣalāḥ, Muqaddima, 137, ms. 057–8. ↩

13. Ibn al-Ṣalāḥ, Introduction, 114–15; Ibn al-Ṣalāḥ, Muqaddima, 160, ms. 069. ↩

14. Ibn ʿAsākir, Muʿjam al-shuyūkh, 236, entry 268, ms. 119. As Devin Stewart notes, this could mean that he sent him an ijāza in a letter. Personal communication, 15 July 2023. ↩

15. Ibn ʿAsākir, Muʿjam al-shuyūkh, 1001, entry 1283, ms. 516. ↩

16. Ibn ʿAsākir, Muʿjam al-shuyūkh, 161, entry 177, ms. 081. His brother was ʿAbd Allāh b. Aḥmad b. ʿUmar b. Abī al-Ashʿath Abū Muḥammad b. Abī Bakr al-Samarqandī: Ibn ʿAsākir, Muʿjam al-shuyūkh, 460, entry 553, ms. 230. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:28–31, ms. 6398–9. ↩

17. See, for example, Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 89, ms. 00039; ID 7249, ms. 02835; ID 15037, ms. 05607; and ID 49896, ms. 17054. ↩

18. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 7249, ms. 02835. ↩

19. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:28, ms. 6398. ↩

20. Mourad, Ibn ‘Asakir of Damascus, 22. ↩

21. As we prepared to publish these posts, we found three additional citations of al-Shirazi’s book through Abu al-Qasim al-Samarqandi that we missed through our search. ↩

22. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 3756, ms. 01529; ID 4157, ms. 01669; ID 61666, ms. 21047; ID 61909, ms. 21148; ID 62238, ms. 21277; and ID 62238, ms. 21277. ↩

23. Ibn ʿAsākir, TMD, 54:159; Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 62238, ms. 21277. ↩

24. Abū Isḥāq al-Shīrāzī, Ṭabaqāt al-fuqahāʾ, ed. Iḥsān ʿAbbās (Beirut: Dār al-Rāʾid al-ʿArabī, 1970), 128, 0476AbuIshaqShirazi.TabaqatFuqaha.Shamela0001031-ara1.mARkdown, ms. 52. ↩

25. Devin Stewart, ‘Women’s Biographies in Islamic Societies: Mīrzā ʿAbd Allāh al-Iṣfahānī’s Riyāḍ al-ʿUlamāʾ’, in Louise Marlow (ed.), The Rhetoric of Biography: Narrating Lives in Persianate Societies (Cambridge, MA: Harvard University Press, 2011), 113–14. ↩

26. For this sense in the TMD, see Ibn ʿAsākir, TMD, 5:104, ms. 1455. Ibn ʿAsākir here describes someone as a musnid among the shaykhs of Basra. Searching for the term musnid requires distinguishing it from the orthographically identical musnad (a Hadith collection organized by transmitter) and musnid in the literal sense of ‘reclining’ or ‘leaning’, which is also attested in the TMD. ↩

27. See Abū Anas Waqīʿ Allāh al-Marwatī’s analysis of the term musnid, quoting al-Suyūṭī and others, in https://www.alukah.net/sharia/0/123404/إطلاق-كلمة-المسند-في-مصطلح-الحديث/. Al-Suyūṭī ranks a musnid behind a muḥaddith and a ḥāfiẓ: al-Suyūṭī, Tadrīb al-rāwī fī sharḥ Taqrīb al-Nawāwī, ed. Abū Qutayba Naẓar Muḥammad al-Fārayābī, 2 vols ([Beirut: Dār al-Kalim al-Ṭayyib], [1996]), 1:29–30, 0911Suyuti.TadribRawi.Shamela0009329, ms. 4–5. ↩

28. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:135, ms. 6446. ↩

29. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:554, ms. 6646. Ibn ʿAsākir himself is called a ḥāfiẓ in this sense by his son, as quoted by Yāqūt (al-Dhahabī likewise uses the term for him). Yāqūt’s text also declares Ibn ʿAsākir one of the imams of Hadith (al-Dhahabī, for his part, calls Ibn ʿAsākir muḥaddith al-Shām). Yāqūt, Muʿjam al-udabāʾ, 4:1697, entry 743, ms. 1199. ↩

30. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 20:135–6, ms. 6446–7. He may be referring to Abū al-Qāsim al-Baghawī’s (d. 317/929) collection of Hadith from ʿAlī b. Jaʿd (d. 230/844–5), published as al-Jaʿdiyyāt, and to Ibn Sufyān al-Fasawī’s (d. 277/890) Maʿrifa wa-l-tārīkh. The third work is selections (intiqāʾ) by Abū Bakr Aḥmad b. ʿUmar b. al-Baqqāl (d. 489/1096)from a work by Abū Ṭāhir Muḥammad b. ʿAbd al-Raḥmān b. al-ʿAbbās (d. 393/1003). The earlier work is called the Mukhalliṣiyyāt. See also Abū Ṭāḥir al-Mukhalliṣ, al-Mukhalliṣiyyāt wa-ajzāʾ ukhrā li-Abī Ṭāhir al-Mukhalliṣ, ed. Nabīl Saʿd al-Dīn Jarrār, 4 vols ([Doha]: Wizārat al-Awqāf wa-l-Shuʾūn al-Islamiyya li-Dawlat Qaṭar, 1429/2008) and al-Dhahabī, Siyar aʿlām al-nubalāʾ, 16:478–9, ms. 5484–5. We capitalise ‘Selections’ to reflect what sounds like a fixed work. ↩

31. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 19:576–8, ms. 6361. Al-Dhahabī quotes Ibn ʿAsākir in his entry on Abu Muhammad al-Akfani. ↩

32. Yāqūt, Muʿjam al-udabāʾ, 4:1698, entry 743, ms. 1199. ↩

33. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 40053, ms. 13726. ↩

34. See within our isnād data set the subset ‘Isnads_with_Search_Terms’, table ‘TC_Shifahan’. ↩

35. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 19:600, ms. 6371. ↩

36. Following Davidson, Carrying On the Tradition. ↩

37. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 5:439, ms. 2360. ↩

38. Al-Dhahabī, Siyar aʿlām al-nubalāʾ, 19:593, ms. 6368. ↩

39. See also Devin Stewart’s definition of this term in Stewart, ‘Women’s Biographies’, 113–14. ↩

40. One of the methods we used for these estimates involved searching for indications of multiple entities in the first position in isnāds (when the names were normalised, such indications might include a comma). ↩

41. They are often cited together with additional persons. We have not disambiguated all of the names that appear in conjunction with the brothers. ↩