Post 4: Ibn ‘Asakir’s Transmission Terms in Isnads

Our previous blog post featured a deep dive into the pool of informants whom Ibn ʿAsākir cites frequently. Now we turn to the big picture of how he says he acquired his information.

Primary terms

We collected isnād citations by splitting the text using the # paragraph marker plus a list of transmission terms. Although we started with a long list of possible terms for splitting the text, we found only the following groups to be useful (insofar as other terms generally produced, for example, not isnāds followed by reports, but only quoted speech).

Term (most common form)	Translation	Count	Individual transmissions
أخبرنا	he informed us	53,841	<1%
أنبأنا	he communicated to us	8,462	<1%
ح[تحويل/حيلولة]	transition [to a new set of isnāds]	6,195	3%
قرأت على	I read/recited to	5,949	90%
حدثنا	he transmitted to us	1,244	33%
كتب إلي	he wrote to me	1,001	100%

Table 4.1: The most common transmission terms in the TMD’s isnāds. By ‘individual transmissions’, we mean instances in which Ibn ʿAsākir reports having received the material individually rather than as part of a group.

By far the most common term that Ibn ʿAsākir uses to describe the way his direct informants transferred information to him is akhbaranā (‘he informed us’). This is the term he uses most often for his transmissions from Abu al-Qasim al-Samarqandi, Abu Ghalib, Abu al-Barakat and Abu Muhammad al-Akfani, four of the five people he cites most frequently in the TMD. It is also the most common term used with all other direct informants if these four are excluded from the count. The next most common term is anbaʾanā (‘he communicated to us’) and its variants. Another noteworthy term is an abbreviation (the Arabic letter ḥ) for the word taḥwīl, which marks the transition from one isnād or group of isnāds to another, new one. It is always followed by another transmission term (most often akhbaranā, ‘he informed us’).¹ We also tracked Ibn ʿAsākir’s uses of kataba ilayya, because the phrase is often followed by the name of a direct informant (and sometimes with an entire isnād).

What is the precise meaning of these terms? What do they reveal about Ibn ʿAsākir’s working methods?

Ibn ʿAsākir does not define his terminology for us. We find some helpful definitions outside the TMD, but caution is in order. We should not assume continuity in meanings across time, nor can we take it for granted that Ibn ʿAsākir shared another scholar’s definitions. That said, Ibn al-Ṣalāḥ, writing less than a century after Ibn ʿAsākir, has much to say about Hadith transmission that speaks to what Ibn ʿAsākir is doing in the TMD. Ibn al-Ṣalāḥ’s Kitāb Maʿrifat anwāʿ ʿilm al-ḥadīth is often taken to reflect the technical vocabulary of Hadith transmission after its crystallisation, and it thus deserves a close reading.²

Akhbaranā, ḥaddathanā and anbaʾanā

Ibn al-Ṣalāḥ describes eight ways of conveying and receiving Hadith, including audition of the speech of the teacher (al-samāʿ min lafẓ al-shaykh) and recitation to the teacher (al-qirāʾa ʿalā al-shaykh). Ibn al-Ṣalāḥ’s comments fit well with Ibn ʿAsākir’s usage of his most common transmission terms in Table 4.1:³

• akhbaranā (‘he informed us’): Ibn al-Ṣalāḥ reports that this term is used extensively and has become, by his time, the primary way of referring to the transfer of information by reciting material back to one’s teacher.

• ḥaddathanā (‘he transmitted to us’): This is a more ambiguous term according to Ibn al-Ṣalāḥ, with some scholars using it even for material transmitted to them through ijāza (written means). Still, it normally carries a ‘strong implication of pronunciation and oral communication’.⁴

• anbaʾanā (‘he communicated to us’): This term is rarely used for material heard via the speech of a teacher.

Ibn al-Ṣalāḥ describes the challenges of interpreting the meanings intended by scholars when they use the terms akhbaranā and ḥaddathanā and concludes:

Differentiating between ‘He transmitted to us’ (ḥaddathanā) and ‘He informed us’ (akhbaranā) has become widespread, predominating among the scholars of Hadith. The attempt to justify the differentiation from the standpoint of linguistic usage is laboured and contrived. The best that may be said for it is that it is a convention among the scholars of Hadith who sought by observing it to discriminate between the two means of receiving Hadith.⁵

He then discusses the issue of passing on isnāds and their specific transmission terms. He advises against replacing the transmission term used in an isnād with another one even when one is certain that the quoted person considered the two terms equivalent. Such changes might be permissible under a principle that allows paraphrasing Hadith but, he says, ‘We think it best to avoid doing anything along these lines in regard to changing the contents of books and collections of Hadith.’⁶

This leaves us better off, though not fully satisfied. Ibn al-Ṣalāḥ is telling us that akhbaranā and ḥaddathanā both generally signify in-person transmission and that akhbaranā typically points to a student reading material back to a teacher. The formulas that Ibn ʿAsākir uses for the sample Hadiths he quotes from his top four direct informants, akhbaranā and bi-qirāʾatī ʿalayhi (‘by my reading back to him’), fit well with Ibn al-Ṣalāḥ’s information. So far so good.

But the meaning of the term ḥaddathanā is less clear-cut. Although Ibn al-Ṣalāḥ argues that it generally implies oral communication, he acknowledges that it can be ambiguous, and for this reason we (following Ibn al-Ṣalāḥ’s translator, Eerik Dickinson) translate it as ‘he transmitted to us’. By contrast, when writing about al-Ṭabarī’s citation practices, we translated ḥaddathanā as ‘he told us’, because al-Ṭabarī’s usage suggested that for him the term denoted the acquisition of material in person from his direct informants. We are less sure that the same is true of Ibn ʿAsākir. Therefore, whether the two terms can or should be distinguished remains an open question.⁷

As we noted in Post 3, Ibn ʿAsākir often uses the transmission term anbaʾanā in situations in which he would have been very young at the time he encountered the direct informant he is quoting. Ibn al-Ṣalāḥ’s understanding of anbaʾanā as generally referring to indirect transmission again seems apt for Ibn ʿAsākir, who may have used the term for the practice of acquiring information via intermediaries and their notes, even if he subsequently omitted these intermediaries from his isnāds.

We gain more certainty when Ibn ʿAsākir uses other transmission terms alongside those in Table 4.1. However, he provides this added precision in only 5% of the citations in our data set, as we discuss below.

Another point of ambiguity is Ibn ʿAsākir’s predominant use of the first-person plural rather than the singular (‘he informed us’, not ‘he informed me’) in his descriptions of how information was passed on to him. Does this mean that he in fact received the information as part of a group? Ibn al-Ṣalāḥ cites al-Khatib as holding that it is desirable but not mandatory to distinguish between ‘me’ and ‘us’ in Hadith transmission. However, he concludes:

Thus, it is permissible for the student, when he hears Hadith by himself, to say, ‘He transmitted to us’, or something similar, because in the speech of the Arabs that is permissible for a single individual. Furthermore, he is permitted, when he hears Hadith as a part of a group, to say, ‘He transmitted to me’, because the teacher did transmit to him as well as transmitting to others. God knows best.⁸

This ambiguity is irresolvable. We cannot be certain why Ibn ʿAsākir favoured the plural – whether this was simply a linguistic preference or indicated actual collective transmission (the latter implying a lesson with audience and witnesses). But the marked predominance of the plural is clearly not accidental: Ibn ʿAsākir chose to express himself in this way, which suggests that he might have sought to convey that transmission typically occurred within groups and had a social dimension.

A final question – perhaps the elephant in the room – is whether Ibn ʿAsākir actually means anything at all with these terms. The fact that he uses different transmission terms for the same individuals, as we saw with his top five direct informants, raises the possibility that the variance is in fact meaningless.

Although the question remains open, we believe that he did generally (though perhaps not always) have something specific in mind when he used one term versus another. This conclusion is supported by his care in collating isnāds, specifying, for example, the different ways in which two different direct informants gave him information (one qirāʾatan and the other shifāhan). A further indication is that he sometimes starts an isnād with one person using one transmission term and then introduces another person with a different term before running through the rest of the isnād up to the report. This practice strongly suggests a deliberate effort to distinguish the respective methods of transmission used by the two people.⁹

It is hard to imagine how Ibn ʿAsākir would have kept track of dozens of transmitters giving him information in different ways. Perhaps he associated each individual’s corpus of material loosely with a particular context and used the appropriate term when writing the TMD. But memory is fallible; did he also record this information in his notes? Incomplete notes and lapses of memory might explain why Ibn ʿAsākir occasionally uses different terms for the same people in different contexts within his work. In other words, he obtained information via different routes and later remembered them with only partial consistency.

Qaraʾtu ʿalā

Two of the transmission terms that Ibn ʿAsākir uses in our isnād data set refer unambiguously to written materials.

Most importantly, Ibn ʿAsākir employs the phrase qaraʾtu ʿalā (‘I read/recited to’) 5,328 times. The plural form of the phrase, qaraʾnā ʿalā (‘we read/recited to’), appears 621 times. This means that on 90% of the occasions on which he refers to reading/reciting material back to a direct informant, he says he was the one doing the reading. The following report via Abu Muhammad al-Sulami provides an example:

قرأت على أبي محمد السلمي عن أبي محمد بن أحمد أنا مكي بن محمد أناابو سليمان بن أبي محمد قال فيها يعني سنة خمس وثمانين مات واثلة بن الأسقع الليثي مات

I read back to Abū Muḥammad al-Sulamī [Abu Muhammad al-Sulami], and he said from Abū Muḥammad b. Aḥmad, who transmitted from Makkī b. Muḥammad, who transmitted from Abū Sulaymān b. Abī Muḥammad, who said: …¹⁰

Six of the people (or pairs of people) for whom Ibn ʿAsākir uses this transmission term most often rank among the top thirty most-cited direct informants. This mode represents a significant portion of his total engagements with them, but it is not the only one.

Direct informant	All citations	# of qaraʾtu ʿalā	Percentage of qaraʾtu ʿalā	Ranking among direct informants
Abu Muhammad al-Sulami	3,224	1,913	59%	5
Abu Ghalib	4,871	1,202	25%	2
Abu al-Fadl b. Nasir	1,011	711	70%	19
Abu 'Abd Allah	2,835	484	17%	6
Abu al-Qasim al-Mustamli	2,687	316	12%	7
Abu al-Fath Nasr Allah b. Muhammad	519	165	32%	32
Abu al-Wafa' Huffaz	148	148	100%	79
Abu Ghalib, Abu 'Abd Allah	1,132	116	10%	16
Abu al-Fadl	124	87	70%	90
Abu al-Qasim al-Khidr b. al-Husayn	130	76	64%	85

Table 4.2: The direct informants for whom Ibn ʿAsākir uses the phrase ‘I read/recited to’ most often in isnāds.

When Ibn ʿAsākir introduces an isnād with the phrase ‘I read/recited to’, one might expect either the direct informant or one of the persons further down the citation chain to be the author of a work. Conversely, one might also expect to find this or a similar phrase referring to a written text in isnāds that contain an author’s name.

But within our isnāds, the sense of reading conveyed by this term does not pertain specifically to reading text associated with or from someone recognised as an author. Of the 5,949 isnāds that begin with qaraʾtu ʿalā, 2,767 contain an author’s name (47%). Of the 31,532 isnāds we have collected that feature an author’s name, only 9% begin with this phrase. In other words, Ibn ʿAsākir uses qaraʾtu ʿalā only slightly more in author-citing isnāds (9%) than he does in other isnāds (8%, based on 5,949/77,231 isnāds), and rarely when an isnād features an author’s name. This illustrates a general point the recurs throughout these blog posts: Ibn ʿAsākir does not seem to employ a consistent and transparent transmission vocabulary for the ways in which his direct informants shared material with him from people we today recognize as authors.

Kataba ilayya

The second and much less frequent way in which Ibn ʿAsākir reports having received material through written means involves someone writing to him, as in this example:

كتب إلي أبو الحسين وابو عبد الله قالا أنا ابن مندة أنا حمد إجازة

Abū al-Ḥusayn [Abu al-Husayn] and Abū ʿAbd Allāh [Abu 'Abd Allah] wrote to me (kataba ilayya), saying that Ibn Manda told them that he was told by Hamad via an ijāza that …¹¹

We have translated the Arabic phrase kataba ilayya as ‘he wrote to me’ in Table 4.1. When we searched for this phrase at the start of paragraphs in the TMD, we found 1,063 results.¹² Ibn ʿAsākir credits a few people dozen of times with having written to him with information (see Table 4.3), but most direct informants who used this method of transmission did so only once, twice, or at most on three occasions. The people he cites most often in this way are generally not heavily represented in his isnāds as direct informants. The exception is Abu 'Ali al-Haddad, who used this method and is Ibn ʿAsākir’s ninth most-cited informant. Given its relative rarity within the TMD, this transmission mechanism does not seem to have played a major role in the work’s composition.

Name	Died	# of kataba ilayya	Total isnāds
Abu Nasr b. al-Qushayri	?	119	168
Abu Zakariyya Yahya b. 'Abd al-Wahhab	512/1118	97	104
Abu Bakr 'Abd al-Ghaffar b. Muhammad	?	49	73
Abu 'Ali al-Haddad	515/1121	44	1,950
Abu 'Abd Allah Muhammad b. Ahmad b. Ibrahim	?	42	128

Table 4.3: The direct informants whom Ibn ʿAsākir describes most often as writing to him with information (kataba ilayya).

If Ibn ʿAsākir obtained material from Abu Zakariyya Yahya b. 'Abd al-Wahhab and Abu 'Ali al-Haddad during their lifetimes, he would have been quite young at the time, since he was born in 499/1105 and both of these informants were dead by 515/1121.

Devin Stewart believes this phrase usually means that Ibn ʿAsākir received an ijāza giving him and isnād back to the work in question in the form of a letter (in other words, an ijāza by correspondence). Perhaps Ibn ʿAsākir’s father or teacher got him ijāzas by correspondence when he was young.¹³ This interpretation seems plausible.

Supplementary terms

In a few instances, representing 4.9% of his citations, Ibn ʿAsākir describes more specifically how he acquired the quoted information from his informants. In 19% of these cases (in 710 isnāds), he uses more than one of the terms listed in Table 4.4.

Term - most common Form	Translation	Count - with a direct informant
idhnan	with permission [to transmit onwards]	859
munāwalatan	by passing along [a text]	264
lafẓan	via [the informant’s own] speech	216
ʿāliyan	with elevation	802
shifāhan	orally	605
qirāʾatan	via reading back	561
ijāzatan	via an ijāza	200
fī kitābihi	in his [piece of] writing/book	976

Table 4.4: Supplementary terms used by Ibn ʿAsākir in conjunction with the primary terms in Table 4.1 to specify how information was transmitted to him.

We have translated these terms on the basis of a variety of sources, including Ibn al-Ṣalāḥ. We read the terms in context and sought to understand what they might signify in conjunction with the primary transmission terms.

There is no exclusive pairing of any primary term with a particular supplementary term. Collectively, these terms most often accompany ‘he informed us’ (akhbaranā): about 94% of the isnāds that contain supplementary terms start with this phrase or its variants.¹⁴ Within the overall data set, 70% of all isnāds begin with ‘he informed us’. The following example illustrates Ibn ʿAsākir’s use of these terms:

أخبرنا أبو العز أحمد بن عبد الله إذنا ومناولة وقرأ علي إسناده أنا محمد بن الحسين أنا المعافى بن زكريا القاضي نا محمد بن الحسن بن دريد أخبرني عمي الحسين بن

Abū al-ʿIzz Aḥmad b. ʿAbd Allāḥ [Abu al-'Izz Ahmad b. 'Abd Allah] informed us, with permission and by handing it over and reading its isnād back to me, that Muḥammad b. al-Ḥusayn transmitted that the judge al-Maʿāfī b. Zakariyyā transmitted that Muḥammad b. al-Ḥusayn …

On 802 occasions, Ibn ʿAsākir highlights the brevity of the transmission chains that connect his direct informants to the originators of the quoted information by using the phrase ʿāliyan, translated literally as ‘with elevation’. The phrase appears relatively seldom, in view of the total number of isnāds in the TMD and the general importance of elevation (that is, short isnāds) within Hadith transmission. Still, it is noteworthy that he refers to the length of isnāds at all. The informants for whom he uses this term most often are all people he cites frequently in the TMD. This suggests that these informants were generally good sources of short isnāds.¹⁵

The following sections discuss four supplementary terms in greater detail.

Shifāhan

Ibn ʿAsākir uses the term shifāhan and its cognate mushāhifatan 605 times in our data set’s isnāds. He uses them more than a hundred times to describe how he acquired information from Abu Muhammad al-Akfani (his fourth most frequently cited direct informant). A typical example is the following:

أخبرنا أبو محمد بن الاكفاني شفاها ناً عبد العزيز بن أحمد لفظاً أنا عبد الوهاب بن جعفر أنا عبد الجبار بن عبد الصمد أنا القاسم بن عيسى العصار نا إبراهيم بن

Abū Muḥammad b. al-Akfānī [Abu Muhammad al-Akfani] informed us orally (shifāhan) that ʿAbd al-ʿAzīz b. Aḥmad transmitted to him verbally (lafẓan) that …

What is the difference between shifāhan and lafẓan? Both could point to an ijāza being granted orally. Ibn al-Ṣalāḥ explains that by his time the phrase ‘he informed us orally’ (akhbaranā mushāfahatan) was used when someone had verbally granted an ijāza (kāna qad shāfahahu bi-l-ijāza lafẓan).¹⁶

But on another occasion Ibn ʿAsākir cites two direct informants using a different term for each, suggesting an effort to draw a finer distinction:

أخبرنا أبو الحسين الأبرقوهي إذنا وأبو عبد الله الخلال مشافهة قالا أنا أبو القاسم العبدي أنا أحمد إجازة

Abū al-Husayn al-Abraqūhī [Abu al-Husayn al-Abraquhi] informed us (akhbaranā) with permission to transmit onwards (idhnan), and Abū ʿAbd Allāh al-Khallāl [Abu 'Abd Allah al-Khallal] informed us orally (mushāfahatan), that Abū al-Qāsim al-ʿAbdī transmitted to them that Aḥmad told them via an ijāza that …

This language is detailed and seemingly specific. But why Ibn ʿAsākir uses them in the ways he does (with different terms for different people) is not clear to us at this point.

Qirāʾatan

Ibn ʿAsākir uses the expression qirāʾatan (‘via reading back’) 561 times within the isnād data set when explaining how a direct informant transmitted information to him.

The term appears to denote Ibn ʿAsākir’s reading text back to a direct informant – or his being in the presence of another person doing so. Its cognate bi-qirāʾatī ʿalayhi (‘via my reading back to him’) is the most common transmission term used in the sample Hadiths he provides in Muʿjam al-shuyūkh. Here is an example from the TMD:

أخبرنا أبو الحسن الموازيني قراءة أنا أبو الحسين بن أبي نصر سنة ثلاث وأربعين وأربعمائة أنا محمد بن يوسف بن يعقوب الرقي نا أبو محمد عبدالله بن شوذب الواسطي نا شعيب

Abū al-Ḥasan al-Mawāzīnī [Abu al-Hasan 'Ali b. al-Hasan] informed us via reading back to him that Abū al-Ḥusayn b. Abī Naṣr transmitted to us in the year 443 that Muḥammad b. Yaʿqūb al-Raqqī transmitted from Abū Muḥammad ʿAbd Allāḥ b. Shawdhab al-Wāsiṭī, who transmitted from Shuʿayb …

Ibn al-Ṣalāḥ describes recitation to the teacher (al-qirāʾa ʿalā al-shaykh) as the second method of Hadith transmission. A reciter reads or recites the transmitted Hadith aloud to the teacher, who confirms its accuracy. There is some latitude in how the recitation might occur:

It makes no difference whether you are the reciter or someone else recites while you listen. You may recite from a book or from your memory. The teacher may have memorised the material recited to him or not. In the latter case he or some other reliable individual should hold his text.¹⁷

If Ibn ʿAsākir’s understanding of the term matches Ibn al-Ṣalāḥ’s, when he writes qirāʾatan Ibn ʿAsākir may simply have been present in a lesson when another person read something back to Abu al-Hasan 'Ali b. al-Hasan.¹⁸

Since the term qirāʾatan suggests the transmission of written material, we sought to ascertain whether its use in the TMD was correlated with citations of people identified by modern scholars as authors. However, a search for author names in isnāds containing the term qirāʾatan revealed no clear-cut correlation. Many such isnāds do indeed include author names, but others do not.

Ijāzatan

The term ijāzatan is used frequently in Ibn ʿAsākir’s isnāds. The following is a typical example:

أخبرنا أبو غالب أحمد بن الحسن أنا أبو الحسين بن الابنوسي أنا عبد الله بن عتاب أنا أحمد بن عمير إجازة

Abū Ghālib [Abu Ghalib] informed us that Abū al-Ḥusayn b. al-Abnūsī [Abu al-Husayn b. al-Abnusi] informed them that ʿAbd Allāh b. ʿItāb ['Abd Allah b. 'Itab] informed them that Aḥmad b. ʿUmayr [Ahmad b. 'Umayr] informed them via an ijāza that …

In another instance, Ibn ʿAsākir provides an isnād that passes through Ibn Manda (d. 470/1077):

أخبرنا أبو الحسين وأبو عبد الله إذناً قالا أنا ابن مندة أنا أبو علي إجازة

Abū al-Ḥusayn [Abu al-Husayn] and Abū ʿAbd Allāh [Abu 'Abd Allah] informed us with permission, saying that Ibn Manda [Ibn Manda] told them that Abū ʿAlī [Abu 'Ali] told them via an ijāza that …¹⁹

In both of these cases, Ibn ʿAsākir mentions ijāzas. But in neither case does he say that he himself received one for the information in question. This is typical: on only 200 occasions does he mention gaining information himself via an ijāza. The rest of the time, the term occurs deeper in the isnād.²⁰

Ibn al-Ṣalāḥ is helpful here, as he has much to say about ijāzas.²¹ He discusses several types of ijāzas, including a licence granted to a specific person for the transmission a specific text. In such a case, ‘the teacher says, “I hereby license book X” – or “the contents of this book list of mine (fihristī)” – “to you”.’²² Other scenarios involve licences for unspecified texts for specific people (e.g. a teacher’s granting a student permission to transmit everything he has heard) or a licence for ‘unspecified person under a general designation’, as when a teacher grants an ijāza to all Muslims in general or to all of his contemporaries. Ibn al-Ṣalāḥ notes that the last type is controversial. He says that although it has been used,

we have not seen nor heard of anyone who is worthy of emulation actually using this form of licensing and relating by it, not even the small group of late-comers who viewed it as valid. Licensing in principle is a weakness. By this extension and relaxation, it becomes a much greater weakness which should not be tolerated. God knows best.²³

Ibn al-Ṣalāḥ thus extends his disapproval of this type of ijāza to a general disparagement of transmission via an ijāza.

Ibn ʿAsākir’s use of the term ijāza indicates that he accepted it as a valid form of authoritative transmission. However, in describing his own knowledge acquisition he chooses to stress his personal interactions with informants and his non-reliance on ijāzas. As seen earlier, Abu Ghalib, Abu al-Husayn and Abu 'Abd Allah all transmitted material to him that relied on the authority of an ijāza. It seems that Ibn ʿAsākir valued antique ijāzas such as theirs. He sought out these vestiges of writing and verified them through a process that was based on oral/aural transmission. These isnāds with ijāzas deserve closer scrutiny, but they very likely have special features. As a musnid, Abu Ghalib, for example, may have developed a reputation for particularly interesting or short isnāds that involved ijāzas.

Ibn ʿAsākir and his predecessors may also have been just plain pragmatic. As noted in an earlier post, Davidson has written about the function of the isnād in the post-canonical period as providing a mechanism for the maintenance of the community of scholars. The term ijāza, in particular, came to refer to a non-oral form of Hadith transmission:²⁴

Considered an analog to oral/aural transmission, in essence the ijāza is a permission granted by a transmitter allowing the recipient to cite and further transmit a text or groups of texts through the granting transmitter’s personal chain of transmission. The ijāza thus allowed scholars to establish a personal chain of transmission for a text when time, distance or other factors made oral/aural transmission of the text unfeasible. In a scholarly culture that expected that a chain of transmission would be cited, the ijāza was a means of meeting this expectation without investing the time in actual, oral/aural transmission.²⁵

The need for ijāzas arose because the demands of oral/aural transmission became too unwieldy; the ‘culture of hadith transmission idealized oral transmission’, but the reality was that the acceleration of transmission and the growing number of works made realising this ideal impossible. The ijāza made it possible for that ideal to be maintained, not only for reports but also for whole books:²⁶

The ability to cite material without having to invest the time in auditing it was an extremely useful tool in a scholarly culture that was increasingly becoming dependent on widely-circulated and often lengthy authored texts but still expected that these sources would be cited with a chain of transmission in the oral idiom.²⁷

Al-Khatib played a major role in the growing acceptance of the ijāza, and he spoke forcefully against the view (in the minority by his day) that an ijāza cited within a chain of transmission broke the chain.²⁸

However, although ijāzas may have been widely accepted in Ibn ʿAsākir’s time, the fact that he refrains from citing ijāzas for his own receipt of materials suggests that he did not want to be seen as availing himself of their conveniences.

Some of the chains in the TMD that contain the term ijāza also feature other terms identifying methods of transmission, as in the example above, which combines ijāzatan with idhnan. In another case, Ibn ʿAsākir credits Abu Bakr Wajih with a transmission ijāzatan and then clarifies that the transmission did not take place via audition (in lam yakun samāʿan).²⁹

Fī kitābihi

The phrase fī kitābihi (‘in his writing’) in Ibn ʿAsākir’s isnāds points to something written by the cited person. But is it used specifically in reference to the writings of authors? The answer is no, as the following example shows:

أخبرنا أبو عبدالله محمد بن الفضل في كتابه عن أبي بكر البيهقي أنا أبو عبدالله الحافظ قال سمعت أبا علي الحسين بن علي الحافظ يقول سمعت إبراهيم بن يوسف الهسنجاني يقول

Abū ʿAbd Allah Muḥammad b. al-Faḍl [Abu 'Abd Allah al-Furawi] told us in his writing (fī kitābihi) from Abū Bakr al-Bayhaqī [al-Bayhaqi], who said that Abū ʿAbd Allāh al-Ḥāfiẓ [Abu 'Abd Allah al-Hakim] said: I heard Abū ʿAlī al-Ḥusan b. ʿAlī al-Ḥāfiẓ saying that I heard Ibrāhīm b. Yūsuf al-Hasanjānī saying …

The term kitāb denotes a piece of writing of some sort or another (perhaps a letter containing an ijāza), and in this isnād it is attributed to Abu 'Abd Allah al-Furawi,³⁰ a direct informant who does not appear in our author list but who plays a large role in the TMD in transmitting the works of others. Al-Bayhaqi is an author, but it is not his kitāb to which Ibn ʿAsākir is referring directly.

Our isnād data set contains 976 instances in which the expression fī kitābihi is used within an isnād. In the vast majority of instances, it refers to the writing of a direct informant in conjunction with a term such as akhbaranā. In other words, Ibn ʿAsākir uses it to say that he (or the group to which he belonged) received information via a direct informant’s writing. The direct informant cited most often in this context is Abu 'Ali al-Haddad (266 times).

To determine whether Ibn ʿAsākir’s use of this phrase was connected to his citation of known authors, we searched the subset of our isnāds that feature the names of individuals whom Scheiner, Mourad and al-Daʿjānī have identified as authors of works (see our discussion of author names in Post 2). We found 376 to contain an author’s name as well as the phrase fī kitābihi, but only very rarely was it used with respect to an author himself; most of the time, as noted above, it refers to the writing of a direct informant.

The phrase fī kitābihi could point to the granting of an ijāza by a direct informant for books he had gained the authority to transmit onwards, as argued by Jeremy Farrell recently for an earlier context.³¹ But were these writings necessarily ‘books’ as we think of them today?³² We believe this is possible (particularly for Abu 'Ali al-Haddad), but also that we must consider a wider sense of written works, extending to excerpts, notes, lists, pamphlets and the like.

A last word on ambiguity from Ibn al-Ṣalāḥ

As noted above, in 5% of his isnāds Ibn ʿAsākir uses vocabulary – supplementary transmission terms – that gives a fairly precise sense of how he obtained his information. He was constrained by the limitations of a notational system that is fundamentally linear; it is easy to imagine, as Aslisho Qurboniev has noted, that traditionists would have preferred to write in more dimensions, using arrows and the like to elucidate the different roles played by two persons cited together, for example.³³

Ibn ʿAsākir’s methods of citation probably would not always have passed muster with Ibn al-Ṣalāḥ. When writing about transmission by handing something over (munāwalatan), Ibn al-Ṣalāḥ mentions with disapproval scholars who used the terms ḥaddathanā and akhbaranā for munāwalatan without making it clear that this is the method they mean. He notes similar ambiguity in some scholars’ use of ijāzas, singling out Abū Nuʿaym al-Iṣfahānī [Abu Nu'aym] (d. 430/1038) and Abū ʿUbayd Allāh al-Marzubānī [al-Marzubani] (d. 370/980) as guilty of this practice. He goes on to state:

The sound and preferred view which the majority follow in practice and which the earnest and scrupulous have adopted is to forbid the unqualified application of ‘He transmitted to us’ (ḥaddathanā), ‘He informed us’ (akhbaranā) and similar expressions to material received by handing over³⁴ or licensing and to designate that material with qualified versions of these expressions which indicate the true situation. The transmitter should say, ‘X informed us’ – or ‘transmitted to us’ – ‘by handing over and by licence’ (munāwalatan wa-ijāzatan), ‘He informed us by licence’, ‘He informed us by transference’, ‘He informed us by granting permission’ (idhnan), ‘Under his grant of permission is’ (fī idhnihi), ‘One of the things he granted me permission for is’ (fīmā adhina lī fīhi) or ‘One of the things which he gave me leave to transmit from him is’ (fīmā aṭlaqa lī riwāyatahū ʿanhu). Or he should say, ‘so-and-so licensed to me’ (ajāza lī fulān), ‘so-and-so licensed to me such and such’ (ajāzanī fulān kadhā wa-kadhā), ‘so-and-so handed over to me’ (nāwalanī fulān) and similar expressions.³⁵

Ibn al-Ṣalāḥ does not mention Ibn ʿAsākir, although he and his works are cited widely across the OpenITI corpus. When Ibn al-Ṣalāḥ quotes al-Khatib, the quotations are of al-Khatib’s Hadith works (especially his al-Kifāya fī ʿilm al-riwāya), not his Tārīkh Baghdād. Ibn al-Ṣalāḥ’s primary concern is with ideal citation, not with its real-world application, and his project thus differs from Ibn ʿAsākir’s in the TMD. Still, it is unlikely that Ibn al-Ṣalāḥ would have wholeheartedly endorsed Ibn ʿAsākir’s way of citing.

1. Ibn ʿAsākir employs the ḥ with akhbaranā or, less frequently, ḥaddathanā or qaraʾtu (or their singular forms) when, after citing an isnād, he introduces another isnād that does not share the names of the preceding isnād. It is included as a distinct category because of the way our script operates on the text of the TMD (we might subsequently revise to include the results within the other terms). ↩

2. A short Hadith manual by a contemporary of Ibn ʿAsākir’s provides a useful comparison. For a description and translation of Abū Ḥafṣ ʿUmar b. ʿAbd al-Majīd al-Qurashī al-Mayyānishī al-Mahdawī’s (d. 581/1185) Mā lā yasaʿu al-muḥaddith jahluhu, see Leonard Librande, ‘The Supposed Homogeneity of Technical Terms in Ḥadīth’, Muslim World 72 (1982), 34–50. There are many other such manuals. ↩

3. Ibn al-Ṣalāḥ, Introduction, 97–109; Ibn al-Ṣalāḥ, Muqaddima, 132–50, ms. 055–65. See also the translations of these Arabic terms in Robson, ‘Ḥadīth’. ↩

4. Ibn al-Ṣalāḥ, Introduction, 102; Ibn al-Ṣalāḥ, Muqaddima, 140, ms. 059. For a view on ḥaddathanā across centuries and contexts (with a slightly different reading of Ibn al-Ṣalāḥ), Devin J. Stewart, ‘On Rhetoric, Reason, and Revelation: Ibn al-Jawzī’s Maqāmāt as an Anti-Parody and Sefer Taḥkemoni of Yehudah al-Ḥarīzī’, Middle Eastern Literatures 19/2 (2016), 207–11. ↩

5. Ibn al-Ṣalāḥ, Introduction, 102; Ibn al-Ṣalāḥ, Muqaddima, 140, ms. 059. ↩

6. Ibn al-Ṣalāḥ, Introduction, 104–5; Ibn al-Ṣalāḥ, Muqaddima, 144, ms. 061. ↩

7. Al-Mayyānishī writes: ‘Most of the learned were of the opinion that there is no difference between the ḥadīth reporter saying ḥaddathanā and his saying akhbaranā. Others concluded that his expression ḥaddathanā meant he had actually heard it from the mouth of its reporter while his word akhbaranā signified he had heard it when the reporter or some other master had read it. It was transmitted to us on the authority of the messenger of God that he had said: “Ḥaddathanā and akhbaranā are the same, since for the Arab there is no difference between his saying ḥaddathanī fulān and akhbaranī fulān.”’ Librande, ‘Supposed Homogeneity’, 42–3. ↩

8. Ibn al-Ṣalāḥ, Introduction, 104; Ibn al-Ṣalāḥ, Muqaddima, 143, ms. 061. ↩

9. As noted in Post 2, we exclude such combined isnāds (of which we found 2,236) from our data set because they challenge our data model. ↩

10. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 72876, ms. 24743. ↩

11. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 74482, ms. 25267. ↩

12. Ibn al-Ṣalāḥ refers to this way of transmitting Hadith as mukātaba (‘passing on’ or ‘transference’), and he addresses it as the fifth method of transmitting Hadith. Ibn al-Ṣalāḥ, Introduction, 122–3; Ibn al-Ṣalāḥ, Muqaddima, 173–4, ms. 076–7. ↩

13. Personal communication, 15 July 2023. ↩

14. Adding in cases of ḥ with akhbaranā and its variants. ↩

15. Ibn ʿAsākir’s top informants for short chains include Abu al-Qasim al-Samarqandi (his first-ranked direct informant who transmits ninety times ʿāliyan), Abu al-Qasim b. al-Husayn (eleventh-ranked direct informant, sixty-six ʿāliyan transmissions) and Abu Ghalib (second-ranked direct informant, sixty-two ʿāliyan transmissions). See Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads_with_Search_Terms’, table ‘TC_Aliyan_NameCount’. ↩

16. Ibn al-Ṣalāḥ, Muqaddima, 171, ms. 75. ↩

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

18. Versus qirāʾati ʿalayhi, which would appear to specify Ibn ʿAsākir’s own reading back. ↩

19. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads’, table ‘TransmissionChains_Normalised’, ID 77305, ms. 23868. ↩

20. A search of the expression ajāza lī (‘he licensed me’) produces 40 results. Though a small number they also are relevant. He says 14 times within biographical entries ajāza lī jamīʿa ḥadīthihi (‘he licensed to me all of his Ḥadīth’). ↩

21. This is the third method of Hadith transmission that Ibn al-Ṣalāḥ discusses. Ibn al-Ṣalāḥ, Introduction, 109–18; Ibn al-Ṣalāḥ, Muqaddima, 150–64, ms. 065–72. ↩

22. Ibn al-Ṣalāḥ, Introduction, 109; Ibn al-Ṣalāḥ, Muqaddima, 150, ms. 065. ↩

23. Ibn al-Ṣalāḥ, Introduction, 111–12; Ibn al-Ṣalāḥ, Muqaddima, 154, ms. 067. ↩

24. See especially Davidson, Carrying On the Tradition, 108–51 (chap. 3, ‘Non-oral Transmission in the Oral Idiom: The Development and Function of the Ijāza’). ↩

25. Davidson, Carrying On the Tradition, 108–9. ↩

26. Davidson, Carrying On the Tradition, 115ff. ↩

27. Davidson, Carrying On the Tradition, 114. ↩

28. Davidson, Carrying On the Tradition, 124–5. ↩

29. Savant and Seydi, ‘Ibn ʿAsākir and His History of Damascus’, ‘Isnads_with_Search_Terms’, table ‘TC_Ijazatan’. ↩

30. Scheiner lists Abu 'Abd Allah al-Furawi (d. 530/1136) as a transmitter in two riwāyas, for works by al-Ḥākim al-Nayṣābūrī (d. 403/1012, Kitāb MuIntikī ruwāt al-akhbār) and Yazīd b. Muḥammad b. Iyās al-Azdī al-Mawṣilī (d. 334/946, Kitāb Ṭabaqāt muḥaddithī ahl al-Mawṣil). According to Mourad, Ibn ʿAsākir sought him out in Nishapur, and he taught Ibn ʿAsākir the Dalāʾil al-nubuwwa of al-Bayhaqī and all the books of the Hadith canon: the Ṣaḥīh of al-Bukhārī (d. 256/870), the Ṣaḥīḥ of Muslim (d. 261/875), the Sunan of Abū Dāwūd (d. 275/889), the Jamīʿ al-kabīr of al-Tirmidhī (d. 279/892) and the Sunan al-kubrā of al-Nasāʾī (d. 303/915). Mourad recounts a story according to which Abu 'Abd Allah al-Furawi was irritated by Ibn ʿAsākir’s pestering until someone reported having had a dream in which the Prophet Muhammad declared Abu 'Abd Allah al-Furawi should be patient with the ‘young dark-skinned Syrian’ (that is, Ibn ʿAsākir) and teach him Hadith. Mourad, Ibn ‘Asakir of Damascus, 23–4. ↩

31. Jeremy Farrell, ‘Jaʿfar al-Khuldī (d. 348/959) and the Early Sufi Textual Tradition: Text-critical and computational approaches’ (PhD Dissertation: Emory University, 2021), 83–99, 130–40. ↩

32. The argument has been made that the phrase alludes to an ijāza for books acquired by written correspondence. Ibn ʿAsākir’s slight citation of ijāzas makes us sceptical that this is what he is saying. ↩

33. Personal communication, 10 February 2023. ↩

34. Dickinson translates munāwalatan as ‘transference’, but we have used the more specific ‘handing over’, following Devin Stewart (personal communication 15 July 2023). Another possibility would be ‘passing along’, following Librande’s translation of al-Mayyānishī’s Mā lā yasaʿu in ‘Supposed Homogeneity’, 43 and note 42. ↩

35. Ibn al-Ṣalāḥ, Introduction, 120–1; Ibn al-Ṣalāḥ, Muqaddima, 170, ms. 074–5. ↩