Why I No Longer Use the Term “Qāḍī-Court Documents”

By Marina Rustow

I came into my graduate seminar on Arabic legal documents with some experience in paleography and diplomatics, but vanishingly little knowledge of the material I was going to be teaching. I knew I wouldn’t always, or even often, have answers about how to read the sources, let alone how the judicial system worked.

But I did have three capable graduate students: Amel Bensalim, a historian of Ḥafsid Ifrīqiya and diplomacy between the Islamic world and Latin Europe; Athina Pfeiffer, a specialist in Islamic law who is planning a dissertation on early Ismāʿīlī law; and Lucia Waldschuetz, who is writing a dissertation on surety contracts in Greek and Coptic from Roman Egypt and has sufficient Arabic to decipher documentary texts.

The seminar felt like a research group. When I hit the paleographic wall, my perplexity showed that documents present difficulties even to professionals, rather than conveying the misleading impression that every question can be answered with hard work (not to mention the ensuing sense of personal shame when hard work yields no good answers). I was honest with them about my limits; they broke us out of impasses with persuasive readings at least as often as I did.

I also entered the seminar suspecting that my relative ignorance of Islamic law could work to our advantage. Being ignorant, I took little for granted about judicial procedure. But I soon discovered that I had taken more for granted than I realized.

What was the role of qāḍīs in legal processes?

For the first half of the semester, I was still speaking of the texts we were deciphering as “qāḍī court documents.” Then, one day, my students asked me why I was using that term, given that to date, we had not seen a single document written or signed by a qāḍī.

What, in fact, was the relationship between qāḍīs and the production of judicial records? It was clear from the documents themselves that notaries were the ones doing the heavy lifting, translating legal acts into watertight documents by remixing formulae. I had assumed that the notaries must be working for qāḍīs who ran the show. Yet when I looked back at the documentary evidence, there was little to support this assumption.

I don’t know where the assumption came from. I have found it in some of the literature, most recently in rereading Émile Tyan’s work on notaries, where he claims that judicial documents were drawn up either in court or else by “the judge’s notary, under his authority.”[1] Wael Hallaq’s influential article on the qāḍī’s dīwān distinguishes between two kinds of judicial documents: maḥāḍir (sing. maḥḍar), which are judicial records containing statements about claims, transactions and settlements; and ḥujaj (sing. ḥujja) or wathāʾiq (sing. wathīqa), which are stand-alone records that remained in the possession of litigants. Hallaq claims that maḥāḍir record statements made “in the presence of the qāḍī, who must sign it before witnesses in order for it to be complete.”[2] In the Fatimid- and Ayyubid-era maḥāḍir that my students and I studied, no claims were made in the qāḍī’s presence.

The qāḍī did, however, enter the document’s life-cycle at a later point. Geoffrey Khan describes the situation as follows. Let’s say the witnesses who attested a document have died. How can the document that they signed continue to be accepted as proof of what the document records? The answer: through recourse to a copy that is beyond suspicion of alteration.

There were two ways to access such a copy. The first was to find one in the possession of a witness to the transaction, in which case it would still need to be validated. Since witnesses were often themselves notaries, this implies the existence of a notary’s dīwān. The second was to find one that had been registered and deposited “in the archives of a court or government office” — the qāḍī’s or the amīr’s dīwān. In the first case, the key to “grant[ing] the written documents probative value” was archiving, retrieval and validation; in the second, it was presentation and registration.[3]

Christian Müller notes, however, that for the Fatimid and Ayyubid periods, “it is difficult to correlate [specimens of] these [documents] with the descriptions of the cadi’s archive.”[4] I think we can afford to be a bit more sanguine.

The first place I searched for signs of judicial archiving and registration was in what Khan calls “court records”: documents that call themselves maḥḍar, as well as others that I assume are maḥāḍir based on their opening verb, ḥaḍara, to appear (before witnesses and offer a statement).[5] Interestingly, however, none of them contains evidence of having been signed by a qāḍī or of having been deposited in a judicial archive. To the contrary: some of the maḥāḍir that have survived must have been given to the parties they name, because those same parties then reused the documents in a rather casual manner.

One of the documents in Khan’s corpus is a statement dated 495/1102 and drawn up in Ibwān near al-Ushmūnayn in Upper Egypt regarding the settlement of a dispute over livestock between two families (Fig. 1). It must have gone home with one of the litigants, a certain Durrī b. ʿAskar al-Qarīṭī, because he later pressed it into service as writing material, having a letter written on the verso on his behalf instructing the recipient to give him ten irdabbs of grain.[6]

Al-Qarīṭī’s testimony never made it to a judicial archive for registration. That stands to reason: documents signed by witnesses were produced in multiple copies, some of which litigants took with them and others of which may well have been registered and/or deposited in judicial archives—but not this one. Two points in the existing scholarship therefore no longer hold: that maḥāḍir are “court records” (Khan); and that they contain statements made “in the presence of the qāḍī, who must sign it before witnesses in order for it to be complete” (Hallaq).

There are, however, documents other than maḥāḍir that show evidence of registration and storage in judicial archives. Some documents refer to judicial or state-run archives; others hint at the procedures involved in maintaining and consulting those archives; still others contain archival notations—physical traces of the qāḍī’s dīwān of Hallaq (and now Müller) fame.

A petition addressed to the Fatimid caliph al-Āmir (495–524/1102–30) that Khan published in 1990 contains unambiguous evidence that some litigants had copies of judicial documents (wathāʾiq) deposited in archives.[7]

The petition is from an anonymous woman who is attempting to collect a debt of 241 dinars (an impressive sum) from the estate of a deceased woman close to the caliphal family, a certain Ṭāhira, known as Sitt al-Ḥasab, daughter of an amīr titled Zaʿīm al-Khilāfa wa-Fakhruhā (“guarantor and pride of the caliph’s rule”). The petition notes that the lender “has registered the documents with the amount of the debt in the noble majlis al-ḥukm” (fa-qad athbatat al-mamlūka al-ḥujaj bi-mablagh al-dayn fī majlis al-ḥukm al-sharīf). The majlis ḥukm in question most likely belonged to the chief qāḍī, a Fatimid state appointee.[8] This is unambiguous evidence that judicial documents were archived in the “assembly of adjudication,”[9] presumably in this case the chief qāḍī’s archive.

A second petition to al-Āmir dating to 516–21/1123–27 refers even more explicitly to the majlis al-qāḍī. The petition, from a Jewish long-distance trader named Mūsā b. Ṣadaqa, seeks redress for the unjust imposition of a customs duty. Mūsā b. Ṣadaqa claims that before he left on a voyage to India and Yemen, he had documents registered (athbata) in the majlis al-qāḍī in order to prevent … (presumably either his paying customs or any dispute about the matter; there is a lacuna in the manuscript).[10] This is explicit evidence not just of judicial archiving but of its utility for making persuasive legal claims—in this case, not even before a qāḍī, but in a caliphal tribunal.

Qāḍīs were, then, judicial archivists, and archiving may have been even more central to their labor than Hallaq argued. Müller, meanwhile, has emphasized the role of the qāḍī in maintaining judicial archives and authenticating the documents they contained.[11] The documents I’ve discussed here suggest that the qāḍī’s dīwān contained authenticated judicial documents whose purpose was to furnish written evidence in case of litigation, and also authenticated administrative or commercial documents one could muster before a caliphal tribunal in case of a dispute with lower officials.

Registering documents ʿinda l-qāḍī

The verb for registration in these documents is athbata. In medieval Arabic documentary practice, verbs derived from this root can mean to register a document’s contents in an official bureau, or to authenticate a document by adding an annotation to it, usually an ʿalāma (a signature consisting of a pious phrase).[12]

Registering a document and authenticating it were linked, and this was true not just in judicial contexts but also in mundane fiscal matters, such as paying taxes.[13] After making note of a document’s contents for the archive (the act of registration), an official noted that he had done so by annotating the document in the upper margin with his ʿalāma (the act of authentication).

An annotation of this kind can be found in the upper margin of a marriage contract datable to the fifth/eleventh century (Fig. 2). The annotation reads Allāhu l-ʿumda, “God is the support,” presumably the ʿalāma of someone such as a qāḍī or another official; and then comes the record of registration: thubita fī dīwān al-ḥukm al-muwaffaq, “it has been registered (or established) in the prosperous judicial archives.”[14]

This is a different use of the ʿalāma from what one finds in Mamlūk-era judicial decisions (aḥkām, sing. ḥukm) issued by judges, or even from the impressive ʿalāmas of caliphs and viziers that one finds at the top of Fatimid state decrees. All these uses of the ʿalāma serve an authenticating function, but the Fatimid judicial annotations—and similar annotations on tax receipts and other state documents—serve an additional administrative purpose: noting that the document has been recorded in an archive.[15]

In the case of this marriage contract, alas it’s unclear who was doing the annotating and archiving. To avoid prejudging the question, I’ve attributed the ʿalāma to “the qāḍī or another official.” Nor, for that matter, is it entirely clear that the “archive of adjudication” (dīwān al-ḥukm) was the archive of a qāḍī, although it would stand to reason.

But in a mid-to-late fifth/eleventh-century draft maḥḍar from Ifrīqiya, two qāḍīs appear clearly as keepers of archives.[16]

The maḥḍar was drawn up following a dispute over an inheritance that arose precisely because the litigants had registered documents in the archives—but the archives of two different qāḍīs.

One qāḍī was involved in executing the will of the deceased. After the qāḍī’s death, his archive passed to his successor and was available for consultation.[17] The second qāḍī, who had also since died, was connected with a pious foundation, and his dīwān could also be consulted.[18] (In neither case do we know what happened to the qāḍī’s archive after his successor’s death, and whether it was in fact discarded, as Hallaq argued such archives were.)

Both qāḍīs’ dīwāns apparently performed the work of authentication of documents: the maḥḍar refers to what’s in these archives as “judicial proof” (bayyina),[19] and the act of consulting the documents in the archives as “muster[ing] proof for” the litigant (ushhidat lahu al-bayyina).[20] Both phrases support the conclusion that a written judicial record acquired the status of proof when a qāḍī recorded the information it contained—or when another official recorded it for him, or in his presence, or in his archive.[21] Frustratingly, the document finesses the question, describing the other document ambiguously as being ʿinda al-qāḍī, “with the judge.” Khan and I both understand this to mean “(registered) with the judge”; if this is correct, it still remains unclear which personnel registered notarial documents in the dawāwīn.

These documents nonetheless provide answers to two important procedural questions: the role of judicial archiving in the production of proof, and the place of archiving among the qāḍī’s duties.

Notarial documents vs. qāḍī-court documents

Now I’ll draw some conclusions before turning the blog over to my students. The glimpses of judicial procedure that emerge from these documents suggest that the term “qāḍī-court documents” is misleading. They should instead be called notarial documents or judicial records.[22] Many documents never came before a qāḍī, and even those that did may have encountered him not as someone who rendered judicial decisions, but as an archivist. The role of the qāḍī—at least in Egypt in the fifth/eleventh–sixth/twelfth centuries—was not to produce documents, nor to oversee their production or even their attestation by witnesses. His role, rather, or one of them, was to authenticate and archive documents so that he and others could use them as evidence.

Witnesses’ attestations and signatures on legal documents were not, then, sufficient to turn judicial records into proof. The act of registration was a necessary step, and that act required officials above the level of notaries and witnesses; it required authentication, registration and record-keeping, ʿinda l-qāḍī.

The famous qāḍī’s dīwān is therefore essential to answering questions about the function and status of documents in Islamic law, but for a reason different from Hallaq’s. In the material I’ve considered here, the qāḍī was not a necessary part of quotidian judicial proceedings. One of the functions of the qāḍī was to make documents deployable as evidence by registering and archiving them; but not everyone who had notaries draw up documents also had qāḍīs authenticate, register and archive them.

For Hallaq, the qāḍī’s dīwān was as evanescent as the “courts” were: the majlis convened and adjourned; the dīwān of the previous qāḍī vanished with his successor. But some of the functions that Hallaq attributed to qāḍīs, such as convoking witnesses to sign documents, in the material I’ve considered here were the job of notaries.[23] Convening the majlis al-ḥukm was, conversely, only part of a qāḍī’s job. The other was administrative: overseeing the validation and storage of documents.


[1] “[Il n’y a] aucune particularité intrinsèque qui distinguerait l’acte dressé en justice de l’acte dressé par un notaire ordinaire. La seule différence consiste en ce que l’acte est rédigé par le greffier du juge sous l’autorité de ce dernier. Pour le surplus, l’acte se présente exactement sous les mêmes apparences que l’acte rédigé par un notaire.” Émile Tyan, Le notariat et le régime de la preuve par écrit dans la pratique du droit musulman (Beirut, 1945), 95.

[2] Hallaq says the judicial archive “consists of a witnessed record of what the maḥḍar contained, together with the qāḍī’s decision (ḥukm) on the case.” This description is based on long-form sources (for which see Wael B. Hallaq, “The qāḍī’s dīwān (sijill) before the Ottomans,” Bulletin of the School of Oriental and African Studies 61 (1998): 420, n.24). He notes that the sixteenth-century Egyptian Ḥanafī jurist Ibn Nujaym adds some interesting nuance: “in the convention prevailing nowadays (al-ʿurf al-ān), the sijill is that which the two witnesses write down concerning the dispute (between them), and which remains with the qāḍī, but does not have on it his own handwriting.” Cf. Christian Müller, “Les actes notariés en droit musulman (VIIIe–XVIe siècles),” Studia Islamica (2022): 15: “Dans la pratique des actes, le mot ‘attester’ (šahida) désigne l’accomplissement d’une action devant le juge.”

[3] Khan, ALAD, 8. Christian Müller has since underscored the second method on the basis of a corpus of 2,400 documents in the CALD database (only 387 transcriptions are available in the public-facing database, all of them previously published): “Keeping an archive was a task that distinguished a cadi from other officials with judicial functions.” He also notes that the qāḍī’s archive contained additional items. Christian Müller, “The Power of the Pen: Cadis and their Archives,” in Manuscripts and Archives, eds. Alessandro Bausi and others (De Gruyter, 2018), 371–72.

[4] Müller, “Power of the Pen,” 372.

[5] See also the statement certifying that a man was a Khaybarī Jew, 654/1256 (Cambridge University Library, Taylor-Schechter [henceforth T-S] NS 327.2c, ed. Khan, ALAD, n°49); and a statement in three drafts, seventh/thirteenth century (T-S NS 305.115, ed. Khan, ALAD, n°51).

[6] T-S Ar.38.56 (ed. Khan, ALAD, n°58; also online through the Princeton Geniza Project database, PGPID 35281). See also T-S Ar.38.71 (ed. Khan, ALAD, n°59, PGPID 8213), a mid-to-late fifth/eleventh-century judicial record concerning a dispute over an inheritance, reused by the long-distance trader from Qayrawān Nahray b. Natan; T-S Ar.34.94 (ed. Khan, ALAD, n°60, PGPID 16244), a judicial record concerning a surgical operation, reused for the draft of a private petition; T-S 8Ja1.1 (ed. Khan, ALAD, n° 57, PGPID 31525), a partnership agreement from 450/1058 reused for Judaeo-Arabic poetry.

[7] Cambridge University Library, Michaelides (charta) A169, line 8, ed. Khan, “Petition to the Fatimid Caliph al-ʾĀmir,” Journal of the Royal Asiatic Society 1 (1990): 44–54 (PGPID 36680).

[8] The qāḍī in question is titled Nūr al-Mulk ʿUmdat al-Aḥkām, which Khan notes is a pattern typical of the titles of Fatimid chief qāḍīs; Khan, “Petition to the Fatimid Caliph al-ʾĀmir,” 50.

[9] A translation of majlis al-ḥukm that I am borrowing from Hallaq; see my previous post in this blog.

[10] T-S Ar.30.273 (ed. Khan, ALAD, n°77, PGPID 9235), lines 10-11.

[11] Müller, “Power of the Pen.”

[12] Müller translates thubūt as “establishing facts”; “Power of the Pen,” 374.

[13] This is evident from the corpus of Fatimid tax-receipts that I am currently editing together with Lorenzo Bondioli, Stephanie Luescher and Naïm Vanthieghem. See also Khan, ALAD, n°s 140–56.

[14] T-S Ar.38.61 (ed. Khan, ALAD, n°34; PGPID 16250). Many other judicial documents have survived with registration annotations in the upper margin from both the Ḥaram al-Sharīf cache and the Geniza. An ʿalāma at the top of a deed of sale from the late fifth/eleventh century reads thubita wa-li-llāhi al-ḥamdu wa-l-shukr (T-S Ar. 53.60r, ed. Khan, ALAD, n°9; PGPID 8314). A registration mark at the top of another deed of sale, this one dated 441/1050 and probably from al-Ushmūnayn, reads thubita ʿindī jamīʿ mā nuṣṣa wa-shuriḥa fīhi, followed by the name of the person who registered it and the date. For this reference and others, see Khan, ALAD, 100 (note to line 1). I am excluding the deed for the purchase of a house from a government bureau (dīwān al-farajī or al-faraḥī, T-S Ar.53.61, ed. Khan, ALAD, n°13; PGPID 8315), since the request for registration at the top refers to the archives of said government office. In Fatimid contexts, dīwān means both a government bureau and the archive associated with it; see Marina Rustow, The Lost Archive: Traces of a Caliphate in a Cairo Synagogue (Princeton University Press, 2020), 347–48.

[15] Müller seems to suggest that there is little evidence for the use of the ʿalāma in judicial procedure before the seventh/thirteenth century. It seems that he has in mind the impressive-looking ʿalāmas that judges began to issue atop aḥkām. See Müller, “Actes notariés,” 17, n.59, and cf. ibid., 18–19.

[16] T-S Ar.38.71 (see above, n. 6). As Khan points out, it must be a draft, as some of the details have been anonymized to fulān and kadhā.

[17] T-S Ar.38.71 (ed. Khan, ALAD, n°59), line 11.

[18] Fī l-waqf, T-S Ar.38.71 (ed. Khan, ALAD, n°59), lines 12–13. It isn’t clear whether the archive was physically in the waqf or the judge was related to the waqf.

[19] T-S Ar.38.71, lines 14 and 17.

[20] T-S Ar.38.71, lines 17–19.

[21] This seems to diverge from Müller, “Power of the Pen,” esp. 369–72, and “Actes notariés,” esp. 18, who argues that before the thirteenth century, although qāḍīs kept documents in archives, no documentary evidence has yet emerged that those documents were valid for the purposes of litigation without first calling witnesses; and that earlier documentary sources suggest that even documents in qāḍīs’ archives had to be validated by witnesses in order for the qāḍīs themselves to draw on them as proof. He does not discuss documentary evidence from the eleventh and twelfth centuries in these articles; I look forward to reading his new book, Christian Müller, Recht und die historische Entwicklung der Scharia im Islam (Berlin, 2022).  

[22] Agreeing with Müller, “Actes notariés,” and with many of the contributors to Intisar A. Rabb and Abigail Krasner Balbale, eds., Justice and Leadership in Early Islamic Courts (Cambridge, Mass.: Harvard University Press, 2017).

[23] For a glimpse of the debates that may have preceded this development, see Mathieu Tillier, Les cadis d’Iraq et l’état abbasside (132/750–334/945) (Damascus, 2009), 398–99.

(Suggested Bluebook citation: Marina Rustow, Why I No Longer Use the Term “Qāḍī-Court Documents”, Islamic Law Blog (Jan. 12, 2023), https://islamiclaw.blog/2023/01/12/why-i-no-longer-use-the-term-qa%e1%b8%8di-court-documents/)

(Suggested Chicago citation: Marina Rustow, “Why I No Longer Use the Term ‘Qāḍī-Court Documents,'” Islamic Law Blog, January 12, 2023, https://islamiclaw.blog/2023/01/12/why-i-no-longer-use-the-term-qa%e1%b8%8di-court-documents/)

Leave a Reply