Documents Entangled: Tax Receipts from the Bamiyan Papers and the Cairo Geniza
Introduction
In 2022, while at a conference in Oxford,1 I delivered a preliminary analytical paper on land management and taxation based on the Persian Bamiyan Papers (aka Afghan Geniza); an 11th–13th-century corpus studied by the Invisible East team in Oxford since 2019. At the Q&A, Lorenzo Bondioli excitedly jumped up to the lectern and showed us an image of an Arabic tax receipt from the Cairo Geniza. It was identical to the Bamiyan receipt I had just shown, not in language but in format, style and scribal marking patterns! What ensued was some kind of epiphany: we had, for the first time, seen direct evidence that state documentation practices, from Nile to Oxus, were intimately intertwined, and this, at a time when Abbasid caliphal rule was but a distant memory.
How could we explain the uncanny resemblances, two to three centuries after the Abbasid Golden Age had lost its shimmer, far into the period of independent caliphal and sultanic rule in Egypt and Khurasan? What was the connection? Moreover, was this just one example of a much bigger phenomenon? In other words, were the commonalities in transactional documentation apparent in other areas of governance and administration, too, or was taxation documentation a special case? Could we speak of “Islamicate documentation” and expand the usual scholarly dialogue, focussed largely on Egypt during this period (see Goitein, Udovitch, Goldberg, Ackerman-Lieberman, and more recently, Wickham)? Could we, by making “the East” less invisible, solidify and diversify the source base for the studies of Islamicate state-sponsored economic regulation?
To answer these questions, we need to backtrack a little, and consider the “long-lost sibling” tax documents, as Bondioli called them, in more detail. Before doing so, some context on the Bamiyan corpus itself may be helpful, which, due to its novelty in the scholarly discourse, is far less understood than the much better studied Cairo Geniza.
The Bamiyan corpus
By the term, “Bamiyan Papers,” I refer specifically to the documents that have been held in the National Library of Israel (NLI) since 2016, as well as a small group of legal and personal documents published decades earlier (Minorsky 1943, Scarcia 1963, 1966, Sarvar Humayun 1965, 1966), all connected, directly or indirectly, with 11th–13th century Bamiyan in present-day Afghanistan. Bamiyan’s history reflects its role as a cultural crossroads and administrative hub in central Afghanistan. Known globally for its towering Buddha statues carved in the late antique period well before the arrival of the Muslims—tragically destroyed by the Taliban in 2001—Bamiyan evolved from a Buddhist centre into a multi-religious, politically significant region under Islamicate rule from the 11th century onwards. Until now, not much was known of Bamiyan’s history during the 11th–13th centuries. This lacuna in the scholarship heightens the value of the Bamiyan Papers as sources for medieval Persianate and economic history, to add to studies of Persianate rural administration carried out by a handful of scholars in the 1960s–80s using data from literary sources on regions further west in contemporary Iran (notably, A.K.S. Lambton).
The Bamiyan Valley, and the empty niche in the sandstone cliffs that used to house the Large Buddha of Bamiyan, 58m tall, until 2001. With permission from Andy Miller
The Bamiyan Papers include documents and literary texts written in Hebrew, Judeo-Persian, New Persian and Arabic. A large number (ca. 220) was purchased in 2013 and 2016 by the NLI. Unfortunately, none of the Bamiyan Papers is provenanced, which poses a significant heritage problem, as well as ethical issues for scholars researching them. The NLI digitised them quickly, and access to this rare cache was opened up to scholars. More are held by antiquities dealers in London and are, sadly, not accessible to the public. The NLI’s Bamiyan Papers include around 125 documents, mainly written in New Persian, that have now been identified as related to an archive of a state warehouse that operated in Bamiyan between 549/1155 and 618/1221. (See my forthcoming book on this). Ofir Haim, who was the first to examine the full corpus, called them the phase-2 set of documents. (An earlier phase-1 set includes less than 20 texts written in the first part of the 11th century during Ghaznavid rule, many written in Judeo-Persian and which probably belonged to a Jewish family archive).
The phase-2 set belongs to the period of Shansabānī Ghurid and Khwārazmshāhī rule (549/1155 to 618/1221–2), when Bamiyan was a provincial capital, during which time many of the Bamiyan Papers were produced. Bamiyan’s vibrant bureaucratic and agricultural life, as evidenced in the documents, was violently interrupted by Chinggis Khan’s destruction of Bamiyan in 618/1221–2, an event that halted the region’s administrative record. The final Bamiyan Paper in the NLI is dated just before the Mongol siege, capturing the last traces of a complex local order—moments before its erasure from the political map.
From internal evidence, we can glean that the state warehouse (lit. anbār-i khāṣṣa) was run by a warehouse manager (anbār-dār), a certain Abū Bakr b. ʿUmar from 549/1155 until at least 581/1185. In documents from 569/1174 onwards, the warehouse manager (anbār-dār) is addressed with the senior administrative titles muʿtamid al-mulk (‘trusted chief of the empire’) and amīn-al-ḥaḍra (‘his majesty’s controller’ or ‘commissioner’). A second important character is ʿAlī b. Abī Bakr, who seems to have been Abū Bakr b. ʿUmar’s son and successor as warehouse manager until 618/1221–2. The case of Abū Bakr and ʿAlī suggests that the position of warehouse manager was passed down the generations within the same family which is perhaps not surprising given that it is quite common to find a son doing a father’s work.
The Bamiyan Papers reveal that anbār-dārs such as Abū Bakr b. ʿUmar had at least five key responsibilities. The first was to receive and store incoming tax payments (mainly grain). Second, they took proactive steps to collect tax arrears directly from farmers. Their third task was to serve as intermediaries in the transfer of grain to processing sites, such as flour mills. The fourth role of the anbār-dār was to manage the granary where grain was kept in reserve to be issued as loans to the cultivators (to be repaid in the future). And fifth, they ensured that administrative protocols were adhered to and appropriate records were kept, and they took remedial action in cases of suspected mismanagement or abuse.
The tax receipts in the Bamiyan Papers attest to the warehouse’s function as a depository of grain tax deliveries on behalf of the provincial state that was based in Bamiyan but covered a much wider jurisdiction along an east–west route of 280km from the Bamiyan Valley to the Ghorband Valley and beyond up to the Panjshīr Valley, as well as in the passes connected to these valleys (see map 1). At each drop-off, the warehouse issued a receipt, and retained copies for the file, which we can surmise from the annotations on the documents written by a later hand, possibly that of an accountant.
NLI, Ms.Heb.8333.1752 is a typical example of about 20 Persian receipts (Ar./Per. barāt) that have survived in the Bamiyan Papers corpus. It confirms payments of what seems to be a tax that is paid in kind, usually in coarse grain (ghalla) or wheat (gandum), occasionally in barley (jaw). The well-known term referring to taxes on land revenue, kharāj, does not fall in the receipts, but ghalla-yi taḥakkumāna does. The compound term was not identifiable to us in the literary record, but using its triliteral Arabic root, ḥ.k.m (ḥukm, i.e. “law” or “rule”) as a basis, we have provisionally translated it to “statutory grain [tax].”
Map 1: showing some key toponyms that occur in the Bamiyan Papers, including Chawqānī and Bamiyan
A close-up view of NLI, Ms.Heb.8333.175
Transcription:
Recto
1. ابو بکر بن عمر
2. ... (؟)
3. تسلیم کرد عمر مودود از جوقانی از غله تحکمانه
4. انبار خاصه گندم
5.<نصفه> اربع3 (؟) منا هشت من
6. این خط حجت باشد بتاریخ فی ذالحجه سنه اربع و ستین و خمسمائه
Verso
1. براه کندم تحکمی
2. چوقانی
Translation:
Recto
1. Abū Bakr b. ʿUmar
2. [Bismillāh?]
3. ʿUmar [b.] Mawdūd from Chawqānī has delivered the statutory grain (ghalla-yi taḥakkumāna) to
4. the state warehouse (anbār-i khāṣṣa) Wheat
5. [Half of it:] Four (?) mann Eight mann
6. This document shall serve as proof, [written] on the date of Dhū al-Ḥijja in the year 564 [September 1169].
Verso
1. Receipt for statutory grain
2. Chawqānī
The document provides all the necessary information one would expect from a receipt: the name of the payer and the payee, the amount, the receiving account, and a date. The documentation includes an archival note at the back referring to the type of payment and the tax unit: a town called Chawqānī (identified as a place in today’s Baghlān district, see Map 1).4
None of the formatting is incidental. The earliest surviving Persian secretarial manuals, Muḥammad b. ʿAbd al‑Khāliq Mīhanī’s Āyīn-i dabīrī and Dastūr‑i dabīrī, which he composed in the 12th century, and which survive in a manuscript copy dated Dhū al‑Qaʿda 585/December 1189–January 1190,5 demonstrate the emphasis given to correct protocols by trained scribes. Mīhanī stipulates, for example, three levels of quill thickness and line spacing, depending on the seniority of the sender (the more senior the sender, the thicker the quill and the greater the spacing). Mīhanī also gives directions regarding the folding of sheets and the importance of leaving empty spaces at the top of each page.6 The modern editor of Mīhanī’s work, Akbar Naḥwī, identified the nisba Mīhanī as related to a place in Khurasan that was destroyed by the Mongols. This means that the text not only is contemporary with the phase-2 Bamiyan Papers but also comes from the same region of Khurasan.7
If we now compare NLI, Ms.Heb.8333.175 with a barāt in the Cairo Geniza, taking T-S Ar.39.168, we find the same kind of information provided, but more importantly, following exactly the same, distinctive documentary conventions. Both documents begin with bismillāh (in the Bamiyan document probably rendered as an abbreviation on the top of the sheet), followed by four statements. The first establishes the payer’s name and residence (ʿUmar Mawdūd from Chawqānī in the Bamiyan document, Mūsā b. ʿAlī from al-Kharas [?] in the Cairo one) and the type of tax towards which the payment is made (ghalla-yi taḥakkumāna in Bamiyan, kharāj in Cairo). The second element is the recipient or purpose of the payment: a grain tax payment to the state warehouse in the Bamiyan Papers and a kharāj payment including, or accruing from, the use of a waterwheel (rasm ʿamal al-sāqiya) in the Cairo Geniza document. Third, the document specifies the exact amount paid in a distinctive column format: the left side gives the full amount, written out in words (eight mann in the Bamiyan case, one dinar in the Cairo one), while the right side specifies the half-amount (tanṣīf). The convention of specifying half-amounts in documents related to exchanges of goods or money is widespread across the Middle East and Asia. Its purpose is to prevent fraud and falsification by providing an easy way to check and confirm the full amount. It is the scribal convention that is interesting because it is distinct to both corpora: the half-amounts are identified by a line drawn over the number written out in Arabic accountancy shorthand (siyāq).8 This is the earliest attestation of siyāq in Persian documentation, previously only identified in documents from the 14th century onwards. The fourth and final item is a closing statement with a date. The Bamiyan document also states that the document shall ‘serve as proof’, a formulaic phrase that appears in many other documents as well. One of the few variations is that the issuer’s name appears in the top left-hand corner of the Bamiyan document, whereas in the Cairo Geniza document the issuer signs off at the bottom of the body of the text.
Finally, the same diagonal jāʾiza stroke has been drawn across both documents. This is not a random line but a cipher (jāʾiza) that was used by secretaries to compare the amounts delivered with inventories of amounts owed. This mark, which Hinz calls a ‘tick mark’ (‘Abhaken’), also appears in dozens of Arabic papyrus tax lists and is mentioned by al-Khwārazmī in his 10th-century manual on standard practice.9 This scribal practice is also described in later accountancy manuals in Arabic and Persian from the fourteenth century, such as al-Qalqashandī’s Arabic Ṣubḥ al-aʿshā from Egypt,10 and the Saʿādatnāma, Risāla-yi Falakiyya and al-Murshid fī al-ḥisāb in Persian.11
Plate 1 Plate 2
Long-lost siblings: Receipts NLI, Ms.Heb.8333.175 of the Bamiyan Papers (left) and T-S Ar.39.168 of the Cairo Geniza (right)
Plate 1: Tax receipt (barāt) in the Bamiyan Papers (NLI, Ms.Heb.8333.175) |
Plate 2: Tax receipt (barāt) in the Cairo Geniza (T-S Ar.39.168)12 |
|
|---|---|---|
| Abū Bakr b. ʿUmar | ||
| [Bismillāh] | Bismillāh | |
| ʿUmar Mawdūd from Chawqānī has delivered as statutory grain [tax] | A valid record has been made for Mūsā b. ʿAlī [for a payment towards] the kharāj for his cultivated land that is incumbent upon him | |
| in al-Kharas (?), in the months of the year 516 [1122–3], to be used for digging [+/− 1] required for operating | ||
| State warehouse wheat | the water wheel | |
| 8 mann [half-amount:] 4 mann | One dinar in cash. Half [..+ 1?..]13 | |
| This document shall serve as proof, [written] on the date of Dhū al-Ḥijja 564 [September 1169] | And it was written by Yūsuf b. Ibrāhīm in the first quarter of Jumādā II of the year 517 [1123–4] | |
| With the mercy of God, the singular one |
Conclusion
How do we explain the congruence between the two corpora? The commonalities across such a wide area from Egypt to Afghanistan may be ascribed to a shared substrate of Abbasid documentation protocols and templates that were transmitted through common training methods, manuals, and specialisation schemes. We can single out the prescriptive manuals for administrators (inshāʾ) and notaries (shurūṭ) written in both Arabic and Persian as examples of vectors through which professional standards implemented throughout the Islamicate lands were transmitted. Then again, while these works do offer templates and talk about style and pens, you would not be able to produce two documents as similar as these two by just reading the manuals. Something else is at work here; administrators may be moving and carrying with them their training and possibly documents themselves, to use them as models. Personnel mobility in the Abbasid period (and beyond) may, therefore, be an even more likely explanation. Mīhanī’s Persian 12th-century manual is one such vector, as is al-Qalqashandī’s 14th-century Arabic Ṣubḥ al-aʿshā. Both will have used preceding manuals, perhaps dating back to the Abbasid Golden Age. Besides these traces of a common heritage, regional variations on the substrate enabled adjustments to local realities in matters of vernacular, ecology, and so on.
There remains much more to be explored and discovered as scholars continue to compare the documentary records in the Bamiyan Papers with the Cairo Geniza, and other contemporary archives across the Islamicate world. The lasting legacy of a possible Abbasid substrate of documentary formats and scribal traditions right up to the 11th to 13th century (and beyond), and its transmission across linguistic communities remains a story to be examined in far more depth and across social and economic sectors. Perhaps the most fascinating aspect of the tax receipts in the Bamiyan Papers is the evidence they contain of the high level of conformity and penetration of “Islamicate” administrative documentary traditions right down to the village level, making the medieval rural “hinterland” of what is Afghanistan today seem far less remote or ungovernable than one might suspect.
Footnotes
1 Invisible East conference, ‘A Hard Row to Hoe: Landowning and Land Management in the Medieval Islamic World (622–1250)’ on 12–14 December 2022.
2 For image and transcription, see https://www.invisible-east.org/corpus/204/.
3 Palaeographically, it may also end in a ة.
4 https://www.geonames.org/1469969/chowgani.html.
5 The manuscript is incomplete and is held at the Fātiḥ Library in Istanbul (no. 4079). The editor used a photo of the manuscript deposited at the Central Library of the University of Tehran (no. 4648). See Akbar Naḥwī’s introduction to Muḥammad b. ʿAbd al‑Khāliq Mīhanī’s Āyīn‑i dabīrī (Tehran: Markaz‑i Nashr‑i Dānishgāhī, 1389/2010–11), نه–ده (ix–x).
6 Mīhanī, Āyīn‑i dabīrī, 9–10.
7 Naḥwī’s introduction to Mīhanī, Āyīn‑i dabīrī, شش (vi).
8 ‘Arabic numerals’, said to have been transmitted from India to the Islamicate world, do not appear anywhere in the Bamiyan Papers.
9 Hinz, ‘Rechnungswesen’, 9–11; Bosworth, ‘Abū ʿAbdallāh al-Khwārazmī’, 128–9.
10 Al-Qalqashandī, Kitāb Ṣubḥ al-aʿshā fī ṣināʿat al-inshāʾ, ed. Muḥammad ʿAbd al-Rasūl Ibrāhīm (Cairo: Wizārat al-Thaqāfa, 1963–70 [1913–19]). I benefited from a presentation by Malika Dekkiche on al-Qalqashandī’s work and comparative discussions with David Durand-Guédy and Frédéric Bauden at the Invisible East conference ‘State Documents from the Medieval Islamicate World’, held at Trinity College, Oxford, on 21–22 June 2023.
11 The Risāla-yi Falakiyya was first recognised by the Turkish scholar A. Z. V. Togan in ‘Mogollar devrinde Anadolu’nun iktisadî vaziyeti’ [The economic situation in Mongol-era Anatolia], Türk Hukuk ve İktisad Tarihi Mecmuası 1 (1931): 3–6, 14–15. Subsequently, it has been discussed by Hinz and Nabipour: Hinz, Die Resala-ye Falakiyye, which was reviewed by Bertolt Spuler in Der Islam 31 (1954): 260; Nabipour, ‘Die beiden persischen Leitfäden’.
12 Khan, Arabic Legal and Administrative Documents, 261.
13 It is also possible that the scribe first wrote “1 dinar exactly (سوا)” and then corrected it to something else, perhaps a fraction: “1 dinar and ¼ (وربع), and it may also be possible to read the half as : “½ and ¼.”
Bibliography (not including primary source references given in Footnotes)
Ackerman-Lieberman, Phillip I. The Business of Identity: Jews, Muslims, and Economic Life in Medieval Egypt. (Stanford Studies in Jewish History and Culture). Stanford: Stanford University Press, 2014.
Bosworth, C. Edmund. ‘Abū ʿAbdallāh al-Khwārazmī on the Technical Terms of the Secretary’s Art: A Contribution to the Administrative History of Mediaeval Islam’. Journal of the Economic and Social History of the Orient 12, no. 2 (1969): 113–64.
Goitein, Shelomo Dov and Mordechai Friedman. India Traders of the Middle Ages. Documents from the Cairo Geniza 'India Book'. (Études sur le judaïsme médiéval, Volume: 31). Leiden : Brill, 2007.
Goldberg, Jessica L. Trade and Institutions in the Medieval Mediterranean: The Geniza Merchants and Their Business World. (Cambridge Studies in Economic History.) New York: Cambridge University Press, 2012.
Haim, Ofir. ‘What Is the “Afghan Geniza”? A Short Guide to the Collection of the Afghan Manuscripts in the National Library of Israel, with the Edition of Two Documents’. Afghanistan 2, no. 1 (2019): 70–90.
Hinz, Walther. ‘Das Rechnungswesen orientalischer Reichsfinanzämter im Mittelalter’. Der Islam 29 (1950): 1–29.
Hinz, Walter. Die Resala-ye Falakiyye des ‘Abdollah Ibn Mohammad Ibn Kiya al-Mazandarani: Ein persischer Leitfaden des staatlichen Rechnungswesens (um 1363). Wiesbaden: Harrassowitz, 1952. Review by Bertolt Spuler, in Der Islam 31 (1954), 260.
Khan, Geoffrey. Arabic Legal and Administrative Documents in the Cambridge Genizah Collections. Cambridge: Cambridge University Press, 2006 [1993].
Lambton, Ann K. S. Continuity and Change in Medieval Persia: Aspects of Administrative, Economic and Social History, 11th–14th Century. London: I. B. Tauris, 1988.
Lambton, Ann K. S. Landlord and Peasant in Persia: A Study of Land Tenure and Land Revenue Administration. London: I. B Tauris, 1991 [1953].
Minorsky, Vladimir. ‘Some Early Documents in Persian (II)’. Journal of the Royal Asiatic Society 1 (1943): 86–99.
Nabipour, Mirkamal. ‘Die beiden persischen Leitfäden des Falak ʿAlā-ye Tabrīzī über das staatliche Rechnungswesen im 14. Jahrhundert’. PhD dissertation, University of Göttingen, 1973.
Sarvar Humāyūn, Ghulām. ‘Kābīn-nāma, ki dar Bāmiyān bi sāl-i 470 h.q. nigāshta shuda’. Āryānā 256 (1343/1965): 1–13.
Sarvar Humāyūn, Ghulām. ‘Takmila bar maqāla-yi “kābīn-nāma”’. Āryānā 258 (1344/1966): 215–20.
Scarcia, Gianroberto. ‘An Edition of the Persian Legal Document from Bamiyan’. East and West 16, no. 3–4 (1966): 290–5.
Udovitch, Abraham L. Partnership and Profit in Medieval Islam. Princeton, N.J.: Princeton University Press, 1970.
Wickham, Chris. The Donkey and the Boat: Reinterpreting the Mediterranean Economy (950–1180). Oxford: Oxford University Press, 2023.
If you enjoyed this Fragment of the Month, you can find others here.
Contact us: genizah@lib.cam.ac.uk
One of the manuscripts in this article is part of the Cairo Genizah Collection in Cambridge University Library. To see more items from this collection visit: https://cudl.lib.cam.ac.uk/collections/genizah/1