What 'Lost' Means in 'Lost Islamic History'

The phrase has become a small industry. Lost Islamic history. Forgotten Muslim heritage. The civilisation the West erased. It is on book covers, conference posters, YouTube thumbnails. It has done useful work — it has drawn readers to material they would not otherwise have looked at — but it deserves a harder look, because the word lost is doing several different things at once, and at least one of those things is misleading.

The records, in the main, are not lost.

What survives, and where

A serious working scholar of Islamic history today has access to an almost embarrassing wealth of primary material. The Süleymaniye Library in Istanbul holds something on the order of 100,000 manuscripts. The Dar al-Kutub in Cairo holds tens of thousands more. The Khuda Bakhsh Oriental Public Library in Patna, the Salar Jung Museum in Hyderabad, the Rampur Raza Library — each holds substantial collections that, taken together, document centuries of South Asian Islamic intellectual life in Arabic, Persian, and Urdu. The Bibliothèque Nationale in Paris, the Bodleian, the British Library, Princeton, Leiden, Yale's Beinecke — Western institutions, acquired by various means in the nineteenth century, also hold thousands of items that have been catalogued, digitised, and in many cases edited.

This is before we count what is still in private family libraries from Fez to Timbuktu to Lucknow, much of it now being slowly catalogued by projects like the Hill Museum's manuscript programme or the Tombouctou Manuscripts Project.

The textual base is not the problem. The problem is what has been done with it — or rather, what has not been done with it, in English, for a non-specialist audience.

The Andalusi case

Consider the historians of al-Andalus. Ibn Hayyan of Cordoba, who died in 469 AH / 1076 CE, wrote a vast chronicle called al-Muqtabis covering the Umayyad emirate and caliphate in Iberia. Substantial portions survive. They have been edited in Arabic, by Spanish and Arab scholars, across most of the twentieth century. Only fragments exist in English translation. An English-language undergraduate taking a survey course on medieval Europe will read Bede, Einhard, William of Malmesbury, perhaps Anna Komnene. They will not read Ibn Hayyan. They will not even be told he exists.

The same is true of Ibn al-Khatib of Granada, whose fourteenth-century history of his city is one of the finest pieces of late-medieval prose in Arabic. The same is true of al-Maqqari, whose seventeenth-century compendium Nafḥ al-Ṭīb is the single best source for the cultural life of Muslim Spain — partial English versions exist, but the full work has never been translated. The same is true of dozens of others.

None of this is lost. All of it is on a shelf somewhere. It simply has not been brought across into the language in which most of the world now reads about the past.

The Mughal court

Move east. The Mughal chancery produced, across three centuries, a body of court chronicle, biographical dictionary, and administrative manual that rivals anything in the early modern world. Abu al-Fazl's Akbar Nama and Aʾin-i Akbari were translated into English in the late nineteenth century by H. Blochmann and H. S. Jarrett, and remain — astonishingly — the standard English versions. The Padshah Nama of Lahori, the official chronicle of Shah Jahan's reign, has been only partially translated. The Maʾasir al-Umara, a biographical encyclopaedia of Mughal nobility, was translated by Beveridge a century ago in a version that needs serious updating.

The contrast with the Tudor and Stuart archives is instructive. Every significant English court chronicle, ambassadorial dispatch, and parliamentary diary of the same period exists in modern critical editions, often in multiple translations, with full scholarly apparatus. The Mughal equivalents largely do not. A graduate student writing on Akbar's religious policy must still, in 2024, work primarily from a Victorian translation produced by a Bengal civil servant.

The Ottoman flattening

The Ottoman case is different in kind. Here the material has been worked on intensively in Turkish, German, and increasingly in English over the last forty years. The flattening is interpretive rather than archival. For most of the twentieth century, English-language writing on the Ottomans was dominated by the decline thesis — the idea that after Süleyman the empire entered a long, slow degeneration, broken only briefly by the Tanzimat reforms of the nineteenth century. This frame has been comprehensively dismantled by a generation of scholars — Cemal Kafadar, Suraiya Faroqhi, Baki Tezcan, Dana Sajdi — but the dismantling has not yet reached the popular narrative. The eighteenth-century Ottoman world, in particular, is one of the most intellectually active periods in Islamic history, and almost none of that activity is legible to an English-language general reader. The poets, jurists, ʿulamāʾ, Sufi commentators, and natural philosophers of Ottoman Damascus, Cairo, Aleppo, and Istanbul in the seventeenth and eighteenth centuries wrote a great deal, and that writing exists, and almost none of it has been translated.

What was actually lost

If we want to use the word lost, then, we should be careful with it. What was lost is not the manuscripts. It is the place this material used to hold in the general account of the past. There was a moment — roughly the long nineteenth century — when the Western academy decided which historical traditions belonged in the core of the human story and which belonged in the specialist annexe labelled Oriental Studies. Islamic history was placed in the annexe. The English-language popular narrative inherited that placement and has never fully revised it.

That is not a story about destruction. It is a story about curriculum, about which books are assigned, which languages are taught, which translations are funded, which scholars are hired. These are recoverable choices. The material is waiting.

The point of saying so is not to scold. It is to set the actual work in front of the people who want to do it. Lost implies that the recovery would require excavation, perhaps a miracle. Untranslated, unread, unassigned — those are the more accurate words, and they are also more hopeful, because they describe problems that a generation of careful readers and editors can actually fix.