How to read the sources a figure page cites
Every one of the 153 figure pages in this atlas ends with a Sources list: 150 figures cite three works and three cite four, 462 citations in all, drawn from 403 distinct entries. That list is the most useful part of the page and the part most readers skip, because it tells you which of the claims above it you can lean on.
Nubala · Published · 8 min read
The short version: a citation to a modern university-press monograph is doing a different job from a citation to a 17th-century Arabic compilation, and both are doing a different job from an Encyclopaedia of Islam entry. A chronicle tells you what a tradition said; a monograph tells you what a historian concluded after weighing the chronicles; an encyclopedia entry tells you what the field currently treats as settled. Knowing which you are looking at is most of source criticism.
The four kinds of entry in this bibliography
Across the 403 entries, four kinds account for nearly everything, and the proportions tell you what sort of reference this is.
The dominant kind is the modern scholarly monograph: 102 of the 462 citations name a university press. These are the works that have already done the hard part - comparing the surviving accounts, discounting the ones with an obvious interest, and saying what can be supported. Salahuddin al-Ayyubi is a good example: his page cites Lyons and Jackson's Saladin: The Politics of the Holy War (Cambridge, 1982) alongside Carole Hillenbrand's The Crusades: Islamic Perspectives (Edinburgh, 1999), which is the single most-cited work in this atlas.
The second kind is the premodern work itself, cited by its own date. There are seven of them, and each appears because that text is the main surviving witness for the figure above it. Mansa Musa cites al-Umari's Masalik al-Absar of about 1340; Askia Muhammad and the other Songhay figures cite al-Sa'di's Tarikh al-Sudan of about 1655 and Mahmoud Kati's Tarikh al-Fattash of about 1519; the Andalusi pages cite al-Maqqari's Nafh al-Tib, a 17th-century compilation; Idris Alooma cites Ibn Fartua's Kanem wars account of about 1600.
The third kind is a translated primary source: 29 citations carry a translator, across 28 figures. These are how a reader with no Arabic, Persian or Ottoman Turkish can go to the text itself - Ibn Battuta through Gibb's Hakluyt Society translation, Ibn Khaldun through Rosenthal's Muqaddimah, Usama ibn Munqidh through Cobb's Book of Contemplation, Babur through Thackston's Baburnama.
The fourth kind is the reference entry: 42 citations to encyclopedias, 28 of them to the Encyclopaedia of Islam and four to Encyclopaedia Iranica. These are the best starting point for a figure you know nothing about, because an Encyclopaedia of Islam entry is a signed, peer-reviewed summary with its own bibliography.
| Kind of source | Citations | What it can establish | What it cannot |
|---|---|---|---|
| Modern monograph, university press | 102 | What the evidence supports, after weighing the accounts | Anything published after it; fields move |
| Encyclopaedia of Islam or Iranica entry | 32 | The field's settled position, and where to read next | Detail, or an argument you can test |
| Translated primary source | 29 | What a participant or near-contemporary actually wrote | Whether what they wrote was true |
| Premodern chronicle or compilation | 7 works | What a tradition recorded, and when it recorded it | Independent confirmation of its own stories |
| Specialist journal article | several | A narrow question answered in detail | The general picture |
| Web archive or UNESCO documentation | 4 | A convenient summary or a heritage designation | The weight of a peer-reviewed source |
The distance problem, with a worked example
The question to ask of any premodern citation is how far the source sits from the event. This atlas contains a case where the distance is nearly six hundred years, and the page is honest about it.
Sixteen figures are filed under Al-Andalus, and six of the citations across the atlas are to al-Maqqari's Nafh al-Tib, written in the seventeenth century, long after the fall of Granada in 1492. Lubna of Cordoba - the mathematician and librarian at the court of Al-Hakam II, dated here to about the 940s to about 984 - is one of those pages. Nearly everything known about her comes through compilations like al-Maqqari's, which preserve earlier material that has not otherwise survived.
That does not make the page wrong. It makes it a report of what a seventeenth-century Moroccan scholar preserved from earlier Andalusi writers, and it sets a limit on precision: her birth is given as a decade, not a year. Notice how her page handles it - the al-Maqqari citation sits alongside two modern works, Jayyusi's Legacy of Muslim Spain and Manuela Marin's chapter on women in al-Andalus, which is the right shape for a claim resting on a late compilation: the chronicle for the material, the modern scholarship for what can be made of it.
Compare that with Salahuddin al-Ayyubi, where the primary witness is Ibn Shaddad, who served him and wrote about him directly. There the distance is nearly zero and the problem is the opposite one: a source close enough to the subject to be partisan. Both situations are normal; they just license different claims.
How to check a single claim on a figure page
A method that works in five minutes, using only what the page gives you.
Start by classifying the claim. A date, a title or an institutional fact - who founded what, who held which office - is normally traceable. A claim about motive, character or a conversation is usually a tradition, and tradition is what chronicles are made of.
Then match the claim to the right citation. Take the encyclopedia entry first if the page has one: it will tell you in a paragraph whether the claim is standard, contested or anecdotal. Take the monograph if the claim is interpretive. Take the primary source if the claim is about what someone wrote or said.
Then check the page's own hedging. The atlas marks uncertainty in the date lines - a tilde for an estimated year, an explicit uncertain where scholarship disagrees, as on Tariq ibn Ziyad's death - and that hedging is a reliable signal that the underlying evidence is thin. See Hijri and Gregorian dates for what each form of date means.
Finally, if the claim matters, go to the work. The citations here are complete enough to find in a library catalogue: author, title, publisher, year. That is the whole point of printing them.
- 462 citations across 153 figures - 150 pages cite three works, three cite four.
- 403 distinct entries. Two of them are the same work under slightly different punctuation, so the number of distinct works is marginally lower.
- The most-cited works are Hillenbrand's The Crusades: Islamic Perspectives and al-Maqqari's Nafh al-Tib, six citations each.
- Only two citations in the whole atlas are to a web archive, both to MacTutor for a mathematician's biography.
- No figure page carries a footnote: the Sources list covers the page, not a specific sentence.
What the bibliography cannot do, stated plainly
Three limits worth knowing before you cite this atlas in your own work.
The citations are page-level, not sentence-level. Three works are listed for a profile of several hundred words, and nothing says which sentence came from which. For a claim you intend to repeat, treat the list as a reading list rather than as a reference.
The biography text itself is an editorial summary. It is written from the cited scholarship, but it is not quoting it, and there is no apparatus to show where a summary has smoothed over a disagreement. Where two serious positions exist, a three-hundred-word profile will usually give one.
And the coverage of the bibliography follows the coverage of the scholarship, which is uneven. The Andalusi, Crusader-era and Mughal pages draw on dense, well-established literatures - Kennedy, Menocal, Hillenbrand, Richards. The Sahelian and Southeast Asian pages draw on a much smaller one, which is why they lean more heavily on chronicles like Tarikh al-Sudan and on a handful of specialists such as Levtzion, Boyd and Reid. That is a fact about the field rather than about the figures, and it is visible in the Sources lists if you look for it.
Questions readers ask
How many sources does each figure page cite?
Three for 150 of the 153 figures, and four for the remaining three - 462 citations drawn from 403 distinct entries. The list sits at the foot of every figure page.
Is a 17th-century chronicle a reliable source for the 10th century?
It is reliable evidence of what the tradition recorded, not of the event. Six citations here are to al-Maqqari's Nafh al-Tib, a 17th-century compilation that preserves earlier Andalusi writers now lost. Pages resting on it, such as Lubna of Cordoba, pair it with modern scholarship and give dates to the decade.
Which source should I read first?
The encyclopedia entry if the page has one - 32 citations here are to the Encyclopaedia of Islam or Encyclopaedia Iranica, and each is a signed, peer-reviewed summary with its own bibliography. Then the monograph for the argument, then the primary source for the text.
Can I cite a Nubala page in an essay?
Cite the works it names, not the page. The citations are page-level rather than sentence-level and the biography is an editorial summary, so the Sources list is a reading list that will take you to something citable.
Why do some regions have thinner bibliographies?
Because the scholarship is thinner. The Andalusi and Mughal pages draw on large established literatures, while the Sahelian and Southeast Asian pages rely more on chronicles such as Tarikh al-Sudan and Tarikh al-Fattash and on a few specialists. The Sources lists make that unevenness visible.
Sources
- Carole Hillenbrand, The Crusades: Islamic Perspectives (Edinburgh University Press, 1999) - the most-cited work in this atlas.
- al-Maqqari, Ahmad ibn Muhammad, Nafh al-Tib min Ghusn al-Andalus al-Ratib, 17th century - cited on six Andalusi pages.
- al-Sa'di, Abd al-Rahman, Tarikh al-Sudan, c. 1655 - cited on the Songhay pages.
- Chase F. Robinson, Islamic Historiography (Cambridge University Press, 2003).
- Franz Rosenthal, A History of Muslim Historiography (Brill, 1968) - cited on the Ibn Kathir and Al-Maqrizi pages.
- Salma Khadra Jayyusi, ed., The Legacy of Muslim Spain (Brill, 1992) - cited on the Lubna of Cordoba page.
Counts in this guide are derived from the atlas data itself, so the prose cannot drift from the corpus. How to read the sources a figure page cites explains how the bibliographies are built and what each kind of work can carry.