Over the past decade, Jean-Charles Ducène has provided students of medieval Arabic–Islamic geography with editions and translations of several lesser-known texts, such as Dalā’il al-Qibla of Ibn al-Qaṣṣ, and comparisons of several manuscripts of the same geographical work, such as the maps of the Nile Delta by Ibn Ḥawqal. Together, his editions and translations bring us closer to understanding the geographical corpus as a developing discourse, which is not limited to a group of canonical texts. By focusing on the manuscripts that have been overlooked, we get a much better feel for the ways in which geographical knowledge and categories were transmitted across the Islamic world over time and space.

Ducène's current contribution is an edition and translation of a geographical work attributed to the Rasulid sultan al-Malik al-Afḍal (r. 1363–77). The work, beyond opening a rare window into geographical knowledge available in the medieval Yemen, is of particular importance for the development of the genre of mathematical geography. Unlike the more familiar tables of co-ordinates, which consist only of numerical data on longitude and latitude, each of the 515 localities in this treatise also receives a brief textual account of the population and main commercial products. This is a marriage of the zīj – purely mathematical tradition, with an emphasis on astrological and astronomical uses – and the tradition of administrative and commercial geography. It seems to precede similar geographical tables compiled by Abū al-Fidā’, and to represent a new genre of geographical literature that became popular from the twelfth century onwards. It is an interesting variation that parallels al-Idrīsī’s more famous attempt at categorizing geographical information within a mathematical framework.

In the introduction, Ducène notes several other Muslim rulers who have taken a keen interest in science, not least Abū al-Fidā’ himself. The Rasulid period in the history of the Yemen is a time of increased maritime trade with India, as demonstrated by Eric Vallet. Commerce may have fostered the sovereign's interest in geography, although al-Afḍal is said to have composed works in several other branches of science, such as agriculture and genealogy. The manuscript as a whole, dated to 777/1375, contains marginal annotations by the Sultan himself, demonstrating that these tables were made for the Sultan's personal use.

The tables appear as two separate sections of the manuscript, which together form a sequence that begins in the Arabian Peninsula and gradually moves away from the central regions of the Islamic world. Given that a significant proportion of localities are from Iranian-speaking regions, Ducène speculates that the original zīj was composed in the entourage of the observatory of Marāgha in the late thirteenth century. The descriptive text of the various localities is largely taken from the Balkhī geographers, in particular Ibn Ḥawqal. It contains some previously unknown information on localities in the Red Sea as well as in Central Asia, although the important innovation here is the way in which the two traditions are brought together. Another distinctive aspect, which the author should perhaps have highlighted more, is the designation of groups of localities by racial categories (Arabs, Turks, Blacks, Rūm, etc.). These are reminiscent of categories utilized in Ismaili missionary daʿwa, and reinforce the association with the observatory established by the Ismaili Nāṣir al-Dīn al-Tūsī.

The edition and translation pose serious challenges, not least because the current location of the manuscript itself is unknown, and the editor had to rely on photographs. Ducène provides very reliable readings and accessible, accurate translation. The numerical values, which so often become distorted in medieval manuscripts, are checked against other works of mathematical geography. This work will be useful for those studying the history of the Yemen in the medieval period, for the development of geographical knowledge in the Islamic world (full indexes in Arabic and in French are provided), and for mapping the history of Arabic–Islamic geography as a discipline.