190 emilie savage-smith
Fraxinetum, the latter a port near Nice under Muslim rule from 888 to 972 (Pinna,
1996: 45).
The best known Islamic representation of the Mediterranean Sea occurs on a
circular world map associated with Muḥammad ibn Muḥammad al-Idrısı̄ ̄, who in
1154 completed his illustrated geography for Roger II, the Norman king of Sicily
(Maqbul Ahmad, 1992; Kahlaoui, 2008a: 167–222). Idrısı̄ is the most familiar to ̄
Europeans of all Arab geographers, for an abridgement of his work, published in
Rome in 1592, was one of the first Arabic books printed in Europe, and a Latin trans-
lation followed 27 years later (Hopkins, 1990: 318). Idrı̄sı was born in Morocco in ̄
1100 and educated in Córdoba. He was a descendant of rulers of Málaga who had
been forced to move to Ceuta. During his lifetime he traveled widely, not only in
Spain, Morocco, and Sicily, but also to the southern coast of France and England.
About 1138, Roger II invited him to his court in Palermo; some scholars have sug-
gested that Roger was more interested in Idrısı̄ ’s political connections with prominent ̄
families in North Africa than his cartographic skills, for Roger had political ambitions
for expanding his influence in Spain and North Africa. Be that as it may, it is for his
illustrated geography—happily titled Entertainment for Someone Who Longs to Travel
the World (Nuzhat al-mushtāq f ı̄ ikhtirāq al-āf āq)—prepared in 1154 for Roger II,
that Idrı̄sı is now remembered.̄
Six preserved copies of Idrısı̄’s illustrated geography open with a circular map of ̄
the world (for examples, see Maqbul Ahmad, 1992: figs. 7.1–7.5 and Plate 11;
Kahlaoui, 2008b). The map is clearly a descendent of the Balkhı school tradition ̄
(Pinto, 2002; Tibbetts, 1992). It is, however, in some respects more sophisticated,
but in others ways less precise. Furthermore, while the general conformation of the
map is derived from the early Balkhı school world maps of the tenth century, the ̄
source for place names is decidedly different.
In this so-called “Idrısı̄ World Map,” the source of the Nile (called the “Mountain ̄
of the Moon”) is indicated by a prominent parachute-like emblem. Within the
Mediterranean itself there are the three islands typical of the Balkhı̄ school, although
Sicily is disproportionately large and the other two are not represented as round.
There are only a few labels on the North Africa coast, almost none in the Levant, and
even fewer occur on the European coast. Inland on the European continent there are
labels for the Alps (jabal al-Muntijūn) and for Gascony (Ghashkūnıyah̄ ). The Venetian
Gulf or Adriatic (al-Khalıj al-Bunduqı̄ ) is labeled, but not triangular in form. No cit-̄
ies in Italy or Greece are named. Even Constantinople is missing.
While this circular world map is frequently reproduced, it does not represent
Idrısı̄’s major contribution to the history of cartography, and in fact Idrı̄ ̄sı does not ̄
ever refer to the circular world map in the course of his treatise. Idrısı̄ ’s treatise is ̄
based on the idea of mapping each of the seven “climes” (klimata) recognized in
antiquity. For this, he employed the classical system by which the inhabited world was
divided into seven regions or strips parallel to the equator, called “climes” (iqlım̄ , pl.
aqālim). They were defined by the length of the longest day of the year, for the length
of daylight on the longest day of the year differed at the mid-point of each strip by
half an hour. Thus in the first “clime” (at the equator) the longest day is 13 hours, in
the second “clime” 13½ hours, and so on. The “climes” were numbered 1 to 7, from
south (the equator) to north. The habitable world was thought to cover an
expanse 160 ° wide—from the Fortunate (Canary) Islands in the west to China in the