the medieval mediterranean 79
trade in the Mediterranean, strongly disrupted by the insecurity caused, in particular,
by Muslim piracy, and by the Muslim conquest of major ports like Alexandria or
Carthage. Using textual sources, Pirenne then noted the disappearance from western
Europe of products characteristic of the eastern trade: Egyptian papyrus, Eastern cur-
rencies, spices or luxury textiles. His thesis therefore placed the end of antiquity and
the beginning of the Middle Ages in the Mediterranean in the seventh century with
the first Arab conquests seen as responsible for the crisis of trade. Pirenne’s conclu-
sions have been challenged by specialists of both Christian Europe and Islam; without
repeating the entire debate, we can focus on two important points: the situation in the
Mediterranean before the Muslim conquests, on the one hand, and the consequences
of those conquests, on the other.
One of the difficulties lies in the scarcity of sources, both textual and archeological,
which makes the first centuries of the Middle Ages a particularly dark period. After the
crisis of the fourth and fifth centuries, the Roman Empire was reconstituted in the
East; in the sixth century, Justinian’s reconquests succeeded in reconstructing an
important part of the Empire around the Mediterranean. New fortifications in port
cities and an effort to develop the fleet made the shores safer and facilitated the
resumption of trade. Some Mediterranean unity was thus reconstituted, marked by a
trade organized in order to supply the new capital of Constantinople with wheat
through the state annona (Wickham, 2005: 76). It was therefore no longer Rome but
Constantinople, in the eastern Mediterranean, which structured the trade networks
into which the western regions were barely integrated. Nevertheless, archaeological
finds, in particular amphorae, show the survival of long-distance exchanges between
the two basins of the Mediterranean, at least until the end of the sixth and beginning
of the seventh centuries. Yet is it possible to deduce, with Pirenne, that Roman unity
and prosperity continued until the eve of the Muslim conquest? First, it should be
noted that the Byzantine presence in the West was mainly confined to Italy, Sicily and
the eastern Maghreb—that is, the central Mediterranean. In the West only a few stra-
tegic points like the Straits of Gibraltar remained under imperial control: their func-
tion was primarily strategic, and they were less integrated into the trade networks. In
addition, there were signs of crisis at the end of Justinian’s reign: the plague that
struck Constantinople in 542 and recurring through the middle of the eighth century
affected demography to a degree difficult to specify; the wars between Byzantium and
the Sassanids weakened the Empire and affected the Eastern cities, where archaeology
shows a decline. At the same time, maritime exchanges tended to develop on a more
regional scale: through the beginning of the seventh century, amphorae carrying
African oil from Carthage are found in Rome, Ravenna, Naples, southern Spain,
Sardinia, and Marseilles, but seldom in the eastern Mediterranean, which already
appears as a separate economic zone.
The Muslim conquests were, therefore, not the cause of the downturn of the
Mediterranean trade; this, however, does not mean they had no consequences. The
most radical critique of Pirenne was the diametrically-opposed position adopted by
Maurice Lombard (1948): that the Muslim conquest and the constitution of the
Islamic empire fed a revival of Mediterranean trade due to the creation of a vast eco-
nomic zone spanning the Mediterranean and the Indian Ocean and connected with
both the north (via the Caspian Sea and Russian rivers) and the south (in particular,
via the African gold trade). For Lombard, Islam restarted the economy of the