SAN GIMIGNANO, WITH ITS FOREST OF PENCIL-THIN Romanesque towers dominating a fortified hilltop, is a famous emblem of Italy's medieval past, but it is only one of thousands of hilltop villages that extend from the Alps to Sicily. The question often asked is why on earth did medieval Italians choose to live on the top of a hill, however spectacular the view, when the Romans had lived in the valleys?
Over the past twenty-five years the study of medieval archaeology has prospered in Italy as, backed by a campaign led by the journal Archeologia Medievale, numerous towns and regions up and down the peninsula have raised funds to uncover evidence of their medieval past. A planned extension to the Uffizi Gallery in Florence by the Japanese architect Arata Isozaki was recently mothballed by the Minister of Culture not for financial reasons but, as he announced in a press release, because it jeopardized important medieval archaeological investigations. The controversy created by the Uffizi decision, however, pales by comparison with the ceaseless stream of conferences and books devoted to pinpointing the origins of Italy's hilltop villages.
The traditional historical view, summarized by the French historian Pierre Toubert in 1974 and based on interpretation of the medieval, mostly monastic, sources, was that hilltop villages evolved, along with medieval Italian life, out of the introduction of a monetary economy by the Carolingians in the ninth century. This economic transformation led, so Toubert argued, to the process of incastellamento--the foundation of castelli, fortified hilltop villages--in the ninth to the tenth centuries, with new villages being established over the course of the next two centuries.
For years Toubert's book was the bible of medieval archaeologists. Nevertheless, it left the period between the collapse of the Roman settlement system in the sixth century and the ninth-century recovery in limbo. What happened to the villa-owners and their tenants in the later Roman period? Did they continue to occupy the old estates on the lowlands or did they move up onto the hills as soon as the Roman empire dissolved? To answer these questions archaeologists needed type fossils--objects, principally pottery, that would make it possible to distinguish sixth- to ninth-century levels in the ruins of Roman sites or in the earliest levels on hilltops beneath later medieval buildings. So throughout the 1980s medieval archaeologists were occupied with identifying pottery types, surveying Roman farms, and driving trenches deep into the earliest occupation levels on hilltops.
The breakthrough came--just as it did in Anglo-Saxon and Frankish archaeology--when the piecemeal digging of small trenches was abandoned for large open-area excavations. One of the first of these excavations was at Montarrenti, a prominent hilltop west of Siena where two elegant towers survive to this day. Here post-built structures of the seventh to tenth centuries were found beneath the late medieval levels on the top of the hill, and, strikingly, beneath the ring of peasant dwellings running around the bottom. The results showing origins as early as the seventh century were presented to a congress in Siena in 1989, but they elicited dismay and scepticism from most of the assembled historians. Plainly, they concluded, the archaeologists had mistaken the dating of the post-holes, which, in any case, were terribly difficult to interpret.
But the authoritative figure of Riccardo Francovich of Siena University was not to be dismissed lightly. With his pupil, Marco Valenti, Francovich launched a campaign of large-scale excavations of hilltop sites all over Tuscany, with compelling results. At Poggibonsi, on an exposed hilltop overlooking the unattractive modern town, they found remains of a cluster of later Roman dwellings. In time, these were transformed first into a village of post-built structures rather like …