For four hundred years the Roman empire reached from the hills of Northumberland to the banks of the Euphrates, to encompass most of Europe and substantial bits of Africa and the Near East within a single state. In the fourth century, the publicists who served its rulers liked to stretch a point and proclaim the 'eternity' of Rome's empire – and in name, at least, Roman emperors would still be ruling in the West until late in the fifth. But already by the early fifth century events were unfolding that would strain the ingenuity of the deftest imperial spin doctor. In the Balkans, Gothic 'barbarians' occupied Roman territory and destroyed an imperial army; other 'barbarian' groupings soon followed, breaking through in the Rhineland to settle in large parts of Gaul, Spain and Roman Africa; and in the summer of 410 a Gothic war party delivered a shocking blow to morale across the empire by occupying Rome itself and plundering it at leisure before departing. The attack was virtually unopposed: the reigning emperor, by all accounts a feeble puppet manipulated by competing courtiers, stayed holed up at Ravenna; when it was announced to him that Rome had perished, the story goes, he imagined at first that his favourite pet bird, a cockerel named 'Rome', had died – 'But he was just eating from my hand!' he exclaimed – and seemed much relieved when the truth of the matter became clear. 

The story is doubtless a fiction, but it points to a deep question that generations of historians of Rome have worried over since Edward Gibbon completed his classic Decline and Fall in 1788. Was the empire's collapse due simply to a brute external factor, an unforeseeable and irresistible sequence of