Mohammed and Charlemagne Revisited: The History of a Controversy (23 page)

S
cholars may argue about the fine details of chronology, but on one thing at least all (including Hodges and Whitehouse and Pirenne’s opponents in general) can agree: archaeology shows that the great transition from classical to medieval occurred in the first few decades of the seventh century. By the 640s virtually all trade between Europe and the Near East (and North Africa) had come to a definitive end. Luxuries which had been common in the West until that time, such as spices and various forms of high-quality ceramics, disappear never to come back. The supply of papyrus comes to an end, and Europeans are compelled to employ parchment for even basic day to day record-keeping. The great cities built by the Romans fall into decay, and in the countryside the scattered and undefended lowland farming settlements that were characteristic of the Roman epoch begin to disappear; to be replaced by secure and easily-defended hill-top settlements – the first medieval castles. Accompanying the abandonment of the lowland settlements, many previously productive regions revert to wasteland: Drained wetlands revert to marsh, and agricultural terraces and ditches are washed away. Harbors begin to silt up and a layer of subsoil forms in valley floors, covering many of the Roman age towns and villages.

One of the most striking of the above developments is the retreat to defended hill-tops. This process, known in Italy as
incastellamento
(“encastling”), marks perhaps the most visible and easily-recognized manifestation of the new and medieval civilization. The Middle Ages was, above all, the age of feudalism and castles. Hodges and Whitehouse mention the movement towards castle-building but offer no convincing explanation of it. “The reason for this shift,” they note, “are many and may never be accurately determined.”
[1]
They point out that in Italy at least, “the shift from open dispersed sites to fortified upland settlements is only explained as a defense against Lombardic invaders [late sixth century].”
[2]
However, “this may be a satisfactory explanation for the change on the edge of the Roman Campagna, but its wider implications have to be assessed.”
[3]
In other words, a process that is observed throughout the whole of southern Europe can hardly be satisfactorily explained by the settlement in northern Italy of a single barbarian tribe. Hodges and Whitehouse suggest that, “Increased taxation by the Byzantine government after Justinian’s reconquest of Italy might account for a phase of rural depopulation in the sixth century. Similarly, we cannot ignore the impact of the Great Plague of 542 which ravaged Byzantium and Europe.”
[4]

But there had been plagues before, and devastating wars. None of these caused the total and permanent abandonment of the lowlands and the retreat of populations to defended hilltop settlements – essentially a return to Iron Age conditions. And whilst castle-building might be explained in Italy by Lombard raids (there is in fact good evidence that the Byzantine and Lombard invasions in the mid-sixth century caused much devastation in Italy), what is the explanation elsewhere, where castle-building also appears in the early seventh century?

Interestingly, in the 1930s Alfons Dopsch had used castle-building as an argument against Pirenne. Without mentioning southern Europe – from which at that time little archaeological data was available – Dopsch noted that since castle-building (in the north) commenced in the middle of the tenth century, medievalization probably had more to do with the Viking onslaught than the Muslim. It is in fact true that castle-building commences in northern Europe in the tenth century – almost precisely three centuries after it begins in southern Europe. Indeed, the three centuries’ discrepancy appears as soon as we leave the Mediterranean coastlands. Thus for example the castles which guard the Pyrenean passes, such as those at Montségur and Lourdes, only a short distance from the Mediterranean, were built in the tenth century, apparently to guard against Muslim incursions from Spain. So, on the Mediterranean shoreline, castles are built in the mid-seventh century to guard against Arab raids, whilst less than 150 kilometers away castles do not appear for another three centuries – and when they do appear they are again to guard against Arab raids. We are therefore presented with a strange dichotomy: In Europe castle-building appears to begin in the seventh century, but then goes into a kind of suspension for three centuries, when it again appears in the tenth century. Furthermore, the defended hilltop settlements, which in Mediterranean Europe form the basis of the seventh-century castles, are precisely the locations of the tenth/eleventh century castles, which appear, for all the world, to be normal and continuous developments from the seventh century fortified settlements.

Here we have yet another instance of that puzzling three-century gap in Dark Age history and archaeology.

Leaving aside the Dark Age hiatus for a moment and returning to the question of the seventh century castle-building, we should note that for Pirenne and his modern acolytes the abandonment of the classical lowland villages and villas and the retreat to defended hilltops has a simple and straightforward explanation: the appearance along the Mediterranean coastlands of Spain, southern France, Italy, and Greece, of fleets of Saracen pirates and slave-traders.

* * *

Clearly related to the phenomenon of castle-building is the appearance, throughout the Mediterranean world, of a layer of subsoil which overlies late Roman sites. This is the stratum referred to by Thomas F. Glick in Spain, who however seemed to be unaware of its occurrence throughout the Mediterranean. According to Claudio Vita-Finzi, who named it the Younger Fill, this deposit is an almost universal feature of the river valleys of the Mediterranean basin in the period roughly corresponding to the final decline of classical cultures, around the sixth to eighth centuries.
[5]
The origin of the Younger Fill “is the subject of considerable debate, and some scholars argue that it is simply the last stage in an intermittent process which began some two thousand years earlier in the Middle Bronze Age.” However, “Vita-Finzi demonstrated that a dramatic geomorphological change took place at the end of classical antiquity.”
[6]

There are two main theories about the formation of the Younger Fill. “The first, proposed by Vita-Finzi, is that it was formed as a result of climatic deterioration, and that it provides us therefore with information on a major, but hitherto unsuspected, cause of the collapse of the Roman Empire.”
[7]
However, “no contemporary chroniclers reported marked changes in climate, and consequently it is difficult to accept this explanation without further evidence.”
[8]
And since weather conditions in the Mediterranean, as well as the crops the region produces, seem to be identical to those of ancient times, this too would apparently rule out the possibility of any dramatic climate change in the centuries since the end of the Roman Empire. The alternative theory, the one which Hodges and Whitehouse subscribe to, “is that the Younger Fill was formed as a direct result of the collapse of the classical agricultural system.” “Failure to repair terraces as the mass-market for olive oil and wine declined,” they say, “led to erosion as previously revetted soils were washed away. It is a familiar process; one sees it in many parts of the Mediterranean today, as farmers plough deeply into terraced hillsides, creating furrows at right angles to valley bottoms, down which the torrential winter rains carry soil at an alarming rate. The implications of this process in Late Antiquity are considerable. It would have led to the degradation of the hill slopes and to marked morphological changes in valleys and estuaries, with implications not only for farming but also for road networks, harbours and towns.”
[9]

Clearly the date assigned to the formation of the Younger Fill is crucial to the whole debate about classical civilization’s fate. The Younger Fill is plainly the geological signature of the end of the classical system of agriculture. Perhaps predictably, Hodges and Whitehouse would like to place its formation in the latter sixth century, several decades before the rise of Islam. They consider the possibility that it may have something to do with the great plague of 542, but they note that at Olympia in Greece, which was covered by the Fill, coins of 565 and 575 were found, “indicating that the city’s demise happened a little later.”
[10]

It should be noted at this point that the plague of Justinian’s time is frequently cited as a suspect in the demise of Graeco-Roman civilization – especially by those who reject the idea that it could have been caused by the arrival of Islam. Yet the evidence from Olympia, by itself, indicates that life continued as normal after the plague, and that it must have been something of an altogether greater magnitude, in the decades after 575, that finally terminated the whole system. But what was this event, and when did it occur?

It so happens that in the East at least a very precise date can be given to the ending of classical agriculture and the formation of the Younger Fill. We have seen that the cities of Asia Minor, which have been extensively studied, show a thriving culture right up to the start of the Persian War in 614. These metropolises, with their enormous populations, could not have existed if the classical agricultural system had been in decay. Their very existence, with their numerous populations, presupposes a thriving agriculture producing very large food surpluses. The destruction of these centers in the years following 614 was final, and none of them recovered. It was then too, in the immediate aftermath of these events, that the Younger Fill appears and harbors begin to silt up.

In short, in the Eastern Mediterranean, which formed the very epicenter of antique civilization, the Younger Fill, and with it the abandonment of the classical system of agriculture, occurred in the years after 614, probably the two or three decades after. And if that is the date in the East, we may be fairly sure that it was the same in the West. Thus it hardly seems open to question that throughout the Mediterranean the retreat to defended hilltops and the breakdown of Roman agriculture occurred in the disturbed years which commenced in the second or perhaps third decade of the seventh century. But this then prompts the question: Who or what caused these momentous changes in the West? In the East, we might suppose they were precipitated by the Persian war; but what about the regions never touched by the Persians – everything west of Egypt? Who caused the destruction of Roman agriculture and the retreat to the hilltops in those lands?

Almost immediately after the conquest of Egypt, Arab raiders and pirates began to scour the Mediterranean. We know for certain that they quickly took to the sea, for they sent a vast fleet to besiege Constantinople in 674. Such large-scale operations were supplemented by hundreds, indeed thousands, of smaller attacks. These unleashed a wave of banditry and lawlessness which may well have been without precedent in the history of the Middle Sea. It is true that in recent years writers such Hodges and Whitehouse have tried to suggest that Arab piracy in the region did not begin until the middle of the ninth century; but their grounds for doing so are spurious.
[11]
As shall be demonstrated in Chapter 13, piracy and slave-raiding were activities fully sanctioned by Islamic law, and have always formed a central feature of Muslim interaction with the non-Islamic world. Pirate raids, often carried out in conjunction with large scale military operations, are recorded from the middle of the seventh century; and continued to cause immense problems in the Mediterranean
until the start of the nineteenth century
. We know that by the fifth decade of the seventh century Arab attacks on Sicily and southern Italy were incessant. A series of assaults on Sicily in 652, 667 and 720 are recorded; whilst Syracuse was conquered for the first time temporarily in 708. Sardinia was Islamicized in several stages beginning in 711, the very year of the Islamic conquest of Spain. The Italian island of Pantelleria was conquered by the Arabs in 700, and was attacked again a century later, when the Arabs sold the monks they captured into slavery in Spain.
[12]

As we might expect, further to the east the Arabs were even more active and at an earlier date. The whole of the Levant was scoured by Arab fleets from the 640s onwards, and the very centre of the Eastern Empire, Constantinople, was not immune from attack. An Arab army, led by Muawiyah I, laid siege to the city between 674-678. Unable to breach the Theodosian Walls, the Muslims blockaded the metropolis along the Bosporus, but their fleet was eventually destroyed by the famous “Greek Fire” of Kallinikos (Callinicus) the Syrian. Although this was a decisive defeat, within just over half a century the Arabs were back. In 718 an 80,000-strong army led by Maslama, the brother of Caliph Suleiman, crossed the Bosporus from Anatolia to besiege the capital of the Eastern Empire by land, while a massive fleet of Arab war galleys commanded by another Suleiman, estimated to initially number 1,800 ships, sailed into the Sea of Marmara to the south of the city. After some desperate fighting, and the use once again by the defenders of “Greek Fire,” this onslaught was also repulsed.

It has to be remembered that, from the years following the decline and abandonment of Europe’s cities – from the 620s and 630s onwards – only a very fragmentary record of events has survived; and the incidents recounted above can only have represented a tiny fraction of the true total: they were recorded precisely because of their scale and importance. Lesser raids, almost certainly at even earlier dates, involving shiploads of pirates and slavers, must have occurred in their thousands. This is certainly the impression gained by contemporary accounts.

The threat posed by Saracen pirates, who often raided far inland, fully explains the abandonment of the Roman fields with their terraces and irrigation ditches, and the retreat of whole populations to hilltop strongholds. Populations along the Mediterranean coasts of Spain, France, Italy, and Greece, were to become all too familiar with these dangers over the centuries, and with time large areas of the coastlands became uninhabited and uninhabitable. The impact of Islamic piracy on the Mediterranean is a question that has never been fully understood or appreciated, in the English-speaking world at least. For a thousand years the Middle Sea, previously one of the world’s great economic highways, was reduced to a hunting-ground for slavers. The cultures of all these regions were profoundly affected by this phenomenon, as were the policies and actions of kings and popes.