This contribution provides an overview of the development of frontiers in Europe from the dissolution of the Western Roman Empire (5th c. CE) to the Carolingian Empire (9th century), focusing on the boundaries between large-scale polities. After a brief introduction reflecting on the contemporary contexts of our interest in frontiers, the paper will first discuss the concepts of the frontier current in the period. What was the terminology and language of the frontier in late antique and early medieval Latin? In what contexts do frontiers appear in historiography, legal texts and other sources? Do these texts indicate linear, spatial, local, symbolical or other concepts of frontiers, and in what ways did they change? Which kinds of practices are connected with frontiers? This includes marking frontiers, guarding them, controlling or blocking movements across them, negotiating or defending them, or emphasising them for symbolical or political purposes.
The main part of the paper deals with the practices connected with frontiers, and in particular, the building and use of walls, dykes and frontier defences in the Early Middle Ages. Were such buildings actually used for defence? Who built such structures in early medieval Europe, and, a question as interesting, who did not? Some cases will be briefly addressed. The first is the Roman limes as a template for later frontiers, its abandonment and its continued uses. Remarkably, the elaborate defensive structures that the Romans built along Rhine and Danube were hardly used in later periods, although some of them are still standing today. Rhine and Danube rarely served as frontiers at all since the end of the Western Roman Empire. One case in which Roman defence architecture continued to be used are the barriers on the southern ends of the pass roads across the Alps protecting a core area of the Empire, the Tractus Italiae circa Alpes, and its eastern section, the Claustra Alpium Iuliarum with their elaborate defensive walls. The Tractus is particularly interesting because we have relatively detailed evidence for its continuing uses, in particular, in the 8th century, when Italian legal texts from the Longobard and Carolingian period prescribe the modalities of its control.
Following on from this case, the second example to be addressed will be Frankish frontiers in the Carolingian period. A few general observations can be offered here. First, external frontiers generally receive little focused attention in Carolingian sources. Only sometimes they are politicized and come in the focus of propaganda. In the expansion period, before c. 800 CE, this mostly happens in the context of preparations for an attack on the respective neighbours. Second, internal frontiers began to matter in the course of the divisions between Carolingian kings, and they were described in quite detailed ways in several division plans. Much more care was taken to delineate these boundaries than those toward external neighbours. Third, the way in which the Frankish kings and emperors organised their frontiers diverged greatly between different regions. Fourth, relatively little effort was devoted to defensive architecture along frontiers, even when, in the course of the ninth century, the Carolingian kingdoms came under attack from the outside.
One of the most elaborate pieces of frontier defence architecture erected in the Carolingian period was the so-called Danewerk. The ‘Danewerk’, a dyke stretching from the Baltic Sea near Schleswig/Haithabu to the North Sea, was, according to archaeological evidence, built in several stages between the 8th and 11th centuries. The Royal Frankish Annals claim it was set up by the Danish king Godofred in response to a threatening Frankish and Saxon attack in 808. That the neighbours of the Franks should set up an ambitious defensive structure is remarkable in comparison with Offa’s Dyke constructed around the same time, reputedly by King Offa of Mercia, which was an expansive power at the time.
The third case is the existence of a good number of dykes and ramparts attested in the steppes of Eastern Europe, known mainly from the archaeological record. It may seem surprising that such structures should be set up by steppe warriors. A long system of walls and ditches, the so-called Csörsz Dyke (or Devil’s Dyke), is found in the east of modern Hungary. It had long been attributed to the Sarmatians, but recent finds point to the Avar period. These dykes do not follow the frontier, but were constructed more inland. Similar dykes are known in the Dobruja, south of the Danube delta, in what used to be the Bulgar realm. Their layout is somehow enigmatic, and their function is unclear. Another long Bulgar dyke was built in a rectangular shape, stretching for many miles around the residence of the khans at Pliska and the land around it.
What was the function of all these walls and other border fortifications? In many cases, they were not designed to withstand massive attacks. Even from the Roman period, there are relatively few examples in which border fortifications were actually defended against a massive attack, and that is surely the case in the early medieval period. Longobards occasionally tried to defend the clusae, the barriers on the Alpine pass roads, against invading Franks, but with little success. We tend to ignore, however, the frequent small-scale raids in frontier provinces, which mostly remain under the radar of the sources. Walls and barriers could provide some protection against raiding parties, and also delay operations on horseback. They also channelled ordinary traffic and thus allowed more efficient control of movements between kingdoms. That was relevant for steppe realms as well, which had an interest to curb the emigration of warriors to their competitors.
Early medieval polities on the territory of the former Western Roman Empire imposed rather strict regulations on mobility; long-distance travellers within and between the countries needed letters by an authority of their place of origin that acknowledged their need to travel, and permits by the king or his functionaries to move through the country. A number of interests converged in this policy: preventing the flight of slaves or dependent workers; catching fugitive murderers or thieves; directing and levying merchant traffic; controlling the movements of monks and clerics who were normally bound to the stability of place; keeping pilgrims on the prescribed routes; curb the activities of spies and informants; and monitor possible oppositional activities. We may doubt that such controls were actually efficient. Yet frontier post or barriers, where they existed, provided good opportunities to control mobility of all sorts.