The dungeon at Montbazon just south of Tours is claimed to be what remains of the oldest castle in France.
Earth-and-timber fortifications proved entirely inadequate to protect western europeans from the Viking and Hungarian incursions of the ninth and tenth centuries. Although motte-and-bailey castles were successfully used by William of Normandy to conquer and subdue England in the late eleventh century, construction of stone castles in London and Colchester seems to indicate that he recognized that these earth-and-timber structures were but temporary defenses for his kingdom. More permanent fortifications, made of masonry, were needed to defend the lands of Europe and their inhabitants from foreign attacks and domestic rebellions.
Yet, despite the obvious need for stone castles and the fact that by the end of the twelfth century they dotted the landscape of every medieval principality, one of the most difficult dates to determine in the history of medieval military technology is when they began to be built in Europe. Unless the historian of fortifications is willing to accept the early-tenth-century Italian charters for stone castles, which are without archaeological confirmation, there are neither written sources nor archaeological evidence to prove the existence of a stone castle before the late tenth century.
This was a period of instability and lack of strong central government throughout all of Europe, a condition that seems to have led to experimentation in fortification construction, experimentation that probably produced stone castles. As in the case of the motte-and-bailey castle, the late-tenth- and early-eleventh-century count of Anjou, Fulk Nerra, may be credited as the initiator of this new defensive construction.
Fulk was faced with almost continual conflict throughout his reign, from both outside and inside his territory, and he countered by building fortifications. Fulk fortified all of his borders with castles, at least 30 major strongholds in all. Frequently, these became the targets of his enemies, but rarely did they fall. Moreover, Fulk’s own offensive forces, quartered in these fortifications, were able to make extensive attacks into his enemies’ lands, ultimately giving him much more territory at the expense of his neighbors.
Most of Fulk Nerra’s fortifications were constructed using earth and timber, and some of them were motte-and-bailey castles. However, at least two, at Langeais and at Montbazon, seemed to have been constructed in stone. Both of these castles played important roles in Fulk Nerra’s castle-building strategy. Langeais was built on the Loire River, and together with the fortress at Amboise, which had been constructed upstream by Fulk’s father, Geoffrey Greymantle, controlled the entire region. Montbazon was built to the east of Langeais, on the Indre River, and provided defense for a southern route of communication with the fortresses of Amboise, Langeais, and Loches.
It has been suggested that the stone ruins at both sites date from Fulk’s reign. The ruins of Langeais reveal a tower keep (also called a donjon) originally 15 to 16 meters high with an outer rectangular perimeter measuring 17.5 by 10 meters. The walls were built on a shallow crushed-rock foundation that in places is only 70 centimeters below the surface. The walls were made of ashlar, and measured between 1.2 and 1.7 meters thick. The fill between the ashlar frame was composed of limestone and mortar. It was built not on a man-made motte, but on a natural rise of relatively low height. No wooden elements survive, but it is apparent that the tower was divided by wooden floors into three levels and that access was gained by a wooden door and staircase.
That these are the ruins of Fulk Nerra’s castles is disputed. Fulk’s castle at Langeais is first mentioned in documents dating to 993/994, and three historians, J. F. Verbruggen, Marcel Deyres, and Gabriel Fournier, claim that what is mentioned there was nothing more than a wooden fortress. The stone keep, in ruins today, did not replace it until the late eleventh century and then was not built as a military structure but as a domicile. Disputing their conclusions are Pierre Héliot, J. F. Fino, Bernard S. Bachrach, and Philippe Contamine, who contend that the stone tower at Langeais was indeed Fulk Nerra’s late-tenth-century fortress and that it was constructed for defensive purposes. As proof, they claim that the extensive and unsuccessful siege operation that Odo I, count of Blois, undertook there in 996 could only have occurred if the structure was a stone fortress.
The dating of the stone fortress at Montbazon to Fulk’s reign is also disputed. Although built in a similar rectangular pattern to Langeais, the keep at Montbazon was much larger. It measured 19.65 by 13.75 meters at the base and stood between 28 and 30 meters high. The walls were also thicker than at Langeais, measuring between 2 and 2.4 meters. There may also have been an exterior (or curtain) wall with a tower built on its northeast corner.
Fulk Nerra’s Montbazon castle dates to the early eleventh century, as a written reference of 1005/6 attests to its completion by that time. Again it is Héliot’s and Bachrach’s contentions that the stone castle that now lies in ruin there today is Fulk Nerra’s, while Deyres claims that it should be dated after 1050, although even he admits that stylistically it could have been built earlier.
The dispute is perhaps inconclusive, but what may ultimately shift the argument in the favor of those who wish to date the stone castles of Langeais and Montbazon to Fulk Nerra’s reign is the large number of stone ruins on sites of other Fulk Nerra castles. Most of these have not been examined as well as the keeps at Langeais and Montbazon, but at least some appear to date from the late tenth or early eleventh century and may have actually been built by Fulk Nerra himself or by his retainers. These include the castles at Loudun, Argenton-sur-Creuse, Melle, and Brosse. Other unexamined stone castles in lands adjacent to Fulk Nerra’s also seem to date from the early eleventh century and may have been influenced by the Angevin count’s building strategy (or perhaps by the stone castles on his borders). Odo II of Blois, William II of Angoulême, Odo of Déols, Boso the Old of La Marche, and even King Robert II of France may have built stone castles. Dating is still imprecise, but archaeological evidence supports these as early stone constructions, as does the fact that many of these fortifications successfully withstood lengthy and harsh sieges during the late tenth and early eleventh centuries, sieges that could only have been endured if the castles were built of stone.
Another possible site for the origin of medieval European stone castles is Catalonia. Since the invasion of southwestern Francia by Muslim armies across the Pyrenees Mountains in the third decade of the eighth century, the area had been contested between Frankish and Muslim Spanish armies. Charles Martel halted the initial Muslim invasion at the battle of Poitiers in 732, but a large number of Iberian soldiers and settlers remained in the region until, at the end of the eighth century, Charlemagne pushed them below the Pyrenees and captured Barcelona. To protect the lands of Provence and Aquitaine from further Muslim conquests, Charlemagne established a royal province along the Spanish March, although it would not acquire the name Catalonia until the twelfth century. After 809, when offensive action by the Franks against the Muslims ceased, this province became a defensive “buffer zone” between Carolingian France and Muslim Spain.
By the middle of the tenth century the situation in Catalonia had changed dramatically. The counts of the province, who had initially been drawn from the Carolingian nobility and military and were tied directly to the Carolingian throne, had gained the same freedoms from central rule that other Frankish nobles enjoyed. Perhaps more importantly, the various Muslim principalities in Iberia began to fight what would become the first of many wars against each other, wars that quickly destroyed the unity of the Muslim Umayyad dynasty in the Iberian peninsula. This allowed the Christian Visigoths in the northeast to establish their own kingdom, León, and to build up their own military strength. By 911 the Christians had conquered one-fifth of Iberia, and although the more populous and powerful southern portions remained in Muslim hands, the call for Reconquista (reconquest) spread throughout the rest of Europe.
Catalonia was to play a major role both offensively and defensively in the early years of the Reconquista. To do so, the counts of the region first had to increase their military strength and then had to build permanent barracks to hold these soldiers. By the end of the tenth century, a large number of castles were built throughout Catalonia. They performed a dual purpose: they provided a defensive formation against any Muslim attacks, and by harboring soldiers, they also provided an offensive threat to the Muslim principalities on the other side of the Pyrenees.
It is not known exactly when these castles began to be built of stone. It seems that initially they were built of earth and wood, although they were clearly not motte-and-bailey castles. By the beginning of the eleventh century (or maybe even earlier), most of these fortifications were built or rebuilt in stone. They were keeps, similar, it appears, in purpose to the castles of Fulk Nerra, but rather than being rectangular in shape, most were circular. They were also massive structures, and in fact dwarfed most castles built at around the same time in France. For example, the tower of Vallferosa was 38 meters in circumference and more than 30 meters high. Most also contained at least three and sometimes as many as five stories with round and elliptical windows.
Whether Fulk Nerra (or one of his neighbors) or the counts of Catalonia first originated the stone castle is not important. Of far greater historical significance is that by the end of the eleventh century stone castles had begun to be built everywhere in Europe. In France, the increase in the number of fortifications was incredible. In Poitou, for example, the number of castles increased from 3 at the beginning of the eleventh century to 39 by its end, and in Touraine, the increase was from 9 to 26 during the same period. In other regions, even more stone castles were built in even less time. For example, in Auvergne the number of castles grew from 8 to 34 between 1000 and 1050, and in Maine the number increased from 11 to 62 between 1050 and 1100.
By the end of the twelfth century the total number of castles that had been or were being built reached a very impressive total. In France, during the reign of Philip Augustus (1180–1223), the king alone held more than 100 castles, with his nobles holding several hundred more, and in England, at the beginning of the reign of Henry II, in 1154, there were an estimated 274 castles held by the king and his barons. While this number cannot compare with the 500–1,000 motte-and-bailey castles built during the reign of William the Conqueror, considering the more sophisticated and elaborate building technology, the need for materials not easily accessible, and the greater expense required to construct stone castles, the number is quite impressive.