The Bulgar Belgrad-Fehervar lying on the right side of the northwestern section of the Maros River was occupied by the Hungarians at the time of the Conquest. The grave of a Hungarian warrior buried with his horse in
side the fortification, as well as other similar graves around it, verify this.
The castle suddenly gained importance when, in the last third of the tenth century, the first Transylvanian gyula established his seat inside its walls.
On the evidence of the mortar mixed with brick-dust, a Byzantine plaster technique, the earliest Christian church, a rotunda, must have been built during the time of the gyulas. There is no evidence of burials around the church in this early period. The commoners were buried in the "pagan"
cemetery northwest of the castle up to the end of the tenth century. The Fehervar of the gyulas, most probably was not a very crowded town, be
cause in its northwestern district, once the site of the Bulgar settlement, there still existed a tenth century "pagan" cemetery inside the walls.
The Fehervar of the gyulas grew further in importance after King Stephen put an end to his uncle's rule of the province in 1003. Its walls of white stone were restored at the beginning of the eleventh century, so that they did not
require further reconstruction until the Mongol invasion. Not long ago, the remains of the first single-nave episcopal church were discovered. It had been built on the site of the old "pagan" cemetery across from the present cathedral during the reign of Stephen: at the time, its length of 19 to 20 m eters would have been considered quite substantial. After the m iddle of the eleventh century the area east of the church from around the Monetaria (Mint) to the walls, became densely built up, with houses even lining the road leading to the castle. The fact that the number of cem eteries around the castle had risen to three by the middle of the eleventh century was obvi
ously related to the increase in the population living around it. The most important of the three was the new cemetery lying to the northwest on top of an old Roman burial ground, one that was used from the time of Andrew I to that of Ladislas I. It was Ladislas I and Coloman who initiated the build
ing of the new, thirty-eight metre long, triple-nave episcopal church at the end of the century. The present cathedral stands on the foundations of that building. The old church of the time of Saint Stephen had been demolished at the time the new one was being built. The earliest example of Transylva
nian Romanesque sculpture, a Maiestas Domini relief on the semi-circular tympanum of the Ladislasean-Colom anean cathedral, has survived as an inside tympanum on the southern entrance of the present, early thirteenth century cathedral together with a few impressive capitals. The date and the H ungarian origin of the low relief is revealed by the figure of Christ sitting with his arms raised on a throne decorated with animals' heads. The com position follows the depiction of seated kings on the royal seals of Ladislas and Coloman. Inhabitants of the castle were obliged to bury their dead in the cemetery around the new cathedral from the twelfth century onward, even w hen that meant that the graves lay within the grounds of the old church! Other churches and cemeteries were built around the castle in the course of the twelfth and thirteenth centuries.
Archaeological evidence suggests that the large fortification com plex known as Dobokav£r came into existence at the beginning of the eleventh century at the earliest and during the course of the eleventh century gradu
ally developed into a great ispan's castle. Repeated destruction and siege necessitated its renovation and enlargement on several occasions (during the reign of Stephen, before 1030, and then in 1068 [?], and again in 1091).
Stone walls replaced the earth-and-timber elements by the beginning of thirteenth century. One of its churches was already standing by the first half of the eleventh century while the other was completed during the reign of Ladislas I. Both were rebuilt several times and had cemeteries around them. A church was also erected in the V&ralja (suburbium) in the twelfth century.
There is still no archaeological proof that Desv&r existed in the eleventh century, although the fragment of a western sword suggests that D6s and the surrounding area was occupied in the eleventh century. The early for
tress and cemetery have yet to be found. Due to later reconstruction, only modest remnants from Kukiill6var are known on the left bank of the Kis- Kukiill<5. The earth-and-timber fortification has quite an ordinary structure with a double moat outside and stone pavement inside, similar to that found
in O-Tordavar and Sajos3rv£r. An eleventh- or twelfth-century vessel re-
162
covered in the area of the fortress has been published. On 4 April, 1241 the castle was destroyed by the Mongols.
Much more instructive are the remaining three of Transylvania's "Seven C astles". All three are classic examples of the castle-to-city evolution pecu
liar to eastern Europe in that the fortifications built in the early years of the consolidation of the Hungarian state were in different locations from the mediaeval towns which, for various economic reasons, grew up in their vicinity, some nearer, some farther away from the castle itself, towns which only subsequently inherited the names of the castle nearby.
O-Tordav&r was built on a flat plateau above today's village of V&rfalva.
In size, shape and structure it is a typical early eleventh century ispdn’s castle. It has not been excavated inside, but the walls have been transected in several places. The cemetery in the suburbium used by the inhabitants of the castle at the turn of the tenth and the eleventh centuries, was excavated in 1912 and its finds published. The oldest sections of the cem etery still exhibit characteristics of commoners' burials, while younger parts reflect the burial rites of the castle's Christian population between the tim es of Stephen I and Ladislas I. The cemetery was discontinued during the reign of Ladislas and the population, was obliged to bury their dead around the new church after the castle was abandoned in the thirteenth century, as village dwellers. This new church was built below the castle during the rule of Ladislas I and is still used today by the Unitarian community. The town of 6-T o rd a was known to be the town of the salt miners who worked in the area of Tordaakna. Here, the earliest cemeteries were opened around the churches. (The eleventh- or twelfth-century cemetery of Tunderhegy inTorda had belonged to the neighbouring village of Szentmiklbs.) The northeastern neighbours of Torda are well known since 1176: the villages of Szind (Scinth) and Kopp^ny (Coppan).
The original Hunyadvar, was built on a promontory called Saint Peter Hill at the confluence of the Csem a and Zalasd rivers, 300 metres from the late mediaeval castle of Vajdahunyad erected on rocks. The castle which was partly ruined during its history has been recently excavated. It was 200 by 70 metres, oval, and a village belonged to it which was built below the hill. The cemetery below the castle, on the northern side of the road leading to Rakod, once used by its inhabitants, was partially excavated and pub
lished in the 1910s. People were buried here from the time of Stephen I until the end of the eleventh century. No later cemeteries from the area are known at present.
Coluswar — castrum Clus — was built 2100 metres to the west of the Rom an city and stood until 1241. It was originally established as an ispdn’s castle in this geographically ideal location: on a hill surrounded by branches of the Szamos River near to a ford and a cross-roads. The finds — for exam ple, coin of Stephen I and pendants from the late Conquest period — show that the earliest earth-and-timber fortification was built at the very begin
ning of the eleventh century. The walls and houses containing the coins of Solom on were burnt down at the time of the Pecheneg (Uz?) attack in 1068, and a higher rampart was subsequently constructed. The size and the struc
ture of the earthwork corresponds to that of the castles built for the ispans;
1 6 3 it is more than unlikely that it was just for the protection of some mansion