Arrabona - Múzeumi közlemények 13. (Győr, 1971)

Gabler D.: Research in the Canabe of Arrabona

RESEARCH IN THE CANABAE OF ARRABONA The author reports on the excavations executed in the years I960—1969 in the Széchenyi Square, Győr, aimed among others at the uncovering of an important part of the canabae, belonging to the Roman camp of Arrabona. The significance of this research may be understood if we consider that there was no large-scale excavation with the purpose of unearthing the remains of the Roman settlement on city territory so far. After an enumeration of the Roman finds which have come to light in and near the Széchenyi Square in the course of non-archaeological works since the last century (their sites are presented in Fig. 1), the author describes the work done by the small research team (D. Gabler —• P. Tomka — J. Topái — M. Albeker — R Pusztai, aided by A. Uzsoki). A surface of the size of 23 X 10 metres has been investigated. The archaeological material uncovered in the course of the excavation, estimated to 15 to 20 thousand finds, needs a considerable time to be dealt with, so this preliminary report is restricted to the description and the stratigraphical connexions of the objects. Research revealed six periods of construction of the settlement, whereas its territory was used for burial in the seventh. Below the only early medieval layer to be ascertained, only finds of the Roman period have come to light to the depth of 450 to 490 cm. (The thickness of the Roman filling in may be summed up as 3 metres.) Prehistoric objects were not found at all. On the whole there were closed Roman strata everywhere, except for one major disturbance by, the digging of a channel in the eighteenth century. The author presents the results of the excavations in a historical-chronological sequence. During the uncovering it was impossible to investigate the phenomena of the layers of the first period (Fig. 2) in their entirety, as only stepped surfaces coud be dug out safely in a relatively deep site, in the neighbourhood of subsoil water. The subsoil was reached in the areas marked by a dotted line in Fig. 2. At the upper borderline of the mud layer above the virgin soil wooden beams and among them wood-fibres made their appearance. Two layers were discovered in section 10 which may be characterized by the remains of wood. A large part of the edifices of the first century vicus may have been constructed of wood alone, but the excavations did not inform us unequivocally on their details; so they may be reconstructed on the pattern of the buildings of Cambodunum. On one place (section 1), near the rests of the wooden huts, a layer with charcoal and pebbles was found in the same depth, above the mud layer. The dating of these phenomena was helped by the terra sigillata material of the Po region at the highest degree. None of the types presented in Fig. 6 is earlier than the Claudian age, nay some of them might be dated even to the Flavian period; single pieces become typical only in the middle of the century. Therefore we may date the earliest constructions of the age of Claudius or of the Flavians, respectively. The cessation of this early wooden building may be attributed to the frequent inundations, which induced the later builders to raise their edifices at a level more than one metre higher than the first. The constructions of the second period are attested by a pavement and the remains of timbers (Figs 8—9), reaching under the later walls of houses in part. Their chronological situation is defined by the wooden house below (Claudian

— Flavian period) on one hand and the stone-based edifice on their top on the other (probably the age of Hadrian) ; thus the short existence of the wooden house is restricted to the last decades of the first century or the first decades of the second century, respectively. The large-size, villa-like building was raised in the third period (Fig. 11). The stretches of the wall, left empty in the ground-plan, bear the marks of a later reconstruction: their lower parts are made of flat, horizontally laid sandstone pieces, while their upper parts were built with the opus spicatum technique (Fig. 12). One cannot observe this dualism in the remaining stretches of the wall (Fig. 13). The foundations were built of pebbles and stone pieces of irregular shape, without any mortar. The foundation trench is shallow, the house was erected almost on the surface of the earth originally. In section 2 the walls, orientated N

—S, or E

—W respectively, were made of sun-dried bricks on a stone foundation. The ground-plan of the edifice seems to have remained unchanged even in the later periods; it may have existed for several centuries, as it is proved by the renewal of floor levels, the strata 52