architechture / investigation

Olga Plamenytska, Eugenia Plamenytska

Kamyanets Podilsky, A Town at the Periphery
of the Roman Empire:

Urban Order and Fortifications


The influence of the Roman Empire on the urban development of Eastern Europe throughout the first millennium A.D. in general still remains insufficiently explored. This pertains especially to Rus', the early medieval principality centered around Kyiv that encompassed much of present-day Ukraine. Not infrequently, therefore, the origins of Rus' and its fully developed architectural traditions are depicted solely (and mistakenly from our point of view) as the product of Byzantine cultural heritage. The level of technical and aesthetic sophistication which was attained in architecture during this period of Ukrainian history could not have been achieved through an almost wholesale introduction of Byzantine culture alone, but had to presuppose the existence of the local traditions of construction, a local practice in building urban complexes, religious monuments, fortifications and the like.

Accordingly, architectural and urban construction sites, whose origins can be traced to a time before the emergence of medieval Rus' are bound to be of great interest and importance. One of these ancient sites is the urban complex of the Old Town of Kamyanets Podilsky. According to historic sources, the site can be dated from the period of Daco-Roman wars (Trajanic wars) and can be identified as one of the Dacian towns noted in the "Geography" of Ptolemy (2nd cent. A. D.). Kamyanets can be identified either as Klepidava or as Methonium, one of the five towns on the left bank of the Dniester river belonging to Dacia and mentioned by Ptolemy (fig. 1). It can also be identified as one of the abandoned Roman fortresses of the Middle Dniester basin described by Constantine Porphyrogenetos in the mid - tenth century. Historians of the 2nd half of the seventeenth century again repeat these foundation stories. These authors of the seventeenth through the nineteenth centuries, such as Cellarius 1659, Cyaneae 1687, Potocki 1805, Marczynski 1820, Przezdziecki 1841,Antoni 1880,CeMenoB 1843,UIeBHM 1855, 1862, CiuiHCbKHH 1927 considered that the Roman legionaries who conquered Dacia during the Trajanic wars, had crossed the Dniester and had founded a line of military settlements in Podillya (present-day south-western Ukraine). The remnants of the so-called Trajanic rampart lines, numerous hoards of Roman coins, (fig. 2) and artifacts of the military Roman life found in Podillya support these references to Roman presence on the left bank of the Dniester as well as the Daco-Roman origin of the town of Kamyanets. Nevertheless, the majority of twentieth century historians, such as I.B����yp, M.������, C.Tpy��������, A.3a��������, I.����ara still reject this evidence as groundless. As a result, important excavated data from Podillya of the first centuries A.D. has not been interpreted adequately and fully.

During more than 30 years (1964 - 1999) we have been investigating both the fortifications and urban complexes of Kamyanets Podilsky. The results of our investigations have allowed us to propose a thorough revision of Kamyanets history. Rather than accepting that the town dates from the Rus' period of the eleventh through the thirteenth centuries, we now claim that a number of defensive structures within the Old Castle limits (fig. 16, 29, 31, 41, 64, 65) and the Castle Bridge itself (fig. 47 - 49, 61) can be dated from the first centuries A.D. Since these defensive structures were connected with the Old Town island by the Castle Bridge it follows logically that there was also a settlement on the island at the beginning of the first millennium A.D. The existence of the later regular Rus'-Polish market located at the center of the island additionally supports the claim of such a settlement. Metrological investigations of the structure of the market allow for a hypothesis which would place the old market-place on top of the 300 m^2 site of the original Roman camp (castrum) (fig. 7,15).

Our architectural and archaeological investigations have demonstrated that the medieval fortifications on the castle mound can be subdivided into three phases. The first dating back to the 2nd - 3rd centuries A.D, is represented by ramparts of two (outer and inner) fortification lines. The outer one (600 m long) enclosed the lower terraces and consisted of small towers connecting to the defensive walls (fig. 16,37,41 - 44,65). The inner line, whose characteristics have not been fully studied, extended across the middle and upper terraces of the castle peninsula. It consisted of ramparts with three towers: a rectangular, a trapezoidal and an oval one (fig. 16,17,29,31,65). They were constructed according to the standard Roman metrology: the measurements of towers in plan are 5.0 m (17 Roman/ Greek ft), 7.4 m (25 ft), 8m (27ft) and the diameter of the oval tower is 6.0 - 7.2 m (20 - 22 ft), the wall thickness is 1.5 - 1.56 - 1.62 m. The towers are typologically similar to the towers of Roman fortresses in Rumania, Bulgaria and the Northern Black Sea coast (fig. 66-72). By the eleventh century, the parts of this old defensive ring were incorporated into the small Rus' castle, situated on the upper terrace, found within the western part of today's castle (fig. 16,28).

The original defensive system of Kamyanets also safeguarded permanent communication between the town on the island and the castle on the plateau via the bridge (fig. 7,47,49). According to the detailed field measurements of the Castle Bridge the core of the bridge, which includes piers and arches, was faced on both sides with solid walls at the end of the seventeenth century. Underneath remains the old core with a five part arcade on piers, comprising the first two phases of construction. The piers belong to the first, Daco-Roman phase which can be dated by ceramics and construction mortars to the 2nd -3rd centuries A.D. The arches belong to the second, Gothic phase (fig. 54). This archeological dating of the first construction phase of the Castle Bridge to the second century A.D., more precisely not later than 138 - 180 A.D., in turn, has redefined our search for comparative bridges. The search for the historical context in which a bridge in the Middle Dniester basin would have been built has focused interest on the Daco-Roman wars (Trajanic wars) and on available sources about buildings from the period.

A rich source documenting the Roman war against Dacia are the bas-reliefs on Trajan's Column in Rome. On the relief Nr. XCIX (Cichorius) a bridge is depicted. According to the generally accepted view this depiction represents the bridge across the Danube river near Drobeta (in present-day Rumania), built in 105 by Apollodoros of Damascus at Trajan's behest (fig. 58). The identification of the bridge on this relief by Cichorius, Coarelli, Aschbach, Petersen, Duperrex, Choisy, Davies, Tudor, Popescu, D. Antonescu, Th. Antonescu, C. Diacoviciu and H. Diacoviciu as the bridge across the Danube is based on a number of serious misreadings and miscalculations. If the depiction of bridges follows the convention depicting temples, where care was taken to show the key features such as the number of columns, then the bridge on the Column would represent a bridge other than the actual one across the Danube. The precise depiction of the bridge shown on the Column corresponds very little to particularities of the bridge across the Danube, whereas the nature of the construction design, the absence of stone arches and the use of a wooden supporting structure of the span appears analogous to the Castle Bridge in Kamyanets.

The bridge across the Danube, according to Dio Cassius' description and according to field measurements taken by Popovici in the 19th century, may be characterized as a twenty-one span construction with the length of its midstream section equal to 1055 m (fig. 59). The axial span length of the bridge was 170 feet (about 51 m) and the support gauge -50 feet (14,8 m). On the bridge piers stood an arch-shaped supporting structure of the span made of wood. Assuming that the height of the bridge balustrade in the bas-relief on the Trajan Column was 4 Greek feet (about 1.2 m) it was possible to ascertain all key dimensions of the bridge. The length of the bridge mainstream section was 5.4+5.8+6.0 + 6.0 + 4.2 + 2.1 x 5 = 37.9 m (128 feet) and the pier gauge - 2.1 m (7 feet).

By comparison, the mainstream five-span section length of Castle Bridge in Kamyanets Podilsky is about 4.8 + 6.3 + 6.6 + 5.4 + 4.5 + 2.1 x 5 = 38.1 m (about 129 feet) and the pier gauge - 2.1 m (7 feet). The correlation of key features and main dimensions of the Castle Bridge in Kamyanets and those in the depiction of the Bridge on the Trajan's column would seem to be sufficient reason for identifying the former with the latter. We would further propose a reconstruction of the Castle Bridge in Kamyanets Podilsky utilizing a wooden supporting span structure on piers as shown on the Column relief (fig. 61).

A comparison of the bridge depicted on the relief with the Danube bridge, on the other hand, shows that the scale, the number and the size of spans, and the bridge structure are about seven (!) times smaller: the number of spans 5 and 18, and their axial span length - about 6.0 - 7.0 m and about 51.0 m (fig. 60). The construction of the wooden supporting span structures shown on the relief for a bridge with small span could not have been realized in the case of the large-span bridge across the Danube. Comparing the depicted bridge structure on this relief with bridges across the Danube and the Smotrich river has led us to question the universally accepted identification of the bridge depiction. In no way does the depicted bridge resemble the bridge near Drobeta. On the other hand, a bridge of this size and the number of spans corresponds to the Castle Bridge in Kamyanets Podilsky.

The fact that the image of the bridge in Kamyanets appeared on the Trajan Column in Rome is a proof that the town of Kamyanets and the whole Middle Dniester basin territory, played an important role at the beginning of our era. Furthemore, the above conclusion indicates that the present views on the way of conducting military operations and the route covered by the Roman legions during the Trajanic wars need reconsidering. Theoretically the depiction of the bridge on the Column can be explained as a depiction of the communication built according the Apolodor's conception of the military roman bridge in the fartherst northern place of the Roman province.

Objecting to the attribution of the bridge on the relief Nr XCIX, the authors put forward a new hypothesis concerning the exterior of the bridge across the Danube. From their point of view the depiction of this bridge can be found on another bas relief on the Trajan Column - Nr LXXXVI (fig. 62), where it corresponds to the stone bridge shown out of perspective. The authors assume that the bridge built by Apollodor of Damascus had not wooden but concrete arches; this was the first approbation of Roman concrete in the constructions with a large span. Not without purpose the antique authors called that bridge "made of stone" and considered it a unique construction that had surpassed all buildings of Apollodor. From authors' point of view this bas relief also represents the portrait of Apollodor of Damascus (fig. 63), depicted near the Emperor Trajan as the priest holding the bull (fig. 62). As to the definition of the relief at present it may be considered preliminary and tentative.

The fact that the image on Trajan's Column does not have to be the Danube bridge and could be taken as another, smaller bridge, like the one in Kamyanets further supports available archeological evidence that the town of Kamyanets, and the whole Middle Dniester basin, were part of the Roman frontier at the beginning of our era. The implication of these findings provide sufficient cause for an urgent reconsideration of Roman military operations and, in particular, of the routes taken by Roman legions during the Trajanic wars. The Daco-Roman origin of Kamyanets advanced by the authors in a 1992 article (O.Pla-menytska, E.Plamenytska. Daco-Roman Period in the History ofKamianets-Podilsky: Toward an Understanding of the Problem // Society of Ukrainian Historical Archaeology. 2nd International conference "Problems and methods in historical archaeolohy". Abstracts. Toronto - L'viv 1992) can now be validated with these further analysis. The shift to an earlier, pre-medieval date for the Castle Bridge as well as and Old Castle fortification should lead to a revision of the geopolitical role of Kamyanets Podilsky; the Middle Dniester basin should now be regarded as the contact zone between the proto-Slavic and Roman worlds.

�� � 4 1999 �.