THE MARBLE BASILICA (B) IN TROPAEUM TRAIANI

IOANA BOGDAN CĂT ĂNICIU

The importance of the city, which developed at the foot of the triumphal monument in Adamclisi, grew
in the Constantin ian period. It is rather difficult to conclude whether the development of the city is the resuit of
radical destruction caused by a Carpian or Gothic attack or by severa) successive attacks, or it is merely an
imperial decision meant to bring new life to a municipium created by Trajan, the emperor taken as a model.
1
Dobrudj a, integrated into the Empire after Dapyx' defeat in M. Licinius Crassus' campaign , was
not transformed into a province, but, alongside the entire Moesia, was the "Kommandobereich"2 of a
legatus A ugusti exercitus. The form of pre-provincial organization 3 lasted relatively long, as in the two
military themes, which made up the border territories on the Rhine. The two Germanies were military
themes under the command of legali A ugusti exercitus and only Domitian and Trajan would provide
these territories with the administrative territorial organization characteristic to a province. It can be
noted that within this development phase Traj an is the first to apply a strategy of municipalization of the
territories in the northeast of the Empire. Traj an took into consideration the form of civilization
developed by the local population in the previous period. We can thus understand why not only Greek
cities maintain their own urban form and the traditional administrative structures as well as the strongly
hellenized Thracians in the southern Balkan s will alsa develop their traditional structures along
Hellenistic lines. Unlike here, in the Getae regions south of the Danube and western Dobruj a - what was
identified as ripa Histri - where the Greek influence was not overwhelming, and Traj an will determine
Roman-type mun icipalization. Only lately has more been discovered about Dobruj a ' s administrative
organization 4, although it is reluctantly accepted that there were civitates peregrinae here, subsequently
transformed into municipia with ius Latii.
Tropaeum Traiani was the first municipium on Getic territory and it probably covered a relatively large
territorium. We can see the first wall of the municipium5 (of probably 7 x 4 actus) oTily as the administrative
and religious center of a Getic population that continued to live in the rural territory of the city. The epigraphic
material6 discovered as building stane in the edifices of the Roman-Byzantine city highlighted the role of the

1 I. Bogdan Cătăniciu, „A propos de Ripa Histri", Novaensia 1 5 , Actes du colloque lnternational - Novae and the
Roman Limes, Swistow - Novae, 1 1 -15 sept 2002, Warszawa, 2004, 73-87.
2 About the meaning of provincia in Res Gestae Augusti and politics in border territories, there is a remarkable

study by Braunert Horst, „Omnium provinciarum populi Romani. . . fines auxi"' (Ein Entwurf), Chiron 7, 1977, 207-209.
3 The absence of clear signs of administrative-territorial organization before the reign of Domitianus, who created
Moesia Inferior, and the rare identification of military presence before Vespasianus, supported the hypothesis of the
lower Danube territories having been „given over" to the Thracian client kingdom.
4 Al Suceveanu and Al. Bamea, La Dobroudja romaine, 1 99 1 , 47-55.
5 I. Bogdan Cătăniciu, in Tropaeum Traiani, Cetatea I, 1 979, 53-54, 60-62; eadem, Tropaeum Traiani, Cetatea,

Electonic-Version, 2003. It seems that these Tropaeum Traiani monographs were of no interest to R. Cîrjan, not even
when, after the presentation of the paper „Tropaeum Traiani un municipe de droit latin en Mesie Inferieure?" at the
Convegno Romanita danubiano-balcanica I-VI secolo d. C, in April 2004, Al Bamea inforrned him that I had shown
evidence of this status of the municipium.
6 See Tropaeum Traiani, Cetatea, Electronic Version-Cimec 2003, chpt. „Introducere Istorică şi Izvoare Antice".

Dacia, NS, tome L, Bucarest, 2006, p. 235-254

https://biblioteca-digitala.ro / http://www.daciajournal.ro
236 Ioana Bogdan Cătăniciu 2

Getic population in the creation of municipal elite as a resuit of assuming magistrate ship. Tropaeum Traiani
confirms the policy of integration of the local population in provincial life and it excludes the hypothesis,
formulated without arguments, of an inferior social status of the Getae7 • The reasons for the granting of the
municipium status must have had to do with the fact that the settlement of the local community was on the
access route from west to east and on the road south. This îs the route the Getae from the north bank of the
Danube took in Domitian's time, and Decebalus' Dacians, allied with the Roxolans, during Trajan ' s first
Dacian war. This geographical position, advantageous in times of peace, facilitated the battles commemorated
by the altar and the funerary tumulus and the triumphal monument dedicated to Mars Ultor, but also the
Costoboci8 attack and then those of the "barbarians" in the 3'd century AD.
As a resuit of Diocletian ' s reforms and the political-economic improvement of the eastem Empire under
the reign of Constantine the Great, Tropaeum Traiani would benefit again from imperial support in order to
develop as the urban and religious center of southem Dobruja. We have sufficient reasons to believe that after
the Milan edict, a bishopric will be established here9, as part of the policy of developing Christian religious
comrnunities on existing municipal structures 10•
The axis of the marble basilica constitutes the central axis of the north-west sector of the Constantinian
city and an area with direct access from the Pomerial Street was probably initially assigned to it. The
Constantin ian wall built a fundamentis included a part of the hill which was left outside the city wall in the
Principate period, at least in the western sector (Fig. 1). The street network of the municipium area probably
developed in the areas included in the large Constantinian city, but the researches carried on the network of the
4th century and of the previous ones have been insufficient. Remarcable, the basilica was built in respect of the
frame of the early Roman city, over the previous edifice, but the baptistery had the long axe parallel with the
city wall of Constantin and Licinius, determined by the shape of the plateau. In this way the baptistery was
planed to harmonize the space. There must have been a pomoerial street, but there are no elements to indicate
how wide it was (the baptistery is at half an actus from the wall), because buildings were erected on it, at the
rear of the wall, as in the case of the eastem gate. In the 5 th-6th centuries, the towers - here, the large supply
tower T l 2 - were reached from a narrow street (3.5 m wide). It is certain that decumanus was the southem
limit of the basilica in the 4th century, but we have no objective elements as to its position and size. The street
was reduced to only 2 . 1 m in the 6th century, and it was paved with bricks. Another street (4.3 m) was revealed
on the northem side of the basilica, by clearing the surfaces investigated at the beginning of the 20th century.
The excavations in the Roman city of Tropaeum Traiani went into decline after 1 977, even though the
researches carried out between 1 968-1 975, with significant resources, succeeded in bringing to light
numerous aspects of Roman-Byzantine life whose study was worth continuing. In the newly established
conditions, priority was given to the initiation of works meant to clarify essential issues regarding the
buildings partially investigated before 1 909, in view of their immediate restoration. The excavations of the
marble basilica in Adamclisi, led by Gr. Tocilescu and, after his death, published by V. Pârvan, were
restarted in 1 979, with the purpose of consolidating the basilica. Unfortunately, they can be not even today
considered concluded. Between the interruption of the research, at Gr. Tocilescu ' s premature death, and its
re-initiation, two World Wars passed over the ruins left at the mercy of the times. WWI left numerous traces
of battle on the territory of the city 1 1 • Part of Gustav von Cube' s site documentation disappeared, and only

7 Al Suceveanu, Al. Barnea, op. cit., 38-39.

8 B . Gerov, ActaAntHung 6, 1968, 326-3 37.
9 The bishoprics were from beginning of urban nature. See J. Zeiller, L' empire romain et l 'eglise, Paris, 1928, 46 ;
Letizia Ermini Pani, Spazio Urbano e organnizzazione ecclesiastica nel Mediterraneo occidentale, în Acta XIII
Congressus Internationalis Archaeologiae Christiana I, Split, 1 994.
10 I. Bogdan Cătăniciu, Les dernieres modifications de l 'urbanisme a Tropaeum Traiani în Acta XIII Congressus
lnternationalis A rchaeologiae Christiana III (Studi di Antichita Cristiana pubblicati a cura del Pontificio lstituto di
Archaeologia Cristiana, LIV, Citta del Vaticano, 1 998, 1 4 1 - 1 48).
1 1 In the central nave, there îs an inscription chiseled into the marble spindle, with the name of severa) soldiers,
Radovici/ Gheorghe / 1 9 1 6 Galati/ Bu----/Pavel / 1 9 1 6 Braila /Niculae IM--- , who had turned the ruins into resistance
nests; at tower T 4, în order to investigate the hypothesis of the existence of a northern gate, the former „dig" tumed out
to be such a machine gun nest - bullets made în 1904 had been used.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
3 The Marble Basilica (B) in Tropaeum Traiani 237

some elements are included in Tocilescu' s manuscript volumes at the Romanian Academy. The museum
created by Gr. Tocilescu in Adamclisi was looted when Dobruja was occupied in 1 9 1 8, and the entire
patrimony disappeared.

�··
.·
. /· ' i f.

·,

,.
,: •'"'-A�
'-..
• •

'
...... . . �....,

.· ....�.... -
( '..„·'
'
'·

\\

, --� <\.
\ -".�·. - ·.,;_ .

.i
,· )

Fig. l. Tropaeum Traiani, the city, restitution (IGFCOT) after the aerial photograph taken in 1973,
the street grid in the IVth c. A.D. as it is now known.

We started our own research with the purpose to reveal the stratigraphic and chronological elements
preserved after the excavations led by Grigore Tocilescu and performed by Gustav von Cube. We intended to
establish the phases and the planimetric evolution of the area of the western city wall, the streets and the
buildings. As we had limited financial resources, we only did probes. As compared to our financial resources,
the results were very significant (Fig. 2) . We established the phases în the existence of the basilica and its

https://biblioteca-digitala.ro / http://www.daciajournal.ro
238 Ioana Bogdan Cătăniciu 4

planimetric evolution throughout all these phases, we concluded the study of the baptistery; we cleared the
previously investigated ruins and we sketched these buildings to scale.

r---T-··�,��fJ - ,��- ��--,-1

; I
. i

:!

Fig. 2. Plan of the 'Marble'Basilica and the ruins in the B sector of the city.

Fortunately, the excavations at the beginning of the 20th century were only canied out in order to reveal
the surface of the buildings and the stratigraphy was affected only where the existence of a confessio was
12
supposed (in the apse and perhaps in the northem nave of the baptisterium) •

12 Even though we believe that the pit which destroyed the basin is one of those dug by soldiers during the First World
War, as G. von Cube would not have partially destroyed it, but he would have discovered the foundation of the basin.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
5 The Marble Basîlîca (B) în Tropaeum Traiani 239

Unfortunately, the research carried out at the beginning of the 2ot1t century did not pay attention to the
superficial constructions in the city' s last phase of existence. Therefore, we can not know whether there was
any form of civilian inhabitation after the definitive destruction of the basilica. The last layer în the basilica,
where G. von Cube's research stopped, was a layer of soii mixed with stane and a Iittle mortar, on which
there was a brick pavement preserved in several places în the central nave. At this levei, there are no traces
of violent destruction by fire, so that we can state that thi s was not the reason why the basilica was
abandon ed.
The resuit of the stratigraphic researches carried out between 1 979 and 1 992 (interrupted between 1 985-
1 99 1 ) in the marble basilica is the knowledge of the phases in the planimetric development of this monument.
There are three phases in the existence of the marble basilica (I-ill ) ; in all three phases, the plan and size of the
naos (25.50 x 1 8 m) and the apse (6.40 m large, 3.3 m in radius) are the same. The east-west interior axis is of
29 .4 m and the exterior dimensions are 3 1 .4 x 1 9 . 8 m. The basilica has three naves and one apse at the east of
the central nave (25.50 x 7.85).
We were able to establish that in antiquity, the natural slope of the plateau descended significantly
westward. The buildings erected in the Roman-Byzantine period used the ground levei difference and reduced
them, so that now the terrain is almost plane.
It must be emphasized that the stratigraphy in the north-westem sector is not the same everywhere. We
could see (Fig. 3) that in section S I (oriented towards the north-south, on the naves of the naos) there is the
same succession of levels as we were able to established în the eastem part of the city. A first Getic
inhabitation, a culture layer of brown soii, without the delineation of treading levels, but which includes a pit
(Fig. 4) (part of a dwelling?). The foundation of a colonnade was built over this pit. (Fig. 5) There are traces of
a fire previous to this construction phase - undoubtedly the Costoboci attack of 170 AD, about whose
consequences we have sufficient epigraphic data 1 3 •

.
.__.._ __.
__····--·-----

Fig. 3. Eastem profile of the section S I, 1980 over the naves of the basilica.

Fig. 4. The east-westem section S V (south profile) în the central nave.

We believe there may be a building, E 1 , from which we only have the foundation discovered in the
northem nave. We postulated this phase based on the fact that levei N II crosses the entire width of the

13 R. Vulpe, DID II, 1968, 1 59, 162- 163.

* See the stratygraphy în the Tropaeum Traiani, Cetatea I, Bucureşti, 1979, 35-40 fig. 8.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
240 Ioana Bogdan Cătăniciu 6

foundatîon parallel wîth the north wall of the central nave. We can assume that this buîlding exîsted at a levei
included in the brown soil layer with traces of bumîng. Ali our observations on the exîstence of thîs buîldîng
14
are hypothetical , as they are only based on two 2 m wide probes.

'" "' '

.-.�-„��·-?
I>

...
_ ......�
. ,

Fig. 5 . The stylobate and the bases for the columns in S V, 1980.

The stratîgraphîc research carried out in basîlîca B revealed a large buîldîng, E 2, previous to the
construction of the basilica. When a "Probe fur Fundamente" was carried out în the apse1 5 , în fact an attempt to
reveal a crypt undemeath the altar, the levei of this construction was reached and one of the column bases was
discovered (Fig. 6) on the stylobate found în the east-west section through the central nave (Fig. 7) . The walls
made of rough stone bound with mortar, 75-80 cm wide, were dîsmantled when the foundations of the naos
were erected. These remains of large buîldings were dated to the 2°d and 3 rd centuries AD 6 . The construction
1
of buîldîng E 2 began at levei N II, superposed onto a levei wîth traces of bumîng. The treadîng levei
correspondîng to buîldîng E 2 was preserved as a thin layer of well pressed yellow soii, corresponding to the
column base discovered în the central nave E 2 C 5 (Fig. 4). The edifice was rebuilt, and this îs proven by the
discovery, next to the plînth of the same column base, of a pit of approximately 25 cm full of mortar and
carved stone. The second levei N m of buildîng E 2 îs approximately 15-20 cm higher. The treading levei was

14 We cannot exclude the possibility that it is indeed a stylobate, parallel to the one discovered in the central nave.
A section may have been discovered between column bases in the northern nave, and thus covered by the treading levei,
from which only the column bases would have been visible.
15
Ms Tocilescu 5 1 29, drawing p. 163.
16
,,Epoca principatului la Tropaeum Traiani, cu privire specială pentru zona basilicii de marmoră", paper on this
inhabitation leve! în the sector of the basilica B, presented at The XXXVl'h Colloquium Pontica, Constanţa, 2004.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
7 The Marble Basilica (B) in Tropaeum Traiani 24 1

not preserved, only a destruction layer consisting of red lhne plaster. The building of the basilica was initiated
after the partial removal of the edifice ruins.
The north-south stratigraphic probe S I, (Fig. 2) which intersected all three naves, was particularly
revealing in the investigation of the initial phase in the existence of the building in the 4th century AD. lt was
noted that it replaced the previous pagan edifice, still incompletely investigated/known. Building E 2 was
systematically dismantled in a period dated by a coin issued in the time of Maximianus Herculius (Fig. 8) 1 7 ,
discovered in the leveling soii which covered the foundation of the wall. A large quantity of pink plaster was
discovered in the demolition layer of the pagan building. Taking into consideration the nature of the building
erected in the znct century, we dare to formulate the working hypothesis that we are looking at a temple and that
we are in the extra mural sacred area of the Tropaeum municipium. We suppose, based on the elements at our
disposal, that building E2 was entered from the east, from within the city.

__.,?

O 1� .�... „!1.;,i.. b:,-.J� I \,.\.\ H... ;..�p,....-v

t'"
4

! ) "°"'"'""� . f'W„:,� .,.. , r'V.v.JVJ... C .�:::

Fig. 6. Page of G. von Cube's manuscript of the excavation report.

The construction layer of the basilica (phase I - Fig. 10) is practically at the levei where the demolition of
the pagan building took place (Fig. 9). We were able to see in all sections that the levei at which construction
work took place was not the same as the treading layer in the finished basilica. Part of building E 2 was not
demolished, above the construction levei of the southem nave - so it was planned to raise the levei inside the
basilica by 0.3 m, in order to emphasize its monumentality. The church must have been at least 7 m tall,
because the foundations went 2.3 m below the construction levei. At the pavement levei of the first basilica,
placed over the filling layer of 30 cm over the leftovers from the construction work, the mortar with which the
southem wall of the southem nave was plastered trickled horizontally over few centimeters, thus marking the
levei, which must have been of stone or brick slabs.

1 7 RIC VI, Cyzicus, p. 58 1 , 15 b a. 295- 299.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
242 Ioana Bogdan Cătăniciu 8

A later levei (Phase II Fig. 1 1 ) is 25 cm higher, on a filling layer of clay with extremely few
-

arthaeological traces, but mixed with average sized quarry stones. This layer must have been covered by
pavement, which was completely removed on the surface we intersected. It must be noted, however, that the
stylobate was so high that it did not allow the cross passing between the lateral and the central naves, except in
places where there were steps. The last treading levei in the basilica (Phase m Fig. 12) is the one where
-

research stopped in the central and southem naves. In the northem nave, previous excavations also went below
the last levei. On clearing the surface of the central nave, we discovered some places in which remains of the
brick pavement were preserved. They also appear in the plan V. Pârvan published (BCMI IV, 1 9 1 1 , Fig. 15).

Fig. 7. Image of the section S V in the apse.

Fig. 8. Coin issued by Maximianus Herculius.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
 The Marble Basilica (B) in Tropaeum Traiani 243

Fig. 9. Part of the building E 2 nondemolished S I, 1 98 1 .

"" •i ""

.. .!1 i"J.:.�'
: ! '

: \ .;.
l�
.
: bi,
�·····
.1...1„...
• . •..

'--··

Fig. 10. Basilica in the Constantinean times - Phase I.

The cancelli (4.40 x 7 m) were built then - their bases made of parallellipipedic marble blocks were
preserved in situ. The orifices for the marble slabs of the railing can be also seen in the bases.
https://biblioteca-digitala.ro / http://www.daciajournal.ro
244 Ioana Bogdan Cătăniciu 10

'U

„..
/c u.�'.
·B '

Fig. 1 1 . Basilica in the middle Fig. 12. Basilica reconstucted in the Anastasius­
of the Vth c. A.D. - Phase II. Justinian period - Phase III.

We believe that the remains of a shell limestone slab discovered 3 . 3 m in front of the cancelli, on
the central axis, can only be identificatethe remains of an Ambon . V . Pârvan, publishing G. von Cube's
research results 1 8 , believes that the basilica went through another rebuilding process, left unfinished, as
the processing of the column capitals was interrupted. If we discard this argument because those capitals
are finished and the ornaments are typical for the period 1 9, we can only wonder whether a superior
treading levei was removed. On the basis of the discoveries made at the eastern gate of the city compared
with others discoveries in the other sectors, we con sider that there are data regarding a rebuilding in the
middle of the 5lh century (N IV B ). The evidence is a coin issued by Theodosius 1120 , and we believe it is
possible that the rebuilding of basilica B to have been initiated in the same period. The Justinian period
(N V) is very likely marked by the last rebu ilding of the basilica in its sumptuous form with specific
marble elements that gave it its name in the literature. Even without the argument regarding an
interruption of the late rebuilding - the fourth phase (Fig. 13) supposed by V. Pârvan - it is plausible that
this phase existed. The evidence may be a late consolidation of the northwestem pillar of the narthex and
the building of the columns in the space created east of the apse. Changes to the shape of the northem

1 8 V. Pârvan, BCMI IV, 1 9 1 1 , 1 80- 1 8 1 .

1 9 I . Bamea, DID, 1968, 484; R . Kautzsch, Kapitellstudien, Leipzig, 1936, no. 566 f; 583; R . Hodinott, Early
Byzantine Churches in Macedonia and Southern Serbia, 1963, pi. 39 K .
20 Tropaeum Traiani, Cetatea I, 225 n. 9.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
11 The Marble Basilica (B) in Tropaeum Traiani 245

wall of the northern portico of the atrium took place in the same sub-phase. Unfortunately, we do not
have sufficient elements to be able to date every rebuilding and addition that can be set in the last
inhabited phase of Tropaeum.
The stratigraphy in the narthex is difficult to decipher due to the numerous ancient interventions in
a limited space. We believe that G. von Cube' s excavations also stopped grosso modo in the narthex at
the last levei of existence. The penultimate levei near the wall of the naos consists of a layer of limestone
over which there is clay with a little mortar, 0.25 m below the current levei. The access stairs to the
central nave were probably built on this clay layer. The narthex existed in this phase with the aspect that
we know today.

V
!oi
-�
b::l'.:-:::.- .d :. J '

;,_ .iL��-�. :--:r

:§: -� . ,.,,,:;,-:;::.
_:;:;

\' .t�:�· "t

!?�::,����

·--··

Fig. 1 3 . Latest (phase III 1 ) changes in the shape of the basilica after the Avars' destruction (586 A.D.).

At -0. 60m near the naos there is average stone similar to the one used for pavement. At - 1 .07 m there
is a sort of pavement made from mortar and stone slabs. It is a layer of yellow soii , stone, brick and lime
mortar with crushed brick in it, which goes down to - 1 .35 m. At - 1 .55 m, there is a layer of pebbles and
mortar - layer which probably marks the time the basilica was built. The construction levei was arranged by
bringing a layer of yellow soil in order to cover the rests of a fire and the walls of the 2nd _3rd centuries,
probably part of edifice E 2. Over the layer of virgin soii there is the layer of brown soil with early Roman
material.
In the first construction phase I, the basilica with three naves and an apse (Fig. 10) to the east could be
entered from the west, by way of a staircase which emphasized its grandeur2 1 , reducing the ground levei
difference in relation to the westward slope (Fig. 14).

21
A coin issued by Constantius II at Nicomedia on revers GLORIA EXERCITVS was found in nartex.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
246 Ioana Bogdan Cătăniciu 12

There is no evidence of the number and interval between the columns, which separated the central nave
from the lateral ones. Nor do we have any way of knowing whether entrances to all three naves or only to the
central one existed from the very beginning.
It is important to mention the fact that the basilica covered the same surface as the previous pagan
buildings, fact revealed by the probes in the narthex. For this initial phase I, the foundations for a propylon
structure were laid over the partially demolished 2°d_3rc1 century wall. In 1 992 we discovered a column base c N 3
(Figs. 15, 16), set on a stone and clay foundation, built on the dismantled 2°d 3 rc1 century wall. Another column
_

base c N 2 uncovered in section S XVIII may be part of the same colonnade if we take into account the fact
that its foundation was built over the remains of the same early Roman wall. This column base (Fig. 17) was
taken out of use and covered by the soii layer which constituted the functioning levei for the later propylon,
built 3 m west from the basilica.

Fig. 14. Coin issued by Constantius II at Nicomedia found in the narthex.

Fig. 15. Profite of the section in the narthex (S XXV); the base of column c N3 .

The surface of the basilica is likely to have been delineated by a temenos, which was very common in the
Constantin ian period. A baptistery was built in this area, to the southwest of the basilica (Fig. 18). At the levei
dated on the basis of 4th century material, the baptistery only had two rooms, approximately equal in size, and it
must have been entered from the southem room (Fig. 10). In the northem room there was an apse embedded
within the eastem wall. The eastem wall of the nave bas a foundation of 2 m deep. The foundation was

https://biblioteca-digitala.ro / http://www.daciajournal.ro
13 The Marble Basilica (B) i n Tropaeum Traiani 247

excavated down to the virgin soil, going past the layer of brown soil, which contained several early Roman
ceramic fragments.
The substructure of the apse included a channel which supplied water to the piscina and a niche, which
was in an advanced state of decay almost a century after it was first revealed. Only the foundation of the basin
was preserved, and only its western half, but its quadrilobate plan can be reconstituted perhaps on the square
area (Fig. 19).

Fig. 16. Image of the section east-west in the pronaos base colurnn c N 3.

Fig. 17. Image of the section S XVIII in the south nave of the pronaos, base of colurnn c N 2.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
248 Ioana Bogdan Cătăniciu 14

The probes 22 in the narthex revealed the exi stence of a wal l, 3 m west of the basil ica, paral lel to
the western side of the naos. The wall, 63 cm wide, i s made of stone and soi i and it breaks off in front
of the entrance to the central nave of the naos. The precarious state of the foundation indicates the
exi stence of only one col onnade in front of the entrance to the naos. It can be as sumed that this i s a
stylobate for a larger propylon. The column base (c N 1 ) d i scovered in S XVIII would be part of this
propylon . It is only for stratigraphic reasons that we can consider this moment as part of phase N IV B ,
dated Theodosius 11 23 .

Fig. 1 8. The baptistery, photograph taken at the beginning of the XX:th century, after BCMI, IV, 1 9 1 1 .

· �:.

-
.- � · �
; �· ,

Fig. 19. The piscina view from the east.

22 We made four probes perpendicular to the Iong side of the narthex: two in 1 984 and another two in 1 992, meant
to complete the data we did not consider relevant. In fact, it would have been preferable to perform surface digs in this
area in order to have certainties, not hypotheses.
23 K. D imitrov, ,,Novae on the Low Danube as Early Christian Centre (5-6 century A.D.)", paper presented at
The XII Internationaler Kongress ftir Christliche Archăologie, Bonn, 1 99 1 a restauration phase of the city after
-

Huns attacks, an the built of the cathedral .

https://biblioteca-digitala.ro / http://www.daciajournal.ro
15 The Marble Basil ica (B) i n Tropaeum Traiani 249

The narthex was built in the form we know today at a later phase, at the end of the 5th century - the
beginning of the 6th century (N V). It is divided into three areas with massive pilasters on both the eastem and
the western walls. The size of these pilasters makes us believe that they supported vaults. On the southem side,
the narthex had an entrance and it had alsa a propylon.
An atrium (Fig. 20) with porticoes on three sides was built în the same time with the narthex. Between
the narthex and the atrium there still was a ground levei difference, so that they communicated through a wide
central staircase and two staircases corresponding to the lateral porticoes. This phase of the atrium functioned
60 cm below the current levei and the porticoes were alsa open to the exterior. Above the foundations, the
porticoes had a quarry stane elevation. We only have data from a probe carried out în the northem portico and
we cannot know what was there before the atrium was built. There might have been a timber building from
which we recovered a pillar pit. The southem portico was entered from the courtyard of the baptistery, and the
western one from a street parallel to the city wall. The probe on the northern side of the northern portico
revealed a subsequent rebuilding of the latter. The northem portico opened ioto a 4.30 m wide street, bordered
by a building which has not been yet investigated and opened ioto the street through a 1 .4 m wide gate. The
threshold of the entrance îs at the same levei as the foundation plate of the northem wall of the portico. The
street might have been paved with stane slabs, two of which were discovered on a 70 x 70 cm area near the
northern wall. It must be mentioned that this entire area was excavated at the beginning ofthe 20th century and
we only cleared the recent deposits. Only a small area in the northwestern corner was untouched by the
previous excavations and we discovered a fragment of a pot decorated with a band of horizontal and wavy
lines, and the fragmented bottom of an inscribed glass vessel.

Fig. 20. Image of the atrium, view from the east, channel which served the phiala.

When the radical reconstruction took place (Phase ID) at the end of the 5 th century and the beginning
of the 6th, three rooms were built east of the basilica. They were the size of the three naves of the basilica
and were used as pastophoria. Later, the apse was redane "a epaulement". The foundations deepened a little
în relation to the levei at which they were built, at the same time; the room behind the apse îs divided by
two columns (or three as in the plan published by V. Pârvan 24), probably for consolidation purposes.
Unfortunately, G. von Cube excavated down to the construction levei of the walls of these rooms and
removed the stratigraphic deposits. The existence of the column base (Fig. 21) at a much higher levei than
the one trodden today, gives us reason to attribute it to the last construction phase of the basilica, (phase ID 1 ) ,
the one Pârvan stated and which we can date after the destruction described by Theophilactus Simocatta.
From what we d iscovered, from an archaeologist ' s perspective, without the necessary knowledge to assess

24 BCMI IV, 1 9 1 1 , fig.20.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
250 Ioana Bogdan Cătăniciu 16

the likelihood of this hypothesis from an architectural point of view, w e note that the apse wall was
destroyed down to a very low levei. lt is possible that this massive destruction took place during the A var
attack and the apse was never rebuilt. The wall is below the levei of the column bases that G. von Cube
discovered which are only 1 .5 m away from the apse wall and 1 m from the eastem wall of the central room
to the east of the basilica. If the apse ceased to exist, it could simply mark (it could serve as a synthronon)
the section reserved to the altar in a rectangular area where there was a deambulatory marked by the
columns, which suggest a circular arc.

Fig. 2 1 . Profile of the section S XVI in the central room built east from the absis.

The stratigraphy în the baptistery area is different from the one in the area of the marble basilica. We
investigated the stratigraphy in the eastem, southem and western areas of the baptistery and inside the three
rooms.
In the perpendicular probe (Fig. 22) on the eastem side of the northem nave of the baptistery - S IV - we
only discovered three levels datable to the Roman-Byzantine period, over the early Roman brown soii. This îs
also confirmed by the survey in tower T 1 3 . Here it was a non-built space.

� rITTîT1 4
�1 11 l.li...LJJ �5

o 1 2 3 m.

Fig. 22. Profile north of the section S IV, 1 984.

The foundations of the baptistery are set in a O. 7 m thick dark brown layer over the virgin soii, with few
Roman ceramic materials from the 1 st_3 rd century AD.
https://biblioteca-digitala.ro / http://www.daciajournal.ro
17 The Marble Basilica ( 8 ) i n Tropaeum Traiani 25 1

Above this layer there is a construction levei defined by ceramic material from the 4111 century25 • The first
phase ofthe baptistery ( 1 1 . 1 x 6.4 m), its long side oriented north south, was bui lt at the levei represented by a
layer of mortar and carved stone. Only the foundations are preserved ( 1 .4 m wide in the northern room and
only 1 . 1 m wide in the southern one and 1 .9 m deep), which were used in the rebuilding ofthe baptistery in its
current form, from quarry stone bound with mortar. A first nave to the north covered an area of3.5 x 3 . 5 m, while
the surface ofthe second room, a vestibule, was of3 .8 x 3 . 8 m. The massive walls ( 1 . 3 5 m wide) ofthe square
northern room suggest the existence of a vault26 •
There were two column bases at this levei outside the baptistery - one of their spindles remained in situ
CV 1 , published (BV5C l ) in the monography2 7 , with observations concerning its role in the structure of the
street porticoes in this area. The second column base C V2 is 2 . 7 m away from the eastern wall of the
baptistery, and the interaxis of the columns is of 2 . 7 m as well. In the 4th century, the street must have been
approximately 8 m wide. Column CV2 seems to be in the corner of the porticoes of the two perpendicular
streets. The baptistery was built at the top ofthe street which ascended from via principalis. It was reduced in
the 51" or 61" century, at the !atest, to only 2. 1 m when a series of stone and clay buildings were erected. The
row of columns is 6 m south of the marble basil ica and it is parallel to it. As a conclusion to the observations
already made, we can confirm the existence of a portico, which functioned in the N IV levei, represented by a
thick stone slab pavement. Only one ofthe slabs was preserved in the immediate vicinity ofthe eastern wall of
the baptistery. It is important that in the first phase, I, the building also functioned as a baptisterium, as under
the treading levei there was a channel (Fig. 23) which supplied water to a cruciform baptismal font (Fig. 19).

O 1 m
1. ...

Fig. 23. Plan and profile of the east-west section over the north nave of baptistery and the space in the west of it.

25 It must be emphasized that this probe also revealed an extremely small amount ofceramic material - proofofthe
fact that we are not within a living space.
26 The same opinion: Sp.Cegăneanu, BCMI IV, 1 9 1 1 , 1 93 .
27 M . Monica M ărgineanu Cârstoiu, A l . Barnea in Tropaeum Traiani, Cetatea I, 1 979, 1 2 1 , fig. 1 05 .

https://biblioteca-digitala.ro / http://www.daciajournal.ro
252 Ioana Bogdan Cătăniciu 18

Level phase I was at a maximum of 5 cm below the current treading level, given the thickness of the
slabs replaced with the current ones. At level II there was a street portico (?) comprising colurnns , C V 1 and C
V 2, which had already been used in the previous levei.
In the initial phase, an apse was built in the eastern wall of the northern room of the baptistery,
recognizable by the opening of the arch carved in block (a on Fig. 23) still preserved in situ.
A smaller niche was built at an inferior levei, embedded within the eastern wall. lts opening was
circumscribed to the arch of the apse. At the time we restarted the investigation of the baptistery, the smaller
apse was already difficult to recognize, as the mortar had deteriorated and the fallen stones had been removed.
We supposed its existence based on the recovery of a stone block (b on Fig. 23) with a suitable arch for such a
niche. This block was 30 cm tal!. The water channel (Fig. 24) was inscribed in the semicircular recess. Its
section was of 30 x 30 cm and 75 cm of its length to the east have been preserved. The falling slope towards
the basin is visible within the wall to the west. In the reconstruction ofthe baptistery, S. Cegăneanu supposed28
the existence of a baptismal font for whose drainage the channel had been built. The completion of the
investigation invalidated this hypothesis. The channel comes from the east and the water came through an
aqueduct, which probably disappeared in the 6th century. At the levei preserved today, there is a slab (a) in the
eastem apse with a 1 O cm deep 25 x 25 m hole in which a stone pillar was placed, against the wall of the apse.
It was probably the leg of a mensa.

Fig. 24. The aqueduct channel in the eastern wall of the baptistery, view from the east.

The baptismal font was used during two phases, after which it was destroyed and probably replaced with
a baptismal font that we have not discovered. A "hiatus" can be seen in the structure of the western wall ofthe
baptistery, with a 40 cm levei difference between the eastern and the western sides of the wall. This anomaly in
the structure of the wall can only be explained by reconstruction, after the basin was abandoned and a drainage
pipe for the baptismal water was no longer necessary. In the north-western corner of the room, in the free area
near the basin we discovered a sort of "channel" built in small stone. The space was closed with an arched lid,
from the same type of stone. Everything is bound with clay. (Fig. 24) We cannot say which one of the two
drainage systems functioned in the first sub-phase of basin use and which in the second. The basin was
demolished in antiquity. The font was used during two phases. It was taken out of use by a pit which was
observed immediately under the levei of the northern and western apse, which transformed the northern nave
into a trilobate area, probably in Anastasius' or Justinian' s period.

28
BCMI IV 1 9 1 1 , 1 80 fig 25.

https://biblioteca-digitala.ro / http://www.daciajournal.ro
19 The Marble Basilica (B) i n Tropaeum Traia11i 253

The foundations of the basin were destroyed by a deep pit which we cleared, discovering fragments of
marble spindles, stones from the smaller apsides which existed in the last phase embedded within the wall
between the northern and the central naves. Ali these deposits are of recent date, as in the photograph
published by V. Pârvan the apsides in the southern wall of the northern nave were very well preserved at the
time of G. von Cube' s investigation.
The preserved remains suggest that the basin had the shape of a Greek cross, more precisely four
semicircles on a square. The axis of the basin measures 3 x 3 m. Only the foundations were preserved. These
are not of the same depth, which suggests the existence of a deeper basin, perhaps with steps in the central
square area. The foundation measures 0.8 m in the central area, and 0.6 m in the semicircular one. The first
1 5 cm of the foundation were made of stane without mortar. In the upper area, with lime mortar, there were
also architectural elements from previous buildings and brick fragments.
It was probably at the time when the building spirit typical of Anastasius' and Justinian's period made
itself strongly felt, that the aspect of the baptistery changed. It adapted to more sophisticated religious
requirements. The baptistery building is divided into three rooms and the eastern apses are built for the first
two rooms resulting from the division of the southern room. The northern nave is triconch, as apses were also
embedded within the northern and western walls. The basin was abandoned, as the baptism of children 29 was
not done by immersio anymore, but by effusio.
The probe in the southern nave revealed the stratigraphic element which sets the moment of change in
the plan of the baptistery: the building of the dividing wall of the phase I southern room 30 cm above the
construction levei for the entire building. The wall was implanted by digging a foundation pit 20-30 cm wider
at this levei, for which there are no absolute dating elements. We noted that in the northern nave, the northern
and western apse of the triconch 30 were built at the levei at which the basin was dismantled.
The baptistery was rebuilt for the last time after massive destruction and fire, datable to the 6lh century.
It appears to have been demolished then, at least in the southeastern area, where burning traces are still visible
under the stones in the apse' s wall, at the treading levei.
Therefore, our observations concur with G. von Cube's. The dating of the three phases in basilica B is
slightly different from that proposed by V. Pârvan 3 1 • We believe that the beginning of the basilica must be set
before the middle of the 4th century AD . As it was well observed3 2, the marble basilica is the first basilica in
Tropaeum, built in the Constantin ian period. The main outcome of our research is the certainty that the baptistery
was built at the same time as the Constantinian basilica and it reinforces our belief that, in this oriental province
too, bishoprics developed on the existing administrative, municipal structures. The existence of the municipal
center, which developed into a strong Constantinian city, allows33 us to state the existence of severa) bishops in
Scythia Minor34 . The one in Tomis was probably recognized as metropolitan from the very beginning.
The destruction of the city by the Goths in Valens' period was not revealed in Tropaeum Traiani,
therefore we cannot assume that the first rebuilding of the basilica might have been caused by possible
destruction. The only massive destruction identified in the area of the marble basilica is of a later date, post
Justinian II, probably the Avar destruction Theophilactus says ended life in Tropaeum. All our researches so
far show beyond any doubt that the Avar attack did not cause the city to be abandoned. The city continued to
be rebuild and the marble basilica existed at least until civitas Tropaeum would gradually be abandoned by its
inhabitants as a consequence of the fall of the "limes" and because of the disappearance of the imperial
power35 , which had ensured a poorer but nevertheless still urban way of life.

29 Augustinus.
30 Anna Gattiglia, ,,L' Architecture symbolique de l'epoque de Justinien dans Ies Balkans et la Region Danubienne",
paper presented at the Xll/ Co11gressus lnternationalis Archaeologiae Christiana I, Split, 1994.
3 1 Ibidem, 1 8 1 .
3 2 V. Pârvan, loc cit.
33 1. Bogdan Cătăniciu, „Notes sur J 'evolution architecturale de la basilique de marbre de Tropaeum Traiani", i n
Akten des X// lnternationalen Kongressesfar Christliche Archăologie, 801211, 199 I, Jahrbuch fi.ir Antike und Christentum
Ergănzungsband 20/ l , 1995, 578-586, pi. 7 1 .
34 Only for a later period - Emil ian Popescu, „Contributions a la geographie historique de la peninsule Balkanique aux
V-VIII siecles de n .e.", Dacia NS 1 3 , 1969, 4 1 2 ; I. Bamea, DID II, 1968, 458-459, 47 1 ; my opinion in loc. cit. 584-585.
35 I.Bamea, DID, 1 968, 4 1 6-43 8 .

https://biblioteca-digitala.ro / http://www.daciajournal.ro
254 Ioana Bogdan Cătăniciu 20

BIBLIOGRAPHY

Armstrong G., „Constantine's Churches: Symbol and Structure", Journal of the Society of Architectural Historians 33,
1 974, 5- 1 6.
Barnea I„ Arta creştină în România, I: secolele /li-V/, Bucureşti, 1 979.
Brown, P. R. The Making ofLate Antiquity, Cambridge, 1978.
Duval N „ Lezine Al., ,,Necropole chretienne et baptistere souterrain", în Cahiers archeologiques X , 1 959, 7 1 - 147
Davies J.G„ The Origin and development of Early Christian Church. Architecture, London, 1952.
Fox Robin Lane, Pagans and Christians, New York, 1987.
Gibbon, Ed., Decline and Fall of the Roman Empire, London, 1 776 (E-book).
Grabar A„ Early Christian Art, New York, 1 968.
Grabar A, Le Baptistere Paleochretien, Mulhouse, 1980.
Kautzsch, R., Kapitellenstudien: Studien zur Spatantiken Kunstgeschichte, Berlin, 1 936.
4
Krautheimer R., Early Christian and Byzantine Architecture , London, 1986.
Pascal Ch„ Les baptisteres pateochretiens de la province romaine de Dalmatie, Diadora, Zadar, 1968, 1 1 - 1 5 8 .
Popescu Em„ „Contributions a l a geographie historique d e l a peninsule Balkanique aux V-VIII siecle d e n .e.", Dacia NS
1 3, 1 969, 404-4 1 5 .
Ward-Park.ins B „ Studies in Roman and Early Christian Architecture, London, 1 994.

https://biblioteca-digitala.ro / http://www.daciajournal.ro