Bridging the Gap: Continuity

and Innovation in Ceramic Kiln

Technology from the 6th c. BC
to the Beginning of the 7th c. AD
in the Territory of Bulgaria
Archaeologia Bulgarica Alexander HARIZANOV
ХХIII, 3 (2019), 15-39

1
Probably the earliest two-chambered Abstract: This contribution explores the archaeological evidence for the development
kiln, found in situ in Bulgaria, was the one of the two-chambered updraught ceramic kiln in the territory of modern Bulgaria,
discovered near the village of Koprivlen, from the 6th century BC to the beginning of the 7th century AD. On the one hand, its
Gotse Delchev Municipality (see Божкова aim is to examine the possibilities for continuation in kiln building technology, from
2006). Another kiln of early date (sec- the Late Iron Age through the Roman and Late Antique periods. On the other, to high-
ond quarter of the 5th c. BC) is one of the light the innovative technological practises, which appear in the region after the Roman
structures, excavated at the ancient town of
Apollonia Pontica (Baralis et al. 2016, 159- conquest, and to assess their effect on the local traditions in ceramic production.
160, figs. 6-7). The data base of this study includes 340 kilns of the time period in question,
2
Parts of nowadays northern Bulga- divided in two groups in accordance with their dating. The first one comprises the
ria were subjugated by the Romans in structures from the Late Iron Age, while the second incorporates the facilities from the
the late 1st c. BC, while the province of Roman and Late Antique periods. The thorough description of their morphological
Moesia was established there cа. AD characteristics made it possible to trace the development of the various kiln building
9-15. The province of Thrace, which techniques, from the time of their supposed appearance. In addition, some obser-
covered most of the territory of south- vations are made on the possible territorial and/or cultural background of the most
ern Bulgaria and a portion of northern influential practises, attested in the territory of Bulgaria.
Bulgaria, was founded on the territory
of the last Thracian kingdom ca. AD
45-46. However, the north-eastern re- Key words: ceramic kiln technology, Bulgaria, Late Iron Age, Roman and Late Antique
gion of today’s Bulgarian lands and the periods.
Romanian part of Dobrudzha were oc-
cupied permanently by the Romans only Introduction
in Flavian times (see Иванов 1999, 17-
28; Tacheva 2004, 53-56; Lozanov 2015, The two-chambered updraught ceramic kiln was introduced in the
78-81; Ботева 2017). territory of modern Bulgaria around the beginning of the Late Iron
3
The terms “Roman” and “Roman-
brought”, used in the description of the
Age1. During that period it was popular mostly in towns with Greek
various techniques in this article, refer inhabitants or in settlements under Greek influence. After the Roman
not to the ethnic origin of the potters, but conquest of Thrace, in the first half of the 1st century AD2, this type of
rather to the territorial and cultural back-
ground of the technology, developed dur-
kiln became the most popular among the ones, used for the firing of
ing the Roman period within the bound- clay-made objects (Harizanov 2016a).
aries of the Empire. So far there is no secure evidence for a continuation in the or-
4
Remains of kiln components (support,
perforated floor, etc.) were found inside a
ganisation of ceramic production from the Hellenistic through to the
pit from the 5th – first half of the 4th c. BC, Roman period in nowadays Bulgaria (Харизанов 2015, 48). However,
within a pit-field, located near the modern there are indications for persistent technological practices that re-
day village of Malko Tranovo, Chirpan mained in use during the second period. Furthermore, the Roman
Municipality (Божкова / Ников 2010,
215). Conversely, fragments of perforated occupation brought a significant number of new techniques, which
floors were also discovered in what ap- seem to have had a profound impact over the local traditions in kiln
peared to be waste pits. They were situated building3.
in the artisanal area of the Hellenistic town
of Cabyle, used during the second half of
the 3rd and the first quarter of the 2nd c. I. The Late Iron Age Kilns
BC, and were probably filled during the
repurposing of the space (Ханджийска /
There are less than 40 two-chambered ceramic kilns from the 6th cen-
Лозанов 2010, 250-254). tury BC to the first decades of the 1st century AD found in the territory
5

The lack of detailed published in- of modern Bulgaria. Parts of several structures have been discovered
formation impedes the total count and
thorough description of all structures
in the filling of pits (waste and deposit?4), while the ones found in situ
from that period. are approximately 325 and originate from 15 archaeological sites.
16 Alexander HARIZANOV

1. Structural components
The basic components of the two-chambered updraught kiln through-
out Antiquity were a combustion chamber, support for the perforated
floor (not necessary for all structures), a perforated oven floor, firing
6
The first such kiln was discovered in
the Greek emporium Pistiros and dated to
chamber, stoking channel and a stoke pit. Although not always found the second half of the 4th – the beginning
in situ, these elements were part of the design of all kilns, discovered in of the 3rd c. BC (Domaradzki 1998, 13;
the modern territory of Bulgaria (Харизанов in press). Попов 2002, 86-87; Танева 2009, 252-
255). The second kiln with quadrangu-
A. Combustion (lower) chamber lar combustion chamber was excavated
Most of the Late Iron Age kilns had circular combustion chambers, with in 2016 at the ancient settlement on the
Chrysosotira peninsula, near the mod-
only two known examples of quadrangular plan6 (table 1). The kiln com- ern day town of Chernomorets (Hristov
ponent in question was usually dug into the virgin soil and plastered with 2017, 51-66). Although it was dated to
clay. Few examples had their walls built of stones or mud bricks, bonded the 1st c. BC, the presented stratigraphic
and plastered with clay. The internal dimensions varied between 0.80 and section of the structure (Hristov 2017,
2.80 m, while the most common in size did not exceed 1.30 m (table 3). 56) provides another possibility for the
time of its operation, which could be set
B. Support for the perforated floor in the Late Roman or Early Byzantine
period. Nevertheless, due to the lack of
The various types of support for the perforated floor of the antique additional information, in this article it
kilns could be classified in two major groups – free standing and at- would be regarded as a Late Hellenistic
tached to the walls of the combustion chamber. During the pre-Ro- structure.
man period the first group comprised a central pillar, a central wall
7
The division between the two types
of support is based on their dimensions.
and two central walls7, while the second group consisted of a “tongue”
The components referred to as “central
wall and a “tongue” wall, which divided both the lower chamber and walls” have length at least twice their
the stoking channel in two parts (fig. 1). width.
There was also a third, “hybrid” variety of construction, which was 8
Milena Tonkova gives a detailed ac-
a combination of a free standing support with another one, attached to count of preserved in situ perforated
floors of this type for the kilns found
the walls of the lower chamber. During the Late Iron Age, it included a
at one of the largest production cen-
central pillar combined with a peripheral wall (ledge) or pilasters, with tres from the Early Hellenistic period
the perforated floor resting upon both. in Bulgaria, located near the village of
The basic building material was in most cases clay, sometimes Gorno Belevo (Тонкова 2002; Тонкова /
combined with shaping the component into the virgin soil. Other ma- Сидерис 2011). For a thorough descrip-
terials used for the construction of these structures were stones, mud tion of such oven floors in Britain see
Corder 1957; Swan 1984, 29-32.
bricks, bricks (only one case) and tiles (again only one case). 9
For the latter see Иванов 1970, 5;
Vagalinski 2011, 220; Varbanov et al.
C. Perforated floor (raised oven floor) 2016, 102.
Most of the preserved perforated floors were made of pre-fabricated ce- 10
In his ground work on the ceram-
ramic bars or segmented ceramic plates8 (figs. 2, 3). Some of the former ic centres near Butovo, Hotnitsa and
were also bonded and plastered with clay on the upper side (fig. 4). Three Pavlikeni, Bogdan Sultov refers to the
kilns with firing chamber placed partly
of the kilns had their oven floors made almost entirely of clay9 (fig. 1). or entirely underground as “dug”, while
the ones that had their upper chambers
D. Firing (upper) chamber built above the ground level are called
Few pre-Roman kilns had a firing chamber partially dug into the ground, “half-dug” (Sultov 1985, 38-39). When
while the majority were built above their ground level10. Of the former, describing the same two categories,
only the underground parts were found in situ and for that reason very Vivien Swan uses the terms “sunken”
and “semi-sunken” instead (Swan 1984,
little is known for the kiln superstructure during that period. 30-31). The latter will also be used in the
The available data indicates that at least some of the semi-sunk- present article.
en structures had temporary domes (Тонкова 2002, 150; Тонкова / 11
For a description of the English kiln
Сидерис 2013, 148-149), while the lower parts of the sunken ones had terminology see McWhirr 1979, 97-99;
permanent walls, lined with clay (see Vagalinski 2011, 220). Swan 1984, 29-32; Hasaki 2002, 77-90.
12
For the Latin terms, used in kiln de-
E. Stoking channel (flue)11 scription, see Cuomo di Caprio 1972,
388-404. Their Ancient Greek equiva-
The stoking channel or praefurnium12 was usually dug together with the lents were collected and presented by
combustion chamber, while its walls were normally only plastered with Eleni Hasaki (2002, 55-63).
 Bridging the Gap: Continuity and Innovation in Ceramic... 17

Table 1. Quadrangular vs. circular plan of the

combustion chamber of the Late Iron Age kilns
from the territory of Bulgaria

Fig. 1. Late Iron Age kiln with a tongue-shaped support and perforated floor made of clay from Rousse, photos
(after Varbanov et al. 2016, 115, fig. 6)
18 Alexander HARIZANOV

Fig. 2. Kiln with a perforated floor made

of pre-fabricated ceramic plates from
the Early Hellenistic site at Halka Bunar,
photo (after Тонкова 2002, 175, фиг. 6)

Fig. 4. Kiln with a double-layered perforated floor, made

of pre-fabricated ceramic bars and clay, from the Early Hellenistic site at Halka
Bunar, drawings (after Тонкова 2002, 166, табло III/3а-б)
Fig. 3. Kiln with a perforated floor
made of pre-fabricated ceramic plates
from the Early Hellenistic site at Halka
Bunar, drawing (after Тонкова 2002,
166, табло III/1)

clay. On several occasions they were built of mud bricks, stones, pre-fab-
ricated ceramic plates or bars, bonded and plastered with clay (fig. 5).
F. Stoke pit (stoking area)
Most of the stoking areas of the Late Iron Age kilns were designed as
simple shallow pits of various shapes, without any clay lining or other
building materials. A common working platform for two kilns was at-
tested at two pre-Roman sites (see Тонкова 2002, 151; Stoyanov et al.
2006, 15-16), while at least one of those pairs of kilns was probably
placed under protective cover (see Stoyanov et al. 2006, 16, fig. 20).
2. Production repertoire
The main purpose of the two-chambered kilns in the territory of mod-
ern Bulgaria during the Late Iron Age (covering respectively part of
 Bridging the Gap: Continuity and Innovation in Ceramic... 19

Fig. 5. Kiln with perforated floor and

stoking channel made of pre-fabricated
ceramic bars, photo (after Тонкова
2002, 176, фиг. 8)

the Archaic, and the Classical and Hellenistic periods), was the firing
13
Most of the hand-made pottery ves-
sels in Pre-Roman Thrace were fired of wheel-made pottery13 (for example Тонкова 2002, 152; Божкова
in clamps or bonfires, while some were 2006, 133; Балабанов 2010, 137-138), both in an oxidised and re-
probably produced in closed single- duced atmosphere (Тонкова 2002, 150-151).
chambered facilities (Ханджийска 2007, Production of building materials (and tiles in particular) was the
9-11). presumed use of the two quadrangular kilns from that time (Попов
14
The accepted margins for the end of
the Roman and Late Antique periods in
2002, 86-87; Hristov 2017, 51).
this article are the first years of the rule
of Emperor Diocletian (AD 284-305) for II. The kilns of the Roman and Late Antique periods14
the former and the beginning of the rule The Roman conquest of Thrace brought large masses of new popula-
of Emperor Heraclius (AD 610-641) for tion, especially to the territories north of the Haemus mons. Roman
the latter.
military units were stationed primarily along the Danubian limes,
15

The catalogue of the most recent
study on ceramic kilns from that pe- while civil immigrants settled in newly established villages and villas
riod comprised 270 structures (one both near the Danube and in the hinterland of the provinces of Moesia
pre-Roman) excavated in the territory and Thracia (Gerov 1988, 12-29; 43-67). It was in that kind of new set-
of Bulgaria (until 2015) and 19 found tlements where some of the earliest ceramic workshops were founded,
in modern Romania. In addition, it in-
probably in the last decades of the 1st c. AD (Харизанов 2015, 39, 43).
cluded basic information for 19 kilns
(one from the Late Hellenistic and 18 The available published data covers no less than 308 kilns15, stud-
from the Roman and Late Antique pe- ied at 72 sites in the territory of modern Bulgaria and dated to the
riods), studied during the 2015-2016 period between the second half of the 1st and the beginning of the 7th
archaeological campaigns in Bulgaria, century AD. Of those structures, 222 had circular or oval plan of the
together with data about at least 90 more combustion chamber, while the remaining 86 were overall quadran-
kilns with insufficient published infor-
mation (Харизанов in press). Only 21
gular16 (table 2).
of the latter, which could be attributed to
specific type of structure (with circular
1. Structural components
or quadrangular plan of the combustion There were no major changes in the general kiln design during the
chamber) and time period (Roman or centuries after Christ. However, a number of less or more important
Late Antique) were added to the other modifications appeared in the course of that period, which affected
289 (excluding those found in Romania)
and included in this article. Several kilns,
the size and capacity of the structures, together with the overall quality
excavated during the 2017 and 2018 ar- of their output.)
chaeological campaigns, were not taken
into account here, due to the lack of pri- A. Combustion chamber
mary publication. The lower chambers of the Roman kilns in the territory of Bulgaria
16
Under the term “quadrangular” are were also dug into the ground, while their walls were either only plas-
described kilns with combustion cham-
bers of several shapes, such as square,
tered with clay or built of building ceramics, stones or mud bricks,
rectangle, trapezium, rectangle with with clay bonding. The internal dimensions of the chambers of the
curved edges, etc. “circular” kilns varied from 0.50 m to 4.00 m, whereas those of the
20 Alexander HARIZANOV

“quadrangular” ones were between 0.65 m and 3.40 m in width, and

0.90 m and 4.70 m in length (tables 3, 4).
B. Support for the perforated floor
The three groups of supports for the perforated floor, attested in Late
Iron Age structures, were present during the Roman period as well.
The free standing group of supports comprised one, two or more than
three free-standing pillars; one or a couple of central walls and a modi-
fication of the structures with cross-walls, where the latter were de-
tached from the walls of the combustion chamber.
Table 2. Quadrangular vs. circular
The class of supports, that were attached to the walls of the lower plan of the combustion chamber of the
chamber, consisted of one or a couple of tongue walls; one or three Roman and Late Antique kilns from the
tongue walls, which divided both the chamber and the stoking chan- territory of Bulgaria
nel in two or four parts; a tongue wall with side pilasters; cross-walls,
which formed narrow arches above a central corridor; cross-arches
above a central corridor17; and a peripheral wall with covered central
corridor.
The “hybrid” group of supports was attested in a number of varie-
ties, which included a central pillar and side pilasters; a central pillar
and a peripheral wall; two central walls and side pilasters; two tongue
walls with two central pillars (figs. 6, 7).
A small but still notable group of kilns with circular combustion
chamber (usually up to 1.30 m in diameter) were designed without an

Table 3. Internal dimensions of the combustion chambers of the circular kilns from
the 6th c. BC – the beginning of the 7th c. AD

17
The division between the two types
of structures is based on the size and
position of the bases for the arches. The
bases of the ones with cross-walls were
larger in length, while those of the ones
with cross-arches were either only arch
steps, attached to the wall of the cham-
Table 4. Internal dimensions of the combustion chambers of the quadrangular kilns ber, or were built in within the latter (see
from the 6th c. BC – the beginning of the 7th c. AD Харизанов in press).
 Bridging the Gap: Continuity and Innovation in Ceramic... 21

Fig. 6. Kiln with a “hybrid” type of sup- Fig. 7. Kiln with a “hybrid” type of support for the perforated floor and a double
port for the perforated floor and a dou- praefurnium from the Roman production site at Nova Nadezhda, photo
ble praefurnium from the Roman pro- (after Harizanov 2016b, 585, fig. 7)
duction site at Nova Nadezhda, drawing
(after Harizanov 2016b, 585, fig. 6)

internal support; the perforated floor rested solely on the top of the
walls of the lower chamber.
The three groups of supports were made by two major techniques.
The first one included shaping the component into the virgin soil, with
the addition of a clay lining.
The second technique was characterised by the use of larger
amount of building materials, like full-sized or fragmented bricks,
tiles or tubes, and also stones, mud bricks and even ceramic vessels.
Building ceramics were the major construction material for the erec-
tion of arches and vaults, which were largely employed in kiln design
after the Roman conquest.
C. Perforated floor
In contrast to the first period, the perforated floors of the Roman and
Late Antique kilns could be categorised in two groups. The first one
comprised constructions with horizontal arrangement of the lower
side of the oven floor, built of full-sized imbrices (fig. 8), tubes, pre-
fabricated ceramic bars (fig. 9) and/or fragments of dolia rims, with or
without clay bonding and lining. These “ribs” were arranged radially18
(fig. 10) or in a “web-like” system19, while the gaps among them were
usually covered with fragments of building ceramics and clay.
The second group of perforated floors was attested in kilns with
arched or vaulted supporting construction. The floors of these structures
were formed of the upper side of the support, with the addition of hori-
zontal layers of building ceramics and clay. In the cases with brick arches,
the gaps among the latter were filled with vertically arranged fragments of
tiles and bricks, usually in a chequered or a zig-zag pattern (figs. 11, 12).
The vents (ventilation openings) of the circular perforated floors
18
Most of the kilns from the ceramic were arranged in concentric circles (fig. 13), while those of the quad-
centre in Varbovski Livadi locality near rangular ones were designed as series of rows (single or double), situ-
Pavlikeni had perforated floors built of ated in the gaps among the supporting pillars, walls or arches (fig. 11).
radially arranged imbrices or water pipes
(Sultov 1985, 39). D. Firing chamber
19
Hasaki (2002, 86) describes such per-
forated floors, found in Hellenistic and The upper chambers of the Roman and Late Antique ceramic kilns
Roman kilns in Greece. have been found in situ far more often than those of the Late Iron
22 Alexander HARIZANOV

Fig. 8. Late Roman kiln with perforated

floor made of imbrices from Cabyle, pho-
to (after Харизанов 2012, 355, фиг. 2)

Fig. 9. Kiln with perforated floor made

of pre-fabricated ceramic bars from
the Late Antique settlement on St.
Atanas cape, modern Byala, photos
(after Yotov / Harizanov 2018, 287, fig. 12)

Age structures. This fact is directly related to the now widely spread
practice of situating this kiln component partly or entirely below its
ground level (table 5). The chambers designed in this manner were
usually only lined with clay plaster. However, in cases of kilns of larg-
er dimensions, additional vertical walls of building ceramics or mud
bricks were attached to the sides of the kiln-pit (fig. 12).
The second group of kilns had their firing chambers placed en-
tirely above ground. The building materials varied in accordance with 20
B. Sultov describes remains of such
the exact shape of the superstructure and the duration of its use, and temporary covers, discovered in the
included full-sized or fragmented bricks and tiles, mud bricks, stones, production centres around Nicopolis ad
clay and probably also wood, straw and turf20. Istrum (Sultov 1985, 35-39).
21
Although none of the firing cham-
The shape of the upper chamber of the circular kilns had the bers of the kilns from Bulgaria were en-
form of a dome, a truncated cone or a cylinder, while that of the tirely preserved in situ, there are indica-
quadrangular ones was respectively close to a cube or was designed tions for such possibilities (Харизанов
as a barrel vault21. in press).
 Bridging the Gap: Continuity and Innovation in Ceramic... 23

Fig. 10. Kiln from the Late Antique settlement on St. Atanas cape, modern Byala, drawing and reconstruction of the perforated
floor (after Yotov / Harizanov 2018, 283-284, figs. 4-5)

D.1. Kiln superstructure

22
For the superstructure of the kilns The superstructure of the kilns of the Roman and Late Antique peri-
in Britain see Grimes 1930, 53-54; ods in the territory of Bulgaria was similar to the one of such construc-
Corder 1957, 14; Bryant 1977, 109-110; tions throughout the Empire22. The analysis of the available data pro-
Swan 1984, 34-35. For the ones in Italy vided four possibilities for the types of cover of the firing chamber of
– Cuomo di Caprio 1972, 401-402; in the structures in the studied territory: 1) a permanent cover with side
France – Le Ny 1988, 23; in Greece –
Hasaki 2002, 91-92; Raptis 2012, 39. For
opening/s; 2) permanent vertical walls and a temporary (flat) cover; 3)
general remarks, following experimental permanent walls in the shape of a truncated cone or dome and a tem-
firings, see Dawson / Kent 1985, 70-79. porary cover; 4) an entirely temporary cover (Харизанов in press).
24 Alexander HARIZANOV

Fig. 11. Rectangular kiln with perfo-

rated floor supported by double brick
arches, from the south end of St. Atanas
bay near Byala, photos (after Yotov /
Harizanov 2018, 288, fig. 13)

Table 5. Sunken vs. semi-sunken kilns

during the Roman and Late Antique
periods
 Bridging the Gap: Continuity and Innovation in Ceramic... 25

Fig. 12. Rectangular kiln with perfo-

rated floor supported by double brick
arches, from the south end of St. Atanas
bay near Byala, drawings (after Йотов /
Харизанов 2017, 495, табла IV-V)

E. Stoking channel
The stoking channel of the Roman kilns was formed by two tech-
niques. The first one employed the previously described method of
digging the component together with the combustion chamber. The
walls of those flues were either only lined with clay or were built of
full-sized or fragmented bricks and tiles, mud bricks or stones, bonded
and plastered with clay.
The second technique was used in structures that were placed al-
most entirely below the ground level. In that case, the stoking channel
was usually tunneled between the pits of the combustion chamber and
the stoke pit. Here the most common (and usually the only) building
material was clay.
F. Stoke pit
The shape, size and position of the stoke pit were directly related to
the position of the kiln itself. The stoking areas of the semi-sunken
kilns were situated close to the surface level and sometimes only the
parts near the stoking channel were located below ground. This type
of stoke pit usually had the advantage of a larger available space for the
potters’ operations.
The stoke pits of the sunken kilns were typically situated at a larger
depth than those of the semi-sunken ones. This variety of construc-
23
For an example of a sunken stoke pit tion benefited from the better insulation and wind protection, pro-
with one supporting wall and a pave-
vided by the surrounding terrain. However, most of the “sunken” stoke
ment, both made of bricks, see Милчев
et al. 2003. pits had smaller dimensions and were probably not that comfortable
24

According to B. Sultov (1985, 40) to work in.
many of the kilns at the centres near Building materials were seldom used in the construction of this
Butovo, Hotnitsa and Pavlikeni were kiln component. Brick, tile or stone walls (fig. 13) were sometimes
placed under protective covers, which
employed as reinforcement for one or more of the sides of the pit, usu-
covered their working platforms as well.
25
The kilns from the sites at Butovo, ally the ones placed near the kiln opening or beneath a slope. On rare
Hotnitsa and Pavlikeni were usually occasions brick or tile pavements were discovered23.
situated in groups of two, three or more Additionally, some stoke pits were placed under protective covers
around a common working platform (sheds), built of wood and covered with tiles (fig. 14)24.
(Sultov 1985, 40). Similar situation has The scarcely attested practice during the Late Iron Age of placing
also been attested at other Roman or Late
Antique sites (Антонова / Атанасов
two kilns around a common stoke pit was established as a customary
1979, 29; Милчев et al. 2003, 430-431; in the course of the Roman period, especially at sites with more than
Борисов 2013, 283). few production facilities that operated simultaneously25.
26 Alexander HARIZANOV

Fig. 13. Kilns with a common stoke

pit surrounded by stone wall from the
Roman production centre near Hotnitsa,
drawings (after Sultov 1985, table XII/1,
digital remastering A. Harizanov)

Fig. 14. Kiln from the Roman production centre near Hotnitsa, drawings and recon-
struction of the possible cover of the stoke pit (after Султов 1969, 13, фиг. 2)

2. Production repertoire
The Roman way of life brought a significant increase in the amount of
items produced by the ceramic workshops in the territory of modern 26

For ceramic production within
Bulgaria. Roman villas in Bulgaria see for example
Александров 1983, 53-61; Николов
During the Principate rural workshops, situated in villas, villages 1984, 29-41; Sultov 1985, 22-25. For ru-
and separate artisanal settlements, produced a large variety of ceramic ral workshops, situated within villages
artefacts, such as fine wares, coarse wares, oil lamps, terracotta figurines, or in separate artisanal settlements, see
flat-bottomed amphorae and pitchers. They were also major producers Sultov 1985, 18-21, 25-30; Иванова
of building materials (bricks, tiles, tubes and pipes of various sizes)26. 2009; Пенчева 2014; Harizanov 2016b.
For the fine ware production during the
The kilns of the ceramic ateliers, located in or near military set-
Principate in the territory between the
tlements (castrae, canabae, vici), were involved in the production of Danube and the Balkan mountains, see
building ceramics (although until now no stamped production has Иванов 2018.
been found in situ near its kiln), fine wares, oil lamps, coarse wares 27
For kilns, excavated in the area of the
and probably also terracotta figurines and flat-bottomed amphorae27. military camp of Novae, see Митова-
Джонова 1966; Димитров et al. 1970;
The output of the urban potteries comprised mainly fine wares,
Dyczek 2005; for production facili-
coarse wares and oil lamps, also flat-bottomed amphorae and terracot- ties from the settlements near the road
ta figurines. The production of building ceramics was scarcely attested station and military fort of Sostra –
during that period (Kalčev 1991; Harizanov 2018). Иванова 2003, 57-64; Hristov 2015.
 Bridging the Gap: Continuity and Innovation in Ceramic... 27

Table 6. Central pillar vs. other types of support Table 7. Central pillar vs. other types of support
for the perforated floor of the circular kilns for the perforated floor of the circular kilns
from the Late Iron Age from the Roman and Late Antique periods

28
For ceramic workshops of the Late The kilns of the Late Antique period were usually situated within
Roman and Early Byzantine period in
Bulgaria see Бацова-Костова 1969;
or in close proximity to towns, fortified and open settlements (within
Дамянов 1975; Minčev / Georgiev 1991; the latter until the end of the 4th – the first decades of the 5th c.) and
Борисов 2013; Боянов 2014; Biernacki were used for the production of smaller amount of items, in com-
/ Klenina 2015; Yotov / Harizanov 2018. parison to the Roman period28. The manufacture of fine wares and
For general information and analysis of terracotta figurines was reduced. Most common products of the
the ceramic workshops from the Roman
and Late Antique periods in the territory
workshops from that period were coarse wares and above all building
of Bulgaria see Харизанов 2015, 28-47; ceramics29.
Харизанов in press.
29
The latter was probably related to the III. Discussion
large scale building activities, initiated
by the provincial and/or state authori- As already stated above, the overall kiln design remained unchanged
ties during the 4th and the 6th centuries throughout the periods in question. However, several major new tech-
(Harizanov 2016a, 7). However, there niques appeared in the course of the Roman period.
is another possibility for the supposed
prevalence of the production of building A. Continuity
ceramics over that of domestic pottery.
The continuity in kiln building between the pre-Roman and Roman
More than half of the studied kilns from
the period between the beginning of the periods could be best traced in several of its features. The floor plan
4th and the beginning of the 7th c. AD had of the combustion chamber remained primarily circular and as a rule
quadrangular plan of the combustion this part of the installation was always situated beneath the ground
chamber, which was sometimes the only level.
reason for their direct association with The main type of support for the perforated floor of the circular
tile kilns, although in many cases there
was no evidence for the exact type of ce-
kilns was the single central pillar. It was found in at least 46 percent of
ramic products of the structures in ques- the Late Iron Age kilns and in no less than 60 percent of the structures
tion. It is also possible that during Late from the Roman and Late Antique periods (tables 6, 7)30. All other
Antiquity many kilns were designed to varieties of support, attested for the first period, were present during
support the heavier loads of fired build- the second one as well.
ing materials, but were also used for
the production of domestic pottery and
The perforated floors, arranged by pre-fabricated ceramic bars
transport containers. or plates usual for the pre-Roman period, remained in use through-
30
This type of kiln construction was out the centuries after Christ. However, this building technique
probably the most widespread variety was attested in a very small percentage of the kilns with preserved
throughout the Mediterranean and was oven floors31. A common practice during the Roman period and the
also present in almost every corner of
Late Antiquity was the use of full-sized ceramic tubes, pipes or im-
Europe during the centuries before and
after Christ (see Henning 1977, 193-198; brices instead, while the gaps among them were more often covered
Cuomo di Caprio 1979, 24-25; Duhamel with fragments of tiles or bricks, with or without an additional clay
1979, 73-74; Peacock 1982, 67-72; Swan lining.
28 Alexander HARIZANOV

B. Innovation
During the Roman period the two-chambered updraught kiln became
the most common type of firing structure, used in ceramic production
(Харизанов 2015, 48).
Although known from the pre-Roman period, the kilns with
quadrangular floor plan of the combustion chamber evolved in terms
of size, architectural forms and most of all in number during the cen-
turies after Christ. Furthermore, in the course of the Late Antique pe-
riod, the number of studied kilns of this variety equalised (or even
superseded?) the number of the circular structures32.
A major innovation in kiln building after the Roman conquest was
the introduction of building ceramics as a main construction mate-
rial33. Full-sized bricks, tiles and sometimes tubes were used for the
erection of the combustion and firing chambers, the stoking channel
and on rare occasions – for that of the stoke pit. However, their best
attested use was in the construction of the perforated floor and its sup-
port.
The second major innovation was exactly the use of Roman de-
veloped architectural forms, like brick arches and vaults, mostly for
the design of the kiln parts that were meant to bear the kiln’s load.
This type of construction provided opportunities for enlargement of
the firing chamber and more uniform distribution of the load’s weight,
whilst leaving large enough space for the fuel inside the combustion
chamber.
Broken and discarded bricks, tiles and tubes or pipes were also
largely employed in kiln building. They were used in the construction 1984, 9-34; Stirling 2006, 2405-2415;
of the chambers, the stoking channel, the perforated floor and its sup- Alaiba 2008, 68; Coll Conesa 2008, 116-
120; Gralak 2009; Le Dreff 2011, 50;
port. On the one hand, this practice was related to the availability of
Бобринский 1991, 189-209). It was like-
this material, especially in urban and sub-urban zones, or in produc- wise the most common kiln type for the
tion areas, where there was an abundance of it. On the other, the large territory of modern Greece during the
scale use of fragments of building ceramics could be explained by the Hellenistic, Roman and Late Antique pe-
physical qualities and overall firing behaviour of previously fired items riods (Hasaki 2002, 153-155; Raptis 2012,
of clay, which provided better thermal insulation and thus enabled the 39-40; Hasaki / Raptis 2016, 212-213).
31
See for example kilns 1 and 5 of the
achieving of higher and constant temperatures inside the kiln34. Late Antique settlement on St. Atanas
The last major technique that became widespread during the cape, near Byala (Yotov / Harizanov
Roman and Late Antique periods was the placing of the firing cham- 2018).
ber partly or entirely below its ground level. As demonstrated above, 32
Identical situation has been attested
this practice appeared in the course of the Late Iron Age, but its fore- for the Late Antique kilns in Greece (see
Raptis 2012, 40-41).
most development occurred after the Roman conquest, when it was 33
It was second only to clay in the
attested in kilns built almost entirely of clay, as well as in structures construction of the kilns from the 1st –
erected of building ceramics, mud bricks and stones. 6th centuries in the territory of modern
Bulgaria (Харизанов in press).
C. Influence 34
A continuation in that practise could
The current state of research indicates that the two-chambered up- be seen in the modern use of firebricks,
which have been employed in kiln build-
draught ceramic kiln appeared in the territory of modern Bulgaria un- ing due to the same reasons (see Hamer /
der Greek influence35. Most of the pre-Roman kilns of “Greek” design Hamer 1991, 185-186).
were semi-sunken structures with perforated floors made of pre-fab- 35
See Tonkova 2002, 151. For one of
ricated bars or plates and clay. The continuity of those Greek derived the earliest sites with studied ceramic
traditions in the craft in the territory of Bulgaria was best preserved kiln, excavated near the modern day vil-
lage of Koprivlen, see Божкова / Делев
on the Western Black sea coast, where it was still present during the
2002, 80-87. For the site at Debelt, where
Late Antique period. Furthermore, the widespread use of the circular another pottery kiln of early date was
kilns with central supporting pillar, as well as of the other types of associated with Greek influence, see
supporting constructions common for the Greek world, throughout Балабанов 2010.
 Bridging the Gap: Continuity and Innovation in Ceramic... 29

the 1st – 6th centuries in today’s Bulgarian lands, is a strong indication

of the continuous impact of potters of Greek origin36 on the ceramic
production in the northeast part of the Balkans37.
In the last centuries of the Late Iron Age, in the region of today’s
Northeast Bulgaria, another variety of kiln construction appeared. The
structures in question were built almost entirely of clay, while their up-
per chambers were situated partly or entirely below the ground level.
Their most distinctive feature was the support for the perforated floor
– an elongated tongue wall that divided the combustion chamber and
in some cases the stoking channel in two parts. Another kiln with a
similar supporting structure, but with a quadrangular plan of the com-
bustion chamber, was found in the central part of Northern Bulgaria
and dated to the Late Antique period38.
This type of construction is often associated with pre-Roman
Celtic traditions in the craft (Duhamel 1979, 59-62; Peacock 1982,
69-72), but has also been found in settlements of Geto-Dacian ori-
gin (Matei 2007). In the centuries after Christ it was found again in
Dacian context (Raţiu 2009, 173-177), in settlements of the Sântana de
Mureş-Černjachov culture (Körösfői et al. 2011), as well as in the ter-
ritory of modern Spain (Coll Conesa 2008, 114-125) and in that of the
Roman province of Noricum (Risy 1994, 22-38). For now it seems safe
to assume that this type of construction was brought to the territory of
Bulgaria from the north or the north-west.
Similar territorial background could be assigned to the practice of
situating most of the kiln components below the ground level, com-
bined with the use of smaller amount of building materials (usually
36
These potters most probably origi- only clay). This technique was common for parts of Western, Central
nated both from the territory of mod- and Eastern Europe (see Swan 1984, 29-38; Gralak 2009; Alaiba 2008,
ern Greece and that of Asia Minor (for 62-64), whereas it was unfamiliar to the Mediterranean potters (for
the latter, see Sultov 1985, 102-103). For
general information on the presence of
example Delcroix / Huot 1972; Hasaki 2002, 71-101; Stirling 2006,
Eastern and most of all Asia Minor im- 2405-2415).
migrants (including craftsmen) in the The development of this practice in the modern Bulgarian lands
territory between the Danube and the could be attributed to potters, originating from the above-mentioned
Haemus mons, see Геров 1954, 339-346. regions of Europe.
37

Another indication is the notable
presence in the territory of Bulgaria of
As it was already stated, the major Roman innovation in kiln
kilns with quadrangular combustion building in the modern Bulgarian territory was the use of brick arch-
chamber and a central pillar (14 struc- es and vaults, together with the enlargement of the overall size of the
tures from 9 sites – see Харизанов in kilns, many of which had quadrangular plan of the combustion cham-
press), which were most common for the ber. Similar phenomenon occurred in many other parts of the former
Greek region of Attica in the Roman and
Roman Empire, after their conquest39. The innovation was most prob-
Late Roman periods (see Hasaki 2002,
168-169; Raptis 2012, 40). ably related to the ceramic production by (or for) the Roman military
38
See Робов 1997. For a detailed analy- contingents, who were usually some of the first to bring the technol-
sis on the kiln and the possibilities for its ogy to the newly conquered territories (see Peacock 1982, 136-137).
precise dating see Харизанов in press. It could be best seen in investigated military workshops, dated to
39

For example in the territory of
the first centuries after Christ, similar to the ones at Holt, Holdeurn,
Roman Britain (Swan 1984, 82-112), also
in those of nowadays France (Duhamel Dormagen, Vransko, Jerusalem etc., where almost all of the used kilns
1979, 63-74; Le Ny 1988) and Austria had perforated floors supported by cross-walls or cross-arches40. Of
(Risy 1994). course, it could be also associated with civil immigrant craftsmen, who
40
See in order of listing Grimes 1930; followed the military and established their workshops in the new ter-
Holwerda / Braat 1946; Peacock 1982, ritories shortly after their conquest.
139-140; Lazar 1999; Arubas / Goldfus
1995, 95-107. For a complex analysis of
During the Roman and Late Antique periods there was an increase
the Roman military ceramic production in the number of kilns, where mud bricks were used as a major con-
see Peacock 1982, 136-150. struction material, especially in the erection of quadrangular produc-
30 Alexander HARIZANOV

tion facilities. This practice might have been at least partly influenced
by immigrants from the Middle East or Asia Minor41, whose settling in
the territories of modern Bulgaria has been attested in the epigraphic
records (for the latter, see Геров 1954, 339-346).
A certain number of kilns, originating from a restricted number of
Roman sites, show distinctive features, common for the western prac-
tices in kiln building, combined with techniques, used in the Greek
world. Another small number of structures have no exact parallel
in other parts of Europe (figs. 6, 7; Харизанов in press)42. The two
groups of production facilities seem to have evolved in the territory
of Bulgaria and could be regarded as evidence for the development of
local practices in kiln technology.
Tracing the territorial or cultural background of a certain techno-
logical practice only through analysis of the used architectural forms
and building materials could lead to unreliable results43. That is why
the above listed facts and suggestions on the background of the various
influences, attested in kiln construction in the territory of Bulgaria,
should be considered most of all as a guideline for future research on 41
Most of the kilns from the Prehistory
the topic. Such research must comprise not only the technological to the Hellenistic period in the East were
issues of the kiln design but also the typological and morphological built of mud bricks (see Delcroix / Huot
characteristics of the local ceramic products, together with the attested 1972). The same material was used in the
types of imported pottery items and the epigraphic evidence for the erection of some of the kilns of the Legio
presence of itinerant craftsmen. X Fretensis at Jerusalem (Arubas / Golfus
1995, 100-104), the ones found at Dura-
It should also be emphasised that the numbers of Late Iron Age Europos (Allara 1992, 110), etc.
and Roman kilns, included in this contribution (tables 3, 4), together 42
As an example of the first type of fa-
with the presented ratio of circular vs. quadrangular installations (ta- cility, could be pointed out the support-
bles 1, 2), reflect only the current state of research and are not to be ing structure, comprising side pilasters
viewed as constants. Nevertheless, they are invaluable for outlining the and a central pillar or two central walls.
The side pilasters were common for the
general trends in the use of two-chambered ceramic kilns in today’s
western traditions in kiln building since
Bulgarian lands throughout the studied chronological period. the Late Iron Age (Le Dreff 2011, 50-59).
During the Roman period they were
Conclusion usually combined with central or tongue
As it was already mentioned, for now there is no secure proof for a walls, or were left without additional
continuation in the organisation of ceramic production from the support (see Cuomo di Caprio 1979, 26;
Swan 1984, 31, fig. 3; Le Ny 1988, 39,
pre-Roman through the Roman periods in the territory of modern
42; Risy 1994, Planbeilage I; Wirtz 1998,
Bulgaria, given that even a topographical continuity has been scarcely 180-182; Coll Conesa 2008, 119, fig. 3).
attested for the time under discussion. However, the presented data for The central supporting pillar, as it was al-
the technological side of the question in mind, and that of kiln build- ready stated, was a distinctive feature for
ing in particular, allows us to “bridge” some of the “gaps” in this issue. the kiln building in the Mediterranean
and especially in the territory of modern
The Greek-derived technological practices in the potters’ craft
Greece.
were fundamental for its development during the Late Iron Age in the A kiln construction with presumably
territory of modern Bulgaria. They were still evident during the centu- local origin comprised two tongue walls
ries after Christ, albeit with some modifications, and their continuous and two central walls, and had a double
impact is not to be overlooked. praefurnium (see Harizanov 2016b, 582).
Still, it is the Roman-brought innovations in kiln technology that
43
This is especially valid for the study
of the ceramic kilns from the Roman
should be regarded as the most essential for the significant advance- period, when the unification of tech-
ment in the production of ceramic artefacts, attested in the course of nological practices throughout the
the Roman and Late Antique periods in this corner of the Empire. Mediterranean reached its peak.
 Bridging the Gap: Continuity and Innovation in Ceramic... 31

Bibliography / Цветкова, Ю. (eds.). ΚΡΑΤΙΣΤΟΣ. Проучване на римската крайпътна

Сборник в чест на професор Петър станция и кастел на пътя Ескус
Александров, Г. 1983. Антична вила Делев. София. 157-168. – Филипополис, том I. Велико
№1 край Михайловград. – Известия Търново. 57-64.
на музеите в Северозападна Боянов, И. 2014. Селище и керамичен
България 8, 37-79. производствен център от IV век в Йотов, В. / Харизанов, А. 2017. Пещ
землището на село Големо Бучино, за строителна керамика (VI век) на
Антонова, В. / Атанасов, Г. област Перник. – Известия на южния бряг на залива Св. Атанас
1979. Пещи за битова керамика Регионален исторически музей край Бяла, Варненско. – Добруджа
в землището на с. Пет могили, Перник 2, 77-95. 32, 487-496.
Шуменско. – Годишник на музеите в
Северна България 5, 28-36. Геров, Б. 1954. Романизмът между Милчев, А. / Койчева, К. / Димитров,
Дунава и Балкана от Хадриан П. 2003. Производствен център за
Балабанов, П. 2010. За стоковото до Константин Велики. Част строителна керамика в Долна Мизия.
производство на керамични съдове II. Продължение. – Годишник In: Сборник в чест на д-р Петър
в древна Тракия. In: Югоизточна на Софийския университет, Горбанов (= Studia Archaeologica 1).
България през II-I хилядолетие пр. Философски факултет 48, 1952-1953, София. 430-450.
Хр. Варна. 134-141. 307-413.
Митова-Джонова, Д. 1966. Пещи
Бацова-Костова, E. 1969. Антични Дамянов, С. 1975. Пещ за строителна за керамика и керемиди от Нове. –
пещи за строителна керамика в керамика до късноантичния град Археология 8/1, 38-45.
Сливен. – Музеи и паметници на при с. Войвода, Шуменско. –
културата 4, 7-9. Годишник на музеите в Северна Николов, Д. 1984. Римската вила при
България 1, 11-17. Чаталка, Старозагорско. – Разкопки
Бобринский, A. 1991. Гончарные и проучвания 11, 5-73.
мастерские и горны Восточной Димитров, Д. П. / Чичикова, М. /
Европы. Москва. Султов, Б. / Димитрова, А. 1970. Пенчева, Е. 2014. Римското селище в
Археологические раскопки в м. „Дана бунар“ (с. Георги Добрево,
Божкова, А. 2006. Пещ за битова восточном секторе Нове в 1966 г. общ. Любимец) в контекста на
керамика от средата на І хил. пр. Хр. – Известия на Археологическия сходните комплекси от Тракия.
при Копривлен, Гоцеделчевско. – институт 32, 55-71. Автореферат на дисертация. София.
Археология 47/1-4, 130-138.
Иванов, Д. 1970. Тракийски могилни Попов, Х. 2002. Урбанизация във
Божкова, А. / Делев, П. 2002. некрополи край с. Сваленик, вътрешните райони на Тракия и
Копривлен и културната общност Русенско. – Известия на Народния Илирия през VІ – І в. пр. Хр. София.
на югозападна Тракия през VII-V музей Русе 4, 3-12.
век пр. Хр. – Известия на Народния Робов, М. 1997. Керамична пещ
музей Бургас 4, 80-96. Иванов, Р. 1999. Долнодунавската в западното подножие на хълма
отбранителна система между „Момина Крепост“ във Велико
Божкова, А. / Ников, К. 2010. Дортикум и Дуросторум от Август Търново. – Известия на Исторически
Археологическо проучване на до Маврикий. София. музей Велико Търново 12, 136-142.
комплекс от ями в землището на с.
Малко Тръново, община Чирпан – Иванов, С. 2018. Типология и Султов, Б. 1969. Новооткрит
югозападен сектор. Предварителни хронология на червенолаковата керамичен център при с. Хотница от
бележки. In: Югоизточна България керамика от производствените римската и старобългарската епоха.
през II-I хилядолетие пр. Хр. Варна. комплекси между Дунав и – Археология 11/4, 12-24.
213-220. Балкана (II-III в.). Автореферат на
дисертация. Велико Търново. Танева, В. 2009. Емпорион Пистирос.
Борисов, Б. 2013. Късноантичната Археологически разкопки в квадрат
грънчарска работилница край Иванова, Б. 2009. Пещ за керемиди Ж 5. – Археологически открития и
с. Караново, Новозагорско, и от римския викус край Горско разкопки през 2008 г., 252-255.
намерената в нея керамика. Абланово, Търговищко. –
– Известия на Националния Археология 50/1-2, 114-119. Тонкова, М. 2002. Новооткрит
археологически институт 41, тракийски център от
281-336. Иванова, M. 2003. Пещ за битова ранноелинистическата епоха при
керамика край с. Лешница, община извора Халка Бунар в землището
Ботева, Д. 2017. Създаването на Ловеч. In: Христов, И. / Китов, Г. / на село Горно Белево. Проучвания
провинция Тракия. In: Попов, Хр. Иванова, М. / Бинев, М. СОСТРА. през 2000 и 2001 г. – Годишник на
32 Alexander HARIZANOV

Археологическия институт с музей 2, Arubas, B. / Goldfus, H. 1995. The e laterizi nell’area italiana, dalla preisto-
148-196. kilnworks of the Tenth Legion Fretensis. ria a tutta l’epoca romana. – Sibrium 11,
In: Humphrey, J. H. (ed.). The Roman 371-464.
Тонкова, М. / Сидерис, А. 2012. and Byzantine Near East: Some recent
Археологически проучвания на archaeological research. (= Journal of Dawson, D. / Kent, O. 1985. Kiln super-
многослоен обект от Късния неолит Roman Archaeology Suppl. 14). Ann structures – the Bickley experiments.
и І хил. пр. Хр. при извора Халка Arbor. 95-107, 273. – Bulletin of the Experimental Firing
Бунар, с. Горно Белево, община Братя Group 3, 70-79.
Даскалови, Старозагорска област. – Baralis, А. / Panayotova, Kr. /
Археологически открития и разкопки Bogdanova, T. / Gyuzelev, M. / Nedev, Delcroix, G. / Huot, J. 1972. Les Fours
през 2012 г., 147-149. D. / Gospodinov, K. 2016. Apollonia dits “de potier” dans l’Orient ancien. –
Pontica (Sozopol, Bulgaria): the results Syria 49/1-2, 35-95.
Тонкова, М. / Сидерис, А. 2011. of the Franco-Bulgarian archaeologi-
Археологически проучвания на cal mission. In: Manoledakis, M. (ed.). Domaradzki, M. 1998. An interim
многослоен обект от късния неолит The Black Sea in the Light of New report on investigations at Vetren -
и І хил. пр. Хр. при извора Халка Archaeological Data and Theoretical Pistiros 1995-1998. In: Bouzek, J. /
Бунар, с. Горно Белево, община Братя Approaches. Proceedings of the 2nd Domaradzka, L. / Archibald, Z. (eds.).
Даскалови, Старозагорска област. International Workshop on the Black Pistiros 2. Prague. 11-31.
In: Тонкова, М. (ред.). Трако-римски Sea in Antiquity Held In Thessaloniki,
династичен център в района на 18-20 September 2015. Archaeopress Duhamel, P. 1979. Morphologie et evo-
Чирпанските възвишения. София. Archaeology. Oxford. 153-179. lution des fours ceramiques en Europe
80-92. occidentale. Protohistoire, Monde
Biernacki, A. / Klenina, E. 2015. Early- Celtique et Gaule Romaine. – Acta
Ханджийска, В. 2007. Tрадиции Byzantine Pottery Workshops in Novae Praehistorica et Archaeologica 9-10,
и тенденции в производството (Moesia II, Bulgaria): Technological, 49-76.
на керамика на ръка в днешна Economic and Historical Aspects. In:
Югоизточна България през VI–I в. Thuillier, F. / Louis, E. (eds.). Tourner Dyczek, P. 2005. A New Pottery
пр. хр. Автореферат на дисертация. autour du pot… Les ateliers de potiers Manufacturing Center at Novae. – Rei
София. médiévaux du Ve au XIIe siècle dans Cretariae Romanae Fautorum Acta 39,
l’espace européen. Publications du 301-306.
Ханджийска, В. / Лозанов, И. 2010. Centre de Recherches Archéologiques
Кабиле през елинистическата et Historiques Médiévales (CRAHM). Gerov, B. 1988. Landownership in
епоха (по данни от последните Brepols. 373-381. Roman Thrace and Moesia (1st – 3rd
археологически проучвания). In: centuries). Amsterdam.
Югоизточна България през II-I хил. Bryant, G. 1977. Experimental
пр. Хр. Варна. 247-270. Kiln Firings at Barton-an-Humber, Gralak, T. 2009. Piec garncarski ze
S. Humberside 1971. – Medieval stanowiska 5 w Polwicy, Pow. Oławski,
Харизанов, А. in press. Пещи за Archaeology 21, 106-123. na tle podobnych obiektów ze Śląska. –
керамика в днешните български земи Śląskie Sprawozdania Archeologiczne
през I – VI век (= Дисертации 15). Coll Conesa, J. 2008. Hornos romanos 41, 209-230.
София. en España. Aspectos de morfología y
tecnología. In: Bernal Casasola, D., A. Grimes, W. F. 1930. Holt, Denbighshire.
Харизанов, А. 2015. Пещи за керамика Ribera i Lacomba (eds.). Ceramicas The Works-Depot of the Twentieth
в днешните български земи през I – Hispanorromanas. Un estado de la cues- Legion at Castle Lyons (= Y
VI век. Автореферат на дисертация. tion. Cádiz. 113-125. Cymmrodor XLI). London.
София.
Corder, P. 1957. The structure of Hamer, F. / Hamer, G. 1991. The Potter’s
Харизанов, А. 2012. Пещ за керамика Romano-British pottery kilns. – The Dictionary of Materials and Techniques.
от Кабиле. In: Collegium Historicum, Archaeological Journal 114, 10-27. 3rd Edition. London / Philadelphia.
vol. 2. Sofia. 351-357.
Cuomo di Caprio, N. 1979. Updraught Harizanov, A. 2018. Urban Ceramic
Alaiba, R. 2008. Meşteşugul olăriei în pottery kilns and tile kilns in Italy in Workshops in the Province of
secolele II-V în spaţiul dintre Carpaţi şi pre-Roman and Roman times. – Acta Thrace during the Principate: The
Prut. Bucureşti. Praehistorica et Archaeologica 9-10, Archaeological Evidence from Bulgaria.
23-31. In: Vagalinski, L. / Raycheva, M. /
Allara, A. 1992. L’îlot des potiers et Boteva, D. / Sharankov, N. (eds.).
les fours à Doura-Europos. Étude Cuomo di Caprio, N. 1972. Proposta di Proceedings of the 1st International
préliminaire. – Syria 69/1-2, 101-120. clasificazione delle fornaci per ceramica Roman and Late Antique Thrace
 Bridging the Gap: Continuity and Innovation in Ceramic... 33

Conference “Cities, Territories and – Philippopolis road (II–III century). Jahrhundert. – Rei Cretariae Romanae
Identities” (Plovdiv, 3rd – 7th October Sofia. Fautorum Acta 29-30, 223-244.
2016) (= Bulletin of the National
Archaeological Institute 44). Sofia. 299- Kalčev, K. 1991. Zur Herstellung der Peacock, D. P. S. 1982. Pottery in the
308. Antiken Keramik in Augusta Trajana / Roman World: an ethnoarchaeological
Stara Zagora. – Rei Cretariae Romanae approach. London.
Harizanov, A. 2016а. Ceramic kilns in Fautorum Acta 29-30, 245-273.
the territory of Bulgaria from 1st to 6th Raptis, K. 2012. Early Christian and
c. AD. Dissertation abstract. – Bulletin Körösfői, Z. / Sófalvi, A. / Nyárádi, Byzantine ceramic production work-
Instrumentum 44, 7. Z. 2011. Töpferöfen in der Siedlung shops in Greece: typology and distribu-
der Sântana de Mureş-Černjachov- tion. In: Gelichi, S. (ed.).
Harizanov, А. 2016b. A New Centre Kultur in Odorheiu Secuiesc-Alsólok, Atti del IX Congresso internazi-
for Ceramic Production from the Siebenbürgen. In: Bemmann, J. / onale sulla ceramica medievale nel
Roman Province of Thrace (2nd – 3rd c.). Hegewisch, M. / Meyer, M. / Schmauder, Mediterraneo. Venezia, Scuola Grande
Preliminary Results. – Rei Cretariae M. (eds.). Drehscheibentöpferei im dei Carmini, Auditorium Santa
Romanae Fautorum Acta 44, Barbaricum. Technologietransfer Margherita, 23-27 novembre 2009.
581-594. und Professionalisierung eines Florence. 38-43.
Handwerksam Rande des Römischen
Hasaki, E. 2002. Ceramic Kilns in Imperiums. Akten der Internationalen Raţiu, A. 2009. Pottery kilns from
Ancient Greece: Technology and Tagung in Bonn vom 11. bis 14. Juni North-Western Transylvania from the
Organization of Ceramic Workshops. 2009. Bonn. 445-456. 2nd – 4th centuries AD. In: Țentea, O.
Doctorate of philosophy (Ph.D.). / Opriş, I. C. (eds.). Near and Beyond
Cincinnati. Lazar, I. 1999. The Roman Tile Factory the Roman Frontiers. Proceedings of
at Vransko near Celeia (Noricum). a Colloquim held in Târgovişte, 16-17
Hasaki, E. / Raptis, K. 2016. Roman and In: Demaine, M. R. / Taylor, R. M. October 2008. Bucharest. 165-186.
Byzantine ceramic kilns in Greece (1st (eds.). Life of the Average Roman.
– 15th c. CE): Continuities and changes A Symposium. White Bear Lake, Risy, R. A. 1994. Römerzeitliche
in kiln typology and spatial organisa- Minnesota. 23-33. Brennöfen in Noricum. Diplomarbeit
tion of the production. In: Cucuzza, N. der Universität Wien.
/ Giannattasio, B. / Pallecchi, S. (eds.). Le Dreff, T. 2011. Fours et ateliers de po-
Archeologia delle produzioni cerami- tiers au second âge du Fer dans l’isthme Stirling, L. 2006. Aspects of Punic and
che nel mondo antico. Spazi, prodotti, gaulois. – Aquitania 27, 19-59. Roman Kiln Design in North Africa.
strumenti e tecniche. Atti del convegno In: Akerraz, A. / Ruggeri, P. / Siraj, A. /
(Genova, 1-2 dicembre 2014). Genova. Le Ny, F. 1988. Les fours de tuiliers gal- Vismara, C. (eds.). L’Africa Romana. Atti
209-229. lo-romains (= Documents d’Archéologie del XVI convegno di studio Rabat, 15-
Française 12). Paris. 19 dicembre 2004, v. 4. 2405-2416.
Henning, J. 1977.
Entwicklungstendenzen der Lozanov, I. 2015. Roman Thrace. In: Stoyanov, T. / Mihaylova, Zh. / Nikov, K.
Keramikproduktion an der mittleren Valeva, J. / Nankov, E. / Graninger, D. / Nikolaeva, M. / Stoyanova, D. 2006 The
und unteren Donau im 1. Jahrtausend (eds.). A Companion to Ancient Thrace. Getic Capital in Sboryanovo, 20 years of
u. Z. – Zeitschrift für Archäologie 11, Wiley Blackwell. 75-90. investigations. Sofia.
181-206.
Matei, S. 2007. The morphology and Sultov, B. 1985. Ceramic produc-
Holwerda, J. H. / Braat, W. C. 1946. De typology of kilns for firing ceramic in tion on the territory of Nicopolis ad
Houldeurn bij Berg en Dal. Centrum the Geto-Dacian Classical period (2nd Istrum. (= Годишник на Софийския
van pannebakkerij en aardewerkindust- century BC – 1st century AD). – Istros университет, Исторически факултет
rie in den Romeinschen tijd. Oudheidk. 14, 279-296. 76, 2, 1983). София.
Meded. Suppl. 26. Leiden.
McWhirr, A. 1979. Roman Tile-kilns in Swan, V. G. 1984. The Pottery Kilns оf
Hristov, I. 2017. ΧΡΥΣΟΣΩΤΉΡΑ 2. Britain. In: McWhirr, A. (ed.). Roman Roman Britain (= Royal Commission on
An Early Byzantine fortress on the Brick and Tile. Studies in Manufacture, Historical Monuments, Supplementary
peninsula of Hrisosotira at the town of Distribution and Use in the Western Series 5). London.
Chernomorets, 5th – 7th century. Empire. BAR International series 68,
Sofia. 97-189. Tacheva, M. 2004. Über die römischen
Provinzen Moesia Inferior und Thracia
Hristov, I. 2015. PRAETORIUM Minčev, A. / Georgiev, P. 1991. (I – III Jh.). In: Ivanov, R. (ed.).
SOSTRA A study of the Roman Marcianopolis – ein Zentrum Archaeology of the Bulgarian Lands.
roadside station of Sostra on Oescus der Keramikproduktion im 2. - 6. Sofia. 49-76.
34 Alexander HARIZANOV

Vagalinski, L. 2011. A new Late La Tène Varbanov, V. / Rusev, N. / Todorova, S. Vergleichende Studie zu Töpferöfen für
pottery kiln with a bread oven on the / Yordanova, R. 2016. Late Hellenistic Gebrauchskeramik). Bonn.
lower Danube (northern Bulgaria). kilns found on the territory of Rousse,
In: Guštin, M. / Jevtić, M. (eds.). The Bulgaria. – Banatica 26, 1, 101-117. Yotov, V. / Harizanov, A. 2018. Ceramic
Eastern Celts. The Communities Kilns in the Late Antique Town on
between the Alps and the Black Sea. Wirtz, R. 1998. Die römische Töpferei Sveti Atanas Cape (Modern Day Byala,
Koper / Beograd. 219-226. Bastion Sterntor/St. Maria in Bonn (= Bulgaria). – Pontica 51, 273-290.

Приемственост
и нововъведения в технологията
на пещите за керамика от VI в.
пр. Хр. до началото на VII в.
сл. Хр. на територията
на България
Александър Харизанов

(резюме)
Статията изследва появата, развитието и разпространението на
двукамерните пещи за керамика в днешните български земи в
периода от VI век пр. Хр. до началото на VII век сл. Хр. От една
страна, изследването има за цел да проучи възможността за на-
личие на континюитет в традициите на изграждане и употреба
на такива съоръжения от късножелезния период в римския и в
Късната античност, от друга – да установи възможни новости в
техниките на градеж и оформяне на този тип съоръжения, настъ-
пили в територията на днешна България след попадането на ре-
гиона под римска власт, както и да определи мащабите на техния
ефект върху местните традиции в производството на изделия от
глина.
Базата данни на настоящото изследване включва 340 дву-
камерни пещи с вертикална тяга, проучени на територията на
днешна България. Съоръженията са разделени на две големи
групи, съобразно тяхната датировка. Първата група включва
късножелезни структури, а втората – такива от римския период
и Късната античност.
Детайлният анализ на морфологичните характеристики на
двете групи съоръжения позволи установяване на използваните
в техния градеж строителни техники и материали. В допълнение
са изказани и предположения за първоизточниците и пътищата
на проникване на регистрираните технологични практики.
I. Двукамерни пещи от късножелезния период
Античните двукамерни пещи с вертикална тяга се състоят от ня-
колко взаимосвързани конструктивни елемента, а именно долна
Bridging the Gap: Continuity and Innovation in Ceramic... 35

камера, подпора (понякога липсва при по-малки съоръжения),

скара, горна камера, канал и предпещна яма. Свързаната с тяхната
употреба технология изглежда навлиза на територията на днешна
България около или след началото на късножелезния период.
1. Конструктивни елементи и особености
Части от конструкции на няколко пещи са открити депонирани в
ями (култови и отпадъчни?) в границите на няколко предримски
обекта. Запазените на място съоръжения от същия период са не
по-малко от 32 и произхождат от 15 обекта. 30 имат кръгла или
овална в план долна камера, докато останалите 2 са с четириъгъ-
лен план на същата (таблици 1 и 3).
Подпорите на скарата са свободно стоящи или свързани със
стените на долната камера (фиг. 1). Разпознава се и трети, „хиб-
риден“ вариант на носеща скарата конструкция, включващ ком-
бинация от свободно стояща и свързана със стената на долната
камера подпори.
Скарата на предримските съоръжения е най-често оформена
от предварително изпечени керамични сегменти, с или без глине-
на спойка и обмазка (фиг. 2-4). По-редки са случаите на оформя-
не само от глина или от глина и дърво (фиг. 1).
Горните камери на пещите от този период се изграждат ос-
новно над съвременната им повърхност. Единични са случаите
на частично или почти изцяло вкопани такива компоненти.
Каналите на съоръженията се вкопават заедно с долните им
камери. Обикновено стените им са само измазани с глина, по-
редки са случаи на градеж с други строителни материали (фиг. 5).
Повечето предпещни ями са оформени като плитки вкопа-
вания, без укрепване или оформяне от строителни материали.
Единични са случаите на обща работна площадка за две съоръ-
жения. Над някои от тези компоненти са били изградени навеси,
за защита от неблагоприятни климатични условия.
2. Предназначение
Основното предназначение на предримските пещи е изпичане на
формувани на грънчарски кръг съдове за бита, като при някои е
възможно термичният процес да е протичал в редукционна ат-
мосфера. За двете четириъгълни в план съоръжения се предпо-
лага експлоатация при производство на строителна керамика.
II. Пещи от римския период и Късната античност
Наличната публикувана информация включва данни за не по-
малко от 308 съоръжения, разкрити в рамките на 72 обекта, да-
тирани в периода между втората половина на I и началото на VII
век. 222 от тези пещи има кръгъл или овален план на долната
камера, докато останалите 86 са с четириъгълен или близък до
четириъгълник план на същата (таблица 2).
1. Конструктивни елементи и особености
В периода след установяване на римската власт на територията
на днешна България се наблюдават редица нововъведения в из-
ползваните строителни техники и градежни материали, които се
отразяват на големината (таблици 3 и 4) и производствения ка-
36 Alexander HARIZANOV

пацитет на съоръженията, както и върху качеството на тяхната

продукция.
Долните камери на римските пещи са също винаги вкопани
в околния терен. Стените им се оформят от глинена обмазка или
строителни материали с глинена спойка.
Трите вида подпори, установени за предримските пещи
(свободно стоящи, свързани със стените на долната камера и
„хибридни“), са застъпени и при тези от римския и късноанти-
чен период, като по това време се наблюдава много по-голямо
разнообразие в тяхната форма и градеж (фиг. 6 и 7). При една
неголяма група съоръжения, с малки размери, липсват подпори
за скарата.
Скарите на римските пещи се оформят по два начина. Първият
е наличен при конструкции, в чийто градеж се използват предвари-
телно изпечени керамични сегменти (фиг. 9), цели имбрици (фиг.
8), керамични тръби, фрагменти от устия на питоси. Елементите
се подреждат радиално (фиг. 10) или мрежовидно, а празнините
между тях често се запълват с фрагменти строителна керамика, с
или без глинена спойка. В градежа на втория вид скари се използ-
ват арки или сводови конструкции, които се оформят върху гор-
ната част на подпорите, от тухли, керемиди или кирпичи, и глина.
В случаите на употреба на арки, празнините между последните се
запълват от вертикално поставени фрагменти строителна керами-
ка и глина (фиг. 11 и 12).
Горните камери на пещите от този период се откриват частич-
но или почти изцяло запазени много по-често от тези от предход-
ния хронологичен отрязък. Този факт е пряко свързан с широко
разпространената по това време практика на частично или ця-
лостно вкопаване на този конструктивен елемент в околния терен
(таблица 5). Оформените по този начин камери се измазват отвъ-
тре с глина, като при съоръжения с по-големи размери се наблю-
дават и варианти на по-масивен градеж от строителна керамика
или кирпичи, и глина (фиг. 12). При конструкциите с надземни
горни камери, използваните материали включват строителна ке-
рамика, кирпичи, камък, глина, а вероятно също пръст, слама и
дърво.
Могат да се предположат общо четири варианта за завършек
във височина за пещите от този период: 1) постоянно покритие
със страничен отвор/и; 2) постоянни вертикални стени и вре-
менно хоризонтално покритие; 3) постоянна част с форма на
пресечен конус или купол и временно покритие; 4) изцяло вре-
менно покритие.
Каналите на римските съоръжения се оформят по два начина.
Първият включва вкопаване заедно с долната камера, като стени-
те се оформят като продължение на тези на последната. Вторият
вариант е наличен при по-дълбоко вкопани структури, като тук
каналът обикновено се прокопава между ямите на пещта и ра-
ботната площадка, а стената му се измазва само с глина.
Позицията, формата и големината на предпещните ями зави-
сят от тези на самите пещи. Работните площадки на съоръжени-
ята с надземна част обикновено се вкопават само частично, поня-
кога единствено в участъка към канала. Ямите на пещи с частич-
но или изцяло подземна горна камера се вкопават на по-голяма
Bridging the Gap: Continuity and Innovation in Ceramic... 37

дълбочина, което ограничава техните размери, но предлага по-

голяма защита от нежелани пориви на вятъра.
Често срещана практика през римската епоха е изграждане
на навес над предпещната яма, като последната много често об-
служва повече от едно съоръжение (фиг. 13 и 14).
2. Предназначение
През римския период пещи и работилници се откриват в грани-
ците на аграрни селища (вили и села), занаятчийски поселения,
градове и селища, свързани с военни части. В тях се произвеждат
разнообразни изделия за бита, включващи трапезна, кухненска
и амбалажна керамика, лампи и теракоти. Строителна керамика
през този период се произвежда основно от извънградски рабо-
тилници.
През Късната античност грънчарските ателиета са с по-огра-
ничени размери и репертоар, като произвеждат основно кухнен-
ска керамика, по-малко амбалажни и трапезни съдове, и лампи.
Обратна тенденция се наблюдава при пещите за строителна ке-
рамика, които през този период изглежда са по-многобройни и с
големи размери. Работилници за битови и строителни изделия се
откриват основно в градове и укрепени селища.
III. Дискусия
Както беше споменато по-горе, цялостното оформление на дву-
камерните пещи се запазва почти непроменено в периода след
установяване на римската власт в региона, като същевременно се
появяват и навлизат широко в употреба и немалко нови техники.
А. Приемственост
Както през предримския период, през римската епоха и Късната
античност сред пещите за керамика преобладават тези с кръгъл
план на долната камера, а най-често срещания вид подпора на
скарата е централен стълб (таблици 6 и 7). Техниката на оформя-
не на скари от предварително изпечени керамични сегменти се
запазва в употреба, но се среща при малък процент от съоръже-
нията. Като своеобразно продължение на тази практика може да
се посочи използване на цели имбрици или тръби за същата цел.
Б. Нововъведения
През римската епоха двукамерната пещ се превръща в най-често
срещания вид съоръжение, използвано за изпичане на керамика.
Въпреки че навлизат ограничено в употреба през предрим-
ския период, пещите с четириъгълен план еволюират като брой,
размери и използвани в градежа им техники през римската епо-
ха. Още повече, през Късната античност тези съоръжения дости-
гат (или дори задминават) по брой структурите с кръгъл план на
долната камера.
Важно нововъведение през римската епоха е масовата упо-
треба на строителна керамика в градежа на производствените
съоръжения, особено при оформлението на подпорната кон-
струкция и скарата.
Втората най-важна иновация е прилагането на строителни
способи, включващи характерните за римската архитектура арки
38 Alexander HARIZANOV

и сводове, при оформянето именно на носещата конструкция на

скарата.
Фрагменти от тухли, керемиди и тръби са също масово упо-
требявани в градежа на пещите от това време. От една страна,
тази практика е свързана с достъпността им във вид на отпадъ-
чен материал, откриван в големи количества близо до обитаемите
зони на селищата. От друга, използването им може да се обясни с
физичните качества на изпечени изделия от глина, позволяващи
по-бързо акумулиране и задържане на топлина.
Последната техника, намерила широко приложение в пери-
ода след средата на I век сл. Хр., е частичното или цялостно вко-
паване на горната камера, което се среща както при пещи с кръ-
гъл план, така и при съоръжения с четириъгълна долна камера.
В. Влияния
На този етап от проучванията изглежда сигурно, че двукамерни-
те пещи за керамика навлизат в територията на днешна България
вследствие на контакти с гръцкия свят. Типични примери за та-
кива съоръжения са кръгли в план структури с централен под-
порен стълб и скари, оформени от предварително изпечени кера-
мични сегменти. Традицията в употреба на подобни пещи се за-
пазва най-дълго по западното черноморско крайбрежие, където
те са в експлоатация почти до края на Късната античност.
В последните столетия на късножелезния период, на терито-
рията на днешна Североизточна България се появява нов вид съ-
оръжения. Малкото на брой открити пещи са иззидани основно
от глина и имат частично или почти изцяло вкопана горна ка-
мера. Тяхна най-характерна особеност е наличието на подпора
за скарата във вид на издължена езичеста стена, която в някои
случаи достига до канала и го разделя на две части. Пещите с по-
добен градеж са свързвани с келтски и дако-гетски предримски
традиции в занаята. През римската епоха тяхната експлоатация
е регистрирана отново в дакийски контекст, също в селища, оби-
тавани от носители на Черняховската култура, както и на тери-
ториите на днешните Австрия и Испания. Засега изглежда, че те
са привнесени на територията на днешна България от север или
северозапад.
Същият произход вероятно има и практиката на ситуиране
на повечето компоненти на пещта под съвременната ѝ повърх-
ност, както и на свързаната с тази техника употреба на минимал-
ни количества строителен материал. Тя е характерна именно за
части от Централна, Северна и Източна Европа, докато от друга
страна е почти непозната на грънчарите от Средиземноморието.
Една от най-важните технически иновации е описаното по-
горе приложение на характерни за римската архитектура арки и
сводови конструкции, което, заедно с цялостното увеличаване на
броя и размерите на откриваните пещи (особено на четириъгъл-
ните в план), е характерна черта за новоусвоени от Рим региони.
Изглежда най-вероятно първите приносители на тази техноло-
гия да са дислоцираните в района военни контингенти и свърза-
ното с тях цивилно население.
По-честата употреба на кирпичи в градежа на пещи от рим-
ската епоха би могло да се дължи на преселници от Мала Азия
Bridging the Gap: Continuity and Innovation in Ceramic... 39

или Близкия изток, докато наличието на пещи с комбинация от

характерни за гръцкия свят и западните римски провинции кон-
структивни особености следва да се приема като доказателство
за развиването на местни традиции в занаята (фиг. 6 и 7).
Заключение
Засега липсват данни за приемственост в керамичното произ-
водство от предримския в римския период. Все пак, направеният
по-горе анализ на наличната публикувана информация дава въз-
можност за достигане на изводи относно традициите в строежа
и оформлението на пещите от двата периода.
Технологичните практики навлезли от гръцкия свят са от
фундаментално значение за развитието на керамичното произ-
водство през предримския период. Те продължават да бъдат при-
лагани и през римската епоха, макар и с малки модификации,
което потвърждава тяхното непрекъсващо влияние.
Въпреки това, нововъведенията в технологията на двукамер-
ните пещи за керамика, настъпили след установяването на рим-
ската власт в региона, имат най-значителен принос за огромното
развитие във всеки аспект на керамичното производство, уста-
новено през римския период и Късната античност в тази част от
империята.

Alexander Harizanov PhD

32 Atanas Uzunov St.
BG-1505 Sofia
a_harizanov@abv.bg