22_Novensia_final_01_21_Novensia 7/13/12 4:09 PM Page 3
NOVENSIA 22
Ośrodek Badań nad Antykiem
Europy Południowo-Wschodniej
Warszawa 2011
22_Novensia_final_01_21_Novensia 7/13/12 4:09 PM Page 4
Projekt okładki
Anna Adamczyk & Janusz Recław
Opracowanie graficzne
Anna Adamczyk
Opracowanie redakcyjne
Piotr Dyczek
Recenzenci
Leszek Mrozewicz
Evgenia Genčeva
Gerda von Bülow
Svetlana Naumienko
Korekta
Tomasz Płóciennik
Niniejszy numer zawiera artykuły w wersji pierwotnej.
© Copyright by Ośrodek Badań nad Antykiem
Europy Południowo-Wschodniej
Uniwersytet Warszawski
Ośrodek Badań nad Antykiem Europy
Południowo-Wschodniej
00–927 Warszawa
ul. Krakowskie Przedmieście 32
novae@uw.edu.pl
Warszawa 2011
ISBN: 978–83–928330–9–3
ISSN: 0860–5777
Wydanie I
Druk: Hussar Books
22_Novensia_final_01_21_Novensia 7/13/12 4:09 PM Page 5
SPIS TREŚCI
TOMASz KOWAL, STEfAN KAROL KOzłOWSKI
zabytki prehistoryczne z Novae i Ostrite Mogili .....................................
7
JERzy KOLENDO
State of research on the inscriptions from Novae (Moesia Inferior) ........
15
JERzy KOLENDO
Symboles des fonctions militaires et des métiers sur les monuments funéraires de Novae, camp de la legio I Italica (Moesia Inferior) ...................
21
ERyK BUNSCH, JERzy KOLENDO
Czternaście nieopublikowanych inskrypcji z Novae (Moesia Inferior) .....
39
JERzy KOLENDO, TOMASz KOWAL
Stamps on ceramic pipes from Novae (Moesia Inferior) .........................
67
TOMASz PłóCIENNIK, JERzy ŻELAzOWSKI
Erotic graffito on the Roman brick from Novae (Moesia Inferior) ..............
77
PIOTR DyCzEK
Observations on marks on rooftiles, bricks and ceramic tiles from sector
IV in Novae (Moesia Inferior) ..................................................................
85
MARCIN BIBORSKI
Ein Schwertgriff vom Typ Khisfine-Köln aus dem Lager der legio I Italica
in Novae (Moesia Inferior) ..........................................................................
109
22_Novensia_final_01_21_Novensia 7/13/12 4:09 PM Page 6
6
AGNIESzKA JANKOWSKA, PAWEł KOzAKIEWICz
Identyfikacja węgli drzewnych i odcisku drewna w opus caementitium
z Novae (Moesia Inferior) ..........................................................................
119
TOMASz KOWAL
Ciężarki z Novae (Moesia Inferior) wydobyte przez ekspedycję archeologiczną Uniwersytetu Warszawskiego w latach 1960–2010 ....................
127
MARTIN LEMKE
The crossing of the Danube near Roman Novae in World War I and the
role of Svištov in military history ..............................................................
149
DOREL BONDOC
Jewelry and clothing on bronze and marble representations from Romula,
capital of Dacia Inferior ...........................................................................
159
GEORGE NUţU
Belt buckles, strap-ends and appliqués from Halmyris (Moesia Inferior /
Scythia) ......................................................................................................
171
Wskazówki dla autorów Novensia .......................................................................
Guidelines for Novensia authors ..........................................................................
201
207
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 171
George Nuţu
belt buckles, strap-ends and appliQuÉs froM
halMyris (MOESIA INFERIOR / ScYtHIA)
abstract: This paper aims to give an overview of the 37 dress accessories found at Halmyris, on
the Danubian limes. These finds were discovered during field surveys in the area close to the
fortress and inside the city walls. The main objective was to show the functionality of the pieces,
establish a typology based on components (belt-buckles, belt-plates, strap-ends, appliqués) and
propose a timeline based on analogies from different geographical areas, at the same time presenting their range.
key words: Moesia Inferior / Scythia, Halmyris, belt-buckles, strap-ends, appliqués
Belt equipment provides an important timeline in dating of archaeological complexes
thanks to the inherent changes of shape and decoration undergone over the ages, from
simple forms to more flamboyant types. Over the last half century many studies of belt
buckles and belt equipment in general have appeared, aiding in a chronological and typological classification of finds from specific archaeological sites and regionally. Collections
of several museums holding buckles from a specific area or heterogeneous finds from different Roman and Byzantine provinces have also been published. Studying these long
neglected artifacts has proved to be an important contribution to a discussion of certain
issues related to type distribution areas, technological methods and production centers.
Ethnicity remains undoubtedly and understandably one of the most difficult and in many
cases questionable issues.
No uniform treatment of the subject for the Dobrudja area has been forthcoming so far.
Most of the known buckles have been published in monographs of the cemeteries from
Callatis and Beroe1 Some finds have been presented in individual archaeological reports.
A rare exception is a series of late Roman pieces (Sucidava, Salona-Histria or Pápá type)
analyzed in studies devoted to these kind of pieces from larger areas.2
1
PREDA 1980; PETRE 1987.
UENzE 1966, pp. 142–181; TEODOR 1991, pp. 117–
138; VARSIK 1992, pp. 77–108; MADGEARU 1997,
2
pp. 39–45; MADGEARU 1998, pp. 217–222; MAD2004, pp. 169–176.
GEARU
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 172
172
Method
Analogies from Scythia have been used in particular to present the finds from Halmyris.
It is not a random decision to refer to the late Roman period in Dobrudja as most parallels
came from the necropolises of Callatis and Beroe, and can be correlated with finds from
sites on the Danube limes, especially Iatrus, Sadovec, Novae and Nicopolis ad Istrum. The
lack of archaeological context in the case of most of the analyzed pieces has determined
a broad dating for many of them, but in a few instances the archaeological context from
the well dated levels of Halmyris could be used for a more exact date. The main objective
was to emphasize the functionality of the pieces, establish a typology based on components
(belt-buckles, belt-plates, strap-ends, appliqués) and propose a timeline based on analogies
from different geographical areas, at the same time presenting their range. finally, in order
to classify these discoveries, I have followed a typology created for different types in a series
of studies with references to various authors who have examined the pieces in question.
survey of the finds
Most of the finds from Halmyris came from an area outside the fortifications where the
civil settlement is assumed to have existed. They were brought to light by several metaldetector surveys carried out in this area in 2000 and 2001. More artifacts were discovered
in the same area in 2009 by the guardian of the site. A series of buckles found during
earlier research in the fortress, in 1981, 1984 and 1991 and more recently, in 2006 and
2008, were added to these. A remarkable diversity of artifacts: brooches, jewellery, military
equipment and utilitarian tools were also discovered. Part of the fibulae have been published.3 Another group of buckles (of ‘military’ type, mostly of the 4th century AD and
later), strap-ends (heart- and amphora-shaped) and appliqués (peltae and volute-shaped,
Propellerbeschläge and one of the Martinovka type) was discovered in the same area in
1998 and 2000.4
catalogue
i. buckles with movable, hinged plate [Pl. 1]
1. ‘bow’-shaped frame buckle (‘b’-shaped frame or ‘kidney’-shaped) and rectangular plate
Keller: type 2, form A.
Sommer: type I, form C, variant b.
Soupault: type II, variant 2a.
Buckle with ‘bow’-shaped frame, rhomboidal section and double movable plate with the
bottom with volutes. The plate has three rivet holes arranged in a triangle.
Copper alloy. Well preserved; the frame made by casting in a bivalve mould, plate by forging while cold.
3
NUţU 2009, pp. 173–190; NUţU 2010a, pp. 99–
108; NUţU 2010b, pp. 64–85.
4
This group consisting of approximately 26 finds
was pointed out by Dr. florin Topoleanu and will be
published separately.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 173
173
Max. frame length: 2.3 cm; width: 3.9 cm; max. thickness: 0.5 cm; for plate: 2.5 x 2.7 cm;
thickness: 0.1 cm; weight: 13.7 g.
Halmyris 2001, civil settlement (east of the fortress), stray find.
A curved upper side of the frame (‘bow’ shape), which is more thickened than the extremities, is characteristic of this type. The ring shape is due to bending while hot, the extremities, in most cases, not welded. The rectangular plate has slotted sides or is decorated
with incised lines; a leather strap is attached with two or three rivets arranged in a triangle.
Such buckles are often found in 4th century AD contexts. The broad range of the type
demonstrates their unpretentiousness, a common and easily fabricated type thanks to its
simplicity. E. Keller dates the type in the middle and second half of the 4th century AD
(340–380),5 while C. Preda narrows the time range to the third quarter of the 4th century
AD.6 The buckle from Romuliana7 was dated even later, in the early 5th century AD, but
the discoveries from Beroe may be dated to the middle and second half of the 4th century
AD;8 the same is true of the buckle from Beške (Srem).9 In addition to finds from the territory of the Empire,10 buckles of this type were discovered in Germania libera.11
Chronology: middle and second half of the 4th century AD.
2. ‘bow’-shaped frame buckle (‘b’-shaped frame or ‘kidney’-shaped frame)
Keller: type 2, form A.
Sommer: type I, form C, variant b.
Soupault: type II, variant 2a.
‘Bow’-shaped frame, oval in section; the plate, now lost, was similar to the one of the previous object.
Copper alloy. The frame was obtained by casting in a mould.
Length: 3.3 cm; max. thickness: 0.3–0.4 cm; weight: 6.2 g.
Halmyris 2009, civil settlement, stray find.
Two similar buckles were discovered at Novae and regarded as pieces of military equipment12 and other buckle frames were discovered at Sucidava13 and Callatis.14 Similar to
the previous object.
Chronology: middle and second half of the 4th century AD.
3. ‘bow’-shaped frame buckle (‘b’-shaped frame or ‘kidney’-shaped frame)
Keller: type 2, form A.
Sommer: type I, form C, variant b.
Soupault: type II, variant 2a.
‘Bow’-shaped frame, rhomboidal in section; the plate now lost.
Copper alloy. The frame was obtained by casting in a bivalve mould.
5
KELLER 1971, pp. 45–46, fig. 18/4, pp. 59–60, fig.
23/3.
6
PREDA 1980, p. 40.
7
žIVIć 2007, pp. 283–284, no. 46b, p. 301, pl. 8.
8
PETRE 1987, pp. 58–59, pl. 104/151d – Grave C67,
pl. 105/153b – Grave C38, pl. 106/154b – Grave
A268.
9
MARIJANSKI-MANOJLOVIć 1987, p. 152, pl. 56/4.
SOUPAULT 2003, p. 43, pl. 10/6–8; ALföLDI 1957,
p. 459, pl. 101/8–12.
11
SOMMER 1984, pp. 22–23, pl. 2/6–8, pl. 3.
12
GENčEVA 2000, pp. 54, nos. 13–14, pl. 1/13–14.
13
TUDOR 1938, pp. 410–411, no. 4, fig. 15/5.
14
PREDA 1980, p. 150, pl. 16/M344.
10
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 174
174
Length: 4.2 cm; thickness: 0.2–0.5 cm; weight: 5.9 g.
Halmyris 2009, civil settlement, stray find.
Similar to the previous object.
Chronology: middle and second half of the 4th century AD.
4. ‘bow’-shaped frame buckle (‘b’-shaped frame or ‘kidney’-shaped frame)
Keller: type 2, form A.
Sommer: type I, form C, variant b.
Soupault: type II, variant 2a.
‘Bow’-shaped frame, rhomboidal in section; the plate now lost.
Copper alloy. The frame was obtained by casting in a bivalve mould.
Length: 3.4 cm; weight: 6.1 g.
Halmyris 2009, civil settlement, stray find.
Similar to the previous object.
Chronology: middle and second half of the 4th century AD.
5. heart-shaped buckle plate
Callatis type.
Soupault: type I, variant 6.
Buckle fragment with movable heart-shaped plate completed by two volutes at the bottom
and by a fastening system consisting of four rings at the top. The edges are defined by
a row of incised pearls and the two fretted crescents in the centre of the plaque. Three
mounting rivets arranged in triangle secured the fastening. Mould cast.
Copper alloy. Well preserved, green patina.
Lenght: 2.8 cm; width: 2.9 cm; thickness: 0.1–0.4 cm.
Halmyris 2000, civil settlement, stray find.
Three buckles of this type are known: two from the province of Scythia, the Roman-Byzantine necropolis of Callatis, and a third found in unknown circumstances in Vârtop village
(Constanţa County, Museum of Constanţa, inventory no. 44665A). The fully preserved
buckle with movable plaque from Callatis was found in Grave 320, together with a simple
bronze bracelet and no dating elements.15 The buckle with fixed plaque from Vârtop, from
the former Matei collection, was dated to the late 4th century AD according to C. Preda.16
Another single find is known from Praesidium Pompei (zindan, near ćićina village on
Morava Valley);17 it is a buckle with movable plaque, identical to those from Callatis and
Halmyris.
Two variants can be distinguished among buckles of the Callatis type discovered so far,
one, A, with fixed plaque, and B, with movable plaque. They share the same cordiform
plaque, a fastening system with three rivets arranged in a triangle and decoration comprising two opposing crescents,18 three double circles on the top (corresponding to
the arrangement of rivets on the opposite side) and a series of circular and semicircular
15
PREDA 1980, p. 40, type d, pls. 15/M320,
84/M320.1; also SOUPAULT 2003, p. 40, pl. 7/2.
16
APARASCHIVEI, CHIRIAC 2009, pp. 72–73, no. 2, fig. 3.
17
RAŠKOVIć 2007, pp. 222–223, fig. 17/3.
18
The opposing crescents decorative motif can be
observed on some balteus pieces, such as one from
Gherla, dated to the 2nd–3rd century AD; see GăzDAC 1995, p. 404, no. 27, fig. 3/27.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 175
175
incisions surrounding the sides and centre of the plaque. The double volute on the bottom
of the plaque is common to both versions, with the exception of the buckle from Callatis.
The sharp-pointed tongue flanked by volutes is common to both variants; the volutes close
the two buttons located on the ring bottom inwards.19 Variant A is represented by the buckle
from Vârtop, variant B by the finds from Callatis, Halmyris and Praesidium Pompei. The
heart-shaped movable plate from Chersonese has an oval frame, circular in section, and
the plate is not decorated.20 As regards the ‘buckle’ plate discovered inside Grave 270 of
the late Roman cemetery from Ságvár and mentioned as an analogy for the finds from
Callatis and Vârtop,21 it can be omitted from this discussion; despite not being fully analyzed by A. Sz. Burger, the ‘buckle’ belongs to a hinged strap-end with plate decorated
with peltae and volutes.22
Overall, this is a rare type of buckle and the range of the finds makes it difficult to pinpoint
its origins and places of production. In terms of type chronology, the buckle from Callatis
was dated by the context to the middle and third quarter of the 4th century AD, while the
find from Praesidium Pompei sets a terminus post quem since the settlement was destroyed
by Hunnic raids between AD 440–450.23
Chronology: 4th century AD, most probably second half of the century.
6. oval buckle plate with engraved decoration
Schulze-Dörrlamm: type B10.
Oval buckle plate, broken in two; the top decorated with vines, grapes and two incised
lines on the edges; three holes for fastening rivets arranged in a triangle. The bottom is
trapezoidal, unornamented, and has one hole for a rivet.
Copper alloy. Well preserved.
Length: 2.5 cm; width: 3.8 cm; thickness: 0.2 cm; weight: 7.3 g.
Halmyris 2001, civil settlement, stray find; ICEM, inv. 45657.
The buckle is fragmented, but parallels help to reconstruct the frame’s solid flattened-oval
shape and the thick bottom of the tongue. Buckles of this type from Argamum,24 Beroe,25
Noviodunum26 and Callatis27 in Scythia have been dated to the second half of the 4th and
first half of the 5th century AD. A buckle found at Novae in the lower Danube is also noteworthy.28 Buckles with representations of fish from Argamum, Noviodunum, Callatis, as
well as engraved plate buckles with religious inscriptions from the Museum of Varna29
and the RGzM30 collection may both be regarded as everyday objects and Christian
amulets, depending on the case.
Chronology: mid 5th century AD.
19
A similar find was found in faimingen; see OL1976, pp. 211–212, 275, no. 977, pl. 74.
20
SOUPAULT 2003, p. 40, pl. 7/3.
21
PREDA 1980, p. 40; APARASCHIVEI, CHIRIAC 2009,
p. 73.
22
BURGER 1966, pp. 127, 223, fig. 116 – Grave 270,
no. 6.
23
RAŠKOVIć 2007, p. 225.
24
NUţU, IACOB 2008, no. 2.
DENSTEIN
25
PETRE 1987, p. 57, pl. 100/147f – Grave D29–D30.
BARNEA 1977, p. 231, fig. 89/2.
27
PREDA 1980, pp. 40–41, pl. 46/6.
28
GENčEVA 2000, p. 58, no. 34, pl. 2/12.
29
MINčEV 1994–1995, pp. 22–23, nos. 3–4, figs. 3–4.
30
SCHULzE-DöRRLAMM 2002, pp. 56–57, nos. 48, 50.
26
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 176
176
7. oval buckle plate
Three holes for fastening rivets arranged in a triangle.
Copper alloy. Hard corrosion.
Lenght: 2.6 cm; thickness: 0.1–0.2 cm; weight: 1.5 g.
Halmyris 2001, civil settlement, stray find.
Although fragmentary, the buckle plate in question can be dated to the 4th century AD
based on parallels from Callatis31 and Beroe.32 Buckles with oval plate have a semidiscoidal
or ‘bow’-shaped frame and, like the type with rectangular plate, they are common finds
in 4th century AD contexts.
Chronology: 4th century AD.
8. circular frame buckle and trapezoidal plate
Circular frame buckle with double trapezoidal plate with central rivet.
Copper alloy. Well preserved.
Lenght: 3 cm; plate length: 1.5 cm; ring thickness: 0.4 cm; weight: 3.9 g.
Halmyris 2009, civil settlement, stray find.
In terms of shape the find from Halmyris is similar to buckles of the so-called ‘Hunnic’
type from Chersonese and Kerch dated to the end of the 4th – beginning of 5th century
AD.33 Two such buckles, made of silver, were presented by M. Sommer.34 A bronze buckle
from Praesidium Pompei was dated rather to the 4th century AD,35 the same in the case
of the object from Callatis,36 while two D-shaped buckles found at Aquileia were dated
earlier, between the mid 3rd and early 4th century AD.37
Chronology: 4th – beginning of 5th century AD.
9. hinged rectangular double plate of a buckle
Double rectangular plate buckle with a hinge in the lower side.
Copper alloy. Well preserved, missing ring and pin.
Lenght: 2.2 cm; width: 1.3 cm; thickness: 0.2–0.4 cm; weight: 2.4 g.
Halmyris 2001, civil settlement, stray find.
In the case of this plate it is difficult to reconstruct the shape and fastening system due to
the fragmented state. The hinge could have been used to suspend another plate in a fashion
characteristic of lorica fastening elements. The simple rectangular shape of the plate is
typical of the early Roman period.38
Chronology: 1st–2nd century AD.
10. oval-frame buckle and semicircular plate
Simpson: group II.
Soupault: type III, variant 1.
Oval frame, triangular in section, ornamented with radial lines; plate semicircular, with
central rivet.
31
PREDA 1980, pp. 38–39, pl. 15/M 352.2.
PETRE 1987, p. 58, pl. 103/151e – Grave A188a–
b, p. 60, pl. 107/159 – Grave C137.
33
SOUPAULT 2003, p. 44, pl. 12/9–11.
34
SOMMER 1984, pl. 77/9–10.
32
35
RAŠKOVIć 2007, pp. 222–223, fig. 15/1.
PREDA 1980, p. 149, pl. 15/M339.
37
BUORA 2002, p. 191, type II/h, pl. 3/25–26.
38
GENčEVA 2000, p. 77, no. 16, pl. 10/15; BISHOP
2002, p. 37, fig. 5.6/1-a, p. 52, pl. 4.
36
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 177
177
Copper alloy. Well preserved, fragmentary frame, missing tongue.
Length: 2.2 cm; thickness: 0.1–0.2 cm; weight: 4.4 g.
Halmyris 2009, civil settlement, stray find.
Characterized by small size, the buckle is similar to a set of three buckles discovered at
Callatis, Tomis and Chersonese, and dated to the 4th century AD. The parallels were classified by V. Soupault as type III, variant 1.39 The buckles from Intercisa,40 Krefeld-Gellep
or Ságvár41 were dated to the same century, while a fragmentary plate from San Giovanni
di Ruoti was dated between the late 4th and the mid 5th century AD;42 the Schulze-Dörrlamm B8 type, larger and with ‘bow’-shaped frame (‘kidney’ or ‘mit nierenförmig
Beschlag’), too.43 On the other hand, two similar but silver buckles found at Nova Nadežda
(Haskovo) were dated later.44 Looking at this group we see a number of heterogeneous
pieces sharing different decorative systems, but all having in common the small semicircular plate, as well as frame, although in the latter case there are also ‘D’-shaped (‘D’förmig) frames. However, the buckle from Halmyris demonstrates a characteristic feature
in the form of a frame ornamented with radial lines in relief (godronné)45 typical of the
period of transition from the 4th to the 5th century AD. In the case of buckles with this
type of frame no parallels for the semicircular plate are known; in fact, the frame is larger
in size than in this case.
Chronology: end of 4th – mid 5th century AD.
11. oval frame buckle
Oval frame buckle, curved slightly outward; the pin’s fastening is much thinner and the
plate is missing.
Copper alloy. Well preserved. full cast.
Lenght: 2.1 cm; width max.: 1.2 cm; frame thickness: 0.1–0.2 cm; weight: 0.6 g.
Halmyris 2001, civil settlement, stray find; ICEM, inv. 45657 or 45651 (?).
The closest parallels for this type of buckle are found at Beroe, in 5th century AD occupational levels;46 at the time of discovery it still retained a small part of its semicircular
plate decorated with dot incisions and attached by two projections that were bent around the
straight edge of the frame in order to form a hinge; now the plate is gone due to corrosion.
Chronology: 5th century AD.
12. oval frame buckle
frame buckle with broken bottom.
Copper alloy. Well preserved, full cast.
Length: 2.6 cm; width: 1.2 cm; thickness: 0.2 cm; weight: 1.4 g.
Halmyris 2001, civil settlement, stray find; ICEM, inv. 45649.
Similar to the previous object.
Chronology: 5th century AD.
39
SOUPAULT 2003, p. 44, pl. 12/1–3.
ALföLDI 1957, p. 462, pl. 104/1.
41
SIMPSON 1976, p. 196, fig. 2/1–2.
42
SIMPSON 1997, p. 32, no. 110, pl. 120.
43
SCHULzE-DöRRLAMM 2002, pp. 52–54, no. 44.
40
44
DASKALOV, DUMANOV 2004, pp. 195, 198, fig.
2/1–2.
45
SOUPAULT 2003, pl. 1/11–13.
46
PETRE 1987, p. 57 – Grave D31, pl. 100/147/f,
p. 64 – Grave D37, pl. 115/173g.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 178
178
ii. single-piece buckles with cast plate [Pl. 2]
13. triangular-shaped plate buckle
Sommer: Gruppe 1, form 3, type E.
Aurrecoechea fernández: Mainz type.
fragmentary buckle with plate consisting of two V-shaped arms and the lower side ending
in a circular button with central rivet. The frame, ‘D’- or ‘bow’-shaped (according to preserved objects), is broken in this case; the top is rounded and the lower side flat. Decoration
of the buckle on the bottom consists of three parallel lines in relief, arranged perpendicularly. Mould casting.
Copper-alloy. Severe corrosion.
Lenght: 4.6 cm; thickness: 0.4 cm.
Halmyris 2001, civil settlement, stray find.
Buckles with flat triangular-shaped plaque are particularly prevalent in Gallic, Rhenan
and Danubian territories and beyond, at Thamusida, in Britannia and possibly in Baetica.
M. Sommer lists 14 finds,47 to which one can add the two copies found in Scythia: the one
from Halmyris and another buckle discovered at an unspecified site in the province and
now in the former Matei collection.48 Areas with high concentration of troops correspond
in some cases to the location of the triangular-shaped plate buckles. Thus, in J. Aurrecoechea fernández’s49 opinion, this type can be considered as cingula militiae, although
the low number and broad scattering of the known finds call for caution in generalizing
this idea. The type is generally assigned to the 4th century AD, but M. Sommer has defined
two chronological horizons based on find contexts in the Gallo-Rhenan area (310 – mid 4th
century AD) and on the Danube (290–400 AD); during this time the type ran with a series
of similar but fully triangular plaque buckles.50 Close analogies to the buckle from Halmyris
are offered by objects discovered at furfooe (Naumur, Belgium),51 Gradine (Drvara),
Szentlászló-Szentgyedpuszta,52 zemun Polje and Ušće,53 Strajata (near Pleven)54 and Baetica
(now in the RGzM collection in Mainz)55 or the buckle mentioned above from Scythia.
Chronology: first half of 4th century AD.
14. single-piece buckles with cast rectangular plate
Buckle with rectangular plate, square in section and fragmentary round frame.
Length: 2.2 cm; width: 1.9 cm; plate length: 1.3 cm; plate width: 1.9 cm; thickness: 0.4 cm;
weight: 5.3 g.
Halmyris 2001, civil settlement, stray find.
The buckle is similar in shape to an example from the Dălgopol Museum and both can be
dated to the 4th century AD.56
Chronology: 4th century AD.
47
SOMMER 1984, pp. 47–48, n. 187; also AURRECOE2001, pp. 128–129, fig. 45.
48
APARASCHIVEI, CHIRIAC 2009, pp. 70–72, fig. 2.
49
AURRECOECHEA fERNáNDEz 2001, p. 129.
50
SOUPAULT 2003, p. 40, type 14, pl. 16.
51
SOMMER 1984, p. 39, pl. 16/8.
CHEA fERNáNDEz
52
VINSKI 1967 (1974), p. 41, pls. 37/12, 38/1.
CRNOBRNJA 1997, pp. 302–303, nos. 2–3.
54
TABAKOVA-CANOVA 1981, pp. 144, 146, fig. 8/28.
55
AURRECOECHEA fERNáNDEz 2001, p. 128, fig. 45.
56
LAzAROV 2001, p. 157, no. 122.
53
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 179
179
15. oval-frame buckle with cast ‘t’-shaped plate
Salona-Histria type.
Buckle with oval frame and T-shaped lower side. The tongue is bent, and has two protrusions, strongly underlined. At the bottom of the foot, the rectangular end through which
the belt passed is missing, being either filed off or most probably attached later, after the casting. On the back it has two loops attached by welding. Made by casting in a bivalve mould.
Copper-alloy. Well preserved, slight traces of corrosion.
Lenght: 3.6 cm; max. width: 1.6 cm; thickness: 0.3–0.8 cm; weight: 6.072 g.
Halmyris 2001, civil settlement, stray find.
Studies by S. Uenze,57 D. Gh. Teodor58 and V. Varsik59 have made the ‘Taschenschnalle
mit Riemenschlaufe’ type well known. They were used as clasps for small bags holding
items of daily use or to secure knives. The buckles from Dobrogea have been analyzed
repeatedly60 and their distribution in the province of Scythia and northern Moesia secunda:
Capidava,61 Sacidava,62 Sucidava (Pârjoaia)63 and Păcuiul lui Soare64 reflects their spread
along the Danubian limes; a buckle is mentioned from Histria on the Black Sea coast65
and another one was found more recently at Argamum.66 Buckles of this type were registered on the middle Danube and in Crimea,67 in settlements and cemeteries. Published
finds from southern Dobrogea include artifacts from Odărci (Dobrič), Vojnikovo (Tervel),
Proučik čunčevo (Kaliakra),68 and further south, at čala – Golyam Izvor (Haskovo).69
Two buckles, from Corinth70 and Saračhane,71 were dated to the 7th century. As regards
the buckle from Halmyris, there are enough parallels to suggest a date between the second
half of the 6th century AD and the first quarter of the next century, while the Salona-Histria
type ran simultaneously with a series of variants. With regard to the latter, the former Matei
collection contains one such example,72 a buckle that can ascribed to S. Uenze’s type
‘Schnallen mit schildförmigem Beschlägende’,73 a variant certainly produced in the workshop at Drobeta.74
Chronology: second half of the 6th century – first quarter of the 7th century AD.
16. shield-shaped plate belt buckle
Sucidava type.
Shield-shaped plate buckle and holes arranged in the form of a mask; the lower side still
retains traces of a fastening hole. Bivalve mould cast.
Copper alloy. Three fragments.
57
UENzE 1966, pp. 142–146.
TEODOR 1991, pp. 125–128, fig. 5/1–8.
59
VARSIK 1992, pp. 80–81.
60
MADGEARU 1997, pp. 42–43.
61
SCORPAN 1978, p. 171, n. 26.
62
SCORPAN 1978, p. 171, no. 55, pls 11/55, 19/55.
63
DIACONU 1974, p. 290, fig. 1/10, pp. 292–293, no. 8.
64
DIACONU 1985, pp. 89–90, fig. 1/8.
65
UENzE 1966, p. 143, fig. 1/13.
66
OANţă-MARGHITU 2006, p. 347, no. 1, pl. 1/1.
67
VINSKI 1967 (1974), p. 24, pl. 17/1–8; VIŠIćLJUBIć 1994, p. 236, no. 32; GARAM 2001, p. 109,
58
pl. 75/1–11; ŠPEHAR 2010, p. 54, nos. 45–47, pl.
2/45; UENzE 1966, pp. 142–146.
68
HARALAMBIEVA 1993, p. 34, pl. 1/1–3, pp. 40, 42.
69
ALADJOV 1989, p. 27; DASKALOV, DUMANOV 2004,
pp. 193–194, no. 2, fig. 1/2.
70
DAVIDSON, HORVáTH 1937, p. 236, fig. 6/E; DAVIDSON 1952, pp. 272–273, no. 2211, pl. 114.
71
GILL 1986, p. 265, no. 565, fig. 404.
72
APARASCHIVEI, CHIRIAC 2009, pp. 73–74, no. 3, fig. 4.
73
UENzE 1966, p. 167, pl. 14; for a variant, see EGER
2001, pp. 341–342, no. 5, pl. 1/5.
74
BEJAN 1976, p. 262, pl. 6.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 180
180
Lenght: 3 cm; width: 2.6 cm; thickness: 0.1 cm; weight: 3.9 g.
Halmyris 2001, stray find; ICEM, inv. 45656.
Sucidava type buckles are one of the most broadly distributed cast plate buckle types in
the Danubian area in the second half of the 6th century AD.75 A massive concentration of
the type in the Danube area and the Balkans in general attests to the production of this
form on the Danubian limes, with workshops located inside fortifications with strong garrisons. Most of the 25 buckles registered from Scythia were found at the late Roman
necropolis in Beroe and (L)Ibida,76 including the Teodor I/1f type, which the buckle under
discussion belongs to.77 Close parallels for the buckle from Halmyris can be found at
Beroe,78 (L)Ibida,79 Negotin,80 Aquis,81 Sadovec,82 Dobrič83 and Pernik.84
Chronology: mid 6th – first quarter of 7th century AD.
17. shield-shaped plate belt buckle
Sucidava type.
Eger: Petropavlovsk type.
Shield-shaped plate buckle with holes arranged in the form of a mask; a small hole at the
top marks the place of the tongue, now lost; two loops can be seen on the back. Cast in
a bivalve mould.
Copper alloy. Well preserved.
Lenght: 2.4 cm; width: 1.9 cm; thickness: 0.2 cm; max. thickness: 0.7 cm; weight: 3.8 g.
Halmyris 2006, PE 8/9, – 0.70 m.
Buckle of the same type as the previous one, but smaller, with shield-shaped plaque with
protrusions at the top; the human figure represented by two holes for eyes and a wider one
for the mouth is more sketchy than in the previous case. As emphasized by D. Gh. Teodor,
different variants of the Sucidava type were in use simultaneously, hence they can be found
sometimes in the same closed contexts, making it difficult to distinguish between the different models.85 The buckle in question belongs to the Teodor I/1g type (= Madgearu II/d)
owing to the styling of the facial motif.86 Buckles with a similar pattern were found at
Beroe in association with a buckle of the previous variant,87 in the former Severeanu collection,88 at (L)Ibida,89 Dobrič and Onogur,90 čala – Golyam Izvor (Haskovo),91
Petropavlovsk,92 in Byzantine Italy (Rimini area),93 but also in Asia Minor, at Sardis94 and
75
TEODOR 1991, pp. 118–125, with bibliography;
MADGEARU 1997, pp. 39–41; MADGEARU 1998, pp.
217–222.
76
MADGEARU 1998, p. 218.
77
TEODOR 1991, pp. 123–124, fig. 4/6, 8.
78
PETRE 1987, pp. 67–68 – Grave C62, pl. 122
bis/189b.
79
OPAIţ 1991, pp. 46–47, no. 45, fig. 18.
80
VINSKI 1967 (1974), p. 37, pl. 31/1.
81
JANKOVIć 1981, pp. 175, 251, pl. 17/2.
82
UENzE et alii 1992, pp. 184–187, pl. 12/4.
83
HARALAMBIEVA 1993, p. 37, pl. 2/12.
84
DASKALOV, TRENDEfILOVA 2006, pp. 383–384,
389, fig. 2/1.
85
TEODOR 1991, p. 125.
TEODOR 1991, pp. 124–125; MADGEARU 1998,
p. 219.
87
PETRE 1987, p. 69 – Grave B45, pl. 126, 200d.
88
GRAMATOPOL 1982, p. 264, no. 9, pl. 27.
89
OPAIţ 1991, pp. 46–47, no. 43, fig. 18.
90
HARALAMBIEVA 1993, p. 37, pl. 3/5–6.
91
DASKALOV, DUMANOV 2004, pp. 193–194, no. 1,
fig. 1/1.
92
UENzE et alii 1992, p. 185, fig. 14/6; also AMBROz
1971, p. 112, pl. 2/7.
93
CAVALLARI 2005, p. 148, no. 15, fig. 118.
94
CRAWfORD 1990, p. 36, M59.45, fig. 119; WALDBAUM 1983, p. 119, nos. 693, 695, pl. 44.
86
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 181
181
Amorium,95 and in North Africa, most probably at Carthago.96
Chronology: mid 6th – first quarter of 7th century AD.
18. shield-shaped plate belt buckle
Sucidava type.
Buckle with bow-shaped ring and shield-shaped plate; the ornament is represented by two
joined holes at the bottom and one for the tongue at the top. On the other side, two loops
for attaching the leather strap. Bivalve mould cast.
Copper alloy. Metal corrosion.
Lenght: 3.3 cm; width plate: 1.8 cm; width ring: 1.6 cm; ring length: 2 cm; thickness: 0.2 cm;
max. thickness: 0.9 cm; weight: 4.9 g.
Halmyris 2006, T12, N 11?, – 1.36 m; ICEM, inv. 45651.
The model is a simplified and sketchy version of the previous variant, but it is characterized
by a longer plaque at the bottom, ending in a sharp protrusion. A strap attachment with
two small loops is preserved and two sharp protrusions can be observed at the bottom of
the ring, as in the case of the earlier buckle. Chronologically, the model belongs to a late
group of Sucidava buckles and as such is seldom encountered. A close analogy can be
found at (L)Ibida.97
Chronology: mid 6th – first quarter of 7th century AD.
19. pierced shield-plate buckle
Teodor: type VI/a.
Buckle with rectangular plate with perforated edges and three circular holes arranged in
a triangle at the bottom; another oval hole is for attaching the tongue, now lost. The buckle
is oval, triangular in section, with two protrusions on the front side top to guide the pin.
Cast in a bivalve mould.
Copper alloy. Well preserved.
Length: 3.3 cm; width: 1.5 cm; frame width: 1.7 cm; frame length: 2.2 cm; plate thickness:
0.2–0.4 cm; frame thickness: 0.2 cm; weight: 3.7 g.
Halmyris 2001, civil settlement, stray find; ICEM, inv. 45652.
Characteristic of this rare type of buckle is the frame, oversized compared to the plate,
with two knobs for attaching the pin; the plate with the perforated sides is similar to
a buckle found in Crimea, but with a different central decorative motif.98 A close analogy
is supplied by a buckle from Bratei.99
Chronology: 6th – first half of 7th century AD.
20. semi-discoidal frame-shaped buckle and rectangular perforated plate
Schulze-Dörrlamm: type G4.
Buckle with semi-discoidal frame, fixed rectangular-shaped plaque and lateral protrusions.
The tongue, gently curved, is in the shape of a bird’s beak and is flanked by two protrusions.
95
98
96
99
LIGHTfOOT 2003, p. 98, pl. 6/13.
EGER 2010, pp. 136–138, pl. 5/2.
97
OPAIţ 1991, pp. 47–48, no. 46, fig. 19.
AIBABIN 1990, p. 219, fig. 40/17.
TEODOR 1991, pp. 131–132, fig. 6/6.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 182
182
The buckle is characterized by a round top and flat bottom. Cast in a mould.
Copper alloy. Excellently preserved.
Length: 3 cm; width: 3.1 cm; tongue length: 2.2 cm; thickness: 0.3–0.5 cm; weight: 8.4 g.
Halmyris 2000, civil settlement, stray find.
There are several parallels for this type of buckle, but with different chronologies. A similar
find from Vraca was recently republished by E. Genčeva and dated to the end of the 2nd
– beginning of the 3rd century AD.100 M. Lightfoot opted for a wider dating, between the
2nd and the 4th century AD, in the case of a buckle discovered in the environs of Afyon,
near ancient Amorium.101 The buckle from Jauja (Córdoba), included in Aurrecoechea fernández’s Teba type, was dated a little later, towards the last quarter of the 4th and beginning
of the 5th century AD.102 The other chronological range considered with regard to the type
is the Byzantine period; such is the dating of a buckle found at Corinth, placed in the Early
Medieval period by G. Davidson without any parallels.103 Recently, with the publication
of a buckle from the Römisch-Germanisches zentralmuseum collection, M. Schulze-Dörrlamm resumed the discussion regarding this type. Thus, based on analogies with some
finds found in well known contexts the G4 type has been dated to the 10th century and
the first part of the 11th century.104
The varying opinions regarding the dating of this type are due undoubtedly to the unknown
contexts in which most of the bronze objects were discovered. This is the case of the finds
from Amorium, Jauja and Corinth, the Römisch-Germanisches zentralmuseum collection
and Vraca, as well as the buckle from Halmyris. On the other hand, the general shape,
similar to early peltiform buckles with rectangular plaque as suggested by Aurrecoechea
fernández, should be emphasized.105 However, M. Schulze-Dörrlamm’s rigorous analysis
has placed the type securely in the Byzantine period and, as far as distribution is concerned,
has pointed to regions under direct influence of the Byzantine Empire. The buckle discovered at Jauja stands as one of the examples the origin of which cannot be reasonably
explained.
Chronology:10th–11th century AD.
iii. simple frame buckles [Pl. 3]
21. ‘d’-shaped frame type
full buckle, oval in section, with flattened fastening and elongated sides.
Copper alloy. Cast in a bivalve mould, well preserved.
Length: 1.8 cm; width: 2.4 cm; thickness: 0.2–0.3 cm; weight: 1.9 g.
Halmyris 1981, S I, square 42, N 2 (= N 12?), – 0.50 m; ICEM, inv. 29578.
The term ‘D’-shaped (‘D-förmig’) refers to a series of small- and medium-sized buckles
100
103
101
104
GENčEVA 2009, p. 15, fig. 1/10.
LIGHTfOOT 2003, p. 87, no. 24, p. 100, pl. 6/30.
The buckle from Saraçhane is dated to the Byzantine period, but it certainly belongs to a different
type; see GILL 1986, p. 266, no. 579.
102
AURRECOECHEA fERNáNDEz 2001, p. 79, no. 34,
pp. 124–129.
DAVIDSON 1952, p. 272, no. 2203, pl. 114.
SCHULzE-DöRRLAMM 2009, pp. 262–264, 359,
no. 576.
105
AURRECOECHEA fERNáNDEz 2001, p. 124.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 183
183
with half-round top and straight bottom specific to the Roman Byzantine era. The small
size of this buckle type found in the Danube and west Black Sea area led S. Uenze to believe
that the buckles had been used for bags (‘Taschen’) and even small knife ‘cases’. Based on
analogies with similar buckles discovered at Sadovec,106 Abritus, Severci107 and in the necropolis at Beroe,108 the buckles from Halmyris can be dated to the 5th–6th centuries AD. At
Novae buckles of the ‘D’-shaped type are dated earlier, in the 4th century AD.109
Chronology: 5th–6th century AD, most probably 6th century AD.
22. ‘d’-shaped type frame
full buckle, oval in section, with flattened fastening and protruding elements for fixing
the strap in place.
Copper alloy. Cast, well preserved.
Length: 1.6 cm; width: 2 cm; thickness: 0.2 cm; weight: 1.2 g.
Halmyris 2001, civil settlement, stray find.
Similar to the previous object.
Chronology: 5th–6th century AD, most probably 6th century AD.
23. large oval frame
Oval buckle, with thicker fastening and the outside front side thinner, slightly bent; oval
in section.
Copper alloy, well preserved. Cast in a mould.
Length: 3 cm; width: 6.7 cm; thickness: 0.6–0.9 cm; weight: 47 g.
Halmyris 2008, T XII, Area C, – 4.35 m.
The large size suggests that it was probably a harness piece, similar to a buckle from
Novae.110
Chronology: 4th–5th century AD.
24. oval frame buckle
full buckle, oval in section, with flattened fastening to secure the pin. The front side of
the frame is thickened and trapezoidal in section.
Copper alloy. full cast in a mould.
Length: 1.9 cm; width: 3.3 cm; thickness: 0.3–0.6 cm; weight: 9.5 g.
Halmyris 2001, civil settlement, stray find.
Buckles of this type feature a simple frame of the same kind as in finds from Novae111 and
Salona;112 however, there are also pieces decorated with incised dots as in Classe
(Ravenna)113 or with lines in relief as in Prahovo (Aquis).114 All buckles mentioned above
have been dated largely to the 6th century AD without any other precise chronology; the
buckle from Sadovec is an exception, being dated to the early 5th century AD.115
Chronology: 5th–6th century AD.
106
UENzE et alii 1992, pp. 192–193, pl. 12/10–11.
HARALAMBIEVA 1993, pp. 40–42, pl. 4/2–3, 5–6.
108
PETRE 1987, p. 65 – Grave E191, pl. 117/176b,
B150, pl. 117/177b.
109
GACUTA 1987, pp. 155–156, no. 308, pl. 27/5.
110
GACUTA 1993, p. 88, no. 305, pl. 58/12.
107
111
GENčEVA 2000, p. 56, no. 22, pl. 2/6.
VIŠIć-LJUBIć 1994, p. 234, nos. 23–24.
113
CAVALLARI 2005, p. 142, no. 11, fig. 94.
114
JANKOVIć 1981, p. 247, pl. 13/16.
115
UENzE et alii 1992, p. 175, pl. 9/18; also at
Novae, see GENčEVA 2000, p. 56, no. 22.
112
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 184
184
25. animal-heads frame buckle
Rectangular buckle with semicircular frame and knobs (horse or dolphin-heads). full cast.
Length: 1.7 cm; width: 2.2; thickness: 0.4 cm; weight: 3.6 g.
Copper alloy. Substantial corrosion in the upper side.
Halmyris 2001, civil settlement.
The pronounced corrosion and the fragmentary state of preservation of this type (with
movable plate or without) makes it difficult to date this buckle. Generally, oval-frame
buckles decorated with animal heads are dated to the 4th–5th centuries AD, but I no close
parallel was forthcoming for the buckle from Halmyris.
Chronology: 4th century AD.
26. small rectangular buckle
Simple small-sized rectangular buckle, triangular in section. The top is faceted, as well as
the pin. Cast in a bivalve mould.
Silver. Excellently preserved.
Length: 1.1; width: 1.4 cm; tongue length: 1.2 cm; thickness: 0.2 cm; weight: 1.7 g.
Halmyris 2001, civil settlement, stray find.
Rectangular buckles are common finds in 4th century occupational levels, at Gräfelfing
(Keller type b, Grave 6 — last quarter of the 4th century AD)116 and Ságvár (Simpson
Group IV).117 Most of them are faceted and bigger than the piece found at Halmyris and
made of bronze, with movable rectangular plate.
Chronology: 4th century AD.
27. oval buckle
Oval buckle, circular in section.
Iron. Significant corrosion.
Length: 4.1 cm; width: 4.5 cm; thickness: 0.5–0.7 cm; weight: 11.2 g.
Halmyris 1984, square15, N 10, – 2.65 m; ICEM, inv. 43649.
The find context, Level 10, corresponds to the first half of the 6th century and probably
ends with the Kutriguri raid of 559,118 producing important data for the secure dating of
this find.
Chronology: first half of 6th century AD.
iv. Varia [Pl. 3]
28. belt buckle
Large-sized rectangular buckle; front side oval, slightly flattened from the body. The
tongue is made of an alloy with different chemical composition. The buckle edges are
adorned with a rope motif; on the top there is a representation of a classic bull’s head and
two floral motifs under each horn. On the bottom, the buckle is concave.
Copper alloy. Well preserved, slightly rusted.
116
117
KELLER 1971, p. 63, pl. 23/4.
SIMPSON 1976, p. 197, fig. 3/5.
118
SUCEVEANU, zAHARIADE, TOPOLEANU 2003, pp.
37–38.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 185
185
Length: 4.5 cm; width: 4.1; thickness: 0.2 cm; weight: 13.9 g.
Halmyris 2000, civil settlement.
There are no analogies for this type of buckle.
Chronology: post medieval or contemporary (?).
v. Miscellaneous buckle fragments [Pl. 3]
29. buckle tongue
Massive tongue characterized by curved sides and a faceted top.
Copper alloy, well preserved.
Length: 3.5 cm; width: 0.5–0.6 cm; thickness: 0.3–0.5 cm.
Halmyris 2001, civil settlement, stray find.
Chronology: 4th – late 5th century AD.
vi. strap-ends [Pl. 3]
30. trapezoidal strap-end
Trapezoidal strap-end, rectangular in section; the narrower upper side decorated with
a square with crossing diagonals and furnished with a loop for attaching a hinge.
Copper alloy. Well preserved.
Length: 5.4 cm; width: 1 cm; max. thickness: 0.3 cm; weight: 9.2 g.
Halmyris 2001, civil settlement, stray find.
This type of strap-end consists of two elements joined by a hinge, with the top attached to
a strap by a rivet. Two strap-ends from Carnuntum119 are perfect analogies for the object
from Halmyris which retains only the lower side, while a fragment belonging to the upper
side of such a piece was discovered at Nicopolis ad Istrum.120
Chronology: 2nd century AD.
31. strap-end
fragmentary rectangular strap-end, square in section; loop in the narrower upper side.
Copper alloy. fragmentary, two pieces.
Length: 3.4 cm; width: 0.5 cm; max. thickness: 0.3 cm; weight: 2.7 g.
Halmyris 1991, sector ?, square 5-7, – 0.45 m; ICEM, inv. 43706.
Although it is characterized by reduced dimensions and different shape, this strap-end
may be associated with several early copies widespread in the Empire.121 It was attached
by leather strap via a double metal plate, sealed with a rivet. This type of strap-end was
used as a belt accessory or cavalry equipment.
Chronology: second half of 2nd – 3rd century AD.
119
120
GRüNEWALD 1981, p. 25, pl. 17/17–18.
POULTER 2007, pp. 23–24, no. 2.58, fig. 2.5.
121
OLDENSTEIN 1976, pp. 142–147, pl. 36.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 186
186
32. heart-shaped strap-end
Keller: ‘Riemenzungen in Herz form’, variant a–b.
Sommer: form A.
Soupault: type VI/1.
fragmentary strap-end (only the bottom is preserved). The body is in the shape of the
heart gradually narrowed to the top. Usually, the strap end ornaments consist of circular
or animal eye motifs. five concentric circles with a dot in the middle are represented
on the piece from Halmyris. Two knobs are preserved on the top, showing that there was
a hinge to connect the piece with the leather-strap.
Length: 3.3 cm; width: 2.4 cm; max. thickness: 0.1 cm; weight: 2.9 g.
Halmyris 2001, civil settlement, stray find.
E. Keller distinguishes two versions of the heart-shaped strap-ends: one with fixed plate
and the other with a plate with hinge. C. J. Simpson on the other hand believes that they
were all attached by a double plate with rivets.122 The latter notes relatively few variations
in the form and ornamentation of the finds.123 However, it should be noted that these are
simple pieces, easy to copy and duplicate, and the unity of form of specimens from different centres of the Empire can be taken as evidence for a broad promulgation of a common pattern ranging in time and produced in many other centres. The fact that some were
found in closed contexts together with the Zwiebelknopffibeln and that their range includes
large zones demonstrates that at least some types of strap-ends can be considered typical
pieces of military equipment or belonging to those who worked in the administrative apparatus. Numerous discoveries of strap-ends have been made in the Danube provinces,
some of them in contexts dated by coins, as in Intercisa,124 Siscia,125 Salona126 and Viminacium.127 In Scythia, heart-shaped strap-ends were discovered in the necropolis of Callatis
(dated by coins of Constantius II, 351–361, and Valens, 367–375)128 and (L)Ibida.129 One
of the latest heart-shaped strap-ends was found in Grave C121 at Beroe, on the Danubian
limes; the date was based on the presence of a Zwiebelknopffibel Keller/Pröttel 6 inside
the same inventory.130 In graves C67 and C38, two ‘bow’-shaped buckles had two strapends attached by rivets as an extension of the rectangular plate.131
Chronology: 4th – beginning of the 5th century AD.
33. heart-shaped strap-end
full heart-shaped strap-end; serrated area with a hole for rivet fastening in the upper part.
Copper alloy. Well preserved.
Length: 2.7 cm; width: 2.6 cm; thickness: 0.1 cm; weight: 3.2 g.
Halmyris 2009, civil settlement, stray find.
Similar to the previous object.
Chronology: 4th – beginning of the 5th century AD.
122
KELLER 1971, p. 64; SIMPSON 1976, p. 201.
SIMPSON 1976, p. 202.
124
ALföLDI 1957, pp. 461, 463, pl. 103/29.
125
KOŠčEVIć 1995, p. 17, pl. 26/227; RADMAN-LIVAJA 2004, p. 98, pl. 50/351.
126
VIŠIć-LJUBIć 1994, p. 231, no. 4.
127
ŠPEHAR 2007, p. 272, fig. 2.
123
128
PREDA 1980, pp. 43, 85, 106, 114, pl. 16 – Graves
10, 260, 356, 359.
129
OPAIţ 1991, pp. 46–47, no. 51, fig. 18.
130
PETRE 1987, p. 61, pl. 108bis/163f; on the
chronology of Keller/Pröttel 6, see PRöTTEL 1988
(1991), pp. 369–372; BUORA 1997, pp. 247–260.
131
PETRE 1987, pls. 104/151d, 105/153b.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 187
187
34. amphora-shaped strap-end
Keller: ‘Riemenzungen in Amphoraform’, variant a–b.
Sommer: form B.
Soupault: type VI/II.
fragmentary strap-end (only the bottom is preserved). Body is in the shape of a cylindrical
amphora with oblong and conical leg ornamented with lines in relief. In the upper side
fragments of ansae fretted as peltae are preserved.
Copper alloy. Well preserved.
Length: 3.9 cm; width: 2.2 cm; max. thickness: 0.2 cm; weight: 2.6 g.
Halmyris 2001, civil settlement, stray find.
C. J. Simpson has established three variants for these amphora-shaped strap-ends
(or ‘Riemenzungen in Amphora-förmig’): cylindrical amphora-shaped, attached by
a hinge; cylindrical amphora-shaped attached by rivets; globular amphora-shaped attached
by rivets.132 Analyzing the finds from the Museum of Aquileia, M. Buora also established
three variants, depending on the manner of strap attachment: by hinge, by rivets or with
collar used for the strap’s direct attachment.133 In terms of chronology, amphora-shaped
strap-ends are dated to the same period as the heart-shaped strap-ends, as E. Keller demonstrated. Based on coins found in a series of graves, he concluded that they were in circulation in the second half of the 4th century AD, a dating which C. Preda subscribed too as
well.134 A strap-end discovered at Beroe might be one of the earliest attested (found together with a Constantine I coin),135 besides a series of discoveries from Pannonia.136 The
distribution range is broad, as in the case of the previous type;137 many finds were discovered at Intercisa,138 Siscia139 (where they attest to local production), Viminacium,140
Novae141 and in Scythia at Callatis,142 (L)Ibida143 and Capidava.144 Although small chronological differences between heart- and amphora-shaped strap ends were demonstrated,
these pieces were discovered in similar contexts and in many cases in association with
archaeological material of the same chronology.
Chronology: second half of the 4th – first decade of the 5th century AD.
35. upper part of an (amphora-shaped?) strap-end
fragmentary strap-end (only the upper side attached to the leather is preserved). The upper
end is serrated and has two rivets on it. Two holes in the shape of schematic peltae form
the amphora’s neck and handles. The bottom is broken.
Copper alloy. Well preserved.
Length: 2.5 cm; width: 1.7 cm; max. thickness: 0.4–0.5 cm; weight: 4.7 g.
Halmyris 2000, civil settlement, stray find.
132
SIMPSON 1976, p. 198.
BUORA 2002, pp. 195–196, pl. 5/45–53.
134
KELLER 1971, p. 65; PREDA 1980, p. 43.
135
PETRE 1987, p. 60 – Grave C77, pl. 108/160b.
136
KELLER 1971, p. 65.
137
SOUPAULT 2003, pp. 49–50.
138
ALföLDI 1957, pp. 461, 463–464, pl. 103/5, 12.
139
KOŠčEVIć 1995, p. 17, pl. 26/223–226; RADMANLIVAJA 2004, p. 99, pl. 50/362.
133
140
TOMIć 1996, pp. 197–200.
GENčEVA 2000, p. 69, nos. 7–8, pl. 5/11–12.
142
PREDA 1980, p. 43, pl. 16/M324.
143
OPAIţ 1991, p. 47, no. 52, fig. 18.
144
COVACEf 1995–1996, pp. 101–102, nos I.3.1–
I.3.2, pl. 3/3, 5.
141
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 188
188
Morphological characteristics and parallels suggest that the fragment probably belonged
to an amphora-shaped strap-end. Similar to the previous object.
Chronology: second half of 4th – first decade of 5th century AD.
vii. belt appliqués [Pl. 3]
36. belt appliqué
Martinovka type.
Shield-shaped appliqué with wavy edges and the bottom ending in a sharp angle. Two rivets
are preserved on the back. The open-work plate depicts a schematic mask consisting of
circular and elongated holes. Cast in a bivalve mould.
Silver. Excellent preservation.
Length: 2.9 cm; width: 2 cm; thickness of the plaque: 0.1 cm; weight: 3.1 g.
Halmyris, unknown year and area/context; ICEM, inv. 44929.
The appliqué in question belongs to a well known and widespread group of belt sets; both
finds from Halmyris share a mask-shaped openwork decoration. In general, the Martinovka
type was produced with one of three techniques: by openworking, by engraving and by
repoussé stamping. The range of the type includes areas controlled by the Avars, the south
Danubian region, the Carpathian area, northern Black Sea, Caucasus and up to the Euphrates and other areas of Europe; it is therefore seen as a ‘multi-ethnic’ or ‘international’
element.145
The chronology of different variants of the Martinovka type has been widely analyzed by
Cs. Bálint.146 It has also been suggested that this type of artifact appeared under double,
late Roman and Sassanid influence.147 This type of appliqués is generally dated from the
second quarter of the 6th century AD until the second half of the 7th century AD.
Discoveries from Dobrudja come from the cemeteries in Beroe and Callatis, Tropaeum
Traiani and unknown contexts at (L)Ibida and Argamum. Among the appliqués from
Beroe, those found in Grave B14 are the most important because of associated archaeological material offering a complete picture of extremely rich grave equipment.148 In Cs.
Bálint’s opinion, the belt set discovered in the aforementioned grave is an indication of
the deceased’s steppe origin, an opinion also based on the decorative characteristics of the
pieces assigned to Crimean workshops. A further argument is an ornament of bone on
a small bag (Taschenverschluß) with good analogies in Avar cemeteries.149 Earlier excavations at Halmyris already produced such a find.150 At Callatis, Grave 132 which included
a silver strap-end, was well dated by a gold coin of Justinian (538–545),151 and at Tropaeum
Traiani a Schulze-Dörrlamm H1 appliqué type was dated around 586.152 The archaeological context of appliqués from (L)Ibida is unknown.153 The same is true of the fischwanzförmigen appliqué type from Argamum, which is similar to the belt appliqué found inside
145
150
146
151
BáLINT 1992, p. 392.
BáLINT 1992, pp. 389–411.
147
KAzANSKI 2003, p. 41.
148
PETRE 1987, p. 71, pls. 130–130bis.
149
BáLINT 1992, p. 405.
ICEM, inv. 43646.
PREDA 1980, p. 72, no. 91.
152
BOGDAN-CăTăNICIU, BARNEA 1979, pl. 165/10.22;
SCHULzE-DöRRLAMM 2009, pp. 265–268.
153
OPAIţ 1991, pp. 46–47, nos. 53–54, 56–57, fig. 18.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 189
189
the Apsile grave from Verchnaja Ešera, in Abkhazia154 and at Dibsi faraj (in Syria, on the
Euphrates),155 the latter being the closest analogy. Mapping of Scythian finds shows even
distribution: two cities on the limes, two on the Black Sea coast and inland, at (L)Ibida
and Tropaeum Traiani. Close parallels for the appliqué from Halmyris can be found at
(L)Ibida (unknown context)156 and čmi (Grave XI),157 but these were made of bronze.
Chronology: end of 6th century – 7th century AD.
37. belt appliqué
Martinovka type.
Rectangular-shaped appliqué with rounded bottom; the ornament is represented by a cruciform motif and a series of holes forming a schematic human face. Cast in a mould.
Copper alloy. Well preserved.
Length: 2.5 cm; width: 1.9–2 cm; thickness: 0.1 cm; max. thickness: 0.5 cm; weight: 1.8 g.
Halmyris 2001, civil settlement, stray find; ICEM, inv. 45651.
Although the ornament is different, the general shape reminds us of the example from the
Apsile grave in Verchnaja Ešera; it is also sketched, but has a rounded rectangular plate at
the bottom.158 A close find was discovered at ‘Kaleto’ near Sredec (Burgas County).159
Chronology: end of 6th century – 7th century AD.
154
BáLINT 1992, pp. 384–389, pl. 51/1.
KAzANSKI 2003, p. 43, fig. 5/3.
156
OPAIţ 1991, pp. 46–47, no. 57, fig. 18.
155
157
AMBROz 1971, p. 115, fig. 5/63.
BáLINT 1992, p. 385, pl. 51/6.
159
BALBOLOVA-IVANOVA 2004–2005, p. 93, fig. 9/1.
158
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 190
190
Pl. 1. Buckles with movable, hinged plate (nos. 1–12)
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 191
191
Pl. 2. Single-piece buckles with cast plate (nos. 13–20)
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 192
192
Pl. 3. Simple frame buckles (nos. 21–27); varia (no. 28); miscellaneous fragments (no. 29);
strap-ends (nos. 30–35); belt appliqués (nos. 36–37)
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 193
193
instead of a conclusion
The main difficulty in studying these artifacts in Scythia is in the end a lack of extensive
publication of belt-equipment from this area; in some cases it is also necessary to re-evaluate
the earlier discoveries. Data on this type of finds is therefore limited from many sites of the
province and a direct consequence is the absence of similar contexts to compare the material
with. Unfortunately, I did not have the opportunity to use archaeological context for dating
belt equipment and was not able to copy for Halmyris James Russell’s excellent study of
everyday life for the city of Anemurium.160
Belt accessories from Halmyris are obviously a heterogeneous group showing settlement
evolution during the late Roman period. At the same time, they are evidence of changes of
equipment and dress in the course of more than three centuries. The discussion obviously
cannot clarify issues relating to ethnic groups inside the city or refine the chronology of belt
sets from the Lower Danube. A serious risk exists especially when trying to define ethnic
groups based on these finds. Martinovka type appliqués suggest the presence at Halmyris of
riders coming from the North Pontic steppes in the late 6th and early 7th century AD, but
keeping in mind the ‘multiethnic’ character of these finds as pointed out by Cs. Bálint places
the discussion in a broader context.
The frequency of belt sets in certain contexts has contributed to the allocation of certain
types and variants to either male or female dress. Adopting this methodology may involve
serious errors by assigning finds to one group to the exclusion of another. In the case of the
finds from Halmyris I did not made this distinction because I felt that there are numerous
exceptions, even if at times it is possible to speak of specific male or female belt sets. The
exclusive attribution of belt sets to military equipment should be noted. Until recently this
was generally the idea in the case of the Zwiebelknopffibeln type, regarded as specific to
military equipment. This is correct to a point. The mixed military and civil character of these
fibulae was rightly noted on various occasions. Given the above we may also identify strapends as finds of a military type, but it is dangerous to consider most buckles as well, especially as the issue of later type was not discussed too often.
Overall, compared to finds discovered at other Scythian archaeological sites and even in
other provinces, the finds from Halmyris represent a large group. Another 26 finds,161 together
with a Pápá buckle type,162 should be added to the finds published so far. This is obviously
one of the largest assemblages published from the province of Scythia so far. Not incidentally,
the number of finds increases with the century until a veritable ‘explosion’ in the late Roman
period, a situation which was attributed by J. Russell163 to a subtle change in antique costume
meaning the replacement of ‘classical’ and ‘regular’ dress elements such as the chlamys and
hymation with the barbarous fashion of wearing trousers (braccae). The presence of military
units with their equipment requiring fastening devices like buckles must have contributed to
a certain extent, as did population flow, which is difficult to quantify in archaeological terms.
In the end, it is the sum of all the above assumptions which explains, at least on a primary
level, the presence of various types and the diverse ethnic origins of the belt equipment from
Halmyris.
160
161
RUSSELL 1982, pp. 133–154.
See note 4.
162
163
MADGEARU 2004, pp. 169–176.
RUSSELL 1982, pp. 145–146.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 194
194
acknowledgments:
As always, I would like to express my gratitude towards Dr. Mihail zahariade (Bucharest)
for his generosity in offering me the opportunity to study the belt-buckles and generally
metal objects discovered in the extramural area of the ancient fortress. Also, I am much
obliged for offering me the belt-buckles discovered in the last years during his archaeological researches inside the fortification. I wish to thank to Prof. Emeritus James Russell
(Vancouver), Dr. Christian Miks (Mainz), PhD Student Sever P. Boţan (Iaşi), Dr. Damien
Glad (Paris), Dr. Vladimir Varsik (Nitra), Dr. Christoph Eger (Munich) who kindly
delivered to me a part of the bibliography to support the present article. Many thanks to
Dr. Martin Lemke (Warsaw) for his kindness and willingness to publish this study in the
prestigious journal Novensia.
abbreviations
ICEM
Institutul de Cercetări Eco-Muzeale (Institute for Eco-Museum
Researches), Tulcea
bibliography
AIBABIN 1990
ALADJOV 1989
ALföLDI 1957
AMBROz 1971
APARASCHIVEI, CHIRIAC 2009
AURRECOECHEA fERNáNDEz 2001
BALBOLOVA-IVANOVA 2004–2005
BáLINT 1992
A. I. AIBABIN, “Chronologija mogilnikov Krima pozdnerimskogo
i rannesrednevekovogo vremeni”, Materialy po arheologii, istorii
i etnografii Tavrii 1, pp. 3–86, 175–241.
D. ALADJOV, Arheologičeski pametnici v raiona na Stambolovo,
Sofia.
M. R. ALföLDI, “Schnallen, Riemenzungen und andere Gürtelverzierungen”, [in:] M. R. ALföLDI, L. BARKóCzI, J. fITz, K. Sz.
PóCzy, A. RADNóTI, A. SALAMON, K. SáGI, J. SzILáGyI, E. B. VáGó,
Intercisa II (Dunapentele). Geschichte der Stadt in der Römerzeit
(= Archaeologia Hungarica 26), Budapest, pp. 456–476.
A. K. AMBROz, “Problemy rannesrednevekovoj chronologii Vostočnoj Evropy”, Sovetskaja Archeologija 2, pp. 96–123.
D. APARASCHIVEI, C. CHIRIAC, “Objets vestimentaires de Dobroudja
romaine-byzantine”, [in:] Studia Antiqua et Medievalia. Miscellanea in honorem annos LXXV peragentis Professoris Dan Gh.
Teodor oblata, ed. D. APARASCHIVEI, (= Honoraria 5), Bucharest,
pp. 69–78.
J. AURRECOECHEA fERNáNDEz, Los cinturones romanos en la Hispania del Bajo Imperio (= Monographies Instrumentum 19), Montagnac.
M. BALBOLOVA-IVANOVA, “Drebni metalni predmeti ot kăsnoantičnata
i srednovekovna krepost ‘Kaleto’ v grad Sredec, Burgaska oblast”,
Izvestija na Muzeite ot Jugoiztočna Bălgarija 21, pp. 91–100.
CS. BáLINT, “Kontakte zwischen Iran, Byzanz und der Steppe.
Das Grab von üč Tepe (Sowj. Azerbajdžan) und der beschlagverzierte Gürtel im 6. und 7. Jahrhundert”, [in:] Awarenforschungen I, ed. f. DAIM, Vienna, pp. 309–496.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 195
195
BARNEA 1977
BEJAN 1976
BISHOP 2002
BOGDAN-CăTăNICIU, BARNEA 1979
BUORA 1997
BUORA 2002
BURGER 1966
CAVALLARI 2005
COVACEf 1995–1996
CRAWfORD 1990
CRNOBRNJA 1997
DASKALOV, DUMANOV 2004
DASKALOV, TRENDEfILOVA 2006
DAVIDSON 1952
DAVIDSON, HORVáTH 1937
DIACONU 1974
DIACONU 1985
EGER 2001
EGER 2010
GACUTA 1987
GACUTA 1993
I. BARNEA, Les monuments paléochrétiens de Roumanie, Vatican
City.
A. BEJAN, “Un atelier metalurgic din sec. VI e.n. de la DrobetaTurnu Severin”, Acta Musei Napocensis 13, pp. 257–278.
M. C. BISHOP, Lorica segmentata, I: A Handbook of Articulated
Roman Plate Armour (= Journal of Roman Military Equipment
Studies Monograph 1), Braemar.
I. BOGDAN-CăTăNICIU, A. BARNEA, “Ceramica şi descoperiri
mărunte”, [in:] A. BARNEA, I. BARNEA, I. BOGDAN-CăTăNICIU, M.
MăRGINEANU-CâRSTOIU, Gh. PAPUC, Tropaeum Traiani, I: Cetatea,
Bucharest, pp. 177–226.
M. BUORA, “‘zwiebelknopffibeln’ del tipo Keller 6 da Aquileia”,
Arheološki vestnik 48, pp. 247–260.
M. BUORA, “Militari e Militaria ad Aquileia e nell’attuale friuli”,
[in:] Miles Romanus dal Po al Danubio nel Tardoantico. Atti del
Convegno internazionale Pordenone — Concordia Sagittaria, 17–
19 marzo 2000, ed. M. BUORA, Pordenone, pp. 183–206.
A. Sz. BURGER, “The late Roman cemetery at Ságvár”, Acta Archaeologica Academiae Scientiarum Hungaricae 18, pp. 99–234.
C. CAVALLARI, Oggetti di ornamento personale dall’Emilia Romagna bizantina: i contesti di rinvenimento (= Studi e Scavi, nuova
serie 13), Bologna.
z. COVACEf, “Accesorii vestimentare, de toaletă şi podoabe descoperite în sectorul estic al cetăţii Capidava”, Pontica 28–29, pp.
95–120.
J. S. CRAWfORD, The Byzantine Shops at Sardis (= Archaeological
Exploration of Sardis. Monograph 9), Cambridge, Mass. – London.
A. CRNOBRNJA, “Pojasne kopče iz Muzeja Grada Beograda”, Glasnik Srpskog arheološkog društva 13, pp. 301–312.
M. DASKALOV, B. DUMANOV, “Metalni predmeti ot perioda VI–VII
v. văv fonda na IM — Haskovo”, Izvestija (Haskovo) 2, pp. 193–
206.
M. DASKALOV, K. TRENDEfILOVA, “Novi postăplenija metalni nakiti
ot V–VI v. văv fonda na IM — Pernik”, Izvestija (Kiustendil) 13,
pp. 381–391.
G. R. DAVIDSON, Corinth XII. The Minor Objects I–II, Princeton.
G. R. DAVIDSON, T. HORVáTH, “The Avar invasion of Corinth”,
Hesperia 6/2, pp. 227–240.
P. DIACONU, “Obiecte romane târzii de la Pârjoaia”, Studii şi
cercetări de istorie veche şi arheologie 25/2, pp. 289–294.
P. DIACONU, “Câteva fibule din epoca romană şi romană târzie de
la Păcuiul lui Soare şi Dervent (judeţul Constanţa)”, Cultură şi civilizaţie la Dunărea de jos 1, pp. 87–92.
Ch. EGER, “Gürtelschnallen des 6. bis 8. Jahrhunderts aus der
Sammlung des Studium Biblicum franciscanum”, Liber Annuus
51, pp. 337–350.
CH. EGER, “Byzantinische Gürtelschnallen aus Nordafrika — ein
typologischer überblick”, Ephemeris Napocensis 20, pp. 129–168.
W. GACUTA, “Przedmioty metalowe z Novae — kampanie
wykopaliskowe z lat 1960–1969, 1971, 1973, 1975 i 1977”, Novensia 1, pp. 75–175.
W. GACUTA, “Przedmioty metalowe z Novae. 6 kampanii
wykopaliskowych Ekspedycji Archeologicznej Uniwersytetu
Warszawskiego w latach 1977–1987”, Novensia 5, pp. 7–178.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 196
196
GARAM 2001
GăzDAC 1995
GENčEVA 2000
GENčEVA 2009
GILL 1986
GRAMATOPOL 1982
GRüNEWALD 1981
HARALAMBIEVA 1993
JANKOVIć 1981
KAzANSKI 2003
KELLER 1971
KOŠčEVIć 1995
LAzAROV 2001
LIGHTfOOT 2003
MADGEARU 1997
MADGEARU 1998
MADGEARU 2004
MARIJANSKI-MANOJLOVIć 1987
MINčEV 1994–1995
E. GARAM, Funde byzantinischer Herkunft in der Awarenzeit vom
Ende des 6. bis zum Ende des 7. Jahrhunderts (= Monumenta
Avarorum Archaeologica 5), Budapest.
C. GăzDAC, “Bronzuri romane de la Gherla”, Acta Musei Napocensis 32, pp. 401–430.
E. GENčEVA, “Metalowe części wyposażenia żołnierskiego z Novae”,
Novensia 12, pp. 49–98.
E. GENčEVA, “Rimskijat ajuren kolan ot Bălgarija”, Arheologija
50/3–4, pp. 11–27.
M. V. GILL, “The small finds”, [in:] Excavations at Saraçhane in
Istanbul, I: The Excavations, Structures, Architectural Decoration,
Small Finds, Coins, Bones, and Molluscs, ed. R. M. Harrison, Princeton, pp. 226–277.
M. GRAMATOPOL, Dacia Antiqua. Perspective de istoria artei şi teoria culturii, Bucharest.
M. GRüNEWALD (mit einem Beitrag von E. WEBER), Die Kleinfunde
des Legionslagers von Carnuntum mit Ausnahme der Gefäßkeramik (Grabungen 1968–1974) (= Der römische Limes in Österreich 31), Vienna.
A. HARALAMBIEVA, “Kolanni toki ot IV–VII v. ot Dobričkij Muzej”,
Dobrudža 10, pp. 32–46.
D. JANKOVIć, Podunavski deo oblasti Akvisa u VI i nočetkom VII
veka (= Arheološki Institut, Kniga 5), Belgrade.
M. KAzANSKI, “Plaques-boucles et garnitures de ceintures du Haut
Moyen âge en Syrie (milieu du VIe siècle – IXe siècle)”, [in:]
M. KAzANSKI, avec la collaboration de P-M. BLANC, S. BOULOGNE,
J.-P. SODINI, V. SOUPAULT-BECqUELIN, Qal’at Sem’an, IV: Rapport
final, fascicule 3: Les objets métalliques (= Institut français du
Proche-Orient, Bibliothèque archéologique et historique 167),
Beirut, pp. 35–48.
E. KELLER, Spätrömischen Grabfunde in Südbayern (= Münchner
Beitrage z. Vor- und Frühgeschichte 14), Munich.
R. KOŠčEVIć, “finds and metalwork production”, [in:] R. KOŠčEVIć, R. MAKJANIć, Siscia. Pannonia Superior. Finds and Metalwork
Production. Terra sigillata (= BAR International Series 621), Oxford, pp. 16–42.
L. LAzAROV, Istoričeski Muzej — Dălgopol. Dreven bronz katalog,
Spasena.
M. LIGHTfOOT, “Belt buckles from Amorium and in the Afyon
Archaeological Museum”, [in:] Amorium Reports, II: Research
Papers and Technical Reports, ed. C. S. LIGHTfOOT (= BAR International Series 1170), Oxford, pp. 81–103.
A. MADGEARU, Continuitate şi discontinuitate culturală la Dunărea
de Jos în secolele VII–VIII, Bucharest.
A. MADGEARU, “The Sucidava type of buckles and the relations between the Late Roman Empire and the Barbarians in the 6th century”, Arheologia Moldovei 21, pp. 217–222.
A. MADGEARU, “A buckle of Pápa type found in the Early Byzantine fortress Halmyris (Murighiol, Tulcea County)”, Peuce S.N. 2,
pp. 169–176.
M. MARIJANSKI-MANOJLOVIć, Rimska nekropola kod Beške u Sremu
(= Vojvodanski muzej. Posebna izdanja 8), Novi Sad.
A. MINčEV, “Kasnoantični toki s hristijanski nadpisi văv Varnenskija
Arheologičeski Muzej”, Izvestija (Varna) 30–31 (45–46), pp. 18–23.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 197
197
NUţU 2009
NUţU 2010a
NUţU 2010b
NUţU, IACOB 2008
OANţă-MARGHITU 2006
OLDENSTEIN 1976
OPAIţ 1991
PETRE 1987
POULTER 2007
PREDA 1980
PRöTTEL 1988 (1991)
RADMAN-LIVAJA 2004
RAŠKOVIć 2007
RUSSELL 1982
SCHULzE-DöRRLAMM 2002
G. NUţU, “A group of strongly profiled brooches from Halmyris
(Moesia Inferior)”, Studia antiqua et archaeologica 15, pp. 173–190.
G. NUţU, “Le ‘zwiebelknopffiblen’ da Halmyris (Provincia
Scythia)”, Quaderni friulani di archeologia 20, pp. 99–108.
G. NUţU, “Roman brooches from Halmyris (Moesia Inferior), I”,
[in:] Rekata i vremeto. Sbornik izsledvanija, ed. V. ATANASOV, Tutrakan, pp. 64–85.
G. NUţU, M. IACOB, “fibbie romane d’Argamum (Moesia Inferiore
/ Scythia). Settore ‘Extra Muros’”, Annuario del’Istituto Romeno
di Cultura e Ricerca Umanistica di Venezia 10, in print.
R. OANţă-MARGHITU, “Argamum între Imperiul roman târziu
şi ‘barbari’. Obiectele mărunte ca ipostaze ale comunicării”, [in:]
À la recherche d’une colonie. Actes du Colloque International
40 ans de recherche archéologique à Orgamè / Argamum, Bucarest
– Tulcea – Jurilovca 3–5 octobre 2005, ed. M. MăNUCUADAMEŞTEANU (= Orgame / Argamum, Supplementa 1), Bucharest,
pp. 345–373.
J. OLDENSTEIN, “zur Ausrüstung römischer Auxiliareinheiten. Studien zu Beschlägen und zierat an der Ausrüstung der römischen
Auxiliareinheiten des obergermanisch-raetischen Limesgebietes
aus dem zweiten und dritten Jahrhundert n. Chr.”, Berichte der
Römisch-Germanischen Kommission 57, pp. 49–284.
A. OPAIţ, “O săpătură de salvare în oraşul antic Ibida”, Studii şi
cercetări de istorie veche şi arheologie 42/1–2, pp. 21–56.
A. PETRE, “La romanité en Scythie Mineure (IIe–VIIe siècles de
n.è.). Recherches archéologiques”, Bulletin AIESEE 17–18, pp. 5–
171.
A. G. POULTER, “The metalwork”, [in:] A. G. POULTER with contributions from M. J. BEECH, T. f. C. BLAGG, z. BOEV, H. BUSH,
J. L. BUySSE, J. CHAPMAN, R. K. fALKNER, M. HENIG, B. IRVING,
J. KENWORTHy, S. A. PARfITT, A. ROBERTS and C. SALTER, Nicopolis
ad Istrum. A Late Roman and Early Byzantine City. The Finds and
the Biological Remains, Oxford, pp. 15–64.
C. PREDA, Callatis. Necropola romano-bizantină, Bucharest.
Ph. M. PRöTTEL, “zur Chronologie der zwiebelknopffibeln”,
Jahrbuch des Römisch-Germanischen Zentralmuseums, Mainz 35,
pp. 347–372.
I. RADMAN-LIVAJA, Militaria Sisciensia — Nalazi rimske vojne
opreme iz Siska u fundusu Arheološkoga muzeja u Zagrebu (= Musei
Archaeologici Zagrabiensis Catalogi et Monographiae 1), zagreb.
D. RAŠKOVIć, “Rimsko naselje Praesidium Pompei — Rimlani
u aleksinačkoj kotlini”, [in:] Niš i Vizantija V, ed. M. RAKOCIJA,
Niš, pp. 205–226.
J. RUSSELL, “Byzantine instrumenta domestica from Anemurium:
the significance of context”, [in:] City, Town and Countryside in
the Early Byzantine Era, ed. R. L. HOHLfELDER, New york, pp.
133–154.
M. SCHULzE-DöRRLAMM, Byzantinische Gürtelschnallen und
Gürtelbeschläge im Römisch-Germanischen Zentralmuseum, I: Die
Schnallen ohne Beschläg, mit Laschenbeschläg und mit festem
Beschläg des 5. bis 7. Jahrhunderts (= Römisch-Germanisches Zentralmuseum. Kataloge vor- und frühgeschichtlicher Altertümer 30),
Mainz.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 198
198
SCHULzE-DöRRLAMM 2009
M. SCHULzE-DöRRLAMM, Byzantinische Gürtelschnallen und
Gürtelbeschläge im Römisch-Germanischen Zentralmuseum, II: Die
Schnallen mit Scharnierbeschläg und die Schnallen mit angegossenem Riemendurchzug des 7. bis 10 Jahrhunderts (= Römisch-Germanisches Zentralmuseum. Kataloge vor- und frühgeschichtlicher Altertümer 30/2), Mainz.
SCORPAN 1978
C. SCORPAN, “Descoperiri arheologice diverse de la Sacidava”,
Pontica 11, pp. 155–180.
SIMPSON 1976
C. J. SIMPSON, “Belt-buckles and strap-ends of the Later Roman
Empire: a preliminary survey of several new groups”, Britannia 7,
pp. 192–223.
SIMPSON 1997
C. J. SIMPSON with contributions from R. REECE, J. J. ROSSITER, The
Excavations of San Giovanni di Ruoti, II: The Small Finds (= Phoenix
Suppl. 34), Toronto – Buffalo – London.
SOMMER 1984
M. SOMMER, Die Gürtel und Gürtelbeschläge des 4. und 5. Jahrhunderts im römischen Reich (= Bonner Hefte zur Vorgeschichte 22).
SOMOGyI 1987
P. SOMOGyI, “Typologie, Chronologie und Herkunft der Maskenbeschläge. zu den archäologischen Hinterlassenschaften osteuropäischer Reiterhirten aus der pontischen Steppe im 6. Jahrhundert”, Archaeologia Austriaca 71, pp. 121–154.
SOUPAULT 2003
V. SOUPAULT, Les éléments métalliques du costume masculin dans
les provinces romaines de la mer Noire IIIe–Ve s. ap. J.-C. (= BAR
International Series 1167), Oxford.
ŠPEHAR 2007
P. ŠPEHAR, “Nalazi metalnih delova vojnog pojasa sa teritorije
Viminačijuma”, [in:] Niš i Vizantija V, ed. M. RAKOCIJA, Niš,
pp. 269–290.
ŠPEHAR 2010
P. ŠPEHAR, Materijalna kultura iz ranovizantijskih utvrđenja u Đerdapu (= Cahiers des Portes de Fer. Monographies 7), Belgrade.
SUCEVEANU, zAHARIADE, TOPOLEANU 2003 – “Stratigrafia”, [in:] A. SUCEVEANU, M. zAHARIADE, f. TOPOLEANU, Gh. POENARU BORDEA, Halmyris I. Monografie arheologică,
Cluj-Napoca, pp. 27–41.
TABAKOVA-CANOVA 1981
G. TABAKOVA-CANOVA, “Kasnoantičnijat nekropol v mestnostta
Strajata kraj grad Pleven”, Izvestija na Nationalnija Istoričeski
Muzej 3, pp. 102–184.
TEODOR 1991
D. Gh. TEODOR, “Piese vestimentare bizantine din secolele VI–VIII
în spaţiul carpato-dunăreano-pontic, A. Catarame cu placa fixă”,
Arheologia Moldovei 14, pp. 117–138.
TOMIć 1996
M. TOMIć, “Bronzani pojasni jezičti u obliku stilizovanih amfora.
Slučajni nalazi sa šire teritorije Viminacijuma i Marguma”, Glasnik
Srpskog arheološkog društva 12, pp. 197–206.
TUDOR 1938
D. TUDOR, “Sucidava I. Première compagne de fouilles et
recherches archéologiques dans la forteresse romaine de Celei, département de Romanaţi”, Dacia 5–6, 1935–1936, pp. 387–422.
UENzE 1966
S. UENzE, “Die Schnallen mit Riemenschlaufe aus dem 6. und 7.
Jahrh.”, Bayerische Vorgeschichtsblätter 31/1–2, pp. 142–181.
UENzE et alii 1992
S. UENzE mit Beiträgen von K. DIETz, J. JURUKOVA, G. KUzMANOV,
M. MACKENSEN, H. TODOROVA, P. VALEV, V. P. VASILEV, V. VELKOV,
J. WERNER, Die Spätantiken Befestigungen von Sadovec (Bulgarien). Ergebnisse der deutsch-bulgarisch-österreichischen Ausgrabungen 1934–1937 (= Münchner Beiträge zur Vor- und
Frühgeschichte 43), I–II, Munich.
VANNESSE 2008
M. VANNESSE, “Le fibule a croce latina: il contributo dell’archeologia per la storia militare dell’Italia settentrionale durante la tarda
antichità (secoli IV–V)”, Quaderni friulani di archeologia 18, pp.
155–165.
22_Novensia_final_01_21_Novensia 7/13/12 4:10 PM Page 199
199
VARSIK 1992
VINSKI 1967 (1974)
VIŠIć-LJUBIć 1994
WALDBAUM 1983
žIVIć 2007
V. VARSIK, “Byzantinische Gürtelschnallen im mittleren und unteren Donauraum im 6. und 7. Jahrhundert”, Slovenská Archeológia
40/ 1, pp. 77–108.
z. VINSKI, “Kasnoantički starosjedioci u Salonitanskoj regiji prema
arheološkoj ostavštini predslavenskog supstrata”, Vjesnik za arheologiju i historiju dalmatinsku 69, pp. 5–86.
E. VIŠIć-LJUBIć, “Pojasne kopče i okovi”, [in:] Salona Christiana,
ed. E. MARIN, Split, pp. 227–238.
J. C. WALDBAUM, Metalwork from Sardis: the Finds through 1974
(= Archaeological Exploration of Sardis. Monograph 8), Cambridge, Mass. – London.
M. žIVIć, “Catalogue of small finds from excavations outside the
fortified palace of Romuliana (2005–2007)”, Starinar 57, pp. 277–
307.
streszczenie
sprzączki, koncówki pasów i apliki z halmyris (Moesia Inferior / Scythia)
Autor dokonuje przeglądu 37 dodatków ubraniowych pochodzących z naddunajskiego
stanowiska Halmyris położonego na pograniczu imperium rzymskiego. Większość
znalezisk pochodzi spoza obrębu fortyfikacji, z terenu przypuszczalnej osady cywilnej,
ale także z wnętrza miasta. zabytki ujawniono przy pomocy wykrywacza metali podczas
poszukiwań przeprowadzonych dwukrotnie, w 2000 i w 2001 roku. Podstawowym celem
było pokazanie funkcji zabytków, ustalenie typologii w oparciu o poszczególne elementy
(sprzączki, płytki, końcówki pasów, apliki) oraz zaproponowanie linii rozwojowej opartej
na analogiach pochodzących z różnych terenów przy jednoczesnym ukazaniu zasięgu.
George Nuţu
Institute for Eco-Museum Researches
Tulcea
nutugrg@yahoo.com