Byzantine Belt Ornaments of the 7th and 8th Centuries in Avar Contexts

Falko Daim
Introduction In many different cultures, regions and periods from the Bronze Age to the present day the belt is an important dress accessory.1 It holds trousers and coats together, is valuable jewellery, but can also carry meaning on a number of different levels. Above all, it is a symbol of social position, rank and wealth.2 This is true, in particular, for Early Byzantium. The East Roman belt fashion of the late 6th to 8th centuries developed from Late Roman military belts, 3 but was also influenced by stimuli from the Sasanian East, whereas influences from the steppe cultures did not play as great a role as had been previously assumed.4 As with other types of jewellery earrings, necklaces, bracelets Byzantine belts were highly attractive for many neighbouring peoples. Therefore, Byzantine belts have been found among the Lombards in Italy, as well as among the Avars, Bulgarians and Moravians. At the same time a diverse local production, sometimes very different from the Byzantine models, developed in these regions. Multi-part belt sets may be studied on a number of different levels. As belt decoration changes relatively quickly, the belt, like brooches, constitutes one of the bases of chronological systems. Also, strap-ends and fittings are often decorated with motifs, which, like large-scale reliefs, mosaics and wall paintings, may be analysed from an iconographical point of view. However, the production techniques used to manufacture belt ornaments are also interesting. With the help of technological studies, we can not only obtain information regarding workshop traditions and the technical abilities of the smiths and metal casters, but also about the systems of social values. It is not only the raw materials used which make a piece of jewellery precious; the techniques used and their combination may also have played an important role. Another important question concerns the social function of these belts. Who wore them and on what occasions? Were there, as in Early Medieval China and Japan, precise rules for their use? If we have no written sources on this topic for Byzantium, can archaeological contexts provide answers? Belts and belt sets are an ideal topic for historical research and cultural studies. With the help of a trans-disciplinary approach, belts can reveal something about social reality in the Early Middle Ages, as well as about the transfer of cultural elements and technologies over territorial borders.This paper will attempt to demonstrate the connections between Byzantine and Avar belts, focusing on the opportunities which the relatively reliable Avar chronology offers for dating Byzantine types. Avar imitations have frequently permitted us to identify and date their Byzantine models. It is nevertheless not possible in every case to determine with certainty whether a particular object is a Byzantine original or a barbarian imitation. If we consider the problem from a general point of view, then we can distinguish between four main types of product, ranging from Byzantine to of Byzantine type: 1. Products made on Byzantine territory, traded and used within the Empire. 2. Products made on Byzantine territory, but brought and used outside the Empire. Here we should mention, for example, diplomatic gifts made especially for this purpose. 3. Products made by Byzantine craftsmen outside the Empire: 3a. according to local tastes, 3b. according to Byzantine tastes. 4. Products made by barbarian craftsmen using Byzantine technology: 4a. according to local tastes (shape and/or motifs), or 4b. according to Byzantine tastes. Possible criteria for the identification of Byzantine objects are: 1. Shapes and motifs which, according to maps showing their distribution, were popular or accepted only within the Byzantine Empire, but rarely outside its borders. 2. Production techniques which, in Byzantium, were thought to enhance an objects value. 3. Evidence for the existence of workshops on Byzantine territory, such as casting moulds, half-finished products, and so on. 4. The use of raw materials, alloys etc., which may serve as evidence for the location of workshops on Byzantine territory. The idea that all valuable objects are Byzantine and all objects of lower quality are barbarian (for example, Avar), is definitely incorrect. Of course, even within the Roman Empire, cheaper and less elaborate objects were manufactured and perhaps this could complicate things there may have been a special production of low-quality objects for the markets outside of Byzantine territory. It is obvious that in each individual case, all available data, both antiquarian and scientific, must be evaluated and only if all observations point in the same direction can we expect to obtain more or less reliable results. However, the effort is certainly worth it. The detailed analysis of archaeological finds and their contexts gives us small insights into the interactions of Avar and medieval-Roman culture, indicating what was accepted, adopted, rejected or altered. We can also discern a number of different phases in the spectrum of Avar-Byzantine cultural contacts. Sources The written sources from the 6th to the 10th centuries mention belts on several occasions. The Latin panegyric In laudem Iustini, written by Corippus to celebrate the occasion of Emperor Justin IIs ascent to the throne, is particularly
Intelligible Beauty | 61

Daim impressive. In this poem, the author describes the careful stage management of an audience which Justin granted to an Avar delegation, only seven days after the beginning of his reign. The splendidly decorated throne was flanked on either side by two goddesses of victory, which held laurel wreaths over the emperors head. The emperor was dressed in splendid, pure white and purple clothing, with a gold coat, a gold belt and purple shoes made of Parthian leather.5 A gold belt is also mentioned in a murder mystery told by Theophylaktos Simokattes. A member of the imperial bodyguard goes hunting with a Gepid. The mans beautiful gleaming dress, his gold belt and the horses gold reins stir the Gepids greed. He killed his hunting partner and fled with his loot, but was later caught and executed. The moral conclusion drawn by Theophylaktos regarding this event is quite revealing. The bodyguards downfall was his gold jewellery, because the ornaments were permanently accompanied by envy and persecution.6 This appeal for modesty corresponds notably with the fact that Byzantine jewellery is characterised more by refined production techniques than by weight and monumentality. A miracle story of St Artemios, written in the late 7th century, also demonstrates the meaning of the belt in Early Byzantine society. This story relates the theft of festive clothes and a belt, as well as of the recovery of the stolen objects with the help of the saint.7 The Vita of Theodore of Sykeon, written before 650, also contains a great deal of information about life in the provinces. For example, it reports that the saint was equipped with expensive clothes and a gold belt when he was only six years old and sent by his mother to Constantinople, hoping that he would be accepted into the emperors service.8 Belts are mentioned in the Tractates and in De Administrando Imperii by Constantine VII Porphyrogenitus (91359). They belong among the gifts of different value which should be taken along on diplomatic missions: various purple and false-purple belts valued at 1 nomisma 4 miliarsia each; and others in false-purple at 1 nomisma, and others at 8 miliarsia.9 If a delegation is sent to the Pechenegs, gifts such as purple clothes, silk fabrics, pepper and authentic Parthian leather, as well as belts, should be taken along and distributed among the foreigners according to strict rules.10 Men with multi-part belts are depicted on a number of mosaics, wall paintings and bronze figures, which have been commented on in several publications recently.11 However, strictly speaking, the only information obtained from studying these sources is that belts with additional straps were worn in the Early Byzantine period in the eastern part of the Empire. The depiction of the family of the rich Roman, Theodotus, in the so-called chapel of Zachary (which he sponsored) in Sta Maria Antiqua in Rome (built between 741 and 752) is an exceptional case in several respects.12 Theodotus was twice consul and dux of Rome that is, he was supreme commander of the troops. Theodotus and his wife stand on either side of the Virgin Mary with, to their right, a little girl and, to their left, a boy, both of whom are marked as being alive by means of a rectangular halo. The boy, presumably Theodotus stepson, is wearing a kaftan closed with a multi-part belt. He later became Pope Hadrian, but when the donators picture was painted, he would have expected to follow in his fathers footsteps. Therefore, he is presumably depicted wearing the uniform of a
62 | Intelligible Beauty

dux, but certainly the dress appropriate for the social status of his family. The main sources for our topic, however, are archaeological finds, even if most of them were not discovered on former Byzantine territory, but outside the borders of the Byzantine Empire, from Lombard Italy, the Bavarian and Alamannic territories in the foothills of the northern Alps and in particular from the Avar Empire, in present-day Hungary and adjacent regions. Although it is likely that a number of Early Byzantine belt ornaments still remain to be identified in the storerooms of museums in Turkey, Greece and Italy, this presumably will not alter the general picture completely, because in the Byzantine Empire, jewellery was only rarely placed in graves, while burials outside the Byzantine borders were well furnished, both with dress ornaments and with grave goods in the real sense (for example: food), up to the middle of the 7th century (Lombards), the end of the 7th century (Alamans, Baiuwarians) and even to the end of the 8th century among the Avars. Therefore, we are faced with the strange situation of having to reconstruct important elements of Byzantine culture with the help of find complexes from adjacent regions on the one hand by means of original products, which reached neighbouring regions in various ways, on the other hand with the help of imitations, which reveal the models of Mediterranean origin, although they were produced locally. The close connections between Byzantine and Avar culture, although these contacts usually moved only in one direction, can also be used to date the Byzantine originals, as well as certain production techniques and motifs. The chronological system of Avar culture is not equally reliable in all periods so it is useful, indeed even necessary, to reflect on this topic briefly. The Avars and Byzantium Due to their steppe-nomadic way of life only slightly modified to suit the entirely different environment in the Carpathian Basin the Avars were always regarded as a non-European exception. Like the Scythians and the Huns before them, they guarded their herds, hunted and fought on horseback. Permanent buildings, settlements and towns had no importance for them, while mobility was crucial. Only very recently have we learned to see the Avars with different eyes. Already more than a 100 years ago their material remains were classified as Late Roman, because they included so many antique shapes and motifs. As a result of international historical research in the past 30 years, we have managed to overcome the much too rigid, traditional concepts of tribes and peoples. This has also facilitated a new understanding of the foundation of the Avar Empire and of its diverse contacts with Byzantium. If it were not for Byzantium, the Avar Empire would never have been founded, because large annual payments from Constantinople and rich loot from raids on the Balkans stabilised the khagans power. Avar representational culture was only created in the Carpathian Basin and in many ways reflects its Byzantine counterpart, or whatever the Avars perceived that to be. The Avars did not adopt techniques, shapes and motifs uncritically, of course. Sometimes, Byzantine models were rejected and for a short period of time in the middle of the 7th century, it seems that a consciously anti-Byzantine orientation prevailed. This seems to

Byzantine Belt Ornaments of the 7th and 8th Centuries in Avar Contexts

Plate 1 Mask fittings. 1, 2 Tolnnemedi; 3 Leobersdorf; 4, 5 Hajduszoboszl. Scale 2:3

Plate 2 Belt set from Kiskrs, burial 9, selected objects. Scale 2:3

be true for the so-called pseudo-buckle belt sets.13 In any case, the material remains of Avar culture are extremely relevant for Byzantine archaeology. Over a period of about 200 years, we can observe in which ways Byzantine culture influenced a neighbouring territory. Remarks on the chronology of the Avar archaeological material (hard and soft facts) For many Byzantine objects in the larger museums and collections no reliable information regarding provenance and find circumstances is available. Therefore, objects found during proper excavations, particularly grave assemblages with their combinations of object types, which may be evaluated using statistics, are of considerable importance for the chronology of the various types of finds and for recording areas of distribution. Avar graves contain a large number of Byzantine object types and therefore Avar typochronology may be able to assist Byzantine studies in a number of ways. But how reliable is Avar chronology at present? For the Early Avar period (568 to 650/670) the situation is still rather disadvantageous. Although we do have a number of (absolute) dates derived from coins, these are only terminus post quem dates for the burials themselves. The time span between the moment when the (youngest) coin in the grave was minted and the time of burial may be considerable. The princely grave from Kungota, which is probably from the first quarter of the 7th century, but contains a solidus of Justinian I, minted after 542, is a good example. In this case, the coin is not really helpful. Modern interpretations of larger cemeteries from the Early Avar period, such as Zamrdi, using statistical methods and the internal chronology of the cemetery itself, will soon improve the state of research. The situation is much better for the Middle and Late Avar periods (about 650/670 to about 800). Due to the large number of Avar grave assemblages and the sometimes huge necropoleis consisting of several thousand inhumation graves, the chronological system for the Avar finds from the Middle and Late Avar periods is quite detailed. Absolute dates for characteristic types of objects can be assigned with the help of a number of 7th century coins and, ultimately with the detailed written sources on the fall of the Avar Empire caused by the armies of Charlemagne just before the year 800. The only question is: how long did it take for the production of decorated

belts within the Avar Empire to stop, once the khagans power had collapsed? The ideas of the scientific community regarding this question differ considerably. Byzantine diversity in the archaeological material of the Early Avar period The archaeological material of the Early Avar period (568 to about 650) is very heterogeneous. Iron pieces of equipment (such as stirrups) and the bow in particular demonstrate close links to Eastern steppe culture. At the same time, there is also a local, Late Roman cultural component as well as a strong Germanic element, presumably due to groups which participated in raids on the Byzantine Balkans. At the end of the 6th and in the early 7th century, brooches from the North Sea and Baltic region are found in the Carpathian Basin and a number of richly equipped womens graves demonstrate that it was possible for people of Germanic origin to gain considerable wealth within the Avar Empire.14 A few years ago, Garam compiled archaeological material from Avar contexts of the Early and Middle Avar periods, for which she assumed an origin from within the Byzantine Empire, or at least a close connection with Byzantine culture.15 Whether these are in fact Byzantine products, objects made by Byzantine craftsmen working for Avar customers, or local imitations, must be examined in each individual case. One should also consider that few types of Byzantine jewellery or equipment were distributed over the entire territory of the Byzantine Empire. Rather, it seems that local customs also manifest themselves increasingly in the use of dress accessories. Well known examples are the Sicilian-Byzantine buckles with animal motifs; these represent a unique, local type, and only the shape of the fittings remains the same on a supra-regional level (see Entwistle, this volume, Pls 2224, 27). Among the oldest Byzantine belt-fittings are the mask fittings, of which only very few examples were found on Avar territory. The more valuable examples of this type are silver castings (Pl. 1). Mask fittings seem to have been most common in the final quarter of the 6th century. Belts elaborately decorated with mask fittings have been found in the northern Caucasus and Sasanian Persia16 and later variations of this type were also found in Lombard Italy. When the phase of large cemeteries in the Avar Empire began, the era of mask fittings appears to have already been over.
Intelligible Beauty | 63

Daim Sheet metal strap-ends with highly stylised linear decoration and scroll ornament are loosely related to the mask fittings. A belt set consisting of sheet metal strap-ends of this type and of cast belt fittings, was found in Grave 9 in Kiskrs (Pl. 2).17 As the combination of sheet metal strap-ends and cast fittings of different shape occurs several times in Lombard Italy, for example in Arcisa Grave 2 as well as in Nocera Umbra Graves 16, 18, 42, 79, 84 and 85,18 it is likely that the belt ornaments from Kiskrs came to the Carpathian Basin from Italy. Such sheet metal strap-ends with linear and scroll ornament are frequently referred to as Martynovka type, after the treasure found in 1907 in present-day Ukraine.19 In the late 6th and early 7th century, they seem to have been distributed widely, from Italy to the Black Sea. The dot and comma ornament, which was later so popular in Byzantium, may have developed from the linear and scroll ornament of the sheet metal strap-ends. According to Garam20 amongst the oldest types of Byzantine belt ornaments is the Fnlak type, which was frequently decorated with dot and comma ornament and of which several complete sets were found in Avar contexts (Pl. 3). One of the most important Avar burials is the smiths grave from Kunszentmrton.21 The grave assemblage points toward Byzantium in several respects, particularly with regard to a precision scale with several weights of different sizes, but also due to numerous formers (positive models) for sheet-metal fittings (P. 4), including a complete set for the production of a belt set of high artistic quality, which bears a certain resemblance to the belt fittings from Kungota (Pl. 5). Characteristically, the central medallion of the former for the main strap end contains an anchor-cross. The gold belt sets of Bcsa-Kunbbony type with the so-called pseudo-buckle fittings were produced using Mediterranean techniques, from many individual components. Their models, however, are found in the steppes of the northern Caucasus and in the woodlands west of the Ural. Therefore, they cannot be considered Byzantine belt sets.22 The gold belt set with pseudo-buckles from Sirmium could be a Byzantine adaptation of this type (see Blint, this volume, Pl. 27). As it consists of heavy, cast components, finely worked only from the front, this could be a Byzantine product made for a potentate in the Carpathian Basin. The only gold belt set with pseudo-buckles outside the Carpathian Basin was found in the princely grave from Mala Pereepino (Ukraine), which may be dated to the middle of the 7th century due to the gold coins which are part of the jewellery. It is likely that those Avar graves with gold pseudo-buckles belong to the same period. In the next section I would like to present two archaeological assemblages with Byzantine belts from the 7th century and several belt sets and individual finds from the 8th century. Kungota (buried after 542, probably in the first half of the 7th century) The princely grave from Kungota, found in 1857, was the first rich Avar burial known to archaeologists.23 The deceased was buried with two horses with decorated bridles. His clothing included various kinds of jewellery, such as eight finger-rings and a belt with gold ornaments (Pl. 5). The sword was covered in gold foil decorated with Dionysian scenes, which may originally have belonged to some kind of casket (see Blint, this volume, Pl. 24). The grave contained a solidus of Justinian I, from 54262. For a long time, it was assumed that the components of the belt set had been pressed using dies. However, technical studies carried out by Bhler in 2001 have shown that each component was produced individually, by chasing and repouss.24 The set consists of a main strap end, four large and four small fittings with long attachment loops, four small strap-ends, a double bow-shaped fitting with attachment loops, a strap loop with a ring attached and a husk-shaped ornamental piece of sheet gold with a semi-circular concavity on one side. The fine ornament on the sheet gold is symmetrical and depicts stylised scrolls (according to Garam) or birds. The central sections are framed by rows of beads and by dot and comma ornament. The medallion in the centre of the main strap end is not decorated with a monogram, but instead with a fine whorl consisting of five rays. The strap loop

Plate 3 Belt set from Keszthely Fenkpuszta, Horreum, burial 15, selected objects. Scale 2:3

Plate 4 Bronze formers (positive models) from the goldsmiths grave from Kunszentmrton. Scale 2:3

64 | Intelligible Beauty

Plate 5 Belt set from Kungota, reconstruction of the mounted belt ornaments. Scale 2:3

Daim

Plate 6 Belt set from Ozora-Ttipuszta, selected objects. Scale 2:3

is decorated with a striking ankh-like cross, which is in fact the only definite Christian cross on Byzantine belt sets from Avar contexts, if we do not take into consideration the anchor-cross from Kunszentmrton (see above) and the medallion from Ozora-Ttipuszta (Pl. 6), which bears some resemblance to a cross. Ozora-Ttipuszta (buried after 669) The graves from Ozora-Ttipuszta are among the most important Avar burials with respect to Byzantine archaeology. They were discovered by accident in 1871. Unfortunately, the find circumstances are not completely clear, but recently Prohszka has identified archive material which shows this archaeological assemblage in a different light.25 As a result, it is now possible to reconstruct how the two graves were discovered in some detail, although many questions still remain. The first grave was discovered accidentally in the course of earthworks. The second was found two or three days later during an investigation in the vicinity of the first grave. In the course of this investigation a skeleton lying face-down was found together with sword fragments, a torc, buckles, a fingerring, small earrings and a chalice-shaped silver vessel. Several pendants were found at the temples of the deceased, the torc with a sheet gold pendant (bulla) was located at the neck. A gold coin of Constantine IV (minted 66974) had been placed on the back of the deceased. Much later, a sword and a silver drinking horn, as well as a glass jug with a copper handle and a silver lid, were also found beside the skeleton. Reputedly, the glass body of the jug was covered with silver on the outside and with gold on the inside. Unfortunately, this object no longer exists. At dawn on the next day, a second bracelet, two small earrings, pieces of sheet gold (presumably the belt set), the attachment plates of the sword, a silver buckle and a piece of sheet gold from the sword handle were found. According to Prohszka, grave 2 could in fact have been a double burial not completely excavated in 1871. In this case the male skeleton would still be in the ground. It is virtually impossible to clarify whether this assumption is true or not. Alternatively, grave 2 could just have been a mans grave which, uncharacteristically,
66 | Intelligible Beauty

also included a torc and bracelets, as well as a necklace. At any rate, both graves contained a number of Byzantine objects, so that if it had been customary for Byzantine citizens to be buried with grave goods one could have mistaken them for burials of Byzantine citizens. The burial customs are Avar and probably the sword, but even this is still open to discussion. Although the belt set is made of sheet gold (Pl. 6), it is not as elaborate from a technological point of view as the one from Kungota. It was produced using formers (positive models), which is why the relief does not have such precise contours on the front. In addition there appears to have been no re-working of details on the decorated pieces of sheet gold before adding the other components of each belt fitting.26 The set consists of a main strap end, one large and three small fittings and six small strap-ends, plus a fitting of double-bow shape and a gilded silver buckle. The individual components are somewhat larger than those from Kungota and the decoration is coarser, although not unattractive. The strap-ends and fittings repeat the motif shown on the main strap end a small tree with large leaves which are attached to the trunk at an oblique angle. In the centre of the main strap end, there is a medallion with a somewhat cross-shaped ornament, which also seems to imitate decoration with precious stones. In this case, the Christian symbol of prosperity and victory (if in fact it is intended as such) is so ambiguous that it could also be interpreted as a geometric ornament. The other objects from the two graves or the double grave are also relevant for Byzantine archaeology, but this is not the place to discuss them. Because the time at which the finds from Ozora-Ttipuszta were buried can be determined with some precision (after 669), one may be permitted to draw some historical conclusions. In fact this case is an excellent example of how archaeological finds may assist the interpretation of isolated historical sources. For, in the absence of the archaeological record, how could we interpret the completely isolated story of an Avar delegation being sent to Constantinople in 678/9 in order to congratulate the emperor on the victory of the Byzantine fleet against the Arabs in the vicinity of the capital? The historian would have many ways to interpret this event. However, only the

Byzantine Belt Ornaments of the 7th and 8th Centuries in Avar Contexts

Plate 7 Belt set from Hohenberg. Scale 2:3

archaeological sources demonstrate that this diplomatic activity occurred in a period during which Byzantine coins and luxury goods entered Avar territory, and the Avar elite almost exclusively followed Byzantine fashions. Because this delegation fits well into a clear archaeological picture, it can be interpreted more precisely from the historians point of view. The process during which the Avar Empire had gradually re-gained its strength had now been more or less completed and, immediately before the foundation of the Bulgarian Empire in 680, Byzantium was looking for allies on its northern border. Belt fittings from Hohenberg to Szeged (8th century) The main classes of multi-part belt sets from the late 6th and the 7th century, whose origins lay within Byzantine culture, were identified and described some time ago. This was due in part to self-evident find circumstances, for instance the treasures from Akalan, Mersin and the finds from the Byzantine stronghold at Sadovec (Bulgaria),27 but also to some of the motifs used, such as monograms and Christian symbols of salvation and victory. For the 8th century, the situation is not quite as clear. Although it seems likely that decorated belts were also used in Byzantium in the 8th century, no actual examples of these were known until recently. The key to solving this problem lay within the archaeological material of the Avar Empire, where it was still customary to bury the dead

with clothing, jewellery and grave goods even in the 8th century. In fact we have more than 50,000 Avar grave assemblages from the 8th century and approximately one in ten mens graves contained a cast belt set, most of which were decorated with obviously Mediterranean motifs: circus scenes, griffins, marine spirits riding on dolphins, imperial portraits, vine scrolls and many more. 28 The enormous quantity of material, but also the many failed castings and half-finished products from Avar settlements demonstrate that there was an active production of belts within the Avar Empire. But where are the models, and how did an Avar bronze caster become acquainted with the many different motifs? Recently, a splendid belt set from Hohenberg in the Austrian Alps has been identified as an Italian-Byzantine product (Pl. 7).29 Apart from the decorated belt, the grave also contained a Carolingian sword of excellent quality. The belts method of production differs from that of the vast majority of Avar parallels. It is made of brass instead of bronze and the main strap end as well as the small strap-ends are complex constructions, consisting of up to 70 individual components (Pl. 8). The historical interpretation of this belt set was assisted by two good parallels which were found recently in Bolzano (South Tyrol, Italy), in an already plundered grave in the church of S. Vigilius, and most notably, by the fresco in the north-western side-chapel of the church Sta Maria Antiqua in Rome (middle of the 8th century), which presumably depicts
Intelligible Beauty | 67

Daim suspect that this is valid also for the 8th and 9th centuries. On the whole, decorated belts appear to have played an important role, both in the representation of high-ranking civilian officials and military officers, as well as in diplomatic relations, and certainly also in trade. It therefore seems reasonable to assume that the high-quality belt set from Hohenberg was brought from Italy to a local potentate by a Byzantine delegation. The example of a small gold strap end now in the Dumbarton Oaks collection, said to have been found at Aleppo, illustrates how the Avars treated Byzantine models (Pl. 10 left).31 Several years ago, the Institute for Pre- and Protohistory at the University of Vienna acquired a small private collection of Early Medieval objects which had previously been bought from antique and even flea markets. The collection included a Late Avar belt set with a main strap end which is strikingly similar, in terms of its construction, to the object from Aleppo (Pl. 10 right). However, the Byzantine bird-motifs have been replaced by quadrupeds. The same observation was also made when studying the decoration of other suspected Avar imitations. The fragment of a beautiful strap end from Mikulice (southern Moravia, Czech Republic) may well be another Byzantine original, because, apart from the birds, which peck at grapes on the vine scrolls, it includes several decorative elements which do not occur on Avar finds (Pl. 11)32 such as half-palmettes and punched dot and comma ornament. Provided this assumption is correct, the fragment proves that high quality strap-ends consisting of two parts which were put together back-to-back were also produced in Byzantium. One of the most expressive Byzantine belt fittings was discovered only very recently, in 2004, during excavations at Szeged-Kiskundorozsma, Kettshatr II, in an Avar grave which had already been robbed in antiquity. The fitting was published by the excavators very soon after its discovery.33 In 20072008, the object was examined in detail at the RGZM in

Plate 8 The construction of the main strap end from Hohenberg

the stepson of the donator Theodotus, wearing a short kaftan and a multi-part belt set of this or similar type (Pl. 9). As noted above, as Theodotus was supreme commander of the imperial troops in Rome, the boy is perhaps shown wearing the official dress of the dux (magister militum).30 The chapel, and with it also the fresco, was built between 741 and 752, which fits in quite well with the date established by means of archaeological criteria for the belt sets of Hohenberg type: the third quarter of the 8th century. Written sources from the early 10th century refer repeatedly to precious belts as diplomatic gifts. There are good reasons to

Plate 9 Wall painting from Sta Maria Antiqua, Rome

68 | Intelligible Beauty

Byzantine Belt Ornaments of the 7th and 8th Centuries in Avar Contexts

Plate 10 Strap end from Aleppo (left) and Avar strap end (right). Scale 1:1

Plate 11 Fragment of a strap end from Mikulice. Scale 1:1

Plate 12 Belt fitting from Szeged-Kiskundorozsma. Scale 1:1

Mainz (Pls 1213). The belt fitting is a little more than 40mm high and weighs about 17g. It consists of numerous silver components, which were soldered together, resulting in a stable construction which was then mercury gilded on the front. The side strip consists of two beaded wires, separated by triangles consisting of nine granules each. The outer edge of the side-strip is decorated all around with a beaded border, worked in repouss, using a frame-shaped piece of sheet silver. Each individual bead was also decorated with triangles and dots, using punches. In the central decorative section of the fitting, the Byzantine emperor is depicted in profile wearing a diadem, as well as a round brooch on his left (!) shoulder and some kind of breast ornament. He is holding a bundle of laurel in front of his body, with another twig or branch placed behind his head. The figural ornament is surrounded by an arcaded frame. The

Plate 13 Belt fitting from Szeged-Kiskundorozsma. Scale 1:1

contours of the relief ornament on the central sheet are just as precise on the front as on the reverse. This can best be seen on the beads of the diadem. As the contours of the beads on the front and the reverse correspond so remarkably well, it seems likely that a two-part bronze die was used that is a (negative) die for shaping the front and a (positive) former for shaping the reverse. After the sheets had been pressed in the two-part die, the recesses of the openwork were cut out and some details re-worked using various types of punches. The centrepiece with figural decoration was then soldered onto the sheet silver frame from the front. Using such a two-part die would have facilitated the production of several pieces of sheet silver with identical decoration. This in fact would have been necessary in our case, as belt sets of this type usually included six identical fittings. Three attachment loops, consisting of narrow strips of sheet silver, were soldered onto the reverse of the belt fitting as a means of attachment to the belt. The next step was to mercury-gild the fitting. We do not know what its hinged pendant looked like, presumably it was semi-circular or pentagonal, like the fitting itself. The belt fitting from Szeged-Kiskundorozsma fits in perfectly with the group of Byzantine belt sets, strap-ends and fittings from the 8th century known at present. In our case, there is more than one Avar adaptation of the type Emperor with a bundle of laurel. A good example is the mercury-gilded buckle from the Avar cemetery of Leobersdorf (grave 69), which may be placed in the middle of the 8th century.34 The Byzantine emperor, as depicted on this 8th-century fitting, does not conform with his image on contemporary
Intelligible Beauty | 69

Daim
(repr. Stuttgart, 1972), VI, 2, 19; M. and M. Whitby, The History of Theophylact Simocatta. An English Translation with Introduction and Notes, Oxford, 1986, 159ff; P. Schreiner, Theophylaktos Simokates. Geschichte, Stuttgart, 1985, 163ff. Recently mentioned in: S. Efthymiades, A Day and Ten Months in the Life of a Lonely Bachelor: The Other Byzantium in Miracula S. Artemii 18 and 22, DOP 58 (2004), 126. I would like to thank Cyril Mango for drawing my attention to this source. V.S. Crisafulli and J.W. Nesbitt, The Miracles of St Artemios. A Collection of Miracle Stories by an Anonymous Author of SeventhCentury Byzantium, Leiden/New York/Cologne, 1997, mir. 18, 11420. A. Cameron, The Byzantines, Oxford, 2006, 117; A.J. Festugire (ed.), Vie de Thodore de Syken, Brussels, 1970, no. 5, cf. no. 12. J.F. Haldon, Constantine Porphyrogenitus. Three Treatises on Imperial Military Expeditions (Corpus Fontium Historiae Byzantinae XXVIII), Vienna, 1990, 11011. R.J.H. Jenkins and Gy. Moravcsik, Constantine Porphyrogenitus, De Administrando Imperio (Corpus fontium historiae Byzantinae I), Washington DC, 1985, ch. 6, 52. Cf. K. Belke and P. Soustal, Die Byzantiner und ihre Nachbarn, Die De Administrando Imperio genannte Lehrschrift des Kaisers Konstantinos Porphyrogennetos fr seinen Sohn Romanos, Vienna, 1995, 75. Cf. R.J.H. Jenkins (ed.), De Administratio Imperio, Vol. II. Commentary, London, 1962, esp. 1415. M. Schmauder, Vielteilige Grtelgarnituren des 6.7. Jahrhunderts, in F. Daim (ed.), Die Awaren am Rand der byzantinischen Welt. Studien zu Diplomatie, Handel und Technologietransfer im Frhmittelalter (The Avars on the Border of the Byzantine World. Diplomacy, Trade and the Transfer of Technology in the Early Middle Ages) (Monographien zur Frhgeschichte und Mittelalterarchologie 7), Innsbruck, 2000, 1544; Cs. Blint, Byzantinisches zur Herkunftsfrage des vielteiligen Grtels, in idem (ed.), Kontakte zwischen Iran, Byzanz und der Steppe im 6.7. Jahrhundert (Varia Archaeologia Hungarica), Budapest, 2000, 99162. A. Rettner, Zu einem vielteiligen Grtel des 8. Jahrhunderts in Santa Maria Antiqua (Rom), in Daim (n. 11), 26782; H. Belting, Eine Privatkapelle im frhmittelalterlichen Rom, DOP 41 (1987), 5569. E.H. Tth and A. Horvth, Kunbbony. Das Grab eines Awarenkhagans, Kecskemt, 1992. M. Menke, Zu den Fibeln der Awarenzeit aus Keszthely, A Wosinsky Mr Mzeum vknyve XV (1990), 187214. For two examples of rich womens graves with a Germanic background of the late 6th or early 7th century, see: A. Kiss, Das awarenzeitliche Grberfeld in Klked-Feketekapu B (Monumenta Avarorum Archaeologica 6), Budapest, 2001, Taf. 2931 and 349. . Garam, Grtelverzierungen byzantinischen Typs im Karpatenbecken des 6.7. Jahrhunderts, Acta Archaeologica Hungarica LI (1999/2000), 37991; eadem, Funde byzantinischer Herkunft in der Awarenzeit vom Ende des 6. bis zum Ende des 7. Jahrhunderts (Monumenta Avarorum Archaeologica 5), Budapest, 2001. Cs. Blint, Kontakte zwischen Iran, Byzanz und der Steppe. Das Grab von Tepe (Sowj. Azerbajdan) und der beschlagverzierte Grtel im 6. und 7. Jahrhundert, in F. Daim (ed.), Awarenforschungen I (Archaeologia Austriaca - Monographien 1), Vienna, 1992, 309496, in particular Taf. 2956. T. Horvth, Die avarischen Grberfelder von ll und Kiskrs (Archaeologia Hungarica XIX), Budapest, 1935, Taf. XXIV. O. von Hessen, Primo contributo alla archeologia longobarda in Toscana: Le necropoli (Accademia Toscana di Scienze e Lettere La Colombaria, Studi XVIII), Firenze, 1972, Tav. 3 and 4; C. Rupp, Das langobardische Grberfeld von Nocera Umbra: 1. Katalog und Tafeln (Ricerche di Archeologia Altomedievale e Medievale 31), Firenze, 2005, Taf. 26, 32, 62, 96, 99101. L.V. Pekarskaja and D. Kidd, Der Silberschatz von Martynovka (Ukraine) aus dem 6. und 7. Jahrhundert (Monographien zur Frhgeschichte und Mittelalterarchologie 1), Innsbruck, 1994, Taf. 313. Garam 2001 (n. 15), Taf. 814. D. Csallny, A kunszentmrtoni avarkori tvssr (Goldschmiedegrab aus der Awarenzeit von Kunszentmrton), Szentes, 1933; F. Daim, Avars and Avar Archaeology. An Introduction, in W. Goetz, J. Jarnut and W. Pohl (eds), Regna and Gentes. The Relationship between Late Antique and Early Medieval

8 9 10

Plate 14 Coat clasp from Dunapataj. Scale 2:3

coins, but rather with his image on 4th-century solidi. The same is true for the bundle of laurel as part of the triumphal ceremony. The only comparable 8th-century example is to be found on the finely worked lid of a small silver box now in the Museo Arqueolgico Nactional in Madrid.35 It seems that the motif has not been understood by the Avars, who copied Byzantine fittings in many cases. Conclusion As the archaeological record shows, a large amount of Byzantine goods reached Avar territory in the late 6th and 7th centuries. Different items of male and female jewellery, vessels of gold, silver, bronze and glass, but also food and drink, as the amphorae show. In the 8th century, however, we find only belt ornaments in graves. It is likely that they served as diplomatic gifts. One exception is the high-quality, gold clasp (= coatfastener) from Dunapataj.36 It was made from sheet gold by chasing and repouss and depicts heads en face and in profile (Pl. 14). A cross, held by the figure on the left, is depicted between the heads in profile. Presumably, this clasp was brought to the Carpathian Basin in the second half of the 8th century, together with a valuable coat. Further research will show whether this is in fact, apart from the small number of belt ornaments identified so far, the only Byzantine object from the 8th century in the Carpathian Basin. Notes
1

11

12

13 14

15

16

A warm thank you to all the colleagues who gave me advice and help while preparing this article and especially to Birgit Bhler for the translation. 2 Most recent research on this topic: C. Schopphoff, Der Grtel. Funktion und Symbolik eines Kleidungsstcks in Antike und Mittelalter (Pictura und Poesis. Interdisziplinre Studien zum Verhltnis von Literatur und Kunst 27), Cologne, 2009, which, however, excludes Byzantium almost completely. 3 C. Wickham, Framing the Early Middle Ages. Europe and the Mediterranean, 400800, Oxford, 2006, 175. 4 J. Werner, Nomadische Grtel bei Persern, Byzantinern und Langobarden, in Atti del convegno internazionale sul tema: La civilt dei Longobardi in Europa (Roma, 2426 maggio, 1971) (Cividale del Friuli, 2728 maggio 1971), (Accademia Nazionale dei Lincei 371), Rome, 1974, 10956. 5 A. Cameron (ed.), Flavius Cresconius Corippus. In laudem Iustini Augusti minoris libri IV, London, 1976, II. 115; W. Pohl, Die Awaren. Ein Steppenvolk in Mitteleuropa 576822 n. Chr., Munich, 1988, 489. 6 C. de Boor (ed.), Theophylacti Simocattae historiae, Leipzig, 1887

17 18

19

20 21

70 | Intelligible Beauty

Byzantine Belt Ornaments of the 7th and 8th Centuries in Avar Contexts
Peoples and Kingdoms in the Transformation of the Roman World (Transformation of the Roman World 13), Leiden/Boston, 2003, 4789, pl. 6; B. Tobias, Frhmittelalterliche Grber mit Schmiedewerkzeugen (RGZM Monographien). In preparation. I. O. Gavrituchin, voljucija vostokq noevropejskich pcevdoprjaek (Kultury Evrazijckich Stepej Btoroj Poloviny i Tycjaeletija n. . (iz Istorii Kostjuma) 2, Samara, 2001, 3186. F. Daim and Z. Rcz, Kungota, Reallexikon der Germanischen Altertumskunde 17 (2000), 48690; . Garam, Katalog der awarenzeitlichen Goldgegenstnde und der Fundstcke aus den Frstengrbern im Ungarischen Nationamuseum (Catalogi Musei Nationalis Hungarici. Seria Archeologica I), Budapest, 1993, no. 106; eadem 2001 (n. 15), Taf. 87. B. Bhler, Der Nachweis der Treibziseliertechnik an goldenem Grtelschmuck der Frh-, Mittel- und Sptawarenzeit (Medium Aevum Quotidianum 45), Krems, 2002, 1512. P. Prohszka, Ozora-Ttipuszta, in F. Daim and J. Drauschke (eds), Das Rmereich im Mittelalter. Studien zum Leben in Byzanz / The Roman Empire in the Middle Ages. Studies on Life in Byzantium (RGZM Monographien 2010). In preparation. Technical study (optical microscope) by Birgit Bhler in 2001. Werner (n. 4), 10956. J. Dekan, Herkunft und Ethnizitt der gegossenen Bronzeindustrie des VIII. Jahrhunderts, Slovensk Archeolgia 202 (1972), 317452. F. Daim, Byzantinische Grtelgarnituren des 8. Jahrhunderts (Byzantine Belt Sets of the 8th Century), in idem (n. 11), 77204, esp. 136ff. 30 31 32 33 See also n. 12. Daim (n. 29), 110ff. Ibid., 122ff. P. Mszros, T. Paluch and C. Szalontai, Avar kori temetk Kiskundorozsma hatrban. Elzetes beszmol az M5 autplyn feltrt lelhelyekrl (Mzeumi Kutatsok Csongrd Megyben 2004), Szeged, 2005, 14462; idem, Avarkori temetk Kiskundorozsma hatrban (Elzetes beszmol az M5-s autplyn feltrt lelhelyekrl), Tatabnyai Mzeum Tudomnyos Fzetek 8 (2006), 97108; F. Daim, Kaiser mit Palmzweigen. Ein byzantinischer Grtelbeschlag aus Sdungarn, JbRGZM 54 (2007), III, 779; F Daim et al., Kaiser, Vgel. Rankenwerk. Byzantinischer Grteldekor des 8. Jahrhunderts und ein Neufund aus Sdungarn, in Daim and Drauschke (n. 25), in preparation. 34 F. Daim, Das awarische Grberfeld von Leobersdorf, Niedersterreich (Denkschriften der sterreichischen Akademie der Wissenschaften, phil.-hist. Klasse 194), Vienna, 1987, Taf. 63; idem (n. 33), 779, particularly 78. 35 Ibid., 78. 36 . Garam, ber das awarenzeitliche goldene Agraffenpaar von Dunapataj, Folia Archaeologia XL (1989), 13753; F. Daim and B. Bhler, Awaren oder Byzanz? Interpretationsprobleme am Beispiel der goldenen Mantelschliee von Dunapataj , in T. Vida (ed.), Thesaurus Avarorum. Rgszeti tanulmnyok Garam va tiszteletre (Archaeological Studies in Honour of Eva Garam), Budapest, 2009 (in print).

22 23

24 25

26 27 28 29

Intelligible Beauty | 71