Back

A comparative analysis of one well preserved composite bow from Georgia

                                               Zoran Pavlović

Introduction

Having examined the photographs of a suprisingly well-preserved bow from Mestia in the province of Svaneti (a region known, among other things, for exquisite archers in history), it seemed that the bow presented in the photographs was a composite bow of common construction specific for a number of presentations recorded in Eastern Christian iconography up to the late 14^th century. Further examination led to complicated riddles that asked for a thorough analysis and comparative methods in order to put the design of the bow into a definite time frame. This paper aims to use all available sources that are indicative of designs analogous to the Svaneti bow. The other bow of much more complex construction consisting of six arms, will not be taken into consideration in this paper .

Although already aware (through written and art works) of a long and rich archery tradition of Caucasus, and including Georgia, from the ancient times to the period when the development of the bow and its use reached its zenith notwithstanding the appearance of firearms, the Svaneti bow is the first item to shed clear light onto the construction, craftsmanship of the artisans, as well as onto the different influences and sharing of experiences via military and trading connections, all due to the location and conditions of this region during the ancient times, the Middle Ages, and even the Modern Age Georgia.

Searching the analogy in written and art sources

The written sources are scarce and offer no detailed descriptions of bows and their construction in Georgia. Furthermore, there are no military manuals offering at least some indications as to the typology of bows, tactics or directions for usage. Neither do the Byzantine military manuals (composed by experienced officers, or found in transcripts) cover this subject, therefore not giving any insight into the construction, designs or typology of bows. Using the existing manuscripts which state the dimensions of 15 or 16 palms or some very scant information about the construction of bows as it is for example 10^th century Sylloge Tacticorum,  Kolias (1988, pp 214-238),  Haldon (1999, pp129-132) , Amatuccio (1996, p.114) and Валери Йотов (2004, pp 17-21 ) all agree on the lengths of bows in the Middle Byzantine period to be in the range between 114 and 125 cm. There are some problems with using these measurements. The palm was sometimes used as a canonical measure, having a fixed value, and at other times it was meant to be related to the individual. This can be seen in Arabic and Persian archery manuals where sometimes they refer to the fist or palm “of the archer” meaning the individual rather than the canonical measurement. The other problem would be, are they talking about the length of the bow around the curves or the length of the strung bow. This is used for measurement in Mongolia where the strung bow was used to measure out archery ranges.

Art sources such as miniature paintings, either from the period of Georgian medieval state or other ones,  offer only schematic presentations which are not a good starting point for lucid conclusions. Religious presentations in wall paintings (frescoes) of numerous churches and monasteries all around territory covered by medieval Georgian state,  abound with secular elements from everyday life, thus with presentations of bows, much earlier that their appearance as a non-canonical element in presentations of holy warriors in Eastern Christian iconography. Nevertheless, in this field as well, bows remain imprecisely presented in spite of the fact that, up to the middle of the 15^th century, they were painted with great details. One would expect more reliability from presentations dating from the period of so called “Paleologian renaissance” (the end of the 13^th century up to the first decades of the 15^th century) in the territories covered by the Byzantine, Serbian and Bulgarian states which include modern states of Greece, Serbia, Bulgaria and Republic of Macedonia. However, this is not the case. There objects were painted in two dimensions, but often in proportion to other objects, thus allowing the spectator to speculate on their inner anatomy, even when objects presented were functional bows, parts of which were painted in an extremely realistic way, but from different perspectives, not to mention compositions where not much attention was paid to proportion or the accuracy of presented elements. Quoting numerous available sources Grotowski concludes that, in Eastern Christian iconography, the bow was a personification of evil and a symbol of infidels, stating the negative view Christians had as to presenting the bow as an inappropriate element associated with secular topics and gods such as Artemis and Eros. According to Grotowski, the bow therefore was not included in presentations of warrior saints until the middle of the 13^th century when it is introduced in scenes with St. Sergius and Bachus riding alongside each other in the monastery of St. Catherine, scenes widely influenced by the Crusader Art. (Grotowsky 2010, pp 371-373). Although, another fact that the bow as a secular element in presentations of Warrior saints in Eastern Christian iconography appears a bit earlier should be added.¹ This is almost a parallel to the Ancient Greek minimization of the value of the bow under a different rationale.

As to the painted manuscripts such as Ionanes Skilitzes’ manuscript and many other Middle Byzantine manuscripts where bows appear in many scenes, they are too schematic to allow any assumptions or conclusions to be made unless there is a direct confirmation in an existing contemporary artefact. Some of 360 folios  from 12^th century Jruchi  Four Gospels II (fig. 1 and fig.2)  abound with presentation of bows which naively painted as they were and with a lot of elements that are doubtful, still partially speak in favour of the Svaneti design. A sufficiently authentic depiction of the ears, as well as the large surface area of bow limbs, are arguments that speak in favour of the bow from Mestia, features that incredibly support the preserved artefact before us. Having in mind that the objects presented date from that period, despite the fact that it is a religious presentation and that it is possible that there are exaggerations as to the proportions, it is clear that the bows are long and used by foot archers in that scene. It is common practice to associate horn bows with horseback archers. Nevertheless, there are many confirmations that composite bows were used by foot archers as well. Byzantine military manuals such as Tactica by Leo VI or  Sylloge Tacticorum from the late 10^th century state that foot archers’ bows were longer, and, according to Tactica, with a longer shooting range (Hurley 2011, p.131; McGeer 1995, p.213). Technically, longer bows are associated with longer draws and more powerful shots. Whether this developed from different traditions for foot archers and horse archers or it was a conscious decision to increase the effectiveness of infantry archers is difficult to ascertain. The bows are quite schematic in many details, making it impossible to assume whether the absence of a deep reflex on the handle is the result of the indifference of the painter of the miniature, or evidence that there was no deep reflex. Thus it makes it an argument against the contemoraneity of the design in question. Presentation from the monasteries of Ikvi and Pavnisi particularly offer no support to our concept. Referring to these frescoes covered by Привалова (1977, pp 111-119) (fig.3)  , Grotowski again tackles the problem of the bow as a symbol, rather than examining its design. (Grotowsky 2010, p372-373). ²

One piece of evidence from art supporting the hypothesis of the time frames of our design is the scene where Esau is coming back from the hunt (fig.4a and fig.5). ³ The older version dates back to the 8^th century. The origin of this version of Pentateuchus  does not match the region of Caucasus, having in mind that it originates from the West. There are also chronological gaps if we are going to place Svaneti bow in later period.  However, the bow that Esau is holding , even though it must be admitted that it was obviously disproportionate to his figure, overwhelmingly resembles the Svaneti design, at least considering its profile.³ In the same manuscript there is detail of Exodus with  seven similarly well depicted bows, having aggressive reflex in the limbs and low brace height (fig.6). A later Byzantine version in Greek dating from the late 11^th  century ⁴,  and there are grounds to claim this shares a lot of characteristics to the type of bow from Mestia (fig.7). Naturally we are talking about the dimensions of the bow Esau is carrying, which is either braced, hooked on his back with the string over his chest, or quite unusually in the obviously wrongly  placed bow case which could not be featured due to the position of his figure. Even though the bow string is not conspicuous, the bow is proportionate to the presented figures and other objects, which is rarely found in miniatures. The deep reflex on the grip and the ears that are slightly curved, admittedly with bulbous tips typical for Avar or Turkish-Khazar lateral antler or bone  plates found in numerous tombs, makes it necessary to speak with reserve about the type of artefact in question. All of aforementioned gives the impression of an aggressive grip reflex which is shown by a setback handle of the bow. Unfortunately, many features are not expressed with the contrast of colours, since the black-and-white microfilm allows only the conclusion that the dark colour was used for the outline of the bow, and therefore it is impossible to make further deductions. However, it does speak in favour of the origin of the bow to be from Mestia, during the Georgian Golden Age (11^th – 13^th centuries). ⁵

Should we make a comparison of the Georgian bow to the presentations of  bows in Eastern-Christian iconography, one of these presentations speaks in favour of our postulation. Namely, it concerns the aggressive reflex of the bow limbs. Serbian mural and other contemporary Balkans painting from the 14th century abounds with presentations of composite bows. Up to the late 1370’s, the so-called Cuman-Mongol type is quite common, albeit there are numerous exceptions. An example of a bow similar to the Georgian bow is the bow of St. Demetrios in the Holy Apostles Church in Pećka Patrijaršija and the same depicted bow in the church of Nova Pavlica, both dating from the 3^rd quarter of the 14^th century (fig.8a, 8b and fig.9) The afore-mentioned bow is highly reflexed, in the manner of the Svaneti bow. It is clearly presented in its whole length, with the siyahs partly covered by the hands of Warrior Saint. A strong asymmetry of the bow limbs is obvious in both of these presentations, and yet this does not have to be the norm. The painter presented the grip area with a contrast of colours to express the bone plates as well as bone plates are exposed at the siyahs having slant notch. We have to bear in mind, however, the fact that in both examples this specific detail was either damaged (the paint that was used for the base) or, and this would be highly unlikely to expect considering the attention the painter paid to all the other details, carelessly painted in this part. In his detailed observation of this presentation, Pavlović (2013, p.51) noticed the material that was used to cover the bow (namely, rivertrout fish skin), as well as many other parts and their dimensions. Nevertheless, the same author assumed that the wooden bow limbs were scarf joined in the area of the grip, thus making the bone plates play the role of stiffeners in the construction, as in earlier nomadic bows. This assumption, however, is not based on firm arguments for some other way of junction, as the painter failed to express many smaller details. Should a method of radio carbon dating be applied to specify the time frame of the Svaneti bow, this would certainly shed a new light on the anatomy of  bows in the territories covered by the Serbian state and other states of the late Medieval Ages in the so-called ‘’Byzantine Commonwealth’’  of this period. When it comes to contemporary artistic representations it is worth mentioning hunting scene from  an old Russian manuscript of 14^th century, Chronicle of George Harmatolos, known as Chronicles of Tver. The form of siyah of the painted bow overwhelmingly mimics the design of bow from Mestia despite the fact that there is no deep reflex in the bow handle. (fig.10) Worth to mention 12^th century stone relief from Dagestan (fig.11) and 13^th century presentation of archer on the doors of Ravello’s Cathedral (fig.12)

On the structure of bow from Mestia

A surprising characteristic of a Svaneti bow, however, is its structural, technical and mechanical mosaic of features that varies from the early Middle Ages, through Late Middle Ages up to the more advanced features that are seen at the zenith of archery of more modern history. It is not rare to see bow limbs put under an extremely large angle, thus giving it a deep reflex on the grip, and such bows appear in the Early Middle Ages. Bows with highly curved angles are seen on coins of rulers from the dynasty of  the Guptas in India. Murray B. Emeneau (Emeneau 1953, p.86) associates these bows with old Sasanid bows, noting their “striking design” and that “nearly all the bows display such angularities of the ears that it must be judged that they are of Sasanid type”. Paterson who has earlier discussed Sasanid bows suggested ''that they may represent interesting stage in development of the composite bows''.  Looking at them as possibly over-development he relies on their efficiency through history (Paterson, 1969, p.30).  (see fig. 13) There is a large number of circular plates from the period of the dynasty of Sasanids from the 4^th to the 7^th century that show rulers on horseback. One of them from 5^th AD shows King Peroz Firouz hunting scene.⁶ However, the feature that differentiates Sasanid bows from the Svaneti bow is the symmetry of limbs. Namely, Sasanid bows which are considered to be Hunnic type of bows with one shorter ear (siyah) do not possess that kind of symmetry. On the other hand, as for the asymmetry of Sasanid bows,  Bede Dwyer considers  this as a problematic assertion. He accentuates that all Sasanid bow reconstructions are based on depictions on silver plates. In the hunting scenes, the round form of the medium forces the upper bow limb to be distorted so that it fits into the shape of the plate. The lower limb in the earlier plates is behind the horse’s neck. There is at least one plate that shows a strung bow held by a standing attendant to the king . It is symmetrical unlike the popular ideas. In regards to the Huns, Dwyer mentions  that archaeologists are aware that often parts of two broken bows were put in a tomb. It can be surmised that there was either an economic reason or a ritual one for this practice. This does not mean that asymmetric bows were not used or that minor asymmetry was a normal part of the sign of some bows, but it suggests that assigning asymmetry to a bow type without surviving example could be unsound. (personal communication, November 3, 2016 ) This statement is supported by the scene of  Khosrau II at the Boar Hunt at Taq-e Bostan in Western Iran and also bows in scenes of contemporary early Byzantine two silver and one ivory plates (fig. 14, 15, 16). The last presentation of a Sasanid bow type dates from the Islamic period in the 8^th century. Furthermore, a careful observation makes it possible to conclude that the limbs of Sasanid bows in the presentations are shorter than the limbs of a Svaneti bow, the ears longer, and probably fitted with bone or antler plates (both, lower shorter and upper longer), and lacking a much more aggressive angle of siyahs that is characteristic of a Svaneti bow. Therefore the bowstring could not lie on the base of “ears” when the bow was braced. We can draw a conclusion that the aggressive reflex of the tips of “ears” (levers) of a Svaneti bow shares the same characteristic of bowstring when the bow is braced, but from different reasons. Bede Dwyer does think that cross sections of the ears are the precursors of the kasan in Turkish bows. It is possible that the development of recurved ears was a development of the Svaneti bow design or that it contributed to the mature form. (personal communication, November 3, 2016).  Another interesting detail is two bone plates glued from both sides of the grip which is oval in cross-section and becomes sharp towards the belly of the grip. In earlier nomadic bows this feature was used to strengthen the joint that formed the grip and attached bow limbs in order to prevent the disjunction of construction due to tension and stress. In other words these fittings led to improving of rigidity of the wooden core. (Werner, p47). There are numerous findings of bone and antler plates in the large territory from Korea, through Central Asia to the Pannonian Plain and even to Britain.⁷ However, there are few well-preserved bows that can be a good starting point for the judgment of their construction to be made. The most common constructions of a wooden core that can be found in earlier nomadic Hunnic types of bows are bow limbs joined in the area of the grip by a horizontal scarf joint with a riser or a vertical scarf joint. (Hall. 2005, 28-36). This made it necessary to strengthen this kind of construction. In such cases the cross-section differed completely from the design of the Svaneti bow. The bone strips may have reduced moisture from the palm being absorbed as well as the deflection of the arrow when shot.

In the case of the Svaneti bow, the way that the limbs are joined together shows a more advanced technology of  joinery which is characteristic of later Chahār Kham bows from the region of Sind in Pakistan (McEwen, 1979, pp 89-96), with the difference that the Chahār Kham bows do not have expressed ridges in their rigid part on the back of the limbs right after the grip, formed by engraving during the process of trimming of the wooden core before the sinewing the bow. As for Georgian design the splice on the grip piece needed to be cut slanting from the back to the belly on each end. It seems unlikely that the grip has a riser from its upper side which would result more or less in an expanded grip which is a characteristic of later Crimean Tatar, Persian, Korean or, in many cases, Sind bows. Most of aforementioned bow types  have limbs of a wooden core joined to a single wooden piece that forms the handle.  Photographs do not offer a clear insight into this junction, but we may conclude that there are two curved wooden strips under which the limbs of the wooden core are joined onto the handle in a V splice. The most likely scenario is that the bone plates played a decorative role as well as to reduce the hand shock due to the large stress from the bow, and in a lesser degree, they served to strengthen the construction of handle. It is most probable that, as opposed to the earlier Turkic-nomadic bows, they were attached after a layer of sinews. Furthermore, the oval cross-section of the grip is another argument in favour of the reduction of the archer’s paradox of the arrow during the release of the string. Technically the archer’s paradox refers to the counter intuitive fact that the arrow, pointed at different angles at the beginning and the end of the draw in relation to the grip, manages to go where the archer is aiming it. The arrow flexes around the bow because of various physical laws and this is what resolves the archer’s paradox. The oval cross section of the grip reduces the deflection of the arrow shaft allowing a stiffer and stronger arrow to be shot.

Instead of the common single horn plate, the belly of the bow consists several horn plates put in a parallel manner along the axis. At present, it is not possible to identify which animal was used for horns in order to make the horn strips. Contrary to the common impression that a single horn strip is used in the making of a composite bow, the facts are quite different. This question was discussed on a larger scale, both on the basis of written sources and the artefacts. The materials that were available often defined the forming of the horn strips and offered solutions. In their examination of the construction of Arab bows, Faris and Elmer (1945, p. 167) note that “The statement that five or six pieces should be contained in the horns of an extra heavy composite bow that is built for the one purpose of achieving the greatest length of cast – which means a competition, or flight, bow – is of both great importance and great obscurity, for the text does not tell us whether the pieces are to be laid in superimposed strata, or set end to end, or arranged in some other manner.” Mc Ewen quotes Sun Yung – Hsing who noted that in the 17^th century a type of long steer horn was not available to the northern barbarians, so they had to use four pieces of sheep horn spliced together to form a horn plate (McEwen 1978, p. 191) . The author of quoted encyclopaedia emphasizes that ‘’ In Kwangtung province, the horns of both water buffalo and yellow cattle are used by the bow maker’’ and ‘’the horn used here is made by joining the toothed ends of two pieces of ox horn together’’ (Sung Ying-Hsing, 1997, p.262) so the fact that the four sheep horn strips are not mentioned as being laid parallel is no surprise because the author is only mentioning the northern barbarians (probably sub-grup of Mongols) to show their inferiority in regards to Ming dynasty army .

Multiple horn strips being an old formula for forming the belly of the bow is proved by the examination of well-preserved earlier bows from the region of Tarim Basin, done by Andrew Hall (2006, pp 65-77) . He discusses that the belly of the bow is made up of numerous horn plates observed in Kotan bows which is discussed at the website ATARN (Asian Traditional Archery Research Network) with the help of photographs containing X-rays published by Stephen Selby, and also Eastern Han dynasty Niya 59MNM001 and 95MN1M4. It seems that a large number of later Modern Age Persian and Indo-Persian bows contained many narrow horn plates parallel to each other, so that Henry Balfour (1890) in his examination of a Persian dissected bow notes that “the belly is composed of a number of narrow strips of horn instead of a single piece.” On the other hand, according to Bede Dwyer,  of the major types of bow used in India, at least one appears to be of standard Persian design and it probably had multiple strips of horn on each limb while the two dissected bows of the crab bow design he knows have a narrow single strip on each limb (personal communication, November 3, 2016).  According to the information provided by Plakhotnichenko later “Buryat  bows consisted of multiple strips of cow horn put together into a mosaic of different colours”. (Plakhotnichenko, 1999).

The territory of Caucasus was very rich in caprides in the Middle ages. Capra caucasica / West Caucasian Tur, Bezoar Ibex / Capra aegagrus, Argali, are some of the endangered animal species nowadays. (Vereshchagin, 1967, pp 360-73). Horn plates offer some reason for one to assume that they come from any specific species of caprides present in this part of the Caucasus. Instead, it seems less probable that they come from a species of buffalo (Azeri or Caucasus buffalo) which was widespread in Georgia until 1940. This particular species of buffalo is characterized by short horns that point backwards and the yellowish color in some parts of exposed horn belly of the bow is quite problematic for making any kind of conclusions. The problem  is the design that is complex, with wide limbs and with the lack of adequate horns, so the craftsmen have found the solution in multiple strips. Worth to mention that Bede Dwyer sees advantages to multiple strips in economy and ease of manufacture, which are not easily discarded (personal communication, November 3, 2016). This element of making bows points to a warrior’s or huntsman’s bow, which opposes the original theory that the Svaneti bow was meant to be used solely by the aristocrats. On the other hand most Georgian aristocrats would have been both huntsmen and bowmen.

Another striking and advanced characteristic of this bow are the ears or the siyah of the bow. The first thing to be noticed is that the ears are curved and joined together in the manner of later bows with wider angle of the tips and with the triangular ridge which makes what modern terminology defines as siyah. In fact Dwyer states the siyah in Arabic only means the rigid ends of the bows. There is no implicit reference to the shape. The term was originally used for the rigid ends of bows like the Sasanid and central Asian Turkish bows. Many of the quotations in books like Arab Archery are not referring to contemporary bows, but date from the 8^th and 9^th centuries (personal communication, November 3, 2016). The triangular cross section of the larger part of ears is not a novelty unless we are dealing with a bow not earlier than the end of the 14^th century. A completely preserved fragment of the lever end with yellowish bone plates of the so-called bow from Belmesa, which was long believed to be part of the armament of the Roman army (Coulston 1985, pp. 233- 34), has a rounded triangular section up to the vertically parallel ends at the place where the bone plates meet. The belly edge has exposed wooden core that shows narrowing down to the very tips. Andrew Hall, comparing a lot of artefacts, put a new light on the origin of the fragment and opened another question making room for considering Belmesa bow fragment as a part of lower limb of Sasanid asymmetric bow (Hall, 2015, pp. 109-112).  Actually, Balfour had already identified the bow as probably Persian in his article on the Homeric Bow. Historically it may have been a relic of the conquest of Egypt by the armies of Khusrou II Parviz in the seventh century. Another later instance is early Mongol Chagan Khaad bow whose profile loses its width and goes over to the lever arms into a rounded triangular profile which, starting from the notch groove, gradually becomes a vertically oriented flat oval. (Becker, Rutschke, 2012, p. 87). Unlike later bows there is no strong transition from the triangular to the rectangular sections. There is a strong correlation in form to the Svaneti bow, but the internal structure is different.

Returning once again to the art presentations,  among a large number of impressive late-Byzantine frescoes, one of them stands out as important for this study. It is a frescoe depicting St. Dimitrius, painted by the most famous fresco painter of the late-Byzantine period Manuel Panselinos in the church Protaton in Karyes on Mount Athos (fig. 17). In this fresco, dating back to the beginning of the 14^th century, there is one extremely detailed part of a bow placed in the bow case, with such features which, carefully observed, lead to the conclusion that the  ears have a triangular shape in the cross section with a bone plate on the belly of the rigid part of the bow. Despite the fact that this artistic technique is limited as for the ability to express every small detail, based on the texture one can assume that the ears are joined together in a V splice, as in a Svaneti bow. There are notable wrapping which decorate the place where there are more than likely wraps of animal sinews, most likely due to the fragility of the bow as a result of stress. The two observed miniatures from the Jruchi Four Gospels do not allow further examination of such details even though their outline resembles the Svaneti bow. Another surprising feature is the triangular cross section that characterises later elements known in literature under the term kasan, which is taken from Ottoman terminology .⁸ The Svaneti bow boasts with an extremely aesthetically attractive carved ridge, which was rarely met with preserved Turco-Mongol bows until the 14^th century, and in artistic presentations kasan painted in small detail is not met before the Italian renaissance painter Antonio Pisano, also known as Pisanello. One of the six sketches that present Byzantine emperor John VIII Paleolog who attended the ecumenical councils in Ferrara and Florence, shows a bow, bow case and quiver. David Alexander deems that this was a bow popular in Syria, also known by the term fahlah, which was one of the most precious Mameluk objects of trade with Europe. (Alexander, 2004, p. 146) (fig.18). On the other hand, Bede Dwyer states that this statement is arguable. Bertrandon de la Brocquière in the 1430’s states that the export of bows was prohibited by the Mamluks and that he had to smuggle his bow out of Palestine (personal communication, November 3, 2016).

Horn tip overlays

Yet another feature of the Svaneti bow comes from a different time mosaic, and that is the additional frontal horn piece at the tips of “ears” where there is a string notch. The largest number of bone and antler plates served either as stiffeners / fittings for a more complex construction (probably for the laminated ears of the bow,) or bone plates in bows of Hunnic type which were introduced in order to prevent the twisting and bending of the ears considering the fact that the ears were extremely long and therefore the bows were extremely sensitive to any stress, as noted by Hall (2006). Archaeology does not remain quiet as to a small number of findings of frontal additional applications of bone or antler, that were used to make the endings of bow ears where there was a string notch. It is easy to assumed that they were added by scarf joint. First we shall look at Medvedev’s observations (Медведев, А.Ф. 1966, p.84) (fig.19), who, beside long lateral plates notes additional frontal bone applications on the edges of the siyah. Table 4 shows additional  frontal bone overlays  with string notches. Флёрова quoting Medvedev states that in the Eastern European steppes, this type of overlays was fixed on the bow remains dated to 12th to 14th centuries. (Флёрова, 2000, p 104).

The lower side which was added to the wooden part is visibly scored / grooved in order to get a better surface for gluing. Dwyer is prone to believe that these types of additional tips do strengthen the tip in a back to belly direction or they might have added the additional features of being easily replaceable and protection from wear on the bowstring (personal communication, November 3, 2016).

 Hudjakov (Худяков 1991, p.27-8) provides a typology and defines these bone applications with string notches into type 5, and in table 11 puts these bows into a time frame between th 9^th and 10^th century, in the territories of the Zabaikalye region (fig. 20). Флёрова (2000, fig. 1, 4-6, p 102; fig. 2, p103) (fig.21 and fig.22) frames these findings into the period from the end of the 10^th to the beginning of the 12^th century. She also states that triangular cross-sectional frontal edge overlays with groove for the string appear from the end of 10^th century . Based on the table assembled by Savin and Semyonov (Савин , Семёнов 1997, p.42, рис 3) (fig.23) it is obvious to note that the tendency of the ears reflex appears with the late Turkic-Khazar type of bow, and they stated that the additional bone applications  have allowed researchers to speak about the presence of specific  so-called Saltovo type of Khazar bows.  (Савин, Семёнов, 1998, p. 294).  . Еxcavations of Ossiniki burial site in the North Altai in a group of burials dated from the 12th to the 13th century include remains of four bows: central lateral bone plates and frontal bone rods for the composite bow. ( Savinov, 1981, p 155) (fig.24)

In his doctoral dissertation of Hungarian weapon from the region of the Carpathian Basin (dating from the 10th to the 11th century) Ádám Bíró (2013) devoted a lot of attention to the typology of bone and antler bow overlays, both lateral and frontal ones. He presented an analysis of Hungarian frontal bone and antler overlays trying to find a connection with similar findings from Inner Asia. These findings were the subject of a large number of Hungarian, Soviet and Russian literature works. Albeit abundant with similarities, it is widely considered that there are chronological and geographical gaps not only between the former Uzuntal type characteristic for the period from the 5th to 8^th century and rarely-found remains of the new type of late-Hungarian bone and antler applications but also between Saltovo and later bumper type that included frontal applications typical for the period between the 12th to the 14th century. Considering the frontal bone rod as a marginal phenomenon in Carpathian Basin, Bíró stated that ''It would be misleading to suppose that frontal type of applications were the main specification of a brand new, more efficient bow type''. The Svaneti bow could open some new issues when it comes to this phenomenon and perhaps shed a light to a new way of junction of the wooden core of composite bows.

The Svaneti bow has horn overlays at the edges of the syah instead of bone, antler or wooden parts.   An interesting find is one Mongol bow found at Chonot Uul, having the end of the ear reinforced with a wood overlay of 80 mm in length and 17 mm in thickness. The bright, dense hardwood could be beech-wood that differs in the light colour from 20 mm thick bow core which was made out of willow. The measured angle of limb and lever end one of 28° (Becker, Rutschke, pp. 83-85). It is interesting that scholars are dating the bow from Chonot Uul back to 12^th century. The thicker core of bow from Mestia could be close to that period which is another fact that suggests approximate time frame of design. Thus the use of additional edge bone parts, and with the absence of side bone plates, can be associated with the appearance of V splice, which is a theory Hall agrees with as well (2006) .

Ventral bone and horn siyah plates

Having examined the bow from Mestia, it is impossible not to notice the ventral bone plate on the great part of the belly on both siyahs, in the place supposed to be a kasan-like ridge. This item is extended by the horn plate that meets the aforementioned horn tip overlays. The V-splice is used to join both of these bone plates to the horn plates and thus stretching to the end of the siyahs.

Dorsal ending bone plates are characteristic of the afore-mentioned bows of Hunno-Bulgar and Turco-Khazar type, as well as of earlier Avar bows with lateral bone plates. Nonetheless, in numerous tombs aside from the frontal tip overlays of Saltovo type, there were also findings of ventral and dorsal tip overlays. In his observations of composite bows from Eastern Europe in the early Middle ages, Kruglov (2005, pp 73-99) cites the remains of frontal bone applications with ventral plates in Alano-Bulgar burials. Savin and Semyonov (1998) present a map of  the prevalence of the  remains of Saltovo type bows on page 274 (fig.25), whilst in their earlier paper they present two technological lines of the evolution of early Middle-Ages bows in Eastern Europe (Savin, Semenov, 1997, p42, fig. 3) (fig.23) with noticeable ventral bone plates in Saltovo  type in the final graphic reconstruction.  Aksenov/ Аксьонов (2005, pp 156-158) (fig.26) engages in an examination of bone tip overlays together with dorsal wedge-shaped plates from the burial of Krasnaya Gorka, stating that their actual length varies between 20 and 25 cm. Worth to mention that most of the  bone tip overlays including those from Carpatian Basin have the scoring on the their upper side in spite of gluing surface along the belly of overlay. It is uncertain were this scoring served wether as a surface for the sinew lashings or dorsal plate itself as there’s problematic assertion wether most of these bone plates are dorsal or ventral.

This paper has already dealt with the structure of the siyah of the partially exposed bow of St Demetrius from Protaton, where the painter used a clear contrast of colours to express not only the cross section but also the ventral bone plate on the belly of the siyah, in the area of the triangular cross section.

It seems that the ventral plates in later Buryat bows were either bone or antler plates, and they spanned over the whole lower part of the siyah between the string bridge and short lateral bone plates added from the side on the ending of the siyah. In his interview  “Нация лука и стрел’’ (http://baikal-info.ru/number1/2007/35/008001.html ), Dylgyr Tsyrendorzhiev notes that the Buryat bow that he reconstructed required usage of fragments of moose antlers in order to elongate the short cattle horns. It seems that the cattle horns were a common construction material in Buriat bows, which is a fact proven by a very old bow that has visible ventral antler tip plates with “bird-eye” patterns and a very long ventral plate on the grip. The cross-section of the siyahs bears obvious resemblance to the cross section of the Georgian bow. Luvsannorov Munkh from Mongolia reports that the bow was brought to Russia by Sukhbaatar, and that the bow in question was 200 years old. (personal communication, October 23, 2016)  It is unclear whether it is  the bow of Ilya Kukshinovich  Shihaev described by George G. Vitt, Jr (1994, JSAA, 66-68).

The Omnogov bow is believed to date from the period of Mongol conquests, in spite of the fact that the wooden material from the bow was examined by a Japanese laboratory stating that it was from 1720, give or take one hundred years. This bow’s cross section of the siyahs resembles the one of the Svaneti bow, but with the difference that it does not have ventral ending plates. (Atex, Menes, JSAA, 1995 pp. 71-75). The Omnogov bow is dated to the empire era by C14 according to  Erdenebat who proposed typological dating of the tomb complex in the 12th - 13th century ( Ahrens et al., 2016, p.86). Bede Dwyer states that the design of the Omnogov bow is old but the bow could be from later period.  Recent bow discoveries suggest that it is a later design than the 14^th century, possibly the late fifteenth or sixteenth century (personal communication, November 6, 2016)

Sinew  wrappings and birch bark

After a thorough examination of the exposed details in the photographs, one can clearly notice remains of animal sinews on the part of the bow’s belly leading towards both bow limbs, especially on the edges, showing the tendency to run to the inner part of the limb. The horn plates (in the vicinity of the grip where a bow limb is fractured) are criss-cross scored. It is most likely that the scoring was placed there in order to enlarge the surface for gluing of sinew lashings. There also is the possibility that the whole limbs of the Georgian bow were wrapped by sinew lashings. Unless we are dealing only with the sides of the horn part of the bow, which is characteristic of bows with one horn plate, wrapping the whole length of the bow limbs with sinew lashings is by no means an earlier formula. This characteristic, albeit not common with well-preserved bows, is met in earlier periods. Jack Farrell and Andrew Hall (2008, pp 86-98), in their examination of two asymmetrical bows from Miran, note spiral wrappings of sinew including the whole bow length reaching up to the nocks. One of these bows has a longitudinal layer of sinew on the back of the bow, underneath the spiral wrappings.  According to Bergman and McEwen, the limbs of later Persian and Indo-Persian bows, which are characterized by incomparably larger width of the limbs than those of their contemporaries - Ottoman and Crimean-Tatar bows, were completely wrapped with sinew. (Bergman,McEwen, 1997, 151). Henry Balfour (1889, pp 220-246) describes the anatomy of the afore-mentioned dissected late-Persian bow noticing a longitudinal layer of sinew along the horn strips made up of a larger number of narrow, parallel strips. Wrappings of sinew at the ends of the grip where there were fixed horn plates in later nomadic bows, were wound with sinew, which is authenticated  by scoring on the bone and antler plates in numerous excavations of Sarmatian, Hun, Avar, Khazar or Hungarian and other burials across Eurasia. Moreover, this characteristic is shared with the preserved Mongolian bows prior to the Conquest Period.  Let us consider the Žagarlant bow as one of many examples.⁹

The Georgian bow from Mestia is exemplary for its sinew wrappings  not only in the area around the endings of the bone plate of the grip and its distinctive ridges, but also in the longer part of the upper bow limb under the position where the siyah with its kasan-like ridge meets the V-splice. Obviously this area is covered with birch bark, which is noticeable from the details on the photographs. Thus, there are indications that the whole bow limb was wrapped with sinew. Moreover, on one part of the ventral bone plate on the belly of the siyah, one can notice remains of layers of sinew, unless we are dealing with some other material.

Let us also tackle the topic of covering the bow. It is indicative of this type of bow that, except in the area of the bone plates, it was completely covered with birch bark. One can easily get the impression that birch bark was not applied with spiral windings as was the case with earlier Mongol and other nomadic type bows , but rather presented the manner of later bows from Qing diynasty period which were characterized by adding a larger number of carefully chosen elegant shorter sheets of bark side by side. This is indicated by the texture on the bark which shows opposite directions of the pattern on birch bark sheets. A short report of  T'an Tan-Chiung (1981, p 191)  leads to this conclusion as well. There may be some confusion here. Some authors refer to spiral wrappings of birch bark, but there is little evidence of the process. It is most likely that small rhomboidal pieces were used giving the impression of a spiral.  Karpowicz goes into some detail on the application of birch bark to Turkish bows. There appears to be an advantage to having the joints between the strips diagonal to the direction of bending. He noticed that usually ‘’the back of the bow is covered in two half-width strips laid side by side, while the grip is covered in as many pieces as necessary’’. Then he concludes that ‘’the use of two pieces side by side are needed in the kasan section, since the bark is not flexible enough to bend smoothly over the ridge’’ (Karpowicz, 2008, p 134)

The surface of birch bark cover, which has a dual purpose (as a protective and decorative material), also shows additional sheets of what is most likely birch bark added to both sides of bow limbs in order to embellish the bow even further. Side strips have a practical value if they are applied over the back and belly strips’ junctions by excluding moisture and protecting the edges of the laminations.

Heart-like decorations placed in the area near the ridges as the extensions of the grip on the rigid part of the bow limb, as well as on the area of the transition of the bow limb and the siyah, will not be commented upon in this paper. Should this be a question of symbolism, it has appeared in many cultures since the ancient times to the Modern Ages, and has been interpreted in different ways. Worth to mention, the bow itself lacks decorations and calligraphy characteristics of many later Ottoman, Persian, or other Islamic realm bows. Specially, there’s no evidence of taliq scripts so called yay kitabeşi either on its back or belly. This could be another argument in favour of the native Georgian bowyer’s work; however, not all the bows were always richly decorated.

Conclusion

Taking into account all of the mentioned arguments, it appears that there is no ground for the bow from Georgia to be considered part of the linear evolution of the composite bow as a whole. One would be more prone to assume that this is a separate local case showing exceptional craftsmanship and knowledge of structural, mechanical and technical characteristics of the Georgian craftsman bowyer who created the bow from Mestia. There is no doubt to the fact that many similar bows must have come out of the same workshop, and yet this is not an argument for the case of the uniformity of archery equipment in the period of time when this bow was used. Unlike nomadic bows, the presented Georgian bow shows artisanship coming from a place with a long armament making tradition. Alongside all of the mentioned observations it is appropriate to note that the design of this bow does not in any case resemble the design of later bows that were under the influence of neighbouring territories such as Safavid Persia that partially occupied Georgia, the Ottoman Empire  or the Crimean Khanate, having in mind that this was the zenith of the usage of bow as armament. Should it be considered that the bow comes from the period of the Middle Ages, which seems highly probable, it makes it possible to assume that this design was rather an advance one, pointing to the transition to later, more advance bow types. In the drawings made by Don Cristoforo De Castelli  (1600-1659, a Roman Catholic missionary that left many written and pictorial documents about contemporary Georgia, there are quite a lot presentations of bows. These drawings, however, do not shed light on some clear conclusion, for the objects appear highly stylized. Several cases, though, do indicate an analogy to the design discussed in this article, notwithstanding numerous analogies to the influences from the neighbouring territories. One of these influences,for instance, was kabak, a competition where horseback archers aimed at a target hanging from a high pole. This was an event characteristic primarily to the Ottoman Empire, but also to the territories of  Iran and the Islamic neighbourhood of  Georgia in general. Hopefully, radio carbon dating will shed some light on the time span of the discussed bow. Nonetheless, remarkable solutions used in the making of the bow itself speak of a unique specimen which mirrors the culture of the Georgian people. This puts Georgia another step forward on the map of territories famous for their rich archery and bowmaking tradition through centuries.

References

Ahrens, Birte; Gončigsuren, Nomguunsuren; Piezonka, Henny. Das mittelalterliche Höhlengrab von Cagaan Chad, Mongolei. Eine Kriegerbestattung am nördlichen Rand der Wüste Gobi, Zeitschrift fur Archaologie des Mittelalters 43(2015) , published by S. Brather, U. Müller und H. Steuer;  Bonn, September 2016

Аксьонов, Віктор Степанович, Луки из погребений салтовского могильника Красная Горка , Проблеми дослідження пам'яток археологи Східної України: Матеріали ІІ-ї Луганської міжнародної історико-археологічної конференції, присвяченої 85-річчю Луганського обласного краєзнавчого музею.  Луганськ, 2005,  pp. 156-158

Notes

1.       See Heather A. Badamo,  Figure 29. St. Theodore Stratelates slaying a dragon, wall painting in naos, 1232/1233,  Monastery of St. Antony at the Red Sea, Egypt.  Figure 75. St. Mercurios slaying Julian the Apostate, wall painting in khurus, 1232/1233, Monastery of St. Antony at the Red Sea, Egypt.  

2.       Scene in question: Saracens attack an icon of St George in the church of St George from second half of 12^th century in Phavnisi, Upper Svaneti doesn’t support design in question.                                                                

3.        Known also as Pentateuque dit d’Ashburnham ou de Tours and kept in Bibliothèque nationale de France, Département des Manuscrits, NAL 2334 f.25r

4.       Vatican City, Biblioteca Apostolica Vaticana, ms. Vat. gr. 747, fol.47r

5.       It is also worth to mention tile showing scene of archer beside 54 tiles on the doors of Ravello’s Cathedral attributed to Barisano da Trani, 13^th century

6.       See p.147 in Wealth of the Roman World, Gold and Silver A.D. 300-700 (edited by J.P.C. Kent and K.S. Painter), British museums publications LTD, London 1977. Also, for presentation of Sasnid bows in Hermitage museum, K.B. Тревер, В.Г. Луконин Сасанидское серебро.Собрание Государственного Эрмитажа.Художественная культура Ирана III-VIII веков. // М.: «Искусство», 1987.

7.       For further details on early Hunnic and Avar bows consult also A.Kollautz, H. Miyakawa 1970, А. М. Савин, А.И. Семёнов. 1992. pp 201-205. Gad Rausing 1967., Coulston 1985.

8.       For further informations consult P. Klopsteg, “Turkish Archery and the Composite Bow”, who was quoting J. Hein “Bogenhandwerk und Bogensport bei den Osmanen” translating Mustafa Kani; Saracen Archery 1970, Karpowitz 2008.

9.       Consult Steppenkrieger pp 184-190 for  reconstruction done by Joachim Rutschke

 Figures

Figure 1. the 12th-century Georgian Jruchi II Gospel,  manuscript H1667, Page 111R., National Center of Manuscripts

Figure 2. the 12th-century Georgian Jruchi II Gospel, Page 186V. fragment. National Center of Manuscripts

Figure 3. Saracens attack an icon of St George in the church of St George, Phavnisi, second half of 12^th century. Scheme (after Привалова and Grotowsky)

Figures 4. and 5. Detail of Esau returns from the hunt scene, Pentateuque dit d’Ashburnham ou de Tours, Bibliothèque nationale de France, Département des Manuscrits, NAL 2334 f.25r (retrieved from   http://gallica.bnf.fr/ark:/12148/btv1b53019392c)

Figure 6. Detail from a miniature showing the Exodus , Pentateuque dit d’Ashburnham ou de Tours, Bibliothèque nationale de France, Département des Manuscrits, NAL 2334, fol .68r, 6th or 7^th century (retrieved from http://gallica.bnf.fr/ark:/12148/btv1b53019392c)

Figure 7. Vatican City, Biblioteca Apostolica Vaticana, ms. Vat. gr. 747, fol.47r: Esau returns from the hunt

Figure 8a.  St. Demetrios, right next to the presentation of St.Sava, the founder of the Serbian independent orthodox church and eigth other warrior saints, in the Holy Apostles Church in Pećka Patrijaršija, 70’s of 14^th century

Figure 8.b St.Demetrios, Presentation of the Virgin Mary Church, Nova Pavlica,  Serbia, memorial of Musić brothers and their mother Dragana, a sister of Duke Lazar. End of 14^th century.

Figure9. St. Demetrios, the Holy Apostles Church in Pećka Patrijaršija, late 70’s of 14^th century, detail

Figure 10. Hunting scene, Chronicle of George Harmatolos ( Chronicles of Tver). 14^th century manuscript

Figure 11. Stone relief from Dagestan, Hunting scene, 12th century

Figure 12. Scene of archer, tiles on the doors of Ravello’s Cathedral, 13^th century

Figure 13. Sasanid Plate of the King Shapur II  hunting lions, 4^th century AD Hermitage Museum, St Petersburg

Figure 14. Byzantine silver plate, scene:David receiving Saul’s armor for his battle with Goliath, c. 629 AD, Metropolitan Museum of Art (source:https://commons.wikimedia.org/wiki/File:David_receiving_Saul's_armor_for_his_battle_with_Goliath,_Byzantine_silver_plate.png)

Figure 15.Plate with Hercules Wrestling the Nemean Lion, 500-600 A.D. Silver. Cabinet de.Médailes of the Bibliothèque nationale de France, Département des monnaies et antiques, Paris.

Figure 16. Egyptian ivory, 6th century

Figure 17. Fresco of St. Demetrios by iconographer Manuel Panselinos, in Church of Protaton on Mt Athos (c. 1290)

Figure 18. Antonio Pisano Pisanello, bow, bow case and quiver of Byzantine Emperor John VIII Palailogos, c. 1438.

Figure 19. Frontal tip overlays , 1-10:  from Sarkel, Beloy Vezhi, Turovo, Bolgar, Bilyar, Kylasova goropisca, Solyanin. (After A.F.Medvedev)

Figure 20. Summary of bows in Bayirqu, Zabaykalye, type 5 (after Hudyakov)

Figure 21. The antler overlays of the bows from Sarkel-Belaya Vrezha hillfort (after Флёрова)

Figure 22.  Stratigraphical distribution of  bow decorations – frontal tip overlay  (after Флёрова)

Figure 23. Two lines of tehnological development of a composite bows in early medieval Eastern Europe (after Savin, Semenov 1997, p. 42, fig. 3): I - ,, Hunnic-Bulgar- Avars "(" Turkic-Khazar- Avar ") line. Methods and schemes of assembling: A - "Аvar type; ГБ – Hunnic-Bulgar (" Turkic-Khazar ") type. II -" Khazar-Saltovo-Hungarian " line. Method and schemes of assembling: X -" Khazar type; C - "Saltovo" type: В- Hungarian type,  variants 1 and 2, БТ – bow from Borotal, MT – bow from Monguno-Taiga

Figure 24. Exavations in Ossiniki, central lateral bone plates and frontal siyah bone rods for the composite bow (after Savinov, 1981)

Figure 25.  map of  the prevalence of the  remains of Saltovo type bows (after Savin, Semenov,1998)

Figure 26. frontal tip overlays, dorsal and ventral wedge-shaped plates from the burial of Krasnaya Gorka (After Аксьонов, 2005)

Zoran Pavlović was born in Prokuplje, in the Republic of Serbia. Devoted to his family life, he decided to stay in his home town and to achieve his personal potential in the field of education, as a history teacher. Apart from his dedication to the younger generations as a teacher and educator, his passion  is his continuing research work in the field of the history of archery, especially archery in the region covered by the Serbian medieval state, as well as other states in the Balkan region from the medieval times. His passion for bows and archery in general as a cultural and historical phenomenon develeoped at the same time as his interest in history and culture of Native Americans, as early as he was a child at the end of the 1970s. Eager to follow new discoveries, he was led to an interesting fact – that archery as part of military history of medieval Balkan states was under-researched, in spite of numerous remains of a rich cultural heritage of medieval Serbian, Bulgarian and Bosnian states, as well as of territories covered by the Byzantine realm as a whole. In view of the broadness of this topic, he focuses on composite bows, the technology of their manufacturing, and the anatomy of bows whereby he artfully uses art presentations and archeological findings, and interacting with different cultural spheres through history, starting from prehistory all the way to the modern ages. His most frequently used method was comparative analysis accompanied by all the available historical sources and literature. His research brought him together with a large number of enthusiasts and top experts of archery as a cultural and historical phenomenon around the world, people who he has excellent co-operation with. These people have influenced his ever-growing enthusiasm, even more so considering the fact that he is not supported by any institutions. So far, he published two research papers connected to archery in medieval Serbia: 'L'аrcherie мédiévale serbe, Si peu connue et si riche’ and  ‘Serbian composite bows up to the third quarter of 14^th century – some views on their origin, structure and design’. He held several successful lectures in museums all over Serbia, presenting archery of medieval Serbia and the Balkans, and he is currently working on new research studies. In recent years he has acted as a member of the Society of Archer-Antiquaries where he published a research paper. In the last 15 years he has actively taken part in discussions on ATARN website. His biggest support are his family and friends. Apart from his Serbian mother tongue , he is fluent in Croatian, Macedonian and Bulgarian, as well as English, French and Dutch, alongside a large number of languages from the Slavic language family such as Russian.

Back