Back

A comparative analysis of one well preserved composite bow from Georgia

       By Bede Dwyer

Introduction

This is a report on my impressions of the bows that I have been made aware of from ethnographic  research in Georgia. Two bows were described in photographs and drawings. One was a composite bow with a strongly pronounced knee where the rigid tip comes off the bending section of the limb. The other was of similar structure, but instead of having two limbs, had six. This is unique in my experience, though multi-limbed bows have a long history.

In this report I do not make specific reference to the find spot of these bows or their dating, both of which will be covered elsewhere by others. The find spot was a church in Svaneti in Georgia, where the bows had been preserved for centuries. The dates of the two bows are yet to be precisely determined.  It is thought they are 11^th to 12^th century CE.

The bibliography has many references for further reading as well as works cited in this report.

The two limbed bow

This bow is of interesting construction showing strong links to reflexed composite bows over a large area of Eurasia. Its major features are as follows.

1.       A strongly reflexed grip.

2.       A very strongly reflexed rigid tip.

3.       A bone reinforced grip.

4.       Sinew reinforcement on the back of the bending section of the arms.

5.       Four strips of horn on the bending section of the limb.

6.       The long tips are not reinforced except for an applied horn strip on the back used to form the nock.

7.       Decorated birch bark covering the sinew for waterproofing

8.       A slight curving of the tip to the back of the bow forming an incipient recurve.

The reflexed grip is an ancient device first occurring in composite bows with the Scythians CITATION Dwy1 \l 3081   (Dwyer B. , Scythian Style Bows Discovered in Xinjiang, 2003). When Scythian bows were superseded by bows with straight rigid tips, the handle area was strongly reflexed using a different construction method from the early bows[1].  One of the earliest bows of this type found at Yrzi already had a reflexed handle. It was built by splicing the two limbs together at the handle and then covering the join with sinew. The bow under discussion had the grip in a single piece, possibly spliced into the limbs. Bone plates enclosed the grip on two sides, almost meeting on the belly. The decorative effect of the bone was secondary to the practical purpose of making a strong narrow grip area. This feature allowed the arrow to pass the bow with minimal deflection which allowed stronger, stiffer arrows to be used.  The grip reflex also produced a setback handle which meant that the bowstring could give a longer stroke to the arrow increasing the efficiency of the bow[2]. By making the handle narrow and deep, the archer had a firmer control of the attitude of the bow. A bow with a setback handle can twist in the hand making it unstable so the form of the handle helps counter this. White gripped handles on bows are recorded in the poems of the Oguz[3].

1 Comparative Cross Sections of Composite bows   The sharp angle between the bending section of the limb and the rigid tip is to increase the force applied to the bending section. In the Yrzi bow and other early examples have the tips curving gradually from the bending sections. However, later bows show stronger angles in this area indicating that bowyers had found a way to make the bows more effective.  The crab bows of India still had this construction in the nineteenth century. The method and use of materials are different in the Indian bows because of a much greater use of sinew in their construction.  External appearances often hide profound structural differences.

There are other differences in the construction of the core which may be significant. The convex cross sectional shape of the core in the bending parts of the limb are similar to both later Turkish bows and the Persian bow in the style called chahār kham (four curved) in Farsi CITATION McE79 \p 89-96 \l 3081   (McEwen E. , 1979, pp. 89-96). In both cases the horn covering of the belly was in single pieces[4]. However, the excavated example shows how useful using strips of horn was for this design. Fitting four strips around the core may have been easier than simultaneously bending the horn along the limb and around its cross section.

The proportion of wood to horn and sinew is also significant. Karpowicz stressed that a bow with a greater thickness of wood in its core was a more stable bow CITATION Kar08 \p 45 \l 3081   (Karpowicz, 2008, p. 45), while Arab authors thought it was an adaption for regions that were wetter[5] .

The use of bone on the grip as a reinforcement was already disappearing at the time this bow was made. Its high point was among the Avars where the grips were completely enclosed by bone plates.  Magyar bows also had grip reinforcements and so did the bows used by the Romans. However, the use of handles either made from separate pieces of wood or with overlays glued onto the core makes it possible to use stronger wood for this area and the need for the bone reinforcement is removed. Bone for decoration clearly survived the new technology, but it gradually faded from use.

Sinew was the preferred backing material for composite bows. Sinew fibers set in glue are excellent for storing energy in tension. Strongly reflexed during manufacture, they are pre-loaded and very efficient.

Persian manuals  CITATION Kho13 \p 75 \l 1033  (Moshtagh Khorasani M. , 2013, p. 75) recommend four strips of horn as a good choice for making a bow[6]. The cross sections of the bow show a rounded belly with the strips arranged as segments of the curve. Using strips rather than a single piece of horn make this matching of the glued surfaces easier. Strips are easier to manipulate because they are individually more flexible[7]. There are other reasons for using strips of horn. Not all horns can supply a wide enough strip to cover the belly of a bow. Ibex horn is a case in point as not all species have wide horns. More of the horns can be used if they are cut into strips than cutting only one or two wide pieces from a horn.

By not having two strips of bone or antler to support the sides of the tip as in earlier bows, the tip weight is significantly reduced. This usually results in a faster bow making it more efficient with lighter arrows which were used for long distance harassment in war and also in sport. The use of a horn strip on the back of the tip is unusual. Antler strips for this purpose have survived from earlier times, but horn has rarely survived because it is easily destroyed by insect action in the grave. The bow from Chonot Uul[8] in Mongolia has an applied wooden nock in the same position as this bow’s horn reinforcement. While horn is heavier than wood in general, the thin tapered overlay in this bow probably does not significantly increase the tip weights. These overlays, considering the long working lives of composite bows, were perhaps replaceable since the animal glues that held them in place would have been susceptible to localized heat.

The sinew layer is protected by a covering of birch bark to waterproof it and control moisture loss. This has been decorated extensively and motives like a stylized heart are repeated on the limbs. Birch trees not only supplied bark, they also produced wood that could be used in bow construction and that was also used for arrows.

One of the more subtle, but significant, features of the bow is found in its gently curving tips. These preview later Mongol bows  CITATION Bec05 \l 1033 (Becher & Rutschke, 2005) which re-introduced the recurved tip[9] with its increase in efficiency. In later bows of the Chinese, Persians and Turks, recurved tips became more prominent CITATION McE91 \l 1033  (McEwen E. M., 1991).

The six limbed bow

The construction of this bow is very similar in detail to the previous bow with the extraordinary feature of having six limbs instead of two. The limbs are more rectangular in cross section in the parts that are meant to bend. This would have made it easier to match the flexibility of each limb to the others. This is a great technical problem. Composite bows rely of great precision in their construction since the normal processes of tillering[10] used on self bows are not easily applied.

The roots of the three limbs on each side of the handle are critical to the stability of the structure. Minor errors at this point could create critical problems for the functioning of the limbs. Composite bows can have their limb profiles affected in the bracing process. In a normal two-limbed bow, this is corrected over the archer’s knee in a period of 30 seconds after the string is attached. During this time the bow is still malleable. With the six limbs of this bow, this process would be very hard to accomplish. The rectangular cross sections would make limb behavior more predictable, but it is still not easy to string one of these bows.

Analysis of the bow structure reveals several important points.

1.       The angle of the ears to the bending sections of the surviving limbs is not as great as in the two limbed bow. This consistent with an earlier date for the bow and also with a desire to make the structure as simple as possible while still making a bow of the normal type.

2.       There are two strips of horn on each limb covering the belly of the bending sections. This suggests bellies with multiple horn strips were the standard construction in the area this bow was produced.

3.       The limbs are not particularly broad in relation to their thickness. This might indicate an adaption to having three limbs alongside each other, but in may also be another standard constructional feature of the area where the bow was produced.

4.       The nocks are carved in an applied horn strip which is also like the other bow and may be a diagnostic feature of bows from this region over a considerable amount of time.

5.       Despite the multiple limbs, the bow appears to be designed to operate as a normal non-contact siyah bow of its era.

6.       The ears are constructed with a triangular cross section that changes into a rectangular one near the nock. This is consistent with the two-limbed bows.  Analysis of paintings or depictions of bows in metalwork like Sasanian Persian[11] examples would not reveal this feature.

7.       Six-limbed construction appears to be unique to this bow and this area.

The bowstring could be made of three individual strings twisted together. About a third of each string could be free at each end with individual loops for each bow tip.  In use the end loops would be pulled together making the bow look as though it had one piece string. It has been suggested that a single string with larger loops could be attached to all three limbs at each end. This would be suitable for a ceremonial object. The main objection is that there would be wear and tear between the sides of the ears where they were pushed together by the bowstring loops. However, if the bow was rarely drawn, this would be a minor problem.

Figure  SEQ Figure \* ARABIC 2  A proposed string shape

The obvious reason for a bow to have three times the limbs of a normal bow is to make it three times as strong or to at least make it appear to be three times as strong. There is a double-limbed bow in the Topkapi Palace Museum [12] which was a present from the Shah of Iran to the Ottoman Sultan. The story that is associated with it is that the Shah sent it as a challenge to see if anyone in Ottoman Turkey could brace and draw it. It was so strong that no one could even put the string on it and it was left hanging on the wall of the Vizier’s palace until one of his gardeners, later to become Deli Hasan Pasha, took it down and strung it. In this case, a multi-limbed bow was a diplomatic gift meant to embarrass the recipient. The complexities of stringing a six-limbed bow may well have tested the skill even an experienced archer.

It is more likely that it was meant as a part of the ceremonial regalia of a prince to show the skill of his craftsmen and the power of his arm. It would be interesting to discover if there are literary references to such weapons. Mediaeval Georgian literature is rich and there might be allusions to the bow that have not been recognised because no one has so far conceived of such a strange weapon.

In most cases, multi-limbed bows were mounted on siege weapons like the giant crossbows called in Persian, kamān-i gav, or ox bow. In that case, three separate bows were mounted on a frame and linked together with cables. In other weapons, bows were mounted above each other and shot individual arrows. In no case is there a hand bow with three sets of limbs.

Methods of Construction

There are three likely methods of core construction in the historical development of straight eared bows.

Type 1

Figure  SEQ Figure \* ARABIC 3 Type 1 bow construction

This would have proceeded in 4 stages.

1.       A suitable piece of timber would have been selected and cut to length and width.

2.       At the points were the tips started to be differentiated from the bending sections, the tips would have been bent in the direction of the back of the bow. Steam or boiling water could be used. It is also possible to choose two matched pieces of wood that had natural bends and join them at the handle with a diagonal splice.

3.       The bow core would be thinned on the belly side where the limbs were meant to bend.

4.       Horn would be glued on the belly side to reinforce the bending areas. Sinew would be glued to the back for strength in tension. It could also be wrapped around the locations where the horn ended to strengthen the bond with the wood.

This is a simplified version of the earliest type of rigid-ended bow. It is also possible to modify this type by bending the core either side of the handle to induce further reflex[13].  Another modification would be to reinforce the tips with bone lathes on either side so that they could be made thinner and lighter. The core wood may have been to flexible otherwise.

Type 2

Figure  SEQ Figure \* ARABIC 4  Type 2 bow construction

This method is more sophisticated because the bowyers take advantage of the different properties of certain species of trees[14]. Some wood is flexible and springy with other wood can be very stiff.

1.       A long flat lathe of flexible wood is cut out to the length and maximum width of the bow.

2.       It is bent in the same manner as the Type 1 bow, but the process is much easier because of its thinness.

3.       Three pieces of harder wood are glued to the belly side. Each tip is reinforced this way and so is the area of the handle. When these sections are narrowed they still retain their stiffness. In the case of the tips, they are significantly lighter than the Type 1 bow. Again they are reinforced by bone strips on the sides. These strips additionally cover the glue lines of the belly laminations. Horn and sinew are added as in the Type 1 bow.

Type 3

Figure  SEQ Figure \* ARABIC 5 Type 3 bow construction

In its final form, the straight eared bow has its various sections constructed of wood of the appropriate mechanical characteristics for their purposes.

1.       Five pieces of wood are cut out to form the two tips, two bending sections and the handle. For the tips and handle, harder stiffer woods are chosen. The bending sections can be made of more flexible wood if desired.

2.       The tips are bent at their bases to create the correct angles for the bow’s profile. The handle is spliced into the two arms of the bow and then the two tips are spliced into the other ends of the bow. Simple “V” shaped joints are all that is necessary for a strong glued construction.

3.       The core is fitted with its horn belly strips and the back is covered in sinew in glue. Since the tips are much stronger than the Type 1 and Type 2 bows, they do not require bone reinforcement on their sides. This type of bow sometimes had a glued-on piece on the back side of the tips so that the nock for the bow string could be carved into it. This could be wood, horn or antler since it did not represent a significant increase in tip weight.

These diagrams are very simple and do not show the layers of sinew or the reflex introduced into the handle areas of these types of bows in some historical examples. This would have created a visually complex and possibly misleading set of images.

Type 1 is represented by the bow found at Yrzi in Iraq CITATION Cou85 \l 1033  (Coulston, 1985)[15]. It is an early example with a diagonal splice at the grip to join its two halves. The second type has been found in Xinjiang in China where an example has been excavated with multiple strips of horn on the belly CITATION Sel03 \l 1033  (Selby S. , Two Late Han to Jin Bows from Gansu and Khotan, 2002). The third type has been excavated in Mongolia and there are detailed reports on the structure of this class of bow CITATION Bec05 \l 1033  (Becher & Rutschke, 2005).

Special Features of the Bows

It is clear that the two bows, aside from the obvious features of the six-limbed bow, represent two different constructions. It can be argued that the six limbed bow is both simplified and slightly more primitive than the two limbed bow. The features that could be considered as simplified would include the rectangular cross sections of the bending sections of the limbs and the simple profiles of the ears. The handle of the six-limbed bow is completely comparable to the handle of the two limbed bow both structurally and in design. The angles of the tips of the six limbed bow are reduced in comparison with the two limbed bow. This might be a decision of the bowyer to make it easier to balance the limbs. It could be also because the bow was never meant to be especially efficient. However, it could also be because the two limbed bow comes from a more developed design.

The two-limbed bow has noticeably curved ears. This presents certain problems regarding how the bowstring sits on the belly side of the tips, which can have an effect on the stability of the bow.  The original photograph of the bow found at Qum-Darya in Central Asia[16] shows this feature though the surviving pieces are not a match either for the photograph or the similar curves of the current bow[17]. Strongly curved ears are shown in some silver plates now in Russian collections. These were made in the seventh or eighth centuries and show the siege of a city from a Biblical story. They contrast with earlier Sasanian Persian plates showing hunting scenes where the bow tips are almost always straight. The existence of the Qum-Darya bow implies this feature is not unique, but it is much more distant in time from the later examples. The Georgian bow sits very close in time to the appearance of the fully developed Turkish and Persian bows. It shares some constructional details with the Persian bows and could represent an origin of the feature the Turks called the kasan.

Conclusion

These two bows are both unique Georgian historical objects and important witnesses to the evolution of bow design. The stunning tour de force that the six-limbed bow represents is not its only value. It is structurally significant also since it shows an insight into how ordinary bows would have been made in its period. The two-limbed bow appears to be a well designed weapon showing that its makers had a profound understanding of the construction of composite bows from wood, horn and sinew. It is also an example of a unique Georgian contribution to bow design. Either of these bows, by itself would have been a major find. That two such significant bows were preserved in an ancient church is remarkable.

Appendix

Bow Terminology

The bow has several unusual uses of common words in its terminology. For the purposes of this report I have prepared a list because the words have other meanings in English and some non-English words are sometimes substituted.

Handle                 The part of the bow held in the hand by the archer.

Tip                         The extreme ends of the bow.

Nock                     The slot or grooved carved into the bow to accept the loop of the bow.

Limb                      The part of the bow between the handle and the tip.

Arm                       Same as the limb.

Knee                     The sharp angle between the bending part of the bow and the ear.

Ear                         The part of the end of the limb that is rigid. It is a translation from Persian.

Guše                The Persian word ﮔﻮﺸﻪ sometimes spelled gushé or gusheh meaning ear.

Siyah                     The Arabic word for the ear of a bow.

Kasan                   The Turkish word for the ridged portion of the ear. It is not synonymous with ear.

Since composite bows were not a native development for the English speaking part of the world a specific nomenclature did not develop for the parts of this type of bow. Surprisingly most languages use parts of the human body as sources for the names of sections of the composite bow.

Bibliography

BIBLIOGRAPHY Becher, H., & Rutschke, J. (2005). Die Mongolenbogen deraustellung "Steppenkrieger" im LVR-Landesmuseum Bonn. Restaurierung und Archaeologie(5), 73-91.

Bergman, F. (1939). Archaeological Researches in Sinkiang. Reports from the Scientific Expedition to Northwestern Provinces of China. Archaeology, I(VII), Pl. XIII A, pl. 18,10.

Cernenko, E. V. (1983). The Scythians 700-300 BC. Osprey Publishing Ltd.

Chernenko, E. V. (1981). Skifskie Luchniki. Kiev: Naukove Dumka.

Coulston, J. C. (1985). Roman Archery Equipment. In M. C. Bishop (Ed.), The Production and Distribution of Roman Military Equipment. BAR International Series (Vol. 275, pp. 220-336). Oxford.

Dennis, G. T. (1985). Three Byzantine Military Treatises. Washington, D.C.: Dumbarton Oaks.

Dwyer, B. (2003). Scythian Style Bows Discovered in Xinjiang. Journal of the Society of Archer-Antiquaries, 46, 71-82.

Dwyer, B. (n.d.). SCYTHIAN-STYLE BOWS DISCOVERED IN XINJIANG . Retrieved from Asian Traditional Archery Research Network: http://www.atarn.org/chinese/scythian_bows.htm

Dwyer, B., & Khorasani, M. M. (2013). An Analysis of a Persian Archery manuscript written by Kapur Čand,. Revista de Artes Marciales Asiáticas, Vol. 8(1), 1-12., 8(1), pp. 1-12.

Dwyer, B., & Moshtagh Khorasani, M. (2012, Marzo). Jāme al-Hadāyat fi Elm al-Romāyat [Complete Guide Concerning the Science of Archery]. Quaderni Asiatici, 97, 45-60.

Dwyer, B., & Moshtagh Khorasani, M. (2013). An Analysis of a Persian Archery manuscript by Kapur Čand. RAMA (Revista de Artes Marciales Asiáticas), 8(1), 1-12.

Faris, N. A., & Elmer, R. R. (1945). Arab Archery An Arabic Manuscript of about A.D. 1500 “A Book on the Excellence of the Bow and Arrow". Princeton: Princeton Univertity Press.

Hall, A. (2006, April 2). The Moshcheveya Balka Bow. Retrieved from Asian Traditional Research Network: http://www.atarn.org/mongolian/moshcheveya_balka.htm

Hein, J. (1925, 1926). Bogenhandwerk und Bogensport bei den Osmanen. Der Islam, 14, 15, 289-360, 1-78;233-294.

Herodotus. (n.d.). The History.

Karpowicz, A. (2008). Ottoman Turkish bows, manufacture & design. Canada.

Khorasani, M. M. (2006). Arms and Armor from Iran The Bronze Age to the End of the Qajar Period. Tübingen: Legat Verlag.

Khorasani, M. M. (2010). Lexicon of Arms and Armor from Iran: A Study of Symbols and Terminology. Tübingen: Legat.

Klopsteg, P. E. (1934,1947,1987). Turkish Archery and the Composite Bow. Manchester: Simon Archery Foundation, Third Edition.

Korkut, D. (1974). The Book of Dede Korkut. (G. Lewis, Trans.) London: Penguin.

Lamotte-Boudot, A. (1968). Contribution A L'Etude De L'Archerie Musulmane Principalement D'Apres Le Manuscrit D'Oxford. Damascus: Institut Francais De Damas.

Latham, J., & Paterson, W. (1970). Saracen Archery. London: Holland Press.

McEwen, E. (1979). The chahār kham or 'four-curved' bow of India. In R. Elgood (Ed.), Islamic Arms and Armour (pp. 89-96). London: Scolar Press.

McEwen, E. M. (1991, June). Early Bow Design and Construction. Scientific American.

McLeod, W. (1970). Composite Bows from the tomb of Tut'ankhamun (Vol. III). Oxford: Griffith Institute.

Moshtagh Khorasani, M. (2013). Persian Archery and Swordsmanship: Historical Martial Arts of Iran. Frankfurt am Main: Niloufer Books.

Moshtagh Khorasani, M., & Dwyer, B. (2012). A Persian Manuscript on Archery, Spear Fighting, Sword Tempering and Lance Fighting and Horsemanship by Šarif Mohammad the Son of Ahmad Mehdi. Pan-Asian Journal of Sports & Physical Education, 4(1), 1-17.

Paterson, W. (1969). The Sassanids. Journal of the Society of Archer-Antiquaries, 29-31.

Phillips, E. D. (1965). THE ROYAL HORDES Nomad Peoples of the Steppes. Thames and Hudson.

Rausing, G. (1967). The Bow: Some Notes on Its Origin and Development. Rudolf Habelt.

Rudenko, S. I. (1970). Frozen Tombs of Siberia The Pazyryk Burials of Iron Age Horsemen. London: J. M. Dent and Sons Ltd.

Selby, S. (2000). Chinese Archery. Hong Kong: Hong Kong University Press.

Selby, S. (2002). Two Late Han to Jin Bows from Gansu and Khotan. Retrieved from Asian Traditional Archery research Network: http://www.atarn.org/chinese/khotan_bow.htm

Selby, S., Karpowicz, A., Baker, T., Dwyer, B., & al., e. (2002, July 17). Letters - The Khotan Bow. Retrieved from Asian Traditional Archery Research Network: http://www.atarn.org/letters/khotan_bow/khotan_discuss.htm

Wang, B. (2001). The Ancient Corpses of Xinjiang. (B. Wang, Ed., & V. Mair, Trans.) Xinjiang People's Press.

Wulff, H. E. (1966). The Traditional Crafts of Persia. Cambridge, Massachusetts and London: Massachusetts Institute of Technology.

Yücel, Ü. (1998). Türk okçuluğu. Ankara: AYK Atatürk Kültür Merkezi Başkanlığı.

Zhuo, X., & Ma, C. (1998). ARCHAEOLOGICAL TREASURES OF THE SILK ROAD IN XINJIANG UYGUR AUTONOMOUS REGION. (Z. Xuejun, & M. Chengyuan, Eds.) Shanghai: Translation Publishing House.

 Biography

Bede Dwyer was born and still lives in Sydney, Australia. He has pursued his interest in oriental archery for approximately 36 years of his adult life. In that time, he has researched the archery traditions of Turkey, Mamluk Egypt, Iran, Northern India, Mongolia, China and Korea. Rather than just compile information from old books, he has actively tested equipment and techniques. He makes archers’ rings, arrows, quivers and bow cases. He was the first person to shoot an arrow guide of the style of a Persian nāvak (Arabic majrā) in the American Flight Archery Tournament at Bonneville Flats in Utah. He also uses the Ottoman siper overdraw in flight shooting and briefly held a world distance record for shooting a hunting arrow with a traditional composite bow.

For more than a decade he has been publishing articles in the Journal of the Society of Archer-Antiquaries on archaeological remains of archery equipment ranging from ancient repeating cross bows to mediaeval closed quivers. He was invited by Stephen Selby to read through and comment on some of the drafts of Chinese Archery and later to present a paper at the opening of the archery collection in the Maritime Museum of Hong Kong. He attended two horse archery festivals in Iowa, where he presented a talk on arrow guides with a demonstration, and met many talented archers and bow makers. Dr Khorasani asked him to edit some sections of both Arms and Armor from Iran The Bronze Age to the End of the Qajar Period and the Lexicon of Arms and Armor from Iran. He also worked with his old friend Dr Charles E. Grayson on his collection in the University of Missouri Columbia and did some research for his book on the collection, Traditional Archery from Six Continents: The Charles E. Grayson Collection. This was particularly satisfying since it was Dr Grayson who introduced him to flight archery with all its challenges.

Over the past few years he has regularly attended the World Traditional Archery Festival in South Korea, on occasion presenting papers at seminars there. This enabled him to meet archers and researchers from the cultures he had been studying. Recently he has been able to set more time aside to follow up on his research and is planning two books on archery to be published over the next few years. In 2012, he attended the traditional archery contest held in Qinghai province in China which enabled him to meet traditional archers from the countryside as well as Uighur archers and inner Mongolian archers.  He was able also to see the products of a new generation of Chinese traditional bowyers and fletchers. There were many competitors from the rest of the world as well to share ideas and compare equipment.

Back