The Archers of IslamAuthor(s): W. F. PatersonSource: Journal of the Economic and Social History of the Orient, Vol. 9, No. 1/2 (Nov.,1966), pp. 69-87Published by: BRILLStable URL: http://www.jstor.org/stable/3596173 .Accessed: 23/05/2011 05:45

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=bap. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Journal of the Economic andSocial History of the Orient.

http://www.jstor.org

THE ARCHERS OF ISLAM BY

W. F. PATERSON

'...archery is dependent on five articles: the bow, the string, the arrow, the thumb guard and the archer'.

Taybughi al-Ashrafi 1) i. Introduction

Though the bow is one of the most important early weapons devised by man and its impact on history has been considerable, it has rarely been given the detailed attention that it deserves. It merits study as it has a degree of complexity which can only be appreciated by actual use and an understanding of its basic mechanics. One may add to this an understanding of the vocabulary and technical terms, as and when they may appear, and a fuller appreciation of the actions of archers in miniature paintings.

Success with the bow results from hard work and regular practice, for only by these means can a degree of mastery be achieved, enabling the archer to draw to the limit of his strength and yet shoot under full control. If one watches an experienced archer, his actions appear smooth and effortless, which is as it should be, but to the onlooker this can give a false impression, as the considerable effort needed to draw a bow is hidden by skill and training.

While the history of the old English longbow represents the finest tactical use of this weapon in a defensive role, the composite bow in the Middle East was of no lesser merit in the hands of the mounted archers when used as a weapon of offence.

2. The Composite Bow

The composite bow, as used by the Islamic horse-archers, represents one of the most efficient forms of this weapon ever devised and its

I) British Museum MS Additional 23489, fo. I9b.

70 W. F. PATERSON

construction is worthy of detailed consideration. The most important source of information is Mustafa Kani, Telchis Resail er-Rumat, Istanbul, A.D. I847, with Joachim Hein's dissertation on this work in Der Islam, xiv (1925), pp. 289-360, on which P. E. Klopsteg based his translation and commentary in Turkish Archery and the Composite Bow, which appeared as a private edition in 1934 with a revision and reprint in 1947. This work is mainly concerned with the flight bow of the Ottoman Turks, which is a specialised form of this weapon devised for sporting purposes. Flight shooting was a favorite pastime among most archers in the Middle East, and its sole object is to achieve maximum distance without regard to accuracy. The techniques involved require special skill, and exponents of this art among the Turks, where it reached its highest perfection, could send their light and delicate little arrows over 6oo yards. However, neither the bows nor the arrows used with them were suitable for war, though this method of shooting has an obvious application in the long-range barrage against a massed target up to a limiting range of about 400 yards.

Valuable contributions have also been made by Henry Balfour ) and Commander H. S. Hamlin, Jr. 2). Their published works give a very complete picture of the internal structure of the Eastern bows. During recent years painstaking and careful dissection of a number of Oriental composite bows have been made by Edward McEwen to determine the exact details of their construction, followed by the making of replicas, using the original materials, to prove the accuracy of his observations. Discussions with him have added considerably to our knowledge of the finer points in the art of the craftsmen who made these weapons.

With reference to fig. I the construction of a typical bow was as follows. The craftsman started with a wood core, the main requirement of which was that it should absorb glue well. Varieties of maple, cornus,

x) Journal of the Anthropological Institute, XIX, 1890, pp. 22zz9-244 and JRAI, LI, 1921, pp. 289-306.

2) "A Study of Composite Bows", Archery (magazine), April, May, June, July, and September, 1948. Palm Desert, Calif.

THE ARCHERS OF ISLAM 71

SIYAK wvH TEM•'oRARY NoCK SIYAH

Ha'RN•/ •:,- t? ; s,\,

/

?•,• i

1'A 1T

SINEW

,,,, It

p1=

CORD

IB7FAN3AQ

PERSIAN

Sii

Fig1 Fig. x

72 W. F. PATERSON

mulberry and similar types are among the better woods for this purpose, and dissected bows often show different woods were used for different sections of the core. This is usually in five sections, two siyahs, two

dustdrs and the grip. The siyahs provide the relatively thick and un- bending ends to the bow, whose purpose will be discussed below, while the dustdrs are thin and flexible, as all the bending when the bow is drawn takes place in this section.

The parts were then glued together with fishtail splices about 34 inches in length. In this way the actual handle, or grip, of the bow is about 5 inches long, but about I2 inches with its splices at either end. The form of the wood core determines the final shape of the bow and the curve of the siyah in relation to the dustdr. In the case of Persian and Turkish bows-using this latter adjective in its wider sense-the siyah is smoothly curved over its length through about 600, while the Indo-Persian ones have a more sharply curved knee near the base of the siyah where it joins the dustdr giving a curve in the siyah through about 900 or, in the case of the so-called 'crab' bow from India of as much as 1200 (see Plate III). To achieve this amount of curve the

siyah is made in two parts spliced together as otherwise weakness would result due to there being no continuous run in the grain of the wood. Craftsmen appear to have preferred splicing to increasing the curve of the wood by heat or of going to the trouble to find a natural growth of approximately the right shape.

The s~yah at this stage was made longer than the final form, as shown in the top left-hand drawing, to allow a temporary nock to be cut on the 'wrong' side and, as will be explained, this was used to control the shape of the bow during manufacture. The siyah, though carved from a single piece of wood, is in two sections. There is the end which is usually oval or triangular in section and is 2 to 3 inches in length and then the ridged part which in section is like a triangle with two concave sides that extend to the point of the splice (see fig. 2).

In designing any good bow there are two important features that must be avoided. One is that there must be no bend in the grip when the bow is drawn. If there is, the bow kicks in the hand when it is shot,

THE ARCHERS OF ISLAM 73

TIP $JEA SYAH rUSTXRL GRI'P

a) ?+

b ) 95~ ?r

c) -?

fl~ WOD S/CW 9

Fig. ?: 2 )Prin b uoPrin c uks

it is unpleasant to use and it is almost certain to be inaccurate. The second is related to the pull on the string, which the archer calls the weight of the bow. This, naturally, increases as the string is drawn further and further back, but if the weight increases rapidly towards the end of the draw it will again result in inaccurate shooting as a fractional variation in draw length will have a more marked effect on the range attained by the arrow than if the weight increases slowly at the end. In this way a rapid build-up in the weight at the end of the draw magnifies any error made by the archer, while a slow build-up tends to eliminate it (see fig. 3).

With the Oriental composite bow the design of the handle eliminates bending in the centre when the bow is drawn and the rigid siyahs, acting as levers during the latter half of the draw, effectively reduce the rate of increase of the weight at this stage. These factors will be con- sidered in further detail below.

The next stage in the construction of the bow was to attach two

74 W. F. PATERSON

strips of horn to the belly, which is the side facing the archer as he shoots. That of long-horned cattle was most commonly used and the strips taper from the grip to the end of the splice where the sjyah is fitted to the dustdr. At the centre of the limb they vary in thickness from about I/8th to 1/4 inch. Both the inside of the horn and the wood core of the dustdr, which is not usually flat, but slightly curved from side to side, were scored along their length to increase the surface area exposed to the glue and thus give better adhesion. The horn was lightly bound to the core to avoid exuding too much of the glue and then left for, perhaps, two months to allow the glue to set. At this stage the bow was a flat 'U' in shape and it was held in this form by a cord tied between the tips of the two siyahs. The curve of the bow was reflex, that is to say the opposite way to which it would be bent for

stringing, and the cord that controlled the amount of bend was firmly seated in the temporary nocks cut in the 'wrong' side of the tips.

With existing Persian bows the horn on each limb is not a single piece but a number of strips glued together and held in position with a binding of fine sinew 1). This practice seems to have been peculiar to the Persians and arose from the design of the limbs which were usually much wider than those of any of the other Middle Eastern bows. Because of this feature a single piece of horn could not be obtained that was flat enough and wide enough to fit the belly. It is of interest to note that a similar width is favoured in modern bow design as this helps to avoid any lateral twisting in the limb as the bow is drawn.

This glueing process took place in the winter, when the cooler and more humid conditions slowed the setting .rate of the glue 2). A slow setting rate gives superior adhesion.

When the glue had set, the next stage was to apply sinew to the back of the bow, which is the side away from the archer as he shoots. Leg tendon of cattle or deer answers well. Payne-Gallwey 3) and F. E.

I) This may be seen in a damaged bow, Ingo Simon collection, Manchester Museum, No. o.7649.

2) Cf. C. Cahen, Bulletin d'Itudes Orientales, XII, p. IZ9. 3) A Treatise on Turkish and Other Oriental Bows, London 1907, p. 4.

THE ARCHERS OF ISLAM 7J

Brown 1) both speak of neck sinew being used, but from practical trials this has proved useless. First, it will not beat out into fine fibres and secondly it is too elastic to provide the desired tension when the bow is drawn. It is sometimes found, cut into thin strips, as additional packing in the region of the siyah, but this would appear to be its only real use in the construction of the bow. After the leg sinew has been cleaned and dried it can be beaten out into fine fibres, and it is in this form that it is impregnated with glue and fastened to the back of the bow.

Glue-soaked sinew cannot be effectively applied in cold weather as it is then difficult to handle and the result will be a poor bow. It is by far the most important component in the composite bow, and even if inferior wood and horn are used such defects in materials can, to a

large extent, be eliminated by the correct application of sinew. This process was normally carried out on a warm spring day 2) and the sinew could be laid on in one, two or even three layers. Two layers were used in making the flight bows of the Ottoman Turks 3) and in this case the reflex in the bow was slightly increased when the first layer was applied. It is not known if the same practice was followed by craftsmen from other regions.

When the final layer was applied the bow was reflexed into a com- plete oval and a cord was led around the temporary nocks cut into the extended siyahs and down to the grip to hold it in this position until the

glue had set (see fig. i, bottom left-hand drawing). The object of this method of construction was to get the maximum tension into the sinew when the bow was later strung and drawn, as high efficiency can only be achieved if the various components are adequately stressed when the bow is shot.

However carefully the two horn strips are butted against each other when applied to the wood core, a gap almost invariably appears between

I) "A Recently Discovered Compound Bow", Seminarium Kondakovianum, Vol. 9, '937, pp. I-1o.

2) Add. 23489, fo. zzb. 3) J. Hein, "Bogenhandwerk und Bogensport bei den Osmanen", Der Islam, 1925,

PP. 347-350.

76 W. F. PATERSON

them when the bow is fully reflexed. If this gap is not plugged unfair shearing strains will be thrown on to the bond between the wood core and the horn when the bow is shot, the slight reflex in the handle and the rigidity of the grip will be lost. This is the reason for the ibranjaq (Turkish: chelik) found at the centre of the grip and fitted between the two horn strips. All existing bows that have been examined have this fitting made from bone or ivory, though in earlier days a suitable hardwood seems to have been employed 1). Elmer suggested that it was a form of shock absorber 2), but he lacked practical experience in their construction.

When the glue had set after about another two months or so, the bow had to be adjusted so that both limbs curved equally and correctly when it was drawn. The excess was cut off the siyahs and the nocks for the bowstring cut into the back. With Turkish and Persian bows a saw cut was made 2 to 3 inches in length and a slim wedge of horn inserted to reinforce and strengthen the tips of the bow. This feature is absent in bows from India and no mention of it has been found in any of the Arabic manuscripts that have been examined, though this does not preclude the fitting as none of the works go into the finer details of construction. On the side of each nock toward the grip, after the tips of the siyahs had been fully shaped, a sinew binding was applied to avoid any risk of the wood splitting under the impact of the bowstring when the bow was shot. Where the horn insert was used it also held

this firmly in place. To string a fully reflexed bow is not an easy task. After removing

any surplus or exuded glue as well as any obvious irregularities with a file, the bow would have been thoroughly warmed to make it more supple. The craftsman, probably with an assistant, would then have carefully flexed both limbs together until a string could be slipped into the nocks. The shape of the two limbs could then be studied and, after removing the string, any section of the limb not curving enough would be filed away until the right shape was achieved. This process, known

I) N. A. Faris and R. P. Elmer, Arab Archery, Princeton, 1945, P. x 5.

z) N. A. Faris and R. P. Elmer, op. cit., p. I62.

THE ARCHERS OF ISLAM 77

as tillering, calls for much care and patience if a first class weapon is to be produced. Having achieved the right shape when the bow was strung, the bow then needed to be partly drawn and studied again to see that the two limbs were still correctly balanced and, if they were not, the file once more had to be applied. This process continued until the bow could be brought to full draw with both limbs achieving as perfect a curve as lay within the skill of the maker to produce. If the two limbs do not recoil exactly together when a bow is shot, inaccuracy is the only result and good shooting characteristics can only be achieved by the slow and careful process described.

Finally, the sinew was covered with thin diagonal strips of bark or thin leather and varnished to make it waterproof. The reason for the

diagonal strips of bark is that this material has virtually no stretch. If they were placed along the length of the limb they would break as soon as the bow was drawn. The horn might also be covered as is invariably the case with existing Persian and Indian bows. Mamluk, Ottoman flight, Mongolian and Sino-Tatar bows, among others, left the horn exposed. The covering was then decorated and many of the designs that can still be seen show a high standard of skill and artistry.

Sir John Chardin, writing of his third visit to Persia, from which he returned in 1677, says: "The Persian Bows are the most valued of all the East: The Matter whereof they are made is Wood and Horn laid over one another, and covered with Sinews, and over that the skin of a tree very sleek and smooth; they paint them afterwards, and Varnish them so admirably well, that one may see one's self in those Bows, and the colour of them is as bright as possible" 1).

3. The Mechanics of The Bow

It is clear that making a bow in this manner was a slow and expensive process, which can, perhaps, be justified by a study of the advantages that it offered.

A bow is a simple machine for storing energy by the action of the

I) Sir John Chardin's Travels in Persia, The Argonaut Press, London I927, p. 270.

78 W. F. PATERSON

archer in drawing the string. When he looses the string it is transferred in the form of kinetic energy to the arrow. No bow can be ioo% efficient as work must be done by the bow in carrying its own limbs forward with the string. This amount is lost as it cannot be transferred to the arrow. It follows that a bow with heavy limbs and, in particular, heavy tips wastes more energy than a bow with light limbs and tips. When the string is loosed not only has the bow to push the arrow forward but it must also carry itself forward and the concept of apparently propelling a weight greater than that of the arrow has been dealt with by P. E. Klopsteg in his Theory of Virtual Mass 1). For convenience, this problem may be considered in two parts. First, how much energy can a bow store and secondly, how much of the stored energy does it transfer to the arrow.

The stored energy depends on a number of factors. There is the physical strength of the archer and thus the weight of bow that he can handle, there is the length of his draw which is determined mainly by the length of his arms and the width of his chest and lastly the cha- racteristics of the bow which determines how the weight builds up as the bow is drawn.

These characteristics of bows may be compared by plotting them on a force-draw curve (fig. 3). This is obtained by measuring the force needed to draw back the string at intervals from rest to full draw and the area under the curve is a measure of the energy put into the bow. The curves are those of an English longbow and a Mongolian com- posite.

Even a cursory examination shows how much more energy the composite bow stores than the more simple wooden bow. With the composite the weight builds up quickly during the first half of the draw and, if it was not for the rigid siyahs, it would continue to build up at much the same rate and thus prove too strong for the archer to handle. During the second half of the draw the lever action of the

siyabs comes into play with the result that the increase in weight is

I) American Journal of Physics, Vol. ii, No. 4, August 1943, pp. I80-I. Also Smithsonian 4565, Washington 1963, pp. 588-590.

THE ARCHERS OF ISLAM 79

5~0

Uf) A-

:3 30 0

~ 20 p.

I..

•DRAW LENGTH IN IN CHES

Fig. 3

relatively small. When shooting this bow it feels, as the shoulder and back muscles come into exerting their power, that the pull on the bowstring reduces, though this is not, in fact, the case.

Sir John Chardin records: "The Goodness of a Bow, as the Persians say, consists in this, viz. That a Bow be hard to bend till the Arrow be laid half over it, and then that it be soft and easy, .. ." .1).

Compared with this the longbow builds up more slowly at the beginning of the draw and more quickly towards the end. In conse- quence the archer feels that considerably more effort is needed to draw such a bow than in the case of the composite. The superior characteristics of the latter can thus be seen as, for a given maximum effort put into drawing the bow, the composite stores more energy.

i) Sir John Chardin, op. cit., p. 270.

80 W. F. PATERSON

The next question is how much of this energy the bow transfers to the arrow. Here, practical performance and efficiency are two very different quantities. From purely mechanical considerations the highest efficiency is achieved with a very heavy arrow that will fully absorb all the thrust from the string. Such an arrow would not travel very far, though its impact at short range would be considerable and enable it to penetrate any armour or mail, if the arrowhead was of the correct type. Against this a light arrow is inefficient and, though it will be discharged at a higher velocity, it cannot fully absorb the thrust of the string. There is an exception to this in the case of the flight arrows, but they are in a special category.

An analogy may be taken by considering three balls, about the size of a tennis ball, one made of lead, one made of rubber and one made of cotton wool. If thrown by hand it could be well that the heavy ball and the light one would go about the same distance, while the rubber one would go very much further. As regards the force with which they might hit an object the lead one, if within range, would obviously give the hardest impact, while that of cotton wool would have least effect. In this way a compromise must be achieved to give the best performance to the arrow when shot from a given bow and its weight must be related within certain practical limits to the weight or power of the bow 1).

If a short stick is taken, one end held and the other bent and allowed to fly back, the speed of recovery when the end is released will be

higher than in the case of a longer stick. In the same way a short bow can impart a higher velocity to an arrow than can a longer bow, assuming that their construction is similar. Thus, for velocity one needs a short bow with a lighter arrow, but for hitting power one needs a longer bow with a heavier arrow. It should also be noted that the short bow tends to be inaccurate as it is more difficult to get the two limbs to move forward perfectly together due to its higher rate of recovery.

i) W. F. Paterson, Bowman's Handbook, Portsmouth I1955, pp. 15-17. See also J. D. Latham and W. F. Paterson, "An Analysis of Arrow-Weights in an

Islamic Military Manual", Journal of Semitic Studies, vol. X, No. 2, I965.

JESHO IX PLATE I

Top: The Mongolian bow used to produce the curve in fig. 3. Lower: A Sino-Tatar bow, similar to that commented upon by Ingo Simon. Just

in front of the grip, though difficult to see in this reproduction from the original in colour, are three thumb rings, two of jade and one of ser- pentine.

JESHO IX PLATE II

Left to right: i. A Sind bow. Note the characteristic double angle in the siyabs and the

marked reflex in the grip. ii. A Turkish bow of unknown origin. Its length suggests it is a target or

hunting bow. The back Is covered with bark strips and the horn of the belly Is left exposed. The grip lacks the hump on the back and is covered with leather.

iii. An Indo-Persian bow. The design of the siyahs shows influence of the Sind bow.

iv-vi. Ottoman Turk flight bows.

JESHO IX PLATE III

Top and centre: Typical Indo-Persian bows. They are good examples of the type of bow carried by the mounted archers.

Bottom: A 'crab' bow from Central India. Note the amount of recurve in the siyahs.

JESHO IX PLATE IV

A fine example of the Ottoman Turk flight bow. The inscriptions on the siyahs show that it was made by Ibratum in 1 69 A.H. The delicate little flight arrows

are tipped with ivory.

THE ARCHERS OF ISLAM 81

Length in a bow tends to give greater accuracy and steadiness in shooting. It also gives a better thrust to a heavier arrow and for these reasons the longer bow is more suitable for warfare.

The Mongolian bow, already mentioned, though accurate, is not a good bow by composite standards, as it is too long. From nock to nock, measured along the side of the bow, it is 63 inches compared with about o50 inches recommended by Taybughi al-Ashrafi 1). Ottoman

flight bows, on the other hand, vary between about 40 to 42 inches. One result of the greater length is that the horn and sinew are in- sufficiently stressed when the bow is drawn and the speed of recovery of the limbs is relatively slow when the string is loosed.

The accuracy of these bows is fully supported by the late Ingo Simon, a noted archer and one of the greatest authorities on Oriental bows. He wrote: "In these latter years I have taken to using mainly one big Chinese Tatar bow. I find that at short range it is unsurpassed. The arrows are too heavy for anything save short range work, but they rarely get broken and are easy to find [afterwards in the ground] on account of their size. Recently I hit a matchbox at zo yards with four successive shots" 2) (see also Plate I).

From these considerations the equipment used by an archer is

usually a compromise, depending on the purpose for which it is intended. For warfare he should have a bow that is not too short, so that it will be reasonably accurate and reliable. He needs to use a heavy arrow if he expects to penetrate armour or mail. However, for use on horseback a shorter bow is easier to handle, but this would require a lighter arrow. With such conflicting requirements it is easy to under- stand how designs have differed, depending on what a given tribe or race considered were the most important factors, coupled with the skill of their craftsmen and the materials that they had available.

This last factor can be of the greatest importance. The question is sometimes asked, why was the composite bow not used in the West, if it was such an efficient weapon? The simple answer to this is that

I) Add. 23489, ff.22b-23a. 2) The British Archer, (magazine) Vol. 8, No. 6, I9r7, p. 231.

JESHO, IX 6

82 W. F. PATERSON

there were no long-horned cattle or goats in Europe from which suitable horn could be obtained. The composite bow was used by the Venetians in their wars with the Ottoman Empire, though even if their own craftsmen made the bows I feel sure that the horn must have been obtained from the Middle East 1).

The most efficient bows ever devised were the flight bows of the Ottoman Turks. They were so delicately made and the materials stressed so near breaking point at full draw that they needed to be warmed, or even heated, before being strung to increase their supple- ness. Such a bow would be useless in war as an archer cannot stop and warm his bow, if it happens to be unstrung, before going into battle. Also, a highly stressed bow can easily be broken and the archer, in war, needed a sturdy and reliable weapon.

Such then are the main technical aspects to be taken into account in determining the design of bows. Climatic considerations also enter into the problem as modifications are needed, particularly under extreme conditions, regarding the optimum proportions of sinew, wood and horn in the bow.

Fig. 2 shows three typical examples. Bows with a thin wood core, such as those found in India are more liable to distortion than those which have a thicker core. Such defects are usually easy to correct by heat treatment, but it is obviously better to have a bow that is less likely to be found unfit for use when it is needed. The Turkish design is superior to the other two.

4. Accuracy and Penetration

The best equipment is of little use unless the archer has achieved a high standard of ability, neither can the archer normally do more than what he has trained to do. For this reason the various exercises need to be studied to assess what the archers might have been expected to achieve in battle.

The basic handling of the bow must be mastered on foot before

I) Cf. G. Giorgetti, L'Arco, la Balestra e le Macchine Belliche, Milan, 1964, pp. 20-23.

THE ARCHERS OF ISLAM 83

attempting to use it on horseback. Target ranges appear to have reached up to about 280 yards on the long field (al-middn al-tawil), whose range was between 140 and 130 outstretched armspans 1). This

compares with the range of 12 score paces, or 240 yards, used by the old English longbowmen and this is near the limit of the war bow employing a full length arrow with a steel head.

It must be emphasised that doubling the weight of a bow does not double the range to which it will shoot an arrow. If, for instance, a bow with a weight of 45 lbs. at full draw will shoot a given arrow 240 yards, a 65 lbs. bow will only shoot the same arrow about 280 yards. The real

advantage of the heavier bow is that it enables a heavier arrow to be used that will hit harder and achieve a higher penetration.

At such ranges accuracy is relatively low and a good shot under calm weather conditions might be expected to hit a target the size of a man on horseback about once in every four shots. However, the application of such long range shooting by a group of archers against a formed body of the enemy needs no amplification.

An arrow will loose much of its velocity in travelling such a distance and should be effectively stopped by good quality armour or mail. During the Crusaders' coast march from Acre in 191, which cul- minated in the Battle of Arsiif, Beha ed-Din records that the Saracen archers could do little against the armour of the Franks:

"their infantry drawn up in front of the horsemen stood firm as a wall, and every foot-soldier wore a thick gambeson and a hawberk, so dense and strong, that our arrows took no effect... I saw soldiers with from one to ten arrows sticking in them, still marching on" 2).

Such evidence, no doubt, led R. C. Small to say in Crusading Warfare, p. 81: 'The bow and arrow appear to have been light weapons. The arrows penetrated Frankish armour, but often without wounding the body of the wearer'.

What is more likely is that the Crusaders' archers forced the Moslem

cavalry to keep its distance at this stage of the engagement and shoot

I) Add. 23489, fo. I27a. 2) S. Lane-Poole, Saladin, London, 1898, p. 30o.

84 W. F. PATERSON

from long range. Coupled with this it may have been that the arrows were too light for such a purpose or that too much use was being made of the light darts shot with the aid of the arrow-guide (majrd).

Later, when the Saracens began to press home their attacks the Itinerarium records:

"That day our own losses and the sufferings of our horses, who were pierced through and through with arrows and darts...". And again: "With deadliest effect they kept launching forth their darts and arrows" 1). Such statements suggest that with the reduced range the arrows became effective due to the higher velocity on impact and supports the idea that the arrows were too light rather than the bows being too weak.

In assessing the ability of an archer to shoot with adequate and consistant accuracy, Taybughi required him to shoot at a target set up at a range of 6o bows *). This would have referred to the strung bow and that specified by him would have measured 45 inches, thus giving a range of 75 yards. At this range he was expected to keep all his arrows within a circle 3 feet in diameter, which is a good and practical standard of accuracy, which an average man can achieve after a few years of

regular practice. From this evidence one may presume that the trained archer, on foot, should hit a man every time at about 60 yards.

When mounted and at full gallop the problem is very different and the ranges at which accurate shooting is possible is very much shorter. In qiqaj the target was basically a mound of earth, or something similar, on the ground at which the archer shot as he went by at full gallop and one might guess that the range was about io yards 3). The other form of shooting was qabaq where a gourd was hoisted to the top of a mast and served as a target at which the archer shot as he rode beneath it 4). This was one of the recognised furusijya exercises and obviously demanded a very high standard of horsemanship as well as skill in archery 5).

i) S. Lane-Poole, op. cit., p. 315 and p. 320. 2) Add. 23489, fo. iooa. 3) Add. 23489, ff.61b-74a. 4) Ibid. 5) D. Ayalon, "Notes on the Furdsiyya Exercises...," Scripta Hierosolymitana, IX,

Jerusalem 196I, pp. 55-56. Also, M. Reinaud, Journal Asiatique, Septembre 1848, pp. 220-221.

THE ARCHERS OF ISLAM 85J

Here, it should be mentioned that it is unlikely that an archer could shoot as strong a bow on horseback as he could on foot. Though the main power comes from the shoulders and the back muscles, with the arms playing a relatively minor part, when shooting a powerful bow an archer normally needs a firm footing and, if the ground is slippery, he may have difficulty in controlling his bow. Some loss of such support would naturally result from shooting in the saddle so that the mounted archer would probably have used a slightly lighter bow on horseback than he could have used on foot.

There is no doubt that the adepts at these exercises achieved a remarkable degree of skill through hard and regular practice and there are some interesting accounts in the pages of Hansard 1). That author, however, was not very good at giving references to his sources of information so it has not yet proved possible to check all his stories. Such tales usually concentrate on the rare feats, which may have been somewhat enhanced by the original writers and the legends that surround a great archer, be he Odysseus or Bahrim Giir, Rustem or Robin Hood, do not enable one to judge what might have been within the ability of the average man.

Accurate knowledge on the power, or weight, of Islamic bows is

lacking though one might draw certain conjectural conclusions from Arab Archery (pp. 14-II5), where Thhir al-Balkhi is credited with

relating the weight of arrows needed with bows of various weights. The heaviest bow that he mentions is one hundred ratls and it could be assumed that this was the most that an average archer could have been expected to handle.

Tihir was one of the so-called Great Masters, but though he is mentioned in most works relating to archery, no biographical details have come to our notice in spite of a careful search. He may have taught under the Samanids or the Ghaznavids and from such scanty information Dr. W. Hinz of Gottingen has kindly suggested that the Baghdad ratl of 406.25 grams might have been the unit that he used.

I) G. A. Hansard, The Book of Archery, London 1840, pp. 34-36, I20-125, 137-142.

86 W. F. PATERSON

This would give his bow of one hundred ratls a weight of 88 lbs. which a man slightly above average strength could master with practice. With this as a basis and from practical experience in shooting with the stronger bows it can be suggested that the majority would have used bows from 65-70 lbs weight. Such a bow should drive a war arrow through any armour or mail up to a range of about Ioo yards given, in the case of armour, a reasonably square hit on the surface of the plate. The surprisingly high penetration of arrows has been shown on many occasions 1).

There remains the problem of the actual forms of the bows used by the Islamic archers. The majority of miniature paintings show two basic types. There are those with the relatively straight tips and, from a side view, such a shape is given by a Persian bow as shown in fig. I. The other type has a strong recurve in the tip which suggests the Turkish design such as that shown in Plate II, ii 2).

The length of such bows seems to have varied quite widely. In the Ingo Simon Collection there is a Bashkiri bow from the Siege of Vienna (Manchester Museum No. 0.7633) and this measures

58 inches from

nock to nock along the side. This would appear to be about the longest of the Islamic war bows. Those from Persia are about 54 inches,

Taybugha specifies one of about 5o

inches, while Indo-Persian bows vary between 44 to 48 inches. These are the shortest if the Ottoman

flight bows--40 to 42 inches-are excluded on the grounds that they were not weapons of war.

Detailed knowledge of many bows is lacking. What of the bows from the Kirghiz, Turkmen and Uzbek areas? Do any of the Mamliik bows still exist? Can the few remaining craftsmen believed to exist in the Mongolian area be contacted before the finer details of their art is lost? Letters have remained unanswered, so may a humble archer make a final plea to those who may have the fortune to come across such

I) See C. J. Longman, Archery (Badmington Series), London 1894; pp. 425-435; Saxton Pope, Hunting with the Bow and Arrow, New York 1923, pp. 46-49.

2) See also, W. F. Paterson, "Persian Archery", Journal of the Society of Archer- Antiquaries, vol.

5, 1962, pp. 23-24, and vol. 6, 1963, pp. 20zo-z22

THE ARCHERS OF ISLAM 87

items? Any details would be most gratefully received and preserved in the archives of the Society of Archer-Antiquaries.

I must gratefully acknowledge the work of Dr. J. D. Latham of Manchester University on an edited and annotated translation of the Kitab ghunyat al-tulldb fi ma'rifat al-ramny bi 'l-nushshib by Taybughi'l- Baklamishi al-Yiin~ni, to be published by the Holland Press, London, and for his kind agreement to use the few brief extracts that appear here. Reference should also be made to the section on archery from an older general treatise on arms to be published shortly by Mr. Antoine Boudot- Lamotte (University of Bordeaux).