THE WEAPONS OF SIXTEENTH
CENTURY WARFARE AT SEA
In the preceding chapters we have outlined the overall parameters of the Mediterranean system of armed conflict at sea in considerable detail. But with the partial exception of the galley, we have said little in concrete, quantitative, terms about the technology involved. We have outlined the strategic framework of Mediterranean warfare at sea in fairly precise terms and have undertaken a general overview of tactics, particularly galley tactics, on the assumption that the strategic characteristics of any system of warfare stem from its tactical characteristics. We shall examine the tactical characteristics of the Mediterranean system of warfare at sea in detail in the next chapter; but these characteristics stemmed, in turn, from the characteristics of the weapons involved and from the abilities of their users.
This last point deserves closer examination. In an age of exotic weaponry and mass armies the second half of the equation user effectiveness has frequently been overshadowed and attention focused almost exclusively on the technological characteristics of the weapons themselves, an emphasis which is inappropriate to our subject. In comparing armies or navies with one another both historians and military analysts have all too often implicitly assumed an equality of user competence and effectiveness which did not, in fact, exist. In part this is a problem of science versus art. While highly sophisticated techniques have been developed in recent years for making a comparative evaluation of the technological characteristics of weapons systems speed, range, vulnerability, destructive impact and, finally, cost the art of comparing the soldiers and seamen of one nation with those of another remains little more advanced than it was in the days of Alexander the Great.
To a degree the modern analysts assumption of an equality of user competence is justifiable. Current weaponry is highly complex, changes frequently and is increasingly designed to be foolproof and easy to use so that men can be trained to use it in relatively short periods of time. When weapons are so designed, there is unlikely to be much difference between what one user and another can get out of them regardless of individual skill. But this emphasis on training is a relatively recent phenomenon.
Until the sixteenth century warfare was dominated by men who were not trained in any meaningful sense of the word, but who learned their military skills in childhood and adolescence as a part of their total cultural environment. The advent of training the imparting of unfamiliar military skills to men who were themselves generally unaccustomed to warfare depended upon the development of relatively simple weapons which could compete effectively with traditional methods of warfare. Gunpowder made this possible on a large scale for the first time.
But gunpowder weapons, particularly individual firearms, were first used successfully on a large scale within what looks very much like the old pattern of military skills acquired through immersion in the total cultural environment.1 At the same time, a successful effort was made in some areas to train fighting men in the use of the more effective of the traditional weapons. The results were strikingly and misleadingly similar. A Spanish musketeer and a sipahi of the Porte of the mid-sixteenth century, a small arms specialist and a trained horse archer, must have had very similar attitudes toward their weapons and their work. This is hardly surprising. New weapons are at first almost invariably used within the conceptual framework established by their predecessors.2 Ultimately, however, the intrinsic characteristics of gunpowder weapons prevailed. A Sicilian naval arquebusier who had learned his military skills in the old way was undoubtedly more effective on campaign and in battle than a trained infantryman of the tercios temporarily embarked on a galley. This point was dealt with in our discussion of the effect on Spanish naval power of the losses incurred at Djerba. But infantrymen of the tercios were, for a variety of reasons, a good deal easier to come by than seaman-arquebusiers of the old type.
The above example represents only the beginning of a process which culminated in the recruiting policies of eighteenth-century Europe where armies came to be composed of guttersnipes and the derelicts of society, swept up by press gangs and forcibly trained to the kings service. This, of course, lies well outside the bounds of our study. Two points, however, must be made: the introduction of effective gunpowder weapons made possible a kind of relationship between weapons, their users, and the culture from which the users came which was, in most parts of the world, entirely new. The tension between this new kind of relationship and that which had preceded it lies at the heart of our study. In this context, the galley propelled in battle by muscle power, but capable of making effective use of gunpowder weapons was an important focal point in the struggle between new and old. Secondly, the relationship between weapon and user was a much closer one in the sixteenth-century Mediterranean than our experience since the industrial revolution would lead us to expect. The apparent modernity of certain of the weapons involved should not mislead us. The weapon which a man carried or served was far more likely to be a reflection of his total cultural environment than merely the product of a bureaucratic decision.
The weapons of sixteenth-century Mediterranean warfare at sea can be conveniently, if arbitrarily, divided into three categories: edged weapons, individual missile weapons, and artillery. A few highly specialized, rare, and obsolete weapons mines, battering rams and catapults are not covered by our categories and the boundary between individual firearms and artillery was indefinite. Individual armor, still undeniably important in warfare, was not really a weapon at all, though it can be logically dealt with in our discussion of edged weapons. Nevertheless, this categorization offers the advantage of accurately reflecting tactical reality.
We shall therefore discuss each of these categories in turn, dealing with the capabilities and limitations of the weapons in each of them in more or less quantitative terms. We will then discuss their tactical employment by the various Mediterranean nations, focusing on galley warfare. In light of the pivotal importance of artillery, considerable attention will be paid to sixteenth-century gunnery and the tactical employment of cannon. We will not, however, center our attention on purely technical matters.
In several key areas important differences existed between nations in the design, construction and use of weapons of the same class. In almost every case these differences, and the differences in tactical and strategic philosophy which they underline, were the result of social, cultural and economic factors rather than purely technological ones. The importance of understanding the interplay between these factors will become apparent in our discussion of galley tactics, the War of Cyprus and the Lepanto campaign. Finally, we will point out the importance of these same factors in governing the adoption or non-adoption of new forms of weaponry. In so doing we will challenge the usual assumption that weapons development proceeds in a straightforward fashion with technological progress steadily providing new and better weapons and rendering older, inferior ones obsolete. In fact, as we will show, the older systems were frequently superior, on a narrowly tactical basis, to those which replaced them.
As the sixteenth century began, edged weapons played an important role, in some areas a dominant role, in the art of war. The battlefields of western Europe in particular were largely ruled by armored men wielding edged weapons: sword, lance, pike, and pole arms of various kinds. By its end, even though direct shock action3 was to remain the ultimate arbiter of the battlefield for a long time to come, individual firearms had clearly wrested cold steel from its position of dominance in Europe just as they had begun to threaten the horse archer and his composite recurved bow elsewhere.
Developments in the Mediterranean world generally paralleled those in Europe. Though the importance of missile weapons in general and artillery in particular increased during the sixteenth century, boarding remained, under ordinary circumstances, the only decisive maneuver of galley warfare and hence of Mediterranean warfare at sea. Missile weapons played an increasingly crucial role in preparing the way for boarding; but boarding itself remained essentially a shock action fought out with edged weapons. Even in siege operations, the other side of the amphibious coin which we shall examine in our treatment of the Ottoman attack of 1565 on Malta, fortified places were still generally taken by storm, not simply surrendered after bombardment had made them untenable.
In land warfare considerable variation existed between the Mediterranean nations with regard to types of edged weapons, the tactical formations within which they were used, and the amount and weight of armor worn. A number of factors acted to level out these differences in warfare at sea. The close confines of a shipboard fight precluded the use of many of the most formidable weapons of land warfare, notably the pike. Variations in tactical formations were virtually eliminated by space restrictions aboard galleys. The greater need for agility at sea posed by the necessity of leaping from ship to ship and clambering over railings somewhat curtailed the use of armor. Turks, North Africans, Spaniards and Venetians alike used halberds, half pikes and other more exotic pole arms of moderate length.4 All used one-handed swords designed according to their national preferences. The full-sized galleys of all of these nations included in their complements at least a small core of armored fighting men. All of this remained essentially constant throughout the sixteenth century.
There were, however, differences. The Spanish and their German mercenaries used long, two-handed swords upon occasion; one of these was wielded by Don Juan of Austria at Lepanto.5 The Spanish and their allies seem generally to have worn more and heavier armor than the Venetians or Turks, who frequently carried small shields by way of compensation.6 Though of good construction, Turkish armor was generally lighter than that of the Christian European nations, as were Muslim edged weapons in general.
In essence, we are primarily concerned here with the size and weight of the specialized fighting complements carried by the galleys of the various Mediterranean nations. Without going into too much detail, these differences can be roughly summed up as follows: the galleys of the Knights of St John of Malta (we know little about the manning of their galleys before expulsion from Rhodes) carried a higher proportion of armored fighting men, men who were also as a rule individually more heavily armed and armored than their adversaries, than any others. Next in the shock power of their fighting complements came the Spanish galleys and those of the Papal States. Genoese galleys in the Spanish service, at least during the second half of the sixteenth century, seem to have carried almost as many armored fighting men as their Spanish opposites, at least when Spanish infantry was embarked; but the picture is unclear. The fighting complements of Venetian galleys included a thin crust of armored Venetian nobles, men who were as heavily armed and armored and who fought every bit as well as their Spanish and Maltese opposites. However, most of the specialized fighting men aboard Venetian galleys, the huomini di spada, or scapoli, were lightly armed Dalmatians and Greek or Albanian mercenaries. Relatively few such specialized fighting men were carried by western standards. This was partially compensated for by the fact that Venetian ciurmi went armed and fought as free men. Though apparently larger, at least on major campaigns, the specialized fighting complements of Ottoman galleys, azabs, sipahis and Janissaries, were somewhat more lightly armed with respect to edged weapons and more lightly armored than their Venetian opponents. Clearly, few Muslim fighting men if any wore even partial suits of plate armor. Christian nobles, gentlemen adventurers and many ordinary fighting men habitually did so.7 Once again, this was compensated for in part by the presence of at least some armed free oarsmen on Ottoman galleys, particularly aboard flagships. Finally, the provision of fighting men aboard the galleys of the North African ghazi states seems to have roughly paralleled Ottoman practice, a somewhat artificial point in view of the North African preference for smaller raiding craft rowed largely by free men.
Though important, distinctions between the size and weight of these specialized boarding parties are largely irrelevant when viewed out of context. The lack of shock power which the complement of an Ottoman or Venetian galley could muster was balanced by an increased emphasis on missile weaponry and, as we shall discuss in the following chapter, tactical mobility.
When we turn to individual missile weapons, the relatively homogeneous picture which we observed in our treatment of edged weapons and armor disappears. The various Mediterranean nations had clear and sharply differing preferences with regard to individual missile weapons, preferences which were solidly rooted in cultural, social and economic differences. There is also an important chronological component. Where the use of edged weapons and armor in galley warfare showed relatively little change during the sixteenth century, important changes occurred in individual missile weaponry. The role of individual missile weapons in galley warfare remained essentially unchanged: to clear the way for a boarding attack and then to support it, to repel hostile boarders, to support skirmishing and siege operations ashore. But gunpowder weapons gained enormously in importance.
In purely descriptive terms the regional differences in weaponry are relatively easy to come to grips with. The Turks habitually used the composite recurved bow, supplementing it with firearms, particularly among the Janissaries. The Venetians also used the composite bow in addition to the crossbow, particularly in the earlier period, though not to the same extent as the Turks. They also turned increasingly to firearms as the century progressed. The others began with the crossbow and went over to the arquebus during the first half of the sixteenth century. The Spanish led the way in this development and went a step further than their competitors by developing the Spanish infantry musket, though never adopting it to the exclusion of the arquebus (see Appendix 1).
But this is oversimplified and leaves many very tactical problems un illuminated. The best way to undertake a comparative analysis of these weapons is to examine two areas: first, the process by which the arquebus replaced the crossbow and by which the arquebus was in turn partially replaced by the musket, and second, the relative advantages of these weapons vis-a-vis the composite recurved bow.
About the origins of the crossbow little is known. All that can be said with certainty is that it made its appearance in Europe toward the end of the Dark Ages. Its use quickly became generalized throughout Europe, but for various reasons it never became important elsewhere except in European hands.8 The crossbow began as a relatively crude weapon, a simple wooden bow set on a staff. It underwent a continuous process of development as the armor which it was called upon to penetrate steadily improved, mounting a composite bow of wood and horn by the eleventh century and a steel bow during the fourteenth. As early as the twelfth century its effectiveness prompted a series of injunctions by the Church banning its use against all but infidels.9 The reasons for this tell us much about the problem of cultural resistance to new and novel weaponry.
The crossbow and its successor, the arquebus, had one important characteristic in common: no great amount of skill was needed to become effective with either. Where a lifetime of constant practice was needed to master the use of the composite bow or for that matter the lance from horseback, a few hours sufficed to teach an ordinary fellow to draw and aim a crossbow. The injunctions against the crossbow cited above were rooted in a realization of this: the crossbow was the first personal weapon capable of killing an armored knight which could be obtained and used by almost anyone without benefit of great skill, practice or wealth. The social implications inherent in this were not lost on the ecclesiastical hierarchy.
The weapon which the Papal injunctions had sought to ban was the composite crossbow of wood and horn. This proved incapable of reliably penetrating the plate armor which became increasingly common during the fourteenth century and was replaced by the steel crossbow.10 In many respects this was a fearsome weapon. The development of an elaborate pulley and windlass system with a considerable mechanical advantage to draw the crossbow permitted the use of a short, very stiff, bow of mild steel into which a great deal of energy could be packed. The energy which a steel crossbow imparted to its projectile was greater than that given to an arrow by even the most powerful bow draw forces in excess of a thousand pounds were common. Because the bow was short, relatively little of the energy was expended in accelerating the tips of the bow and most of its considerable force was applied directly to the bolt.11
But there were two great disadvantages to this system: it was intrinsically inaccurate and the winding operation took a great deal of time and attention. The reasons for the crossbows inaccuracy are somewhat involved. They begin with the mechanics of the release mechanism (see Fig. 4). Where an archer, by precisely controlling his release, could ensure that the energy in his bowstring was smoothly transmitted to the arrow, the crossbow release mechanism released the cord abruptly and somewhat erratically. Instead of being smoothly accelerated in a carefully controlled direction, the crossbow bolt began its voyage lying loosely in its trough, and was then slapped into flight with enormous force. Crossbow bolts had to be made short and thick with a flat base in order to prevent the tremendous impact of the cord from reducing them to splinters.12 In view of the need for strength and the basic inaccuracy of the crossbow, war bolts were often very crudely made, having a single leather fin set into a slot sawed across the base of the bolt. The aerodynamic inefficiency of the resultant shape sharply increased drag and therefore reduced the maximum range. This was aggravated by the considerable and unpredictable vibration which the impact of the cord imparted to the bolt. By further and inconsistently increasing the aerodynamic drag of the bolt this vibration additionally reduced both range and accuracy.
From the point of view of the sixteenth-century fighting man aboard ship, the steel crossbow was a murderously effective short-range weapon. Armed with a good stout crossbow, he need fear neither man, devil nor armored knight if he could hit him. But to be sure of getting a solid hit, the crossbowman had to wait until his target was quite close: about 75 yards at most and even less if he were armored.13 This was well and good if the crossbowman were shooting from a secure place where he could reload at will. But in a boarding fight he had to make his first shot tell: it was the only one he was likely to get off. Drawing, cocking, and loading a crossbow was a lengthy process.
First the windlass had to be attached to the butt of the stock (see Fig. 4). Then the tackle (a four-pulley arrangement was most common) had to be relaxed and attached to the cord. Then the cord was drawn back with ten or twelve brisk turns of the crank (a foot stirrup was provided on most crossbows to hold it steady during this operation). This accomplished, the notched spool which held the cord was rotated into position behind it and the trigger cocked. The drawing tackle was then relaxed and removed and the bolt laid in its trough. Only then was the crossbowman ready to take aim and shoot. All of this took time: about one shot per minute was the maximum rate of fire possible.14 It was best not accomplished in the company of rough fellows even friends who were continually jostling you about. When all else failed the crossbow did not even make a very useful club.
The maximum range of the steel crossbow was a respectable 400 yards or so; but the maximum effective range was a good deal less, depending on the conditions and the target.15 The bolt lost most of its destructive energy to aerodynamic drag in the first 150 yards and was effective thereafter only against unprotected men or horses. The basic inaccuracy of the cross bow made shooting at long ranges a chancy business in any event.
The spread of the crossbows use was influenced by the sort of social factors that we have already hinted at. It had been popular with European mercenaries men who were trained to fight and civic militias from its inception as an effective weapon, and was adopted by sailors at an early date. It was the weapon par excellence of those who, by nature of their work, had to fight upon occasion, but who had neither the wealth nor the leisure time for the practice needed to become archers or men at arms. The crossbow reigned supreme among such groups until the first practical firearms began to appear about the end of the third quarter of the fifteenth century. These did not immediately render the crossbow obsolete, but slowly supplanted it as they became technically more efficient. The point to be made here is that the sort of social groups to which the crossbow was a useful and practical weapon were precisely those which turned to firearms most naturally.
The process by which the arquebus replaced the crossbow was an almost entirely technical one, with perhaps some retardation due to force of habit and custom. The cost of the two weapons must have been about the same since the value of materials needed and the amount of labor involved were closely comparable. When the matchlock arquebus appeared on the scene it fitted into a niche in the social-military system already established by the crossbow. Even more important, the crossbow had already accustomed the ruling elite of Western Europe to a certain tolerance for the existence of effective weaponry in the hands of the lower social orders. The initial shock among the lay and ecclesiastical hierarchy had slowly given way to a grudging acceptance of the crossbow. It had proved useful against the infidel: Richard the Lionheart of England had defeated the redoubtable Saladin at Arsouf in 1191 largely with its assistance.16 Nobles had found it entertaining to hunt with the crossbow (light ones which even their ladies could use were made for use against birds and small game),17 and had found it useful to keep a supply of them in their castles to turn the kitchen help into serviceable soldiers in the event of an unexpected siege. Thus the onset of effective firearms, though universally decried by those in high places, was not without precedent in Europe.
In the East, however, things were different. The crossbow, although known to the Arabs, had never been much used. The heavy steel crossbow was unsuitable for use on horseback and would in any case have been totally outclassed in range and rate of fire by the composite recurved bow. Here, on the great plains of Central Asia, the Ukraine, Anatolia and the Fertile Crescent, the traditional weapons of antiquity dominated. The critical importance of strategic and tactical mobility to cavalry warfare on open plains precluded the use of heavy armor, and the accuracy of the bow at long ranges could tell. The technical superiority of the composite recurved bow was never effectively challenged and the military supremacy of the traditionally trained man on horseback remained secure. So did the social and political dominance which that military superiority conferred. The violent opposition of Eastern cavalry elites to the use of the arquebus magnificently documented for the Mamluks by David Ayalon18 was directly connected to this situation. Only the Ottoman Turks showed the social flexibility needed to accept and foster the use of individual firearms in war.
The genesis of individual firearms was at first a slow process. Hand-cannon had made their appearance as early as the mid-fourteenth century, but were awkward to serve and were useful only at sieges and the like.19 It was not until the development of the matchlock mechanism, apparently conceived in Germany about the middle of the fifteenth century, that individual firearms came into serious competition with the crossbow.
Whereas the old hand-cannon had been fired by a gunner thrusting a hot coal or a length of smouldering slow match20 into the flashpan by hand, the matchlock accomplished this mechanically. The tip of the match was grasped in a pair of adjustable jaws atop a pivoted arm (see Fig. 5). The arm was connected to a trigger so that the match could be brought into contact with the priming powder in the flashpan using only the fingers of one hand.
The matchlock was not an ideal arrangement. The burning match had to be lit before use, which took time, and then kept lit (a costly business when we think in terms of whole armies) and continually adjusted as it burned down. Tactical disadvantages were numerous as well: the glowing match was sure to give you away in a night surprise and the matchlock was manifestly unsuited for use on horseback.21 But the matchlock was simple, cheap and foolproof. These qualities kept it in almost universal service until it was finally replaced by the flint lock late in the seventeenth century.
The fully developed matchlock was on a rough parity with the crossbow in Spanish armies by 1500.22 In 1518 it replaced the crossbow on Venetian war galleys by order of the Council of Ten.23 By 1530 crossbows were growing scarce on the weapons inventories of Spanish warships and are hardly ever encountered after 1540.24 All in all, it is safe to say that the matchlock arquebus was known throughout Europe by the 1520s, though in many areas it was not in general use. According to Venetian sources French galleys in 1552 still carried 40 crossbowmen and 60 arquebusiers.25
The superiority of the arquebus over the crossbow was based almost entirely upon technical considerations. No more accurate than the crossbow, it had at best a marginal superiority in short-range penetrative power. Its advantage in maximum effective range, if any, was small. Its rate of fire was, on the average, faster; but this was not the crucial factor. Most important was its greater ruggedness and simplicity. The arquebus had no elaborate winding tackle which could get tangled, broken or mislaid. The essential mechanism of the arquebus consisted of a simple tube of wrought iron plugged at one end and provided with a touch hole and a flash pan. There was little to get out of adjustment and nothing much to wear out. As far as weight and bulk were concerned the crossbow and arquebus were about on a par, though the arquebus was no doubt easier to carry and made a much better club in a pinch.
The greatest limitation on the power of the arquebus was the amount of recoil which a man could take. This increased sharply not long after the beginning of the sixteenth century when arquebusiers began butting their weapons against their shoulders instead of against their breastbones as had been the practice hitherto, a practice which seems to have originated in Spain. This permitted the use of larger and more powerful small arms and led directly to the development of the Spanish musket. (For the rather complex series of cause and effect relationships involved, see Appendix 1.) In its fully developed form the musket had a considerable advantage over the arquebus in killing power at long ranges. The Spanish musket could reliably penetrate armor at a hundred yards and kill an unprotected man or horse at five hundred.26
In part this advantage was gained by sheer size. The typical sixteenth-century arquebus, if there was such a thing, weighed ten pounds or less and had a bore diameter of about 60 caliber (sixty hundredths of an inch). Spanish muskets, however, weighed at least eighteen pounds and generally had bore diameters of from 70 to 85 caliber, with a few specimens larger than 90 caliber.27 This meant that in place of the arquebus ball of about a half ounce, the musket fired a full two-ounce lead ball. The recoil of these massive weapons, even though partly absorbed by their own great weight, must have been prodigious. Only large and powerful men could fire them, a limitation which prevented their universal use even in Spanish armies. Though it was no more accurate than the arquebus hitting an individual man at 75 or 80 yards would have been an exceptional feat28 the muskets long effective range and great stopping power made it an ideal individual missile weapon for siege warfare and shipboard use. The Spanish used the musket extensively and Ottoman shoulder arms, whose effectiveness at sieges was particularly noted, approached musket proportions.29
The last and most misunderstood of the individual missile weapons is the composite recurved bow. Though traditionally derided by European historians as antiquated and obsolete, the Turkish bow of the sixteenth century was capable of delivering a higher volume of accurate and effective fire at longer ranges than any competing weapon. It is plain, in fact, that the composite recurved bow in the hands of a strong and experienced archer a crucial qualification was, for most purposes, far and away the outstanding individual missile weapon of the sixteenth century.
The sixteenth-century Turkish war bow seems to have been of moderate size, perhaps because of its heritage as a cavalry bow. The bulk of evidence, including contemporary pictorial representations, suggests that the typical Turkish war bow was about 40 to 45 inches long and shot an arrow weighing about an ounce.30 Such a bow could have driven its arrow to an extreme range of about 500 yards and, given a square hit, through almost any armor at 100 yards.31 Because of its superior aerodynamic characteristics an arrow did not lose its velocity and destructive energy as rapidly as a crossbow bolt or an arquebus ball. Only the size and brute power of the musket gave it a slight advantage in maximum effective range.
A trained archer, when fresh, could get off six aimed shots in a minute with ease. Perhaps the best modern estimate of his long-range accuracy comes from W. F. Paterson, himself an experienced archer. Paterson concludes that a good shot on a calm day could be expected to hit a target the size of a man on horseback at 280 yards about once in every four shots.32 This represents an impressive standard of accuracy and is not far below what a reasonably experienced modern rifleman can achieve firing offhand at an unknown distance over iron sights. At a hundred yards such an archer would have been able to pick his man. At fifty he would have been sure death.
This is not to say that the composite bow, even in the narrowly tactical sense, was the ultimate weapon. Even in the hands of an expert it had its disadvantages. Draw forces of 150 pounds or so taxed even a very strong man and the initial rate of six shots per minute could not be sustained for long. The composite bow was more delicate than firearms. Humidity and dampness had a detrimental effect on the composite bows performance, a significant problem in war at sea, though one which does not seem to have been insurmountable or even particularly serious under ordinary circumstances.33 Finally, the bow presented difficulties in firing from cover and through loopholes. Once his weapon was loaded an arquebusier or musketeer could fire over, under, or through almost anything, a consider able advantage in siege operations and warfare at sea; but an archer had to stand erect, if only for an instant, to draw, aim and release.34
But the Achilles heel of the composite recurved bow was neither technical nor tactical. Archery did not consist of a series of mechanical movements which anyone could be trained to execute, but of a whole complex of cultural, social, and economic relationships. Though the Ottomans could, and did, train archers an immensely time-consuming business the real basis of the composite recurved bows existence was the traditional life style of the horse nomad of the steppes. The time of the steppe nomad was almost exclusively spent herding and guarding livestock, activities which kept him in the saddle but usually not otherwise occupied. He therefore had an infinity of time in which to perfect his traditional military skills. The tradesman, sailor, or farmer, by contrast, would have found it difficult to allot even a few hours a day to military exercise without suffering serious economic or physical consequences. Surrounded from birth by men who gained prestige and status from their ability to fight and shoot from the saddle, the young nomad engaged in constant archery practice from childhood. In view of the number of targets human and animal, edible and otherwise which the steppe presented, constant practice made economic sense. Constant practice was necessary, moreover, simply to master the basic rudiments of the art. Drawing a bow the weight of which approached or exceeded 150 pounds involved immense skill as well as immense strength. Reversing the bend of an unstrung recurved bow to string it was thought for many years to be impossible without the use of mechanical aids, yet the Turks did this as a matter of routine with even the most powerful bows.35 Where a few days and a good drill sergeant might suffice to train a reasonably good arquebusier, many years and a whole way of life were needed to produce a competent archer.
The same point applies with equal validity to the construction of bows. Individual firearms could be built quickly by methods which approached mass production. Bullets could be cast rapidly and in great numbers by unskilled workers. But the construction of a composite bow was a time-consuming ritual which took years (see Fig. 6). Some of the glues took more than a year to dry fully and had to be applied under precisely controlled conditions of temperature and humidity.36 The final shaping and dynamic balancing of a composite bow involved a level of skill which could only be learned by many years of constant work under the supervision of a master bowyer. Even the arrows had to be shaped, fletched, and balanced with almost as much precision as went into the bow itself. There were no short cuts and productivity per man hour must have been extremely low. All in all, the manufacture of composite bows was labor-intensive in the extreme and would have been highly vulnerable to any upward movement of the wage-price spiral.
When we compare personal firearms with the composite bow, most of the advantages of the latter were lessened by the peculiar conditions of war at sea. The crews of warships, unlike cavalry in the field, fought at least partially under cover until the moment of boarding. Protection was provided by the sides of the ship itself or by temporary bulwarks erected before combat. Venetian galleys, for example, were fitted with pavisades (large shields along the sides of the rowing benches) as were Spanish galleys early in the sixteenth century, probably with the menace of long-range archery in mind.37 On Spanish and Papal galleys, the raised fighting platform at the bow was covered on the sides and front before combat with a temporary rampart of wood, often reinforced with hides and excess cordage.38
At close ranges, a heavy musket ball fired at relatively high velocity would have been more effective than an arrow at smashing through the light planking of such a temporary rampart and doing damage on the other side. Though an arrow has impressive penetrative ability, this is largely the product of its superior shape. Whereas the arrow penetrates because of the concentration of its kinetic energy behind a narrow cutting edge, the musket ball smashes through by brute force and ignorance. In penetrating planking an arrow lost much of its destructive energy dragging its shaft and feathers through the hole. But at close ranges a musket ball could penetrate light planking handily, get through with enough energy to leave a gaping wound, and carry along with it a host of splinters and wood fragments which were capable of injuring eyes and other unprotected parts.
At ranges long enough for pavisades or the temporary bulwarks of a war galley to provide effective protection against musket or arquebus fire, the bow was no better off. Its supreme advantage, long-range accuracy, was lessened by the difficulty of firing from a rolling ship and the problem of accurately estimating ranges at sea.39 Its other primary advantage, rate of fire, would have been lessened in a protracted fight since archers tired more rapidly than musketeers. All the same, the composite Turkish bow in the hands of an expert was a formidable weapon. A handful of archers with a vantage point over or on the deck of a hostile ship with a clear space in which to draw their bows would have been capable of executing great slaughter with their rapid and accurate fire, even against armored men.
Unlike the arquebus and the crossbow, the composite bow and the musket were not equivalent weapons. The niches which they occupied in the military history of the time overlapped, but were not synonymous. Each had its strengths and weaknesses, even where expert users were involved. An indication that these were recognized and deliberately exploited is given by the report of a Spaniard who had served as a galley slave for the Paşa of Vélez in North Africa and escaped to tell his tale.40 He recorded that the Paşas galiots were manned with sharpshooters (tiradores), each one of whom carries an arquebus [escopeta] and a bow. Presumably the bows were used for close-in, rapid-fire work aboard enemy ships after boarding, for raiding ashore, and for sniping at extreme ranges, while the arquebusses were used in the opening stages of an attack on a Christian ship.
What was the cumulative effect of these regional differences in individual weaponry? Before the advent of gunpowder weapons the tactical edge had unquestionably gone to the Islamic nations among whom mastery of the composite recurved bow was a traditional skill. They clearly benefited from the considerable advantage in fire power which the composite recurved bow enjoyed over the crossbow. Because of the absence of suitable materials for its construction41 and because of the absence of social and economic conditions which would have fostered the practice of archery (except in England, a special case which has no direct bearing on our subject) the composite bow was never used in Western Europe except by the Venetians.
The Turkish composite bow had less short-range stopping power than the windlass-drawn European steel crossbow of the fifteenth and early sixteenth centuries, but this was the crossbows only technical advantage. The impact energy of a crossbow bolt dropped off rapidly with range, giving the composite bow firing an arrow of reasonable weight an increasingly marked advantage in killing power beyond approximately 150 yards. In addition, the composite recurved bow in skilled hands had an enormous superiority in accuracy and rate of fire. The crossbows main significance lies in the fact that it paved the way for individual firearms socially and intellectually.
The introduction of effective individual gunpowder weapons changed the tactical balance, but only gradually. The early sixteenth-century arquebus may have had a slight advantage over the composite bow in short-range stopping power; but it was markedly inferior in all other categories. The full-blown musket, however, had a number of advantages. Though its rate of fire was somewhat less than that of the arquebus and its accuracy not much better, its immense smashing power and long effective range served it well in naval engagements and sieges. While pavisades or a reinforced barrier of light planking provided reasonably good cover from arrow fire, they were of little value against a two-ounce lead ball impacting at close to 1,000 feet per second.42
It seems clear that gunpowder weapons had, by the middle of the sixteenth century, redressed at least the Spanish disadvantage in individual weapons vis-à-vis their Muslim opponents. This was particularly true in formal head-on naval engagements where both sides had time to set up their pavisades and barricades on the bows and sterns of their galleys, and in siege warfare where both sides fired from behind cover most of the time and where the duration of engagements and health conditions made fatigue an important factor. But these narrowly tactical considerations, though important, were not critical.
In the end it was the expanded scale of warfare as much as anything else which shifted the balance in favor of the West. The composite bows dependence on highly skilled manpower was its Achilles heel. While the Spanish could convert American gold and silver almost directly into arquebusiers and musketeers at will, many years of lead time were needed to raise a force of archers.43 Likewise a force of archers lost in combat was far more difficult to replace if it could be replaced at all. Archery depended upon the existence of a traditional socio-economic structure, in which such special skills were nurtured. Once changing conditions began to undercut the direct economic benefits of archery to a society, it could be kept alive only by direct and permanent government support. The Ottomans seem to have done this to a degree, but such a system had little elasticity. The difference can be seen in the effect upon Spain and upon the Ottoman Empire of a major defeat accompanied by severe manpower losses. The effect of Djerba on Spains pool of trained military manpower, though severe, was temporary. There are indications, however, that the Turks never quite recovered from Lepanto. Certainly they never again displayed the same tactical aggressiveness at sea.
It was not, in sum, the technical superiority of gunpowder weapons which made the difference. Under most circumstances gunpowder weapons, even the mighty Spanish musket, were quite inferior to the composite recurved bow. Only in the specialized circumstances of a formal head-on clash between galley fleets or a siege on land did the muskets superior impact energy give it the edge. But because archery depended totally upon the economically vulnerable values of a traditional society the nations relying upon it were deprived of resilience in the face of adversity.
Having dealt with individual shock and missile weapons, we now turn to artillery, in many ways the most important of the three. It is generally accepted that changes in cannon technology played an important role in bringing about the decline of the Mediterranean system of warfare at sea. But this was not, as has been frequently asserted, because the introduction of effective heavy artillery rendered the war galley obsolete. The mechanism involved was far more complex and was not exclusively technological in nature. Cannon were readily absorbed by the Mediterranean system of warfare at sea up to a point. Just what that point was will be the subject of our final chapter.
Since this whole matter is intimately bound up with tactical considerations, we will include in the following pages an examination of the tactical characteristics of sixteenth-century artillery. We will not go into the technical history of artillery except to show the origins of the main types of artillery used in the sixteenth-century Mediterranean and their relative economic and tactical advantages.
The first cannon made their appearance around the beginning of the fourteenth century, small, crude, pieces throwing a weird variety of ill assorted projectiles which were incapable of accomplishing much in the way of physical destruction. This situation prevailed until the third quarter of the fourteenth century. Then a major technological breakthrough occurred with the development of the hooped bombard, made up of wrought-iron staves and hoops much like a barrel and designed to throw a spherical projectile, usually of cut stone. This led, by the 1380s, to the construction of relatively large pieces, mostly breech loaders with detachable powder chambers of reduced internal diameter for structural reasons, throwing large cut stone balls. These quickly became gargantuan in size and were the first gunpowder weapons to have any real impact on the art of war. Their use as siege weapons became increasingly important and the traditional dates for the first use of artillery in the Mediterranean world come from this period.44 By the second decade of the fifteenth century German cannon founders, at least, had learned to successfully duplicate the shape of the hooped bombard in bronze.45 Cast as muzzleloaders with an integral powder chamber of smaller diameter than the bore, these were stronger and safer: but bronze was far more expensive than iron and the wrought-iron bombard hung on for economic reasons as long as labor remained relatively cheap. By the 1450s effective siege artillery for the state with the fiscal power needed to command the resources was a reality. Charles VII of France consolidated his kingdom with the assistance of the Bureau brothers siege train from the late 1440s and Constantinople fell to the power of Mehmed IIs bronze siege bombards in 1453. Effective shipboard artillery made its first, hesitant, appearance in this period.
Also about this time frame long guns firing cast-iron balls appeared in competition with the older, but steadily improved, designs which threw balls of cut stone. Though at first of hooped wrought-iron construction, long guns soon came to be made of bronze as well, an innovation which endured to produce the premier cannon of the early sixteenth century. The reasons for the success of long bronze cannon are curiously backhanded: increasing the length of cannon barrels yielded no significant increase in range (see Appendix 2); but the increased length helped to overcome the limitations of the imperfect casting techniques and metallurgy of the day, resulting in safer cannon which were less likely to burst. This was because the density and strength of bronze gun metal increase as a function of the pressure under which it is cast. European founders cast their cannon with the muzzles up; hence the density and strength of the metal at the breech where internal pressures were greatest and where guns commonly burst was a direct function of the length of the barrel (see Appendix 3). Though long guns required more bronze for the size and weight of ball thrown, they used cast-iron cannonballs which were cheap and easy to produce. Cannon designed to throw stone cannonballs, on the other hand, consumed less expensive bronze in their manufacture a fact which was generally recognized but fired projectiles which were quite expensive in terms of highly skilled labor.46
None of the developments which we have outlined, at least none subsequent to the development of the hooped bombard, established any clear-cut technical superiority. Tactical advantages were generally offset by economic considerations and vice versa. Thus all of the technical features which we have discussed survived into the sixteenth century, often in odd permutations and combinations. Generally, however, sixteenth-century Mediterranean artillery can be divided into four basic categories: large, wrought-iron pieces, small swivel guns, cast-bronze muzzleloaders and cast-iron artillery.
First were the large wrought-iron cannon. These were, for both technical and economic reasons, approaching obsolescence as the sixteenth century dawned. Large breechloaders had dropped out of favor as the basic manufacturing problems involved in making a wrought-iron barrel permanently sealed at one end were solved. The removable powder chambers of cannon designed to fire a ball weighing more than a few pounds were simply too heavy and unwieldy, particularly for shipboard use.47 Equivocal references to large pieces of this sort are very rarely encountered as late as the 1530s; but by and large they were replaced by muzzleloaders early in the century.48 Large wrought-iron muzzleloaders designed to fire cast-iron cannonballs were made early in the century; but these were ultimately no more successful than the breechloaders. Built of multiple layers of staves and very closely spaced hoops which gave them a relatively smooth outer appearance, they must have been inordinately expensive in terms of labor. Apparently for this reason they were quickly replaced by cast-bronze guns. We have shown evidence that the Ottoman basilisks used at Jiddah in 1517 were of this type; but this is our last relatively unequivocal evidence of their use. Numbers of old wrought-iron bombards (by definition stone-throwers) soldiered on into the sixteenth century; but corrosion took a steady toll and by the middle of the century they were found only aboard small merchant ships and in fortifications. Such pieces tended to be unsafe, partly because of the effects of saltwater corrosion and partly because they were built for stone projectiles (which were increasingly hard to find), and the high internal pressures which resulted from the use of iron cannonballs caused them to burst. Consequently wise artillerists underloaded them and their effective ranges must have been very short, probably no more than 200 yards at the outside. Their importance in warfare at sea, never great, diminished throughout the sixteenth century.
More successful were the small wrought-iron breechloaders, called versos by the Spanish and Portuguese, which appeared late in the fifteenth century. Incorporating many of the characteristics of the very earliest successful cannon these, because of their cheapness and tactical efficiency, lasted on through the sixteenth century to be increasingly refined and even copied in bronze. Made originally of a single layer of wrought iron staves with a dozen or so reinforced hoops shrunk on to the reverse-tapered barrel at regular intervals (see Fig. 7), one of these could have been made by a competent blacksmith and his helpers in perhaps a week. With bore diameters ranging from under two inches to over eight, these pieces were generally about four to six feet long. Their weights varied from under a hundred pounds to over four hundred. Invariably swivel mounted, they came equipped with two or three removable powder chambers which closely resembled beer mugs in form. These could be loaded and fired in rotation, permitting a maximum rate of fire of perhaps two shots per minute for short periods. Though the size, weight and basic characteristics of these swivel guns remained essentially unchanged, their design and construction was progressively refined: smooth wrought-iron and cast bronze construction were introduced early in the sixteenth century and slowly supplanted the earlier breechloaders of hooped construction ( Fig. 7). The swivel-mounted breechloaders with barrels made of finely wrought iron like large musket barrels tended toward smaller bore diameters and proportionately longer barrels than their hooped predecessors. In Spanish naval service these were called, in order of decreasing size, versos dobles, versos, ribadoquines and esmeriles, the esmeril being roughly equivalent to the Turkish prangi.
With cast-bronze swivel guns the picture was more complex. The bronze verso was much like the iron verso; but the naval medio cañon, though clearly a swivel piece larger than a verso (see Appendix 6), was probably a muzzle loader and the morterete (the Turkish darbezen) was a very short piece designed to throw scatter shot exclusively. Confusingly, the term falconete was more or less simultaneously applied to a cast bronze swivel piece (possibly a muzzleloader) intermediate in size between a medio cañon and a bronze verso, a large wrought-iron swivel gun of hooped construction, and an orthodox small bronze muzzleloader of the culverin class used ashore. Similarly, the term mosquete (musket) designated, in addition to the shoulder arm with which the term is usually associated, a larger naval swivel piece of similar construction and a small long, cast-bronze swivel piece for naval use.
Because of their unquestionably modest muzzle velocities, the product of the reverse taper of their bores, the effective ranges of the early hooped versos must have been about 200 yards. That of the short morteretes would have been even less. The later verso types with longer untapered bores and tighter chambers would have had considerably longer effective ranges when used with solid shot, probably approaching 500 yards against personnel in the open or men lightly protected by pavisades.
Swivel pieces of this general type were used as railing pieces on large merchantmen and were the basic armament of smaller ones.49 Loaded with scattershot and fired down into the rowing benches of a raiding galiot at close range they must have been horribly effective, particularly in light of their high rate of fire. The larger swivel guns formed an integral part of the bow artillery of the war galleys of all the Mediterranean nations while the smaller types were mounted at the stern and along the railings.
Before turning to our discussion of cast-bronze artillery proper, a few words on gunnery and ballistics are necessary in order to lay the tactical groundwork for our discussion of technical and economic considerations. Here, as in many other areas, modern preoccupations have influenced our view of sixteenth-century Mediterranean conditions and it is necessary to set the record straight. The modern fixation on long-range accuracy is an example. Modern works dealing with sixteenth-century ballistics and gunnery have devoted considerable attention to the theory of shooting at long ranges, yet it is probably accurate to say that most sixteenth-century gunners knew little (at least little that was correct) and cared less about accurate long-range shooting. The reason for this is not hard to find: sixteenth-century artillery was quite ineffective at long ranges.
It is fair to say that the maximum effective range of sixteenth-century artillery was between two and five hundred yards, depending on the conditions and the nature of the target. Any sound bronze cannon larger than a swivel piece could throw its projectile several thousand yards; but for a number of reasons this was of little practical consequence. First of all, a smooth-bore gun firing a spherical projectile is intrinsically inaccurate, not least because spherical projectiles which spin in an unpredictable fashion do not fly true. On top of this, undersized cannonballs were used to compensate for variations in ball size and shape, for bores which were slightly out of true and for the hard powder deposits which quickly built up on the inside of the bore with firing. This meant that the cannonball rebounded from one side of the barrel to the other and left the muzzle at a slight and unpredictable angle from the centerline of the bore. In addition it acquired an indeterminate amount of spin from dragging against the walls of the barrel and, like a sliced golf ball, had its trajectory curved in one direction or the other by aerodynamic forces. Secondly, because inaccuracy prevented concentration of fire and because a spherical projectile quickly loses its destructive energy to aerodynamic drag, inert cannonballs could not cause much damage to any target which they were capable of hitting at long range. Exploding cannonballs were known; but since they were fused with a simple piece of slow match which had to be lit before firing a hair-raising procedure which called for delicate timing and consistently good luck they were used only in short-barreled mortars and then rarely.
Consequently, long-range artillery fire played a very small part in sixteenth-century warfare. Like his successors for the next two and a half centuries, the sixteenth-century gunner did the overwhelming majority of his work well within musket range. Occasionally massed formations in the field presented a worthwhile long-range target: at the siege of Malta in 1565 gunners firing from Fort St Angelo helped to repel an assault on Fort St Elmo nearly 1,000 yards away across the Grand Harbor.50 Venetian gunners at the siege of Famagusta in 1570-1 disrupted a general review of troops by Mustafa Pasa, the Ottoman Commander, with fire from 60-pound culverins at an estimated range of three miles.51 But these were exceptional occurrences. At sea an escape might be made good or a prize captured by means of a long shot which brought down some important part of the enemys rigging; but there was an important element of luck in such exploits. Significantly, the English expression a long shot has come to be associated primarily with gambling.
The gunners on Venetian galleys, all in all probably the best naval gunners in the Mediterranean, seem to have had a reasonable expectation that they could hit an oncoming hostile galley, given calm sea conditions, at about 500 yards, about the effective limit of naval gunnery. More will be said about this in chapter 5; but it is clear that Venetian gunners could hit at ranges which were considered excessive by others.52 For tactical reasons which we have already touched on most artillery fire in galley engagements was within what the Spanish, with characteristic directness, refer to as clothing-burning range (quemaropa).
Nor were effective gunnery ranges much longer on land. For battery, the breaching of fortress walls by cannon-fire by far the most important role played by artillery through the middle of the sixteenth century at least about 60 yards was considered the optimum range.53 Good heavy artillery was rare and massed counterbattery fire was seldom encountered, so it made good sense to go straight at the main fortress walls from less than a hundred yards out for maximum destructive impact. These tactics, expensive in casualties and cheap in material and munitions, stand in direct contrast to siege tactics in the days of Vauban when cannon were relatively plentiful and manpower expensive. More will be said about this in our examination of the siege of Malta.
For both siege and naval warfare simple aiming techniques sufficed. Sighting down the top of the barrel between his upthrust thumbs,54 a good gunner could get all the accuracy that, practically speaking, his cannon was capable of producing. If the target was at or near point-blank range (from the Spanish punto de blanco, pointed at the target), usually about two to three hundred yards, the gunner simply aligned the top of the barrel directly with the target (see Fig. 8). This system, called sighting by-the-line-of-metals (por el raso de los metales in Spanish), continued to be used almost exclusively by naval gunners everywhere through the second decade of the nineteenth century,55 an eloquent commentary on the intrinsic inaccuracy of smooth-bore artillery. If the target were beyond point-blank range the gunner simply elevated his line of aim slightly above the target, judging from experience how much elevation was required, and let drive. At any range where there was a reasonable expectation of hitting the target at all this system was quite accurate. Though the gunners quadrant (see Fig. 8) was known, it was rarely used in practice. In the absence of any remotely realistic data connecting range with angles of elevation, it must have been of little or no value to the practical gunner.56
The sixteenth-century gunner did, on the other hand, exercise considerable care and precision in the loading operation. Christian gunners, at least good ones, seem to have been proficient in using elaborate rules to compute the correct ball size and powder charge for a cannon, depending upon its type, size, proportions and condition. Careful measurements of the cannons dimensions were converted, according to these rules, to a maximum safe powder charge which was then adjusted for the cannons condition. A ladle was constructed to throw a charge of the computed weight using measurements based on simple multiples of the ball diameter depending on the type of gun. Luis Collados Platica Manual de Artilleria is devoted primarily to such calculations and references by Biringuccio and Tartaglia among others make it evident that similar methods were actually used by ordinary gunners.57 Similar calculations of this type were made for Ottoman guns by arsenal personnel who engraved the correct ball size and weight of powder directly on the breech.58
The reason for these elaborate calculations, often derided by modern writers because of their lack of any theoretical foundation,59 was a very real and compelling one. Sixteenth-century cannon-founding techniques were imperfect, particularly early in the century, and improved only slowly (see Appendix 3). Many cannon failed proof (burst under test with an oversized charge) at the foundry; but many more passed, only to burst on active service with terrible consequences for the gunners involved.
Thus while the sixteenth-century gunner was not much worried about long-range accuracy, he was intensely concerned with internal pressures. He went to great lengths to predict them, after a fashion, and to control them. His careful measuring of the proportions of cannon and the size of cannonballs, his careful construction of powder ladles to throw the proper charge and his care in loading his gun, were all directed to this end. His computations were entirely empirical and were ultimately based only upon past experience. All that can be said for them is that they usually worked and that is a great deal.
Having dealt briefly with gunnery, we now turn to the guns themselves. As the sixteenth century dawned, heavy artillery of cast bronze had established, if only just, a decisive superiority over its older wrought-iron competition. Bronze cannon of the sixteenth century can be divided into two large categories: those designed to fire a stone projectile and those designed to fire a projectile of cast iron. The first of these, called pedreros by the Christians, were cast with an integral powder chamber of reduced diameter, usually about a third to a half that of the bore (see Fig. 9). Because the sectional density of a stone cannonball is only about a third that of an iron one, the internal pressure and therefore the amount of gunpowder needed to produce a given muzzle velocity was less. This was reflected in the practice of sixteenth-century gunners who, at least in the West, loaded a pedrero with one half the powder charge of a cannon of battery throwing an iron ball of the same weight.60 The barrel walls of a pedrero could therefore be made thinner than those of an iron-throwing gun,.with attendant savings in weight and in the amount of expensive bronze required. The Turks, at least, cast their pedreros muzzle down, thus putting the strongest, densest, metal where the barrel was thinnest and relying on sheer mass for strength around the chamber. The moulds contained a large bell above the breech (this was cut off after the metal had cooled) to increase the pressure of molten metal during casting. This permitted relatively short barrels of about eight calibers (eight times the bore diameter, not counting the chamber) without sacrificing strength (see Appendix 3). As a result, pedreros were considerably shorter and lighter than cannon designed to throw an iron cannonball of equal weight.
A comparison of the main types of cannon used in Mediterranean warfare at sea
(a) Ottoman basilisk, ca. 1453, cast for a stone ball of about 930 lb. Guns similar to this may have been used by the Ottomans at Zonchio and Jiddah. After an existent cannon in Rumeli Hisar, Turkey.
(b)Ottoman 120 pdr, wrought-iron basilisk. After an existent piece in the Deniz Musesi Park, Istanbul. The Ottoman basilisks used at Jiddah in 1517 were possibly similar pieces, perhaps 40-50 pdrs, of the same construction. This monster was probably a specialized siege/fortress gun.
(c) Spanish 36 pdr culverin, mid-sixteenth century. After existent pieces and Luis Collados proportions.
(d)Spanish 50 pdr full cannon, mid-sixteenth century. After existent pieces and Collados proportions.
(e) Venetian 52 pdr full cannon, based on existent pieces.
(f) Ottoman 56 pdr bas topu, mid-sixteenth century. After an existent piece in the Askeri Musesi Cannon Park.
(g) Ottoman 56 pdr pedrero, ca. 1570. After an existent piece in the Askeri Musesi Cannon Park.
(h)Spanish 14 pdr half culverin, mid-sixteenth century. After a gun by the German founder Gregory Leofner in the Museu Militar, Lisbon. This is the gun referred to in Pig. 17. Similar pieces by Spanish founders tended to be larger and heavier for their bore diameters.
(i) German 12 pdr half culverin, ca. 1516. After an existent piece in the Askeri Musesi Cannon Park, Istanbul. This piece is probably representative of many similar Spanish pieces by German founders. Earlier Spanish-German pieces tended toward greater bulk and length and more elaborate ornamentation.
(j)Venetian 12 pdr sagre, early to mid sixteenth century. An Ottoman ayka topu (sidepiece) would have been of similar dimensions.
(k)Venetian 12 pdr aspide, mid to late sixteenth century. Note the short length of this later piece.
(l)Spanish 6 pdr half sacre, mid-sixteenth century. Most sacres used on galleys would have been somewhere between this piece and the 14 pdr half culverin (h) in size and proportions.
Pieces (c) to (g) are probably representative of most cannon used as main centerline bowgun on Mediterranean war galleys from the 1530s until Lepanto.
The design characteristics of pedreros made it possible to cast them quite large: fifteenth-century Ottoman siege and fortress pieces designed to throw a ball weighing in excess of a thousand pounds are still extant61 and siege pedreros which threw balls of six and seven hundred pounds must have been common in the sixteenth century. Smaller pedreros, usually from twelve to fifty pounders, were used extensively at sea. Luis Collado speaks highly of them in this context and no less an expert than John Hawkins seems to have used them by preference.62 The modern belief that pedreros were anti-personnel weapons whose stone projectiles shattered on impact with a wooden hull must be discounted entirely. Evidence that the impact of a stone cannonball on a warship was at least as destructive as that of an iron cannonball of equivalent weight is overwhelming and explicit.63 Nor did the pedrero suffer a disadvantage in range: the limiting factor in naval warfare was not range, but as we have explained, accuracy which affected iron- and stone-throwing cannon alike. In siege warfare a stone projectile at least a large one made up in close-range smashing power whatever it lacked in penetrative ability.
The pedreros weak point was the amount of highly skilled labor which went into the manufacture of stone cannonballs. A cannonball cutter had been a very highly paid artisan from the earliest times.64 It is reasonable to suppose, therefore, that the wage-price spiral of the sixteenth century acted to increase the price of stone cannonballs more sharply than the prices of less labor-intensive products and, at the same time, to reduce the cannonball cutters real wages.65 The cannonball cutter, like the maker of composite bows, was thus caught in an economic squeeze, particularly where free market conditions prevailed, with predictable long-term results. It is a matter of record that pedreros dropped out of use first in those areas where the upward movement of wages and prices began England and north-western Europe and hung on longest in economic backwaters where the onset of the wage-price spiral was delayed Portuguese India and the Ottoman Empire.66
In the sixteenth-century Mediterranean the pedrero was fighting a rearguard action, its tactical advantages being slowly but surely undercut by its economic vulnerability as bronze and gunpowder became increasingly cheap relative to skilled labor. Large stone-throwing guns for siege and fortress use were employed extensively only by the Ottomans and by the Portuguese in the east. Both, however, continued to use them throughout the sixteenth century. The smaller pedreros extensively used by all the Mediterranean nations as shipboard artillery slowly became less common as the century wore on. The Ottomans and curiously the Knights of Malta seem to have clung to them longest and the Venetians to have abandoned them first.
Cannon designed to fire a cast-iron cannonball fell into two categories. We have already discussed the origins of the first of these, the long guns or culverins.67 Culverins came in a wide variety of sizes, from two-pound falconetes weighing less than 150 pounds (exclusive of the carriage) to sixty-pound double culverins, with barrels weighing close to 9,000 pounds; but all had straight, unchambered bores which were at least 26 calibers long (26 times the bore diameter) and sometimes a great deal longer. In order of increasing size, the most common sub-types within the culverin class were the falconete (usually a two or three pounder), the half sacre (about a six pounder), the sacre (about a ten pounder), the half culverin (usually about a twelve pounder), and the culverin (normally a 24 to 26 pounder).68 Double culverins, bastard culverins, and the like, of considerably greater size and sometimes as large as 70 and 80 pounders, were occasionally encountered.
In competition with the culverins were the cannons (here we use the term in its specific, technical, sense), or cannons of battery as they were known in the sixteenth century. Like culverins, cannon were of cast bronze and fired iron cannonballs; but their bores were shorter, usually about 18 calibers long, their barrel walls were thinner and they customarily used a smaller powder charge. The fact that many early cannon were chambered, that is fitted with a powder chamber of reduced diameter, suggests a bombard ancestry. This feature made loading difficult (with pedreros, which were also chambered, this was less of a problem since pedrero barrels were relatively wide-mouthed and short) and became progressively less common as the century wore on. Cannon were less elaborately categorized than culverins, the main sub-types being the half-cannon (normally a 22 to 24 pounder whose barrel weighed from 4,000 to 5,000 pounds) and the full cannon (usually a 40 to 60 pounder with a barrel weighing from 5,000 to over 8,000 pounds in the sixteenth century).69
References to quarter cannon, roughly equivalent to a half culverin, but shorter and throwing a larger ball, are occasionally encountered. Monster double cannon and cannon royal were cast to throw balls of from 80 to as much as 150 pounds, particularly early in the century; but these weighed more relative to their projectiles than smaller cannon, were considered unwieldy and inefficient in the West, and are rarely heard of except in the Ottoman siege train.70
The main advantage of the culverin early in the sixteenth century was its reliability and the confidence which it inspired in gunners. Whether this was explicitly recognized or not, the culverins great length gave them added strength at the breech. This was initially a very important safety factor. That many early cannons were chambered is clear evidence that weakness at the breech was recognized as a major problem. That this feature became less common as time went on, finally disappearing toward the end of the sixteenth century, must be attributed to gradual improvements in smelting and casting techniques and in the quality of cannon metal .
Though initially less reliable and more dangerous, cannons required only about two-thirds the bronze needed for a culverin cast for the same ball weight and used a powder charge which was smaller by at least a third, powerful economic incentive for further development.71 Not surprisingly, as manufacturing standards improved throughout the sixteenth century, culverins became progressively less popular and the barrels of cast-bronze cannon in general became shorter.72 It is difficult to avoid postulating a cause and effect relationship between these developments and the general improvement in manufacturing standards.
This progressive reduction of barrel lengths has been variously interpreted by naval historians, usually as an effort to improve ease of shipboard handling by sacrificing range.73 Instead it seems evident, in light of our knowledge of the characteristics of cast bronze, that length was an undesirable but initially necessary means of strengthening the breech. Because of the intrinsic characteristics of black powder, increasing the length of a cannons bore beyond about 12 calibers yielded only a slight increase in muzzle velocity and range and increasing it beyond 18 calibers yielded effectively no increase at all (see Appendix 2). The maximum ranges of all sixteenth-century cannon firing cast-iron projectiles were therefore essentially the same regardless of barrel length (the extreme ranges of pedreros were somewhat less because of the greater aerodynamic drag of a stone cannonball per unit of mass; but as we have indicated, this was of little practical significance).
The cannons ultimately triumphed over the culverins because of their greater economic efficiency. The relative unwieldiness of the longer, heavier culverins was undoubtedly a consideration; but there can be little doubt that the crucial factor was the saving in bronze and gunpowder. Using considerably less of both, a cannon could throw an iron ball of equal weight just as far and just as hard as a culverin. The completeness of the cannons victory is illustrated by the application of the term cannon to virtually all modern gunpowder weapons larger than a machine gun.
Though relatively little systematic study has been devoted to the subject, there were a number of clear-cut regional differences in the construction and use of bronze ordnance at sea by the Mediterranean nations. Though we will go into the matter in more detail in our discussion of galley tactics, a few general observations are in order here. Venetian cannon, regarded by contemporary experts as the best in the Mediterranean,74 were markedly shorter and lighter than equivalent Spanish or Ottoman pieces, a fact of considerable tactical significance in galley warfare as we shall see. The difference in bulk and length between equivalent Spanish and Venetian naval pieces was particularly marked, Venetian culverins being almost equivalent to Spanish cannons and vice versa. A Venetian 52-pound cannon of 1556 cited as typical by Cristoforo da Canal had a bore length of only 16½ calibers and weighed 5,180 pounds.75 We can take it as a fairly representative example since Venetian guns unlike Spanish ones were reasonably closely standardized. A contemporary Spanish naval cannon which weighed almost exactly the same, 5,050 pounds, and about which we have precise information was just over 21 calibers long and threw a ball of only 40 pounds.76 Spanish cannon (technically half cannon) of about the same weight and vintage survive which were bored for a ball of only 26 pounds.77 The general tendency of Spanish guns toward relatively great length and weight is explicitly confirmed by Luis Collado, an experienced and informed Spanish artillerist.78 Interestingly, while Collado considered Spanish cannon to be sound, if somewhat unwieldy, he considered Genoese cannon, apparently of about the same proportions, to be of defective construction and positively dangerous.79 Portuguese naval artillery was among the best in the world, if not the best, early in the sixteenth century. The Portuguese, however, hardly ever intervened directly in Mediterranean affairs after the 1530s. French artillery, particularly field artillery, was highly regarded, but had little impact on Mediterranean warfare at sea the use of French artillery by the Barbarossa brothers to reduce the Spanish-held Peñon of Algiers in 1529 is the only major example following the French abandonment of Venice in 1499 which comes readily to mind.80
The Venetians seem also to have had a clear-cut superiority in the efficiency of their naval gunners, particularly aboard galleys. Their galleys were designed, as we shall see, with the efficient use of their heavy ordnance as a primary consideration. Ottoman and Spanish naval gunnery seem to have been roughly on a par with each other. Though the general level of practice was apparently not as high as that of Venetian naval gunners, the best Turkish and Spanish gunners were very good indeed. Ottoman siege batteries seem to have been particularly well handled, techniques having been developed to combine systematically the superior penetration of iron cannonballs with the tremendous smashing power of the large stone-throwing cannon.81
One final point about sixteenth-century gunnery must be made. The sixteenth-century gunner was not, unlike his eighteenth- or nineteenth-century successors, a mere private soldier or able seaman who had been trained to perform certain mechanical movements under fire. Gunnery was still like race-car driving before World War I or commercial flying in the 1920s an uncertain business with a certain grimy glamor and a tangible aura of mystery about it. It was all very much seat of the pants. Experience, basic intelligence, and a feel for the business paid off more surely than precision in following a rigid set of procedures. Sixteenth-century gunnery was an art, not a science; and attempting to evaluate it as if it were a science can lead to serious errors. Its finest practitioners were men of unusual ability and daring who were attracted to gunnery by the mystery of the esoteric knowledge involved, by the danger, and by the prestige. It is worth noting that a number of great artists were cannon founders and gunners as well Benvenuto Cellini and Leonardo da Vinci come immediately to mind. It is not surprising, then, that the best sixteenth-century gunnery surpassed that of the eighteenth and early nineteenth centuries, the lack of any scientific apparatus notwithstanding.
But the elitist character of sixteenth-century Mediterranean gunnery was essentially a cultural holdover from pre-gunpowder days, soon to be smothered by an outpouring of cheap cast-iron cannon from the north. As long as cannon were unique, individually made pieces, few in number and extremely expensive, this sort of relationship between gun and gunner was viable. A single good cannon, properly placed and skillfully served, could make the difference in almost any sixteenth-century engagement on land or sea. The applicability of this observation to galley warfare is obvious. But when cheap, standardized, cannon became available in large (numbers, crude mass replaced selective precision and the Mediterranean war galley found itself in serious trouble, militarily as well as economically.
This brings us to our last type of artillery, properly almost beyond the bounds of our subject cast-iron cannon. Successfully made for the first time by an English parson in the service of Henry VIII in 1543,82 artillery of cast iron had, by the 1580s, become common in northern waters, the Dutch, the Swedes and the Germans in turn having acquired the technology from the English. The Mediterranean nations, however, did not do so until much later.
Cast-iron ordnance was not superior to bronze; it was heavier, bulkier and far more dangerous to use. But it had one overwhelming advantage it cost only a third as much.83 The manufacture of iron ordnance in the north on a scale approaching mass production from the 1580s on sounded the death knell for the traditional methods of acquiring military skills, however effective they may have been in the narrowly tactical sense. It also sounded the death knell for the war galley, marking the ultimate triumph of gunpowders essential characteristics over those of the Mediterranean system of warfare at sea.
chapter is excerpted from:
GUNPOWDER AND GALLEYS:
CHANGING TECHNOLOGY AND MEDITERRANEAN WARFARE AT SEA IN THE SIXTEENTH CENTURY
JOHN FRANCIS GUILMARTIN JR.
CAMBRIDGE UNIVERSITY PRESS
Published by the Syndics of the Cambridge University Press
Bentley House, 200 Euston Road, London NWI 2DB
American Branch: 32 East 57th Street, New York, N.Y.10022
Library of Congress Catalogue Card Number: 73-83109
ISBN: O 521 20272 8
First published 1974
Printed in Great Britain by
Alden & Mowbray Ltd.
at the Alden Press, Oxford
1The Hussites must be noted as a partial exception. The Hussites were forced, by their social complexion and the nature of the religious wars on which they had embarked, to break with their old cultural patterns and take up new methods of warfare. It was precisely their lack of traditional military skills and time which forced them to train in the use of firearms. The reference here is to the Ottomans and the Spanish.
2This truism has been noted frequently with regard to socially and culturally based resistance to the correct tactical employment of new and novel weapons; but it is equally true with regard to expectations as to the sort of training to be employed. It is commonplace to note that aircraft were at first employed as cavalry for reconnaissance. But it is equally significant that replacement aircraft were at first casually handed out to units just as remounts were assigned in the cavalry and that French pilots in World War I were allowed to teach them selves to fly by the Bleriot system without benefit of an instructor. Like a boy learning to handle a docile old plug before progressing to more spirited mounts, the tyro pilot began on an underpowered, clipped-wing, aircraft which was capable only of fast taxiing. He then progressed to an underpowered trainer which was capable of short hops and, by stages, to full-fledged aircraft, never flying with anyone but himself.
3By shock action we mean combat in which the attacker attempts to gain the upper hand by bringing his troops and/or their mounts into physical contact with the defenders in order to benefit from the violent impact the shock of their onslaught. Pure shock action, as represented by a medieval European cavalry charge or the attack of a Swiss pike square, is encountered relatively rarely in military history; nevertheless the distinction between shock and missile action is a valid one. Though too much can be made of it, troops who are exceptionally good at the one are rarely particularly adept at the other.
4These came in a bewildering variety under an even more bewildering variety of names. Basically, they were all variations on the half-pike (for thrusting) and the halberd or pole axe (for hacking), with the optional addition of a hook on the reverse side for pulling an opponent off balance and other gruesome looking projections for aesthetic appeal.
5This weapon, some six feet long, is on display in the Museo Naval, Madrid.
6Many of these, both Venetian and Turkish, are on display at the Ducal armory and in the Museo Correr, Venice.
7The tapestries depicting Charles Vs Tunis campaign of 1535 in the Armeria Real, Madrid, are our most concrete evidence of the general trend. Spanish nobles, like Venetian nobles and the Knights of St John, fought in essentially full plate armor: breastplate and backplate, helmet, arm and shoulder defenses, tassets and complete defenses for at least the upper leg. Many, if not most, ordinary Spanish infantrymen and German and Italian mercenaries in the Spanish and Imperial service wore at least an open helmet of some sort plus a breastplate and tassets. Ottoman timariot cavalrymen, who served dismounted aboard ship on major campaigns, wore an open helmet and fairly complete suits of chain mail reinforced with plate at strategic points, often including a complete breastplate; but Janissaries and azabs seem to have worn very little armor, most of it chain mail, though helmets were not uncommon and breastplates were apparently worn by a few.
8I have ignored the Chinese crossbow since it had no influence on the events with which we are concerned here. The Arabs knew the crossbow; but only seem to have used it in siege warfare.
9The use of the crossbow against all but infidels was first interdicted under penalty of anathema by the Lateran Council in 1139. The prohibition was confirmed at the close of the twelfth century by Pope Innocent III, and Conrad III of Germany (1138-52) forbade its use in his territory. See Sir Ralph Payne-Gallwey, The Crossbow (London, 1903), p. 3.
10According to Payne-Gallwey, The Crossbow, pp. 32, 90, the steel crossbow first appeared about 1370. This date corresponds closely to the appearance of effective plate armor on a large scale. See Sir Charles Ffoulkes, Armour and Weapons (Oxford, 1909), pp. 30-68. The transition from chain mail to full plate began in Western Europe about 1300 and was completed shortly after 1400.
11The energy stored in a tensed bow, when expended, drives not only the arrow, but also the bowstring or cord and the mass of the bow itself. It follows that a bow with less mass will be capable of driving its projectile (assuming a sufficiently small projectile mass) with greater velocity. The easiest way to reduce the mass of the bow while holding the force applied to the projectile constant is to make the bow stiffer and shorter.
12Payne-Gallwey, The Crossbow, pp. 14-15, describes shooting an ordinary arrow from a heavy steel crossbow with just that result.
13Payne-Gallwey, The Crossbow, p. 10, gives 80-300 yards as a moderate range for the destruction of a single enemy. When he says this, however, he is thinking primarily of penetrating power and not accuracy. Against a moving target on the battlefield we should shorten this considerably.
14Payne-Gallwey, The Crossbow, p. 154.
15The main variables were the size of the enemy force and the degree to which the hostile troops were protected by bulwarks, armor, etc. Payne-Gallwey, The Crossbow, p. 20, gives 370-80 yards as his estimate of the extreme range attainable with the ordinary military crossbow of the fourteenth century. This estimate applies only to normal field weapons. On p. 14 he describes shooting a heavy bolt to a range of 460 yards with a large, specialized, siege crossbow. This crossbow, however, was a monster weighing 18 pounds and still having a draw force of 1,200 pounds after four centuries of deterioration.
16C. W. C. Oman, A History of the Art of War in the Middle Ages, vol. I, 2nd ed. (London, 1924), pp. 305-19, gives a superb account of this most instructive battle, as does W. F. Paterson, The Archers of Islam, Journal of the Economic and Social History of the Orient, vol. IX, Parts I-II (November, 1966), pp. 83-4, with the addition of a thorough technical analysis. R. C. Small, Crusading Warfare (1097-1193) (Cambridge, 1956), has perhaps the best tactical analysis of Arsouf. Most of Richards crossbowmen seem to have been mercenaries from the great Italian mercantile cities.
17Sporting crossbows made for noble owners were a completely different proposition from the ordinary military crossbow and the best of them were undoubtedly more accurate. Some powerful sporting crossbows were fitted with cranequins, an expensive gear and ratchet arrangement which could be used on horseback. But the occasions on which finely made weapons of this type appeared in the field were so few and the numbers in which they were used so small as to justify omitting all mention of them from the text. The use of cranequin crossbows by Francis Is mounted bodyguard of 200 men at Marignano in 1515 (mentioned by Payne-Gallwey, The Crossbow, p. 134) is the only example which comes to mind and a very late one at that.
18Gunpowder and Firearms in the Mamluk Kingdom, a Challenge to a Medieval Society (London, 1956). The strong resistance by mounted elites to the adoption of firearms was by no means unique to the nations of Asia and the Muslim Near East. The persistence of the lance as a primary cavalry weapon in European armies until World War I, and even after in some cases, makes this clear.
19Once again we must cite as an exception the use of firearms in the field by the Hussites in the 1420s and 1430s. The Hussites solved the problem of a slow rate of fire and the need of pre-matchlock small arms for multi-man crews by giving their handgunners cover in their wagon laagers.
20Slow match was ordinary string soaked in a saltpeter solution and then allowed to dry. It burned gradually with a bright, smouldering tip like a punk stick or a small, intense, cigarette end.
21Highly strung cavalry horses disliked having a length of smouldering match swinging around their ears. More important, the matchlock arquebus required two hands, one to hold and one to adjust the match and pull the trigger, leaving none for the reins. This problem was partly solved by the invention in the early sixteenth century of the wheel-lock, a clockwork mechanism which rotated a serrated wheel against pyrites to throw sparks into the flash pan. Though fairly reliable, this mechanism was delicate and its great cost placed it beyond the means of the average soldier. Though it made the cavalry pistol possible, it had no discernible effect on war at sea.
22There is an abundance of evidence to support this point, for example Christopher Columbus request of January 1495 for 100 espingardas (equivalent to arquebusses) and 100 crossbows to equip 200 troops who were to accompany him to the New World, cited by James D. Lavin, A History of Spanish Firearms (New York, 1965), p. 43. Bernal Diaz del Castillo, in his account of Cortez expedition, The Discovery and Conquest of Mexico 1517-1531, A. F. Maudslay trans. (New York, 1956), habitually speaks of firearms and crossbows in the same breath.
23Alberto Tenenti, Cristoforo da Canal: la Marine Venetienne avant Lépante (Paris, 1962), p. 39.
24Francisco Felipe Olesa Muñido, La Organización Naval de los Estados Mediterraneos y en Especial de España Durante los Siglos XVI y XVII (Madrid, 1966), vol. I, pp. 326-8. The crossbow is still encountered in the regulations of 1550 of the Casa de Contractación of Seville establishing the standards of armament required for merchant ships trading with the Indies. These, however, undoubtedly contain an element of bureaucratic conservatism and probably represent a minimum standard of armament rather than what was actually carried. A dock yard requisition for the outfitting of 20 galleys at Barcelona in 1530, Colección Sanz de Barutell (Simancas), Articulo 3, dto. 25, fol. 87, lists arquebusses but not crossbows.
25Tenenti, Cristoforo da Canal, pp. 39-40.
26Robert Held, The Age of Firearms (New York, 1957), p. 39, estimates that the Spanish musket could penetrate armor handily up to 125 yards and stop in his tracks by sheer impact what ever man or beast it might hit at ranges well over 200 yards, an estimate which seems quite reasonable.
27This statement is based upon examination of the numerous examples on display in the Museo del Ejercito, Madrid.
28Payne-Gallwey, The Crossbow, p. 26, cites a letter from an English soldier in the Napoleonic wars which mentions the picking off of an enemy sentry with a musket at 80 yards as an exceptional feat. The accuracy of all smooth-bore small arms was about the same.
29The statement regarding the size of Ottoman small arms is based upon an examination of Muslim muskets in the collections of the Armeria Real, Madrid, and the Ducal Armory and Museo Correr, Venice. The effectiveness of Ottoman musketry in siege warfare is attested to by Franscisco Balbi di Correggio, The Siege of Malta, 1565, E. Bradford trans. (London, 1965), pp. 49-50. Balbi contends that the great length of the Muslim muskets limited their rate of fire.
30Heavier arrows give increased stopping power and penetrative power. But because of their weight they can only be driven by a relatively long bow which, because of its greater mass, is limited in the velocity and hence range which it can impart to them. Short bows, on the other hand, are capable of producing higher velocities and longer ranges for the same draw force, but only if used with a sufficiently light arrow since they lack the power to drive a heavier one. The rather complex trade-off between bow size and arrow weight, depending upon the range and toughness of the target, is explained by W. F. Paterson, The Archers of Islam, Journal of the Economic and Social History of the Orient, IX, Parts I-II (November 1966), 81. A bow of the size and weight described here was optimized toward the light, long range end of the spectrum. The portions of my discussion dealing with the construction, use and range of the Turkish bow are based ultimately upon Mustafa Kani, Telchis resail er-rumat (Excerpts from the writings of the Archers) (Constantinople, 1847), written during Turkish archerys final revival during the reign of Murad III. Kanis work was translated and published with limited commentary by Joachim Hein, Bogenhandwerk und Bogensport bei den Osmanen, Der Islam, XIV (1925), 289-360. This, in turn, was the basis for an expanded and technically much more knowledgeable treatment by Paul E. Klopsteg, Turkish Archery and the Composite Bow (Evanston, Ill., 1947), which I have used as my primary reference.
31This estimate of penetrative ability is from Paterson, The Archers of Islam, p. 86, and is based on modern experimentation. The estimate of extreme range is my own, based upon extrapolation of data given by Payne-Gallwey on pp. 18-20 Of his Appendix on the Turkish bow in The Crossbow,. I consider Payne-Gallweys data to be somewhat conservative since it is based upon the skill and strength of an archer who would have been at best a dilettante by sixteenth-century Ottoman standards, and upon bows which were preserved by a random process of selection and are unlikely to have been representative of the best and most powerful even when new.
32Paterson, The Archers of Islam, p. 83.
33Klopsteg, Turkish Archery, p. 53.
34This disadvantage of the bow was seized upon by the opponents of archery in the late sixteenth-century debate between English professional soldiers over the relative merits of the longbow and firearms. See particularly Sir Roger Williams, A Briefe Discourse of Warre (London, 1590), and, for the pro-longbow side, Sir John Smythe, Certain Discourses ... Concerning the Forms and Effects of Divers Sorts of Weapons ... (London, 1590). The anti-bow arguments used by Sir Roger in this highly instructive debate are frequently quoted out of context to imply that the bow was rendered obsolete by inherent technical defects. In fact, as careful analysis of this debate shows, the longbow itself had a considerable advantage over late sixteenth-century firearms in accuracy, range and rate of fire, but good archers could no longer be found in sufficient numbers to justify their recruitment.
35Payne-Gallwey, A Treatise on Turkish and Other Oriental Bows of Medieval and Later Times, in The Crossbow.
36Klopsteg, Turkish Archery, pp. 50-51. These glues, made of boiled-down cattle tendons and skin with small amounts of fish glue added to increase fluidity and flexibility, were as strong as the best modern glues, but were slow-drying and very difficult to handle and apply. A great deal of art was involved in their preparation and application, much of it characterized by a mystical, semi-religious approach.
37Olesa Muñido, La Organización Naval, vol. I, p. 211. Admiral Fincatis reconstruction of a Venetian galley of the 1530s in the Museo Storico Navale, Venice, has these too. Pavisades are listed in an inventory of material needed for the construction of to galleys in the stocks in Barcelona in 1539),Colección Sanz de Barutell (Simancas), Articulo 4, vol. I, dto. 122, fol. 383-7.
38Olesa Muñido, La Organización Naval, vol. I, pp. 188, 211-12.
39In the absence of mechanical range-finding devices or landmarks whose distance from the shooter is known, accurate shooting at long ranges depends directly upon accurate visual range estimation. Range estimation by eye is particularly difficult at sea where there are no terrain features between the shooter and his target.
40Colección Sanz de Barutell (Simancas), Articulo 6, dto. 45, fol. 117-18, entitled An account of what the Alcalde of Vélez has done with his galiots against the Spanish who navigate the coasts of Andalusia ... from the first of August to the end of October of 1561, related by Francisco Yañez, his captive.
41Cattle and goats with horns of sufficient length and adequate quality to provide the horn backing for the composite bow were not bred in Western Europe. Paterson, The Archers of Islam, p. 82.
42In tests conducted at the H. P. White Laboratory, Eel Aire, Maryland, on 1 July 1970, an 85 caliber lead ball of 890 grains (about 2.3 ounces), driven by 215 grains of black powder (0.49 ounces), typically produced a muzzle velocity of about 1,100 feet per second. Sixteenth century Spanish musketeers almost certainly used a considerably larger powder charge, as heavy as the weight of the ball according to Jorge Vigon, Historia de la Artilleria Española (Madrid, 1947), Vol. I, p. 236.
43This point is clearly related to that made by Braudel, La Méditerranée, Spanish ed. (Mexico City, 1953), p. 190, to the effect that gunpowders ultimate effect was to shift the military balance between nomadic and sedentary peoples decisively in favor of the latter.
44Mediterranean writers almost universally point toward Germany as the birthplace of artillery which we can take in context to mean effective siege artillery and to about 1300 as the date. See, for example, the statement made ca. 1453 by Kritovolous, History of Mehmed the Conqueror, Charles T. Riggs trans. (Princeton, 1954), p. 46 . Later Italian writers (e.g. Biringuccio, Tartaglia and Guicciardini) were also of this opinion, echoed by Luis Collado, Platica Manual de Artilleria (Milan, 1592), fol. 6.
45Heinrich Müller, Deutsche Bronzegeschützröhre 1400-1750 (Leipzig, 1969), pp. 28ff.
46Collado, Platica Manual, fol. g. Sir Henry Brackenbury, Ancient Cannon in Europe, Part II, Proceedings of`the Royal Artillery Institute, V (1865-6), 8-9.
47L. G. Carr Laughton, Early Tudor Ship Guns, Michael Lewis ed.,The Mariners Mirror, XLVI (November 1960), 242ff, 258 .
48L. G. Carr Laughton, Early Tudor Ship Guns, pp. 242ff. See below, Lepanto, pp. 230-31, for evidence of large wrought-iron cannon mounted aboard galleys ca. 1536 .
49Colección Navarrete, vol, XII, dto. 76, fol. 271-9, a Spanish government inventory made in 1564 of privately owned ships of certain of the Biscay ports and their armament. Fourteen of the 36 ships inventoried (they were mostly quite small, ranging from 35 to 150 tons burden) carried something larger than small arms, mostly wrought-iron railing pieces. Eight of the fourteen carried something larger than a verso. Of the eight, three carried a single piece of cast bronze plus from four to eight wrought-iron bombards on deck mounts. Two carried bombards only. Other Mediterranean nations may have armed their smaller merchant ships less heavily, but versos were clearly used extensively as railing pieces on the larger ones. See also Colección Navarrete, vol. III, dto. 4, fol. 55-132, a compilation of royal ordinances written in 1543 and amended in 1554 and 1556, prescribing the standards of armament required for merchant ships trading with the Indies. The ships of the largest class, 220 to 230 tons burden, were to have a single bronze cannon or half culverin, two sacres and a falconete of bronze, 10 wrought-iron bombards and no less than 24 versos. This document appears to mirror accurately the actual availability and use of ordnance, at least as of the earlier date.
50Francisco Balbi di Correggio, The Siege of Malta, 1565, E. Bradford trans. (London, 1965), pp. 81-2. Note that the gunners were firing at a known range, a considerable aid to accuracy.
51From a contemporary Italian source cited by Sir George Hill, A History of Cyprus, vol. 111 (Cambridge, 1948), p. 997.
52Alberto Tenenti, Piracy and the Decline of Venice (Berkeley, 1967), p. 79, gives an account of an engagement between Venetian galeasses and Dutch freebooters early in the seventeenth century in which the former were able to cannonade the latter into submission by accurate gunnery from beyond the Dutchmans range in a flat calm. This must have involved a range of at least 500 yards.
53Collado, Platica Manual, fol. 43. This was not some theoretical ideal, but the range actually used wherever possible. The use of Ottoman siege batteries 70 to 80 paces (about 60 to 65 yards) from the walls of the final Spanish redoubt on Djerba in 1560, already noted above, is a case in point.
54This technique is described by Vannoccio Biringuccio, The Pirotechnia, Cyril S. Smith and Martha T. Gnudi trans, based on the 1540 Venice edition (New York, 1941), p. 419.
55Sir Howard Douglas, A Treatise on Naval Gunnery (London, 1860), p.100, states that the practice of firing by the line-of-metal was still common in his day.
56For methods of aiming see Biringuccio, Pirotechnia,, pp. 419-20, and Collado, Platica Manual, fol. 41-2, 51. The gunners quadrant was used, if at all, only as a means of duplicating a correct angle of elevation found by trial and error. Tartaglia, a mathematician and not a gunner, has misled many with his emphasis on the gunners quadrant. As is explained in Appendix 2, many of the so-called range tables (tables purporting to give the relationship between angle of elevation and range) which were attached to most early works on artillery have little or no discernible basis in fact. Such tables have nevertheless been religiously repeated and taken as gospel by many modern authors.
57These rules are painstakingly explained by Collado for each of the three basic types of cannon: the culverins, the cannons and the pedreros, Platica Manual, fol. 14ff., 29ff., and 35ff.
58These markings can be seen on all the Ottoman cannon and many of the captured Christian pieces in the cannon park of the Askeri Musesi near St Irenes Church, Istanbul.
59For example, A. R. Hall, Seventeenth Century Ballistics (Cambridge, 1952), p. 36, says that [sixteenth-century] gunners were weighing and measuring charges and distances before measure became the great instrument of physical science, in fact we can see in their activities a conspicuous example of the truth that accuracy with balance and ruler will prove very inadequate without the framework of an intellectual system. In terms of long-range accuracy essentially a modern preoccupation Hall is quite correct; in terms of internal ballistics and safety he is much less so.
60Collado, Platica Manual, fol. 14, 29, 36.
61Many of these are still on display in Istanbul, particularly in the small park near the Deniz Musesi and in the fortress of Rumeli Hisar on the Bosporus.
62Collado, Platica Manual, fol. 9, 11. Careful examination of the Spanish inventory taken of the ordnance of Hawkins flagship, the Jesus of Lubeck, after her capture at San Juan de Ulloa, given by Michael Lewis, The Guns of the Jesus of Lubeck, The Mariners Mirror, XXII, No. 3 (July 1936), 324-35, and Of a contemporary painting of the ship by Anthony Anthony, reproduced by D. W. Waters, The Elizabethan Navy and the Armada Campaign, The Mariners Mirror, XXXV, No. 2 (April 1949), 96-7, makes it clear that three of the eight main broadside pieces were pedreros.
63See, for example, Michael Strachan, Sampsons Fight with Maltese Galleys, 1628 ,The Mariners Mirror, LV, NO. 3 (August 1969), 281-90, who quotes Captain William Rainsborow of the Sampson to the effect that the 25 pound stone shot from the galleys, unlike the smaller iron projectiles which did little damage, staved on our lower decke two barrels of beef, two of pease, a Butt of Wine; some of the shot passing through nineteen inches of planke and timber ... They bare in whole plankes, shooke and split the timbers and ... made us leakie.
64Sir Henry Brackenbury, Ancient Cannon in Europe, Part II, The Journal of the Royal Artillery Institute, V (1865-6), 8-9, cites a French document of 1375 giving the cost of a small stone cannonball, probably of no more than six inches diameter, since the cannon for which it was cut weighed only 500 pounds, as two shillings and sixpence. The most costly wrought iron used in the manufacture of the cannon involved cost only sixpence a pound, the difference in price surely being attributable entirely to labor costs.
65See B. H. Slicher van Bath, The Agrarian History of Western Europe A.D. 500-1850 (London, 1963), Pp. 113-15, for a concise explanation of the wage-price movements of the sixteenth century.
66Pedreros are rarely heard of on English ships after the 1560s and were an anachronism in England, and probably Holland as well, by 1600. Pedreros continued to be cast in Portuguese India through the mid-seventeenth century (many excellent examples are on display in the Museu Militar, Lisbon) and lasted on in the Ottoman Empire into the eighteenth.
67From the Italian colubrina (snake), the term culverin or its equivalent was used in all major Mediterranean languages. See Henry Kahane and Andreas Tietze, The Lingua Franca in the Levant (Urbana, Ill., 1952), entry 210, colubrina.
68This simplified and somewhat arbitrary categorization reflects fairly accurately mid-sixteenth century Spanish practice and, with minor modifications, Venetian practice. See Olesa Muñido, La Organización Naval, vol. I, pp. 318-23. The reader should be cautioned that the use of these terms, even by contemporary Mediterranean authors, was anything but consistent. Not only were there regional variations in the size and type of gun to which a particular term was applied, but usages changed with time. There is only the most general correlation between the definitions given here and contemporary English usage. The Spanish sacre, Venetian sagre and Turkish ayka topu, for example, while roughly equivalent to each other, were equivalent not to the English saker as we might expect, but to the English demiculverin.
69These values represent Mediterranean practice and are confirmed by the observation of surviving examples in the collections of the Museo del Ejercito, Madrid; the Museu Militar, Lisbon; the Askeri Musesi and Deniz Musesi, Istanbul; and the Museo Storico Navale, Venice. Then, as now, the term cannon had a general meaning which overlapped the specific, technical meaning. The Turkish equivalent, ba topu, simply meant main cannon.
70Collado, Platica Manual, fol. 32.
71Biringuccio, Pirotechnia,, p. 477, Collado, Platica Manual, fol. 11-12, 14-18 This is also confirmed by observation of surviving specimens.
72M. A. Lewis, Armada Guns (London, 1961), pp. 199-201 presents a thorough marshalling of the evidence for this change in English ordnance. Olesa Muñido, La Organización Naval, vol. I, pp. 302ff., presents fragmentary data which suggest that a similar process began somewhat later in Spain and (p. 318) data suggesting that such a process may have begun in Venice as early as the 1540s.
73e.g. Lewis, Armada Guns, p. 201.
74Collado, Platica Manual, fol. 8.
75Cited by Olesa Muñido, La Organización Naval, vol. I, p. 318. The cannon in question, the main centerline bow gun of a galley, was a 50 pounder; but the pounds in question were large Venetian pounds of 477 grams as opposed to the modern English pound of 453.6 grams.
76Ibid. vol. I, p. 318.
77For example the four matched half cannon in the Museo del Ejercito, Madrid, by the German founder Gregory Leoffler. These beautifully executed guns, cast between 1543 and 1546, vary in weight from about 5,100 to 5,300 pounds and are bored for balls varying in weight from 26 to 28 pounds.
78Collado, Platica Manual, fol. 27-8, indicates that barrel lengths of 26-28 calibers were common for Spanish half cannons and 22-24 calibers for Spanish cannons. Olesa Muñido, La Organización Naval, vol I, p. 318, says that Spanish cannons and half cannons with barrels as long as 26 and even 30 calibers were not unusual.
79Platica Manual, fol. 8: The cannon founding of Genoa is the worst in Europe.
80Olesa Muñido, La Organización Naval, vol. II, p. 1117.
81Collado, Platica Manual, fol. 13, 32.
82H. R. Schubert, The First Cast Iron Cannon Made in England, Journal of the Iron and Steel Institute, CXLVI (1942), 137.
83Carlo Cipolla, Guns, Sails and Empires (New York, 1965), p. 42, n. 3. See also Braudel, La Méditerranée (Paris, 1966), p. 460. Braudels data, drawn from English and German sources, shows the price of raw copper slowly descending to a low of about three times that of iron in the 1520s, then rising steadily to about nine times that of iron by 1600. This cannot be taken as a quantitative indication of Mediterranean conditions; but there was something approaching a world market in iron and copper, particularly the latter, and Braudels data is surely indicative of general conditions.