My Battle of Hastings - part two

An analysis of potential and battlefield army strength

But Harold's forces did not increase. In fact, they diminished. Florence of Worcester says: "But inasmuch as the English were drawn up in a narrow place, many retired from the ranks, and very few remained true to him."(36)

How many is "many," and how few is "few?" It is now time to make an examination of the strength of the respective armies. To pursue this we shall back up before we move on: let us make a determination of how large king Harald Hardrada's army was likely to have been; this will give us a benchmark with which to gauge the size of the opposing English army led to Stamford bridge by the Godwinsons, which was without a doubt much larger than the Viking army.(37)

The ship levy of Norway in the mid 10th century was 120 x 20-benchers and 116 x 25-benchers (and one notably large 30-bencher, perhaps the king's personal drekar).(38) The levy of king Harald in 1066 was 200 (perhaps 240) ships of war, with other supply and smaller craft; knörrs, which are beamier and therefore more capacious than a longship. His total fleet, therefore, has been estimated at around 300 (but some would give him as many as 500 vessels -- based on Florence of Worcester). The maximum levy of Norway, if eight to ten percent of the population were fielded, would have been c. 20,000 men. Hardrada took half his strength on the campaign to England in 1066 -- or c. four to five percent of the Norse population: Rather than the 9,000 men estimated for his army, which does not include the crews of the supply ships, his total forces, when added to Tosti Godwinson's, probably numbered at least 12,000.(39)

Now to the English. There is a roundabout way to approach the problem of estimating the strength of Harold's army, or armies. It involves computing the number of hides (carucates and sulungs) in England -- some 70,000 -- and applying the "five-hide rule" to come up with the number of fyrdmen: which is therefore around 14,000 for the entire realm.(40) Add in the housecarles of the earls and the king, a total which is fairly easy to estimate at 4,000.(41) Without some way of determining how many there might have been, we would be at a loss to estimate the numbers of the other class of warriors in the fyrd known as geneats. The geneat -- "companion" -- was a ceorl who had been promoted by his lord. He held land and increased privileges, in exchange for which he had sworn the oath of vassalage, which included service in the thegn's retinue.(42)

That is about as far as we can go using the five-hide rule. There is a better way, and it works for virtually any period and any locale. Here is an old maxim: "No state, without being soon exhausted, can maintain above the hundredth part of its members in arms and idleness."(43) Should we be able to arrive at a population figure for England, our problem is well on the way to being solved. Because of the miraculous survival of Domesday Book, such an estimate is possible. In fact, using Domesday, scholars have made similar studies of medieval France. Domesday England has been estimated at between 1.25 million and 2 million souls: France at the same time had around 6.2 millions within her 1794 boundaries.(44) One percent of 1.4 million is 14,000; a figure which supports the estimate of 14,000 thegns on 70,000 hides: while adding in the housecarles is one percent of 1.8 million. The only truly full-time warriors in England were the housecarles: and even they, by the time of the Conquest, had begun to be settled on estates like the Anglo-Saxon nobility. The thegns, geneats and some housecarles functioned as participants in the economy when they were not at war, so cannot fairly be counted as being kept solely "in arms and idleness." Therefore, there may have been more of them than one percent of the population; perhaps at most twice that.

It has been said that "the whole strength of central and southern England took part in that great campaign."(45) And in the Anglo-Saxon Chronicle Harold's muster produced "a land force larger than any king had assembled before in this country" (version C); "King Harold. . .assembled a naval force and a land force larger than any king had assembled before in this country" (version D). There is no doubt that the size of Harold's army impressed everyone. The "whole of central and southern England" -- East Anglia, Northamptonshire, Hampshire, Berkshire and the "home counties" -- has been estimated at 29,000 hides, or some 5,900 fyrdmen out of the whole 14,000.(46) Added to this, of course, are the Godwinsons' housecarles -- 3,400 at most. Thegns were accompanied by a retinue of geneats, well enough off to have a pony to ride to battle. If the same proportion of central and southern England answered the call to arms as had mustered for king Harald Hardrada -- four to five percent of the Norse population -- then the 5,900 thegns from that region of England, representing c. one percent, must be multiplied by four or five to 23,600 or even 29,500 total men. Aside from the thegns, many -- or most -- of these troops would have been armed like the light English infantry depicted in the Bayeux Tapestry, shown defending a piece of high ground against Norman cavalry: these are not peasant levy, because you can see that they are all well-armed with sword, spear, axe and shield. How many of these potentials had mounts with which to keep up on the march? No one can say. But if the heriot assessment of a thegn was a mail shirt, two swords and four horses (or two horses),(47) it hardly seems unreasonable to allow any geneat at least one nag to ride away from his farm in the retinue of his lord. This argues for a huge body of mounted infantry, and I am not sure if I believe it. But further research may discover that the common ceorls who had been raised to companion status by thegns -- on a ratio of perhaps five to one -- were provided for with horses by their lords, or at least borrowed for a given campaign.

An army of 30,000 men-plus would certainly fill the description of: "Where he goes he leads forests (of spears) into the open country, and he makes the rivers through which he passes run dry!"(48) Or: "If an author from antiquity had described Harold's army, he would have said that as it passed rivers dried up, the forests became open country."(49) Probably many of these geneats were without horses at all. But it seems very reasonable, that those geneats who were part of a thegn's retinue would ride one or even two of his spare horses:(50) the best geneats at any rate, and the rest would be left in the dust to trudge along on foot. There must have been at least an equal number of mounted geneats as thegns. Lastly, an unknowable number of housecarles were by now land holding tenants; those that were would have brought their own mounted vassals in a similar proportion to the thegns'. It seems, therefore, that the total strength of king Harold's army that rode north to confront the Norse invaders was at least 15,000 to 16,000 men.

The number of well-armed men available to king Harold to repel invasions was considerable. Added to this unprecedented effort at national defense would be the northern earls, Edwin and his brother Morcar, with their thegns and housecarles and local levy of geneats and ceorls. This army was on a par with that of the Norse for the battle of Gate Fulford, for the English "caused them heavy casualties."(51) Hardrada had probably left one-third of his total force with the ships at Riccall, making his land army c. 8,000 men. The earls of Mercia and Northumbria, with peasant levy added in, would have had at least as many: King Harald's Saga declares that the earls were waiting at York "with a huge army."(52)

"Heavy casualties" in any army can justly be laid down as twenty-five to thirty percent. That is a reasonable estimate of duke William's losses at Hastings -- a most hard-fought and long battle indeed(53) -- and Hardrada's casualties at Gate Fulford were probably only slightly less than this. When Harold surprised the Norse at Stamford bridge, their army was even smaller than in the first battle. Snorri says: "He (Harald Hardrada). . .divided his forces, choosing which of them were to go with him and which were to stay behind: from each company two men were to go for every one that was left behind (to guard the ships)." Thus two-thirds of an army which had already sustained c. twenty percent casualties was what faced down the Godwinson brothers on 25 September. The odds against the Norse were above two-to-one -- counting the king's army only. But in fact, as Harold had drawn near, he would have added to his strength the local peasant levies without horses -- living close enough to march on foot to Stamford bridge and arrive in time for the battle -- and such dregs of Morcar's and Edwin's army who would have been eager to redeem themselves after their defeat at Gate Fulford five days before. The battle, though fairly lengthy, was interspersed with long periods of maneuver, parley and resting, during which time the ranks were set in order. The intervention of Orri's ship-guard, after a forced run of a dozen miles, was little more than butchery. The casualties to the English from this exhausted and even more outnumbered force would have been negligible.(54)

When Harold received word of the Norman invasion, he took the road back to London in all haste. His army was largely unscathed -- except for the housecarles, who had sustained the brunt of the fighting. And if a few eager northern thegns joined their king, preferring not to wait for their youthful lords, the sons of Ćlfgar, to gather strength before taking the road south themselves -- which, in fact, they never did -- perhaps Harold's army was at nearly the same strength as when he had set out for the north on the 20th of September.

So how strong was the English army at Hastings? It was comprised for the most part of troops who had ridden with the king to Stamford bridge. So allowing at least this much, I compute Harold's army strength as he left London and headed for the Hastings peninsula at no less than 13,000 mounted men, broken down as follows (and this includes a ten percent reduction for casualties from Stamford bridge, except the housecarles who might have lost twenty-five percent or more): 5,310 thegns; rather more mounted geneats, riding borrowed -- spare -- thegns' horses; 2,000 to 2,500 housecarles. The total strength could very well have been greater, when we consider that other thegns from distant shires had arrived at London too late to catch the king's muster for the ride north; and these would have been accompanied by their retinues. During the few days he rested his army at London Harold received reinforcements as they rode in, and even troops on foot would be there by then. But these latter were left to follow as best they could as Harold rode swiftly south. He arrived late at night on the 13th of October. There he met the rustica gens of Sussex and Kent already awaiting his coming at the mustering place on Caldbec hill, at the "hoary apple tree." These men were ceorls, peasants without fyrd status at all. But they burned to revenge themselves on the invaders. How numerous were they? A few hundreds or a couple thousand perhaps. Thousands potentially could have been there. But only the most angry and brave would be. They formed only a fraction of the king's army on the morrow.

Having now established Harold's potential army strength (and we shall soon show reasons for reducing the size of it considerably), we turn finally to duke William's composite army. Again, an attempt at numbers is best left to a potential, based on one percent to five percent of the population of Normandy or north-western France; since at least half of his army came from the duchy and the rest mainly from Brittany, then Maine, Boulogne, Ponthieu, the Île de France, Poitou, Anjou and Flanders. The most distant recruits, if the Carmen speaks true, were Normans from southern Italy and Sicily. But as no other source, original and otherwise, mentions volunteers from so distant a place, there must not have been a very large contingent of them.

Calculating the relative size of Normandy to the rest of France (within her modern boundaries), for the late 11th century, we arrive at a figure somewhat under 300,000 souls. When the duke's vassal states of Maine, Brittany and Ponthieu are added in the number increases to around half a million.(55) Using the one percent maxim, duke William's total forces of enfeoffed knights maintained in the duchy and her environs would be 5,000 to 6,000 men. Normandy alone had nearly 3,000 enfeoffed knights.(56) Like the Norse and English, Normandy and the vassal states were able to support an army probably equal to four or five percent of the population for a rare major campaign. So a potential force of 24,000 to 30,000 men was not impossible for duke William to muster; especially since half were volunteers from lands other than his own, who came with their own arms. William's army that gathered at the coast of Normandy was unusually well equipped. It was not in any sense a feudal force. His own Norman vassals were the largest portion, probably over fifty percent of the whole, but they served because he had convinced them with his propaganda and promises of reward, not out of a sense of obligation to their lord. The volunteers were all heavy knights on horse, mostly light marksmen using bows or crossbows and other infantry armed in hauberks.(57) Promises, propaganda and his excellent abilities to meet the heavy logistical demands of a large army kept them together throughout the summer. Meanwhile, he drilled his mercenaries and volunteers with the ducal mesnes.

But his problem was shipping. Even with all his collecting and building of a large fleet, he lacked enough ships to transport his entire army in one passage of the Channel (the reinforcements came at the end of October). For the Hastings campaign, he packed his ships as full of troops as possible. Each knight had at least two horses; probably the lords had three or even four. Commentators reason this problem backwards and forwards to no avail. The evidence of the Bayeux Tapestry is not literal but artistic, and this is just as true where ships are depicted. The largest number of horses shown being transported is ten, the fewest is three per ship. The ships vary in size from the almost-realistically proportioned to the comically stubby. (The Tapestry has not been called the world's largest comic strip for nothing!) More sound evidence than the Tapestry's pictures of ships is a comment like William of Malmesbury's, that in 1142 the earl of Gloucester embarked c. 350 milites onboard 52 vessels: this works out to seven men and horses per ship.(58) But perhaps the knights had more than a single horse apiece, or some did, like the lords and privileged knights. No one argues that ten horses per ship is impossible, or even unlikely, but it is still guessing.

The problem of trying to calculate the size of William's fleet, using all manner of mathematical and logical methods, is not to be resolved. I believe we should go with the number for the fleet given in the sources which comes up the most consistently: make use of the advice, "in the mouth of two or more witnesses," and have done with the fruitless arguing. That number is either side of 700.(59) So we'll go with that.

Since ships of a supply-carrying nature varied from c. 45 to 55 feet in length, we will require an average-sized vessel of 50 feet: or one about the size of a knörr with 10 to 11 oars per side. If used to carry troops, this ship could accommodate up to 65 men -- including a couple of steersmen -- and would probably be constructed with oar tholes (as shown on some of the ships in the Tapestry). But Normandy was many years beyond her Viking days, and seafaring had become, if not a lost, then at least a diminished art for the Normans: Longships of any description we shall seek for in vain in the Tapestry or anywhere else. If we take the same proportion of supply ships to troop transports as we have for the Norse fleet (200 to 240 warships out of 300), William's fleet of 700 had some 518 ships to carry his men and horses. At 10 horses or 65 men per ship, this gives a total of 16,835 men and 2,590 horses -- after cutting both figures in half to "share" space for man and beast. But there is no question that the duke's assembled army had far more horses than this. If we allow for a balanced army, relying on missile power in conjunction with cavalry shock, then he must have embarked at least 3,000 to 4,000 horses. In light of the fact that William had two or even three horses killed under him during the battle, and that the remounts he used belonged to other knights and were not remounts of his own, it seems evident that the knights traveled "light" and left most (or all) of their spare horses behind, to come along with the reinforcements later. Each spare horse brought along in the initial crossing would mean one less effective cavalryman. If the duke himself did not bring along a spare mount, his troops certainly could not.(60) That the proportion of light infantry to heavy equipped with mail was about three to one is born out pictorially by the Bayeux Tapestry, which shows one mailed archer in a group of four. Of mailed spearmen the Tapestry shows none positively, but we assume from William of Poitiers that infantry who could throw spears (i. e. javelins) were there, and fought in the first assault with the marksmen.(61)

William's army of invasion differed from Hardrada's and king Harold's in another way: all of the Vikings, thegns, geneats and housecarles were effectives in war, but the same could not be true of William's host. He had to take along sailors, carpenters, cooks, smiths, servants and clergy -- this last being a very important "arm" because of his vaunted championship of Holy Church. It seems impossible to determine what percentage of his 16,835 men were warriors and what part were noncombatants: at the very least, he had a total of 2,800 sailors from his 700 ships, but possibly quite a few of them were also fighters. The same could be true of the cooks, grooms and so forth: many must have been men who served in camp and fought as infantry. All of the truly noncombatants would have been in the fortified camps at Pevensey and Hastings on the day of battle -- except, perhaps, some to manage the sumpter beasts with supplies of missiles, and those brave and dedicated clerics who wished to witness from a safe distance God's handiwork as they slaughtered each other. Taken together, they may have equaled a third part of the whole infantry. Any infantry left behind to garrison the camps would have been minimal. Some troops may have remained unaccounted for, since William had to call in his foraging parties during the night before the battle.

Having belabored the subject for far too long, let me say that the consensus allows for up to 10,000 or so men per side, but the popular stand to take is still with the smaller numbers. However, the foregoing analysis allows for a maximum of 12,228 effectives for the Franco-Norman army: mounted knights: 3,500 at the most and 2,500 at the least: and 4,703 to 9,728 infantry -- after subtracting for noncombatants and a couple of hundred foot soldiers each to garrison the two fortified camps.

Whatever the reasons for it, the two armies at Hastings were roughly even in terms of numbers.(62) Harold could have had his 13,000-plus there, but because of what Florence of Worcester says -- "inasmuch as the English were drawn up in a narrow place, many retired from the ranks, and very few remained true to him" -- he could not have kept his full strength. As for "many" retiring from the ranks, I picture them (mostly local levy) fading away back into the Andresweald, rather than follow the king out onto that barren ridge. Trying to pack 15,000 and more troops onto it -- including local peasant levy -- would have made it a "narrow place" indeed. The same source has just stated that Harold "(advanced) with all speed into South Saxony against his enemies. . .though he well knew that. . .one half of his army had not yet arrived:" 30,000-plus mustered is "half" of the kingdom's full military manpower of 56,000 to 70,000 thegns and their retinues. As for Harold joining battle "before a third of his army was in order for fighting:" this can be explained easily by the great numbers of fyrd struggling to arrive on foot: 8,000 to 10,000 present out of a known muster in excess of 30,000 is certainly "a third." Nothing in the wording of Florence of Worcester requires us to interpret what he said as meaning that only a third of the English troops on the battlefield were formed when duke William appeared; but were instead arriving -- mostly on foot -- throughout the day. Hasty to come to grips, Harold must own: but surprised by the invaders he was not.

The sources and my analysis argue for an English army of equal or slightly larger size. So, here follows my tentative order of battle:

The Franco-Norman army has 2,500 to 3,500 heavy cavalry, armed with sword, lance (which can be thrown) and a javelin; 2,500 to 5,000 marksmen, mostly unarmored infantry, using bows and one-fourth with crossbows; and enough infantry armed with sword, spear and a javelin each, to bring the numbers up to around 8,200 to 12,200.

The English army has 2,000 to 2,500 housecarles, heavy infantry armed with three javelins apiece,(63) sword, spear and two-handed axe; 2,000 to 5,000 thegns, mostly medium infantry, some heavy and some light, armed like the housecarles; 2,000 to 5,000 geneats, mostly light infantry, some medium, armed with two javelins each, and spears mainly, with fewer swords and two-handed axes; 1,000 to 2,000 peasant levy, maybe half with a shield, the rest with no protection; armed with one hand missile (a few throwing spears and axes, but mainly stones tied to pieces of wood) and the rudest collection of weaponry, short hunting spears and improvised farm implements predominating. Archers amongst the peasant levy would total no more than ten percent -- the usual reason given for few archers in Harold's army being that, as poor men, they have no horses and therefore cannot keep up with the fyrd.(64) The total English strength is somewhat more than that of the invaders.

My own collection of wargames miniatures follows the foregoing analysis -- at 64 men per infantry and 50 men per cavalry figure. Other variations are just as speculative:

The invaders:

Left wing, Bretons, Angevins, Poitevans; commander-in-chief count Brian:

768 marksmen; 1,152 heavy infantry; 800 cavalry. One command of three units which can operate separately of each other.
Total: 2,720.

Center, Normans of Normandy; commander-in-chief duke William:

1,408 marksmen; 2,048 heavy infantry; 1,550 cavalry. One command of 4 to 6 units which can operate separately of each other: the cavalry may form two or three units, one (or two) in line, with the reserve under the duke (which can be positioned behind, or in line with the rest); the heavy infantry form a single unit, or can be divided into two units; the marksmen are a single unit.
Total: 5,006.

Right wing, French, Flemish, Boulognnais, etc; commander-in-chief count Eustace:

640 marksmen; 896 heavy infantry; 700 cavalry. One command of three units which can operate separately of each other.
Total: 2,236.

Total invader army: 9,962.

The English defenders:

Right wing; commander-in-chief earl Leofwin:

512 heavy infantry housecarles
512 medium infantry thegns (some heavy and some light)
512 medium infantry geneats (some thegns)
512 light infantry geneats (some thegns)
512 light infantry geneats
512 light infantry ceorls
512 unarmored infantry ceorls
One command; one combined unit; housecarles remain with fyrd.
Total: 3,584

Center; commander-in-chief king Harold:

1,152 heavy infantry housecarles
1,024 medium infantry thegns
1,024 light infantry geneats (with some light infantry and medium infantry thegns)
128 unarmored infantry ceorls
One command; housecarles in line remain with fyrd; but housecarles of Harold's reserve/bodyguard are a separate unit.
Total: 3,328

Left wing; commander-in-chief earl Gyrth:

512 heavy infantry housecarles
512 medium infantry thegns (some heavy and some light)
512 medium infantry geneats (some thegns)
512 light infantry geneats (some thegns)
512 light infantry geneats
512 light infantry ceorls
512 unarmored infantry ceorls
One command; one combined unit; housecarles remain with fyrd.
Total: 3,584

Total defender army: 10,496.

References and commentary:

36. Ibid, page 31.

37. Ibid, page 181. Richard Glover: ". . . .Harold Godwinson took a very large army to Stamford Bridge. . . .Saxon, Norman, Scandinavian and other continental sources are . . .strikingly unanimous in the impressive picture they paint of the great size of Harold's forces at Stamford Bridge. 'The whole strength of southern and central England took part in that great campaign,'" he says, citing Freeman. No such claim is made for the Norse army in any of the sources.

38. The Gulathinglaw, quoted in The Vikings, Ian Heath, Osprey Publishing Ltd 1985, page 7.

39. King Harald's Saga, page 139, fn 2. The ASC E version gives Hardrada 300 ships, SM 23. Naval Warfare Under Oars, 4th to 16th Centuries, W. L. Rodgers, Naval Institutes Press 1988 ed, pages 69-88: At one man per oar, plus two steersmen, 240 longships -- with 118 x 25-benchers -- produces an army of 11,260, not counting Tosti's reinforcement or the c. 100 supply ships. The ASC D and E versions state that Tosti had only "twelve small vessels" with him: the crews, Frisians, Flemings and men of Orkney, would have numbered between c. 384 to 1,152, assuming an average ship of 12 to 18 benches (too small to be classed as a "modern" warship), with one to three men per oar. Hardrada's longships would not have had a full war complement of two or three men per oar, because of the need to carry supplies, and because for open sea voyaging Viking crews relied on sail power: not to mention the fact that the army was making room for plenty of loot on the return journey! Hastings 1066 The Fall of Saxon England, page 43, shows a knörr, typical of both Hardrada's and William's supply or troop carrier ships. A minimal crew of four to six men would suffice, but up to 32 men each would be possible with supplies aboard.

Levies this big could not stay away from home for extended periods or else the economy would suffer, both from neglect and from the need to maintain so large a body with pay and provisions. An army this large would expect to be paid in booty, and rather quickly too. In this period, the greatest danger was not the drain on the economy posed by an unprofessional army larger than one hundredth part of the population: it was the vulnerability the realm would suffer if the army took heavy losses. Hardrada's defeat hurt his people at home and freed Denmark's king Swein for years to come to expand his own interests without fear of intervention from Norway. Hardrada's son king Olaf was known as "Olaf the Quiet" for good reason. See the last page in King Harald's Saga.

40. Richard Abels, in SM, page 74 fn 4.

41. Nicholas Hooper, Anglo-Saxon Warfare on the Eve of the Conquest: A brief Survey, Proceedings of the Battle Abbey Conference 1978, estimates 2,500 employed by Canute who maintained some 40 ships. Edward disbanded his fleet in 1051, but must have retained housecarles to the end of the reign, because he sent a force of them, which included "a large fleet," with earl Siward to take out king Macbeth in 1054 -- Richard Abels, in SM, page 77. Harold Godwinson, his brothers and the other earls each employed their own housecarles. An earl's unit of these mercenaries and sworn companions probably numbered around 250 to 300: Tosti's 200 men killed in York were housecarles, and some survived. -- Ibid, page 76 fn 15. Earls Edwin, Morcar and Waltheof had similar-sized households. King Harold maintained a large fleet, and his housecarle following has been estimated at 3,000: presumably lumping the housecarles he already had as earl of Wessex together with those he assumed with the throne, i. e. the royal housecarles -- Ben Levick, Who Were the Huscarls? Regia Anglorum Publications 1995; but this is probably too high. I have opted for a total from all sources of 4,000.

42. Richard Abels, in SM pages 60-61. "If we examine the duties enumerated in chapter 2 (of the Rectitudines, an early 11th century work), we find that the relationship between the king and his thegn paralleled in some respects that which existed between the thegn and his geneat. It would seem that the author of the Rectitudines viewed bookland (land tenure by royal decree) as a privileged but still dependent form of tenure. The thegn was the lord of the estate, but he held his land de rege, rendering service to the king for it, just as his own man, the geneat, held of and rendered service to him."

43. Edward Gibbon, The History of the Decline and Fall of the Roman Empire, Folio Society, vol 1 page 113.

44. N. J. G. Pounds, An Historical Geography of Europe 450 BC - AD 1330, Cambridge University Press 1973, page 244. The Domesday Book, England's Heritage Then and Now, 1985 CLB International, page 17.

45. Freeman, Norman Conquest, iii (2nd edn), 362.

M. K. Lawson also puts forth very convincing arguments for a much larger English army. For instance, the Bayeux Tapestry's "water course" scene, and the Carmen's description of Harold's position as occupying "not only a mons (hill) and rough ground, but also a vallis ('valley')", as evidence for the English army being large enough to cover the entire length of the ridge and at the same time have the right flank anchored on the low marshy ground to the west. The Battle of Hastings 1066, Tempus Publishing 2002, page 143.

46. Abels in SM page 74 fn 4. His use of the word fyrdmen implies that these warriors who held tenancy over the hides were the only members of the fyrd. He calls them "soldiers" without trying to differentiate between thegns or wealthy ceorls, or geneats.

47. The Anglo-Saxon Fyrd 878 - 1066 A. D., Ben Levick, Regia Anglorum Publications 1995. "In addition, the heriot, that is the death duty paid to a lord when a thegn dies, was set at four horses (two with saddles), two swords and a coat of mail. Since the heriot represented the return of the gifts of a lord to his retainer, we can see that this was the equipment a thegn would be expected to possess. That the thegns did possess this equipment is borne out by the fact that, although it was possible to commute this payment to cash, the payment was almost always made in the form of these arms." But Richard Abels says the heriot debt was two horses, one saddled, SM page 69.

48. Carmen, page 23.

49. William of Poitiers in SM, page 12.

50. Ben Levick, The Anglo-Saxon Fyrd: "The reason for the large number of horses may be explained by the fact that it represented a mount and remounts or pack-horses for the thegn and a retainer who would look after the thegn's horse while he was in battle, or carry messages for him, etc."

51. SM, page 24.

52. Page 142. It seems unlikely that Hardrada would leave a third of his forces to guard his ships after his victory of the northern earls, and not do the same before his victory when his strategic position was in doubt.

53. Hastings 1066 The Fall of Saxon England, page 81. "It is said that almost 15,000 men perished there," van Houts, pages 168-169: Orderic Vitalis, using William of Poitiers as his main source, like all the Anglo-Norman chroniclers of the 12th century, describes the slaughter at the end of the battle, as they fell foul of the 'malfosse' while pursuing the fleeing English. William of Poitiers gave 60,000 as duke William's army strength, SM page 10. Other sources speak in language which paints a picture of great carnage and slaughter amongst the victorious French: William of Malmesbury: ". . . .by frequently making a stand, (the English) slaughtered their pursuers in heaps. . . .they slaughtered them to a man. . . they trod under foot such a multitude of their enemies . . .that they made the hollow level with the plain, by the heaps of carcases." William of Jumičges: "The battle began in the third hour of the day, and continued amid a welter of carnage and slaughter until nightfall." ASC: ". . .there were heavy casualties on both sides." Florence of Worcester: ". . .after great slaughter on both sides. . .the king . . .fell." Carmen: ". . .by measureless slaughter, Harold was forcing the masters of the field (the Normans) to go the way of (all) flesh."

54. Bradbury, page 165: "It is probable that the battle had not been quite so prolonged as later sources said. The nature of it, with the surprise attack resulting in victory, normally would speak of a relatively brief conflict. It had been prolonged by the arrival of reinforcements from the coast, but the English army must have escaped without enormous losses. Had Stamford Bridge been too damaging on Harold's men, he would not have been able to contemplate another battle. The signs are that the victory had been so great that few men were lost."

I wish to add here an aside, apropos of Snorri's account of the battle of Stamford bridge. Let's be fair: Snorri's narrative of the battle seems like a twisted version of Hastings, because it is in such close proximity to the more famous battle; but would an Icelandic poet care two pins about a Norman victory? Would no oral tradition in the north exist describing Stamford bridge? Would anyone narrating the battle have to make up everything out of whole cloth, or crib from Hastings? In fact, Snorri's account of Stamford bridge does not follow Hastings in sequence nor setting. His tactics for the English mounted archers and javelin-men are also not at all like the Norman cavalry at Hastings, who, "Disdaining to fight from a distance. . . rode into battle using their swords." Snorri says nothing at all about hand-to-hand weapons for the English cavalry, but rather: "The English cavalry kept charging them and falling back at once when they could make no headway." Neither does he suggest a feigned flight to lure the Norsemen out, but rather feints at charging are employed, hoping to make the Vikings break and run. They did not oblige, being wise to that trick. English horses were not war-horses nor trained to close combat and Harold did not use them that way. So the battle continued with "only skirmishing. . .so long as the Norwegians kept their formation." This shows clearly that there had been no close combat fighting thus far: therefore, no cavalry charges to break through the shield-wall as the Norman chivalry attempted repeatedly at Hastings. The last time Snorri mentions mounted troops at all is when "the English rode down on them from all sides, showering spears and arrows on them." This was in immediate response to the Norsemen impatiently breaking out of their static defensive formation and launching an attack. It reads as though the Norse were trying to reach the mounted troops who had been skirmishing before them, but it must mean that Hardrada moved to attack the main English body of infantry, which was standing passively nearby and letting their mounted brethren continue to harass the Norse. After that the battle is all hand-to-hand and no horses are mentioned. The cavalry unit -- at least the armored men -- had dismounted and entered the fighting on foot beside the main English force. Had they retained their horses, surely they would have fled when: "It looked then as if the English were on the point of being routed." (Note, that earl Ralf the Timid in 1055 -- ASC page 181 -- tried to make his men fight on horseback: but was this an experiment in the continental tactics Ralf was familiar with? The English routed "before any spears were thrown." This reads more like disaffection with the earl, rather than unfamiliarity with skirmishing forms of mounted warfare. He was prepared to fight an opening skirmish action, like his fyrdmen were familiar with; but because they wanted to be elsewhere in the first place, having horses made this easy and they abandoned the field forthwith.)

55. Using the 1901 census for France, we find that the population is c. 40.7 million. This figure is before the emigration and shift to the industrial districts that occurred beginning in the first quarter of the 20th century: I don't have earlier population statistics, but assume that one hundred years earlier the relative distribution of the population would have been virtually the same as in medieval times. The districts of Normandy and her medieval vassal states are c. 3.8 million, or 9.4% of the entire nation: Normandy herself is only 4.7%. Calculating on the same percentages of the whole -- assuming that in such an old region of settlement as France is, where population ratios tend to remain constant as the population total increases -- and assuming around 6.2 millions for late 11th century France, we arrive at the above numbers. The exact numbers are, for Normandy: 291,400; for the total of Normandy and vassal states: 582,800. But this includes the county of Boulogne (in the department of the Somme), which must therefore be discounted.

56. But only a fraction of this number shows up in the servita debita of Normandy: 10 were owed to the king of France from his vassal the duke of Normandy, but the duke could demand 1,500 knights to serve him from his own vassals. Beeler, Warfare in Feudal Europe, Cornell University 1971, page 37. Douglas, William the Conqueror, pages 101-103 and 282-283, shows that in Normandy, at least after the Conquest, the number of knights owed to the duke was far less than the number maintained by each of his tenants.

57. William of Poitiers, in SM pages 12-13. Bayeux Tapestry: The light English geneats already mentioned, most of the marksmen of the Norman army, a lone English archer, a ceorl being executed by an unhorsed knight, peasants (or Norman noncombatants?) looting the dead, and the fugitives fleeing into the Andresweald at the end of the battle are the only light-armed : the preponderance of warriors on both sides are all armed exactly alike, horse and foot, in conical helms with nasals, mail armor and carrying shields, mostly kite-shaped, but with round or oval shields present too.

58. C. M. Gillmor, in SM page 117.

59. The Brevis Relatio ship list, which shows the quotas William extracted from his magnates gives a total of 782 (776) ships, and "expressly (states) the fleet as 1,000, a number suggesting the presence of smaller vessels," SM pages 114-115. And the Roman de Rou of Wace claims that his father reckoned the whole fleet at "700 ships less four," Bradbury, page 136 en 41. But the fleet lost ships making for the Somme estuary. Could the discrepancy between c. 700 and the ship list be explained by those ships which were wrecked?

60. The Carmen has William forcibly take his second horse from a craven knight of Maine, who ignored the duke's command to dismount like a loyal vassal so that his lord could ride; and then after having that horse killed under him too, count Eustace gave up his own horse so the duke could remount (the count's sacrifice was made up by one of his knights giving over his own horse to Eustace); Carmen, pages 30-35. The claim that William had three horses killed under him is put forth by William of Poitiers (SM, page 14), who seems to have used the Carmen as a source for his own sycophantic history of William the Conqueror, and augmented material in the Carmen to magnify William's achievements and personality.

61. SM pages 12-13. William of Poitiers describes troops using "spears from a distance." The knights can, if they wish, also "fight from a distance;" but in the battle they "disdained" doing so. If one accepts the evidence of the Bayeux Tapestry and William of Poitiers -- the two most viable and original sources -- then the proportion of light infantry to mailed was three to one: but these are nowhere described as spearmen. Poitiers states clearly that the knights "came to the rescue" of the foot, who opened the battle with missile weapons; which he has just described as bows and crossbows. He only mentions men "(throwing) spears. . . from a distance" after reiterating that the knights were the ones who "dared to attack (the English) with swords." So the reference to spear throwing would be more applicable to those knights who refused to remain up close to the melee and preferred to fall back and throw their lances as javelins. Modern commentators on Hastings have construed William of Poitiers' narrative to postulate spearmen in the Norman army: but no other source even alludes to them at all. The Tapestry gives full coverage to the archers: and later 12th century historians mention the use of bows and arrows (especially Henry of Huntingdon); the Carmen gives emphasis to the presence of bows and crossbows, even differentiating between their destructive power. But in the original sources the only troops mentioned or depicted are knights and marksmen. ("Florence" - John - of Worcester mentions "foot-soldiers" in addition to slingers and archers; but this is a 12th century Anglo-Norman source, apparently based on a no-longer existing version of the Anglo-Saxon Chronicle; ergo not an original source per se.) This could be a case for playing out the battle as a war-game with only these two arms. It's something I have never done. But if I did so, I would make one-fourth of the marksmen "more steady and armed with hauberks in the rear" where Poitiers places them: i. e. a second line of marksmen to back up the first attacking line. If there were spearmen, they were not necessarily heavy infantry only, but also may have included lightly armed; and the marksmen must have outnumbered the spearmen, if one considers the Tapestry's lack of pictorially showing Norman spearmen at all, and the large numbers of bowmen shown. The first group of four bowmen is the only main panel depiction of troops positively not part of the Norman chivalry (a couple of figures might be unhorsed knights or heavy foot using swords).

"The evidence for the Norman period strongly suggests that the normal infantry weapon was the bow." Armies and Warfare in the Middle Ages -- the English Experience, Michael Prestwich page 129. In fact, there is the distinct possibility that the later archery tradition in England under the Plantagenet dynasty received its genesis with the Norman Conquest.

62. "(Wace) says that in his day many have explained the defeat by saying that Harold had a small force, but others say, 'as do I, that he and the duke had man for man', and adds that William had more knights and more archers." Bradbury, page 192.

63. Note the "bundles" of spears/javelins carried by the English in the opening scene of the battle in the Bayeux Tapestry. I take the number literally, and allow three spare javelins: the fact that one of the two figures holding "bundles" of javelins is actually throwing a francesca merely emphasizes the presence in the English army of many hand missiles. If both "bundle"-carrying housecarles were depicted with throwing axes, I would have considered this as sufficient evidence of a tactical form and armed housecarles with three javelins and one throwing axe. In fact, the francesca, because of the manner in which it is thrown, actually has a longer effective range than the javelin. The spear -- which could also be thrown, but was usually retained -- was the main weapon of the English army, not the famed two-handed axe. The axe or sword was only resorted to when the combat spear was broken or otherwise lost. Note also, that in the rest of the scenes, no English troops are shown with more than a single spear apiece. The proportion of axemen to spearmen in the opening battle scene is 2 to 15: but in the last battle scene, just before the final rout, the four housecarles still on their feet are using two axes and one sword and one spear. This is a neat little accuracy of detail, graphically showing how the spears had nearly all been broken and thrown by the end.

On the use of the English axe, see Mark Harrison, Anglo-Saxon Thegn 449 - 1066 A D. Osprey 1993, page 45. It is easy to visualize here what the author suggests. Those warriors most brave and skilled would step forward from the front ranks of the shield-wall, where they could wield their axes with murderous effect; while at the same time, the spearmen behind covered the axemen and defended them from counter-attacks while they recovered themselves -- the follow-through always left the axeman vulnerable. Such combined arms would be effective: the axes to shatter shields or behead horses, while at the same time the hedge of spears worked to dissuade cavalry and guarded the axemen. The natural evolution of these cooperative tactics led soon to the adoption of various pole-arms, combining the features of axe and spear into a single weapon.

64. And yet we see mounted English archers at Stamford bridge, according to Snorri Sturluson -- writing realistically for his 13th century audience. So where are they at Hastings? Perhaps they are the few inferred by the single archer figure shown in the opening scene of the battle in the Bayeux Tapestry. Certainly, not even counting the archers in the lower margin, the English archers were outclassed by the firepower of duke William's army by at least four to one. Later, the Andresweald produced a peasantry of bowmen (Bradbury, page 103). But there is no evidence for them in Anglo-Saxon times. (David Howarth, 1066 the Year of the Conquest, page 105, claims that the bow was the sole possession of the aristocracy and a "strictly guarded mystique." This would explain the rarity of archers in English armies; but he does not substantiate his view with any references, and others disagree: Jim Bradbury, pages 101-104, goes into various reasons why archers were lacking at Hastings and never mentions the restriction of bows to the nobility: rather, he postulates that, since most of the very rare pieces of evidence mentioning Anglo-Saxon archery are from the north, archery was regional: bows were rare elsewhere in England.) Such archers as king Harold had with the fyrd all remained on horseback to open the battle of Stamford bridge, thus comprising a noticeable proportion of the mounted unit. They must have been no more numerous, however, than 200 men. At Hastings, their feeble return fire was negligible.