by Paul Hill
Law code 51 of King Ine of Wessex (688–726) spells out the consequences of neglecting fyrd service: ‘If a gesiðcund man who holds land neglects military service, he shall pay 120 shillings and forfeit his land; [a nobleman] who holds no land shall pay 60 shillings; a cierlisc [free peasant] shall pay 30 shillings as penalty for neglecting the fyrd.’ This law is thought to reflect the structure of the earlier Anglo-Saxon armies in that when a king called out his fyrd it would consist of his own landed companions, those who owed service to these men through some other gift exchange, and the free peasants who served them.
However, by the eighth century the rise in the power and wealth of the Church had begun to throw a spanner in the works. The Church required its land in perpetuity and not just for the lifetime of the king who gave it. A system came into being called Bookland. This grant of land to the Church came with no obligations from the Church to the king except, it seems, for the securing of his soul in heaven. But for the king, it meant that the land was permanently removed from his financial and administrative system and given to his ultimate lord, God himself.
Bookland was a problem that took time to materialise but which had inevitable consequences for the kings of middle Saxon England. The more land you gave away free from military service dues, the smaller your warrior base became and potentially your kingdom would become vulnerable to other kings or raiders. The answer to the problem was found in the kingdom of Mercia. King Æthelbald (716–57) decreed that churches and monasteries could hold their land free from all dues except bridge building and the defence of fortifications. The mighty King Offa (757–96) extended this to the all-important fyrd duty. Thus, the three necessities, or ‘common burdens’, were founded. These burdens became the bedrock of military recruitment across the Anglo-Saxon kingdoms of England until they were modified by Alfred the Great (871–900) in his grand reforms of the ninth century.
At around the same time as the Bookland issue was being resolved there was a change in terminology being employed. The duguð was a term in decline and seems to have been replaced by the word ‘thegn’, which soon came to be a widely used term across the country. The thegn held the Bookland from the king in return for military service. The nobleman who held the large estate known as the ‘scir’, or ‘shire’, would be the ealdorman.
So, the nature of recruitment on the eve of the Viking invasions at the end of the eighth century was fairly straightforward. The armies of Anglo-Saxon England ranged from small warbands of local thegns to larger more territorial units commanded by ealdorman who were protecting their shires. Above this was the royal host called out by the king himself which in effect was made up of numerous groups of thegns, ealdormen and their retinues. All of the constituent parts of these forces came to the battlefield through the duties imposed upon them by lordship ties bound by a mixture of gift giving and land tenure.
The activities of the larger hosts throughout this period are well documented since they are the instruments of the king. Wars between Saxon kingdoms such as Wessex and Mercia, or between Mercia and the Welsh kingdoms, feature in the Anglo-Saxon Chronicle. But when Alfred the Great first fought the Great Heathen Army alongside his brother King Æthelred I, the chroniclers make a revealing statement about how armies were then organised. The entry for 871 says that during that year there were nine ‘general’ engagements between the West Saxons and the Danes. This figure did not count the expeditions that the king’s brother (Alfred), ealdormen and king’s thegns often rode on. The message here is clear: senior noblemen, ealdormen and king’s thegns were each capable of mounting their own military expeditions outside of the activities of the royal host. The king’s host would of course take weeks to gather together. Importantly, the chronicler mentions that these smaller forces ‘rode’ on their campaigns.
So, how did military provision change during Alfred’s reign? The answer lies in the years after the watershed Battle of Edington (878) when Alfred finally rid himself of Guthrum the Dane in a famous encounter. After Edington, over the next two decades, Alfred instigated a series of fortified places throughout his kingdom resulting in an entirely new type of defence: a defence-in-depth system in which no two forts were more than around 20 miles apart. The impact of these forts and how they were manned is examined below (see pp. 88–93), but the most important point of the reforms was that the fyrd was organised into a three-part system.
We are told by the Anglo-Saxon Chronicle in the entry for 894 in a passage that describes the strategic manoeuvring against the renewed Viking threat in south-east England, that ‘The king had separated his army into two so that there was always half at home and half out, except for those men who had to hold the fortresses. The raiding army did not come out in full from those positions more than twice.’ Ample evidence, then, for a system that had its obvious benefits. With a strategically gifted commander such as Alfred, and with numerous manpower resources spread deliberately over the landscape, the new kingdom of the Anglo-Saxons was to be a well-defended place indeed. This kingdom was born out of the combination of Wessex and the English parts of Mercia. Eventually, it would expand under King Athelstan (924–39) to become the Kingdom of the English. The new mechanism meant that there was always one part of the force at home on their estates, another in the field at a state of readiness and a third part on fortress duty. They could act in coordination with one another or mount their own expeditions. It was a plan based on Alfred’s own observations of the Carolingian model, which in itself probably borrowed from a papal Italian arrangement for the defence of Rome. The point is that it worked. Despite some notable problems when one Anglo-Saxon force’s tour of duty had expired before its replacements could take over from it–as happened in 893 at Thorney Island –it remains the case that the Anglo-Saxon armies of southern England were vastly better organised than they had been when the Vikings first descended upon Wessex.
Alfred’s reorganisation still required strong leadership to put into action. Moreover, one fundamental thing remained unchanged and that was the basic make-up of the fyrd. It still comprised the same people. The cost of all this, however, was immense. Inevitably, the system was doomed to be a hostage to neglect. After the reign of King Edgar (959–75) during which the Anglo-Saxons had achieved an unprecedented level of power in the British Isles, the recruitment system in part fell back upon makeshift levies based on land tenures measured in hidage. The grand fortification schemes employed by Alfred, his son Edward the Elder (900–24) and his grandson Athelstan (924–39) had in places fallen into disrepair, leaving England vulnerable once again to Viking attack. This is not to say that King Æthelred (979–1016) did not attempt to rectify the situation–in fact, far from it. Æthelred’s attempted reforms, particularly in respect of naval provision, were ambitious to say the least (see pp. 70–1). Between 1008 and 1013 the king instigated reforms which included the construction across the country of a navy of around 200 ships and the provision of mailcoats for thousands of warriors. However, all the time the king had to pay increasingly burdensome sums of money to pay off the Danes and soon he even took on a Danish contingent of his own which required feeding and provisioning.
The 5-hide unit is usually associated with a thegnly rank, this being an amount of land that qualifies its holder in that rank. However, it is clear from Domesday Book that there was great variety in the land-based obligation, varying wildly from region to region. A nobleman holding just a few hides might be required to provide a warrior, while an estate of well over 5 hides might only be obliged to provide just a single warrior. It depended on the nature of the arrangement between the landholder and the king. If an estate had to provide more than one warrior, the head of the estate would need to recruit from his tenants or find a way of replacing what he could not provide with a money payment so the king could hire a mercenary.
What seems not to have changed was the bond through lordship that drove the obligation. As if to reinforce this notion at a time when service seems to have been widely based
on land-holding arrangements, the Danish king of England Cnut (1016–36) came up with a stringent law for deserting one’s lord:
Concerning the man who deserts his lord. And the man who, through cowardice, deserts his lord or his comrades on a military expedition, either by sea or by land, shall lose all that he possesses and his own life, and the lord shall take back the property and the land which he had given him. And if he has book-land it shall pass into the king’s hand.
By the time of the Battle of Hastings, it is still the case that King Harold’s army would have been recruited largely through the bond of personal lordship ties. Land tenure based on the 5-hide unit certainly came into it, but so too did the royal purse and the provision therein for mercenary hire (pp. 63–6). While not as rigidly organised as the armies of the later Alfredian period, those of 1066 were just as capable of providing results in the field.
What do we know, however, of how long it took any of these armies to gather a force into the field? We have precious little evidence to answer this question, but what little we do have points to the reasons why it was so easy for Scandinavian coastal predators to achieve so much in a short space of time. We know from the 871 Anglo-Saxon Chronicle entry that units of mounted men under the command of local nobles were able to put into the field almost immediately, but were small in number. Larger hosts, however, took longer to gather. Alfred’s famous call to arms of 878 issued from Athelney to the remainder of his core supporters in surrounding shires bore fruition in the seventh week after Easter, but it is not certain when the call went out. It was a nervous wait for Alfred, but when the men of Somerset, Wiltshire and parts of Hampshire answered his call, they joined him at his camp at Egbert’s Stone and were ready to march with him the next day to Illey Oak and thence to Edington.
The Danish assault on Norwich in 1004 gives a clearer idea. The Danish force seems not to have been of a size that any mounted rapid-reaction-style force could deal with. Instead, Ulfcytel Snilling, the local East Anglian leader, was obliged to play for time by buying peace from the army while he raised his force. The Anglo-Saxon Chronicle is specific that Ulfcytel had not had time enough to gather his own army at this point. The Danes stole away to Thetford and Ulfcytel’s order for their ships to be broken up fell on deaf ears. It was three weeks since their first assault on Norwich when the Danes torched and sacked Thetford. The following morning they prepared to return to their ships when Ulfcytel’s East Anglians fell upon them, forcing a battle. It was a hard-fought battle which Ulfcytel ultimately lost. However, it is explicitly stated that had Ulfcytel’s numbers been up to full strength the Danes would never have made it back to their ships. So it would seem then that three weeks is barely enough time to raise a sizable force to match that of the invader. This gives an idea as to the desperation of the year 1016–a year of endemic warfare in England–when King Edmund Ironside was compelled to call out the ‘national’ host at least five times for extensive campaigning.
The year of 1066 provides its own clues, although we have to be mindful of the possibilities that Harold had with him at all times a household force of a size enough to deal with all manner of problems, probably in the form of his own and his brothers’ retinues and his Danish mercenaries. Nevertheless, he decided to disband his army in the south of England on 8 September 1066, but then heard of the Norwegian invasion in the north and arrived at Tadcaster on 24 September 1066. Similarly, on his dash to London–a journey of some 190 miles–the passage of time is around two weeks, but it is clear from the sources that this was not enough time for a full host to be properly gathered. Harold’s reasons for his actions and his failure to wait at London for reinforcements may be perfectly justified based on what his strategic thinking was at the time, but once again the lesson is that two or three weeks is barely enough for a full host to form.
Quite how the fyrd was mustered is another question that puzzles historians. Messengers will certainly have been used, but did they use any devices to show the king’s will? It has been suggested from very little evidence that token wooden swords may have been used to symbolise a summons to the host. This has been based on finds of such items, sometimes with runic inscriptions from Frisia, Denmark and the Low Countries. The era is not the same (these being late Roman Iron Age or Dark Age discoveries), nor is it directly the same culture, but the intriguing possibility remains that the Anglo-Saxon messenger, when arriving at the residence of our thegn who was to prepare for war, may well have left a physical token of his visit.
The Size and Structure of Armies
The size of the armies of Anglo-Saxon England has been a subject of controversy for many years. It is generally thought that during the Migration period of the fifth and sixth centuries the average size of a warband could scarcely have numbered more than a few hundred. This much is gleaned from the historic and archaeological evidence from what is known of the Danish late Roman Iron Age Continental bog finds of warrior weaponry and what is drawn from the words of renowned ancient Roman authorities such as Tacitus. As early as the late seventh century King Ine of Wessex (688–726) was moved to categorise numbers of armed men: ‘We call up to seven men thieves; from seven to thirty-five a band; above that is an army.’ There is not a great deal that can be deduced from this code, however. Ine’s army was in fact a here. It is probably the case that fyrds and heres of the eighth century were no larger than an extended warband, perhaps numbering just a few hundred. Not until the later Anglo-Saxon period do things take on a quantifiable dimension, if only to confound us with the probability that we are only ever looking at an inherent manpower capability of a kingdom that probably never put all its available men in the field at any one time. Except, that is, for the post-reform period of Alfred’s reign.
A good example is the document known as the Burghal Hidage. Datable probably to the tenth century (see pp. 88–93), it gives specific garrison strengths for thirty-three strongholds built across Alfred’s Wessex and parts of Mercia. It even gives a formula for calculating the garrisons in term of the length of fort’s walls and the numbers of hides providing men to garrison them. The upshot of all this is that across his kingdom Alfred had a burghal garrison strength of around 27,000 men. The figure for all the land south of the Thames was probably larger in reality because the Burghal Hidage does not list the obviously cooperative Kentish strongholds such as Canterbury and Rochester, nor does it cover Cornwall. Moreover, Alfred’s kingdom of the Anglo-Saxons expanded into Danish Mercia during the tenth century to become the Kingdom of the English under Edward the Elder (900–24) and Athelstan (924–39). By the time of King Edgar (959–75), whose kingdom stretched from the south coast to the borders of Scotland, it must surely be the case that the king and his regional ealdormen were capable of raising forces in the field whose numbers were at the very lowest in their thousands.
Writing about the famous events in London in 1016, the German Bishop Thietmar of Merseburg claims he heard there were 24,000 ‘byrnies’ (mailcoats) in London. Even he thought this was incredible, so it is unlikely to be a casual exaggeration. Quite what has happened to all these mailcoats is a matter only for speculation (see p. 171). But the most obvious place to go if we want to make a guess at the near un-guessable, is to look at the Domesday Book and throw around a few figures. Judging from what we have shown so far about the apportionment of land and its division into hides, we might note that the Domesday Book records a total of 80,000 plough teams in the kingdom of England. The duty imposed by Æthelred II (979–1016) in 1008 requiring a ship from every 300 hides and a helmet and byrnie from every 8 hides would indicate on the face of it a ‘royal’ navy of 267 ships and an army comprising around 10,000 mail-clad warriors. Add to these armoured men their lesser armoured retainers at a ratio of say 2:1 and you have an army of 30,000 men. But even this figure is conservative given our knowledge of the Alfredian garrison strength of Wessex alone. Quite how many of these men took to the field at any one time when required is another matter. The actual numbers in any on
e army of the period must therefore have varied wildly.
One last observation on numbers goes to the Battle of Hastings. Here, the English army is harder to quantify than that of the Normans. The Norman army is well recorded and much research has been carried out as to the likely numbers who came to England in September of 1066. These figures, which are based on calculations relating to the number of recorded ships, the actual available men in Normandy and William’s mercenary contingents, in most people’s estimations amount to a force of between about 5,000 and 7,000 men. Although it is the result of educated guesswork, the consensus has been that the division of Norman forces was probably along lines of around 2,000 cavalry, 800 archers, 3,000 infantry, 1,000 sailors and logistical support men.
But what of the Anglo-Saxon army at Hastings? William apparently was told that he was likely to be swamped by English numbers when Harold’s army arrived. The English army was led by Harold and his brothers Leofwin and Gyrth, each with personal retinues numbering probably in their hundreds. We cannot be sure how many Danes were in the army, although it is suggested by William of Poitiers that these were ‘considerable’ and had been sent by the king of Denmark to help the English. To the Danes we must add the numbers of men gathered (or at least warned to attend the host) along the way. While these may have taken time to join Harold’s force and get to London, there is mention in the twelfth-century Robert Wace’s account of the regions from which the king recruited his force. These comprised the men of London itself, Kent, Hertford, Essex, Surrey, St Edmunds, Suffolk, Norwich, Norfolk, Canterbury, Stamford, Bedford, Huntingdon, Northampton, York, Buckingham, Nottingham, Lindsey, Lincoln, Salisbury, Dorset, Bath, Somerset, Gloucester, Worcester, Hampshire and Berkshire. We cannot be sure how authentic this evidence is, or whether each of the regions answered their call. It does not represent the entire ‘national’ host, either, which may be why William of Malmesbury suggested Harold’s numbers were not as high as they might have been. However, if we assign an unlikely low figure of 200 men per place mentioned, we come up with a force of 5,400 men before we have even added our Danes and earl’s retinues. It is small wonder then that when William of Poitiers (borrowing his style from the ancients) described the English army emerging from the Wealden forest to the north of Senlac Ridge he said, ‘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. For from every part of the country large numbers of English had gathered.’ If the numbers themselves are difficult to pinpoint, the evidence outlined above must surely suggest that the Anglo-Saxon military system, however allegedly archaic and obsolete by 1066, was capable of sending a force of up to 10,000 or more men to Senlac Ridge that October morning. We should not underestimate this capability.
