Castle Warfare in the Gesta Stephani
By Sarah Speight
Chateau Gaillard XIX: Actes du Colloque International de Graz, 1998 (2000)
The reign of the English king Stephen (1135-1154), popularly known as `the Anarchy’, is a fruitful source of study for historians and castellologists, providing evidence for the deployment of `adulterine’ castles and the conduct of siege warfare. It is a reign that has been considered by several recent authors, including most notably Matthew Strickland.1
Stephen succeeded his uncle, King Henry I of England, in 1135. This was a controversial succession seeing as Henry had a daughter living, Matilda, to whom his barons had sworn three oaths of allegiance. Matilda, furthermore, had an infant son by the time of her father’s death (the future Henry II) and so it could be argued that she should have inherited the throne if not in her own right then at least as reigning guardian for her child. Initially, the kingship of Stephen was accepted by the majority of lords in England and Normandy. However, as the reign progressed. the concerns over his claim to rule were used as excuses for dissent. By 1138 this dissent was out in the open with the defection of Earl Robert of Gloucester, eldest illegitimate son of Henry I, to the party of his half-sister Matilda, and her hus–band Count Geoffrey of Anjou.
In September 1139 Matilda landed at Arundel on the south coast and the war for the throne of England began. It continued, on and off, until late 1153. By the end, many of the key protagonists were dead and others had refused to fight.2 The `anarchy’ was over. But, it would be wrong to imagine a country torn in two by bitter factional fighting. First, this was a war of borders; conflict was very much restricted to particular disputed zones and was not endemic. Secondly, this was no free-for-all; there were rules of warfare in place. Whilst it would be ambitious to claim that there were `laws’ of war acknowledged and enforced by both sides, there were certainly customs, acceptable and unacceptable standards of behaviour, and set routines, for instance, in siege situations.3
These customs and standards of behaviour litter the pages of the most significant source for the reign: the Gesta Stephani. The Gesta spans the entire reign of Stephen and shows us a war dominated by sieges – even the battles tend to follow on from sieges, as at Lincoln in 1141. A general, although not unchallenged consensus is that it is the work of Robert of Lewes, Bishop of Bath and Wells, who travelled with the king in the 1140s and who wrote up the early years in about 1148 and the concluding section after 1153.4
The Gesta and other contemporaneous accounts of 12th century warfare can be used to support the current view within castle studies that the status/symbolism of a castle was more important than its military role.5 The castle was a deterrent, the medieval Polaris missile, designed to be physically used only as a last resort. To this end, an elaborate ritual of warfare evolved to lessen confrontation.
2. Siege Warfare
Accounts of the sieges of Faringdon, Bedford, Exeter and Wallingford in the Gesta reveal that to hold a castle against the king was to challenge his lordship, a challenge that could not go unanswered or the king’s prestige would suffer (his subjects might be suffering for instance) and men might transfer their allegiance to his rival. If the castle was well-stocked, the attacker might chance an assault but, in most cases, would expect to install a blockade to prevent further supplies reaching the besieged. Siege-engines could be used to batter the castle, perhaps more in the hope of damaging garrison morale than of damaging the fabric. Chemical warfare (poisoning the water supply, catapulting in rotten carcasses), psychological torture (threatening a hostage) could be used, as at Ely in 1131.6 If a stalemate ensued a deal might be struck to the effect that, if a relieving army did not approach the castle within so many days, the defenders would be allowed to surrender without bloodshed and return to join their main force. This was basically to keep the war moving. It was, however, essential to secure permission to surrender. The Gesta provides this example from Plympton, Devon, in 1136:
Then when all were striving hard, the one side to overcome the besieged, the other to defend itself from the besieger unsubdued, Baldwin’s men, to whom he had given the charge of watching his castle of Plympton, despairing of their lord on account of the invincible army which they had heard was with the king, and fearing more than they should have, being utter cowards and irresolute, to run danger to their lives, sent secretly to the king about surrendering the castle and coming to a peaceful agreement.7
The Plympton garrison does not expect their lord to be able to reach them and so they negotiate a secret surrender of the castle. The main interest is that, despite the fact that the surrender is to the advantage of the king, the royalist author of the Gesta is horrified by what he perceives to be the cowardice of the garrison. They are breaking the rules of correct behaviour – they should not surrender without the permission of their lord.
3. Formal Defiance
There seems to be a general feeling that information must be shared and that secrecy is unacceptable. William of Malmesbury, for instance, makes his hero Robert of Gloucester send a formal defiance or diffidatio to Stephen just after Whitsun 1138 before he launches his rebellion:
He sent representatives and abandoned friendship and faith with the king in the traditional way, also renouncing homage, giving as the reason that this action was just, because the king had both unlawfully claimed the throne and disregarded, not to say betrayed, all the faith he had sworn to him.8
Malmesbury proceeds to comment on the number of churchmen who had advised Robert that he would suffer eternally if he did not keep his oath to his father (to support Matilda). The chronicler also mentions a papal bull which besought Robert to maintain his oath although, most suspiciously, Malmesbury promises to quote from the bull only in his next book.9 The key word in the quotation is traditional. There is no need for written laws when both sides share a kind of unofficial Geneva convention. Both sides know what is the `done’ thing, as we see most clearly in the Christmas truce of 1142:
Turning aside thither, then, at the suggestion of their sovereign, (since the soldiers who had remained at her departure, after delivering up the castle (Oxford), had gone away without molestation, and the holidays admonished them to repose awhile), they resolved to abstain from battle, and retired to their homes.10
The whole diffidatio episode seems to be Malmesbury’s attempt to justify his hero’s rebellion against an annointed king. It brought the conflict out in to the open – Robert was now free to make a deal with the Angevins, while Stephen could now attack Robert’s lands without incurring censure. In a similar vein, looking back at the siege of Exeter Castle in 1136, Stephen was besought to show mercy to the men of Baldwin de Redvers because the besieged had not sworn allegiance to the king’s majesty, and had taken up arms only in fealty to their lord.11 They did not need to issue a diffidatio whereas Robert, who had sworn homage to Stephen, did.
Yet, it is important to remember that rigid definitions of the feudal bond between lord and vassal are a relatively modern construct. The idea of a feudal contract is a late one.12 Robert’s diffidatio may have been due to his personal relationship with Stephen rather than to any feeling of legal obligation.
4. Siege or ‘counter’ castles
A major feature of the war was the use of siege or `counter’ castles. These temporary fortifications were used for a number of purposes, one being to form a defended base from which a besieging force could attack. The Gesta provides numerous examples of such siege-castles, many of which utilised churches which caused something of a dilemma for a clerical chronicler:
…he [Geoffrey Talbot] recklessly brought in a throng of armed men and turned a house of prayers and a place of atonement for souls to a confusion of strife and a haunt of war arid blood [Hereford Cathedral 1140]. It was indeed dreadful and intolerable to all men of righteous feelings to see a dwelling of life and salvation transformed into an asylum of plunderers and warriors, while everywhere the townsmen were uttering cries of lamentation, either because the earth of their kinsfolk’s graveyard was being heaped up to form a rampart and they could see, a cruel sight, the bodies of parents and relations, some half–rotten, some quite lately buried, pitilessly dragged up from the depths; or because at one time it was visible that catapults were being put up on the tower from which they had heard the sweet and pacific admonition of the bells, at another that missiles were being shot from it to harm the king’s garrison.13
Here we see the psychological pain inflicted upon the Hereford community by the desecration of their graveyard to create an offensive base from which attacks could be launched on the castle.
Temporary fortifications were also used to blockade castles; not necessarily to allow storm but rather containment. Hence, earthworks were built close to the target castle, normally just on the edge of contemporary cross and long-bow range (250 yards).14 The `cat and mouse’ game could then be played to perfection – neither side could quite hurt each other but the besieging force could nibble away at the garrison’s morale and speed up the process of surrender.
Other temporary castles were built up to fifteen miles away from the enemy castle. Their function was to protect lands that had been subject to raiding (thus preserving good lordship) and to make it difficult for the enemy garrison to forage for supplies. Matilda followed this policy in the Thames Valley in an attempt to outflank the royalist garrison at Oxford:
Matilda fortified castles in various places, wherever she most conveniently could, some to keep the king’s men more effectively in check, some to give her own more careful protection.15
Her engineers made use of whatever local feature was available be it a church or a pre-existing ancient earthwork. She built a series of structures for defensive and offensive purposes but with one unifying function; to challenge the lordship, the dominatio, embodied in the royalist castle, usually a castle that had existed since the late 11th century and had embodied royal authority since that time. She was not out to destroy such symbols, rather to obtain their peaceful surrender with the minimum of damage so that such castles could then become the symbols of her rightful authority.
In 1145 Philip of Gloucester persuaded his father, Earl Robert, to fortify Faringdon Clump as an Angevin base from which to challenge the royalists in Oxford. The Gesta reports that the Earl:
Built in it [Faringdon] a castle strongly fortified by a rampart and stockade, and putting in a garrison that was the flower of his whole army he valorously restrained the wonted attacks of the king’s soldiers, who had been coming out from Oxford and other castles round about to harass his own side. And they, being indeed most straitly confined, as they could not even leave their castles because of the foes who hemmed them in, did the only thing that remained, by letters and messengers they immediately begged for the king’s help.16
The construction of Faringdon was a challenge to the king’s lordship. Fifteen miles from Oxford, it was a bold statement of the westward advance of Angevin power. Despite the speed of its construction and its short life, there are still traces of it today: clay revetments of the sandy soil and foundations of stone walls.17 Faringdon was regarded as a serious threat by the royalists, hence the response of Stephen. He raised reinforcements in Oxford and prepared for a long siege:
Then he instructed his men to busy themselves with a wonderful task and not without profit, namely, surrounding themselves with a rampart and stockade, that a sudden attack of the enemy might not break in to their confusion but, ensconced in a sure refuge of their own, they might both provide more securely for themselves and go to meet the enemy more safely and more boldly when occasion required. And without delay, setting up engines most skilfully contrived around the castle, and posting an encircling ring of archers in very dense formation, he began to harass the besieged most grievously. On the one hand stones or other missiles launched from the engines were falling and battering them everywhere, on the other a most-fearful hail of arrows, flying around before their eyes, was causing them extreme affliction; sometimes javelins flung from a distance, or masses of any sort hurled in by hand, were tormenting them, sometimes sturdy warriors, gallantly climbing the steep and lofty rampart, met them in most bitter conflict with nothing but the palisade to keep the two sides apart. It was in fact like this that the king’s men harassed the besieged by daily onslaughts; they, on their side, defended themselves manfully without giving way, until those who were chief in command, without the knowledge of the others, sent secretly to the king and made an agreement conceding his den land for the surrender of the castle.18
The sequence of siege warfare is clear. The building of the castle is the threat to Stephen’s authority, not only by its illegal construction, but perhaps principally because its very existence implies that the local villages will be raided for supplies, thus the king’s subjects will be injured. The king is obliged both to respond to any request for help from his beleagured Oxfordshire garrisons and from his subjects. If the king does not respond, then he cannot object to the surrender of his men as he has fai–led to be a good lord.
Stephen does respond and a siege ensues including an opening storm. The Gesta account reveals the siege of Faringdon to have been an active one with siege engines in use and even occasional hand-to-hand fighting – a significant effort was made in the early stages of the siege to demoralise the garrison. As the Gesta comments, the besieged defended themselves manfully until the leaders made a secret deal with the king. Although the Gesta is a supporter of Stephen (at this stage in his chronicle), he is clearly disgusted at the underhand deal (as he had been at Plympton).
In planning resistance and surrender a delicate balance had to be struck. The commander who surrendered too early was a coward who had betrayed his lord – surrendering too early was often measured by the amount of food left within the castle. The cowardly lord was despised by his own side and by his enemy likewise. Alternatively, the die-hards were a major nuisance and unlikely to receive mercy on final surrender. The cunning commander found the middle ground – if he could not expect aid from his lord then he needed to hold out long enough for the honour of his party and to earn the respect of his opponents. This would normally ensure the lives of the garrison and an honourable withdrawal, often with all equipment, to fight again another day.
The secret deal which saw Faringdon surrender gave the king hostages to ransom and a store of arms and booty. To the chronicler Henry of Huntingdon this marked an up-turn in the king’s fortunes.19 His swift action in besieging Faringdon, the fierce initial attack that may have decided the leaders to opt for a secret deal without telling the garrison, these raised his prestige and encouraged the waverers (eg. Ranulf of Chester) to throw in their lot with the king once more.
Wallingford Castle, the most easterly Angevin outpost on the Thames, was more or less under continual siege throughout the fourteen years of the war.20 In 1152 the third major phase began. Stephen had built a siege-castle at Crowmarsh on the western bank of the river, and had also fortified Wallingford Bridge, thus cutting the garrison off completely from the eastern side of the river. Following conventional practice, the garrison sent an appeal to Duke Henry begging him to come to their aid or, if he could not, to permit them to surrender.21 A truce may have been arranged while they waited – until the summer of 1153.
On his arrival, Henry quickly took the bridge works and then concentrated upon Crowmarsh:
…coming to Crowmarsh, a castle rising on a very high mound in front of Wallingford with only the river in between, he ordered his troops to attack it with great vigour on every side. When, behold, the king’s men, who on hearing of the duke’s arrival had withdrawn to places where they could not be seen, though a few kept up a show of resistance in the outer part of the castle, burst out in small parties from different hiding–places and made a gallant charge on those who had already climbed the mound and entered the outer part of the castle, and capturing some and killing others they compelled by their valour the whole body to give way…22
Crowmarsh was clearly a double-enclosure site (as it still appears on the ground). Henry approached the outer bailey, with its high rampart necessary to protect the inmates from the missiles despatched from Wallingford Castle. Stephen’s men divided into two groups – those retreating to the inner enclosure, those hiding in the outer. Those royalists who burst out in small parties from different hiding-places and made a gallant charge on those who had already climbed the mound were repelling those Angevins who had managed to scale the outer rampart. Witnessing the unusual strength of Crowmarsh, Henry fortified his own encampment between it and the bridge:
….commencing the important and difficult enterprise by digging a deep trench round the walls and his own camp, so that his army had no egress but by the castle of Wallingford, and the besieged had none whatever.23
There is a hint here that Henry may have worried about deserters, ensuring that any of his men who tried to retreat had to do so under the eyes of the Wallingford garrison. Certainly, both king and duke were beginning to have problems inducing their men to fight. Those inside Crowmarsh found their main entrance blocked but they were being supported by cavalry skirmishers despatched from Oxford by Stephen.24 Any Crowmarsh men unlucky enough to be captured by the Angevins were killed; Robert of Torigny reports the beheading of sixty royal archers by the duke.25 Such treatment was justifiable and even acceptable because the garrison had refused demands for their surrender – but note that fatal treatment was reserved for the hated archers whereas the knights were ransomed. War did not upset the social pattern of society – those of the highest class looked after each other, whatever side they might take. On Stephen’s approach from the Oxfordshire bank, Henry dismantled his encampment and retreated back over the bridge into Berkshire. He could not afford to leave a fortified camp so close to the bridge although he did leave a small force to guard its Oxford end as long as possible.
At this point the bishops, led by the Archbishop, managed to secure a five-day truce, during which Stephen was to evacuate the eighty men garrisoning Crowmarsh and demolish it. Due to a baronial refusal to fight on and to church pressure, this truce was to be the beginning of a peace-making process, hence Stephen’s responsibility to dismantle the hostile, temporary castle of Crowmarsh. His reward was the safe removal of his remaining garrison. As Robert of Torigny said, these conditions `really were honourable ones’, forming a normal part of peace negotiations.26
The term `anarchy’, so glibly used by modern students of the reign of Stephen, is clearly inappropriate. This was a period of conflict but, as the Gesta reveals, it was tightly controlled conflict which recognised fights and status at all times – the victims were as usual the civilians and footsoldiers. Mercy was frequently on offer because it avoided breaking the local social, structure and could ensure future loyalty. The laws and conventions of warfare may not yet have been written down, but they existed and the higher ranks knew what they were. And it reveals overwhelmingly the function of the castle as a symbol of status and privilege – that is why so much effort, time and manpower was invested in siege warfare because only by the subduing of castles could each side retain their rank.
Twelfth-century castles function in an elaborate power-play game, in war and peace alike. Very few are taken by physical force or straightforward bombardment. Instead, they changed hands due to knightly agreements and dirty deals which some times shock the chroniclers. The niceties should be observed. The castle is first and foremost the symbol of lordship – the siege castles which are thrown up are temporary aberrations of lord ship which need to be demolished for peace and the status quo to be restored. They are not `castles’ in our current usage and really do need different nomenclature.
Bradbury, J., The Medieval Siege (Suffolk, 1992)
Bradbury, J., Stephen and Matilda: The Civil War of 1139-53 (Stroud, 1996)
Dixon, P., “The Donjon of Knaresborough: the castle as theatre”, Chateau Gaillard 14 (1988), p.124-140.
Dixon, P., and Marshall, P., “The great tower at Hedingham Castle: a reassessment”, Fortress 18 (1993), p. 16-23
Greenway, D. (ed.), Henry of Huntingdon, Historia Anglorum (Oxford, 1996)
Harper-Bill, C. and Holdsworth C.J. (eds.), Studies in medieval history presented to R. Allen Brown (London, 1989)
Higham, R. and Barker, P., Timber Castles (London, 1992)
Holdsworth, C.J., “War and Peace in the Twelfth Century: the Reign of Stephen Reconsidered”, War and Peace in the Middle Ages, ed. B.P. McGuire (London, 1987), p.67-93.
Howlett, R. (ed.), The Chronicle of Robert of Torigni, in Chronicles of the Reigns of Stephen, Henry II and Richard I, Rolls Series, 4 volumes (no.4) (1884-9)
King, E.(ed.), The Anarchy of King Stephen’s Reign (Oxford, 1994)
Leeds, E.T, “An Adulturine Castle on Faringdon Clump, Berkshire”, Antiquaries Journal 16 (1936), p.165-78
Leeds, E.T., “An Adulturine Castle on Faringdon Clump, Berskhire (second report)”, Antiquaries Journal 17 (1937), p.294-8
Morillo, S., Warfare under the Anglo-Norman Kings 1066-1135 (Suffolk, 1994)
Potter, K.R. (ed.), William of Malmesbury, Historia Novella (Edinburgh, 1955)
Potter, K.R. (ed.), Gesta Stephani (Oxford, 1976)
Reynolds, S., Fiefs and Vassals: The Medieval Experience Reinterpreted (Oxford, 1994)
Strickland, M., War and chivalry : the conduct and perception of war in England and Normandy, 1066-1217 (Cambridge, 1996)
1. Strickland, 1996
2. Potter, 1976
3. Strickland, 1996, p. 34-46
4. Potter, 1976, xviii-xxxviii; Bradbury, 1996, p.141-2.
5. Dixon, Marshall, 1993, p. 16-23; Dixon, 1988, p. 121-140.
6. Potter, 1976, p. 78-9
7. ibid, p. 34-6
8. Potter, 1955, p. 23
10. ibid, p. 51
11. Potter, 1976, p. 42-3
12. Reynolds, 1994, chapter 2
13. Potter, 1976, p. 53
14. ibid, p. 42
15. ibid, p. 69
16. ibid, p. 180-1
17. Leeds, 1936, p. 165-78; Leeds, p. 294-8
18. Potter, 1976, p. 182-3
19. Greenway, 1996, p. 746-7
20. Slade, 1960, p. 33-43; Spurrel, 1995, p. 257-270
21. Greenway, 1996, p. 758-9
22. Potter, 1976, p. 236-8
23. Greenway, 1996, p. 766-7
24. Potter, 1976, p. 238-9
25. Howlett, 1884-9, p. 173-4
26. ibid, p. 175-6.
This article was first published in Chateau Gaillard XIX: Actes du Colloque International de Graz, 1998 (2000). We thank the Centre de Recherches Archéologiques et Historiques Médiévales for their permission to republish this article.