16
When Gruffudd ap Cynan set sail from Dublin for northern Wales in 1075, he was taking a calculated risk. Gruffudd only had a handful of men with him, generously supplied by his mother’s relatives, the powerful Viking rulers of Dublin. With their support, he hoped to establish himself as ruler of Gwynedd. The hope was not entirely vain. For Gruffudd’s father was a prince of the royal line, who’d been forced into Irish exile in 1039 when Gruffudd ap Llywelyn, a member of a rival dynasty, seized the throne. Now the younger Gruffudd ap Cynan was out to exact revenge for his father’s mistreatment.
His chances were helped by the political situation in northern Wales. He’d begun preparations upon hearing of the death of Bleddyn ap Cynfyn, Gruffudd ap Llywelyn’s half-brother and successor. Still, time and distance were not on Gruffudd ap Cynan’s side. And while he moved fast, by the time he arrived off Anglesey a few months later, Gruffudd found that Trahaearn, the cousin of Bleddyn, had established himself as ruler in his stead. Gruffudd ap Cynan now made contact with a few local allies, before sailing east to the castle of Rhuddlan, in north-central Wales. Here he secured the support of the local lord, Robert. The cousin of Hugh, the new Norman earl of Chester, Robert of Rhuddlan had been extending his influence into northern Wales for some time. He and Hugh were only too happy to support Gruffudd ap Cynan and, with their assistance, Gruffudd was able to make quick work of his opponents. Finally, Gruffudd ap Cynan was able to sit on his ancestral throne.1
At a glance, these events may seem rather unspectacular. One provincial Welsh ruler replaced another, in one of those complex dynastic struggles that had long characterised the region. Yet when we look more closely, differences appear. Gruffudd ap Cynan was only able to secure the throne with Robert’s support; and once established, he immediately moved against his Norman backers. The Normans were now an important new force in Welsh politics.
By this point, the Normans had been making incursions into Gwynedd and its southern neighbours for a number of years. The difference between these undertakings and the events of 1066 is that they were private ventures, taken at the initiative of the new lords of the march (as the English frontier with Wales was known). With a kingdom and duchy to his name, the Conqueror had little interest in extending his direct rule into the poorer regions to England’s west and north. For his aristocrats, however, the situation was different. Conquest had whetted their appetite; and they were hardly going to stop at the frontier. For his part, William was happy to cheer on his men. If Norman influence could be extended without any cost to the Crown, William stood to gain as much as his men.
It was out of these processes that the marcher lordships of Wales slowly emerged.2 The term ‘march’ refers to a border or frontier (much like the German Mark). In its Old English form, it had already given rise to the name of the Mercian people of the West Midlands (literally, ‘the people of the frontier’). These regions had thus long possessed a frontier character, and the great eighth-century Mercian ruler Offa famously erected a dyke running north–south along the Welsh-English border. Low-level conflict between the groups was common, and the dyke may originally have been intended to prevent cattle rustling. At a political level, however, stability generally reigned. English kings often exerted a degree of overlordship over their Welsh neighbours, but made little effort to conquer them or rule directly.
The years before the Conquest had started to see this change. In the tenth and eleventh centuries, Wales was generally divided into two realms: Gwynedd in the north (also encompassing Powys in north-central Wales) and Deheubarth (literally meaning ‘the southern part’) in the south west. These were often ruled jointly and remained distinct from Morgannwg (Glamorgan) in the south east, which enjoyed closer connections with the neighbouring English.3 In 1055, the northern ruler Gruffudd ap Llywelyn secured control over Deheubarth and also over these south-eastern domains. He now presented a very different kind of threat, a threat exacerbated by ongoing factional conflicts at the English court. These years saw the ascendency of the Godwins, which was bitterly resented by the earls of Mercia, represented by Ælfgar. And Ælfgar soon became a firm friend and ally of Gruffudd ap Llywelyn, his immediate neighbour to the west. It was the Welsh king who offered Ælfgar a safe haven when he was exiled in 1055. In the following years, the two often made common cause. Gruffudd ap Llywelyn married Ælfgar’s daughter Edith. And when Ælfgar was exiled a second time in 1058, it was with Gruffudd’s assistance that he was restored.4
Ælfgar’s death in 1062 left Gruffudd exposed, however, and the Godwins were quick to take advantage of the situation. Having obtained royal sanction for a pre-emptive strike, Harold marched along the northern coast of Wales in late December 1063, ravaging as he went. The plan was to catch Gruffudd unawares. This failed, although Harold took the opportunity to torch Gruffudd’s residence at Rhuddlan. The following spring, he gathered forces in Bristol and began a joint land and naval assault with his younger brother Tostig. Gruffudd was now driven into Snowdonia, where his men turned on him. The Welsh king’s decapitated head was then sent to Harold as a grim peace offering.
The result of Gruffudd’s fall was turmoil, which Harold happily exploited. The regions between the Dee and Clwyd were now brought under English control (and integrated into an enlarged Cheshire). The main gains, however, were made in the south, where Gruffudd’s power had always been weakest. In the early to mid1050s, the earldom of Wessex had been expanded to incorporate the south-west Midlands, which bordered on the old kingdom of Gwent in eastern Morgannwg. Gruffudd’s fall allowed Harold to extend his authority here, building a hunting lodge at Portskewett. He also married Gruffudd’s widow (Ælfgar’s daughter Edith). This was partly a move to forge an alliance with the earls of Mercia. But it was also an act which resonated in Wales. As in other parts of Europe, it was common here to wed the widow of a defeated foe; Harold was symbolically enacting victory.5 Despite this interruption, the alliance between Gruffudd’s and Ælfgar’s families lived on. And when Ælfgar’s sons Edwin and Morcar rebelled against the Conqueror in 1068, they were supported by Gruffudd ap Llywelyn’s successors, his half-brothers Bleddyn and Rhiwallon.
The West Midlands had itself seen significant Norman influence before the Conquest. It was here that Ralph of Mantes’ modest earldom was to be found, and it was here that the earliest Norman castles were constructed in the 1050s. Their numbers multiplied rapidly after 1066, as Norman aristocrats were introduced to the region in ever larger numbers; and the Welsh march soon boasted a higher density of fortifications than any other part of the British Isles.6 The Anglo-Saxon dyke had given way to the Norman motte and bailey. Yet the new lords of the march had more than modernising the defences in mind. The castle was primarily an instrument of lordship, and it was this that was at stake: control over men and land.
The emerging march attracted (and bred) a particularly aggressive sort of lord – men accustomed to high risks and high gains. Few better exemplify this spirit than Hugh d’Avranches, the first earl of Chester.7 Hugh’s father had been a major contributor to William’s invasion fleet in 1066. And though Hugh was too young to participate, he crossed the Channel soon thereafter to assist in consolidating Norman rule. Hugh was initially rewarded for his father’s loyalty with command of the castle of Tutbury in Staffordshire. But Hugh was marked out for greater things, and was soon entrusted with the strategic town of Chester and the newly created earldom of this name.8 Chester was one of the main English ports on the Irish Sea, facilitating trade between the Midlands, Wales, Ireland and Scotland. It was also located just on the English side of the frontier with Wales, controlling traffic up the Dee. From his seat, Hugh could keep tabs on the northern Welsh kings, who’d long been the most powerful (and dangerous) of the native dynasties. Edwin and Morcar had recently looked to Wales for support for their 1068 uprising; the Conqueror was now placing a trusted man on the frontier in response.
What drew Hugh and other like-minded lords to the region were the opportunities afforded by the frontier. The Norman conquest of England had been remarkably swift. Most new lords were in place within a decade of 1066; and by the time the Domesday Survey was compiled in 1086, the process was all but complete. This meant that ambitions would now have to be directed outward, towards the independent realms to England’s west and north. It was only here that new lands and lordships were to be gained.
Hugh soon set about realising these prospects. Shortly after his appointment, Hugh’s men secured the lowland areas to the west and south of Chester which Harold had won in 1064. Some of these went to new Norman lords, but most were earmarked for Hugh’s cousin (and right-hand man) Robert of Rhuddlan. It was Robert who, under Hugh’s aegis, would now drive Norman expansion in the region. He was following in Harold’s footsteps; and by 1073, Robert had set himself up at Gruffudd ap Llywelyn’s old residence at Rhuddlan. A formidable fortress was now erected on royal orders.
Hugh and Robert both participated in a raid on Llŷn, probably in the mid-1070s. And by 1078, they’d extended their authority to Degannwy, overlooking the Conwy. Most of these gains were made by Robert, but Hugh was present at crucial junctures, taking the lead in a foray of 1081 and returning again in the mid-1090s, upon Robert’s death. When the Domesday inquisitors came to Cheshire, they included the entire region up to the Clwyd within the survey; in their eyes, this was effectively England. They also did not ignore the lordships established by Robert beyond this. North Wales – as they term Gwynedd – may not have been England proper, but it was evidently within the English orbit. From these territories, they record that Robert paid an annual tribute of £40 – the same sum owed by the Welsh ruler of Deheubarth in the south west.9
By the early 1090s, Hugh and Robert had erected castles at Aberlleiniog (on Anglesey), Caernarfon and Bangor; there may also have been similar structures in Tomen y Mur and Tomen y Bala in Meirionnydd. By 1092, they were in a position to appoint a Norman to the bishopric of Bangor, in the heartlands of the old kingdom of Gwynedd. As in England, claims to religious, political and cultural domination went hand-in-hand.10 When the southern ruler Rhys ap Tewdwr fell in 1093, local chroniclers record that all Wales had fallen with him. It’s easy to see why they thought this. The north had been in Hugh and Robert’s hands for some time; now, it was joined by Deheubarth to the south. This was a view shared on the other side of the frontier. In the West Midlands, John of Worcester likewise reports that ‘kings ceased to rule in Wales’ at this point.11 Norman influence was everywhere; it seemed but a matter of time till this was formalised.
This is not to say that Norman incursions went unresisted. One of the main opponents of Hugh and Robert was Gruffudd ap Cynan, with whom we began. Gruffudd is the only Welsh ruler of the Middle Ages for whom we have a contemporary biography, so we can follow his career in unusual detail.12 Initially, Hugh and Robert were happy to support Gruffudd ap Cynan, doubtless in the hope of establishing a pliable client king. Yet once established, Gruffudd turned against his Norman allies. Fortune did not, however, favour the new king. Before 1075 was out, Gruffudd had been defeated and driven out by Trahaearn ap Caradog, whom he’d just replaced. Gruffudd did not, however, give up hope. In 1081, he returned in force, backed by the southern ruler Rhys ap Tewdwr. The two of them met Trahaearn at Mynydd Carn (a site somewhere north of St Davids), and there they defeated and slew him.
Gruffudd now marched into Gwynedd again, targeting the lands of his opponents along the way. Upon arrival, he secured the throne. Yet his second reign would prove to be not much longer than the first. This time, it was the Normans who brought Gruffudd down. Earl Hugh had been happy to use Gruffudd against other northern Welsh rulers in the mid-1070s, but he had little desire to let him establish an independent regime of his own. Hugh therefore enticed Gruffudd to a meeting at Rug in Edeirnion, just west of modern Corwen. Here the unsuspecting prince was captured by his old ally, Robert of Rhuddlan. Gruffudd now spent over a decade in an Anglo-Norman prison. During these years, Hugh’s own attention was frequently drawn elsewhere, as the division of the Conqueror’s massive cross-Channel realm between Robert Curthose and William Rufus (II) threatened his Norman patrimony.
Soon enough, however, the earl was forced to return to Chester. In 1093, Robert was killed by a Welsh raiding party, leaving Hugh’s lands dangerously exposed. The following year, a full-blown rebellion was under way. After two decades of steady advance, the Normans now came to appreciate the difficulties presented by the fractious world of Welsh politics. Like southern Italy, Wales was a mountainous landscape in which rapid conquests and centralised states were the exception rather than the rule. Around this time, Gruffudd himself escaped – indeed, he may have been responsible for Robert’s death (the later chronicler Orderic Vitalis mentions a ‘Gruffudd king of the Welsh’ in this connection). Certainly by 1097, Gruffudd was on the loose and looking to make good his claim to his north Walian patrimony.
The 1090s was a decade of contestation. Initially, the momentum lay with the Welsh. A rising against the Normans had begun in the north, where it was led by Cadwgan ap Bleddyn, the son of Gruffudd ap Llywelyn’s half-brother and successor. But rebellion spread and 1094 saw castles fall across Gwynedd, Ceredigion and Dyfed. Hugh’s authority was pushed back beyond the Conwy, never to be fully resuscitated. So serious was the situation that William Rufus appeared twice, in 1095 and 1097. But the Welsh refused to be drawn into battle and the gains were few.
In the end, the Normans did prevail. But it would be the marcher lords who secured the victory. In 1098, Hugh joined forces with Hugh de Montgomery, the earl of Shrewsbury. Together, they marched along the northern coast of Wales, driving back the men of Gwynedd and Powys, led by Cadwgan ap Bleddyn and Gruffudd ap Cynan. The latter two were not natural allies. But with Normans just across the Menai Strait, they were willing to co-operate. Even so, there was no stopping the advance, and both Cadwgan and Gruffudd were forced into exile across the Irish Sea.
Yet Norman success was to be partial and fleeting. For at this moment in 1098 the restless king of Norway, Magnus ‘Barelegs’, appeared off the coast of Anglesey. Magnus was following in the footsteps of his own Viking forebears, who’d long dominated the Irish Sea from bases at Dublin, Man and the Isles. Sailing over from Norway, he’d first established himself on Man. From there, northern Wales was a natural target, particularly in the light of recent political instability. But just as Magnus’ ships approached land, a skirmish broke out with the Normans guarding the shoreline. Our sources give contradictory accounts of the conflict, but most agree that Magnus felled Hugh de Montgomery with a spear thrown from the deck of his ship. This threw the Normans into disarray, but the Norwegians were in no position to press their advantage. Magnus only had six ships at his command and may already have overextended himself. The Norwegian king therefore retreated to Man, while Hugh de Montgomery’s men returned to Shrewsbury, where they laid the earl to rest seventeen days later.
Hugh d’Avranches likewise retired to his base at Chester. Anglesey was simply too far from Chester (or Rhuddlan) to make outright conquest viable. The people here remained loyal to their native lords, and would do so well into the thirteenth century. Hugh therefore made contact with Gruffudd, offering to restore him in exchange for his submission. Similar agreements were made elsewhere, with Cadwgan being established in parts of Powys and Ceredigion under the oversight of the new earl of Shrewsbury, Robert de Bellême.
These arrangements were a significant step back from the ambitions of earlier years. But we should be wary of branding them failures. Men such as Hugh and Robert were out for fame and fortune, and willing to use any means to achieve them. The risings of 1093 had demonstrated the limits of Norman power and control. Provided a degree of overlordship could be maintained, Hugh was happy to cut his losses and leave matters at that.
Hugh was, in any case, not much longer for this world. In late 1100 or early 1101 he fell ill. And sensing the end was nigh, he took monastic vows. Writing at Saint-Evroult in Normandy in the 1130s and 1140s, Orderic Vitalis paints a vivid picture of Hugh. Orderic had grown up on the march, so was well acquainted with the earl’s stomping grounds. He also had a more direct source of information, in the form of Robert of Rhuddlan’s brother, who was a fellow monk at Saint-Evroult. According to Orderic, Hugh was the pre-eminent lord of the Conqueror’s realm (a slight exaggeration) and a ‘lover of this world and earthly pomp’ (no exaggeration). The earl was a great warrior, who always fought from the front; and a man of prodigious generosity and lover of luxury. Here we see the ideal attributes of the Norman marcher aristocracy – bravery, generosity and joie de vivre.13 These characteristics are reflected in the monikers Hugh earned. To many, he was ‘Hugh the Fat’; to others, he was ‘Hugh the Wolf’. Yet as his deathbed conversion reveals, Hugh was also possessed of piety. Hugh’s ambition was rivalled only by that of his cousin Robert of Rhuddlan, who was driven by ‘pride and greed’ to ‘unrestrained plunder and slaughter’ (or so Orderic has it).14
What marks the conquests of Hugh and Robert out is their enterprising nature. It was they, not the Conqueror and Rufus, who drove Norman expansion into northern Wales. William I only came to Wales once (in 1081), and then to have his (nominal) overlordship acknowledged. William II was a more regular presence, but only in the crisis years of the later 1090s. Norman gains in Wales therefore took on a very different character to those in England. Here the king was at most a distant overlord. This was, of course, part of the appeal. Marcher lords were men of independent means, not simply agents of the Crown. As the later saying went, the king’s writ did not run in the march. Sometimes, marcher lords interpreted this quite literally. When Walter Clifford received a royal missive which was not to his liking in the thirteenth century, he made the messenger eat his words – ink, parchment and all!15
Over time, this independence came to be formalised. The result was the marcher lordships of the central Middle Ages – or the March proper (with a capital M).16 What characterised this was independence from the Crown. As one contemporary noted, here each lord was ‘like a king’ (quasi rex) within his own domains.17 This meant not only oversight of justice, but also control of what we might now call ‘foreign affairs’. Marcher lords were free to make peace and wage war with each other and their neighbours. This was the ‘Wild West’ of medieval Britain, where lordship was asserted at the end of a spear point. It’s no coincidence that the John de Warenne who produced a rusty sword as evidence for his tenure in England in the late thirteenth century was of marcher stock.
The marches were also hybrid in character: these were Anglo-Norman lordships erected on Welsh foundations. Elements of local practices and power structures were adopted, and marriages and alliances with Welsh rulers were common, particularly in earlier years. The March was thus a point of contact between England, Normandy and Wales, a culturally and politically dynamic region in which bi- and even trilingualism was common.18 This is not to suggest that the March was some sort of modern multicultural paradise – quite the reverse. The flip side of cultural contact was cultural chauvinism, as the dismissive attitudes of Anglo-Norman writers to the Welsh reveal. The March was a place where identities blurred, but were also reasserted.