By tradition, the Battle of Hastings was fought at Battle Abbey. We have been sceptical since visiting the abbey as school children. There is no physical supporting evidence. Indeed, no one has found so much as a battle related button to give a hint about any aspect of the invasion or battle. Instead, the orthodox narrative is based on a dozen or so equivocal contemporary accounts. They might have been misinterpreted.
We spent twenty years scouring the Hastings Peninsula for a hill that better fits the descriptions in the contemporary accounts. There was a serious danger of us not living long enough to reach any useful conclusion. Then we saw Nick Austin present “Secrets of the Norman Invasion”.
Austin pointed out that the traditional translations and interpretations of the contemporary accounts - that most enthusiasts, including us, had been working from - are based on a predetermined assumption that the battle happened at Battle Abbey. We explain the basis for this assumption and why we think it is flawed here.
If the battle did not happen at Battle Abbey, there are lots of ways that the contemporary accounts can be translated and interpreted. Austin used his own interpretations to support his theory that the battle was fought on Telham Hill. We explain why we disagree with his conclusion here, but we stole his approach.
A knowledge of archaic languages is one of our least useful skills. Or at least we thought so until we listened to Nick Austin. We went back to first principles, considering alternative translations and interpretations of the contemporary accounts. We also regressed the geography of East Sussex back to medieval times. We used these to work out the likely landing place, which we used to work out the likely camps. We used those to work out the likely battlefield. In retrospect, we realise that they are not necessarily linked, but it provides a structure to present our investigation.
There are a couple of points we need to make before getting into the detail. Perhaps most importantly, what we are about to reveal is completely at odds with the traditional Battle of Hastings narrative. The landing, the camps, the combat and the battlefield will be unfamiliar and new. We know it sounds incredible. It is difficult to believe that historians could have got all the events completely wrong. But it is not as far-fetched as it might seem.
The one unambiguous locational clue that is corroborated across multiple contemporary accounts is that Battle Abbey was built on the battlefield. Historians have taken this as sacrosanct, then worked back to deduce the events that preceded it. But if the battle did not happen at Battle Abbey, the reasoning chain collapses. We explain here why we think that all the 'Battle Abbey on the battlefield' statements are unsound. If they are wrong, most probably, so too are the orthodox battlefield, landing site, camps and allied events. If we are right, historians did not make dozens of errors, but just one, from which the others cascaded.
John Grehan and Nick Austin have already been down this road, proposing that the battlefield was at Caldbec Hill or Telham Hill respectively. Their approach is to present an interpretation of the contemporary accounts that tries to prove their proposed battlefield location beyond reasonable doubt. In our opinion, both arguments leave a lot of doubt, as we explain in our Alternative Battlefield blog. In our experience, the contemporary accounts are too ambiguous to prove anything significant. We are not even persuaded beyond reasonable doubt by our own theory.
Our objective is not to educate. Professionals do a far better job of explaining history than we ever could. Rather, we want to tap our readers' knowledge and inspire you to think about the clues. Everything we know and nearly everything we suspect is in these blogs. We have done all we can, yet fallen short. If the real battlefield is to be found, our readers need to provide the corroborative evidence. We are optimistic. No one has looked for evidence at our proposed landing place, camp sites or battlefield. We are hopeful that someone has unknowingly collected or deduced evidence to corroborate or refute one or more parts of our theory.
We found nearly 100 clues about the landing place, camps sites and battlefield location in the contemporary accounts, more than half of which are novel or have never been taken seriously. We have devised an interpretation of each clue that matches our theory. Some of them will doubtless be proved wrong. It makes no difference to the credibility of our theory. When we apply the same clues to the other battlefield candidates they barely match more than any random hill on the Hastings Peninsula. Even if half of our interpretations are proved wrong, our proposed battlefield would still be the best match for the contemporary account clues. Regardless, we want to be as accurate as possible. We would love to hear about any errors we might have made and about improvements to our translations or interpretations.
Our main interest is in evidence about our proposed landing place, camps or battlefield. We would also love to hear readers' thoughts about some of our novel conjectures, including the following: The Roman fortress, burh and church that we think were at Cadborough; The Frankish monastery that we think was near Sedlescombe crossing; The Alfedian burh that we think was at Winchelsea; The motte for William's kit fortress that we think was at Monks Lane where it joins to Refectory Lane; The island harbour that we think was the original pefenes ea; The levee and tiled water channels that we think lined the north bank of the Brede.
We are also interested in three fundamental differences between our theory and the others: 1) The other battlefield candidates are on the Hastings Peninsula, whereas ours is not. 2) The other proposed battle theories have straight shield walls, whereas we think it was enclosed; 3) The other proposed battlefields need the English flanks to be protected by impenetrable woods or impassible streams, whereas ours relies on protection from a line of hollows that are still there.
We would like to hear theories (or opinions) for and against the first two of these. For the third, we would love to hear from anyone that visits the site whether they agree that the hollows would protect the north end of the battlefield and what might have caused them. We are told that there is no such thing as an impenetrable mature forest in temperate latitudes. We would like to hear from experts that can confirm or refute this. We would like to hear from anyone that has a credible reason why the Norman cavalry could not just ride through the streams that supposedly protect the English flanks at Battle. It is easy to check because both streams are crossed by public footpaths.
We are equally happy to hear from anyone who suspects that any of our conjectures are wrong. One expert contacted us about a minor flaw in our conjecture that Harold planned to annex Normandy and Brittany after he had defeated William. It is exactly what we hoped for. Eliminating possibilities worked for Sherlock Holmes. It can help here too. We are only amateur hobbyists. We have doubtless made countless mistakes. We would love to be corrected.
Our investigation might be more Clouseau than Poirot. Our conclusions are linked in time, but not in consequence. We might have fingered the right battlefield even though we got the wrong landing place and/or camps. Any or all of them could be wrong. Any or all of them could be right, though not necessarily for the reasons we think. We urge you to finish, even if you vehemently disagree with some of our intermediate conclusions. This, after all, might be your opportunity to contribute to the most monumental discovery of the 21st century.
It seems to be standard practice in studies about the Battle of Hastings to include a section about medieval society, the Church, feudal land tenure, Anglo-Saxon England, Normandy, William, Harold, Edward the Confessor’s succession, and the events following his death. Not here. We expect readers to have this knowledge. We do not provide it. Others do a better job than we ever could. The Wikipedia entry for the ‘Norman Conquest’ covers the basics. There are dozens of books and websites that go into more detail. You could do a lot worse than English Heritage’s website or Teresa Cole’s book entitled ‘The Norman Conquest’.
Throughout this blog we refer to a ‘Sedlescombe crossing’. It is important for our battlefield theory that it was easy to cross the Brede at Sedlescombe. But the Brede valley is 200m wide at Sedlescombe, which makes some historians think it was only traversable by ferry. In the early Holocene, when the Brede was tidal beyond Whatlington, it would have been 200m wide at Sedlescombe at high tide. By Roman times, glacial rebound had dropped the Brede’s normal head of tide to Sedlescombe. This has to be so because the Rochester Roman road crossed the Brede at Sedlescombe. The Romans shipped a lot of iron products across the Brede. We guess they channelled the fluvial Brede and crossed the valley with a causeway and timber bridge. The abbeys of St Denys and Fécamp also shipped a lot of iron products across the Brede. We guess that they restored the Roman bridge. Failing that, we are convinced there would have been a solid shallow shingle ford.
A few words about ethnicity too. For convenience we will refer to the adversaries as Norman and English, but it is not accurate. Perhaps a half of William’s army were Bretons, Franks and others. The defenders were only English insofar as they were defending England. They are often referred to as Saxon, but this is not right either. Harold's mother was Danish. His children had Danish names. His father was Saxon but came to power as an ally of Danish King Cnut. Harold thought of himself as ethnic Danish, as did the majority of his barons, his elite guards and the most loyal of his subjects. Wace lists the English army's home counties. There were at least as many ethnic Danish Jutes and Angles as there were ethnic Jastorf Saxons.
The Traditional Landing
Everyone knows the traditional Battle of Hastings narrative. The Normans landed near Pevensey. The knights rode to modern Hastings. Everyone else sailed to a port below modern Hastings, then made their way up the cliff to join the knights in a camp at modern Hastings. Like everything else to do with the invasion, there is no tangible evidence. Instead these events have been derived from landing sites and camp locations named in the contemporary accounts, all of which sound something like Pevensey or Hastings. We are far from the first to point out that the tradition is preposterous.
Figure 5: 1066 Hastings coastline with traditional landing locations; southern boundary of Andredsweald shown in green dots
A deliberate landing at Pevensey is implausible. It held the only major garrison between Lympne and Portchester. Surely William would not aim to land at the one place on the coast opposite Normandy that was liable to be well defended. It was in a saltmarsh. Surely William would not land where his cavalry would be impotent. It was at the end of a narrow-necked peninsula that had no running fresh water. Surely William would not land where the wells might have been poisoned or where a few hundred determined English defenders could have blockaded the Norman army until they ran out of food. William’s destination was the Hastings Peninsula. Surely, he would not aim to land 30 miles away (by horse), on the far side of the dense hostile Andredsweald forest.
Several contemporary accounts say that the Normans landed near a port without specifying its name. Pevensey was an associate member of the Cinque Ports in the 13th century, so it did become an important port. It was not one in the 11th century. According to Andrew Pearson, the materials to build Anderitum (the Roman fort at Pevensey) and its access road were brought by boat. Once the road was finished, they used that instead. We think the wharf fell into disuse. Pevensey had few local inhabitants, few natural resources and no farm produce. What would they distribute and to whom? Who would pay for its maintenance? A J F Dulley spent four years excavating there in the 1960s. He found the remains of a 13th century quay and some evidence of 13th century Continental trade, but nothing dated before 1200.
Several other contemporary accounts say that William headed for a safe anchorage that was well known to his sailors and navigators. Pevensey, or at least Pefenesea, which many believe to be an earlier name for it, is mentioned four times in the Anglo-Saxon Chronicle as a maritime refuge. It sounds like a safe anchorage that Norman sailors would know. Something must be wrong.
Figure 6: Pevensey Lagoon in medieval times, based on map by Tom Chivers
According to Salzman and others, medieval Pevensey Levels was not a marine bay so much as a shallow coastal lagoon, like Venice, covered in tidal mud flats. It was retained by a shingle spit known as The Crumbles that roughly followed the modern coastline.
Note how Tom Chivers' map (Figure 6) shows the lagoon dotted with place names ending ‘ea’, 'ey’ or 'eye’. In this area, as local name expert Simon Mansfield once told us, 'ea' means island. These are surviving places that were inhabited islands during Saxon times.
Pevensey was 2km inland of the proper coast. Its name ends 'ey' but it was never an island, because a Roman road ran over land that connected it to the mainland. Ships could have been deliberately beached on the mud flats near Pevensey, but it would have been a risky business. We explain below why we think that Pefensea did not refer to modern Pevensey but to somewhere more out to sea.
It is implausible that the Normans would intentionally land at Pevensey and William took steps to avoid landing anywhere unintentionally. Poitiers says that the Normans moored on a sandbank off St Valery to avoid any risk of arriving at an unfamiliar or dangerous anchorage in the dark. Poitiers, Carmen and Malmesbury say that they moored again off the English coast to wait for daylight and the tide, presumably to make sure they avoided cliffs, sand banks and muds flats. Even if we imagine that adverse weather or poor navigation did force the Norman fleet into Pevensey Lagoon, they would simply have landed on the east bank which was on the lee side of the inlet, close to their destination and accessible along the deep channel of the Ash Bourne.
A transfer from Pevensey, or anywhere else, to the Bourne estuary and/or the Priory valley below modern Hastings is equally implausible. The Bourne was a short gorge with no strand. The Priory was a marshy inlet with less than a mile of strand, unable to fit more than a third of the Norman fleet. The mouth of the Bourne and Priory are at the windward end of 4 miles of perilous sea cliffs. A minor sailing error, a change of wind or a sudden stiff breeze could have dashed half the Norman fleet. Those same sea cliffs could have trapped the Norman fleet inshore for weeks on the prevailing breeze. It might have been disastrous if they came under siege, which is more than likely given that the place the Normans are supposed to have camped, at the end of a steep sided spur, was perhaps the most siege prone place on the Hastings Peninsula.
The only explicit evidence that the Normans landed or camped near Hastings is in the Chronicle of Battle Abbey (CBA) which says that the Normans camped at Hastinges ‘not far’ from where they landed, then implies that Hastinges was in a direction from Battle Abbey that points to Hastings Castle. The direction description is in a section of CBA that cannot be trusted - for reasons we return to below – but even if it is right, it could be a misleading anachronism. We explain in the Place Names section below why we think that the Normans referred to Hæstinga port as Hastinges until some time in the 12th century when Hastings Castle took over that meaning. It is possible then that the first reference was to Hastinges’s meaning at the time of the invasion whereas the second was for its meaning when CBA was written.
Historians also have toponymic evidence. Most of the contemporary accounts say that the Normans landed and/or camped at Hæstinga port, or somewhere that sounds like it. It is widely believed that Hastinges castle and rape, from which modern Hastings gets its name, was named after Hæstinga port. Perhaps they were both named after a nearby settlement of the Jutish Hæstingas tribe. In either case, the argument goes, they were probably adjacent. Thus, if the Normans landed at Hæstinga port, it must have been below modern Hastings in the Bourne estuary or the Priory valley or both. The argument is credible. Probably 90% of the places in southern England with names ending -ing or -ingas derive from similar sounding coterminous Anglo-Saxon settlements. But there are exceptions.
At least three contemporary accounts suggest that Hæstingas referred to a region rather than a settlement. If so, Hastinges castle and Hæstinga port could both have taken its name without any implied juxtaposition, apart from that they were both in that region. The other accounts do imply that Hæstingas, or somewhere sounding like it, referred to a Saxon era settlement, but this does not mean it was coterminous with Hæstinga port or modern Hastings. Winchelsea and Romney, just up the coast from modern Hastings, are both named after Anglo-Saxon settlements that were several miles from their modern namesakes. It is quite feasible then that Hastinges castle, and therefore modern Hastings, was several miles from a Hæstingas settlement and/or Hæstinga port. Turning this around, Hæstinga port could have been several miles from Hastings Castle and modern Hastings.
There is no shortage of arguments that Hæstinga port was not in the Bourne/Priory valleys. As we say above, they are at the windward end of 4 miles of sea cliffs, which would jeopardise landings and which might trap ships inshore for weeks. According to the Hastings EUS, the area around Hastings has been subject to 22 archaeological excavations since 1968, plus dozens earlier, without showing any evidence of a Saxon era port or of Saxon era occupation. De Viis Maris, written in the late 12th century, says that there was no port at Hastinges, explaining that the nearest ports were 7 miles to the east at Winchelse and 8 miles to the west at Penresse. If there was no port at Hastings after 100 years of urban development, it seems implausible that there was one when it was virtually uninhabited in 1066.
Above all, a major international port like Hæstinga port needed goods to export and/or a densely populated hinterland to draw imports. The Bourne/Priory valleys had neither. East Sussex was sparsely populated with less than a thousand families according to Domesday. The closest city was Canterbury, 35 miles away, which had its own port. East Sussex did have the biggest iron bloomeries in the country and the greatest salt production in southern England. All of it was shipped afar by sea. But the only active iron and salt production in medieval times was in the Brede basin, on the far side of the Hastings ridge. There is no credible reason why bulk goods like iron or salt would be carried up, over and down the Hastings Ridge when they could be shipped out on the Brede.
In summary, the Normans probably did land and/or camp near Hæstinga port. The orthodoxy that it was in the Bourne or Priory valleys below modern Hastings is based on the flimsiest of evidence. It has not been contested because there has been even less evidence that Hæstinga port was anywhere else. We will explain where we think it is more likely to have been in the next section, 'The Landing'.
A profusion of place names
Of course, none of the contemporary accounts actually say that the Normans landed or camped at Pevensey or Hastings, because none of them were written in modern English. Most were written in Latin. The exceptions are Roman de Rou and Benoît which were written in Old French, and the Anglo-Saxon Chronicle which was written in Old English. In those pre-map days, written place names were transliterated from the way they were spoken. Every author had their own stab at it. Those mentioned in relation to the landing include:
Pevenesæ, Pefenesea, pefnes ea, Peneuessellum (one 's’ or two), Pevenessellum, Peuenesea, Peuenesel, Penevesel, Penress; Hæstingas, Hastingas, Heastinga, Heastingum, Hestingan, Hestinga, Hestenga, Hastingae, Hastingum, Hastinges, Hastingis, Hastingues
These are the place names that get translated into modern English as Pevensey and Hastings to form the tradition that the Normans landed at Pevensey and camped at Hastings. The only justification for the translations is that modern Pevensey and Hastings are in the vicinity with similar sounding names. Lazy translation in our opinion. Perhaps we can do better.
The ‘P’ names can be narrowed down. Latin had no 'v’ sound, the closest being f. Before the Renaissance, the Latin letter v was pronounced u. Thereafter u was added to the alphabet. The choice was down to when and who copied or transcribed the original manuscript. In this document we will generally substitute v by u in Latin transcriptions of place names. Also, Latin long e and Latin short i were pronounced similarly and were interchangeable in Latin transcriptions of place names. The Old English letters f and v were allographs, used depending on whether the sound was in the middle or ends of a word. We will generally use f. Applying these substitutions and removing declensions leaves three root names: Pefenesea, Peuenesel and Peneuesellum. They have enough similarities that they could refer to one place, and enough differences that they could refer to two or three places.
The 'H’ names are easier to whittle down. There is a widespread consensus - i.e. Wikipedia says - that the name is Old English, deriving (as we mention above) from the Jutish Hæstingas tribe. They were the first Anglo-Saxon settlers of the Hastings Peninsula. -ingas -inga -inge -inges and -ingum are Old English declensions of the same stem. There was no diphthong æ in the Classical Latin alphabet. It was often substituted by a or e in transcriptions and transliterations of proper nouns. As above, Latin long e and Latin short i were interchangeable, and declensions can be removed. Thus, most (or perhaps all) of the place names in the ‘H’ group might refer to the same place. It cannot be modern Hastings.
An inland landing
Figure 7: Tapestry Panel 39
Our first objective was to locate the landing place. Our starting point was Bayeux Tapestry Panel 39 (Figure 7). It shows horses being unloaded as the invaders arrive in England. To the right are a row of empty boats. Their masts are down and they are on the land side of the esquire. They must have been dragged up onto a beach or riverbank ... well, apart from the two that seem to be self-levitating, perhaps.
Tapestry Panels 6
Tapestry Panel 34, left; Tapestry Panel 36, right
Panels 6 and 34 (Figure 8a and 8b left) show how the Tapestry depicts anchors being used in shallow water. Panel 36 (Figure 8b right) depicts empty boats in shallow water that are tied to poles. All of these must be on a marine shore where there was a danger of them floating away on the tide, or getting blown away by a storm. The boats shown in the invasion are not tied or anchored. They are sheltered and above the tide, which probably means they are in an estuary or inlet.
This is not a new idea. Nick Austin uses exactly this argument in “Secrets of the Norman Invasion” to support his theory that the Normans landed in Combe Haven. It was pointed out to us on the Reading Museum Tapestry replica while it was on display at Battle in 1966. The guide just (wrongly) assumed that they were in the bay at Pevensey.
An inland landing would only contradict one contemporary account, Orderic Vitalis, who specifically says the Normans landed on the seashore (‘littus maris’). He has to be wrong. The only coastal strand in the vicinity was at modern Bexhill. It was a three-mile wide narrow-necked peninsula in those days. It looks siege prone, perhaps even by a garrison size force. In general, Orderic repackaged other accounts. We guess he got the wrong end of the stick. Not only would a Bexhill landing invite a disastrous siege, but it would give away the potentially huge benefit of splitting the defence (as discussed in the Wargames blog).
Several of the other contemporary accounts support an inland landing. Carmen says: “Since leaving the sea behind, you seize a sheltered strand”. It uses the term ‘litora’ rather than ‘littus maris’ which usually means an inland strand. The only ‘sheltered’ places are inlets and estuaries. Carmen also says that the landing was in a “calm basin”, which means an estuary or inlet. Baudril of Bourgueil quotes William saying: “Whither would ye flee? Our fleet is far from the shore: we removed all hope of escape when we moved away from that”. The Warenne Chronicle reports that: “without any resistance between the forts of Hastinges and Pevenessullum he entered the land of the English". If he entered the land of the English still aboard ship, he sailed into an estuary or inlet.
Wace mentions the Norman ships being at anchor, which would be unlikely if they were in an estuary or inlet, but it is a misunderstanding. What he actually says is: “together they cast anchor and ran onto dry land; and together they discharged themselves". They cast anchor before running aground. We interpret this to mean that drop anchor to form a line astern while still in the centre of the estuary, then they simultaneously sail, row or pole ashore. This is exactly what we describe in the Wargames section as the best method to split the defence.
Wace makes it sound like they let out their anchor lines as they came ashore, presumably in case they had to quickly haul themselves back into the river after an ambush. Tapestry Panel 39 does not show any anchor lines. This is understandable for the boats that have already been unloaded, because they would weigh anchor before being dragged up the bank. Perhaps there should be an anchor line on the ship that is unloading horses. Maybe the artist or embroiderers missed it. But the ship is being held steady by a man with a pole. This would not be necessary if the ship were still anchored. We think it more likely that they weighed anchor before reaching the shoreline.
Landing site candidates
Wace says that the Norman fleet lands together. It makes military sense for them to land together, thereby stretching the defence to give the best chance of establishing a bridgehead. This was as true for William as it was for D-Day 900 years on.
A landing in some of the smaller estuaries and inlets around the Hastings Peninsula can be eliminated by calculating how much landing space they needed. That depends on the number and size of the boats, which in turn depends partly on how many troops and horses they carried. We explain our calculation in our book. In summary, we think the Norman army had roughly 2,000 cavalry and 4,000 infantry, and that they arrived on 700 longships plus several hundred cargo skiffs. We estimate that the longships had an average 4m beam.
The only practical way to effect a simultaneous mass landing in an estuary or inlet was to anchor line-astern midstream, then to simultaneously sail, row or pole ashore. The ships would therefore be separated on shore by at least the difference between their length and width. They would need to be separated by at least 3m anyway, in order to make space to unload horses and cargo over the side and/or to deploy oars if they needed to leave in a hurry. 700 longships with an average 4m beam separated by a minimum of 3m, means that the Norman battle fleet would have needed roughly 5km of landing space. Cargo vessels might have needed another 2km, although some of them could have landed elsewhere.
Here is Figure 4 again. Note that Pevensey Lagoon (now the Pevensey Levels) and Combe Haven were open to the sea in those days and that the estuaries were deeper and much wider than they are now. The only estuaries or inlets surrounding the Hastings Peninsula that were big enough to hold the Norman fleet are the Brede, the Ash Bourne, Combe Haven and, at a squeak, Hooe Haven. We are convinced that the landing must have been in one of them. Hooe Haven, at about 4km in length, is marginal. We will give it the benefit of the doubt for now.
Primary source abbreviations
Thirteen of the contemporary accounts hold credible unique information about the invasion. They are usually referred to as the primary sources. Here is a list with abbreviations we sometimes use:
- ASC = Anglo-Saxon Chronicle (reasonably contemporary with events)
- ASC-C, ASC-D = The versions of ASC that cover the invasion
- Tapestry = Bayeux Tapestry; finished c1077
- Benoît = Chronique des Ducs de Normandie; Benoît de St-Maure; c1170
- Carmen = Carmen de Hastingae Proelio; c1067
- CBA = Chronicle of Battle Abbey; c1170
- Chronicon = Chronicon ex Chronicis; John of Worcester; c1125
- CKE = Gesta Regum Anglorum; William of Malmesbury; c1135
- Domesday = Domesday Book; 1086
- Orderic = Historia Ecclesiastica; Orderic Vitalis; c1125
- Wace = Roman de Rou; Master Wace; c1160
- Warenne Chronicle = Chronicon monasterii de Hida iuxta Winton; c1200
- WJ = Gesta Normannorum Ducum; William of Jumieges; c1070
- WP = Gesta Guillelmi; William of Poitiers; c1072
Primary source landing and camp descriptions
All these primary sources specify place names and/or geographical descriptions associated with the landing. This is what they say in English but with untranslated place names (except for some v/u substitution):
- WP says that after leaving St Valery the fleet moors offshore, waiting to ensure that they do not arrive in England before dawn at a “dangerous or unknown anchorage". It means their destination was a familiar and safe anchorage.
- CKE says of William that: "The earl himself first launching from the continent into the deep, awaited the rest, at anchor, nearly in mid-channel. All then assembled round the crimson sail of the admiral's ship; and, having first dined, they arrived, after a favourable passage, at Hastingas". Malmesbury is saying that they moored off St Valery, sailed, ate breakfast, then arrived at Hastingas.
- Carmen says that: “the looming rocky coast" did not discourage William’s invasion.
- Orderic says that, upon hearing of Tostig’s invasion, Harold: “withdrew his ships and troops from Hastingas et Peneuesellum, and the other seaports opposite Normandy".
- Carmen (Kathleen Tyson translation) says: “On the open sea you moor offshore; You caution to take in the sails, awaiting the morning to come; But after the dawn spreads red over the land, and the sun casts its rays over the horizon; You order the sails set to the wind to make way.”
- CBA says the Normans: “Arrived safely near castrum Peuenesel. The Duke did not remain long in that place, but went away with his men to a port not far distant named Hastinges".
- CBA (Lower and Searle translations) says that Hechelande, which it describes being northwest of and adjacant to Telham on the ridge, is in the direction of Hastingarum from Battle Abbey.
- WP says that William's ship lost contact with the rest of the fleet: "In the morning, a lookout at the top of the mast declared that he could see nothing but sea and sky. They anchored at once." William had breakfast. By the time he had finished, the rest of the fleet was in sight.
- Warenne Chronicle says: “without any resistance between the forts of Hastinges and Pevenessullum he entered the land of the English"
- WP says: “Borne by a favourable breeze to Peneuessellum, he disembarked with ease and without having to fight his way ashore"
- WP (Foreville translation) says: “The rejoicing Normans, once they had landed, occupied Peneuessellum, where they built their first camp, and built another at Hastingas to provide a refuge for themselves and a shelter for their boats".
- Brevis Relatio (our translation) says that Duke William and his fleet: "arrived in England, by the favour of God, near the fortress of Pevensel. After a short delay he arrived with his whole army at another port not far away named Hastingas"
- WJ says that William: “Landed at Peneuessellum where he immediately built a castle with a strong rampart. He left this in charge of some troops and, with others, hurried to Hastingas where he built another".
- Orderic says: “... and reaching the coast of England, where they met no opposition, joyfully came ashore. They took possession of Peneuesellum and Hastingas, the defence of which was entrusted to a chosen body of soldiers to cover a retreat and to guard the fleet".
- Benoît de Sainte-Maure says the Normans: “Arrived at Pevenesel, at a port/harbour beneath a fortress handsome and strong"; and later: “The Count came to Hastinges without staying" (i.e. William did not stay at Pevenesel)
- Tapestry Panel 38 is captioned: “Duke William in a great ship crossed the sea and came to Pevenesæ";
- Tapestry Panel 40, showing a scene immediately after Panel 38, is captioned: “The knights hurried to Hestinga"
- Chronicon says that William: “Had moored his fleet at a place named Pefnesea"
- ASC-D (Ingram translation) says: “Meantime Earl William came from Normandy to pefnes ea on the eve of St. Michael's mass; and soon after his landing was effected, they constructed a fortress at Hæstinga port”.
- ASC-E says: “Meanwhile Count William landed at Hestingan on Michaelmas Day"
- Wace says: “The ships steered to one port; all arrived and reached the shore together; together cast anchor, and ran on dry land; and together they discharged themselves. They arrived near Hastingues each ship ranged by the other's side."
- Wace says that a messenger tells Harold: “The Normans are come! They have landed at Hastingues"
- Carmen (Kathleen Tyson's translation) says of the landing: “... the happy land owed to you embraced you and yours in a calm basin".
- Carmen says: “One Englishman kept hidden under the sea cliff". He watches the Normans disembark, then rides off to tell the King.
- Wace says that an English knight: “posted himself behind a hill" to watch the Normans disembark, then rides off to tell the King.
- Carmen says:“You restore the strongholds that were lately destroyed"
- Chronicon says that Harold: “Gave them battle nine miles from Heastinga, where they had built a fort”.
- Wace says: "They [the knights] formed together on the shore, each armed upon his warhorse. All had their swords girded on, and passed into the plain with their lances raised .. When they [the carpenters] had reached the spot where the archers stood, and the knights were assembled, they consulted together, and sought for a good spot to place a strong fort upon". They then assembled a kit fortress they had brought with them. It was complete by that evening.
- Wace says that on their first day after landing they went on a raid. “They held their course along the coast; and on the morrow came to a fortress named Penevesel", which they plundered. He goes on to say that: “the English were to be seen fleeing before them, driving off their cattle and abandoning their houses. All took shelter in the cemeteries."
There are some post-invasion documentary clues about the landing place names too. Four contemporary accounts describe Odo’s rebellion in 1087 during which he was besieged at Anderitum (Pevensey castle). Three refer to it as the fortress of Pefenesea, the other as the fortress of Pevenesel. Norman writs such as A.4171 refer to Anderitum as 'castrum Peuenesel'. The area around Pevensey castle became known as ‘rap de Peuenesel’. These well-known references suggest that Pefenesea and Peuenesel were cognates, and that both were alternative names for Pevensey.
It is no wonder then that historians interpret the invasion accounts as compelling evidence for a landing at Pevensey. Three of them imply the landing was at Pefenesea, which is thought to be an early name for Pevensey. Two imply it was near the fortress of Peuenesel, which is thought to be another name for Pefenesea. Three more say the landing was at Peneuessellum, which could be a Latin corruption of Peuenesel and therefore yet another name for Pevensey. If so, eight apparently independent references seem to say that the landing was at Pevensey. It sounds conclusive, yet somehow or another, it has to be wrong.
Assuming the contemporary invasion accounts are reasonably accurate – and there is no obvious reason they would not try to be accurate about landing or camp place names - they must have been misinterpreted. Our analysis is here. It is rather arcane and we do not want to discourage casual readers, so here is a summary of the conclusions:
- Hæstingas was the Old English name for the Hastings Peninsula. This is its meaning in Saxon Charters and Anglo-Saxon accounts (ASC and the Tapestry). Even though the Tapestry was created for Norman viewing, we think its Hestinga and Hestenga referred to the Hastings Peninsula, perhaps because it was embroidered by Saxon nuns under instruction from Normans who could not read.
- Hæstinga ceastre was the Old English name for a former Roman fortification and probable Alfredian burh on the Hastings Peninsula. Norman accounts do not mention Hæstinga ceastre directly, but some mention pre-existing fortresses in/at Hastingas.
- Hæstinga port was the Old English name for the international port on the Hastings Peninsula. Its docks and warehouses were at Winchelse on a shingle spit at the mouth of the Brede. Its harbour, dry docks, maritime supplies, mint and dock administration were at Winchelsea.
- Hastinges, often Latinised to Hastingas or portus Hastingas, was the Norman name for Hæstinga port.
- Pefenesea was the contraction of pefenes ea, an island harbour some 2km southeast of modern Pevensey. The term pefenes ea means 'pefen's isle' in Old English, referring to a Saxon named pefen.
- Peuenesel was the Frankish translation and transliteration of pefenes ea (i.e. the contraction of peuenes îles) used by Franks and Normans to refer to the island harbour some 2km southeast of Pevensey.
- Castrum Peuenesel was the Norman name for the castle at Pevensey, also known as Anderitum by the Romans and Andrades ceaster by the Saxons.
- Rap de Peuenesel was the Norman name for the area surrounding pefenes ea. It included the newly refurbished castle at Pevensey.
- When pefenes ea was destroyed by storms in the 12th century, the local population relocated to the castle at Pevensey, taking the name of their settlement with them. This is how the settlement around the castle came to be known as Pefenesea (and ultimately Pevensey). It is exactly analogous to what happened at Winchelsea.
Figure 16: East Sussex medieval place names
This place name schema matches nearly all the landing accounts. It sounds contrived and implausibly complicated. It is not. Hæstinga, Hæstinga port, Hæstinga ceastre, pefenes ea and Andrades ceaster followed the normal Saxon naming convention. Of these, before 1066, most Normans only dealt with Hæstinga port and the harbour at pefenes ea, which they referred to by the Norman translation Portus Hastinges, usually abbreviated to Hastinges or Hastingas, and the Frankish translation Peuenesel, respectively.
Norman ex-pats in England assimilated, gradually adopted local place names. They must have adopted the name Wincenesel for Winchelse and Iham for Winchelsea because they are listed as ports in the Pipe Rolls and Ship Service. Presumably, they narrowed the meaning of Hastinges from Hæstinga port to some small part of Hæstinga port because it has a separate sub-listing in Domesday. We guess that it was taxed differently from the rest of Rameslie, perhaps because it had a military or government purpose after the Conquest, at least until the stone castle was complete at Hastings.
But the meaning of these place names evolved through the 12th and early 13th centuries with the changing geography and politics of the region. The major geographic change was the gradual erosion and eventual 13th century destruction of the Camber and Crumbles shingle spits. Ports and harbours on those spits were forced to move to the new coast. Among them Winchelse, Pefenesea and Romney, all of which played a part in the Norman landing. They took their occupants, businesses and place names to Winchelsea, Pevensey (then known as Peuenesel) and New Romney respectively.
The major political change was that William constructed castles along the East Sussex coast, each controlling a rape. The administrative control of the Hastings Peninsula passed to Hastings Castle. When Domesday was compiled, the rapes were known by the name of the Norman baron that controlled them. During the 12th century, those rapes took the name of their castle. We think from its listing in the the Pipe Rolls that the settlement around Hastings castle was known as ‘Nova Hastings’ , presumably because the name Hastinges was still being used to refer to part of Hæstinga port. By the 13th century, the castles at Hastings and Pevensey, and the settlements around them, had got the familiar names that have lasted through to the present day.
Place name evolution mainly affects the Anglo-Norman accounts - Historia Anglorum, Chronicon ex Chronicis and the Chronicles of the Kings of England - all of which were written in the early 12th century during the transition. We try to interpret case by case whether place names refer to their meaning at the time of the invasion or at the time they were written. Some of them were mixed. The same is true of CBA and Wace, for which there was a 100-year gap between the invasion and when they were written. We have to be particularly careful about place names in first-hand quotations, which presumably had the place name meanings at the time they were said rather than when their chronicle was written.
The occasional and inconsistent expansion of Hastinges and Hastingas to Portus Hastinges and Portus Hastingas is another source of confusion. As far as we can see, it is totally arbitrary. Normans in Normandy had no reason to use the 'port’ suffix because they had no dealings with anywhere on the Hastings Peninsula apart from the port. This is exactly analogous to their use of the name Douvres to mean the port of Dover and to our modern use of the term Calais to mean the port of Calais. We guess that the occasional switch to the formal Portus prefix was for the sake of those that were unfamiliar with English geography.
The major remaining discrepancies between our landing narrative and the orthodox narrative is caused by historians' understandable desire for simplicity. They knew from the most trustworthy accounts that the Normans had an initial and a secondary landing, building a fortress at each. Twenty different place names are used in the landing accounts. They are listed at the beginning of this text, with names including Hæstinga port, Hastingas, Hestinga, Pefenesea, Peuenesel and Peneuesellum. The obvious solution was that the ‘P’ names referred to the first landing site, and the ‘H’ names to the second.
The orthodoxy is wrong because most of the ‘P’ accounts refer to the Norman arrival off the coast of England rather than their initial landing. The real narrative is that the Norman fleet arrived near pefenes ea (often contracted to Pefenesea and known to the Normans by its Frankish translation Peuenesel), an island harbour on the coast about 2km southeast of Pevensey fortress. The Normans waited there for sunlight and the right tide, then sailed on and landed at Peneuesellum. They stayed for just one day, before moving to Hæstinga port on the south bank of the River Brede, where they stayed until a week after their victory.
So, all the primary source landing accounts are accurate, as far as they go. None are complete. This is understandable. The Normans only moored for a few hours near pefenes ea and they only occupied Peneuesellum for one day. Nothing happened at either place. All the accounts that omit the arrival and/or the Peneuesellum landing were heavily abridged. It makes sense that some would redact these events. We liken it to Ellis Island. Nearly all of the 12 million U.S. immigrants that landed at Ellis Island would naturally have reported that they landed at New York.
Peneuesellum and the first Norman landing
One important place has been omitted from the place name definitions above: Peneuesellum, where three of the earliest and most trusted primary sources specifically say the Normans landed. One of them says that William returned there six months after the invasion, reiterating that it is where the Normans landed. We feel sure they are right, but they provide few clues about where it was.
Historians think that Peneuesellum referred to Pevensey. The only evidence is that they were both in East Sussex and that Peneuesellum looks like a Latinised corruption of Peuenesel, which became the Norman name for Pevensey. We cannot accept it. Franks and Normans consistently referred to pefenes ea as Peuenesel before and after the Conquest. There is no obvious reason these three accounts would buck the trend, and the Normans could not have landed near Pevensey for the many reasons we list above.
The first thing to note is that Peneuesellum only appears in Norman accounts. It is a Latin format name but with no obvious Old English place names in the vicinity upon which it might have been based. The only likely reason that the Normans might have their own name for somewhere in Sussex that has no English counterpart is that it was part of the land belonging to the Abbey of Fécamp. In this vicinity, this means that it was part of Rameslie manor.
There are four references to Peneuesellum. WP and WJ say that the Normans landed at Peneuesellum then went to Hastingas, making it sound like they were adjacent. Orderic says that Harold withdrew his ships and troops from “Hastingas et Peneuesellum”. Later he says that the Normans occupied “Peneuesellum et Hastingas” upon landing, leaving a body of men to guard them both. Note the reversal of names, so at least one of them was not a proper noun. Orderic's statements only make sense if Peneuesellum was adjacent to Hastingas. Warenne Chronicle says that the Norman fleet passed between a fortress at Hastingas and a fortress at Pevenesellum. These are all Norman sources (well, Orderic was half-English but wrote for a Norman audience), so their Hastingas meant Hæstinga port. Warenne Chronicle's Hastingas fortress was Hæstinga ceastre, which we place at Winchelsea. Assuming its Pevenesellum meant Peneuesellum (note the n/v switch), this only makes sense if Peneuesellum faced Winchelsea across the Brede.
So, Peneuesellum was adjacent to Winchelsea and perhaps faced it across the Brede. In the first edition of this book, we mused that Peneuesellum referred to Rye. There is no evidence of Saxon era occupation of Rye, but 150 years later it had 20% the trading volume of Hæstinga port, which was the busiest port east of Southampton. That is astonishing growth. We think Rye was in Rameslie manor and was therefore a possession of the Abbey of Fécamp. We speculated that Hæstinga port ran out of capacity after the Conquest, because the Normans hugely increased trade with Normandy, so the Abbey of Fécamp had to develop a small Saxon era harbour at Rye. We said that they probably referred to that harbour as Peneuesellum. We were wrong.
Soon after we published our book, Kathleen Tyson published her Carmen translation, in which she suggests that the name Peneuesellum means ‘fort in the wash’. Her derivation looks good to us. There is no evidence of Saxon era occupation at Rye, let alone a fortress. If Kathleen is right, Peneuesellum was somewhere else.
If Peneuesellum was not Rye, we made another mistake. In the first edition of this book, we suggested that ‘Hastingas et Peneuesellum’ was probably a composite proper noun, referring to a port named ‘Hastingas et Peneuesellum’, which might have been a Norman name for the harbours at Hæstinga port and Rye. It was always a bit tenuous because of Orderic’s later reference to the Normans occupying ‘Peneuesellum et Hastingas’, with the names reversed. Rye was the only feasible harbour adjacent to Winchelsea, Winchelse aside, because there were no natural inlets upstream on either bank. If ‘Hastingas et Peneuesellum’ was not a proper noun, then Orderic referred to a place named Hastingas and a place named Peneuesellum.
This means that Orderic was trying to say that Harold withdrew his ships and troops from Hastingas and from Peneuesellum. If Peneuesellum had standing troops, they were probably stationed at a garrison fortress, which would corroborate Peneuesellum referring to a fortress.
Kathleen Tyson thinks Peneuesellum was at modern Udimore village. She notes it was where William built a grand manor house now known as Court Lodge. She speculates that it was built on the ruins of a fortress, and suggests that Udimore would be a 'magnificent' place for a navigation beacon and an ideal place to tax cargo passing from Hæstinga port to Kent on a causeway across the Brede. We are sceptical. Udimore has shown no archaeological evidence of pre-invasion occupation, let alone a fort. It has a severely restricted sea view pointing to Boulogne in what was Hauts-de-France rather than Normandy. It is difficult to believe that the Saxons had the wherewithal or skills to construct a pioneering 2km tidal causeway, especially when there was a low-water ford and a bridge a few miles upstream. Udimore was 6km from the end of the Udimore peninsula, so it was barely ‘in the wash’.
WP, WJ and Orderic make is clear that Peneuesellum was near to Hæstinga port, probably adjacent. ‘In the wash’ suggests it was somewhere that is lapped by tidal waters on both sides, which means an island or a peninsula. Peneuesellum's name suggests it was part of the Abbey of Fécamp's Rameslie manor. So, Peneuesellum was on a peninsula adjacent to the Brede and close to Winchelsea but it was not Rye.
By our reckoning, Peneuesellum must have been on one or other bank of the Brede towards the seaward end of the estuary. Tapestry Panel 40 says that the knights hurry to Hestinga to forage for food. The knights would not have gone to a port to forage for food, so this Hestinga cannot mean Hæstinga port. The only alternative is that it referred to the Hastings Peninsula, in which case the knights were not hurrying from the Hastings Peninsula. Therefore, they must have been hurrying from somewhere on the north bank of the Brede. Therefore Peneuesellum, where the Normans first landed, was somewhere on the north bank of the Brede, towards the seaward end of the estuary.
It is interesting to speculate why the Abbey of Fécamp coined their own term for part of the Brede’s north bank. We guess that it relates to the Saxon manor of Bretda. Charter S982 clarifies that the manor of Bretda was included in Cnut’s gift to the Abbey of Fécamp. The obvious reason for this clarification is that it was previously in dispute. We suspect that Bretda originally contained the salt-pans that were gifted to Abbey of St Denys then to the Abbey of Fécamp. Neither of those Charters mentioned Bretda, perhaps leaving the rest of Bretda manor with its original Tenant in Chief. Therefore, the land that contained the salt-pans needed a new name.
Figure 17: Yeakell & Gardner Cadborough in 1770
Perhaps we can be more specific about Peneuesellum’s location. It makes no difference to our theory if we are wrong but we think that Peneuesellum referred to Cadborough. Yeakell & Gardner (above) label Cadborough as Caresborough in 1770. The Brythonic term ‘caer’ means fortress. If the name was Brythonic, it must have predated the Saxon invasion, which means it must be Roman. The earliest ‘boroughs’ were Saxon burh fortifications. These burhs were often built at former Roman fortifications on promontories overlooking places that the Vikings had previously raided. Cadborough fits the bill, at the eastern tip of the Udimore peninsula, where it was definitely ‘in the wash’ and it had the widest sea view of anywhere on the Udimore Peninsula.
We are not suggesting that the Normans actually landed below Cadborough. It had a dangerously narrow shore and steep cliff. Rather we think that it was the closest place to the landing that had a name, at least one that Normans would recognise. Most likely, they landed upstream of Cadborough, between Float Farm and Brede ford. This would be below Court Lodge, so Kathleen Tyson may well be right that Court Lodge commemorated the place where the Normans landed.
Figure 18: Brede estuary 11th century place names
A Brede landing
Commercially and toponymically, as we explain above, it looks like the Norman fleet passed though Hæstinga port at the mouth of the Brede and landed on the north bank of the Brede estuary. What about the other clues?
Logistics is unhelpful. The Brede was crossed by the only paved Roman road in the region, which would have been ideal for easy plundering and foraging, but the other candidates were close to ancient trackways. If those tracks were well maintained, they might have been almost as good. All four candidates were close to a freshwater stream. Each had at least one nearby hill/ridge on the Hastings Peninsula that would have made a good camp: Hooe Haven and the Ash Bourne had Standard Hill; Combe Haven had Upper Wilting or Green Street; the Brede had Cackle Street and Cock Marling to the north, Winchelsea, Snailham, Starlings and Cottage Lane to the south.
William would have seen the “looming rocky coast” of Beachy Head wherever he landed. All the landing site candidates were in a “calm basin". None of the landing site candidates has surviving remains of a fort. No confirmed 11th century mottes have survived. There are sea-level moats at Barnhorne and Wartling, and a side-hill moat at Lower Snailham on the south bank of the Brede, but they are not necessarily mottes and there is no reason to believe they were earlier than 13th century. These clues do not narrow down or preclude any of the landing site candidates.
The rest of the non-place name clues point to a Brede landing. Carmen says that an English spy hid under a sea-cliff to watch the landing. The Brede was the only candidate overlooked by sea cliffs, namely Cadborough Cliffs. It had the only paved road for freight distribution and perhaps a low-tide canal. It is the only candidate with a strand long enough to fit the Norman army. Warenne Chronicle says that the Normans passed a fortress at Hastinges on their way to land. There is a Roman enclosure at Wilting in Combe Haven, but the Normans would not have passed it on their way to land. If Combe Haven had a Roman enclosure for protection and administration, the far bigger port in the Brede must have had one too. Its most likely location is the hilly peninsula of Winchelsea.
Carmen says that the Norman fleet arrived at 'safe landing grounds' at the third hour of the day. It cannot be referring to the actual landing grounds, because the word 'safe' would be extraneous and misleading, being that they expected it to be defended. We interpret it to mean safe from natural hazards, which along the East Sussex coast most likely refers to the sea cliffs between Hastings and Fairlight. These cliffs would not jeopardise a landing in the Ash Bourne or Hooe Haven, but they would jeopardise a landing in Combe Haven or the Brede estuary, especially in a southerly breeze.
A possible alternative interpretation of Carmen's 'safe landing grounds' is where they became protected from storms. This applies to all four landing site candidates. The Brede was sheltered behind the Camber shingle spit. Hooe Haven and the Ash Bourne were sheltered behind 'The Crumbles' shingle spit, which enclosed the Pevensey Lagoon. Combe Haven was sheltered behind an island known as Bulverhythe. Assuming the overnight mooring was somewhere between pefenes ea and the Royal Sovereign Shoals, and that the Norman fleet left at dawn, three hours would be about right for the 20-mile reach to Winchelse on a southerly breeze against the tide. It seems improbably long for a ten-mile run on a southerly breeze to Bulverhythe, let alone a six-mile run to The Crumbles.
In our opinion, all the clues point to a Brede estuary landing. This brings us back to the spit crossing. The main entrance to the lagoon was 10 miles up the coast at Old Romney (Figure 15 and Figure 19). We know that the Normans did not use it, because Wace reports that several ships landed there by mistake; their crews were killed by the local inhabitants. If they landed in the Brede, they must have used a channel or canal across the spit. Andrew Pearson, Bernard Leeman and others show the crossing at Winchelse (Leeman refers to it as Old Winchelsea).
Figure 19a: Romney Marsh in medieval times; Andrew Pearson
Figure 19b: Romney Marsh in medieval times; Bernard Leeman.
It is interesting to check the order of events versus the tides and sunlight. Sunrise on the 28th September was at 5.57, sunset at 17.44, low tide at 09:45, high tide at 16:00. Carmen suggests that the Normans left their mooring place near pefenes ea at sunrise. They would have aimed to disembark simultaneously at high tide when the estuary took them as high as possible up the bank. At any other time, there would have been a risk of horses and armoured men getting stuck in the mud and perhaps drowning. Tapestry Panel 39 depicts some men and horses disembarking. They are high up the riverbank on solid ground. The lead ship would need to get far enough up the Brede estuary - say to Brede Place - before high tide. Do the timings work?
According to Carmen, the Norman fleet reached their 'safe landing place' at the third hour of the day. Whether it means safe from sea cliffs or safe from storms, it would refer to Winchelse for a Brede landing. The third hour of the day would mean the lead ship arrived at Winchelse around 09:00, 45 minutes before low tide. If they needed a minimum of, say, 30 minutes to organise the midstream anchoring before landing, the lead ship had roughly six hours to get 10km to Brede Place. Their ability to achieve this depends in part upon how long it took to cross the Camber spit, which in turn depends upon its width and when it opened.
If the Normans arrived just before low tide, it suggests that the cross-spit channel was open soon after low tide, perhaps because it was dredged. It seems likely then that the idea was to drift through the channel on the flood tide, steering with a pole. If Chesil Beach is a good model at 200m across, it would have taken around 10 minutes to make the crossing. Pearson and Leeman both depict the spit as roughly 1km wide at Winchelse, which would have taken more like 30 minutes. Worst case, assuming the channel opened soon after low tide, the lead ship had 5 hours to get 10km to Brede Place on a rising tide. The tide had to come through the main lagoon entrance at Old Romney. It would be 30 minutes late and slow; more like a lock rising than a tidal bore. Perhaps it contributed 5km, leaving the lead ship to make 5km in five hours. The Viking Ship Museum in Roskilde, Denmark, told us that a fully manned Snekka could row at 2km per hour for five hours in a neutral wind. Even the Karvi could probably row 5km in five hours. The landing would therefore be achievable in anything other than a westerly breeze.
But five hours of rowing after a sleepless night is far from ideal for a fighting force. As it happened, the English garrisons were empty. William was not to know. He expected the landing site to be defended. He expected to fight ashore and/or face a garrison counterattack. Even though his troops would have been fit and strong, William would surely have hoped for some help from the wind as they made their way up the Brede. Reaching into even a modest southerly breeze or running a gentle easterly breeze, along with the flood tide, would comfortably take the lead ships to Brede Place in five hours without any exertion.
Several primary sources say that the Norman fleet was stuck in Normandy for a month while they waited for a southerly breeze. It worked out well for William because Harald Hardrada dragged the defenders away during the wait. William was not to know. He would surely have wanted to invade earlier if possible, to take advantage of better weather and longer daylight, and to prevent his troops getting restless. Yet they could have crossed the Channel on the prevailing south-westerly breeze at almost any time.
Note that historians often interpret this to mean that the Normans were trapped in port for a month by a constant northerly breeze. They are wrong. There has never been a month of northerly breezes in the Channel in September since records began, and the nature of the weather in our latitudes suggests that there never will be. Moreover, Heimskringla makes it clear that the weather in the middle of September was mild and dry – i.e. not an arctic blast - because it was too warm for the Norse army to wear their armour on the day of the Battle of Stamford Bridge.
A Brede landing explains why the Normans had to wait for a southerly breeze. They needed to head northeast. Any northerly breeze would have scuppered a landing at any of the candidates because the Karvi would have become detached from the fleet on the Channel crossing. Any other breeze would have allowed a landing at Combe Haven, Hooe Haven or the Ash Bourne. But not the Brede. A south-westerly would have forced the close hauled Karvi to beat on the Channel crossing. An easterly breeze would have detached close hauled Karvi on the easterly run to Winchelse. A westerly breeze would have prevented the lead ships from getting far enough up the Brede estuary before high tide. If William intended to land in the Brede, he had to wait for a southerly or south-easterly breeze and a low tide between 08:00 and 12:00.
So, in our opinion, a Brede landing (as shown in Figure 20) best fits the primary source descriptions and it is logistically feasible. There is still the question of why? A mishap in the Winchelse spit crossing might have blocked the entire fleet. The tide timing was stringent and critical in the Brede and it needed a southerly breeze. A landing in Combe Haven, Ash Bourne or Hooe Haven had no such dependences. If William had been prepared to land at any of them, the invasion could have happened several weeks earlier, when his troops were less fractious, daylight longer and weather perhaps more favourable. The Brede must have had at least one compelling attraction as the landing site.
We have already mentioned one of the Brede's advantages: it was the longest estuary or inlet around the Hastings Peninsula, which made it the best place to implement the midstream anchoring ploy, outlined above and described in the Wargames blog. But Combe Haven and the Ash Bourne were nearly as good. It is a modest return for the risk of traversing the shingle spit at Winchelse. There must have been more. We think the Brede had two more unique advantages, both tied to the bank upon which they landed.
Rameslie manor, which we think lined both banks of the Brede (for reasons explained in our place name blog), is listed in Domesday with 100 saltpans. This is the greatest concentration of saltpans in southern England. Why there? One reason is that the Camber shingle spit protected the Brede estuary from storms and flooding. Another is that it has an east-west orientation. Saltpans are ideally placed on a south facing bank of a wide estuary or inlet, where they are never shaded by hills or vegetation, even when the sun is low in the sky. Most, if not all, of Rameslie's saltpans would have been on the north bank of the Brede.
Figure 21: Salt plain landing upstream of Peneuesellum
We guess that the evaporation ponds were created by the Romans using a levee between Float Farm and Brede Place and by channelling the streams running off the Udimore Ridge. Over the centuries, the ground would level and compact. The ponds behind the levee would be flooded in the Spring and evaporate over the Summer. By the end of September, the concentrated brine would have been harvested to leave a wide flat straight-edged dry plain just above the estuary bank (Figure 21). It would be perfect for a simultaneous mass landing; firm enough underfoot to support mounted horses and level enough to assemble a kit-fortress without a motte.
The Brede's third advantage was that the Udimore Ridge narrows to a pinch point just 100m across at Sowdens. William was expecting a counterattack from the garrisons at Lympne and/or Pevensey. It would come from the west, either along the ridgeway or along the strand. William had to block both. He had brought a kit fortress, but it could not block the ridgeway and the strand. Moreover, there was no time to dig a motte, so it had to be constructed on clear level ground, which means not the ridgeway. We think that William blocked the ridgeway at Sowdens with barricades and the strand with the fortress, as depicted on Figure 26. This would explain why they went to so much effort to bring and assemble a fortress that was only used for one day.
Figure 22: Tapestry Panel 37
Tapestry Panel 41 (Figure 23), at the Norman landing site, offers some corroboration of this theory. Before explaining why, we need to point out that the vast majority of the Tapestry has bobbles on the baseline, like the righthand side of Panel 37 (Figure 22). This happens to be the coast of Normandy, but Pontieu, Mont Saint-Michel and the other coasts and riverbanks are the same.
Presumably then the bobbly base represents fields, meadows, dunes and scrub. The non-bobbly baseline is usually reserved for the areas in and around buildings. But the characters in Panels 40, 41 and 42 are on the baseline without a foreground building or motte in sight. It is the only riverbank or coast on the baseline. Indeed, it is the only outdoor scene on the baseline, apart from the sea camp motte.
Figure 23: Tapestry Panel 40 & 41
We interpret this to mean that the landing site was as flat and smooth as a road or motte. Wace confirms that the landing site was flat: “All had their swords girded on, and passed into the plain with their lances raised”. There were no natural plains around the periphery of the Hastings Peninsula in those days. We think the answer to both puzzles is that the landing site was a smooth plain of dried-out salt evaporation ponds.
Figure 24: Tapestry Panel 41
The huts in the background of Panel 41 (Figure 24) offer some further corroboration. One has a shingle roof, one a plank roof and one a tile roof. This at a time when nearly all dwellings in England had thatched roofs. We think they were evaporation chambers, where they boiled away water from concentrated brine to crystallise salt. If so, they would have been beyond the evaporation ponds, on the lower slopes of the ridge, exactly as depicted.
Kathleen Tyson favours a Brede landing too, through a different argument. She translates Peneuesellum to mean 'fort in the wash’. She independently worked out that the 'wash' referred to the tidal waters around the Udimore Peninsula, which means the landing must have been in the Brede.
A Brede landing also fits Jo Kirkham’s theory that the invasion was planned by the monks of Fécamp Abbey. They were Norman and therefore probably loyal to their patron William. They may well have resented Harold for having dispossessed and banished Normans from England. Wace tells us that William brought some monks from Fécamp to act as interpreters. The only place they could have learned Old English was at Rameslie, the manor around the Brede (their other property in Steyning was too recent). Having lived there long enough to learn the language, they would have known the area intimately. William would surely have tapped their local knowledge to identify the best landing place and to help develop a battle plan.
If we are right about most of the above, it solves two puzzles and perhaps an enigma. The first puzzle is why serial owners of Hæstinga port would gift such a valuable asset to monasteries: first to St Denys in the 8th century, then to Eynsham Abbey (S911) and finally Fécamp. In St Denys' case, forged charters say that a Saxon baron named Bertoald gave them the port in gratitude for their healing services. Tommyrot. We think the reason was commercial.
The administration of the entrepôt was too complicated and too expensive for normal barons. It needed quays, jetties, wharfs, canals, roads, dredging, ferries, security, warehouses, barges and bridges that were provided as an operational overhead. It was a capital-intensive business before there was an easy way to raise capital. There was no central power to provide these services once the Romans had left. The Church was the nearest substitute. They had the funds and the skills to run major infrastructure projects. Clerics alone could read, write and sum, the essential skills for keeping records and ledgers. We think that the port’s owners had to give it to one abbey or another, in exchange for a cut of the tolls, to prevent it decaying into disuse with no revenue.
The second puzzle is Wace’s description of the first raid. He says that the Normans follow the coast, then loot a fortress named Penevesel while the locals drive off their cattle and hide in cemeteries. This has always been interpreted to be ‘castrum Peuenesel’ (i.e. Anderitum at Pevensey). But Pevensey would have been too marshy for cattle and too sparsely populated for loot or cemeteries or even a church. Also, Wace uses the ‘n/u’ spelling Penevesel, as in Peneuesellum, rather than the ‘u/n’ spelling of Peuenesel. He would not have used the Latin -um suffix, because he was writing in Old French. We think he was referring to Peneuesellum, at Cadborough. It was just along the coast from where we think they landed, exactly as Wace describes. We think Cadborough had a dilapidated Roman fortress and a church, and perhaps a burh, so it would have fitted Wace’s description of the first day raid.
After our book was published, a Sowdens resident told us of a local lore that St Mary's Church was moved to Udimore in the 12th century from somewhere closer to the sea. The oldest part of St Mary's is indeed 12th century. No older foundations have been found in the vicinity. We guess that the original St Mary's was at Cadborough. We would love to hear from anyone that knows more.
The enigma concerns the Sedlescombe coin hoard, which was found north of Sedlescombe bridge in 1876. The latest coin in the hoard is dated 1064, which has made some think that it was buried long before the Normans arrived. We guess that the mint had not changed their coin stamp, so the coins might have been minted up to soon before the invasion. Regardless, something traumatic and lasting must have happened to bury such a valuable treasure and not return to collect it. The invasion looks culpable.
The hoard is often said to be Harold’s war chest, but it seems unlikely. There was a mint at Hæstinga ceastre. It produced less than 1% of England’s coins, but two-thirds of the coins in what remains of the hoard. We cannot think of a plausible reason why Harold would have brought so many coins minted at Hæstinga ceastre.
We guess that the mint was melting down foreign coins taken as taxes, tolls, and fees by the port, then stamping and re-issuing them. Presumably, some of those coins were used to pay the port’s warehousemen, stevedores, ferrymen, and hauliers. We think the hoard’s collector was taking payments, directly or indirectly, from them. An inn, brothel, or toll house perhaps.
But if the Normans landed anywhere other than the north bank of the Brede, the hoard’s owner had two weeks to move the coins to safety. If, on the other hand, the Normans landed on the north bank of the Brede and immediately rode to the Hastings Peninsula to get food, as we suggest above, they would have ridden over the Brede at Sedlescombe. The hoard’s owner would have been in immediate danger. They might have buried the coins as soon as they saw the Norman knights, then fled.
One reference seems to contradict a Brede landing, the Warenne Chronicle which is usually translated: “unopposed he entered the land of the English between the fortresses of Hastinges and Pevenessellum". It uses the u+n spelling which our rule defines as Peuenesel (i.e. Pevensey). If the translation and spelling are right, it is saying that the Normans entered the land of the English in an estuary or inlet between Pevensey and Winchelsea, which is likely to mean Combe Haven. We previously speculated that it had erroneously transposed the u and n. This is still possible, but we now think it more likely that the translation should be: "unopposed between the fortresses of Hastinges and Pevenessellum he entered the land of the English". In other words, the Norman fleet was unopposed as it sailed from pefenes ea to Winchelse, where it entered the land of the English by sailing up the Brede estuary.
Reconciliation with the contemporary accounts
This then is what the other contemporary accounts are trying to say:
- WP: William wanted to avoid a “dangerous or unknown anchorage". The Norman fleet moored outside St Valery then ran downwind on a southerly breeze heading for the well-known harbour of pefenes ea. To muster the fleet back together, and perhaps to obfuscate their intended destination, they moored on shoals several miles off the English coast.
- CKE says of William that: "The earl himself first launching from the continent into the deep, awaited the rest, at anchor, nearly in mid-channel. All then assembled round the crimson sail of the admiral's ship; and, having first dined, they arrived, after a favourable passage, at Hastingas". The Norman fleet waited mid-channel off St Valery, crossed the Channel, ate breakfast, then eventually arrived at Hæstinga port, which CKE referred to as Hastingas.
- Carmen: “the looming rocky coast" did not discourage the invasion. The Normans would have seen the sea cliff at Beachy Head as soon as it started to get light.
- Orderic says that, upon hearing of Tostig’s invasion, Harold: “withdrew his ships and troops from Hastingas et Peneuesellum, and the other seaports opposite Normandy". Harold withdrew his ships from Hæstinga port and his troops from Hæstinga port and Cadborough.
- Carmen: “On the open sea you moor offshore; You caution to take in the sails, awaiting the morning to come; But after the dawn spreads red over the land, and the sun casts its rays over the horizon; You order the sails set to the wind to make way”. The Norman fleet moored off the English coast, then beam reached on a cross wind to Winchelse.
- CBA: The Normans “Arrived safely near castrum Peuenesel. The Duke did not remain long in that place, but went away with his men to a port not far distant named Hastinges". The Norman fleet arrived near the island harbour of pefenes ea, within sight of Anderitum, which CBA refers to as castrum Peuenesel [the castle of Peuenesel]. They moored for a few hours, then sailed to Winchelse, which CBA referred to as Hastinges
- WP describes the scene on William’s ship: “In the morning, a lookout at the top of the mast declared that he could see nothing but sea and sky. They anchored at once.” William was on the newest and biggest Snekka in the fleet, and it carried no horses. His ship would be much faster than the rest. He moored off the English coast and had breakfast while waiting for the fleet to catch up. By the time he had finished, the rest of the ships were in sight.
- Warenne Chronicle: “without any resistance between the forts of Hastinges and Pevenessullum he entered the land of the English". The Norman fleet was unopposed as it sailed from near the fortress of Pevensey to the fortress as Winchelsea where they drifted into the Brede estuary.
- WP says: “Borne by a favourable breeze to Peneuessellum, he disembarked with ease and without having to fight his way ashore". The Normans landed unopposed on the north bank of the Brede, which WP refers to as Peneuessellum.
- Brevis Relatio says that Duke William and his fleet: "arrived in England, by the favour of God, near the fortress of Pevensel. After a short delay he arrived with his whole army at another port not far away named Hastingas". The Norman fleet arrived at the island harbour of pefenes ea, near the fortress of Peuenesel, then sailed to Hæstinga port, which Brevis Relatio refers to as Hastingas.
- WP says: “The rejoicing Normans, once they had landed, occupied Peneuessellum, where they built their first camp, and built another at Hastingas to provide a refuge for themselves and a shelter for their boats". The Normans landed on the north bank of the Brede, which Normans referred to as Peneuessellum, then moved to Hæstinga port, which Normans referred to as Hastingas.
- WJ says that William: “Landed at Peneuessellum where he immediately built a castle with a strong rampart. He left this in charge of some troops and, with others, hurried to Hastingas where he built another". William landed on the north bank of the Brede, which WJ refers to as Peneuessellum, where the Normans assembled a fortress. He then hurried to Hæstinga port, where he assembled another.
- Orderic: “They took possession of Peneuesellum and Hastingas, the defence of which was entrusted to a chosen body of soldiers to cover a retreat and to guard the fleet". The Normans occupied the north bank of the Brede and Winchelsea, which Orderic referred to as Peneuesellum respectively. The fleet moored in the harbours around Winchelsea. Men were stationed at Winchelsea to guard the fleet and to cover a retreat.
- Benoît: The Normans “Arrived at Pevenesel, at a port/harbour beneath a fortress handsome and strong" and “The Count came to Hastinges without staying". The Normans arrived off the island harbour of pefenes ea, which Benoît refers to as PeveneselIt was within sight of Anderitum. They sailed to Hæstinga port, which Benoît referred to as Hastinges.
- Tapestry Panel 38: “Duke William in a great ship crossed the sea and came to Pevenesæ". The Normans arrived off the island of pefenes ea, which the Tapestry referred to as Pevenesæ.
- Tapestry Panel 40: “The knights hurried to Hestinga". The Normans disembarked on the north bank of the Brede, then the knights rode around the Brede to forage for food on the Hastings Peninsula, which the Tapestry refers to as Hestinga.
- Chronicon: William “Had moored his fleet at a place named Pefnesea". The Normans moored off the island harbour of pefenes ea, which Chronicon refers to as Pefnesea.
- ASC-D: “Meantime Earl William came from Normandy to pefnes ea on the eve of St. Michael's mass; and soon after his landing was effected, they constructed a fortress at the Hæstinga port”. The Normans moored off the island harbour of pefnes ea on Michaelmas Eve, then effected a landing on the north bank of the Brede. Soon after – i.e. the next day - they moved to Winchelsea, part of Hæstinga port, where they constructed a fortress.
- ASC-E: “Meanwhile Count William landed at Hestingan on Michaelmas Day". William landed Winchelsea on the Hastings Peninsula, which ASC-E refers to as Hestinga[n], on Michaelmas Day, having spent the previous day on the north bank of the Brede.
- Wace: “The ships steered to one port/harbour; all arrived and
reached the shore together; together cast anchor, and ran on dry land; and together they discharged themselves. They arrived near Hastingues each ship ranged by the other's side." The Norman fleet steered towards the port of Hæstinga port, arrived together in the Brede estuary, cast anchor together midstream, then landed and discharged together on the north bank, near Hæstinga port, which Wace refers to as Hastingues.
- Wace: A messenger tells Harold “The Normans are come! They have landed at Hastingues". The Normans occupied Winchelsea, part of which Hæstinga port which Wace refers to as Hastingues.
- Carmen (Kathleen Tyson's translation): “the happy land owed to you embraced you and yours in a calm basin". The Normans landed on the north bank of the Brede, which was in a calm basin and which William might have thought had been illegally stripped from the Norman Abbey of Fécamp by Harold.
- Carmen: “One Englishman kept hidden under the sea cliff". The spy was at the base of Cadborough Cliff on the north bank of the Brede.
- Wace: An English knight “posted himself behind a hill" to watch the Normans disembark. The spy posted himself behind the spur at Float Farm on the north bank of the Brede to watch the Normans disembark.
- Carmen: “You restore the strongholds that were lately destroyed". The Normans patched up the Romain/burh fortresses at Winchelsea and Cadborough that had recently been destroyed by Tostig, as well as constructing their own kit fortresses.
- Chronicon says that Harold: “Gave them battle nine miles from Heastinga, where they had built a fort”. The Normans assembled a fortress at Winchelsea, part of Hæstinga port.
- Wace: The knights “Formed together on the shore, each armed up on his warhorse. All had their swords girded on, and passed into the plain with their lances raised … When they [the carpenters] had reached the spot where the archers stood, and the knights were assembled, they consulted together, and sought for a good spot to place a strong fort upon". The Normans assembled a motte-less kit fortress on a plain which was adjacent to the strand.
- Wace says that on their first day after landing they went on a raid. “They held their course along the coast; and on the morrow came to a fortress named Penevesel", which they plundered. The Norman raiding party followed the north bank of the Brede to its eastern tip, then marched west along the ridgeway to raid Cadborough, which Wace referred to as Penevesel.
Wace mentions only one Norman camp, near where they landed. ASC-D mentions only one camp, at Hæstinga port. Carmen mentions only a ‘sea camp'. WP and WJ mention camps at Peneuesellum and Hastingas. The Tapestry depicts two camps, the second of which is captioned as at Hestenga or 'Hestenga ceastra'. CBA mentions a camp at the 'port of Hastinges' and a battle camp. We think there was a bridgehead camp at Peneuesellum, a sea camp and a battle camp.
The bridgehead camp at Peneuesellum
Figure 25: Tapestry Panel 42
WP and WJ say that the first Norman camp was near where they landed at Peneuesellum. Both say that William did not stay long before moving to Hastingas. How long? Orderic, ASC-D and others say that William came to pefenes ea on Michaelmas Eve, landing later that day, presumably at Peneuesellum. ASC-E says that William landed at Hestingan - referring to the Hastings Peninsula, we think - on Michaelmas Day. It looks like ASC E simply redacted the temporary landing at Peneuesellum because their stay was short and uneventful. If so, the Normans only stayed one day at Peneuesellum. It was not that William changed his mind about staying at the first camp because he would not otherwise have brought the second fortress. He always planned to move the next day.
Peneuesellum was on the north bank of the Brede, east of Broad Oak. As we mention above, it had some compelling benefits for the Norman landing and bridgehead, not least its long dry salt plain and the Sowdens pinch point, which made it the only inland strand near the Hastings Peninsula that could accommodate the entire Norman fleet, and the easiest bridgehead to defend. Presumably, then, it had some equally compelling drawbacks that made it unsuitable to stay more than a day.
In our opinion, the long-term problem with the Udimore Peninsula was being too small and too short of tactical possibilities. William needed to lure Harold to the vicinity then ambush or trap him. The eastern end of the Udimore Peninsula offered no opportunities. Harold might have been tempted to make his operations centre within Norman striking distance, at Broad Oak or Cackle Street, but the Udimore Peninsula was too small to hide the Norman cavalry. English scouts were almost bound to spot them and warn Harold off. More likely, and almost certainly if he discovered the Norman cavalry, Harold would blockade the Udimore Peninsula and conduct his campaign from Starvecrow Cottage or Pelsham Farm on the north bank of the Tillingham where he was out of reach. William could not take the risk. He had to move.
Tapestry Panel 42 (Figure 25) shows cooks working in front of a towered structure. This is invariably assumed to be the first Norman fortress. It does not look like that to us. Carmen says the Norman fortress was surrounded by palisades. This building has open sides. It could not have come as a kit from Normandy because it has stone towers with windows, foundations and cupolas, whereas a kit would be wood with no foundations or adornments. Also, it is depicted on a bobbly base whereas a motte-less kit fortress would have needed to be on a smooth level base. We reckon the kit fortress is out of shot.
The structure in Panel 42 looks like a simplified version of William's palace from Panel 35, with the roof being held up by an arch and cross beam. We are drawn by the bobbles. It is one of only two buildings on the entire Tapestry that are depicted on bobbles. The other is the Saxon house being burned in Panel 47. Perhaps it was done by mistake, but we suspect that the bobbles indicate that both buildings are far distant, beyond bobbly fields. If so, the building in Panel 42 might be the Roman fortress and burh on the Udimore ridge, at Cadborough we think (Figure 26). Otherwise, perhaps it is a Saxon salt warehouse or fishing net dryer.
Figure 26: Bridgehead fortress location
In the afternoon and evening of their one day in Peneuesellum, WJ says that the Normans built a fortress with a strong rampart. It might be possible to work out where it was. Wace says they brought the fortress as a kit from Normandy. He goes on to say that the knights and carpenters join the archers on a plain at the edge of the strand where they: "consulted about where would be a good place to build a strong fortress".
It sounds like the fortress was on the plain or close to it. Tapestry Panel 41 (Figure 23) depicts the first camp at a treeless plain. A motte-less kit fortress could only be assembled on flat level ground. It would be most useful on a treeless plain because there would be no natural shelter. We think it was somewhere on the dried-out salt evaporation ponds.
Wace says that an English spy watches the fortress being constructed from behind a hill. Carmen says that he previously watched the landing from the base of a sea cliff. If the landing was in the Brede estuary, this has to be Cadborough Cliff. He would not have gone too close for fear of getting captured. The obvious observation place was Float Farm, south of Cock Marling, where a spur came close to the water’s edge (eye in Figure 21). We guess that the fortress was at A, B or C, where it would protect the part of the camp to its east.
The only English garrisons in the vicinity - Lympne and Pevensey - were empty on the day of the invasion, but William was not to know. He would have planned to defend against a garrison counterattack in order to buy enough time to unload his ships and establish a strong bridgehead. The English troops would have come east along the Udimore ridgeway, shown as a cyan line on Figure 26.
In the first edition of our book, we thought the kit fortress was most likely to have been at A, where it would protect an evacuation if the English army turned up before the horses had been unloaded. We have changed our minds. We now think that was more likely to have been at C, where it worked in conjunction with a blockade of the Sowdens pinch point (x) to protect the landing site. It would have been positioned midway across the salt-plain and east of a stream that drained runoff from Sowdens. Even if the fortress were only 30m square, it would have been horribly difficult to get past with perhaps only 50m either side for the Normans to defend.
One other interesting point here is that, before their arrival at the second camp, CBA says that William "burned the greatest part of his ships" in order to get his more lily-livered supporters to focus on their goal. Wace says that William: "commanded the sailors that the ships should be dismantled, and drawn ashore and pierced, that the cowards might not have the ships to flee to". Both statements are usually dismissed as poetic license. After all, WP says that an English messenger finds William inspecting his fleet, which would be pretty pointless if it was burned or holed.
In our opinion, William almost certainly did burn the greatest part of his fleet at Peneuesellum. We guess that: 1) There was not enough space for the entire fleet at their second camp; 2) The horse carriers and cargo ships - which comprised more than half the fleet - were no longer needed because William had no intention of ever returning the horses, fortifications or provisions to his ships; and 3) Harold might have found the cargo ships useful in a blockade, so William wanted to ensure those left behind were not seaworthy.
It also makes sense that William ordered the bungs to be removed from the ships that went to the second camp, in order to discourage deserters. However, we think the bungs were readily available and could be used at short notice if William ordered the ships to leave quickly. We will return to why it might be relevant when discussing William's plans.
The Norman sea camp
Carmen says that, after the battle, William “returned to his sea camp", hence we refer to it as the 'sea camp'. ASC-D says that: “soon after his landing was effected, they built a fortress at Hæstinga port". Carmen says that after the battle William spent five days in his “camp at Hastinges portus". CBA says that William went away with his men to a port named Hastinges where, “having secured an appropriate place ... he built a fortress of wood". WJ says the Normans assemble a second fortress at Hastingas. Chronicon says it was at Heastingam, then that: “William, however, returned to Heastingam” after the battle. We explain in the place name blog why CBA's Hastinges, WJ's Hastingas and Chronicon's Heastinga also referred to Hæstinga port.
Thus, all the sources that mention the second Norman camp say it was at Hæstinga port. We explain earlier why we think that Hæstinga port comprised Winchelse on the Camber shingle spit and Winchelsea, then known as Iham. Panel 45 (Figure 27) depicts the second camp. It shows a fortress on top of a hill. It cannot be on a shingle spit, so it must be at Winchelsea. This makes sense anyway. The Normans were hardly likely to make their camp on a flat, flood prone, siege prone shingle spit with only rain for water, and a few hens and goats to eat. Winchelsea is close enough to the coast to be a ‘sea camp’.
Figure 27: Tapestry Panel 45
Tapestry Panel 45 (Figure 27) depicts the second camp. It is captioned: “He ordered that a motte should be dug at Hestenga ceastra", which is usually interpreted to mean ‘at their Hastings camp’. This cannot be right for the same old reason that Hastings did not exist at the time. It might mean 'at their Hastings Peninsula camp', on the basis that we think the Tapestry's other references to Hestenga referred to the Hastings Peninsula. But the caption is odd.
'AT HESTENGA CEASTRA’ is Old English whereas the rest of the Tapestry captions are in Latin. The traditional explanation is that the embroiderers were Saxon nuns, so they used the Old English term by mistake. It would be an uncharacteristic error considering how fastidious they were. We believe that they used Hestenga ceastra as a proper noun and indicated it by prefixing Old English ‘at’. Thus, it seems to us that the Tapestry is saying that the Norman sea camp was at Hæstinga ceastra, on the summit of Winchelsea. If so, the Tapestry is corroborating and narrowing the description given by the other contemporary accounts.
Winchelsea was very different in the 11th century, a narrow-necked steep-sided peninsula, sitting in a tidal lagoon. It was as good a defensive location as there is in the region. It had sea cliffs to the north and east, a steep slope down to the sea to west and a narrow causeway entrance to the south. As we said earlier about Sowdens, pinch points are double edged. They are good for defence but siege-prone. In this particular case, we suspect that is exactly what William wanted. He needed to lure Harold to the Hastings Peninsula in person. Placing himself somewhere distant, passive and siege prone might have given Harold the confidence to come in person. On the other hand, an actual siege would have been disastrous for William and Normandy.
William and his barons and his guard would have been ensconced at Winchelsea. Most of his infantry would have been spread around the eastern cape of the Hastings Peninsula, guarding the fleet and ready for a counterattack. We guess that he had one division at Hog Hill to protect the causeway, a second at Winchelsea harbour to guard the fleet, and a third at Icklesham to block the ridgeway. Kathleen Tyson, who thinks the Norman sea camp was at Icklesham, is therefore probably right: Part of the sea camp probably was at Icklesham, although we think its focus was at Winchelsea.
Carmen describes William restoring ‘dilapidated strongholds’. Quedam Exceptiones, an epitome of WJ written in the early 12th century, says that William: “restored the most strongly entrenched fortification” (Kathleen Tyson’s translation) on Peneuesellum.
We think these fortresses were at Winchelsea (Hæstinga Ceastre) and Cadborough (Peneuesellum). Both fortresses could have been damaged when Harold raided the area 1052 or when Tostig raided it earlier in 1066. Indeed, they might have been prime targets since at least one of them had a mint that would have held gold. None of this would contradict any primary sources if, as we think, the Norman kit fortresses were in addition to dilapidated burh fortresses that already existed at Winchelsea and Cadborough.
The exact location of the Norman sea camp fortress
There is a lot of confusion about Panel 45 (Figure 27). It is captioned: “He ordered that a motte should be dug at Hestenga ceastra". Nearly everyone assumes that the Normans had already assembled their kit fortress on the top of the hill and that they are digging a moat or ditch around the bottom of the hill. This is often based on the mistaken belief that ‘motte’ is another word for 'moat'.
A motte is the raised level ground upon which to build a keep. William is therefore ordering his men to raise and level the ground as a foundation for his kit fortress. They are still digging. Clearly then, they had not yet started to assemble the kit fortress. Yet there is already a fortress on the top of the hill. It must have been there before the Normans arrived. It has the word ‘CEASTRA’ embroidered right in the middle of it. We think that the hill-top fortress is Alfred’s burh at Hæstinga ceastre. If so, the burh fortress, William’s kit fortress and William’s sea camp were all at Winchelsea.
If the hill-top fortress on Panel 45 was Alfred’s burh fortress, why did William need a kit fortress? Carmen says that the pre-invasion fortresses had recently been damaged. Perhaps it was too dilapidated to offer a good defence, although Carmen goes on to say that William had it restored. Even if it was too damaged to be fully restored, why did William not build the kit fortress at the top of the hill inside the old burh wall? The answer is that the kit fortress had another purpose, to guard the only weak point of Winchelsea's defence.
Alfred’s burh fortress would have been on the top of the hill, where the graveyard of St Thomas now lies. It is shown as a green square in Figure 28. It would have been inside the burh wall, which was roughly 600m long. As we say earlier, there were sea cliffs on the east and north sides of Winchelsea and a steep slope leading down to the sea to the west. The south slope was shallow and vulnerable. William’s fortress, shown as a red square, would have controlled the narrow neck – 150m across – that provided the only easy point of attack. The Tapestry’s viewpoint is looking north from the magenta arrow.
Figure 28: Artist’s view for Tapestry Panel 45
The Tapestry's tower
What about the tall thin tower to the fortress’ right? It is square. It has a polygonal pyramid roof and a small window either side of the middle beam. There are no windows at the top, at least on the visible side. It seems to have a load-bearing timber frame; the middle crossbar protrudes at the side. Apart from the first Tapestry Panel, which is usually thought to represent Westminster, all the other towers in the Tapestry are stone. This makes some sense, because the Panel 45 tower and Westminster would be Saxon whereas the others were Norman or Carolingian.
Towers were rare in pre-invasion England. There were probably only a hundred or so in the entire country. Most of them were bell towers attached to monasteries. The only surviving record of what Saxon timber towers might have looked like is Greensted Church in Essex. It is a broad-based boxlike affair. There were some Saxon stone bell towers, a dozen or so of which survive, but they are broad based too. Moreover, a bell tower should have big windows evenly spaced around the bell stage, to let out the sound, whereas the highest window on the Tapestry tower is not much above the middle. And if it was a bell tower, there is no sign of its monastery. Not a bell tower then.
There are only four other types of building it could be: a watch tower, a message tower, a stair turret or a lighthouse. Tall thin stair turrets were exceedingly rare in Saxon times, but they do exist. Two examples survive at Brigstock and Hough-on-the-Hill. They are both stone, which makes sense because they were attached to monastery bell towers which had been designed and built by European stonemasons. There are no known timber stair turrets, although this does not mean that they never existed. Never-the-less, we think the Tapestry tower was not one of them because it has no adjacent belltower or monastery.
There are no known Saxon lighthouses, so this possibility does not seem promising either. But Winchelsea would have been an obvious place to build a lighthouse cum navigation beacon. Winchelse was a busier port than Dover in Roman times and the Romans built two pharos there. One survives. There was a dishevelled lighthouse at Winchelsea that was already old in the 13th century, because Nicolas reports that a tax was levied on ships using the port to pay for its renovation. By 1300 a snazzy new stone lighthouse tower featured on Winchelsea's seal (Figure 29), standing between its two churches. It is possible then that the seal lighthouse replaced a 12th century stone lighthouse that had replaced the Tapestry's timber lighthouse that might have replaced a Roman pharos.
Figure 29: 1300 Winchelsea seal showing its two churches and a lighthouse tower
Saxon watchtowers were rare too, but each of Alfred’s burhs would have had one. Burhs are often said to be fortifications to defend against Viking attacks. It was presumably more important to provide an early warning because if the locals hid themselves and their valuables, the Vikings would go away. It would therefore be helpful if the burhs sent early warnings to each other. They were too far apart to have had line of sight, so it is widely believed they were augmented by a network of message relay beacons or message relay towers.
If the Panel 45 fortress is indeed Alfred's burh at Hæstinga ceastre, we are pretty sure it would have had a watchtower. The one depicted is not it. It is outside the burh and too low down the slope. It looks like the burh tower got lopped off the Panel 45 by the Tapestry's top banner.
We think that the Tapestry tower had a commercial purpose. Assuming the tower was 10m high, the pole on the top would be visible from roughly 25km. The Udimore ridge to its west was 25m higher than the pole and 3km further away from the sea, so it was also visible from roughly 25km. A coincidence? We think not. We guess that the idea was to line up the pole with a landmark on the Udimore ridge to guide ships into Hæstinga port at Winchelse. Presumably, the navigator would pick different landmarks depending on the wind.
This would explain one of the many perplexing features of the Tapestry, which is why they built a tower so far down the slope that its viewing position was below the level of the fortress. Under any other circumstance, the Tapestry tower would be pointless; they might as well look out from the fortress rampart. But in this case, we guess that the additional elevation and the fortress walls were high enough to block the view of the Udimore Ridge beyond, so the navigation tower had to be built down the slope.
Figure 30: Greyfriars, Winchelsea in 1728 and now
This would also explain Winchelsea's enthusiasm for towers. After the Tapestry tower, Nicolas's dilapidated tower and the Winchelsea seal tower, another one was built at Greyfriars monastery (Figure 30). They were all similar dimensions and similar heights. They all had poles on the top. We guess they all served the same navigation purpose and that they were all in the same place. That place is 50.921802, 0.710274, which we know because the remains of the Greyfriars monastery tower are still there. It is marked with a yellow dot on Figure 28, in exactly the right place for the tower depicted in Panel 45.
If the tapestry tower was only for navigation purposes, there would be no point giving it stairs or windows. We think it had another important purpose, which was to watch for incoming commercial vessels. Its position 8m down the slope only makes 2km viewing difference to the horizon - 22.6km vs 24.7km - which was presumably less important than its line of sight to the docks at Winchelse. The idea would have been to send signals down to the docks, instructing them to prepare outbound cargo, to get stevedores out of the pub, to clear the cross spit channel, to make space in the warehouse, or to prepare the inshore barges. This could not be done from the burh tower because the higher cliff, greater distance from the cliff edge and the burh walls would have blocked the view between it and the docks.
Figure 31: Monastery tower from Panel 48 (L); Sea camp tower from Panel 45 (R)
Panel 48 depicts a monastery from where the Normans launch their attack. It is so grand that it is surely where the senior Fécamp Abbey clerics lived and where the port administrators worked. We think it was near Sedlescombe. Next door is a cigarette shaped tower (Figure 31 left). Being adjacent to a monastery, it seems likely to have been a belltower, but it is too thin and it lacks bell stage windows at the top.
A monastery tower at Sedlescombe would have had line of sight to the sea camp tower at Winchelsea. A close up of the tower in Panel 45 (Figure 31 right) is therefore of great interest. At a distance, the top left infill appears to be stone or shingle. On closer inspection, it is obviously a wall of crosses. There are too many for them to be structural or to be holding up bells. They would be too high to have had a defensive purpose. When we first saw the Tapestry, we guessed they were wall anchors, but we never came up with a plausible explanation why a timber tower would need so many, and only on one part of one side. One possibility is that top left nearside wall has fallen off, leaving wattle fixings visible on the inside of the far wall. Another possibility is that they were simply for artistic effect. We think not. We think the real answer is more impressive.
We guess that the sea camp tower crosses are some sort of messaging system. Hardly anyone could read in those days. We think the crosses are pictorial representations of coloured symbols that carried some sort of coded message, like naval flags. The window would be where the operator views responses and new messages from other towers. We guess that the sea camp tower sent messages to the port administrators at Sedlescombe about incoming vessels, warehouse space, the need for barges and the like. We would love to hear from anyone that knows more about this.
Before moving on to the Norman battle camp, it might help to think about the leaders’ plans. Harold was doubtless incensed that William had invaded his realm, threatened his monarchy, questioned his integrity, ransacked his ancestral land and attacked his personal manor. All the primary sources that mention his frame of mind say that he wanted to get to the Hastings Peninsula as quickly as possible to destroy the invaders in an immediate nocturnal or surprise attack. Doubtless that was his instinct, but he was no fool. According to Orderic, Harold was quick witted and smart.
In our opinion, Harold would only have gone to Sussex if he thought he was safe; he would only have tried a surprise attack if he thought the Normans were unprepared; he would only attack at all if he felt certain of victory. If the Normans had any chance of surviving a battle, Harold would have worn them down by a scorch-earth siege. Wace specifically says that he heard tell that Harold ruled this out in a discussion with his family, saying that he could not plunder his own subjects. Dodgy provenance again. This one is plainly bunkum.
Harold's family and advisors would surely have recommended a scorch-earth siege. It is clearly the right strategy. If Harold rejected their recommendation, we cannot believe it was because he feared plundering his own subjects. This was his ancestral homeland. He kept it in his own name. There were only 300 or so families living on the Hastings Peninsula at the time and they nearly all worked on Harold's manors. They would all have fled. The only person that could be plundered was Harold himself, by the Normans.
Sure, Harold was indignant, but why would he even leave London? True, warrior Kings were expected to lead their army in those days. True, Harold was the best person to repel the invasion. He knew Norman battle tactics and he knew William intimately after having fought with him in Normandy. Also, he was the most experienced English commander and this was his home, so he knew the terrain better than anyone. But why take the risk? According to Wace, Harold’s family recommended that brother Gyrth should lead the army. It would have been so much more logical. William might even have gone home if the English army arrived without Harold.
We are convinced that Harold wanted to move quickly and in person because he planned to invade Normandy. He knew that William had made a disastrous error; that William’s chances of victory were minuscule. The overwhelming odds were on the invaders getting annihilated. William had brought contingents from all the surrounding Dukedoms. Normandy was virtually undefended. Anjou, Maine and the part of Brittany controlled by Count Eozen had lost roughly half of their barons and knights. We think Harold was expecting to annihilate the invaders and then to annex Normandy, and perhaps Anjou and Maine. William had even brought the ships that Harold needed to transport his army across the Channel.
Having decided to lead the army in person, Wace, Carmen and CKE say that Harold did what anyone sensible would have done in his shoes: To get an idea of the size, composition and position of the enemy by sending spies and messengers to scout the enemy camp. This suggests to us that he was still open minded about an attack or siege.
Wace reports that the Count of Flanders was Harold’s most trusted court spy. Unfortunately, it seems that the Count was poorly informed – or perhaps duplicitous - because Wace says that the Count repeatedly told Harold that William could only bring a feeble army with few horses. From Harold’s subsequent actions, his messengers must have told him much the same, because he expected a quick crushing victory.
Harold's route to Hastings
Figure 1: Medieval southeast England
One important early choice was whether the English army should head south down the Margary 13 Roman road from Rochester to the Hastings Peninsula or use Margary 14 from Peckham to Lewes then cut across the Andredsweald on the LIN-129 ridgeway from Uckfield – these are shown on Figure 1. Margary 13 would have been quicker and easier, but Simon Mansfield, Simon Coleman and others think Harold used Margary 14 and LIN-129.
Mansfield cites two main reasons that Harold would not have used the Rochester road: 1) That it would remove the element of surprise because William’s scouts would see the English coming; 2) That Harold would be delayed by the ferry across the Rother. Both points are valid, but we think they are incidental.
True, Harold wanted to catch William by surprise, but they had been exchanging messages. William knew Harold was coming, knew where he was coming from and had been preparing his defences for two weeks. There was no chance of William being taken by surprise and Harold knew it. He arrived in the battle theatre on 12th October. A third of his troops were not due until the 15th at the earliest (otherwise they would have arrived in time to turn the tide of battle). He would have wanted to act quickly and decisively as soon as they arrived. His haste is most likely to have been because the earlier he arrived in the battle theatre, the more time he had to scout the enemy position and devise his plan.
True, the English army would be delayed at the Rother, but the alternative route through the Andredsweald on LIN-129 would not have been quick. A typical medieval chain/rope ferry might carry 100 troops. The Rother crossing and return might take 15 minutes. At that rate it would take 16 hours to get the army and supplies across the Rother. It is a worst case. Harold or his delegate could requisition every boat on the river to ferry the English army across the Rother. If pushed, we guess the Rother delay is unlikely to have been more than 8 hours, which is considerably less than the time it would take to get an army of supply carts 30km through the Andredsweald on a narrow rutted track. As a reference, a coach took 14 days to make the 600km from London to Edinburgh on partly maintained pre-turnpike roads. It would have taken two days at best to get through the Andredsweald on LIN 129. If, as seems probable, the track was only one cart wide, the entire train would get blocked every time a cart had a mechanical failure. It might have taken a week.
It is not even obvious that the LIN-129 trackway across the Andredsweald from Uckfield to Netherfield was still usable in the 11th century. It does not appear in any court or Church travel records. There were no manors and no active bloomeries in Saxon times, which removes two reasons to maintain the trackway. There are no Saxon Charters charging anyone to fix the ruts. There were no settlements where broken carts could be fixed, which would discourage freight. Hardly any Saxon archaeology has been found. Wolves and bears lived in the Andredsweald, which would discourage travellers. Unless it had some otherwise unknown military purpose, we suspect that it had become overgrown by the 11th century.
Even if LIN-129 was still usable, there are several good reasons why Harold would not have used it. It would be impossible to protect him from snipers, blockades and ambushes in 30km of narrow woodland trackway. There were no farms to provide food for the troops. The Kentish troops would arrive on the Rochester road. The huscarls would arrive by ship where the Rother met the Rochester road. If they had to muster with the rest of the army at Netherfield, they would be vulnerable as they headed west from Cripps Corner. Harold would again be vulnerable to snipers, ambushes and blockades if the English army then headed southeast on the narrow LIN-130 ridgeway to Battle, especially as it crossed the isthmus at Sprays Wood.
If the English army did cross the Andredsweald on LIN-129, an easier and safer alternative would have been for the main English army to continue east through Netherfield to muster with the Kentish troops and huscarls at Cripps Corner, where the Udimore ridgeway crossed the Rochester road. But if the plan was to muster at Cripps Corner, it would have been quicker and safer to for the main English army to use the Rochester road in the first place.
ASC-D says that Harold: “assembled a large army and came against him at haran apuldran”. Whitelock translates haran apuldran as ‘hoary apple tree’. It seems implausible to us. More than a dozen haran apuldre (or similar) are mentioned in Saxon charters as boundary markers, all in insignificant places that readers would not know. We think it means a nationally well-known landmark. Ingram proposes ‘estuary of Appledore’. Stenton notes that ‘haran’ is used with features that cannot possibly be ‘hoary’, like pools and rivers. He and Jepson reckon that haran usually means ‘boundary’ in place names. Haran apuldran might therefore mean the ‘boundary of Appledore’. Kathleen Tyson reckons the ‘anchorage of Appledore’. All mean the Rother. This makes sense. Most of the huscarls would have arrived by ship on the Rother. Kentish troops would have arrived via the Rochester road. The supply carts might have taken a week to arrive on any route other than on the Rochester road. The obvious place to muster was at the junction of the Rochester Roman road and the Rother, near modern Bodiam.
William of Jumièges says that Harold rode through the night to meet his troops at the English camp. It is plausible, but he would not have ridden through the Andredsweald at night on LIN-129. If Jumièges is right, he too arrived on the Rochester road.
We feel confident that Harold and the English army arrived at the battle theatre on the Rochester Roman road. It seems likely to us that Jumièges is right that Harold did not arrive until the English camp had been established and fortified. Otherwise, the Normans might have attacked Harold in the English camp before it had been fortified. It does not impact our theory if we are wrong about both. But the route the English army took is crucial to us.
Our entire battle scenario depends on the English camp being beside the Rochester Roman road. If the English army came through the Andredsweald on LIN-129, then branched off to Battle on the LIN-130, they would have bypassed the Rochester road and would not have camped beside it. We remain confident. There are only two possibilities. If Harold was with the English army before they camped, he would not have risked the sniper prone and ambush prone LIN-130 between Netherfield and Battle. If Harold arrived later, as Jumièges suggests, he would not have ridden through the Andredsweald, especially at night. In either case, the English army would have ended up at Cripps Corner on the Rochester road. From there, they had to choose a camp location which is likely to have been beside the Rochester road. We will return to this momentarily.
William needed to kill Harold quickly. He had six months to prepare. He did so meticulously. We cannot believe that the best military plan he could come up with was to wait on the Hastings Peninsula in the hope that Harold would come in person with a powderpuff army, then recklessly attack and fight a losing battle to the death. The chance of success was negligible. Conversely, the chance that William and the entire Norman army would get besieged and starve were huge. Normandy, shorn of its leaders and fighting age men, would have been subjugated by someone. Even though the potential rewards of victory in England were huge, so were the risks.
William had six months to prepare. He did so meticulously. We cannot believe that the best military plan he could come up with was to wait on the Hastings Peninsula in the hope that Harold would come in person with an understrength army, then walk into an ambush on land he knew well, or recklessly attack on disadvantageous terrain and fight a hopeless battle to the death. The chances of success were negligible. Conversely, the chance that William and the entire Norman army would get besieged and starve were huge. Normandy, shorn of its leaders and fighting age men, would have been subjugated by someone. Even though the potential rewards in England were huge, so were the risks. William must have had a better ‘Plan B’.
William's easy options were out of the question. England had four times Normandy's population with hardly any disloyal barons. The Normans had an impossibly long supply chain. They stood no chance in a prolonged campaign against a determined defence. They stood no chance if Harold adopted a Fabian strategy. They could not catch Harold if he chose to be elusive. They stood no chance of a successful siege on London or Winchester. Defeating the English army without killing Harold could easily backfire if Harold switched to a Fabian strategy.
The Normans’ only advantage was mobility. The fyrds, which comprised the majority of Harold’s armed forces, were part-time infantry who plodded around on foot and whose fighting skills were honed for repelling Viking axemen. Normans on the other hand, could whizz around on longships, fight on mobile tanks or pick off the English leaders with snipers.
If we were in William’s shoes and the English army arrived without Harold, we would quickly withdraw and ride/sail around to the Thames, hoping to trap him at Westminster (which was outside the city walls). If Harold was not there, we would lay siege to Winchester, hoping to draw him out before he raised an overwhelming army. If the English army turned up at either place without Harold, we would raid coastal settlements - just as Harold himself had done after his exile and as Tostig had done a few months previously – in the hope of humiliating Harold into premature battle.
There is one small clue in the primary sources that this was William’s backup plan. Orderic and WP say that he leaves a body of soldiers to guard the fleet and to cover a retreat. The only obvious reasons for an organised retreat are if Harold turned up with an overwhelming force or the English army turned up without Harold.
Figure 32: East Sussex topography, roads and trackways
So much for Plan B. Plan A was to lure Harold to the Hastings Peninsula then ambush or trap him. Harold was in danger from the moment he crossed the Rother. The Rother Peninsula isthmus at Vinehall Street was barely 1km long. If the Normans sailed up the Rother and occupied the isthmus ridge, the English would be trapped. The stretch of Roman road between the Rother and Cripps Corner (C on Figure 32) went through Lordship Wood, which could harbour snipers. Cripps Corner was at the junction of the Roman road and the Udimore Ridge, which was a good place for a blockade and/or ambush. There was another ambush/blockade opportunity where the Roman road intersected with the Great Sanders ridge (G) and the best of all where it crossed the Brede at Sedlescombe (S).
The further Harold ventured south, the greater the danger he was in. If we were in William’s shoes, we would let Harold venture as far south as he wanted to go, hoping that he eventually tried to cross onto the Hastings Peninsula. If Harold did try to cross onto the Hastings Peninsula, William would have had the choice of whether to ambush him at the crossing point or to allow him across.
One possible drawback of allowing the English army onto the Hastings Peninsula was that defensively sound hills at Cottage Lane ridge, Starlings hill and Doleham hill (labelled in magenta as O, T and D on Figure 28) were within a few hundred metres of the Brede crossing points. If the English chose to camp at any of them, they would have been difficult to defeat before their reinforcements arrived. William would have fancied his chances of blocking the English reinforcements at any of the crossing points, but there was a level of uncertainty. They might have landed on the coast, or they might have crossed the Ash Bourne and come through an unknown forest path, for instance.
Another is that the north slope of the Hasting Ridge was lined by dense woodland all the way down to the coast, and there was virtually nothing but dense woodland to the northwest for 80 miles. If the English were allowed to get more than a mile from any of the Hastings Peninsula crossing points, they might have melted away into the woods where Norman armour, lances, horses and archers would be worse than useless. If this happened, the invasion would almost certainly have failed.
On the other hand, the three Brede crossing points – Sedlescombe (S), Brede ford (BF) and Whatlington ford (W) – were ideal for an ambush; low lying, boggy and backing onto a bottleneck bridge or ford. The LIN-130 ridgeway from Netherfield to Caldbec Hill – whether Harold arrived there from Uckfield or from Cripps Corner – was almost as good. The isthmus between the Brede basin and the Ash Bourne basin at Sprays Wood is barely 500m across. William could have waited for Harold to pass, then sent half of his troops up to Cripps Corner, west to Netherfield, then south to block the ridgeway. Harold would have been trapped on the isthmus ridge (IR) with nowhere good to defend.
By tradition Harold immediately crossed onto the Hastings Peninsula and camped at Caldbec Hill (CH). The tradition starts with seven primary sources that imply that Battle Abbey was built on the battlefield. If the battlefield was at Battle Abbey and the English army had moved south from their camp less than hour earlier, as specified in several contemporary accounts, then their camp must have been a kilometre or so to the north. Caldbec Hill is the only candidate. But, as we explain in our Traditional Battlefield blog here, we think that the 'Battle Abbey on the battlefield' accounts are unsound, so the argument chain is faulty. Even if this were not so, it seems implausible to us that Harold would camp at Caldbec Hill. To get there, the English would have had to cross onto the Hastings Peninsula and then head 5km in the wrong direction by an unfathomably circuitous route to make a defensively poor camp.
We do not believe that Harold ever tried to cross the Brede. This was his ancestral homeland. He would have known the ambush risk better than anyone. In our opinion, he would not have crossed or circumvented the Brede before the far side had been thoroughly scouted. That would have taken at least a day. In the meantime, the English had to camp somewhere north of the Brede. William’s only chance of victory in this eventuality would have been an early attack on the English camp, in which case the English would not have had the chance to cross the Brede.
One interesting question is why Harold ventured close enough to the Norman army to jeopardise his campaign and his life. The only plausible answer is that he thought he was safe. Wace says that Harold only discovers the enormous size of the Norman cavalry on the morning of the day of battle. This would explain it. If Harold believed that the Normans were mainly footbound and they tried to sortie over the Brede at Sedlescombe or Whatlington, he would have thought that the English could swoop down to trap them on boggy ground against the Brede. If they tried to loop around via Netherfield, he could blockade the isthmus. Harold would have thought himself safe behind the front lines. But none of this was possible because the Norman cavalry would catch the English in the open and cut them down if they tried to move anywhere away from their camp.
The real puzzle then is why Harold did not realise the Normans had brought so many war horses. It is odd because most of the contemporary accounts describe their exchange of messages. Wace says they had been communicating by "messengers, clerks and knights" while Harold was still in London. At dawn on the day before battle, according to Wace, Harold and brother Gyrth scout the Norman camp. They hear Norman horses, see the knights’ armour and look down on the enemy huts and tents. The strength of the Norman army comes as a shock, so Harold dispatches two spies to get more details. They were captured, again according to Wace, taken on a tour of the Norman camp and sent back to Harold. Yet Harold is still shocked by the size of the Norman cavalry on the day of battle.
Harold blames the Count of Flanders: “The Count of Flanders hath betrayed me: I was a fool to trust him when he sent me word by letter and assured me by messages that William could never collect so great a chivalry”, according to Wace. But had heard the Norman horses himself. He must have thought they were not too numerous. His messengers and spies must also have reported that the Normans had few horses. Two of them were taken on a tour of the Norman camp.
It seems to us that, until the day before battle, William deliberately enticed Harold close by giving the impression that his army was puny, toothless and siege prone. Then on the day before battle, he seems to have tried to intimidate Harold with a show of strength, still without revealing his cavalry. Our interpretation of this is that William was sandbagging, trying to lure Harold into a trap. Then on the day before battle he tried to scare Harold into fleeing, presumably having positioned his cavalry to intercept anyone trying to flee.
WP reports that one of Harold's messengers finds William inspecting his fleet. He must have been at his sea camp, at Winchelsea we think. This sounds like the one leg of his sandbagging ruse. If William and some barons, a few horses and perhaps half his infantry positioned themselves at the sea camp, they would seem unthreatening to Harold. Not only would they be at least eight miles away from the Brede crossing points, but Harold's scouts on the Udimore ridge would see them leave and give Harold plenty of warning. The second leg would have been to conceal his cavalry and the rest of his infantry somewhere near the Hastings Peninsula crossing points, where they could ambush Harold if he tried to get onto the Hastings Peninsula. This would explain why Harold's messengers and spies did not see the Norman cavalry and why Harold expected a crushing victory.
Perhaps we are giving William too much credit. WP says that the Norman cavalry were out foraging when the English army entered the battle theatre. If they were out foraging every day, it would also explain why Harold's messengers and scouts did not see them. Perhaps it was just luck. We doubt it. It seems more likely that the Normans did all the foraging they needed on the first day because they would have expected the locals to drive off their livestock and burn their grain stores as soon as they knew the Normans were there. We think that William deliberately sent his knights out foraging every day, even though they probably found meagre pickings, in order to conceal their existence.
We will return to the land scenarios momentarily. William could not be sure that Harold's arrival at the battle theatre was not a diversionary tactic to take his attention away from a sea borne attack. Harold might have landed his main army at Fairlight Cove, marched up through Pett and Guestling to attack or besiege William at Winchelsea. William had it covered. In this scenario, the Norman cavalry, positioned out beyond Westfield, would have coalesced to occupy the ridgeway at Guestling Thorn, thereby trapping the English army on the Winchelsea Peninsula.
Norman battle camp at Hechelande
Figure 33: Leuga around Battle Abbey
CBA names the Norman battle camp Hechelande: “The Duke came to meet him [Harold], surrounded by units of cavalry. Arriving at the hill named Hechelande, which lies towards Hastingarum, while donning their armour ...”. In other words, William and some of his knights came from their sea camp to a battle camp on a hill named Hechelande where they dressed for battle. As we explain in the section on Hechelande (page 59), other locational clues in CBA place it just to the west of the Hastings Ridge, a few hundred metres northwest of Telham, on the land now occupied by Loose Farm (Figure 33). This clue is too good to ignore, but can it be trusted?
CBA was written to protect the Abbey’s wealth and independence. Their case was based on their claim that the Abbey was built on the battlefield. No surprise then that their description for the location of the Hechelande battle camp is consistent with the Abbey being on the battlefield. In the Traditional Battlefield section, we list many reasons why we believe the battle was not fought at Battle, in which case the Abbey is not on the battlefield. Other parts of this same CBA passage, notably William’s vows to build a monastery on the battlefield, are known to have been fabricated. It gives reason to question CBA’s description of Hechelande’s location.
There are four possibilities: 1) There was no Norman battle camp; 2) The Norman battle camp was not at Hechelande; 3) The Norman battle camp was at Hechelande, at the place described in CBA; or 4) The Norman battle camp was at Hechelande, somewhere other than the place described in CBA.
In the next section we list four clues that there was a Norman battle camp, so we discount (1). It is possible the there was a place named Hechelande at the location described in CBA at the time of the battle, yet the Norman battle camp was elsewhere. But if, as we suspect, the monks of Battle Abbey were trying to give the spurious impression that the battle was fought at Battle Abbey, this would be risky. Another contemporary battle account might have listed the place name of the real Norman battle camp, which would undermine CBA’s veracity. They would be better off simply not naming the Norman battle camp, so we think (2) unlikely. What then of the third possibility, that the Norman battle camp was at Hechelande, at the location described in CBA, even though the battle was not fought at Battle?
Figure 34: CBA land holdings in its Leuga - Duniford Wood in green; Petley Wood in blue; Bodeherste Wood in teal; Hechelande Wood in orange; the 37 acres in purple; the wist between Bodeherste Wood and Hechelande in cyan; the huge uncultivated plain in yellow; Battle Abbey, Bodeherste manor house and Hechelande manor house are labelled A, B and H respectively.
It would help to flesh out Figure 33 with some other features described in CBA (Figure 34). It says that there were four woods inside its Leuga: Bodeherste, Hechelande, Petley and Duniford. Petley Wood survives. It says that Bodeherste was due east of Battle Abbey. A place named Bothurst Wood - a Bodeherste cognate according to Lower - is shown on the 1770 Yeakell and Gardener map to be coterminous with modern Great Wood. Chevalier reckons that Duniford Wood was north of Caldbec Hill. It must have been west of the Whatlington Road, because CBA says that Uccheham was to the east of the Whatlington Road and south of Petley Wood. Presumably, Duniford Wood spanned the Line in order to get the 'ford’ part of its name. It was therefore north and northeast of Caldbec Hill. CBA says that Battle Abbey held a wist and 37 acres that are between Bodeherste Wood and Hechilande Wood and between the infirmary and Chapenore. We interpret this to mean that it was either side of the ridgeway from Telham to Battle Abbey. Finally, CBA says that Battle Abbey held a huge uncultivated plain between Bodeherstegate and the road adjacent to Hechelande. The only road known to be adjacent to Hechelande was the Hastings Ridge ridgeway.
It is obvious to us that the Norman battle camp cannot have been at the Hechelande location described in CBA, unless the battle was fought at Battle, which we think implausible. The English army can only have been coming from Sedlescombe, Whatlington or Netherfield. The view from CBA’s Hechelande to Sedlescombe is blocked by Bodeherste Wood, that to Whatlington is blocked by Petley Wood and that to Netherfield by both Hechelande Wood and Beechdown Wood. It is feasible that the Normans wanted to stay hidden from the English army until they were dressed for battle, but this would contradict Carmen, Wace and Brevis Relatio, which say that the English position was visible from the Norman battle camp.
By a process of elimination, then, we think that the Norman battle camp was at Hechelande, somewhere other than the location described in CBA. Based on an idea Nick Austin shared with us long ago, we think that the monks of Battle Abbey invented a placed named Hechelande near to the Abbey because they knew that the real Norman battle camp was at Hechelande and too far away for the battle to have happened at Battle. There was a danger that another contemporary battle account would name the original Hechelande as the Norman battle camp, making it obvious that the Abbey was not on the battlefield. By renaming somewhere within the Leuga as Hechelande, if this were to happen, it would endorse the Abbey’s claim to be on the battlefield.
CBA’s Leuga boundary description hints this is what happened. It lists ten places around the boundary. All were outside apart from Hechelande which it says was inside. It would have been more consistent to have named Telleham instead. They could then have said elsewhere that Hechelande was near to and northeast of Telleham. It seems to us that they were trying to draw attention to Hechelande, perhaps because they knew that there was another Hechelande elsewhere.
Norman battle camp
Four clues from the contemporary accounts, in addition to the specific clues in CBA and Brevis Relatio, suggest that the Normans had a battle camp:
- The term 'sea camp’ comes from Carmen, which says that William returns there after the battle. The fact that it needed the 'sea’ adjective implies that he had another camp which was not by the sea. Carmen does not mention the bridgehead camp at Peneuesellum, so this other camp was probably somewhere else on the Hastings Peninsula, further inland.
- When Harold and Gyrth go to scout the Norman camp at dawn on the day before battle, Wace says that they see huts, tents, pavilions and gonfanons. If they had been scouting the sea camp, they would surely have taken more note of the burh fortress and William’s kit fortress and the sea.
- According to Orderic, William took possession of Hastingas and Peneuesellum on landing. He says that William left a body of men to cover a retreat and to guard to the fleet. It sounds as if they were at the sea camp. If they were at the sea camp and got left, the rest of William’s army must have gone to another camp elsewhere.
- According to WP, the place where they build the second fortress was a refuge and shelter for their boats. A refuge is a safe place to retreat. Orderic says the men at the sea camp were there to cover a retreat. We interpret this to mean that most of the Norman army moved onto another camp that was less safe and from which they might have to retreat.
Having arrived at the battle theatre, Harold had four options: 1) To attack immediately; 2) To enter the Hastings Peninsula to affect a siege; 3) To send saboteurs across the Brede to scorch-earth around the Norman camp; or 4) To stand off the Hasting Peninsula, scouting the enemy position while awaiting reinforcements.
William would have been prepared for all Harold’s contingencies. He needed the ability to ambush Harold at any of the four possible entry points; the ability to intercept scouts and saboteurs; and the ability to cut off an English retreat, then attack the English camp. Troops at the sea camp (2) would be too far distant and too slow. English spies on the Udimore Ridge would see them leave, giving Harold ample time to respond.
In our opinion, as soon as Harold entered the battle theatre, William’s army coalesced up near the Brede crossing points, at a battle camp. It had to be close to the Roman road between Sedlescombe and Westfield (black line on Figure 35). Not only was this road the most likely way for Harold to try to enter the Hastings Peninsula, but it was also the quickest route to the other crossing points and to the ridgeways that circumvented the Brede and to the Norman sea camp. Carmen says that William’s monk emissary leaves the Norman camp on a road (Latin 'iter'). Latin 'iter' usually means a paved road: the Roman manuscript which listed the paved roads in Britain, was known as the Iter Britanniarum. The road between Sedlescombe and Westfield was the only paved road on the Hastings Peninsula (although there were other metalled roads).
CBA and Brevis Relatio describe the Norman battle camp being ‘a parte Hastingarum’, which Lower and Searle translate to mean ‘in the direction of Hastings’ or ‘towards Hastings’, respectively. These translations are viable, but we feel that ‘to the side of Hastingarum’ is more natural. We think this Hastingarum referred to the Hastings Peninsula although it might have referred to Hæstinga port at Winchelsea. In the first case, it would mean somewhere on the south bank of the Brede. In the second case, it would mean somewhere west of Winchelsea, which also means somewhere on the south bank of the Brede.
One place stands head and shoulders above any other battle camp candidates. It is the ridge above Pestalozzi International Village, along which Cottage Lane now runs (shown in magenta on Figure 34). It probably was a heathland hill, so it might have been known as Hechelande. Once again, we think William planned meticulously with expert local advice from the monks of Fécamp. They would have told him that this ridge had these advantages:
- It overlooked the Rochester-Sedlescombe Roman road along which Harold and his troops would arrive
- It was midway between Whatlington and Brede village where there were ford crossing points
- It had a fast metalled track to the Hastings Ridge via Beauport Park, which could be used to intercept Harold if he tried to enter the Hastings Peninsula using ridgeways via Netherfield
- It gave the opportunity to attack/ambush Harold or his reinforcements on the Roman road before they reached the Hastings Peninsula
- It was ideal for spying on English troop movements, including scouts or saboteurs trying to cross onto the Hastings Peninsula
- It would have allowed the Normans to cut off an English retreat along the Roman road back to Bodiam
- It was adjacent to Oaklands Roman iron bloomery, with everything needed to make and repair weapons, armour and saddlery
- It was at the crossroads of the Sedlescombe to Westfield Roman road and the newly discovered Roman track between Oaklands and Beauport Park, which would have made it the best place on the peninsula for raiding, foraging and driving rustled livestock
- It had a view all the way down the Brede to where the Norman fleet had been beached
- A branch off the Sedlescombe to Westfield Roman road went to Winchelsea, linking the two Norman camps.
If most of this is right, the sequence of events is straightforward. As soon as the Norman scouts spot the English crossing the Rother, William and his barons leave their sea camp to muster with the rest of his army at the Cottage Lane battle camp (O). It is a 4½ mile march for the English to Sedlescombe, an 8-mile ride for William to Cottage Lane – see Figure 35. Even if Harold did not pause, William would have got to the Norman battle camp in plenty of time to dress for battle and prepare an ambush as soon as Harold crossed the Brede.
In practice, it seems implausible that Harold would be stupid enough to try a short-handed attack on a position that the Normans had been fortifying for two weeks. Although this seems to be what most of the contemporary accounts suggest, they were severely abridged. We guess that they simply redacted the events between his arrival at the battle theatre and the battle. The two most detailed accounts, Wace and Carmen, say that he made camp on Thursday and dispatched scouts and messengers to check the Norman position. The Normans were not to know. They dress for battle anyway, pray and receive William’s pep talk. In time, they realise that the English are camping overnight and stand down.
Figure 35: Battle theatre showing topography and roads
There are four points to clear up. First, CBA is the only primary source that specifically mentions a battle camp and none of them mention a possible Norman ambush. They would not. They are written from William’s point of view and he only got dressed at the battle camp. They only reported events, not contingencies. The battle camp played no direct role in the battle and nothing happened there. We think it was redacted, even from Wace. The only reason it was reported in CBA was to support their ‘Abbey on the battlefield’ claim.
Second, why did William not attack as soon as the English crossed the Rother? The Norman cavalry could get to the south bank of the Rother in 30 minutes without breaking sweat. The English army would have been demolished if they chose to fight cavalry on flat dry open ground. We guess that William was swayed by two potential issues. One is that the ground would have been boggy near the riverbank. The English army might have scattered where horses could not chase. The other is that Harold might have got back into a ferry and rowed to safety over the Rother. William might have won the battle, but it was no use if Harold escaped.
Third, some people, Simon Mansfield for one, think the English could not have come down the Rochester road, because the manors between Cripps Corner and the Rother are flagged as ‘wasted’ in Domesday. They presume that the Normans had been foraging there before the battle. If this was so, they reason, the battle would have happened somewhere north of the Brede. It makes no difference to our theory because we think the battle did happen north of the Brede. Anyway, we think it was the English army that wasted those manors to feed themselves at their camp.
Fourth, why did William not attack the English camp at dawn on Friday the 13th? Wace says that the Norman barons urge William to attack as soon as possible because English reinforcements were arriving constantly. William delays by a day. As we have hinted several times before, we think the English were effectively trapped. We guess that Norman scouts at the Rother reported that no large contingent of English reinforcements was imminent. We think William tried to scare Harold into fleeing, having set a trap along the route away from the English camp. It was worth a try. If Harold had been killed fleeing, the Godwinsons might have lost support through his apparent cowardice and William might have taken the crown without a fight.
A Battle of Hastings not on the Hastings Peninsula?
It is natural to assume that the Battle of Hastings must have been fought on the Hastings Peninsula because of its name, but if William had a battle camp at Cottage Lane, as we propose above, it would be virtually impossible. Either Harold tried to get onto the Hastings Peninsula, in which case William would have ambushed him at the crossing point; or he did not try to get onto the Hastings Peninsula, in which case William would have attacked the English camp. Either way, the battle did not happen on the Hastings Peninsula. This is a major departure from orthodoxy and seems to contradict the battle’s name.
Yet the battle's name cannot mean what it implies. There was no fighting at Hastings because it did not exist at the time. Up to and including the time of the invasion, Normans used the term Hastinges to refer to Hæstinga port. There was no fighting there either, or the sea would have been a major factor, and the Normans/French have always referred to the battle location as Senlac anyway. Up to and including the time of the invasion, Anglo-Saxons used the term Hæstingas to refer to the Hastings Peninsula, but the Old English accounts do not use the term 'Battle of Hastings’.
Indeed, there is only one early reference to the Battle of Hastings: Domesday mentions it three times, once each for Ælfwig, Ælfric and Breme who it says died in the 'bello de Hastinges'. There is no clue to what it meant by the term, and it looks like all the subsequent references to the Battle of Hastings took their lead from Domesday.
One possible origin for Domesday's bello de Hastinges is that it referred to Hastings Rape, which covered the entire Rother Peninsula and therefore anywhere that the battle could possibly have been fought. Rapes almost certainly date back to early Anglo-Saxon times, probably having had some military purpose. They became important military divisions after the Conquest, each controlled by a castle. But Domesday refers to the Sussex Rapes by the lord that controlled them, 'Terra Comitis de Ow' in the case of Hastings Rape, rather than to their castle. It is therefore unlikely to have invented the term 'bello de Hastinges' to mean the Rape. It probably took the term from common parlance, but that is unlikely to mean the Rape either, because there are no known Norman references to Sussex Rapes that predate Domesday.
The other likely origin is that bello de Hastinges referred to the place the Normans knew as Hastinges (i.e. Hæstinga port). Perhaps it was named after the Norman camp at Hastinges where William and his barons spent all but three days of the campaign (the exceptions being one day in Peneuesselum, one at their battle camp and one at the battlefield). Or, most likely in our opinion, Hastinges was the nearest named place to the battlefield that William and other Norman aristocrats would recognise. All three of these possibilities are plausible. None of them fix the battlefield on the Hastings Peninsula. Each of them could mean anywhere south of the Rother, which might be up to four miles from the Hastings Peninsula.
Historians put more faith in two traditional clues. Above all, the primary sources that imply Battle Abbey was built on the battlefield. We explain in the Traditional Battlefield blog why we think these references were all fabricated by monks of Battle Abbey to defend their wealth and liberties. The other traditional clue is the list of 'wasted manors' - i.e. those manors that lost most of their value - in Domesday. The argument is that the manors near the battlefield would have been destroyed and most of those that were wasted were on the Hastings Peninsula. The only significant exceptions - Drigsell, Higham and Salehurst - were too low lying to have been the battlefield. We are unconvinced. This was not a WWII artillery battle. The greatest damage would have been close to the Norman camp and wherever they raided rather than at the battlefield, and there is no doubt that the Norman camp was on the Hastings Peninsula.
On the contrary, there are four clues that hint the battlefield was close to the Hastings Peninsula but not on it: Chronicon, Brevis Relatio, ASC-D and Tapestry Panel 48 (Figure 36).
Chronicon is the most straightforward. It says that Harold: “Gave them battle nine miles from Heastinga, where they had built a fort”. As we explain in the Place Names blog, we think that Chronicon's Heastinga referred to Hæstinga port at modern Winchelsea. We assume Chronicon meant Roman miles, which equates to about 8 modern miles. No battlefield candidates on the Hastings Peninsula were less than 11 Roman miles from Winchelsea. The other Norman camp candidates are at Hastings and Wilting but everywhere on the Hastings Peninsula bar Winchelsea is within 7 Roman miles of Hastings or within 6 Roman miles of Wilting. Indeed, 9 Roman miles from either of them would push the battle beyond the Tillingham estuary or out into the Andredsweald, which is a pretty good argument against either of them being the location of the Norman camp.
Brevis Relatio (Dawson translation) says: “Accordingly, coming to a hill which was on the side of Hastingas, opposite that hill upon which Harold with his army was, there under arms, they halted for a short time, surveying the army of the English." This might be the key passage in finding the battlefield, if only it could be understood. It was written by a monk at Battle Abbey. Depending on his ethnicity, this Hastingas could therefore refer to the Hastings Peninsula or to Hæstinga port. We doubt the latter. The Norman sea camp was on a hill at Hæstinga port. It would be unusual to have a hill on the side of a hill and it sounds like they had travelled some distance before coming to it. We think this Hastingas probably referred to the Hastings Peninsula. If they were looking at the English army from a hill to the side of the Hastings Peninsula, the English army was not on the Hastings Peninsula. The term ‘opposite’ hints mirror image, which sounds like there was a body of water between.
ASC-D, in the Ingram and Giles translations, says that “[Harold] came against him at the estuary of Appledore”. We interpret this to mean that the estuary of Appledore – i.e. the Rother – was the closest named place to the battle with which ASC readers would be familiar. They were presumably familiar with Hæstinga port which was not only one of the biggest ports in England but also named in the ASC as the place where the Normans built their second fortress. If the battlefield was anywhere south of the Brede, we think ASC would have said: "Harold came against him near Hæstinga port" or something similar.
Figure 36: Tapestry Panel 48
Panel 48 - Figure 36 - is captioned: “Here the knights have left Hestenga and have come to the battle against King Harold". Panel 40 also mentions Hestenga, albeit spelled with an i rather than an e. It says that the Norman knights go foraging for food at Hestinga. As we explain above, they would not have gone foraging for a few hens and goats at Hæstinga port. They would have gone to the richest farmland in the vicinity, which was south of the ridge on the Hastings Peninsula. For this and other reasons, we think the Tapestry’s Hest[i]enga meant the Hastings Peninsula. If it is being consistent, Panel 48 is saying that the knights left the Hastings Peninsula to attack Harold.
This argument could be corroborated (or debunked) by the likely location of the building on Panel 48. It looks like a stone church or abbey with a rounded apse, clerestory, single storey lateral aisles, a thin stone tower and a giant door. It could be an invention of the artist, but it is difficult to imagine why they would invent somewhere so elaborate. Perhaps they just copied an abbey that was familiar to them. If it is even an approximation of the abbey that was there, it is not Saxon and it is not modest.
The building on Panel 48 looks like it was built by a rich foreign abbey, which in this area can only mean St Denys or Fécamp. The architecture is unhelpful. It is reminiscent of the late 8th century Benedictine Abbey of St John in Val Müstair, which was contemporaneous with the Romanesque Carolingian architecture of Frankish St Denys, but early Norman monasteries were of similar design.
We have spent 40 years examining the 'door' next to the Norman soldier. If it is a door attached to the monastery, why is it so big? On the other hand, if it is part of the next panel, why is it so small? Is it even a door? Note that it is behind the tower's foundations. We guess it is the monks' dormitory, stepped back from the monastery which makes it look small. The horizontal bars make it look like a door. We suspect that the nuns who embroidered the Tapestry were working from a sketch and got confused. We would love to hear from anyone that has other ideas about it.
Considering the Brede side settlements at the time, the only candidates for this monastery's location are Winchelsea, Sedlescombe and Cadborough. Cadbrough, on the north bank of the Brede, would be well placed to manage the salt pans, but 10 miles from the Rochester road, 12 miles from the bloomeries and 20 miles by foot from the port. Moreover, as far as we know, Fécamp Abbey did not hold Rameslie long enough (probably 30 years) to construct such an elaborate building. We think the monastery was probably built by St Denys Abbey at Winchelsea or Sedlescombe, then occupied by the monks of Fécamp when they acquired the land.
Kathleen Tyson thinks this building is the main Fécamp monastery on the site subsequently occupied by St Leonard’s at Winchelsea, thereby substantiating her theory that the Normans were leaving their sea camp at Icklesham. Her only evidence, as far as we know, is that Fécamp Abbey once held this land in Winchelsea. They still held it at the ‘Dissolution of Alien Priories’ during the reign of Henry V. They must have had a building in Winchelsea to administer tolls. But we doubt it was the building shown in Panel 48.
For one thing, if Panel 48 is depicting Winchelsea and Icklesham, we would expect it to show a fortress or two, the tower, the hill and/or the sea. The horse is on the baseline rather than on bobbles, which suggests it is on a riverbank or road. Then there is the nature of the medieval foreign cells, whether they were from St Denys or Fécamp. The clerics were businessmen rather than religious zealots. Businessmen, unless they have changed a lot over the intervening centuries, would want to live somewhere sheltered, comfortable, well connected, safe from Viking raids and close to lots of peasant girls, which would disfavour Winchelsea.
Fécamp Abbey held land on both sides of the Brede estuary, and at Horse Eye, Eastbourne and Steyning. St Denys held land south of the Brede and at Rotherfield, Peuenisel (near Pevensey) and London. In either case, if we were the first abbot, we would commission our HQ beside the Rochester Roman road, south of the Brede at Sedlescombe, where it would be well connected to the port and to our other lands. It would also be close to farmland and running fresh water from the fluvial Brede, for food and water. Perhaps most importantly, it would be at the end of the road down from Beauport Park, the biggest iron bloomery in the country, where the output could be assessed and taxed. We think it offered the best place in the region to minimise theft, smuggling and toll avoidance.
We have no evidence to support this theory, but we wonder about the tower in Panel 48. It looks too thin for a bell tower and it has no windows at the top. We guess it was a watchtower used to communicate with other outposts. This makes it unlikely to have been at Winchelsea, which already had two of towers. If it was on the Udimore Ridge exchanging messages with the Panel 45 tower at Winchelsea, it would not have needed to be any taller than the surrounding vegetation. Likewise if it were north of the Brede at Sedlescombe. But if it was south of Sedlecombe bridge, it would have needed to be 20m high to see the tower at Winchelsea over the shoulder of the Cottage Lane ridge. Perhaps it exchanged news of incoming trading vessels, recently dispatched cargo barges, or of imminent Viking raids.
We are reminded of Frank Johnson’s discovery of huge timber foundations at Old Orchard, south of Sedlescombe bridge. He thought they were from a wharf, but they were 50m from the river. We wonder whether they were the monastery's foundations. And the point of all this is that if the building behind the Norman knights on Panel 48 was at Sedlescombe, the knights were leaving the Hastings Peninsula.
The English camp
Figure 37: East Sussex topography, roads and trackways
By tradition, the English camped on Caldbec Hill (CH on Figure 37). English Heritage have a plaque at the park entrance that provides some details. The tradition is based on some translations of ASC-D (Dorothy Whitelock, for instance) that say: “[Harold] assembled a large force and came against him [William] at the hoary apple tree". Historians think this tree was a hundred junction marker on Caldbec Hill. We dispute the translation and we do not think the English could have camped at Caldbec Hill anyway. We discuss this in the Alternative Battlefields Theories blog.
Harold seems to have deduced from court spies and messengers that the Normans were weak, footbound and too far distant to pose a threat outside the Hastings Peninsula. Having crossed the Rother by ferry, he probably scooted through to Cripps Corner (C on Figure 37), just in case Lordship Wood harboured Norman snipers. What then?
Two early and trusted accounts are traditionally interpreted to be saying that the English were attacked as they approached the battle theatre, perhaps trying a surprise attack. WP says that Norman scouts: “announced the imminent arrival of the enemy, because the King in his fury had hastened his march … He intended to crush them in a surprise or nocturnal attack … The Duke put on his hauberk reversed to the left … The speech with which he rallied the courage of his troops, though brief due to the circumstances, …”. WJ says that: “Hastening to take the Duke by surprise, Harold rode through the night and arrived at the battlefield at dawn. But the Duke had taken precautions against a night attack. He had ordered his men to stand by until dawn. At first light, having disposed his troops into three lines of battle, William advanced undaunted against the terrible enemy”. CKE, Chronicon, Orderic, Benoît and CBA do not mention anything happening between the English march and the battle, which gives the impression that the events were contiguous.
Wace and Carmen, on the other hand, say that the English arrived at the battle theatre on Thursday and camped there for two nights. Both give detailed accounts of negotiations between William and Harold on Friday. Wace describes both sides’ scouting activities and tactical planning, as well as a detailed description of the English camp.
WJ is not saying what it seems. He says that Harold arrived at dawn, then that William had set a night guard to be ready for a nocturnal attack, then that William deployed his troops to attack at first light. The narrative sounds like the events were contiguous, but they cannot have been. Harold arrived at dawn. There would be no point mentioning the Duke’s night guard precautions for the night Harold arrived because the night was over. There would be no point mentioning that the Normans attacked ‘at first light’ if they attacked as soon as Harold arrived because it had just said that he arrived at dawn. The night guard must have been for a subsequent night. The Norman attack must have been on a subsequent dawn.
WJ, ASC, CKE, Chronicon, Orderic, Benoît and CBA are histories or chronicles that cover hundreds of years. Each allots just a few paragraphs to the battle. WP is about William’s life. It has a few pages on the battle. Wace and Carmen are dedicated to the conquest. They have far more detail. Admittedly, they are less than scrupulous, but they had no incentive to invent events that did not glorify Norman culture or denigrate Saxon culture. Which is more likely, we reason? That Wace and Carmen invented the English camp, scouts and messengers, or that the others redacted details that are not relevant to their narrative? We believe Wace and Carmen.
As for WJ and WP, we guess their term ‘battlefield’ meant what we would now refer to as the ‘battle theatre’. If so, they are not wrong. They are trying to say that the Normans did indeed prepare for battle as Harold approached the battle theatre in case he kept marching onto the Hastings Peninsula. Only he did not. He camped somewhere north of the Brede, so the Normans stood down. William set a night guard on Thursday night. The Normans prepared again for battle on Friday, but Harold did not attack, so they spent the day scouting the English camp. The Normans set another night guard expecting a nocturnal attack on Friday night. Again, it did not come, so they prepared yet again for battle on Saturday and attacked at first light.
By tradition, the English were on the march when they were attacked. Carmen provides some details. It says that as the Normans approached, the English emerged from a wood: “Suddenly the forest poured forth troops of men, and from the hiding places of the woods a host dashed forward ... there was a hill near to the forest ... they seized possession of this place for the battle."
But then WP and Wace give the impression that the English were attacked in their camp. WP says that when the Normans attacked: “... the English were camped on higher ground, on a hill close to the forest through which they had come". Wace says that Harold marched from Westminster to: “where the abbey of the battle is now built. There he said he would defend himself”.
Wace confuses matters with his reference to Battle Abbey. Stent explains that a medieval commander’s most important task, and the greatest factor in a battle’s outcome, was choosing the battlefield location. Harold would not have chosen to fight where the Abbey now stands. It was a miserable defensive position. In the unlikely event that he crossed onto the Hastings Peninsula, he would have headed for Cottage Lane or Lower Snailham. We explain in the Traditional Battlefield blog how we think Wace was misled. A brief summary. Roman de Rou is based on first-hand accounts taken by Wace's father. We think that Wace Senior recorded first-hand accounts that the Normans attacked the English camp. He could not have recorded any first-hand reports that it was where the Abbey would be built because the Abbey was not started until seven years after they left. We guess that Wace Junior read that the Abbey was built on the battlefield and linked the two, wrongly deducing that the English camped at Battle, where they were attacked and defeated.
Having arrived at Cripps Corner, we are pretty sure that Harold would have done what anyone sensible would have done in his shoes: to send scouts to check the Norman strength, deployment and food availability. At this time, still ignorant of the Norman cavalry, he was probably thinking of an attack on the Norman camp as soon as his reinforcements arrived a couple of days hence. Meanwhile Harold needed a camp that could be used as a forward operations base. The contemporary accounts give us some clues:
- Carmen says that William's monk messenger goes to the English camp on a road; Latin 'iter', which usually means a paved road. The only paved road in the region was the Rochester Roman road. The English camp is likely to have been near it anyway. It was whence the English reinforcements would arrive: He would not want them getting lost or ambushed when they left the road. It was the only route to safety. It was the only easy or dry way to get on or off the Hastings Peninsula. It was both Harold’s best route of attack and the best place to ambush a Norman infantry sortie.
- Carmen reports a conversation between William and his returning messenger. William asks: "Where is the King?" The messenger replies: "Not far, you can see his standards". The Norman and English camps cannot have been more than two miles apart, probably less, with a treeless valley between.
- Wace says that Harold and Gyrth reconnoitre the Norman camp at dawn on the day before battle. Leofwine wakes early, spots they are missing, and goes to find them. He meets them on their way back to camp. It is still early. Harold and Gyrth cannot have ridden more than a few kilometres.
- Wace says that Harold and Gyrth can see the Norman huts, tents, gonfanon and armour from their scouting location. They can hear the Norman horses. They were unguarded and cannot have been more than one kilometre from the Norman camp. The geography must have given them some sort of protection to feel safe that close to the Norman camp.
- Brevis Relatio talks about the initial encounter (Dawson translation). “Accordingly, coming to a hill which was on the side of Hastingas, opposite that hill upon which Harold with his army was, there under arms, they halted for a short time, surveying the army of the English." We explain above that we think this means the Normans were on a hill at the edge of the Hastings Peninsula looking at a hill with the English army, probably across a river.
- Just before the Normans leave their camp to launch their attack, Brevis Relatio says that William asks a nearby soldier where he thinks the King might be. The soldier replies that he thinks the King: "was in the midst of that dense array, which was before them on the top of the hill, for as he was thinking, he saw Harold's standard there". It sounds like the entire English deployment is visible from the Norman camp and not much more than a mile away.
- WJ, Orderic and Chronicon say that the battle started at the third hour of the day. Given an hour to armour horses and knights, 15 minutes for William’s pep talk, 15 minutes to assemble into line and 15 minutes to disassemble into divisions, the camps could not be more than an hour’s march apart, probably a lot less.
All of this suggests that both camps were beside the Roman road and separated by no more than two miles. The fact that the English Standards were visible confirms that the English could not have camped any further north than the Udimore/Rother-Isthmus ridge, which was two miles away from the Brede. But did the English camp on the Udimore/Rother-Isthmus ridge or south of it?
A camp on the ridge would have been the safer, being just two downhill miles to the Rother. A camp south of the ridge would have been more positive and aggressive. Harold’s nature suggests the latter, but the primary sources provide no clues.
An English camp at the battlefield?
It did not make sense to us that the English would leave their fortified camp to fight a defensive battle somewhere else. The traditional explanation is that they were caught trying a surprise attack. Poppycock. Harold knew of the Norman cavalry by the day of battle. He and William had been exchanging messages, so each knew the other’s camp. William had prepared for a surprise attack. There is no way Harold would have tried a shorthanded attack on the Norman camp if the Normans were prepared. Having discovered the Norman cavalry, Harold would either retreat to Bodiam or fortify the English camp while they awaited reinforcements. If they had retreated, the battle would not have happened. We therefore assumed they fortified their camp and stayed there, exactly as Wace suggests.
We originally concluded that Carmen was either mistaken about the English emerging from a wood as the Normans approached – albeit an oddly detailed mistake – or it was trying to say that reinforcements were arriving from a wood when the Normans attacked the camp. Wace does say that English reinforcements were arriving all the time, so perhaps some of them poured forth from a wood to augment those that were already in their camp.
If the English were attacked in their camp, it had to fit descriptions of the battlefield. All the primary sources agree that the ground was steep and difficult. Carmen says it was untilled; probably heathland. WP says that at the start of their attack: “The Duke and his men, in no way daunted by the difficulty of the place, began slowly to climb the steep slope". CKE says that the English: “roused with indignation as the Normans strived to gain the higher ground, drove them down into the valley beneath, where hurling their javelins and rolling down stones on them as they stood below, destroyed them to a man". Carmen says: “In summo montis vexillum vertice fixit", which Morton & Muntz translate as: “On the highest point of the summit he planted his banner".
Historians have always interpreted these statements to mean that the battlefield hill is high, steep, conical and topped by a distinct summit. Their reasoning is not complicated. Carmen says that Harold planted his banner on a ‘montis’, Latin for mountain. There are no mountains in southern England - presumably why Morton & Muntz ignore it - but it sounds like it must be a high hill. “Slowly began to climb the steep slope" implies that the hill is high and steep. Carmen seems to confirm this by going on to say that the Duke: “boldly approached the steep slope". For stones to be rolled as weapons the hillsides must be steep. The Normans would not attack up a steep slope if the battlefield hill had alternative shallow slopes. The only hills that do not have shallow slopes are conical or ridges. Ridges seldom have distinct summits. Q.E.D.
Figure 38: English camp candidates
We looked for a high steep conical treeless hill that was close to the Roman road and close to a freshwater stream. There are none. There are a few raised sections along the Rother Ridge near its intersection with the Roman road at Cripps Corner. By definition, they are conical near the top. We examined them. Compasses Hill (B on Figure 38) is most similar to our traditional image of a steep conical battlefield hill, but it is only 200m across and it is covered in trees. Hook’s Beech (C) is a better size for an English camp site, but it too is covered in trees. Yeakell and Gardner's 1770 map and MAGIC suggest that both have been woody since ancient times. That left ‘The Beacon’ (A).
We spent weeks investigating The Beacon (apart from its battle possibilities it has some fascinating WWII anti-tank placements). It is a plausible camp and battlefield. It has sparse tree cover, a flattish top, several nearby lakes and it is steep around perhaps 240° of its circumference, especially to the west. On the other hand, it is big and flat-topped to be defended by 6,000. The enclosed shield wall around the flat-top shown in green on Figure 39 is 1750m long, nearly twice the length at the traditional battlefield. Also, it is at the ‘T’ junction of the Udimore Ridge and the Rother Ridge, which gives it three shallow ridge-crest approaches: to the north, east and southwest.
Figure 39: Possible troop deployments at The Beacon
Harold might possibly have protected the northern approach with a fosse across the ridge crest (shown in red). If so, he could have covered the shallow east and southwest approaches eight deep and still leave enough men to be two deep elsewhere. Perhaps he had more men than we calculate. Our bigger issue is that there was almost certainly a ridgeway (yellow dots) on the Rother Ridge that crossed the western side of the summit. It would have given the Normans direct access to the shield wall on an easy shallow dry slope, quite unlike the primary source descriptions of steep slopes on difficult untilled ground. It also lacks some of the more enigmatic details mentioned in the primary sources, such as it has nowhere that might be described as a “a narrow place", no way to “enclose the battlefield" and it is not above a plain.
Try as we might, we could not come up with a scenario that matched The Beacon to the primary source camp or battlefield descriptions. William would clearly have split his forces to attack from different directions, whereas the primary sources say that they attacked in three divisions from the same direction. They would have attacked on the shallow approaches which would contradict the primary source descriptions of a steep untilled slope. There was no valley in javelin range. Stones would not be dangerous on any slope the Normans would use to attack. The English would not have entered the battlefield in a column. It has nowhere that might be described as a “a narrow place", no way to “enclose the battlefield" and it is not above a plain. We could think of no way, given what was at stake, that a fosse could not be bypassed. The line would be too thin away from the shallow approaches. We are convinced the Normans would have contrived a way to break through on the shallow southern slope.
Reluctantly, we abandoned the combined English camp/battlefield theory. This meant we would have to come up with a plausible explanation as to why the English might leave their fortified camp to fight a defensive battle somewhere that was unfortified; and we would have to explain why Poitiers and Wace imply that the English were attacked in their camp if they were not. But it did mean that the English could have camped south of Cripps Corner on a hill that was not high, steep, conical or treeless.
English camp on Great Sanders ridge
There are three candidates for the English camp between the Brede and the Udimore/Rother isthmus ridge: Woodmans Green (H on Figure 38), Cackle Street (F) and Great Sanders ridge (D). Each faced a hill across the Brede from where the English standards would have been visible.
We concluded in the Wargaming section that Great Sanders ridge was the most likely English camp, because its proximity to the Roman road made it the best place to receive reinforcements and its position midway between the three Brede crossing points gave it the best offensive and defensive options.
If we are right that the Norman battle camp was at Cottage Lane, Great Sanders ridge is the only candidate from where the English Standards would have been visible. It is the only candidate from where Harold and Gyrth could have gone scouting the Norman camp without dropping down into the Brede valley. They could have ridden out to Balcombe Green or Brede Barn Farm, both of which were hills overlooking the south bank of the Brede just 500m away.
Besides, if the English camped at Cackle Street or Woodman’s Green, the Normans would have marched up the Roman road to Cripps Corner, then out along the Udimore or Rother ridge respectively to attack from behind and above. They did not. If the English camped north of the Brede, Great Sanders ridge is the most likely place.
Harold or his delegates did not know about the Norman cavalry when they chose the English camp. Given what they did know, Great Sanders ridge would have seemed an ideal place for the English for an operations base and camp. It is a good defensive location. It was adjacent to and controlled the Roman road. It is just a mile from the Brede, ideal to watch all the Hastings Peninsula egress points. It is close enough to blockade those egress points or to ambush the Normans if they tried an infantry sortie. It was well placed for an English attack over any of them. It was 2km from where we think the Norman battle camp was located at Cottage Lane with nothing to impede the view, comfortably close enough for William to see the English standards. It was within striking distance of where Harold’s messenger had found William at his sea camp. We are convinced it is where the English camped.
An English camp on the Great Sanders ridge solves two other vexing puzzles. One is that Wace says that Harold had the English camp surrounded by “a good fosse, leaving a well-guarded entrance on three sides". Previously we did not believe Wace on this point. Harold was only at the English camp for a day. Spades were tiny in those days, like lawn edge-cutting tools. Even if they brought five hundred of them, which does not seem likely, how could they dig a useful 2km fosse in a day, especially among tree roots. But the Great Sanders ridge has a steep slope to the north and great gouges out of its rugged southern slope, as if someone has ravaged it with a 20m wide ice-cream scoop. These form a natural fosse. The man-made fosse would only have needed to block the vulnerable eastern and western ridge ends. If they had enough spades, they could have surrounded a camp on Great Sanders ridge with a useful fosse in a day.
The other puzzle is that, according to Wace, Harold and Gyrth went alone to reconnoitre the Norman camp on the day before battle. He says they: “rode on, viewing and examining the ground, till from a hill where they stood they could see those of Normans, who were near. They saw a great many huts made of tree branches, well equipped tents, pavilions and gonfanons; and they heard the horses neighing". The route from the camp to the viewing hill must have been secure underfoot, along a ridge perhaps. The view to the Norman camp must have been unimpeded by trees. They must have been no more than 1km from the Norman camp, in order to hear the horses. So, how could Harold and Gyrth have felt safe to be out in plain sight on their own – well, apart from a few squires perhaps – and that close to the Norman camp? The answer at Great Sanders is that the Brede was in between. We guess they rode along the mid-west spur-crest to Balcombe Green (B on Figure 34).
In summary, we are convinced that the English camped at Great Sanders ridge. It was a defensively sound camp of the right size for 6,000 troops. It overlooked all the Brede crossing points. It had a view of the LIN 130 ridgeway, along which the Normans would have to march to circumvent the Brede. It was strategically well placed to attack a Norman camp at Winchelsea and to defend against a Norman sortie from Winchelsea. Or at least it would have been if, as Harold thought at the time that he chose the camp location, the Normans had no cavalry. It is the only plausible camp location that provides a credible explanation for why Harold did not withdraw back to Bodiam when he discovered the Norman cavalry. It is the only credible camp location that would lead to a battle on Saturday.
This brings us to a puzzle that baffled us for thirty years. Harold could not have known about the Norman cavalry when he was at Bodiam, or he would not have crossed the Rother. He could not have known when the English camp was chosen, or he would not have allowed a camp south of Cripps Corner. But he must have known by the time of the battle, even if he only found out that morning, or he would not have fought a defensive battle with no chance of victory. So why did he not immediately leave to summon reinforcements or lead the English army back to safety at Bodiam as soon as he finds out?
According to Wace, Harold and Gyrth get an idea of the strength of the Norman army when they scout the Norman camp at dawn on the day before the battle. Harold suggests to Gyrth that he, Harold, should return to London for reinforcements. They have an argument. Gyrth replies that abandoning his troops would be viewed as cowardly; that he would permanently lose their respect. Harold accepts Gyrth’s argument and decides to stay.
We are sceptical about Wace's provenance. He says that Harold and Gyrth went scouting alone. Both died in battle. They would not have told anyone some of the details that Wace reports. Perhaps they were accompanied by squires that survived to tell their story. Or perhaps Wace invented some or all of the English side of his narrative.
It makes no odds. Harold must have known about the Norman cavalry by the morning of battle, or he would not have deployed his forces as an enclosed loop (see below). Harold might not have felt able to leave his troops, but why did he not withdraw the entire army? At Hurst Lane, it was not too late to retreat when he saw them line up at the bottom of the hill. They only had to get to Cripps Corner, 1500m away. If Harold's scouts told him about the strength of the Norman cavalry as they crossed the Brede at Sedlescombe, the English still had time to get to Bodiam 4 miles away.
Historians often say that an English retreat would be deemed cowardly. We disagree. Harold just had to spin a plausible reason for retreat that did not make him look cowardly. People were superstitious in medieval times. He could say that God had come to him in a revelation to tell him to withdraw, or that his soothsayer had dreamed it was the best way to win, or that the Normans had brought dastardly Greek-fire against which they needed extra supplies from Bodiam. He could have told them that the Normans were sailing around the Rother to attack from the north, so the English needed to get there first. There was no Internet in those days. He could spin whatever fibs he like with no chance of getting caught out.
Why then did the English not retreat? The only plausible explanation we can come up with is that the English army was effectively trapped and that Norman scouts informed William that no major English reinforcements were imminent. Thus, William was better off spending Friday trying to intimidate Harold into flight (having laid a trap for him), scouting the English camp and devising a good plan of attack.
William must have been worried about Harold and/or his brothers fleeing successfully. It was a huge risk to the success of the invasion. On the day of battle, Baudri of Bourgueil explains: “Backing up the enemy line, at a distance, were horsemen waiting to intercept anyone trying to flee”. We think that William had riders waiting to intercept anyone trying to flee from the moment Harold entered the battle theatre.
Figure 40: William's trap
Figure 40 shows Harold's predicament. Great Sanders ridge, with the English camp shown in cyan, is surrounded by the Udimore and Isthmus ridges, shown in white dots. We guess that William left riders at his Peneuesellum camp specifically so that they could spread out along the Udimore ridge as soon as Harold passed Cripps Corner. Failing that, he could have sent them around to the Udimore ridge via the ford at Whatlington or Brede. Their main job would be to catch Harold or his brothers if they tried to flee.
A few hundred riders on the Udimore Ridge could not hold the entire English army if Harold decided to retreat. But the main cavalry at William’s battle camp, shown in magenta, could canter to Cripps Corner before the English could run there. If the English army tried to retreat, they would get caught on open ground and would have been annihilated.
As soon as Harold discovers the Norman cavalry, regardless how he found out, battle was inevitable on the 14th. William needed to slay Harold before English reinforcements arrived. Harold could not retreat in person or lead the English army to safety. He could not attack because the English would get ambushed as they crossed onto the Hastings Peninsula. He could not blockade the Hastings Peninsula egress points, because the Normans were already guarding them.
Roughly two hours after dawn, the Normans left their battle camp to attack the English camp on Great Sanders ridge. This camp does not fit the primary source descriptions of the battlefield. Harold must have moved the English army elsewhere when he saw the Normans coming. In this way, all the primary sources are right: the Normans did attack the English camp, only the English left to occupy a nearby hill when they saw the Normans coming.
The English could not have gone far in the time available. We walked around the outside of the Great Sanders ridge pondering why the English might have left and where they might have gone. Eventually it crossed our minds that a landslide from Great Sanders ridge might have filled in the northern slope of a former conical hill, thereby leaving it more like a spur with only three sides. We thought about the hollows on the south side of the ridge and wondered whether the land that once filled those hollows had slipped down towards the River Brede.
To check, we walked out onto the footpath (at 50.9424,0.5434) between Hurst Lane and Sedlescombe to look up at the ridge. There was no obvious landslide. We looked around. It was a bit like when Harry Potter realises that he is descended from Ignotus Peverell, opening a chain of discovery about the Deathly Hallows.
Then, another revelation. Wace says that Harold issues orders that: “all should be ranged with their faces towards the enemy”. Once again, we previously thought this must be a mistake. Why would anyone face away from someone trying to kill them? But it makes perfect sense at Hurst spur because most of the English were fighting back-to-back. Harold was telling them to hold their position in the line whatever was going on behind.
Then a new revelation. This must be what John of Worcester meant by: “the English were drawn up in a narrow place”. It could not have been that narrow or the Normans would not have been able to deploy in three divisions. It could not have been narrow in both directions or it would be too small to hold the English army. The alternative is that it was a ridge or spur, narrow in one direction but deep in the other. But why then would the Normans attack the narrow front of the shield wall rather than the long sides, where they could use an oblique order attack? The answer at Hurst Lane is that the long sides were protected by boggy streams. If the Normans tried to attack the long sides of the shield wall, they would get shield charged into the bog. They were forced to focus their attack on the narrow front of the English shield wall, which was barely 100m across and protected by a fosse .
One more. Wace says that William chooses to: “fight in the middle throng, where the battle shall be hottest". This never made any sense to us. At any normal place, William would have employed an ‘oblique order’ attack, where the fighting should be more intense on a flank, but not at Hurst spur. At Hurst spur the Norman flanks would have done little real fighting, because the slope, rugged ground, gloop and missiles would have made it virtually impossible for them. Their main job would be to occupy the English flanks to prevent them sending reinforcements to the melee point at the front. At Hurst spur, the fighting would have been hottest in the middle throng.
And yet another. CKE says that the English throw javelins at the Normans in the valley, destroying them to a man. How can this make sense? Lots of primary sources talk about their slow climb up a steep slope to get to the English shield wall. They must have started in the valley. The climb sounds like it was a kilometre or more. So how could the English be killing them with javelins? But it all makes sense at Hurst spur. CKE was talking about the valleys at the sides of the battlefield, which were no more than 150m from the English shield wall. The boggy banks probably prevented anyone getting within 50m of the streams. With the slope and the ability to run out of the shield wall, the Norman flanks were in range of English javelin throwers even if they were in the valley.
That was five of the most perplexing puzzles from the primary source battlefield descriptions solved in five minutes. As we went through the other primary source battlefield clues, we came to realize that there was a credible explanation for all of them at Hurst spur. We will run through them shortly. But this battle scenario feels wrong. The battlefield and troop deployments are too different from tradition, too different from how we imagine it. We will explain the major discrepancies.
We believe that the English shield wall was a plectrum-shaped enclosed loop (Figure 41). The orthodox narrative depicts the shield wall as a straightish dogleg, gently bending around the south slope of Battle Hill. It is more how we imagine shield walls to be. That is because they usually faced enemy shield walls. Victory was achieved by breaking or outflanking the enemy line, which required a straightish shield wall to maximise the men engaged. But an archerless shield wall cannot defeat cavalry. It can only survive.
The best chance for infantry to survive a cavalry attack is to enclose a space with spears facing outwards. This was a well-known tactic employed by Roman legions against cavalry attacks and was still effective during the Napoleonic wars. It was used two weeks previously at the Battle of Stamford Bridge by Harold Hardrada to defend against the large English cavalry. The Norse army just needed to survive too, while they awaited reinforcements. The English found it impregnable until the wall lost discipline. If Harold did not know already, he would surely have learned from that experience and would have deployed is troop similarly. The contemporary accounts concur. Not one of them suggests the English deployed a straight or open shield wall. Seven of them suggest an enclosed shield wall, some multiple times. When we list the statements, it seems so obvious.
There are four more clues that suggest the English flanks bent back on themselves, without going so far as to suggest they joined to make an enclosed loop. We have already mentioned the first three.
All these clues require some degree of interpretation and, to some extent, are ambiguous. Not so for one final clue, perhaps the most explicit and telling of them all. Baudri of Bourgueil says: “The enemy, discarding their horses, form themselves into a close wedge”. So, the English shield wall was an enclosed wedge-shape.
It is little wonder that no one has considered the hill along lower Hurst Lane as a potential battlefield before. It is bisected by a road which disguises its topography. At no more than 67m above sea level it is puny by hill standards. It is not even a proper hill, but a spur off the Great Sanders ridge.
Yet a spur battlefield would perfectly answer why the English were deployed in a wedge shape: they followed the contours to ensure they were all on rising ground. Moreover, it would resolve two apparently inconsistent clues.
CBA reports the monks of Marmoutier discussing the placement of William’s abbey, which is usually translated: “having viewed the scene of the battle, judged it an unsuitable site for so noble a building, but thought a lower place on the western side of the hill more eligible.” Again, a tautology: anywhere away from the summit of a hill is lower down. As with Carmen, we think it is trying to say that the more eligible site is lower down the crest of the spur on the western side of the hill.
But a spur has a major and obvious drawback compared to a conical hill: it has a downhill approach from its parent ridge. The Normans would not attack up a steep and slippery slope if they could instead get onto the parent ridge to attack from above. Admittedly, this is less of a drawback compared to the orthodox English battlefield deployment, or indeed compared to any of the alternative battlefield deployments that have been suggested hitherto, because they all depict the English shield wall as a straightish line part way down the slope. They all rely on implausible impenetrable woodlands or untraversable streams to prevent the Normans outflanking or looping behind the English shield wall. Still, they all have the option to deploy an enclosed English shield wall that is entirely on rising ground.
Hurst spur has a trick up its sleeve. It is protected upslope by giant gouges, presumably the legacy of former iron working activity, that look as if they have been excavated by a 20m wide ice-cream scoop. This can be seen in the hill profile shown in Figure 42. If the Normans tried to attack from the English camp on Great Sanders ridge, they would be fighting uphill on a horrible 45° slope. Worse, they would be trapped in the dip and vulnerable.
Hurst spur has another major advantage compared to other spurs in the region, which is that it is narrow, bounded on both sides by boggy streams. Today they are only 1m wide. A lady that lived in the area as a child told us that they were sometimes 3m across. The gap between them is 400m to 450m. The boggy banks were about 25m wide on each side following the dry autumns of 2014 and 2016. In medieval times the ground would have drained less well, and the water table would have been higher. We guess that the boggy banks might have extended 50m or so from the streams at the time of the battle, leaving just 300m to fight at the pinch point.
Hurst spur has a flattish top some 100m wide at the pinch point, 300m wide at the top. Assuming the boggy banks stretched 50m from the bounding streams, the fighting area would have been just 100m to the west and 125m to the east at the pinch point, narrowing to nothing at the top of the slope. In other words, the further the Normans ventured up the battlefield slope, the narrower the good ground, the more likely they would be shield charged into the boggy stream banks. The danger was not only getting their clothes dirty. Backing into boggy ground almost inviably causes stumbles, with those on the bottom getting crushed or drowned, while those on the top would get hacked to bit by English axes.
While the sides of the English shield wall would have been dangerous for infantry, they would have been almost impossible for cavalry. Horses would be under constant risk of slipping or tripping. A dismounted rider would find it virtually impossible to get back on. The further a rider strayed from the bog, in the hope of better ground, the greater their danger from javelin and hammer throwers. The rugged ground would prevent them getting up any speed and the steepness of the hill at the contact point would have blunted a traditional Norman couched lance attack.
The 18th century Yeakell & Gardner hachure map (Figure 43), despite its inaccuracies, perhaps gives a better impression of the battlefield, like a finger pointing down the river.
The changes in relief are not big enough to show on standard OS maps. Tapestry Panel 54 (Figure 44) perhaps gives a better impression. It is trying to show a front elevation of the hill from William’s point of view. The proportions between the ground, the horses and the figures has gone awry, but it shows a reasonable interpretation of the flattish top and steep sides at Hurst spur, with a flat top and the west slope being steeper than the east.
The narrowness of the battlefield answers four tricky clues in the contemporary accounts. We have already mentioned John of Worcester’s statement that “the English were drawn up in a narrow place” and Wace’s that William chooses to “fight in the middle throng, where the battle shall be hottest".
The fourth is that Wace says no Norman barons died at the battlefield. We used to think that this was totally implausible, but Hurst spur has an explanation. The front of the shield wall was protected by a fosse so effective that Wace reports that no one survived its crossing. The sides of the shield wall were no safer; so narrow, the further a rider ventured away from the streams looking for better ground, the more they were in range of English missiles. In practice, after sacrificing a few junior riders at the start of the fray, we guess that the Norman cavalry, including all the Norman barons, did not participate in any fighting until the dénouement.
The Battle of Hastings battlefield, according to received wisdom, was big, high and steep. It is one of the few things upon which the contemporary accounts seem to agree. Hurst Lane spur, on the other hand, is small, low and shallow, apart from 10m in front of the shield wall. Ostensibly it seems an unlikely battlefield. Yet a couple of clues from the contemporary accounts suggest that the traditional picture of the battlefield is out of scale.
If the battlefield is a lot smaller than we have been led to believe, there must be something wrong with the traditional interpretation of the sources that suggest the battlefield is high and steep.
First, Poitiers and Carmen. Poitiers says: “The Duke and his men, in no way daunted by the difficulty of the place, began slowly to climb the steep slope", Carmen says that the Duke: “boldly approached the steep slope". Poitiers makes it sound like the battlefield was especially difficult because of its steepness. The slope from Brede Lane is a shallow 1:20, with a 50m plateau below our proposed shield wall. It does steepen to 1:8 in front of the shield wall, which is perhaps what Carmen is trying to say, but that would not apply to Poitiers’ context.
Perhaps Hurst spur is steeper than it sounds. Crecy was also described by French commentators as a steep hill that hindered cavalry charges. It too was only 1:20, and despite a Summer downpour before the battle it is unlikely to have been as wet and slippery as coastal England in Autumn. We think Carmen is accurate while Poitiers was trying to say that the hill was steep enough to be a difficult climb for the Normans infantry who would have been barefoot or wearing slick-soled leather sandals on ground that was wet, slippery and tangly. Moreover, we interpret Poitiers’ statement to mean that the battlefield was difficult in ways apart from its steepness. Hurst spur has plenty of other difficulties. The ground close to the streams would have been boggy. Runoff from the Great Sanders ridge would make the ground away from the streams wet and slippery. The slopes would have been tussocky, covered in gorse, heather and other dwarf shrubs. It would have been steep where it mattered: 15% or so at the contact points. And, as we have said, Hurst spur would have been dangerously narrow for cavalry.
In summary, we think there is a plausible explanation that matches Hurst spur to the contemporary account statements that are traditionally interpreted to mean that the battlefield is big, high or steep, even though it is none of these. Conversely, there are at least three sources that imply the battlefield was small, tight and narrow, just like Hurst spur.
Searle’s ‘This place’ might refer to where they built little huts or it might refer to the battlefield. Every professional historian that has looked at this passage interprets it to mean that the huts were at Herste, so they reject it as a useful locational clue for the battlefield. Porter says: “Searle’s translation uses a standard usage known as a connective relative to differentiate between the monks in the first sentence and the battlefield in the final sentence”. We have no idea what he is trying to say. ‘connective relative’ is a figure of speech not writing. We will return to the Latin momentarily. For the time being it is just worth noting that Folio 12 might be saying that the battlefield was at Herste or that huts were at Herste, although they were ‘not far off’ from each other, so there is a good chance that they were both in Herste.
The confusion comes from Professor Searle’s translation. The original says: “Qui memoratum belli locum considerantes cum ad tam insignem fabricam minus idoneum, ut uidebatur, arbitrarentur in humiliori non procul loco, uersus eiusdem collis occidentalem plagam, aptum habitandi locum eligentes ibidem ne nil operis agree uiderentur mansiunculas quasdam fabricauerunt. Qui locos, hucusque Herste cognominatus, quandam habet spinam in huius rei monimentum.” Searle’s translation is above.
Note how she breaks down one Latin sentence into three English sentences. When she translates ‘Qui locos’ to ‘This place’ it gives the impression that it is referring to the last of these sentences, to where the monks built their little huts. But this would be ridiculous. Why would CBA bother noting the name of a place where some monks built temporary little huts rather than the name of battlefield? It is clear to us that ‘Qui locos’ refers to the subject of the Latin sentence, the battlefield. It is saying that the battlefield and the little huts are in Herste. But is it believable?
The logic is pretty much the same. The monks of Battle Abbey were trying to protect their wealth and independence by claiming that the Abbey was built on the battlefield. It would be dangerous for them to claim spuriously that the battlefield was at Herste, whether or not it was near Battle Abbey, in case another contemporary account said that the battlefield was somewhere other than Herste. We therefore believe that the battlefield was at Herste, but somewhere other than Battle Abbey.
If this is right, we can work out where the huts and battlefield marker should be. Whether the battlefield marker was a low stone wall is less certain - we have never seen ‘spinam’ translated to ‘low stone wall’ but it would be an odd mistake for Professor Searle to make. Her translation of ‘collis’ to ‘ridge’ is clearly influenced by her belief that the battlefield was at Battle Abbey. ‘Collis’ usually has the more general meaning ‘hill’. If, as we think, the battlefield was at Hurst spur and the battlefield marker is lower down on the western side of the hill, it should be near where Brede Lane crosses the stream that drains the western side of Hurst Lane.
Wace, Carmen, Jumièges and Orderic, in one way or another, say that the battle started at the third hour of the day. Given the time needed to don armour, assemble and receive a pep talk and commands, the battlefield cannot have been much more than a one-hour march from the Norman battle camp. Hurst spur is a one-hour march from Cottage Lane where we think the Normans had their battle camp.
Poitiers says that even late in the day after having suffered heavy casualties the English position was still “very difficult to surround". At Hurst spur, William would not have had enough men. The tightest encircling loop William could have formed - shown as the yellow loop on Figure 46 - would have been some 700m in diameter which would have put the English out of arrow range. Over 2km long, he did not have enough men to make it secure. And those to the east and north would have been in a wood where horses would have been a liability and the infantry would have been vulnerable to skirmishing attack from Harold’s newly arriving reinforcements.
A Hurst spur battlefield provides a possible answer another interesting question, which is the toponymy of Killingan Wood. This wood is some 200m west of Hurst Lane. It is a Saxon era name. ‘Kill’ looks like it comes from Old English ‘cwellen’, to kill. ‘ing’ and ‘an’ are Old English suffixes. We wonder whether some of the English troops took shelter in the wood as they fled from the battlefield, only to be get caught and slaughtered there.
One more. Battle is the only place on the Hastings Ridge that is visible from Hurst Lane, through a narrow gap between Petley Wood and what was known as Bodeherste Wood (now Great Wood). Perhaps this influenced the placing of the Abbey.
Finally, there is the Malfosse. Poitiers describes it as a “labyrinth of ditches”, which implies iron ore mining to us. Hurst Lane was only 200m from iron ore mines that have recently been found in Killingan Wood and less than 1km from Footlands which was one of the biggest sources of iron ore in Britain. Both of them would have been covered by a “labyrinth of iron ore mines”.
There are two outstanding puzzles.
The Normans sortied from their camp on Cottage Lane at dawn to attack the English camp on Great Sanders ridge. The English scouts saw them leave. Harold ordered his army to occupy Hurst spur, some 400m. The route was so narrow that they moved in an infantry column, then fanned out to form a wedge-shaped shield wall.
Why would the English leave their fortified ridge-top camp to occupy an unfortified position lower down. First, they were at the camp no more than two days, so the man-made fortifications could not have been great. Second, William had prepared to attack the camp, so his plans would be set awry. Third, the camp had defensive weaknesses with three shallow access points, two of which held the Roman road. Fourth, it was woody. Trees obscure the line of sight and hinder the ability to command. We think Harold reasoned that he stood a better chance of controlling the battle and keeping the shield wall intact at one of the Great Sanders ridge spurs.
The far west spur had the Roman road up its crest. The far east spur was shallow, wide and woody. The mid-west spur was too wide to defend. None of the others had hollows protecting the top slope. We think Harold judged his best chance of surviving one day was on the mid-east spur upon which Hurst Lane now runs. It had excellent line of sight with very few trees and it was narrow and boggy which would have provided protection against the Norman cavalry. The spur’s only weak point – the shallow southern end - could be adequately protected by a hastily made fosse.
The Normans had to stream slowly across a narrow Brede crossing. The Norman commanders were still at Cottage Lane when the English deployment became apparent. They changed their point of attack, branching off the Rochester Roman road along Long Lane, skirting around the stream head at the reservoir, then forming into three divisions roughly along modern Brede Lane. Then then slowly climbed the steep slope.
The engagement started with a crossbow attack, killing and injuring many English troops at the front of the shield wall. The English had no crossbowmen to return ammunition, so the crossbowmen retired. Then an archer attack, ineffective because arrows could not pierce the English shields. Then an infantry attack at the front of the shield wall, where the Normans were smashed by axemen in the second rank. Then a cavalry attack at the front of the shield wall, also repulsed by the barricade and fosse. Meanwhile the Normans tried infantry flank attacks which were smashed at the shield wall or shield charged into the flanking bogs. This might have taken an hour or so by which time William would have realised that the Normans stood no chance of victory unless the shield wall could be disrupted.
By tradition, many thousands of men were killed at the battlefield. Some of the contemporary accounts suggest tens of thousands. Yet the Normans buried their dead in one night. Wace says that no Norman barons died in the main battle. We guess that after the opening skirmishes there was very little fighting at the battlefield and very few new casualties. William’s cavalry did not participate again in the battle until the denouement. Harold had no archers to return ammunition, so the Norman archers and crossbowmen would have become impotent after the first few minutes too. In effect, for most of the day, William had 3,000 infantry fighting an uphill battle on hugely adverse terrain against 6,000 more appropriately equipped rivals. Harold presumably predicted this when he chose to defend Hurst spur.
Harold knew the English were safe as long as the shield wall held. William would have realised it too after the opening melee attacks. William’s instinct would have been to encircle the English army and starve them into submission. But even that was not feasible Hurst spur because William did not have enough men.
We suspect that the rest of the day involved a lot of shouting and goading but very little combat. There may well have been a lot of casualties at the start of the battle, but we guess there were only a few dozen more casualties at the main battlefield. The feigned retreat broke the shield wall. There may well have been many hundred casualties as the Norman cavalry harried the fleeing English troops.
Here is a list of battlefield clues, including many previously thought to be impossible or conflicting, with how they match Hurst spur.
Then there are eleven clues that suggest the English shield wall had tightly refused flanks, in a horse-shoe, wedge or enclosed loop deployment.
We are convinced that the battle was fought on the spur at Hurst Lane. We think it best fits the primary source accounts, the local geography, contemporary military tactics and common sense.
Just to re-iterate what we said long ago: the real battle narrative is as simple as it could possibly be (see below). The English camped on Great Sanders ridge, close to the Hastings Peninsula, yet at what Harold thought to be a safe distance from the Norman army. To his horror, he discovered the next morning that William had brough a significant cavalry (although he still did not realise how big) that put them in range of attack and prevented an organised withdrawal. The English army was effectively trapped. William spent that day scouting the English camp and devising a plan of attack. He sortied to attack the English camp at dawn the following day. Harold moved the English army to a nearby hill, which he thought stood a better chance of survival. Unlike other battle theories, this does not rely on idiotic battle tactics, suicidal surprise attacks or camping at ridiculously poor defensive positions. It just needs Harold to have a small but crucial intelligence oversight.
We have been lucky enough that the main land owners on both sides of Hurst Lane have allowed us to metal detect the prospective battlefield and dig up some finds. The eastern side of Hurst Lane has been intensively irrigated and farmed. There is a lot of stuff in the ground, but the signals from medieval artefacts are difficult to separate from 20th century farming activities. The western side is teeming with finds but a public footpath crosses the battlefield, so it has a lot of modern junk in the ground. We did find some lumps of distressed iron and lead which we sent to the Royal Armouries Museum in Leeds. They confirmed some of them are medieval, but cannot identify any of them as being military.
We are only amateurs. Hopefully, one day we will find professional metal detectorists and/or archaeologists who can do a better search.