return Return

Battle of Hastings Place Names


All the contemporary accounts name Hastingas (or somewhere that sounds like it) in the initial or secondary Norman landing. A few say that it was in Sussex. CBA alone hints at where in Sussex, but it is untrustworthy on this subject. Hæstingas also appears in some ASC entries and in several Saxon Charters without adding any details beyond that it was on the coast. We will try to work out where on the coast.

If it were not for some Norman invasion accounts, there would be little doubt that Hastingas referred to a county-like area on the south coast of England. Saxon Charter S318, dated 857, refers to a gift of coastal land in Hastingas, which only makes sense if it is an area. Malmesbury says that William’s other monastery was in Hastingas (“Hastingis”), which can only refer to an area. Huntingdon refers to ‘planis Hastinges’ (the plains of Hastinges), which only makes sense if it is an extensive area. The ASC entry for year 1011 says that Vikings overran the land south of the Thames, which it defines as Kent, Hæstingas, Sussex, Hampshire and Surrey. It implies that Hæstingas was a county-like area on the south coast between Sussex and Kent. It would have included the Hastings Peninsula, although it would not necessarily have been bounded by it.

Two related places are mentioned in the ASC before the invasion: ‘Hæstinga port’ and ‘Hæstinga ceastre’. ‘Ceastre’ was the Old English term for a Roman fortification, so these places sound like the major port in the area of Hæstingas and the major Roman fortification in the area of Hæstingas. Hæstinga ceastre is listed as one of Alfred’s burhs. These were the forerunners of modern boroughs. There were no organisational units between a borough and a county. If Hæstinga ceastre was inside Hæstingas, it also suggests that Hæstingas was a county-like area.

Yet Hastingas is never mentioned as a county in Domesday or elsewhere. So, what was it?

Briggs explains that the Old English suffixes -ing and -ingas mean ‘followers of’ or ‘dwellers in’, depending on whether the stem is a person’s name or a landscape feature. Thus, Hæstingas is thought to mean a place inhabited by followers of Hæst or Hæsta. There are many places in East Sussex with -ing suffixes that might previously have had -ingas names, including Guestling and Wilting on the Hastings Peninsula. None of them were bigger than hundreds. If Hæstingas referred to the Hastings Peninsula, it was different somehow.

We guess that Hastings Peninsula was like Dengie Peninsula and Tendring Peninsula in Essex, and the Isle of Wight, all of which took their names from early Anglo-Saxon tribes: the Dænningas, Tendringas and Wihtwara, respectively. They were all physically isolated. They were all small enough and passive enough to stay under the radar. Their occupants were among the earliest Anglo-Saxon settlers, which perhaps helped them retain racial integrity. The last three left a vestigial geographic term for the peninsula or island upon which they lived. We think the Hæstingas tribe left a vestigial geographic name for the Hastings Peninsula.

John Blair has a theory that might explain why they are different. In early Saxon times, he thinks that -ingas sometimes referred to sub-kingdoms or statelets, so called ‘regio’. The two he researched were ‘Woccingas’ and ‘Godhelmingas’, which became modern Woking and Godalming. There are others. The founding territories of Mercia and East Anglia were known as Iclingas and Wuffingas. The land occupied by the Dænningas was known as Deningei, meaning ‘the territory of the Dænningas people’. The land occupied by the Wihtwara was known as Wihtland, meaning ‘the state of the Wihtwara people’. Both sound like statelets. It makes no difference if we are wrong, but we guess that Tendringas and Hæstingas were also former statelets that gave their name to their ancestral homeland.

Most of the contemporary invasion accounts tell a different story. WP and WJ say that the Normans built a fortress at Hastingas, as if it were a settlement. CBA says that the Normans constructed a fortress at “a port named Hastinges”. Brevis Relatio says that William: "arrived with his whole army at another port nearby named Hastingas". CKE says that William: “built another monastery near Hastingis, dedicated to St. Martin”, which only makes sense if Hastingis referred to a settlement near Battle. Benoît says that after the battle: “William placed his best knights to guard the fortress at Hastinges”, which only makes sense if Hastinges is a settlement. Chronicon says that Harold: “… gave them battle at a place nine miles from Heastingam”, which clearly refers to a location rather than an area. Moreover, it must be a location at the eastern tip of the Hastings Peninsula – and therefore potentially a port - because nine Roman miles would otherwise take the battle out into the Andredsweald.

We note that all the accounts that suggest Hastingas was a settlement or a port were written by Normans or in Norman monasteries. Our interpretation is that Saxons used the term Hæstinga to mean the Hastings Peninsula whereas Normans used the term Hastinges – often Latinised as Hastingas - to mean Hæstinga port, at least until the castle was established. These names take their Old English, Old French or medieval Latin declensions.

Our interpretation is not unreasonable. Foreigners tend to think in terms of their own experience. So, most Europeans refer to ‘Manhattan’ as ‘New York’ because they never go anywhere else in New York. Normans traded with Hæstinga port. There was nothing else on the Hastings Peninsula that they would care about. It makes sense that they would drop the ‘port’ part of its name. It is the same as their use of the name ‘Douvres’ to mean the ‘port of Dover’, and indeed, ours of Rotterdam to mean the ‘port of Rotterdam’.

Domesday is an exception. It lists a place named Hastinges in the manor of Rameslie. A series of charters note that there was a port in Rameslie. It looks like Domesday’s Hastinges probably referred to the port, like the other Norman accounts, but it is listed with only 4 burgesses and 14 bordarers. It is too small to be the port, and there is no obvious reason the port would be broken out from the rest of Rameslie. Indeed, the port was one of Rameslie’s main attractions and an important reason the Abbey of Fécamp coveted it. We think it was broken out because it was an area within Rameslie manor from where the region was administered before the completion of the castle, and was therefore not charged to the Abbey of Fécamp. The most obvious candidate is the fortification of Hæstinga ceastre.

There are other problems. Some later Norman accounts refer to 'portus Hastingas’ or 'Hastinges portus’, which would be tautologies if Hastinges/Hastingas meant the port. Also, the castle was known as Hastinges in the 13th century and beyond. Some of the late 12th century accounts might have been referring to it rather than the port. It seems that Norman use of the name evolved through the 11th and 12th centuries.

Pre-invasion charters (S318 and S686) suggest that ex-pat Frankish monks of St Denis assimilated local place names, including Hastingas. There is no reason why ex-pat Norman monks of Fécamp would not do the same. We think that pre-invasion Normans in Normandy, including the authors of the early invasion accounts, used the name Hastinges to mean Hæstinga port. By Domesday, Normans in England seem to use Hastinges to mean Hæstinga ceastre. Then they completed the stone castle at modern Hastings. It was referred to in the 1182 Pipe Rolls as “castelli Noue Hasting”, presumably because Hastinges still referred to Hæstinga ceastre. The castle grew in importance at Hæstinga ceastre’s expense, dropping the Noue by the early 13th century. This is exactly what happened simultaneously at Windsor, whose castle was initially known as New Windsor.

Normans in Normandy probably ignored parochial events in obscure English outposts. We guess that their meaning for Hastinges switched from the port to the castle in the mid-12th century, forcing them to refer to the port as 'portus Hastingas’ or equivalent.

We feel confident that the early Norman references to Hastingas – WP, WJ and Carmen – referred to Hæstinga port. The late contemporary accounts – CBA and Wace – were reporting events that were originally recorded soon after the invasion. Hastinges in quotes probably means Hæstinga port, whereas in editorial it might mean the castle. Most of the other accounts were written by authors with English mothers and Norman fathers, so they might adopt the Norman or Anglo-Saxon convention. We will look at them case by case.

Wace was not Norman and his references to Hastingues are ambiguous. They could refer to an area or to the port. He is quite pedantic about ports. He uses the term ‘port de Lune’ for Bordeaux, ‘port de Saint-Morin’ for Morin, ‘port de Hantone’ for Southampton, etc. He uses Hastingues rather than ‘port de Hastingues’ which hints that he was referring to an area, but his references are quotes recorded soon after the invasion, so we think they referred to the port.

Six of the primary sources are Anglo-Norman, not necessarily following either convention. Orderic was English but lived and worked in Normandy. He refers to Hastingas as a seaport, following the Norman convention. John of Worcester was English but lived in a Norman monastery. All his references to Hastingas seem to refer to the port. Malmesbury and Huntingdon had Norman fathers and English mothers. They seem to switch between references to an area or settlement by convenience. CBA has two references to Hastinges. The first specifically says it was a port. We will return to the other in the section about Hæstinga port. The Tapestry is more complicated.

Figure 10 : Bayeux Tapestry Panel 40

The Tapestry was embroidered in England by Saxon nuns, but probably commissioned by a Norman aristocrat. Tapestry Panel 40 (Figure 10) is captioned: “ET HIC MILITES FESTINAVERUNT HESTINGA UT CIBUM RAPERENTUR”; [here the knights have hurried to Hestinga to seize food]. This Hestinga cannot mean the port because they needed to feed 10,000. They would not have wasted their time chasing a few goats and hens around a port. They needed to secure a month’s worth of food on the first day because, given half a chance, the locals would have driven away their livestock and burned their grain stores. We think they raided the biggest grain stores and rounded up sheep, cattle and poultry from the richest farmland.

Domesday lists resources by manor. The ten manors between the Brede and the Rother only had 35 acres of meadowland between them; barely enough to sustain the Norman army for a week. The four manors between the Hastings Ridge and the Brede estuary only had 6 acres of meadowland between them. We think the knights headed straight for Hooe, Filsham and Crowhurst, which had 116 acres of meadowland between them; enough food for a month. Therefore, we think that the Tapestry’s Hestinga referred to the Hastings Peninsula or, less likely, just to the part south and west of the Hastings Ridge.

Figure 11: Hastings and surrounding peninsulas in 1066

We mention earlier that the Anglo-Saxon area of Hæstingas might include land beyond the Hastings Peninsula. It cannot extend west of Pevensey Levels, because Pevensey was in Sussex. It cannot extend east of the Rother, which was in Kent. The most likely extensions are the adjacent peninsulas of Wartling to the west and Udimore to the north (Figure 11). Doubtless members of the Hæstingas tribe spread from the Hastings Peninsula and established communities outside. Perhaps the Hæstingas statelet included these extensions. But, by the time it only had a geographic meaning, we think Hæstingas was bounded by the Ash Bourne and Pevensey Lagoon to the west, and by the Brede estuary and the Andredsweald to the north.

Kathleen Tyson has a different theory about Hastingas. She thinks that it was the Frankish name for the Brede basin, which was bounded by the Udimore and Hastings ridges (see Figure 2). She reports it as fact, but then contradicts herself by saying that Hastingas was the cape between Winchelsea, Icklesham and Fairlight. We could not find her evidence for either argument. Both seem unlikely. The Brede basin and the Hastings Cape are too small to be the county-like place mentioned in ASC 1011 and too big to be a settlement or the port. We think our theory is more credible.

Pefenesea and Peuenesel

Pefenesea is universally understood to have been the Anglo-Saxon name for the place that eventually became modern Pevensey. Historians interpret the ASC, the Tapestry, Chronicon, Brevis Relatio and Benoît to be saying that the Normans landed there. We think that both notions are wrong.

First, we should explain the relationship between Pefenesea, pefenes ea, Peuenesel and Pevenesel. These four place names are really two: Pefenesea is the contraction of pefenes ea; ‘u’ and ‘v’ were sometimes interchangeable during medieval times as the Latin alphabet developed. The two pairs of names are more similar than it might seem because the closest Latin sound to the Old English ‘f’ was ‘u’ or ‘v’, so they were often substituted in place name transliterations.

Despite the different word ending, there is good reason to believe that Peuenesel (with ‘u’ or ‘v’) was another name for Pefenesea (one word or two). For example, three accounts of Odo’s rebellion - ASC, Chronicon and Symeon refer to Anderitum as the fortress of Pefenesea while a fourth, CKE, refers to it as the fortress of Peuenesel.

R G Roberts has a plausible explanation in his 1914 book ‘The Place-Names of Sussex’. In this vicinity, Old English ‘ea’ means island. Roberts thinks that the ‘el’ at the end of Peuenesel is the Frankish root for the modern French word île meaning ‘island’. The only other known Saxon, Latin or Norman place name ending ‘el’ (apart from places ending ‘dell’ which would not apply here) is ‘Wincenesel’, the Frankish and/or Norman name for Winchelse. The most plausible explanation is that Pefenesea and Winchelse were part of Bertoald’s 8th century gift (attested in Charter S133 and S318) to the Frankish Abbey of St Denys. If they owned Pefenesea, it makes perfect sense that they would transliterate Pefenes to Peuenes, and translate ‘ea’ to ‘el’, to make the name ‘Peuenes el’. If so, the Normans probably would adopt the name.

Back to the orthodox landing at Pevensey. We give a list of reasons in the introduction for why it would have been a bad idea for the Normans to land at or near modern Pevensey. And none of the primary sources actually say that they did. The Tapestry says that Duke William: “came to Pevenesæ”. ASC-D says that: “Earl William came from Normandy to pefnes ea”. Benoît says the Normans: “Arrived at Pevenesel”. Chronicon says that William: “moored his fleet at a place named Pefnesea". None of them mention a landing. Brevis Relatio is sometimes translated to say that the Normans landed at Pevenesel but it uses the Latin verb 'appello' which can mean 'landed’ or 'arrived’.

Two other sources corroborate Chronicon that the Normans moored off the English coast. Carmen says: “On the open sea you moor offshore”. WP says: “having reached shallow water off the English coast, William drops anchor to wait for the rest of the fleet to catch up”.

In summary, we believe that the Norman fleet headed for Pefenesea and moored nearby but did not land. Kathleen Tyson, who researched this independently, has found an ingenious ‘sea harbour’ interpretation of Carmen to explain how the Normans kept their fleet together during this time.

The evidence that Pefenesea referred to the place that became modern Pevensey is more difficult to dispute. As we just say, three accounts of Odo’s rebellion refer to the fortress of Pevensey as the fortress of Pefenesea. There is compelling evidence that modern Pevensey was known as Peuenesel from at least the 13th century. Saxon Charter S527, dated 963, gifts saltearns at ‘pefenes ea’ and land at ‘hanecan’ (later named ‘hacanan hamme' ) near a place named ‘glindlea’. Glindley and Hankham survive, not far from Pevensey. Birch therefore associated Hanecan with Hankham and pefenes ea with Pevensey.

Yet, we are as sure as we can be that 11th century Pefenesea was not the place that became modern Pevensey.

Domesday lists the manor of Peuenesel with 110 burgesses and a mill in 1086, whereas a comprehensive excavation by Dully in the 1960s found no evidence of pre 13th century civilian occupation at Pevensey. The Old English suffix 'ea’ always means 'island’ in this vicinity, whereas Pevensey was a peninsula connected to the mainland by a Roman road. The only pre invasion Saxon reference that almost certainly refers to Pevensey names it Andredesceastre, which is consistent with Saxon references to other Roman fortifications in England. If the Saxons knew the fortress as Andredesceastre and there was no civilian population, there is no reason why it would also be known as Pefenesea.

Pevensey’s founding Charter, issued in 1207, has the answer. It says: “… we have granted to the barons of Pevensel and confirmed by this our present charter that they may build a town on the headland between the port/harbour of Pevenesel and Langeney, which lies is within the liberties of the Cinque Ports, to keep and maintain according by which our subjects of the Cinque Ports possess.” In other words, the new town that became Pevensey was not the same as the place known as Pevenesel (aka Pefenesea) at the time of the invasion. The new town was between Langeney and Pevenesel. Langney survives to the southwest of modern Pevensey. Turning the description around, the pre-13th century place named Pevenesel must have been east of modern Pevensey.

Pefenesea is listed as Penress/Peneness in De Viis Maris, which says that it was eight miles west of ‘Hastinges castellum’ and two leagues east of the unreliable port of La Crumbie. La Crumbie is lost but must have been east of Eastbourne. Distances are rough and ready in De Viis Maris. It might have got the distance from Hastings wrong by a mile or two, but it is unlikely to have got the distance from La Crumbie wrong by more than 50%. Pefenesea must have been east of Pevensey but not as far as Southeye.

Kathleen Tyson has come up with two more clues. She resolved the name Pefenesea to mean ‘near-the-ness island’. At least two of the references to Pefenesea in the Anglo-Saxon Chronicle say it was used as a maritime refuge or harbour. Kathleen interprets one of them to mean that it was at a fordable break in a shingle spit.

One final clue is that Benoît says that the Normans: “Arrived at Pevenesel, at a harbour beneath a fortress handsome and strong”. We think there were a few former Roman fortresses dotted around the East Sussex, but the only one that might warrant the description ‘handsome and strong’ is Anderitum at Pevensey.

As far as we can judge, all the clues point to 11th century Pefenesea being an island between Langney and Southeye on the 'Crumbles' shingle spit between Bexhill and Eastbourne  (Figure 12). It would have been eight miles west of Hastings Castle, five miles north of the sheltered anchorage of Royal Sovereign Shoals, one mile south of the ness that was Pevensey peninsula, and three miles from Hankham and Glindley. It probably had a long spit that would make an ideal harbour and refuge. Its landward side would have been sheltered enough for salt production. It could easily have had a low-tide ford to Langney.

Figure 12: Pevensey Lagoon with the location of pefenes ea

Our proposed location for Pefenesea contradicts the orthodoxy. The biggest source of confusion is that the inhabitants of Pefenesea on The Crumbles moved to modern Pevensey in the 13th century, taking the name with them. This is exactly analogous with what happened to Old Winchelsea and Old Romney. The 11th century Pefenesea mentioned in the invasion accounts was abandoned. All evidence of it has probably been washed away. Pefenesea on The Crumbles is only a mile further from Glindley and Hankham, still close enough to comply with S527. We guess that Pevensey castle was known as the fortress of Pefenesea/Peuenesel because that was the nearest major settlement. This is analogous to Windsor castle taking the name of a settlement four miles downstream because it was the nearest major settlement. And like Windsor Castle, the town eventually moved to the castle.

Kathleen Tyson has looked at the same clues, yet concluded that Pefenesea is more likely to have been an island on the Camber shingle spit near to Lydd. It is plausible. It does match most of the clues. But it conflicts with the important and specific clue in De Viis Maris that Pefenesea was eight Roman miles west of Hastings Castle. It would be a peculiar interpretation of Pevensey’s foundation Charter, equivalent to describing Devon’s location as between Cornwall and East Anglia; not wrong but unhelpful. It would be a similarly peculiar interpretation of S527. It makes Benoît’s description of ‘arriving below a fortress handsome and strong’ look wrong. On balance, we remain confident in our conclusion.

We should explain our thinking with the rest of Figure 12. There must have been at least one break in The Crumbles, to allow fresh water to drain into the sea. Pevensey Haven (aka Hurst Haven) drained the land to the north and west. Waller’s Haven drained the land to the east. Waller’s Haven was redirected in the 16th century. Its original route can be followed along ‘Old Haven’ on OS maps. It makes an inland confluence with Hurst Haven several hundred metres to the east of Pevensey. It had to drain into the sea somewhere to the southeast of that. It is common for silty estuaries to form spits at their mouth. We think pefenes ea had just such a spit tail, somewhere around modern Beachlands.

A Pefenesea island at Beachlands does not contradict the traditional evidence. It would have been the biggest and most important settlement in the region in Saxon and early Norman times. It makes sense that the ‘rap’ was named after it. Anderitum was in that rape, so it would be referred to as castrum Peuenesel in A.4171 and in CKE’s Odo rebellion account. It would have had a sheltered tidal shore, so it might well have had saltearns to fit S527.

We discuss the international port in the next section. It is just worth noting here that Pefenesea on The Crumbles was not it. A port has a wharf to load/unload cargo, whereas a harbour does not. Pefenesea had no natural resources, no hinterland, no roads to distribute goods, and hardly anyone to distribute them to. It was, as ASC suggests, a sheltered harbour rather than a port. We say earlier that there was no port or settlement at Pevensey at the time of the invasion. It is accurate. Pefenesea on The Crumbles was not at Pevensey and it was not a port.

This sheds a new light on S133, the Saxon Charter dated 790 which gifts land in East Sussex and elsewhere to the Frankish Abbey of St Denys. The gift included a port: “de portu super mare, Hastingas et Peuenisel" [the coastal port of Hastingas and Peuenisel]. Note ‘port’ singular. It is ambiguous. It could be trying to say: “the coastal port of Hastingas et Peuenisel”, with ‘Hastingas et Peuenisel’ as a noun, or “the coastal port of Hastingas, and Peuenisel”, with Peuenisel somewhere other than the port. In the first edition of this book we speculated that that the former was more likely, because we thought that it could be synonymous with the port mentioned by Orderic as Hastingas et Peneuesellum. We have subsequently refined our understanding of Peneuesellum – see below - which makes this unlikely. We now think that the latter is more probable. The attestation in S318 suggests so, in that it refers to the gift as land in/at Hastingas and land in/at Peuenisel.

One source of confusion is that ASC-D, the Tapestry and Chronicon immediately pass from the arrival at Pefenesea to the construction of a fortress or, in the case of Chronicon, to the battle. Historians read into this an implication that there was a landing at Pefenesea. We think the journey to the landing place and the landing itself were redacted. All three of these accounts are heavily abridged, covering the invasion in a few paragraphs. They had to redact uneventful details like the journey to the landing site and the landing.

There are a few puzzles about the Norman Channel crossing and about Pefenesea that we should try to answer.

Why did the Normans moor near Pefenesea if their ultimate destination was the Hastings peninsula? We guess that it was standard practice for Norman trading ships to moor off Pefenesea before docking at Hæstinga port because the Sussex coast was a treacherous place to sail in boats with no centreboard. WP explains that William wanted to avoid sailing in dangerous or unknown waters. We think they moored near the Royal Sovereign Shoals, several miles off The Crumbles. From there they could safely run or reach into the port on any wind, bar a north-easterly which is rare. Wace says that the fleet steered towards a port/harbour, which was presumably well known to his sailors and navigators. We think that that harbour was Pefenesea and that they used Beachy Head as a navigation beacon.

Why did William delay the invasion – as Wace and WP say – to wait for a southerly breeze? We think his main reason was the weatherliness of his ships. Viking style longships had no centreboard. Snekka were long and narrow, which made them fairly weatherly. Karvi, which carried the horses and cargo, were not. They slipped horribly when reaching and were inefficient beating. Moreover, the ships would sail at different speeds, depending on their style, size and load. If they had to beat to hold their line on a prevailing south-westerly breeze, it would have been impossible to prevent the fleet getting separated, there would have been a lot of collisions and there would be a danger of the Karvi slipping east of their target. William’s only safe option was to run downwind on a southerly breeze. We suspect that a secondary reason is that, having left the sea, they headed west up an estuary and William wanted to sail rather than row, which would not be possible on the prevailing south-westerly breeze.

Why did the 13th century barons of Peuenesel want to create a new borough at modern Pevensey? According to the founding Charter, the new borough was a market town and Cinque Port equivalent. The implication is that Pefenesea harbour was silting up and/or that the local Normans barons wanted to drain the marshes and encourage settlement in the area. To do that they needed a full-service port with access to Anderitum’s Roman road.

How did Anderitum and Peuenesel Rape end up being named after Peuenesel? Domesday’s Peuenesel borough was the only significant settlement between the Hastings peninsula and the Cuckmere River. All the Sussex Rapes took the name of their main settlement. Most of the contemporary Norman accounts and post-Conquest writs refer to Anderitum as ‘castrum Peuenesel’. In other words, it was the major fortification in the region of Peuenesel. At some point before the 14th century, Pefenesea harbour was closed due to storms and longshore drift, then The Crumbles washed away. The new borough around Anderitum would have been left as the major settlement in the Rape and the only settlement associated with the name Peuenesel.

In the first edition of this book, and the blogs upon which it was based, we speculated that Peuenisel with an ‘i’ – as mentioned in S133 - might have referred to somewhere other than Peuenesel with an ‘e’. Refinements to our understanding of the name Peneuesellum have persuaded us otherwise. We now think they both referred to pefenes ea.

Rameslie manor

Rameslie was a big and wealthy manor in Guestlinges hundred, which it shared with the manors of Guestlinges and Ivet. Manors in the same hundred were not always contiguous, but most are, especially as in this case, when one is dominant over the others. Guestling survives as a settlement south of the fluvial part of the River Panel. Ivet was very small. There are no other hundreds in the vicinity. Therefore, it is safe to assume that Rameslie filled the gap between the Panel and the Brede, an area we refer to as the Winchelsea peninsula.

Domesday lists Rameslie manor with 100 saltpans, 35 ploughlands, 7 acres of meadowland, 2 woodland swine renders and 5 churches. The Winchelsea peninsula was nowhere near big enough to hold it all. It must have had a lot of other land, including perhaps four other significant settlements.

King Cnut married Emma of Normandy in 1017. As a dowry, presumably at her instigation, he gifted the manor of Rammesleah in Sussex to the Norman Abbey of Fécamp. Rammesleah looks like an alternative spelling of Rameslie, and Rameslie manor was held by the Abbey of Fécamp in Domesday. We will assume they were one and the same.

Cnut’s gift is described in a Charter (S949). It says that Rameslie had an unnamed port. Historians assumed that this port was Hæstinga port and that it was below modern Hastings. Therefore, by tradition, Rameslie manor sprawled along the coast from Winchelsea to Hastings and perhaps beyond. We think it unlikely. We explain in the next section why there is no likelihood of a significant Saxon era port anywhere within five miles of modern Hastings. This removes the only justification for thinking Rameslie extended to modern Hastings. Moreover, Dawson and Taylor have revised Ivet’s probable location from Lidham to modern Pett, in which case Rameslie could not have extended south of the Pannel.

If Rameslie did not stretch south along the coast, it must have stretched west and/or north. Matthew concludes that Rameslie spanned the Brede. We think he is right. Medieval salt evaporation ponds in this region average 35m across. They are best placed on the north strand of a wide east-west estuary, where they get reflected sunlight, no riverbank shade and have no need for deforestation to prevent tree shade. Better on an east flowing river, where they are protected from storm surges or bores that might flood the saltpans. Most estuaries on the south coast are south flowing. The east flowing Brede was the only east-west estuary on the south coast that was long enough to hold 100 saltpans. If these saltpans were predominantly on the north bank, Rameslie must have stretched at least as far west as Brede Place on the north bank of the Brede.

This is still nowhere near big enough to have five churches. Iham (which became modern Winchelsea) had one. The north bank of the Brede east of Brede Place is unlikely to have had more than one. That leaves at least two settlements to find.

One must have been at the port. S949 says that Rameslie had a port. Winchelse, the busiest port between Dover and Southampton, was on the Camber shingle spit which stretched 20 miles northeast from Pett (Figure 4). S982 attests Cnut’s gift to the Abbey of Fécamp, then immediately authorises them to raise ‘two parts’ of the tolls at ‘Wincenesel’. Wincenesel looks like a Frankish translation of Winchelse and the Abbey of Fécamp built St Giles church there. It was exactly the sort of infrastructure asset that they coveted. We have no doubt that the Camber spit was part of Rameslie and that St Giles accounted for another church. St Thomas, which moved to modern Winchelsea, was there too.

One church to find. S982 confirms that the manor of Bretda was included in Cnut’s gift to the Abbey of Fécamp. Manors that are worth coveting usually have a church. Bretda is never mentioned again, so presumably it was absorbed into Rameslie. The fact that it had to be confirmed means that it was not specified in the earlier S949 Charter. This means that Bretda did not incorporate the port or the saltpans. Its name makes it sound like it was beside the Brede, in which case it was west of Brede Place on the north bank, and/or west of Guestling on the south bank.

Bretda’s location could be narrowed down if there were some other Brede side manors. The only other Domesday manors in the vicinity were Sedlescombe and Dodimere. Sedlescombe sounds like it would have had estuary frontage, but it was a mile upstream in the 11th century (see Figure 33). That leaves Dodimere.

By tradition, Dodimere manor surrounded the settlement of Udimore on the Udimore ridge. East Sussex HER says that Dodimere was a dispersed ridgetop hamlet on the Udimore Ridge. This seems unlikely because ‘mere’ is the Old English term for a body of water, which would not apply to a ridgetop settlement. The manor is not listed with any saltpans, which implies it did not have Brede estuary frontage. It was in Babinrerode hundred, whose only other manor was tiny Kitchenham (2 households) on the Rother. If Dodimere was on the Udimore Peninsula, Goldspur hundred would have separated it from Kitchenham. Divided hundreds are not uncommon, but it would be very odd for one that only has two manors when the other is tiny. Something must be wrong.

Dodimere manor is associated with Udimore because Robert Count de Eu was Lord of the manor and Dodimere sounds like Udimore which was named after him. But he was Lord or Tenant-In-Chief of over 100 East Sussex manors any of which might have been named after him. We suspect that Dodimere and Udimore were different places that were independently named after him, and that Dodimere manor spanned the Rother Peninsula north from Beckley Furnace.

If we are right, Rameslie manor lined the entire northern bank of the Brede estuary, as far west as the tidal limit at modern Sedlescombe. What then of the south bank?

Beauport Park was once the third biggest source of iron in the Roman empire. IHRG recently found a metalled Roman road from Beauport Park to the Rochester Roman road, with a junction at Pestalozzi Children’s Village on the south bank of the Brede near Sedlescombe. Another major Romano-British iron bloomery was nearby at Oaklands, also on the south bank of the Brede. Kathleen Tyson has found evidence there was once a low-tide canal from Sedlescombe to Winchelsea, which presumably had barges that took iron blooms from Beauport Park, Oaklands, Footlands and Chitcombe to the port at Winchelse. Iron blooms from Footlands and Chitcombe would have been brought by cart to the Brede on the Rochester Roman road.

We will shortly explain why we think some of these iron workings were still active in the 11th century. If so, they would have needed jetties, wharfs, barges, warehouses, ferries, road maintenance, a river crossing and, perhaps, a low-tide canal. This stuff does not come cheap and there are no surviving Charters to make anyone responsible for them. It was a capital intensive infrastructure business before there was an easy way to raise capital. The only people wealthy enough to operate these services were monasteries. In our opinion, the Abbey of Fécamp provided all these infrastructure services as part of what was effectively an entrepôt paid for by traffic tolls.

A lot of the Brede’s infrastructure services would have been on the south bank and up near the head of tide. Given that there were no other manors in the vicinity, we believe that Rameslie must therefore have stretched all the way along the south bank to modern Sedlescombe.

Figure 13: Brede side manors

If we and Taylor are right, Rameslie manor lined both banks of the Brede up to Sedlescombe (Figure 13). It controlled the saltpans, the access roads to Beauport Park, Chitcombe, Footlands and Oaklands. It controlled the Rochester Roman road along which all road traffic would pass, as well as the iron products coming out of Footlands and Chitcombe. It was bounded to the south by Westfield, Guestling and Ivet (aka Luet), all of which would have had access from the Hastings Ridge ridgeway.

Cooper must have gone through similar reasoning 170 years ago, because he worked out – without saying how – similar locations for Rameslie’s five churches. He says that one was at Brede village, one in Rye, two in Winchelse (St Thomas and St Giles), and one at Winchelsea (St Leonards). We think he is right, other than that his church at Rye (see Rye below) is more likely to have been at Cadborough in 1066.

We were unsure where Bretda manor might have been on the Brede estuary. Kathleen Tyson told us that she thinks it was on the north bank. It could not have been east of Brede village because the saltpans were included in the original Rameslie gift to the Abbey of Fécamp. We have no evidence to the contrary, so we depict it west of Brede village.


Before discussing Hæstinga port, it important to explain about Hechelande, the only specific clue to its location. CBA has five references to Hechelande, variously spelled Hecilande and Hechilande. One explains that it is shy 1½ miles southeast of Battle Abbey, between Bodeherste and Croherste. This places it adjacent to the Hastings Ridge, near to Telham. Two more references are consistent with the first. A third says that it is a wood. The other that it is a hill.

There are no conical hills northwest of Telham and the crest of the Hastings Ridge would probably have been unwooded. We interpret the clues to mean that CBA is describing a manor on a west spur of the Hastings Ridge that contained an eponymous wood. It would be on the land now occupied by Loose Farm. That land contains Bushy Wood. Professor Searle translates Hechelande as Hedgland. Might be a link between Bushy and Hedgey?

The CBA reference about the hill has the key Hæstinga port location clue. Professor Searle translates it to be saying that Hechelande ‘lies towards Hastingarum’ (Hastingarum being a Latin declension of Hastinges). CBA had previously said that Hastinges was a port. A line from Battle Abbey through Loose Farm hits the coast somewhere near Hastings Castle. It is as clear as any clue in any of the contemporary accounts, yet we think it is misleading.

For one thing, this part of CBA cannot be trusted. It was written to defend Battle Abbey’s wealth and independence. Its case was based on their claim – spurious, in our opinion - that the Abbey was built on the battlefield. CBA had previously said that the battlefield was at a place named Herste and that the Norman battle camp was at Hechelande. As we explain in the Traditional Battlefield section, we think these were indeed the manors containing the battlefield and the Norman camp, but they were some distance from the Abbey. If references to those names appeared in other Conquest accounts, it would undermine the Abbey’s claim to be on the battlefield. We think the Abbot turned this risk into a potential endorsement by renaming two places near the Abbey as Herste and Hechelande. One reason to think this is that CBA’s description of its Leuga is a list of ten places, all of which are outside the boundary apart from Hechelande. It would have been more natural to have named Telleham instead. It looks like they are drawing attention to Hechelande, perhaps because there was another Hechelande elsewhere.

For another thing, we are unconvinced by the translations. CBA says that Hechelande was ‘a parte Hastingarum’. Searle translates to ‘lies towards Hastingarum’, Lower to ‘in the direction of Hastingarum’. Both are valid, but uncommon. CBA has other passages that describe the ‘direction of a place’ and ‘towards a place’ that do not use ‘a parte’. On the other hand, it has 30 or more uses of ‘parte’ where it means ‘side’, as in the ‘south side’, the ‘opposite side’, the ‘side of the church’, and so on. We think the most natural translation of ‘a parte Hastingarum’ is ‘to the side of Hastingarum’. CBA’s Hastinges was a port, so if Hechelande was to its side, it could not be near Telham or anywhere else inland.

Even if CBA did mean that Hechelande was in the direction of Hastinges, it might be an anachronism. As we explain above, we think that the Normans referred to Hæstinga port as Hastinges at the time of the invasion but that they referred to Hastings Castle as Hastinges by the time CBA was written. CBA’s description of Hechelande’s location might therefore accurately refer to a newly created Hechelande at Loose Farm being in the direction of Hastings Castle, without saying it was there at the time of the invasion.

In our opinion, nothing can be assumed from CBA about Hechelande’s location at the time of the invasion. It suits our argument if it is saying that Hechelande is to the side of Hæstinga port, but this is just an interpretation.  

Hæstinga port

Locating Hæstinga port is important. One primary source says it was where the Normans landed, three say it was where they camped. Five more say that the Normans landed and/or camped at Hastinges or Hastingas, which we believe to be the Norman and Norman Latin names for Hæstinga port.

By tradition, Hæstinga port was in the Bourne and/or Priory valleys below modern Hastings. This seems to be what CBA is saying, as we describe in the Hechelande section above. Pre-invasion Normans referred to Hæstinga port as Hastinges, which sounds like it might have given its name to Hastinges castle and thence to modern Hastings. If this is right, Hæstinga port was below modern Hastings in the Priory and/or Bourne valleys.

We think that this orthodox argument is flawed. CBA’s locational description for Hechelande and the port of Hastinges cannot be trusted, for reasons we explain above. Hastinges castle could have taken its name from the port of Hastinges – i.e., Hæstinga port - even if it was several miles away. Or Normans, having assimilated after living in England for 50 years, might have named their castle after Hæstingas, the Old English name for the Hastings Peninsula.

There are plenty of reasons to think that Hæstinga port was not in the Bourne and/or Priory valleys. Serial excavations have unearthed no archaeological evidence of Saxon era occupation. Access and egress would be dangerous, with treacherous sea cliffs on both sides of the estuary mouth. There was no local population centre to draw imports, negligible natural resources in the basin to export, and no known access roads to get freight up or down the precipitous valley walls. Hæstinga port is the location of the Norman landing according to several contemporary accounts but they are unlikely to have landed in the Bourne and/or Priory valleys for a host of reasons we list earlier. And De Viis Maris specifically says that Hastings did not have a port.

If Hastings is excluded, there are four Hæstinga port candidates, namely Monkham Wood and Bulverhythe in Combe Haven, Northeye at the mouth of Hooe Haven and the Ash Bourne estuary, and Winchelse at the mouth of the Brede estuary.

De Viis Maris lists the ports, even the bad ones, from which crusaders might leave for the Continent. Between Folkstone and Beachy Head it has entries for Lympne, Romney, Hythe, Winchelse, Penress/Peneness (pefenes ea) and La Crumbie (probably Hydney, now in Eastbourne), from which the Crumbles gets its name. It would be strange if one of these was not Hæstinga port, the only international port between Dover and Southampton. No ports are listed between Winchelse and pefenes ea. This gap includes Redgeland Wood, Bulverhythe and Northeye, implying they did not have international ports in the 12th century. They cannot have silted up during the intervening 130 years because they were active ports in the 13th century. De Viis Maris therefore implies that Hæstinga port was at Winchelse.

The Hastings Peninsula and its surrounds were too small, too sparsely populated and too short of a hinterland to have had a major international port that did not ship a significant volume of natural resources. If Hæstinga port was the only significant international port in the region, it must have shipped most of its natural resources. This suggests two more methods to corroborate De Viis Maris: 1) From references to the region’s major port; and 2) From proximity to the major natural resource production centres.

Domesday says that Rameslie manor in the Brede basin had 100 saltpans; the greatest concentration in the south of England. The rest of the Hastings Peninsula combined had 35. It must have been shipped from, or been used at, the port of Winchelse at the mouth of the Brede.

S982 authorises the Abbey of Fécamp to take two-thirds of the tolls from Winchelse. There would be no point in making this provision if the tolls were not substantial. Yet the Abbey of Fécamp would not be taxing their own salt. Something valuable other than salt must have been shipped out of Winchelse. There were not enough people to make anything valuable. It must have been some sort of natural resource.

Fish is one candidate. The Isle of Wight had huge herring shoals. The catch would have been gutted and salted at sea but would have been landed for packing and onward delivery. The logical process would be for the fishing boats to land their ready-salted catch and take on new salt at the port closest to where it was produced, most likely then at Winchelse.

Timber is another possibility. The south coast of England and most of its estuaries were lined by woodland in Saxon times. But uncut timber would have been almost impossible to move on the rutted tracks that were typical at the time. The Brede estuary, uniquely for the region, was lined by steep slopes on both banks. Timber could have been slid down the water on log chutes. Local historian Mark Freeman has found what looks like a medieval log chute in Steephill Wood. If timber was a major export commodity in the region, nearly all of it would have shipped from Winchelse.

In Roman times, the region’s most valuable resource was iron. Cleere reckons that the Brede basin bloomeries at Beauport Park, Chitcombe, Footlands and Oaklands accounted for 80% of the region’s iron output. He says that there is no evidence of Wealden iron smelting during Saxon times. Robert Turgoose of the Wealden Iron Research Group agrees. But there is hardly any evidence of Saxon era iron production anywhere in Britain, even though there was huge demand. Thomas Birch notes that some finished iron products were imported, and some iron was recycled. The rest must have come from domestic iron ore deposits. Turgoose told us that evidence of bloomery dates is rare and indirect, through pottery or coins found in the slag heaps. We guess that there must have been major domestic iron production, albeit too small to leave pottery or coins, and the most likely source is the former Roman bloomeries in the Brede basin.

In summary, the Brede basin produced 70% of the region’s salt, and would have had a similar proportion of any fish being packaged and shipped. It would have had close to 100% of any iron and/or timber being produced for export. All of it would have been shipped from Winchelse.

The only unambiguous evidence soon after the Conquest is in Domesday which shows that Rameslie manor was far more populous with 189 households than the manors around possible ports at Redgeland Wood or Northeye. Indeed, Wilting manor around Redgeland Wood is listed with only 14 households. Hooe is listed with 73 households but most of them would have been occupied on its huge farmland.

Based on proximity to natural resource production centres, Domesday’s population figures and De Viis Maris, the port at Winchelse was at least an order of magnitude bigger than any other port in the region. The Pipe Rolls of 1204 records that Winchelse was the biggest port between London and Southampton, and the third biggest on the south coast. The relative size of the south coast ports can be corroborated from ‘Ship Service’ records, but they are an interminable source of confusion, so we will try to explain.

Ship Service refers to a deal whereby the King could requisition ships and crews from local fleets in exchange for liberties; the more valuable the liberties, the more ships. It was established by Edward the Confessor and reinstated by the Plantagenets. A 1227 Charter reproduced by Jeake defines Hastyng, Doverr, Romone, Hethe and Sandwich – Hastings, Dover, Romney, Hithe and Sandwich, the original Cinque Ports - as ‘Head Ports’, charged with getting their apportionment of ships from ‘member’ towns in their vicinity. They were not chosen because of their port but because they were the administrative hub for their section of the coast. The Ship Service is really saying that the Count de Hastinges, for instance, has responsibility to supply ships from the coastal manors around him. It does not necessarily mean that Hastings provided any ships or, indeed, that it had a port.

Jeake’s Charter demands 57 ships, listed as 21 from Hastings, 10 from Winchelse, 5 from Rye, 5 from Romney, 5 from Hithe, 21 from Dover, 5 from Sandwich. Lots of historians have looked at these figures and inferred that the port of Hastings was more than double the size of Winchelse and, crucially, that it was somewhere other than at Winchelse. They are not mathematicians. These apportionments add up to 72.

Jeake explains that the sums only work if the 'Ancient Towns’ - i.e. Winchelse and Rye - are included in Hasting’s 21, and they are described as ‘members’ rather than Head Ports. The 57 then, are 21 from Dover, 21 from Hastings, 5 from Romney, 5 from Hythe and 5 from Sandwich. Within Hastings’s 21, there were 10 from Winchelse, 5 from Rye and 6 from the other ports which are listed as Seaford, Pevensey, Hydney, Northeye, Bulverhythe, Iham, Beaksborne, Grench and, perhaps, Hastings. Iham was the old name for modern Winchelsea, so the Brede estuary (Winchelse, Rye and Iham) provided more than 15 ships. On average, each of the other Hæstinga port candidates provided less than one. Exactly as expected, the combined ports at the mouth of the Brede were more than ten times bigger than any other Hæstinga port candidate.

It does not necessarily follow that Winchelse must have been the region’s major port at the time of the invasion, just because it was a hundred years or so later, but nothing significant had changed. Doubtless England’s new masters required enormously more wine imports but Winchelse was an export hub. Sylvester reports that at the turn of the 14th century Winchelsea exported 15 times as much as it imported, and that is after the huge increase in wine imports. Iron, salt and timber production techniques did not change significantly through the dark ages. If Winchelse was the region’s dominant port when Domesday was collated, and at the time of the second crusade, and in 1204 and 1227 and later, we are convinced it would have been the region’s dominant port at the time of the invasion, and therefore the most likely place to have been Hæstinga port.

Hæstinga port is the only coastal port in the ASC or Saxon Charters with a ‘port’ suffix. The other coastal ports – Sup-hanton [Southampton], for instance – were just known by their settlement’s name. Hæstinga port would have needed dozens of stevedores, ship chandlers, sailors, barmen, prostitutes and others. It must have been at a named settlement but seems to have a different toponomy. Assuming it is not just a naming quirk, the most rational explanation is that Hæstinga port was an entrepôt covering two or more named settlements. This would uniquely match Winchelse, Rye and Iham, all within a one-mile radius at the mouth of the Brede.

It is not unreasonable that Hæstinga port was an entrepôt, even though it was 800 years before the concept existed, because Winchelse left a lot to be desired as a port. It was on a shingle spit that would have had no running water apart from rain, no fresh food apart from fish and chickens, no fuel, no road access and it would have been horribly flood prone. It was presumably adequate for Hæstinga port’s docks and warehouses, but we guess that the administrators and professionals – running the mint, port administration, dry docks, ship repairs, etc - were somewhere more comfortable, presumably modern Winchelsea (then known as Iham). This would explain why Hæstinga port had the look of a modern entrepôt and why, unlike all the other ports in the country, its name did not refer to a specific settlement.

Even though there is no mention of a port at Hastings in the Pipe Rolls or Ship Service records, Hastings did have a port. It is listed in half a dozen Norman Charters as Portus de Hasting. It has to be at Hastings because Winchelse, Rye, Bulverhythe and all the other ports in the region appear on lists alongside Portus de Hasting. Presumably, it was in the Priory valley, as it is often depicted, beside the priory. It never amounted to much. Even when it was servicing Hastings Castle in the 14th century, Sylvester reckons Portus de Hasting only had 15 ships, compared to Winchelsea’s 115.

Hæstinga ceastre

Place names ending -ceastre were fortified Roman settlements. The earliest reference to Hæstinga ceastre is as one of Alfred’s 33 'burhs’, which were fortresses to give early warning of Viking raids. Alfred liked to build them at former Roman fortifications. The next reference was in 928 to license a mint. There could hardly be a better place to house Hæstinga port’s mint, toll house and garrison than inside the walls of Hæstinga ceastre.

Chronicon repeats ASC entries about Hæstinga ceastre and Hæstinga port, in which it refers to both as Heastinga. Some coins from the Hæstinga ceastre mint were stamped 'Hestingport’. Presumably, the mint restamped foreign coin and bullion taken as payment for iron and salt from Hæstinga port customers. The ASC 1050 entry says that boats and boatmen from Hæstinga ceastre defeated two of Sweyn’s ships on behalf of the King. All of this suggests that the ceastre and port were either adjacent or coterminous.

Alfred liked to build his coastal burhs on promontories overlooking places that the Vikings had previously raided. The Romans liked to build their fortifications on promontories too. The only promontory adjacent to Winchelse was Iham (modern Winchelsea). If Hæstinga port was at Winchelse, Hæstinga ceastre was probably on the summit of Winchelsea. Hæstinga ceastres wall, calculated from the Burghal Hidage, would have been some 760m long, spreading 190m north and west from the junction of Back Lane and St Thomas Street. This would explain why Winchelsea’s first church, St Leonard’s, was consigned to the remote northwest corner of the town rather than holding the naturally prominent position on the summit.

Winchelsea would have been a typical location for a Roman fortress, on a promontory where it overlooked and protected the Brede basin’s port and natural resources. It would have been a typical location for an Alfredrian burh too, on a defensively sound bluff at the only place on the Hastings Peninsula that had an uninterrupted view of Old Romney which was a previous Viking incursion point. It was also the only place north of the Hastings Ridge that had a view of the Fire Hills, which from its name was presumably the location of a medieval messaging beacon.

Martin White made a well-reasoned submission to the Bexhill Bypass commission that Hæstinga ceastre was at Wilting in Combe Haven. This is the site proposed by Nick Austin for the second Norman camp. White uses much the same Hæstinga port juxtaposition arguments as us, only with Hæstinga port being in Combe Haven rather than the Brede estuary. He has three additional items of evidence. One is the impression of a Roman enclosure he has found on a LIDAR scan of Wilting which seems to comform with Hæstinga ceastre's listing in the Burghal Hidage. Second, he believes that nearby Silverhill might have taken its name from the Hæstinga ceastre mint. Third, he has found a nearby area of land named 'Burghs' in the 1847 Hollington tithe map.

The first known reference to Silverhill was as Salver Hill in the 18th century Yeakell & Gardiner map. We suspect its name was changed by an enterprising Georgian estate agent. It is only to be expected that the Romans would build an enclosure to protect their port at Monkham Wood. They did the same for most of their ports. Bulverhythe and Filsham nearby became a moderately important port after the Conquest, not least because they serviced the new Norman castle at modern Hastings. Like many international ports it would have received royal liberties and would therefore be known as a burgh. Its market would naturally have been on the high ground at the top of Gillmans Hill, on the land known in the 19th century as the Burghs. But this does not mean it was a Saxon burh at the time of the invasion and it is 1500m from the proposed Hæstinga ceastre burh location. White's additional evidence is credible but, in our opinion, not compelling and nowhere near enough to overcome our scepticism that Hæstinga port was in Combe Haven. 

Keith Foord thinks that Hæstinga ceastre was adjacent to Hæstinga port on the Camber shingle spit. This would have been the best place to defend the port, but it seems implausible to us. We cannot believe that the Romans would fortify anywhere on a shifting shingle bank. If they did, it is unlikely to have survived 500 years until Alfred needed somewhere for his burh. Even if it survived, Alfred would not have built his burh at sea level because it was supposed to be a lookout. And Tapestry Panel 45 shows William’s camp at ‘Hestenga ceastra’ at a hill, so it could not have been on a shingle spit.

Kathleen Tyson thinks that Hæstinga ceastre was at Icklesham, a few miles upstream of Winchelsea. She notes that it had a Roman bloomery, so the Romans were there. She has a reasonable argument, but we think that there is a better argument for Winchelsea, which was closer to the sea and with a wider field of vision that included the important Romney marsh access point at Old Romney.

Winchelse (aka Old Winchelsea)

Winchelsea’s founding Charter and De Viis Maris and S982 say that Winchelse was a port on the Camber shingle spit. The Camber would have looked rather like the spit part of Pagham Rife, just along the coast. The spit extended 20 miles northeast from Pett (see Figure 4 and Figure 15) in medieval times. No trace of it has ever been found, but Cooper lists some reasonably specific coordinates. He says that it was roughly 6 miles NE of Fairlight Cliff, 3 miles ESE of modern Winchelsea, 2 miles SSE from Rye and 7 miles SW of Old Romney. Jeake says that it was more than a mile east of modern Winchelsea (see Figure 14).

Figure 14: Cooper’s coordinates for Winchelse

Whole number of miles and 1/16th compass points leaves plenty of room for interpretation. Dugdale reckons Winchelse was somewhere under Rye Harbour Nature Reserve; Cooper reckons to the east side of the east pier head, which would put it under Rye Golf Course. The distances intersect near Rye harbour; the directions intersect 2km southwest near Watch House. At least one of the directions or distances is rogue. The only way to fix a location to within 11° and 0.5 miles of all Coopers clues with one change is to revise the distance from Winchelsea to 2 miles. This would place Winchelse at 50.92112, 0.74778, roughly 1.75 miles east of St Thomas’s, Winchelsea.

Winchelse was a thriving place with 300 homes - and its own hospital, two churches and a friary - when it succumbed to the sea in the 13th century. It was not that big in 1086 when the whole of Rameslie had 189 households. Allowing for the farm and salt workers, we guess it had somewhere between 100 and 150 households before the invasion. That is still an enormous number of people to be living on a flood prone shingle spit. But Winchelse was not a normal port.

Figure 15: Romney Marsh in medieval times; after Andrew Pearson

Pearson hints at Winchelse’s secret in Figure 15, where he marks ‘Possible tidal inlets’ on the Camber spit. Cunliffe and Green analysed the flow of the Brede and Rother from Roman times. One of their investigations is why the Brede had the greatest concentration of saltpans in southern England while the Tillingham and Rother further north had none. It is odd because the main sea water access was through a breach at Old Romney which was north of all of them. Sea water from Old Romney would have had to pass through Guldeford and Walland marshes to get to the Brede, but neither show signs of marine encroachment. Their conclusion is that there must have been at least one other breach in the Camber spit near the mouth of the Brede. It would be around what is now Rye Golf Course, which is where Cooper suggests Winchelse was located.

De Viis Maris, written in the late 12th century, is more explicit. It says: “and further up the Winchelse inlet is a good town called Rie”. It is saying that there was an inlet through the spit at Winchelse which led to Rye.

This would explain Winchelse’s success. Presumably, the settlement grew up to service ships and barges passing through the breach. Why else would anyone build a settlement on a shingle spit with little food and no running fresh water that was prone to flooding and too unstable for building foundations? Cooper reports that Winchelse had bridges. There is no reason for bridges on a shingle spit, other than to cross a channel or canal. Note that if Winchelse was at this breach, Pearson is right to suggest that there was a third breach towards Pett because Winchelse has the ‘ea’ sounding suffix associated with islands in this region. Jeake and Cooper both reckon it was an island.

Green thinks that the cross-spit breach was narrow and blocked at low tide. If he is right, ships must have docked on both sides, either to be loaded, unloaded or wait for the tide before making the spit crossing. This was doubtless profitable for bars – of which there were apparently 11 - and brothels, but they cannot account for more than a fifth of the people that lived there.

Even though a channel or canal would be a lot faster than the route via Old Romney, it would still be inefficient, requiring a high tide to get in and another to get out. From the number and type of people that are named in the 1288 Charter, we think the Romans developed a better logistics system that was still in use at the time of the invasion, and which employed the balance of Winchelse’s population.

Winchelse seems to have been a rudimentary transhipment hub. The commercially sensible process would be for inshore barges to take iron and salt from jetties on the banks of the Brede to Winchelse where they would wait for the tidal current to whisk them across the spit. On the seaward side of the spit, the barges could be unloaded directly into sea-going vessels or into warehouses. If the channel was temporarily blocked by storms or longshore drift, perhaps cargo was carried across the spit in carts.

Some of the iron and salt would have been exported to the Continent. Rather than return empty, presumably the freight vessels returned wine, oil, fruit and cloth, most of which would have been destined for London or Winchester on tidal drifters, or to Canterbury and the hinterland on the Rochester Roman road. This would explain how Norman navigators got to know the treacherous and ever-changing offshore sand banks around East Sussex.

Cunliffe explains that the most likely reason for the Camber spit breach at Winchelse to remain open was if freshwater channels were feeding through from the Brede and Tillingham. He still seems a little perplexed that eastward shingle drift did not block the breach or create its own spit. We have a simple explanation: We think it was dredged and had been since Roman times.

Wealden iron ore was a major reason for the Roman invasion of Britain. We think that the Romans, who hated bendy transport systems, cut the breach in the Camber shingle spit to accelerate exports of iron and salt, then established the docks to facilitate loading. Those docks would be within two miles of the location described by Ptolemy – i.e. mid-longitude between what is now Cannon Street in London and South Foreland in Kent – for ‘Portus Novus’. If we are right, Winchelse was the major port in East Sussex from Roman times right through to the 13th century.


Rameslie manor is traditionally modernised to Rye. We cannot find the origin, but it is odd. Rye has been subject to dozens of excavations without finding any evidence of a Saxon era occupation. It was an island, perhaps with a low tide causeway, in the 11th century, which makes it an unlikely hub for a huge and hugely wealthy manor like Rameslie. If it was within Rameslie manor, we doubt it had any significance before the invasion.

Yet it had become a major port by the time of the Ship Service in 1227, being charged to provide five ships, as many as established ports like Romney and Hythe, half as many as Winchelse, and five times more than any other port around the Hastings Peninsula. It cannot have grown to half the size of Winchelse in less than 100 years. A Charter assigning it additional liberties was issued in 1191, which means it was already well established by then. Rye liberties were later sold by the Abbey of Fécamp, so it clearly belonged to them and must have been established by them.

We believe that the Abbey of Fécamp established some sort of administration and port service centre at Rye soon after the Conquest. Perhaps this was forced upon them when military administrators moved into Hæstinga ceastre at Winchelsea (as we suggest above). Perhaps they just needed more capacity to handle extra passenger traffic with Normandy.

Regardless, this might solve one last Domesday enigma, the ‘novus burgus’ entry in the manor of Rameslie. Domesday’s boroughs generally refer to somewhere that has liberties and/or rights to toll, typically a port or harbour. In our opinion, it was probably Rye.

Just as a matter of interest, it is usually thought that the name Rye has an Old English root, with its ‘ye’ ending being typical of islands in this region. And we think it was an island in Saxon times. Yet Rye was originally spelled ‘Rai’ which, as far as we know, would be a unique structure for an Old English place name. It was in the Abbey of Fécamp’s manor of Rameslie. We think they named it Rai after a place near to Rugles in Normandy, and probably did so soon after the Conquest.