History Files

Please help the History Files

Contributed: 175

Target: 400

Totals slider

The History Files still needs your help. As a non-profit site, it is only able to support such a vast and ever-growing collection of information with your help, and this year your help is needed more than ever. Please make a donation so that we can continue to provide highly detailed historical research on a fully secure site. Your help really is appreciated.



Anglo-Saxon Britain

Who were the Hwicce?

by Edward Dawson, 1 March 2009. Updated 24 November 2018

One aspect of historical study involves words and their meanings. As I am fascinated by words and their meanings let's take a closer look at one name which has puzzled and annoyed me: that of the Hwicce.

The apparent meaning of the word 'hwicce' in Anglo-Saxon (AS) is 'locker, chest, trunk'.

However, it's very hard to see a group of aggressive Saxons deliberately calling themselves that. It neither proclaims them to be powerful (for example, Goth, ie. the gods, or Alemanni, 'all men'), nor does it attempt to belittle them (for example, wiking or viking, a waterborne thief who hides in inlets).

But there may be a solution to this oddity. What if Hwicce is a degenerate form of Gewissae (Anglo-Saxon Chronicle: the certain, knowing or trustworthy)?


The Gewissae were laeti who were embedded among Britons, and who knows what liberties they and their neighbours took with mispronouncing the name? It's easy enough for the 'g' of Gewissae to shift into a 'k', and the Briton habit was to pronounce a 'k' as 'ch' (as in the Scots word, 'loch'), then the Saxons could soften it more into an 'h'. And the 'ss' of Gewissae may well have been spoken as an 'sh' by Britons, which the Saxons could alter into a 'ch'. And as we have no idea which dialect the Belgae spoke, all of this could produce 'Hwicce' by natural processes.

So where did 'Gewissae' come from in the first place? An individual named Gewis is claimed as a descendant of Blaedaeg, son of Woden, almost certainly an invented name. 'Gewisse(n)' is a plural form, with the personal name being backwards formed from it by removing the plural. The prefix is 'ge-'. This indicates more than one person is doing something together, or it is something else acting which is not a single thing. 'Ge-' gradually degenerated over the centuries into 'y-' and 'i-' (for example, the Anglo-Saxon 'Sumer is icumen in'), and then later into 'a-', and is now entirely archaic and has been dropped. When you read the old sentence 'a-hunting we will go', the 'a-' on the front indicates that at least two people, maybe more, will go hunting. Thanks to the 'ge-' on the front of 'wisse' there is no way that it can be a personal name. Instead, 'wis' and 'wit' mean 'to know'. It is still used in modern English as deliberately archaic slang, for instance when someone says 'to wit'. It can also be seen in 'wise'.

The Gewissae were the 'Educated People', probably a reference to the laeti being Roman citizens after the first generation and receiving a level of Roman education. It seems that these Saxons could have become the dominant force in the Thames Valley Saxons, prior to conquest by the West Saxons. It's also possible that they were the Saxon migrants into later Hwicce land, perhaps moving to avoid West Saxon control, (see the relevant lists for more information).

An intriguing alternative possibility is that their name was a British pronunciation of 'guest' (ASC: 'gaest, giest') which is 'gwestai' in modern Welsh. The Britons may have taken the second pronunciation of 'giest', added their habitual 'w' after the 'g' and dropped the 't'; and with an ending vowel added by the Saxons. This would produce the name Gwiesa or Gwiese - just add the Latin plural (nouns ending in '-a', with the plural being '-ae') and you have Gwissae. One must always keep in mind the fact that although the Britons spoke their Old Welsh, they wrote in Latin.

The River Dee
The River Dee, Deva to the Romans, or Afon Dyfrdwy to the Welsh, forms part of the modern border between England and Wales, but in the sixth century it may have formed another border, that between Powys to the west and Rheged to the east

As stated, it's easy enough for the 'g' of Gewissae to shift into a 'k', then to a 'ch', and then to soften into an 'h'. Similarly, the 'ss' of Gewissae could have become 'sh' and then 'ch' with the Britons. This could produce Hwicce or something close to it. Once it had mutated that far, the Saxons would have tried to justify it or make it sensible again by using the nearest word which sounded approximately the same, one which means 'chest' or 'trunk'. In a culture with no written records at the time, there would have been no references to use for going back to the original meaning.

Fluctuating borders

When I look at a modern conjectural map of kingdoms of Britons and Saxons during the Dark Ages, I am struck by the neat, even, tidy lines drawn between kingdoms. That there were many exact boundaries I have no doubt. That there were many inexact borders between kingdoms I am equally certain

Why? Because of non-arable or marginally arable land. The period in which the Saxons managed to conquer the majority of Britain was marked by persistent cold weather, apparently punctuated by severe storms. The Roman Warm Period had ended during the fourth century and it became much harder for a farmer to bring in a crop every year. What is seen today during our modern warm period as useable land was often not worth living on in AD 500. Uplands would be abandoned in many if not most cases, and the people farmed the valleys. A mid-fifth century plague and Saxon raids most likely cut the British population down enough so that the survivors simply moved into the suddenly extra-important river valleys.

For example, I have seen maps of Powys which show a northern extension of the kingdom only on the western side of the River Dee (Deva). In my opinion that northern extension of Powys prior to the battle of Chester was the Dee Valley itself, on both sides of the river. Whoever claimed as their own any unused or slightly used nearby uplands was probably of little relevance. What was important was the valley itself and the agriculture the valley supported.

The Thames Valley path to the west

So too were other British kingdoms formed around arable river valleys. The kingdom which has been conjectured to have been centred on Caer Gloui dominated the River Avon valley (one of many such rivers named Avon, from the Brythonic word 'afon' for 'river'. I have to wonder if the original name of this particular river was Afon Gloui, the River Glou). Caer Baddan was the hill fort above the valley containing Bristol (another Avon). Caer Ceri stood above the westernmost headwaters of the Thames watershed.

The Thames watershed has its origin at Thames Head (or nearby; there is some disputation on the subject), near the modern village of Kemble in Gloucestershire. Kimble is 6.5 kilometres from Caer Ceri (modern Cirencester).

Caer Ceri's location at the western end of the Thames placed that kingdom in an awkward position: the Thames Valley was occupied and controlled by Saxons, some of whom appear to have been settled there as laeti by the Roman administration during the fourth century. Eventually the Saxons would show up on their doorstep as they expanded upriver...

Roman Aquae Sullis (modern Bath)
Shown here is an artist's impression of the Roman city of Aquae Sulis, a stronghold during the time of Ambrosius but now buried under modern day Bath in Somerset

If the Gewissae were indeed laeti who were settled in the uppermost reaches of the Thames, they would have been perfectly positioned to send their younger sons a few kilometres to the west to settle in a West Saxon-controlled Ceri, Gloui, and Baddan (modern Bath). I would argue against the Gewissae being located in a coastal area such as Hampshire because seaborne barbarian laeti would be easier to control if they were settled away from the sea. What seems to be missing from the historical record is any mention of sub-Roman British kingdoms in the Atrebates tribal area. I suspect the Gewissae were settled in the north to north-west corner of the Atrebates territory, on the fringes of the postulated British territories of Caer Gwinntguic and Caer Celemion. So far the archaeology in the region has found traces of Saxons farther east than I am postulating, in the Abingdon area.

Emergence of the 'Hwicce'?

I perceive the pattern of conquest of the three cities of Caer Gloui, Caer Baddan, and Caer Ceri to have happened roughly like this:

1. The West Saxon army defeats the three cities at the battle of Deorham (Dyrham) in 577 and kills the magistrates and their knights. Most British nobles flee, abandoning their serfs.

2. The Wessex king rewards his most prominent and effective soldiers with land. He assigns each of the three cities (and also Worcester) to high ranking Saxon leaders, and smaller subdivisions to lesser soldiers in his army. If these soldiers were descended from the Gewissae then it would explain this entire territory belonging to the 'Hwicce'.

But this leaves vast tracts with Britons on them which have no Saxons...

3. Younger sons, and young men with no prospect of inheriting land, head west to join the West Saxon soldiers mentioned in Point 2, forming personal bodyguards, and slowly or quickly (more likely quickly) they are each assigned a hide of land or more in return for military service. Lower caste Saxons use the conquest as an opportunity to better their conditions. These Saxons are the Gewissae, moving in; from adjacent settlements. At first they settle the Caer Ceri area, then later filter into Gloui and Baddan.

4. Within a few years the three cities area has changed, being managed by a Saxon nobility on whose land lives both Saxon soldiers and British serfs/slaves.

The majority of the Saxons in Point 3 would come from the nearest Saxon settlements; these would be in the upper Thames Valley from where I am postulating the presence of the Gewissae. A possible major settlement of Hwicce Saxons among Britons is Wanborough, mentioned below in a quote from the Anglo-Saxon Chronicle. The Roman town is less than thirty-two kilometres to the south-east of Caer Ceri down the Roman road called the Ermin Way (not to be confused with Ermine Street).

Wanborough was likely to be another of several Roman towns who, because they held onto the civitas system of Roman-style leadership elections, selected a Saxon as magistrate. Saxons acquiring Roman titles appears to have been the case at Lincoln of the Lindisware, because Saint Paulinus reported meeting a Praefectus Civitatis there who was a Saxon. I doubt that Lincoln was a single isolated case.[1] One could argue that much of Roman Britannia didn't actually 'fall' to the Saxons. Many civitas capitals may have embraced the protection of a strong Saxon leader and his war band, legally electing the Saxon as magistrate. Only in places where the British 'magistrate' was mere window dressing for a hereditary kingship did the Britons have a chance to remain British.

If the Hwicce were indeed the Gewissae, they would have been living among Britons for centuries. They would view Britons as a familiar people to be exploited, not foreigners to be killed. As was pointed out to me in a discussion on this subject:

[1] Like the former three cities area of Caer Baddan, Caer Ceri, and Caer Gloui, Lincoln apparently retained a prominent population of Britons and many British influences on the Anglian kingdom which later emerged.

...there's a strong element of the survival of British culture and worship practises in the Hwicce lands. Maybe they just accepted their new, apparently benevolent, masters over the old. The period of serious conquest, with Britons fleeing before the swords of the invaders, was on its way out by 577. Now it was more a game of shifting political patterns (with the British kingdoms usually losing out), so a British defeat like 577 may have been nothing more than losing the top level of Britons in exchange for a top level of Saxons. The British farmers carried on ploughing their fields regardless. [2]

Final note

That there was some sort political interaction between Saxons and British citizens is implied in a fascinating (and chilling) entry in the Anglo-Saxon Chronicle:

AD 591. This year there was a great slaughter of Britons at Wanborough; Ceawlin was driven from his kingdom, and Ceolric reigned six years.

It sounds as though Ceawlin was supported by the British at the Roman town of Wanborough, and that when Ceolric took control he killed Ceawlin's supporters... British supporters.

Since Ceawlin had been the leader at the battle of Deorham, then it seems the Hwicce were Ceawlin's (probably loyal) supporters. I can see the likelihood of Ceawlin operating with British-tolerant political support, and apparently direct support by the Britons of Wanborough. To remove Ceawlin would require pacifying his supporters, especially 'foreign' (Welsh) ones, whose deaths would not cause a reaction from angry Saxon relatives demanding weregild.

Were the Hwicce the (Saxon laeti) soldiers of Wanborough's Britons, descended from the Gewissae who had been settled by the Roman administration two centuries earlier?

[2] Peter Kessler, 15 November 2008.


Main Sources

Greek references - Dictionary.com website

Latin references - Bantam New College Latin & English Dictionary by John Traupman, and the University of British Columbia Maths Department website

Ammianus Marcellinus - Res Gestae Libri XXXI available online at 'Early Church Fathers - Additional Texts' website

Ptolemy - Geographia

Map at Rootsweb (the main map is very silly because it has Concani and Gangani adjacent in the Thomond area, when anyone can see they are the same tribal name with different spellings)

Map perhaps not under copyright: Alexander G Findlay's Insulae Brittanicae, produced in 1849 in A Classical Atlas of Ancient Geography, available online at the University of Texas Libraries



Text copyright © Edward Dawson. An original feature for the History Files.