Bury Hill, Approximate Site of Ariconium (Photo Courtesy Wikipedia)
Before I offer my solution to the etymological problem posed by the tribal name Hwicce, I would urge my readers to seriously peruse the following excellent study on the subject by Dr. Richard Coates:
I feel everyone is working way too hard on trying to come up with an acceptable etymology. English Hwicce, 'chest, coffin, ark', is just fine. I would guess that it is an English rendering for Ercing or Ergyng, whose name is to be directly derived from the Roman settlement-name Ariconium. While I've elsewhere supplied the proper etymology for Ariconium itself (see http://mistshadows.blogspot.com/2016/10/a-new-etymology-for-roman-settlement.html), the early English could easily have applied their own meaning to the Welsh kingdom-name.
Ariconium and the Dobunni
(Map Courtesy Barry Cunliffe)
(Map Courtesy Barry Cunliffe)
Ariconium and the Hwicce
(Map Courtesy Della Hooke)
(Map Courtesy Della Hooke)
The GPC lists the Welsh name for ark, chest, coffin as follows:
[bnth. Llad. arca, Crn. C. argh, Llyd. C. arch, H. Wydd. arc]
eb. ac yn eithriadol eg. ll. eirch, archoedd, archau.
a Cist, coffr; blwch i gladdu neu amlosgi corff ynddo, coffin, ysgrîn; hefyd yn ffig.; ?creirfa:
chest, coffer; coffin; also fig.; ?shrine
Rivet and Smith, in their THE PLACE-NAMES OF ROMAN BRITAIN, have the following entry on Ariconium:
Antonine Itinersary : 4853 (Iter XIII) : ARICONIO.
DERIVATION : The name is formed from British *are- (*ari-) 'in front of ' and *conio-, of unknown meaning but perhaps the same as in Viroconium. Among names formed in this way (Holder I. 188) are possibly British Argistillum (and divine name Arnemetia), and abroad Armorici (Aremorici, the people 'in front of the sea'), Arelaunum silva, Areduno vico > Ardin (Deux-Sèvres, France); these do not help to guess a meaning for the present name, though Jackson observes that thc prefix is 'usually used in place-names of regions beside some feature such as a forest, a marsh, the sea, etc.' (Britannia, I (1970), 68).
IDENTIFICATION. The Roman settlement at Weston under Penyard, Herefordshire (SO 6423).
Note. The name has interesting survivals. In EPNS, XL, 192, its first part, with Anglo-Saxon -ingas attached, is recorded as Ircingafeld in the ASC (918) and Arcenfelde in DB, modern Archenfield, a deanery of the diocese of Hereford. There is also Ergyn(g), the Welsh name for a district in Herefordshire. Possibly the first element survives in a different form in the name Yartleton; the Roman site is some three miles to the north-west of this place.
Eilert Ekwall in his CONCISE OXFORD DICTIONARY OF ENGLISH PLACE-NAMES adds early forms Erchenefeld, Erchin, Ercincg. More forms/spellings can be found at http://www.melocki.org.uk/places/A.html:
circ. 380 Ariconium, Iter Anton. ?
915 Ircingafeldes, Yrcingafeld, Iercingafeld, A.S. Chron.
1086 Arcenfelde, Arcenefelde, Dom.
circ. 1120 Jerchynfeld, Glos. Cart.
circ. 1130 Ergyng, Ercincg, Ergin, Erchyng, Erchynfeld, Urcenevelde, etc., Lib. Land.
1138 Erchenefelde, Glos. Cart.
circ. 1147 Erging, Geof. Mon.
circ. 1150 Herchenefeld, Brec. Cart.
1243 Urchenefeld in Wallia, T. de Nev.
1291 Irchenefeld, Yrcheneshome, Tax. Eccl.
circ. 1550 Herchinfield, Leland.
no date Ierchenfeld, Herchenefeld, Glos. Cart.
According to https://eprints.utas.edu.au/19778/2/whole_GrimmerMartinRae2003Vol2_thesis.pdf,
"The name Ergyng derives from the Roman town Ariconium (Weston-under-Penyard, Herefords.). As this is to the east of the Wye and the later region of Archenfield, some historians have suggested that the centre of British Ergyng had retreated westward under Anglo-Saxon pressure (e.g. Sims-Williams, Religion and Literature in Western England, p. 45)."
From "Religion and Literature in Western England", 600-800 by Patrick Sims-Williams:
"It could just have been on the margins of Dobunnic territory."
Della Hooke, however, insists that
"The Welsh district of Ergyng, a district around the Roman centre of Ariconium, lay outside the kingdom of the Hwicce. This is discussed by Bruce Coplestone-Crow in his BAR book (British Archaeological Reports, British series 214, 1989) Herefordshire Place-Names."
In her thesis, Dr. Sheila Waddington makes the case for Ergyng having once extended all the way to the Severn (see http://etheses.bham.ac.uk/4738/9/Waddington13PhD_-.pdf). Included in that study is a map displaying the original extent of this kingdom, drawn from Coplestone-Crow's HEREFORDSHIRE PLACE-NAMES:
Figure 7.11 Coplestone-Crow’s depiction of the extent of Ergyng in the sixth and
seventh centuries, extending from the eastern banks of the Wye, Monnow and Dore
across to the western bank of the Severn. The territory of Dunsaete may have been
within the district ‘Cantref Coch’, shown hatched red. Coplestone-Crow,
Herefordshire Place-Names, 3, with additions.
If this is correct, then there are some adjoining boundaries and even "overlap" territory for Hwicce and Ergyng along the Leadon and to the west of the Severn. Ariconium itself, however, would still be somewhat to the west of the Hwicce's western boundary.
Now, the Kingdom of Ercing in Wales not only has major Arthurian associations*, it bordered on Uther Pendragon's/Illtud's "Llydaw", i.e. the Vale of Leadon. And the Vale of Leadon was anciently within the Dobunni tribal territory (a fact confirmed by Barry Cunliffe). The Leadon later formed the western boundary of the Hwicce (a surmise based on the boundary of the Diocese of Worcester). As the Hwicce appear to have been the sub-Roman inheritors of the Dobunni kingdom, and their name accords with a folk etymology applicable to Archenfield, I would suggest that at some point in the history of Ercing this kingdom had extended its sway over the region once controlled by the Dobunni. Or, conversely, Ercing was merely a sub-kingdom of the larger Dobunni tribal area and, eventually, that area took on the name Hwicce from Archenfield. This may have happened merely because Ercing supplied a ruler or rulers to the larger region, or because Ercing happened to be the most powerful sub-kingdom at the time.
My solution to the Hwicce name problem has the advantage of relying upon a known place-name (Ariconium) and kingdom-name. On the other hand, past theories have no foundation in any known place-name or tribal group. Even Coates must admit that his hypothetical hygwych does not exist in Welsh. I would add that I've been unable to find any other tribal or place-name constructed like hygwych; there simply is no precedent for such a formation.
In conclusion, we can surmise that the Welsh also did not know the real etymology of Ariconium/Ercing/Ergyng. There is no reason whatsoever to suppose that they did! They may have linked the name to their own word arch, 'ark, chest, coffin', with the plural eirch, and so prepared the way for the translation into English Hwicce. As modern scholars have not been able to parse Ariconium (until I related it to a Gaulish word meaning "lord"), why should we think the Dark Age Welsh were able to do so?
* Eigyr, Arthur's mother, is the daughter of a king of Ercing in Welsh tradition. Illtud's mother was also from Ercing. A son of Arthur is placed there, albeit in the form of a personified stream/wellhead.