an accidental excursion into monmouthshire journal 2019-2020rev7...bemerton, wiltshire, (the poet...

204THE CASTLE STUDIES GROUP JOURNAL NO 29: 2015-16THE CASTLE STUDIES GROUP JOURNAL NO 33: 2019-20

William Coxe, Thomas Morrice andthe Castles of Monmouthshire -

An Accidental Excursion intoMonmouthshire

Jeremy KnightIn August 1798, the historian and travellerRevd. William Coxe (1747- 1828), vicar ofBemerton, Wiltshire, (the poet GeorgeHerbert’s old parish outside Salisbury), wasstaying with his friend and patron Sir RichardColt Hoare (1758-1838) at the latter’s houseat Stourhead in Wiltshire. The anniversary ofthe death of Sir Richard’s wife wasapproaching and this may have encouragedthe two to embark on ‘an accidentalexcursion into Monmouthshire’. They stayedwith Colt Hoare’s friend James Green M.P.at Llansantffraed outside Abergavenny, andit was there that the three friends devised theidea of a history of the county. When the twovolumes of Coxe’s An Historical Tour ofMonmouthshire appeared in 1801, it was,Coxe wrote in his dedication to Colt Hoare,‘commenced in your company, written atyour suggestion and embellished by yourpencil’. It was also, he might have added,funded by Colt Hoare’s considerable wealth.(Coxe 1801, dedication). A panel of experts was enrolled to assist.Coxe’s friend Thomas Leman (1751-1826)advised on Roman remains, the pioneer (iferratic) Celtic scholar William Owen Pughe(1759-1835) on Welsh place names,1 a newmap of the county was commissioned fromNathaniel Coltman, and a Cardiff landsurveyor Thomas Morrice (1727-1812),normally employed on the canals and tramroads of the industrial revolution, providedsurveyed plans of towns, ancient earthworksand castles. As a result, Coxe’sMonmouthshire is one of the first historicaland archaeological works to be illustratedwith professionally surveyed ground plans ofhillforts, earthworks, and castles.

William Coxe, canon of Salisbury cathedralfrom 1803, and archdeacon of Wiltshire from1805, was a documentary historian ratherthan an antiquarian. His work on eighteenth-century European history has been praised bymodern historians of the period. (Plumb1966).2 However, whilst a careful andaccurate observer, his knowledge ofprehistory, or of medieval architecture, wasthat of his time. Round-headed arches were‘Saxon’ (or ‘pre-conquest’) or ‘Norman’.Pointed arches were ‘gothic’ and later.Michael Thompson has shown that the truedate of such ‘Saxon’ buildings was alreadyknown, as demonstrated by Colt Hoare’s useof the term for places such as LlanthonyPriory, whose origin was known fromdocumentary sources (Thompson 1983, 25,233). However, the use of ‘pre-conquest’suggests that some measure of ambiguity still

William Coxe (1747-1828). Engraving by W. T.Fry, 1904, based on an original picture by SirW. Beechey, RA, drawn by J. Jackson, in ‘AHistorical Tour of Monmouthshire’ published byDavies and Co., 1904.

An Accidental Excursion into Monmouthshire


existed, at least among more traditionalhistorians. Coxe’s account of the lords ofChepstow and the history of Chepstow castle(Coxe 1801, Vol II, 375-378) covers muchthe same ground as the modern Cadw guides.He saw that the castle, whilst of Normanorigin, had been altered and added to atdifferent periods. He dismissed the idea thatthe early Norman hall block (then ‘thechapel’) was of Roman origin (‘Somefanciful antiquaries have attributed theconstruction of the castle to Julius Caesar’Coxe 1801, 365) and recognised that the hallwas largely built of re-used materials (Coxe1801, 365-391) but the comparativearchitectural detail necessary to relate thestructural history of the castle to thedocumentary evidence was not yet available.The situation with rural earthwork castles waseven more fluid. Eighteenth-century travellers, educated inthe classics, often saw rural earthworks andfield monuments as relics of the Roman army.The motte-and-bailey at Walterston (‘Coed yCrusel’), on the Hereford side of theMonmouth border, but surveyed by Morrice,lay on the approximate line of the RomanAbergavenny to Kenchester road. Coxe,though he says nothing of the motte, thoughtthe semi-rectangular bailey to be Roman.(Coxe 1801, 23 and plan). Indeed, it ispossible that the motte and inner ward weresuperimposed on an earlier earthwork (fig. 1).Similarly, discussing the castle mound builtby William II next to St Gwynllyw’s churchat Newport (‘Twyn Gwynllyw’), Coxe citedthe earlier opinion of Revd. William Harristhat it was ‘an arx speculatoria or watchtower, which the Romans always constructednear their camps’ (Harris 1773, 7; Coxe 180155).3 The name Twyn Gwynllyw(’Gwynllyw’s mound’) follows an older folktradition which saw mottes as burial mounds.At Trellech in Monmouthshire, the sundialerected by Lady Magdalene Herbert in 1689

with sculptured representations of theantiquities of the village shows the motte(‘Magna Mole‘) with the caption O Quot HicSepulti ‘Oh, How Many are Buried Here?.Of the prominent motte of Twyn Barlwm onits high ridgeway siting, Coxe, afterdiscussing theories of its origin and date,concludes: ‘It might contain the ashes ofsome valiant chief of the Silures, who fell indefending his country against the Romans’(Coxe 1802, 75-76 ).4

The Afterlife of Castles - Chepstow andMonmouthMost castles in Monmouthshire were alreadyruinous in Elizabethan times, as the travellerThomas Churchyard, writing in 1588,lamented:-

Most goodly towers, are bare and nakedlast, / That cov’red were with timber andgood lead. /.. The shell of this, I meanethe walls without, / the worthie work, thatis so finely wrought../ the firme freestone,that was so derely bought, / Makes menlament, the loss of such a thing /.....To seeso strong, and stately worke decay’.(Churchyard 1776, 54).

Three castles were still maintained as seatsof power by the Somersets of Raglan, Earlsand Marquesses of Worcester and laterDukes of Beaufort. Raglan was their mainseat; Monmouth the county and assize townand Chepstow a port and gateway to thecounty, whether by land or sea. Chepstowwas also important as the outlet for the supplyof Forest of Dean timber for the navy and forBeaufort’s lucrative ironworks aroundTintern. Raglan and Monmouth castles wereboth slighted at the end of the Civil War andthe Somersets removed across the BristolChannel to Badminton in Gloucestershire,signalling their wider involvement in nationalpolitics. During the First Civil war (1642-6),Chepstow castle was held for the king, but




surrendered to Parliament in October 1645after a short siege, when the walls werebreached by two brass culverins and an ironpiece. During the Second Civil War (1648-49) local royalists seized the castle in May1648. Cromwell, on his way to put down arising in west Wales, left Colonel Ewer tocomplete the siege. An attempt to storm thecastle gatehouse failed, with the loss of aMajor killed by a stone dropped from above.Ewer ‘raised (razed) the battlements of theirtowers with our great guns’, silenced thecastle’s artillery and bombarded the interiorwith mortar fire. Finally, after a fortnight’ssiege, when the south curtain has beenbreached ‘so low that a man may walk into it’the castle was stormed (Knight 2005, 99, 125-6). The story of the earlier part of the periodbetween these events, when all three castlesstood siege, and the visits of Coxe and ColtHoare, when they appear in their new role assubjects of antiquarian study is related to abitter post-war feud in the county. This wasbetween Henry Somerset, third Marquess ofWorcester, first Duke of Beaufort from 1682,and a group of former Cromwellians. It wasrooted in opposition both to Somerset’s familyhistory of Catholicism and his felling of wood-land in Wentwood Forest despite long estab-lished rights of common, and owed more tolocal than to national factors. It culminated atthe time of the so called ‘Popish Plot’ of1678-79, when four Catholic priests wereexecuted in Usk, Cardiff and Hereford. Pre-war Monmouthshire had probably the highestproportion of Catholic recusants in England orWales, and though Henry Somerset was now aProtestant, he was widely suspected of being a‘crypto Papist’ and had many Catholic alliesand clients. These factors were to affect the postwar history of Chepstow and Monmouth castles.

Chepstow was used for state prisoners,whether the Anglican divine Bishop JeremyTaylor under Cromwell; the regicide HenryMarten at the Restoration; or the

nonconformist Nathan Rogers and otherformer Cromwellians at the time of theMonmouth Rising under James II. During theDutch Wars, many prisoners of war were heldthere. (C.S.P.D Addenda 1660-1670, 364,384 ) Following the 1648 siege, the southerncurtains were lowered in height andthickened to resist gunfire, several towersfilled with earth and embrasures made forcannon. The south curtain wall of the LowerBailey was rebuilt, backed by a substantialearth rampart, revetted by a stone wall, witha gun port and a series of musket loops alongthe parapet. These works can be related torecorded expenditure of £300 in 1650 and£500 in 1662 (C.S.P.D, 1649-50, 176, 308,381; 1654, 53;1655 256; 1661-2, 490, 521,Turner 2000, 20) 5

Under Cromwell, in times of crisis 400-500locally recruited horse and foot could beassembled at Chepstow to reinforce thegarrison and armed from the stores within thecastle, In the looming uncertainty of 1659,the garrison was increased to 100. In 1660Henry Somerset, son of the second Marquessof Worcester, was appointed governor by thenew royalist government. There was aproposal that the castle be demolished, butSomerset wrote to Charles II that it was ‘thekey to the four adjoining Welsh counties...(and) a bridle to the ill-affected, who aboundin those parts’. He offered to pay for areduced garrison of 60 men, with a Governor,lieutenant and ensign, so putting the castleunder his direct control, though nominallystill a royal garrison (C.S.P.D. 1661-2 184,73). Two muster rolls of ‘His MajestiesGarrison of Chepstow’ survive, from 1662and 1664, listing 2 sergeants, 3 corporals, agunner, a drummer and 60 men. Somersethimself acted as Captain or Governor, with aLieutenant Governor, Thomas Nansen in1662, in actual command. There was still apost-war nervousness and Nansen wrote toSomerset in December 1661 that a former


Fig. 2. The ‘Inside view of Chepstow, looking eastward’. From an engraving of the drawingby Paul Sandby, 1777. A view of the Lower Bailey with a view toward the main gatehouse.At the time it was being used as a farmyard with (left) the glass factory and malt-housebuildings south of the Great Hall, from which extends the elaborate chimney.

Fig.1 Walterston, Hereford-shire. Plan of the motte-and-bailey, surveyed by ThomasMorrice. (T. Morris del etsurvd, Harding sc.). CoxeHistorical Tour in Mon-mouthshire (1801). Coxethought the squarish baileymight be Roman (Coxe 1801,23)



parliamentarian living in Chepstow had‘rebuilt and fortified a strong wall’ aroundhis house, originally part of the parliamentarysiege works, which commanded the southface of the castle. (C.S.P.D. 1661-2, 184, 73). Two inventories, of 1672 and 1679, list thearmaments of the castle in detail. In 1645 theyhad included eighteen cannon and twoarquebuses (The True Informer, 22 Oct. 1645). The 1679 survey (Geear et al 2006, 230-34)lists some 20 brass and iron guns in variouslocations within the castle, plus two mortarsand thirteen‘ wall-pieces and murtheres’ outof repair. Many were on ‘old decayedcarriages ’or ‘lying on the ground’, andthough still serviceable, were in a state ofneglect. There were also quantities ofmuskets, pikes and even bundles of arrows,many of which had clearly been there forsome time. The only surviving firearm fromChepstow castle is a copper-alloy ‘pocketdag’ or pistol of c. 1600, shaped like aminiature cannon and now in the NationalMuseum of Wales (Redknap 1996). Henry Somerset had become third Marquessof Worcester in 1667. At the time of theSuccession Crisis over Charles II’s Catholicbrother and heir to the throne, James, and therelated Popish Plot, there were renewedaccusations of Papist sympathies againstSomerset, his Deputy Governor CaptainFrances Spaulding and the Chepstow garrison(C.S.P.D. 1679-90, 20-21, 1683, 180). Later,in 1685, following the Monmouth Rising,James, now king, disbanded the Chepstowgarrison under Beaufort’s command andreplaced it with one of his own. Beaufort’sindignant letter to Clarendon shows thesomewhat ambiguous role of the castle as aroyal garrison at this time :-

‘I am therefore no more to havethe command of my own house, orthe garrison in it, ....the main pur-pose of the castle as a magazine for

armaments in the district wouldcease to exist.... how ammunitionand stores could be .... continuallyre-checked is not explained.... gun-ners are necessary for so manyguns, of which many are my ownproper goods’ (quoted Durant1973, 63).

Eventually, in 1690, after the fall of James IIChepstow’s remaining guns were shipped toChester and thence to Ireland for serviceagainst James in the Williamite War. The castle however remained occupied. Onhis first visit in 1798, William Coxe met aneighty-year-old lady who lived in the LowerBailey as a tenant of the Dukes of Beaufort.She remembered two elderly former maids ofHenry Marten, who lived in the upper roomsof the tower which now bears his name, withMarten and his wife on the floor below. Aview by Paul Sandby of 1777, (fig. 2), showsa lively scene in the Lower Bailey. The viewis framed by the tower of the gate to theMiddle Ward (then used as a nailer’sworkshop) on the left and Marten’s Tower,still roofed and with a now vanished gable, tothe right. Roger Bigod’s Great Hall had beenre-roofed as a glassworks, for the manufactureof wine bottles (Turner 2000, 20-21). Smokeissues from a chimney and a square tower-likelouvre with a pyramidal roof is behind it. Thescene is animated by the usual picturesquefigures, including cattle, perhaps artist’slicence, and the walnut tree in the middle ofthe ward, which survived until the 1950s, isslender, but already well grown (Turner andJohnson 2006, p. 211). The castle was already attracting manyvisitors and by Coxe’s time there was adisplay of decorated medieval tiles ‘orna-mented with birds and flowers’ on a wall inthe Lower Bailey. (For similar tiles see Lewis1999, nos. 44-50 and p. 232). Sandby’s printand Colt Hoare’s drawings in Coxe’s Tourshow what has been lost, particularly from



the upper works of the castle. Marten’s Towerstill retained its roof and gabled upper storey,and ‘no less than twenty four ancient chimneys’remained, the main one, on ‘the inhabitedbuildings’(Roger Bigod’s ‘Earls Chamber’)handsomely decorated on the outside’. This canbe seen on the Sandby drawing (fig. 2). Theview of Marten’s Tower also shows the adja-cent, now vanished, well-house - a circularstructure with a conical tiled roof, a wellchamber approached by a flight of steps, and alean-to porch. (Coxe 1801, 367 and plate oppo-site). One can also see some of the lean-tobuildings against the interior of the curtain wall,whose fireplaces remain visible. Morrice pro-vided an excellent surveyed plan of the castle,together with internal elevations of the walls ofthe early hall block, and a view of its ‘Saxon’doorway, with details of its chip carving.Monmouth Castle had a different but related

post-medieval history. Speed’s plan shows aninner ward with a circular great tower similarto e.g. Skenfrith or Caldicot and a hall com-plex, comprising a twelfth-century hall-keep,with an early-thirteenth-century great hall atright angles (fig. 3). The great tower wasdemolished. In December 1647 the congrega-tion at nearby St. Mary’s Priory were startledby a crash when ‘ye Tower of ye castle ofMonmouth fell down upon one side whilst wewere at sermon’ (Pye Diary 22 Dec. 1647).This was probably the hall-keep, one side ofwhich is missing. The great hall was retainedas the County Assize court, with accommoda-tion for prisoners, until the Assizes weremoved into Great Castle House around 1700. Henry Somerset still had need for a powercentre of appropriate status in Monmouth,particularly as he was no longer resident inthe county. In the new post-war world, theconflicts within the county elite, fought outduring the Civil War, were now continuedin parliamentary elections, at times with filesof musketeers from Chepstow garrisonbrought in to overawe electors. The answer

was Great Castle House, built within theward of the demolished castle, on theapproximate site of the Great Tower, in 1683. John Newman has described Great CastleHouse as ‘A house of splendid swagger, outsideand in… intended for official and ceremonialpurposes by the Marquesses’ (Newman 2000400-401). Its baroque splendour includes lavishplasterwork and it was clearly (to borrow aphrase of Gibbon) intended for ostentationrather than for use. Newman has suggested thatthe ashlar of its facing was re-used from thegatehouse of the castle, traditionally the birth-place of Henry V. At first-floor level, above thecentrally-placed main door, is an appearancewindow of similar size to the door below,topped by an elaborate shell-shaped pedimentwith three vase-like finials. Like its counterpartsin a number of medieval castles, it was designedfor Somerset to display himself to crowds gath-ered in the castle yard at elections and similarceremonial events. By the time of Coxe’s visitthe house was a girl’s boarding school. Later, it served as the base for the countyMilitia, of which the Duke of Beaufort wasColonel (Kissack 1991). The Royal Monmouth-shire Royal Engineers, the senior regiment ofthe British army reserve, are still based in GreatCastle House, making Monmouth the last Welshcastle still in military occupation (other than forregimental museums). In the nineteenth centurythe Regiment was known locally as ‘The RoyalMonmouthshire Militia (The Pope’s Own)’ dueto the many Roman Catholics in its ranks.


Fig. 3. John Speed’s plan of Monmouth castle,c. 1610. Detail from his map of Wales.


Caerleon CastleThe large motte near the angle of the Romanfortress walls at Caerleon had beenrecognised as medieval since the time of theElizabethan traveller Thomas Churchyard:-

But chiefly for to note/ There is a castlevery old that may not be forgot/ It standethon a forced hill/ Not far from flowingflood. (Churchyard 1776, 24)

The castle occupied the area between thewalls of the Roman legionary fortress and theriver Usk. The motte was sited outside theeastern angle of the fortress, just clear of itsditch. It overlies the ruins of an extra muralbath building, the remains of which werefound under it in an exploratory tunnel, stillvisible, in 1878 (Woollett 1878). The walledarea of the legionary fortress, containing themedieval church and settlement, thus servedas a kind of outer bailey (fig. 4). The road running from the Roman andmedieval river bridge to the south-east gateof the fortress must originally have skirtedthe bailey to pass through the Roman PortaPrincipalis, the main entry to the Romanfortress. Later, it reverted to a direct line andpassed under the castle gatehouse on the lineof the present High Street. The gatehousesurvived until 1800, but during Coxe’s lastvisit the remains were being taken down:- ‘Inthe street leading from the bridge and nearthe passage to the castle, are the remains of aportal, which seems to have formed theentrance of the castle works. Parts of a roundtower still remain, with a groove for aportcullis, and a public house called the GateHouse marks is situation’ (Coxe 1801, 89).Earlier, the road had passed beneath the archof the gateway. In 1849, John Edward Leewrote: ‘It is a singular fact that there arepersons now living at Caerleon who canremember the main street of the town runningunder the gateway of the castle, above whichwas a room large enough to be inhabited’

(Lee 1849, 73-74, 1862, 134). The site ismarked on early editions of the OrdnanceSurvey 25 inch map, as a ‘Roman Gate’.(Bradney 1923, 188). Lee was uncertainwhether the gateway belonged with theRoman walls or the medieval castle, but thepresent High Street is not in the exactposition of the Porta Principalis and theportcullis confirms that this was the medievalcastle gatehouse, not a surviving Romangateway (fig. 5). The south-east angle of the bailey is markedby the early-thirteenth-century HanburyArms Tower, next to the hotel of that name.This, with its long arrow-loops similar tothose in the Garrison Tower of Usk Castle,is usually attributed to William Marshal theElder, following his seizure of Caerleon inthe autumn of 1217, but, Neil Ludlow haspointed out that this is merely a terminus postquem and that an attribution to the youngerMarshals is possible, particularly sinceMarshal would only have had time for onebuilding season before his death in 1219.(Knight 1987, 79-80 and fig. 15. Ludlow2018-19, 236). David Cathcart King, by awholly uncharacteristic error, in a lecture textand relying on the evidence of earlyengravings, suggested that the post-medievalforced opening in the outer face of the towerwas a lateral postern like those at Pembrokeand Caldicot (King 1977, 165). The opening,now with modern blocking, opened on to theadjacent quay of what was, until the buildingof the present Newport bridge, a busy port,as surviving port books show. Until thenineteenth century there was a weekly marketboat from this quay to Bristol and the towerserved for a while as the town’s post office.It was probably breached to provide, alongwith the adjacent two-storey range shown inColt Hoare’s 1800 drawing, storage space forgoods from the quay. A later illustration, ofc. 1838 by J. S. Prout shows the quay still inuse for small boats (figs. 6, 7).6



ABOVE: Fig. 4. Caerleon castle (from Knight 1963).BELOW: Fig. 5. Caerleon - Thomas Morrice’s plan of 1800, showing the areas of the Romanlegionary fortress and the castle. (Coxe 1801, 81).



Fig. 6. Caerleon castle - Hanbury Arms Tower, drawn by T. Tudor for Sir Richard ColtHoare (Coxe 1801, 89), T. Tudor del., W(illiam).B(yrne 1743- 1805) dirext.Fig. 7. Caerleon castle - Hanbury Arms Tower c. 1838 by J. S. Prout. An almostidentical version appears in his ‘Castles and Abbeys of Monmouthshire’ (1838).



From the base of the motte, the foundationsof a twin-towered gatehouse found in the1830s and still visible, and a ‘mass ofmasonry’ serving as the abutment of adrawbridge across the ditch, led to stairs upthe mound to either a shell keep or a GreatTower. Its precise form is unknown. It isusually assumed to have been a shell keep,on the basis of a survey of 1622, recording‘there is a castle w’thin the towne of Carlion,decayed and utterly ruinated’: ‘WilliamThomas of Carlion, gent, holdeth the forsaidcastle of Carlion with a great round hill anda fold thereunto adjoining, together with thecastle bailey and all the lands belonging tothe said round hill or castle’ ( Bradney 1933,197). Another survey, of 1653 refers to the‘castle....decayed and utterly ruinated..…witha round hill, together with the castle baileyand gardens’. (Gwent Archives D 4165/24-copy of 1791). Both surveys are primarilyconcerned with documenting property rightsand though they mention the motte, say littleof any structures on it. If the ‘fold thereuntoadjoining’ refers to the motte top, this mightimply a shell-keep, but Coxe’s descriptionmight suggest something more substantial: ‘In the middle of this century’ (i.e. c. 1750)‘the walls of the tower were not less thanforty foot in height, but they were loosenedby the severe effects of the frost in 1739 andfell down in enormous fragments. Withinthe memory of the oldest inhabitants were.... a flight of stone steps’ (ascending themound). During my last excursion, somemassive foundations were discoveredtowards the summit. The greater part..... wassold to Mr Williams.… who had built ahouse with the materials. The remains whichI observed were not less than twenty feet indepth, ten in breadth, and thirty inheight….of large stones bedded in mortar’(Coxe 1801, 88). Thomas, the son of aCaerleon man, Walter Norman, who died atan advanced age about 1762, told the vicar

of Caerleon that his father, as a boy, used to‘mount the summit of the walls’, from whichhe claimed (rather improbably) that he couldsee the hills of Somerset.7 This mightsuggest the possibility of a circular GreatTower on the motte top, in the manner ofLongtown, as Bryan O’Neil once suggested(O’Neil 1949, 135, but see Ludlow 2018-19,267, n. 37). The motte was landscaped inthe nineteenth century, with a spiralpathway to the top, which was surroundedby a raised walk. It is now in a walled andgated residence and difficult of access.Surveying a CountyHistorically, the most important illustrationsto Coxe’s Tour in Monmouthshire are thesurveyed plans of the towns of the county,showing individual houses and land divisionsin urban landscapes which in many caseshave changed almost beyond recognition inthe intervening two hundred years. Thearchaeological survey plans include a seriesof Iron Age hillforts; twenty castles; the CivilWar earthworks at Penrhos outside Caerleonand the remains of early ironstone mining atCraig Y Garcud (‘the rock of the hidden fort’)near Usk. The castles include threeearthworks, at Walterston (actually inHerefordshire); Twyn Barlwm; and RumneyCastle, along with Skenfrith, Grosmont,White Castle, Chepstow, Caldicot, Newport,Llangibby, Usk, Raglan, Abergavenny andthe ‘Wentwood’ castles of Penhow, Pencoed,Llanvair Discoed and Striguil.Twyn Barlwm. This massive castle moundin its ridgeway setting at a height of 419metres (at ST 242926) above Risca andCwmbran is a landmark over much ofsouthern Gwent, even visible in goodweather from parts of northern Cardiff (Fig.8). The mound is 45m in diameter and 8mhigh. A low mound outside the lip of theditch on the bailey side may mark a bridgeabutment. It stands at one end of a large oval



hilltop enclosure, often identified,particularly from air views, as an Iron Agehillfort. Seen on the ground, the squaresectioned rock cut ditch is quite unlike anyprehistoric work and almost certainlymedieval. It is also clearly unfinished. Notonly are there gaps in the rampart, asMorrice shows, but some sections have onlya marking out bank and heaps of spoil fromditch digging. It is an instructive example ofan unfinished castle earthwork. There is noindication of date, but the remote uplandlocation recalls Gilbert de Clare’s MorlaisCastle of the early 1290s outside MerthyrTydfil. The name Twyn Barlwm ‘The moundof the bubble’ probably refers to theappearance of the large motte, as seen froma distance, but it could almost describe thisephemeral unfinished work in its waterlessuphill location. (Ray Howell in Olding2016, 44-45, 76-77).Newport Castle. The ditches of NewportCastle had been filled with spoil from thedigging of the adjacent Monmouthshire Canalbefore Coxe arrived, but Morrice’s planshows the curtain walls around the threelandward sides before they were demolishedand the area of the bailey lost to urbandevelopment (fig. 9). His plan confirms thatthe impressive three-towered façade along theriver Usk, (where such defences were hardlyneeded), was matched on the landward sidesonly by what one guidebook called ‘acommon wall, without any flank of defence’ Newport castle had been moved from itshilltop location next to St Gwynllwy’s churchto the crossing of the Usk by Hugh D’Audelebetween 1327 and 1347. His daughter marriedRalph, Earl of Stafford (d. 1372), a founderknight of the Order of the Garter, who tooklittle interest in the town. His son, Hugh, Earlof Stafford, founded a house of AugustinianFriars there, granted its burgesses a charterand may have walled the town. He may havecontinued to rebuild the castle, but in 1385,

after the murder of his son, he went onpilgrimage to Jerusalem and died in Rhodes.Thereafter there was a long period whensuccessive heirs were minors and the lordshipwas in the hands of the crown (Knight 1991). This may explain the curiously unfinishedstate of the castle. The exile and death of HughStafford may have halted work on theunfinished castle, followed by franticemergency building in April – May 1405during the Glyndŵr rising, with 36 masonsand with a large body of labourers working onthe ditches (Knight, 1991, 24). Earlier, over1,000 ‘fishes called hakes’ were shipped toNewport and Cardiff ‘for the men-at-arms andarchers there’, (Calendar of Patent RollsHenry IV, 1401-1405 , 296-7), part of a muchwider similar provisioning of Welsh castlesagainst Glyndŵr. Had Stafford lived, a moreconventional and symmetrical fourteenth-century castle might have emerged, with alandward range mirroring the existingriverward façade (Coxe 1801, 49, Morgan1885, Knight 1991), (figs. 10, 11). Whilst many of the other castle plans are,at first glance, similar to those in recentguidebooks, closer scrutiny often showsdetails now vanished or obscured. Thus theplan of Usk Castle shows the South Tower,now reduced to foundations, and usefullydepicts the Outer Ward in the area of thepresent gardens and the outer hornwork tothe north (fig. 12). At Skenfrith the Buck planof 1739 shows a round-headed entrance arch,with the windows of a first-floor chamberabove, the whole on the verge of collapse.Coxe’s plan shows that these had gone, butalso hints that the gap had been widened toallow access for carts or the like to extractbuilding material. Morrice’s plans are insome cases (e.g. White Castle) sufficientlydetailed to show things like draw bar holes,whilst his plans of Llanvair and Striguilcastles are valuable records of their plan andstate before two centuries of neglect.



Fig. 8. Twyn Barlwm, Risca. Survey of the unfinished castle earthworks. Coxe thought the massivemotte could be the burial mound of a Silurian chief, who fell in battle against the Romans. Coxe1801, 75. (T. Morrice survd et del, Harding sc)Fig. 9. Newport Castle 1800. Plan by Thomas Morrice, Coxe 1801, 49. (T. Morrice suvd et del,Harding sc).



Fig. 10. Newport Castle - plan. (Knight 1991)Fig. 11. Newport Castle - the buildings in the east range (Knight 1991, 32)



With Coxe’s Monmouthshire published, ColtHoare and his friends planned a similarvolume on Wiltshire, with Coxe as theintended author. He recommended a youngartist and surveyor employed by the OrdnanceSurvey, Philip Crocker (1780-1840), whoseplans of earthworks in Ancient Wiltshire arethe successors of Thomas Morrice’s of Mon-mouthshire sites. Coxe also introduced aHeytesbury woollen merchant, William Cun-nington (1754-1810) to Colt Hoare. Progress however was slow. Coxe, though hesometimes indulged in barrow digging, was adocumentary historian, not an archaeologist.He once wrote to Owen Pughe enquiring if‘there was anything in the ancient WelshChronicles that might say when Stonehengewas built’. Monmouthshire lacked the wealthof prehistoric sites of Salisbury Plain andwhereas Colt Hoare excavated Wiltshirebarrows ‘in the hope of meeting somethingwhich might supersede conjecture’, Coxe, as a

clergyman, may have been constrained by theBiblical account of Creation, with its implieddate of 4004 B.C.. After something of a crisismeeting at Stourhead, Colt Hoare took over thework. ‘Now we have got the business....out ofthe hands of my friend Coxe’ he wrote toCunnington in 1804 ‘we shall go on ‘...morerapidly’(Cunnington 1975, 63). Coxe continuedto write scholarly works on eighteenth-centuryhistory until his death in 1828. These were translated into several Europeanlanguages and Napoleon took one volume oncampaign with him. Colt Hoare’s TheAncient History of South Wiltshire waspublished in 1812, with a companion volumefor North Wiltshire in 1821. These were verydifferent books to Coxe’s antiquarian tour ofMonmouthshire. Colt Hoare’s programme ofbarrow excavations with Cunnington laid,as Richard Atkinson put it ‘the veryfoundation of British prehistory’(Cunnington1975, xiv).

Fig. 12. Usk Castle, survey by Thomas Morrice: Coxe 1801, 49. (T. Morrice suv’d et del. Harding



BibliographyBradney, J. A. 1923, A History of Monmouth-shire 3, part 2 Hundred of Usk______________ 1933, A History of Monmouth-shire 4, part 1 Hundred of CaldicotChurchyard, T., 1776 The Worthines of WalesA Poem. A true note of the auncient castles,famous monuments ..... fine townes andcourteous people that I have met in the noblecountriese of WALES (London 1588, reprintedThomas Evans, London)Coxe, W., 1801, William Coxe An HistoricalTour In Monmouthshire (Cadell and Davies,London, 1801). New edition, with introductionby J. K. Knight, Merton Priory Press, Cardiff1995. Coxe D.N.BC.S.P.D. Calendar of State Papers DomesticCunnington, R. H., 1975, From Antiquary toArchaeologist; A Biography of WilliamCunnington 1754-1810 (Shire Publications,Aylesbury)Durant, H., 1973, Henry, 1st Duke of Beaufortand his Duchess Mary (Pontypool, Hughes andSon).Geeare, G., Priestly, S., and Turner, R., 2006,‘After the Restoration’ Turner and Johnson (eds.)229-242Harris, W., 1773, ‘Observations on the JuliaStrata and on the Roman Stations, Forts andCamps in the Counties of Monmouth,Brecknock, Carmarthen and Glamorgan’Archaeologia 2, 1- 24.King, D. J. C. 1977, ‘Pembroke Castle:derivation and relationships of the domedvault of the donjon, and of the Horseshoe Gate’Château Gaillard 8, 159-169Kissack, K, 1991, ‘Life in the MonmouthMilitia, 1778-1812’ Monmouthshire Antiquary7, 71-82.Knight, J. K. 1963 ‘The Keep of Caerleon Castle’Monmouthshire Antiquary 1,part 3, 71-72_________,1987 ‘The Road to Harlech: Aspectsof some early thirteenth century Welsh castles’Castles in Wales and the Marches: Essays in

Honour of D. J. Cathcart King ed. John Kenyonand Richard Avent (Cardiff, University of WalesPress), 75-8___________,1991 ‘Newport Castle’ Mon-mouthshire Antiquary 7, 17-42.___________, 2005 Civil War and Restorationin Monmouthshire (Logaston Press)___________, 2006 ‘Civil War and Common-wealth’ Turner and Johnson (eds) 221-228;Lee, J. E.,1849,‘Roman remains lately found atCaerleon’ Archaeologia Cambrensis IV, 73-82__________,1862 Isca Silurum (London ).Lewis, J. M., 1999,The Medieval Tiles of Wales(National Museums and Collections of Wales,Cardiff)Ludlow, N., 2018-19 ‘William Marshal,Pembroke Castle and Angevin design’Castle Studies Group Journal 32, 209-92.Morgan, O., 1885 ‘Newport Castle’Archaeologia Cambrensis 5th series, 2, 70-79.Newman, J., 2000, The Buildings of WalesGwent/Monmouthshire (Penguin / University ofWales Press.Olding, F., 2016 (ed.) Archaeoleg UcheldirGwent. The Archaeology of Upland Gwent RoyalCommission on the Ancient and HistoricalMonuments of Wales (Aberystwyth)O’Neil B.,1949 ‘Castles’ in V. E. Nash Williams(ed) A Hundred Years of Welsh Archaeology(Cambrian Archaeological Association)Plumb, J. H., 1966, Men and Places Penguin,Harmondsworth)Pye, M, 1899, ‘Diary for 1647’, printed W. M.Warlow The History of the Charities of WilliamJones (Bristol), 92-99Redknap, M., 1996, A ‘Pocket Dag’ from ChepstowMonmouthshire Antiquary 12, 56-61Thompson, M. W., 1983, The Journeys of SirRichard Colt Hoare through Wales and England1793 - 1810 ed. M. W. Thompson (Alan Sutton,Gloucester)Turner, R., 2000, Chepstow Castle (Cadw,Cardiff).



Pembroke Castle and Angevin design’

Turner, R., and Johnson, A., 2006 ChepstowCastle: Its History and Buildings (LogastonPress)Woollett, R. F. 1878, Notes on the CastleMound, Caerleon and the recent Excavations(Newport reference library Ms Pf M260 900W00 Monmouthshire Collection)

Notes1. The correspondence of Coxe and William

Owen Pughe, 1799-1802 is in Aberyst-wyth, National Library of Wales, Mss13222C and 13224B

2. J. H. Plumb bracketed Coxe with Macaulayand Samuel Johnson as the fathers ofmodern biography and wrote ‘the works ofArchdeacon Coxe, a far too neglectedhistorian, mark the beginning of modernhistorical biography’ Plumb ‘ThomasBabington Macaulay’ in 1966, 284.

3. At the time, Iron Age hillforts weregenerally thought to be Roman, Saxon or‘Danish’. Harris considered them Romanmilitary works, but Coxe recognised thatsome at least were earlier.

4. The writer was assured more than once byfarmers in upland Gwent that the motte ontheir land was the burial place of a ‘Romanlegion’.

5. Warrant to pay Lord Herbert of Raglan £500on account of repairs to Chepstow Castle.Some of these works may date from theFirst Civil War. In 1662 fifteen iron cannonwere brought from Shrewsbury, but werereturned there the following year (C.S.P.D1661-2, 425).

6. October 1662 - Society of AntiquariesLibrary, London Ms 390/1 (Wakeman Collec-tion) March 1664 -National Library of Wales,Bradney Papers, printed Bradney 1933, 17.Alan Aberg discovered the Prout engravingin the Library of the Society of Antiquaries. Iam deeply grateful to Julian Mitchell foridentifying the artist and informing me of anear identical version in Prout’s Castles andAbbeys of Monmouthshire (1838).

7. For the Norman family see J. A. BradneyHistory of Monmouthshire 3.2 Hundred ofUsk, 1923, 204. They were the ancestors ofSir Arthur Evans, excavator of Knossos.

Illustrations1. Walterstone, Herefordshire Plan of themotte and bailey, surveyed by Thomas Morrice.(T. Morris del et surv’d, Harding sc.). CoxeHistorical Tour in Monmouthshire (1801). Coxethought the squarish bailey might be Roman(Coxe 1801, 23)2. Chepstow Castle - 1777. Paul Sandby.Lower bailey with a view toward the maingatehouse. At the time being used as a farmyardwith (left) the glass factory and malthousebuildings south of the Great Hall, from whichextends the elaborate chimney.3. Caerleon Castle (from Knight 1963)4. Caerleon - Thomas Morrice’s plan of 1800,showing the areas of the Roman legionaryfortress and the castle. (Coxe 1801, 81).5. Caerleon Castle - Hanbury Arms Tower,drawn by T. Tudor for Sir Richard Colt Hoare(Coxe 1801, 89), T Tudor del., W(illiam). B(yrne1743- 1805) (dirext).6. Caerleon Castle - Hanbury Arms Towerc. 1838 by J. S. Prout. An almost identicalversion appears in his Castles and Abbeys ofMonmouthshire (1838).7. Twyn Barlwm, Risca. Survey of theunfinished castle earthworks. Coxe thought themassive motte could be the burial mound of aSilurian chief, who fell in battle against theRomans. Coxe 1801, 75. (T. Morrice surv’d etdel, Harding sc).8. Newport Castle 1800. Plan by ThomasMorrice, Coxe 1801, 49. ( T Morrice surv’d etdel, Harding sc).9. Newport Castle - plan. (Knight 1991).10. Newport Castle - the buildings in the eastrange (Knight 1991, 32).11. Usk Castle, survey by Thomas Morrice Coxe1801, 49. ( T.Morrice suv’d et del. Harding sc )Surv’d - surveyed; del - drawn by (delineated);dirext - directed (as head of workshop) Sc-engraved.


an accidental excursion into monmouthshire journal 2019-2020rev7...bemerton, wiltshire, (the poet...

Documents