Aber Castle

It is perilous to precisely date any medieval structure, with or without archaeological or documentary evidence.  This is because it is difficult to be sure what structures documents refer to and even what exactly archaeological evidence actually signifies.  This is as true for castles as for llysoedd.  In each case it is still a matter of weighing possibilities and deciding likelihoods.  However, this does not mean that such dating should not be attempted, merely that it should not be set in stone, especially while much of the evidence is yet to be evaluated.

At Aber we have a still partially ditched motte roughly 120 feet in diameter and a little over twenty feet high.  Such mottes literally abound throughout the British Isles.  The summit is approximately fifty feet in diameter and shows clear signs of once having supported either a small shell keep or large, probably round, tower.  The motte was almost certainly surrounded by a ditch as was common practice.  This is most noticeable to the south, although the drop in height to the houses on the site of the bailey now marks its position elsewhere.  The motte appears to have been surrounded by an eye-shaped bailey approximately 550 feet north to south by 350 feet east to west at its maximum extent.  This bailey itself would appear to have been divided roughly in half.  The dividing ditch survives mostly to the south, while to the north it may be discernable in the aerial photograph.  Most of the bailey defences to the north and west have been obliterated by later houses, but the distinct drop in height strongly suggests the line of the northern bailey.  The bailey defences to the south and east have also been largely erased, possibly by ploughing, but more likely by the deliberate destruction of the rampart which was probably used to fill the ditch.  Such degradation of the military earthworks of the site should make us very careful of the modern suggestion that this was the site of the hall complex of the later princes of Wales.

Within the northern bailey excavation has uncovered a structure which has been ‘identified as the llys or princely court recorded here through the thirteenth century'.  There are many problems with this identification and it would appear - certainly no evidence has been advanced that any proper historical research has been undertaken and certainly none worthy of the name has been published - that this assertion has been made without adequate historical research or taking llys and castle sites in context.

The building alleged to be a royal llys was uncovered twice, once in 1993 and once in 2010.  This structure was initially claimed to be approximately 37 feet east to west by 26 feet north to south internally, with walls some 2½ feet thick.  Projecting chambers, each about twenty feet by thirty feet, have been claimed by the excavators as additions built on both sides of ‘the hall'.  It is therefore necessary to examine the remains to see if they justify such an interpretation.

The aerial photographs of the dig site - and a short personal inspection - would suggest the site has a complex history which archaeology shows continued into the nineteenth century. From the excavation reports we can judge that we are not yet anywhere near to fully understanding the ‘modern' history of the castle site after the Middle Ages.

Firstly, a few things have to be said of the reports that have appeared and which can be summed up in the last government published report.  These contain many claims, but no historical research, while the few solid facts that they do contain are used in juxtaposition with dubious identifications to ‘prove' their cause.  Such a report even resorts to ‘straw man' arguments, viz: ‘Some believe that... Pen y Bryn... [was the llys] quoting evidence from place names, antiquarian writers, local tradition and interpretation...'.  Such misrepresentation of the facts from original thirteenth and fourteenth century documentation does no real favours to history and is an abomination from a government sponsored body.  However, using this final ‘preliminary report' as a basis for what has been uncovered in the castle bailey, it is possible to make the following observations.  The south-western corner of a later ‘masonry building' has clearly penetrated the wall of the claimed long house of the Welsh princes.  For some reason the excavators make this penetration out to be an original doorway of their hall of the princes.  If this was a doorway - of which there seems no evidence - then why was a later building apparently ‘of the fourteenth century' built into it?  What was this later ‘building' that shows evidence of metal working going on within it?  Was it a building or a simple corral wall around an industrial complex approximately 60 feet by 50, built after the demolition of the alleged hall.  Further, why is there no historical evidence of this change of use of the site of a royal palace into an industrial complex as has been uncovered by excavation, unless of course this is not the site referred to in any of the documentation?  Remember that the capital messuage, or caput of Aber estate, as Garth Celyn was, is mentioned as a functioning estate down to 1417 and is still being granted by the Crown as such as late as 1485 and Pen y Bryn is mentioned as the capital messuage of the estate of the Thomas family during the sixteenth century.

Of the claimed ‘hall' itself, understanding its north-eastern section is even more problematic due to the denuded nature of the remains, which are even worse at this corner.  The northern section of the primary building has been almost totally obliterated with the north-eastern corner totally lost.  The junction of the north-west corner with the claimed ‘north wing' is not clear, but the better quality mortar in the ‘wing' wall would suggest that it abuts onto the primary chamber wall.  The rest of the so called ‘north wing' appears to be illusory, but more will be said of this later.

Two entrances have been claimed into the ‘hall house'.  The first, to the east, is less than three feet wide and consists of a simple break in the wall without a doorstep.  As this is overshadowed by the claimed ‘wing' to the south and a later wall which abuts to the north, it probably was an entrance of a very poor kind.  It would also appear to have been covered by an outbuilding or porch judging by the remains.  This is hardly the great porchless ceremonial entrance which appears on the imaginative reconstruction of the surprisingly misnamed ‘Ty'n y Mwd' hall.  The second ‘entrance' to the west, which we have already examined above, is even more imaginative.  This is simply a gash carved through the wall where the later, apparently industrial complex was cut through what therefore appears to be an abandoned primary building.  If the building was occupied and of a high status it would not have been pierced by such a lowly structure.

The building claimed as the southern ‘wing' of the ‘hall' is 35 feet by 16 feet internally.  The foundations of this chamber are mostly intact, although much of the east wall has gone.  It appears to all be of one build except for a later external buttress added roughly half way down the southern wall.  That the eastern wall of the ‘hall' penetrates the northern wall would suggest that it post dates this structure.  However, it could just be a change in building plan that happened virtually contemporaneously with the building of the ‘hall'.  A ‘bronze ring brooch... of thirteenth to fourteenth century date' was recovered ‘from the interface of the old ground surface within the south wing of the building'.  This would ‘suggest' that the brooch was lost after the building was abandoned and before much soil built up.  It is hardly satisfactory dating material and if anything points to a pre fourteenth century date for the abandonment of this structure.  If this assumption is correct, it shows that this could not have been the palace of the princes as these were still in occupation in the fifteenth century.

The northern part of the excavation site shows at least four or five phases and has obviously had a more complex history than the two structures to the south, identified by Gwynedd Archaeological Trust (GAT) as a hall and its later wing.  This northern section is also the best preserved part of the masonry and the thickest, with the wall approaching six feet thick.  In front of the northernmost wall was a ditch which was not fully explored by the excavators.  This is a shame as it would appear to have been the ditch dividing the northern bailey from the southern one, which would make the northern wall of the excavated complex the curtain wall of the southern bailey of the castle.  This purported curtain wall would appear to have been rebuilt with a new, narrower wall topping the remains of the earlier one, of which only the northern front can now be seen.

The northern ‘wing' of the alleged palace seems more to have been drawn with the eye of faith rather than from evidence on the ground and if there is an eastern return wall it would appear to be west of the eastern wall of the primary chamber.  In other words this is hardly a wing.  Further east from the northern ‘wing' are the remains of what appears to be a long narrow building which partially underlies the secondary ‘curtain wall'.  This structure, and the claimed north ‘wing' were all said to have been built with lime mortar.  The rest of the masonry uncovered was said to be clay or earth mortared walls.

To the west of the southern half of the main excavated structures just described is a large rectangular enclosure that has already been mentioned as its foundations have pierced and obliterated a portion of the west wall of the ‘hall'.  This structure has slightly thicker walls, that are not as well constructed as the walls of the south ‘wing'.  It is approximately 55 feet east to west by sixty feet north to south externally.  Excavation shows that it contained at least six pits as well as burnt soil.  As such it would appear to have been an industrial site which postdates the ‘hall' to the east which has been claimed by GAT as Llywelyn's llys.

The official government agency summary of the site is twofold and date from 2007.  Under the heading: The Llys at Aber, House Excavated at Pen y Mwd comes:

The llys or princely court at Aber was one of the principal residences of the princes of Gwynedd through the thirteenth century.  Repairs are recorded in 1289 and 1303, following the English conquest.  Remains were still visible in the early sixteenth century.  Excavations in 1993 recovered the plan of a hall with crosswings at either end, associated with thirteenth and fourteenth century material.  The hall was 11.2m by 8.0m internally.  The site lies close by the foot of a castle mound (NPRN 95692).  There are several other instances in north Wales of apparently unfortified houses associated with castle mounts, for example Castell Prysor (NPRN 308964), Crogen (NPRN 306558) and Rug (NPRN 306598).  In these cases the mount can be regarded as an adjunct to the house, conferring a certain status and arguably furnishing a refuge.

A second entry has: Aber Castle; Pen-y-Mwd Mound:

A mound thought to be a medieval castle mount associated with a medieval mansion excavated at its foot (NPRN 309171).  This is a sub-circular steep sided mound, roughly 36m in diameter and 6.6m high.  It has a level summit about 17m by 14m.  There are traces of a ditch on the south side, but no further defensive features have been identified.  The mansion was excavated in the field to the south-east.  It produced thirteenth and fourteenth century material and is identified as the llys or princely court recorded here through the thirteenth century.  There are several instances in north Wales of castle mounts associated with apparently unfortified houses, for example Castle Prysor (NPRN 308964), Crogen (NPRN 306558) and Rug (NPRN 306598).  It is possible that the mount was raised as an adjunct to the court.

In the official mind there is no doubt that the buildings uncovered by excavation in the castle bailey are those of the llys of the princes of Gwynedd, which is rather unfortunate as this case is devoid of clear documentary evidence.  Indeed what original evidence there is (medieval royal documents from the 13th to the 15th century) tends far more to point towards the extensive site at Pen y Bryn on the Garth (roughly 250 feet square), rather than to the site in one of the baileys of Aber castle (a roughly triangular shape roughly 150 feet by 120).  Most llys sites, Llys Gwenllian, Rhosyr and the tentative site at Aberffraw, all seem to be rectangular and in the region of between 200 and 250 feet long.

What we certainly have in the southern castle bailey at Aber are the remains of what is a series of structures quite unlike those excavated at Rhosyr llys and apparently unlike any of the remains found at other houses of the thirteenth and fourteenth centuries.  Indeed the only ‘Caernarfonshire halls of the fourteenth century' which looked even remotely like the Aber bailey site was a debatable reconstruction plan of Penrhyn (above right).  The current walls all lie within the southern bailey of Aber castle, stand no more than a course or two high, and show no sign of mortar other than poor leached remains seen to the north.  What is left bears no resemblance to a ‘high status building'.  It appears more like the jumble of buildings that would be expected in a castle bailey.  It should also be noted that these remains are inferior to the possibly thirteenth to eighteenth century hafod buildings uncovered by excavation in 1961.  It is further quite clear that the poor quality of the remains is not solely due to stone robbing.  These structures were never of any great standing and probably were only one storey high, judging by the thinness of the walls and the paucity of the ‘mortar'.

Apparently the pottery remains at Aber would suggest a thirteenth to fourteenth century usage for the buildings.  Yet again the amount of pottery and coins found on such a small site is extraordinary, especially as royal sites were always well maintained and kept scrupulously clean.  For example, the fourteenth century details for the cleaning of the royal castle of Berkhamsted have survived and it would seem unlikely that other royal sites would have been allowed to become so unkempt as the alleged palace in the bailey of Aber castle is said to have been.  Thus we find in 1351 the porter of Berkhamsted castle was allowed all the litter found within the castle buildings whenever they were cleaned, which appeared to be a yearly business.  Although masses of such minutia have not survived from most habitations, the cleanliness of royal sites when excavated shows that such agreements were widespread.  Indeed, even the baronial castle of Hen Domen at Montgomery was kept so clean during its two hundred odd years of occupation that the excavators were appalled by the lack of dateable evidence found.  This therefore adds to the impression that Aber castle was not the royal house used by Edward I and II and their Welsh predecessors.

It is a shame that excavation did not take place on the motte which would have shown if the masonry of the keep was similar to that uncovered in the bailey.  If it had been we may have been able to tell if the whole structure had been revamped after its destruction in 1094 when all the castles of Gwynedd were certainly destroyed.  This might have told us a great deal about the site and the dates of its occupation.  A small dig upon the motte may well still show us the life span of the motte and bailey castle.

It has been asserted that the foundations uncovered in Aber castle bailey can be related to the rebuildings carried out for Prince Edward in the early fourteenth century and that antiquarians often state that this was the site of the llys.  Neither of these arguments stand up to serious consideration.  The best preserved part of the structure is to the north where one ‘wall' has been overlain by several large river-worn boulders.  The whole could be little more than sleeper walls for a wooden structure.  Certainly to describe the foundations as they appear as a mansion or royal hall seems rather grand and the reconstruction drawing of ‘the castle' as it was said to have been in the early fourteenth century is positively misleading, especially when compared to the one drawn for Rhosyr.  In the reconstruction at Aber the petty east entrance into the primary building has lost its porch, while the low foundations which have the appearance of sleeper walls have been imagined into a two storey structure which positively dwarfs the motte and ignores the industrial compound to the rear as well as the wall and ditch between it and the motte.

It is worth noting that Aber castle motte and bailey stands immediately west of the fast running Afon Aber, just at the place where the river valley widens out into the coastal plain.  It therefore controls the river crossing and is in a lowland position.  It should again be emphasised that it is a well recognised general principal that Welsh castles tend to dominate the highlands and Norman castles the lowlands, although both sides on occasions used the others' fortresses.  It can therefore be seen that the remains uncovered by excavation at this site are in accordance with what has been found and is expected at other Norman sites, but they do not meet with the criteria found at other llys sites, viz Rhosyr.


Copyright©2016 Paul Martin Remfry

  • Index

  • Home