2007:838 - Mullamast, Kildare

This site was excavated as part of the N9/N10 Kilcullen to Waterford scheme, Phase 3: Kilcullen to Carlow. This site was identified as having significant archaeological potential in the environmental impact statement because of its proximity to a suspected castle site (KD036–018). This building is shown on the Down Survey map of 1650 as being to the west of Prospect House, the farm immediately to the east of the road-take zone at the southern end of the site, and it is also shown on a subsequent map of the 1740s. An article by Fitzgerald in the Journal of the Kildare Archaeological Society for 1906 mentions that stones taken from the castle site, from the ‘old town’ field immediately to the west of the farm, were used in the building of part of Prospect House, although further supporting documentation about it is hard to find. Prospect House first appears on the Francis map of 1783.
A geophysical survey of 3.16ha within the road-take zone and beyond it to the east and west, undertaken in 2002 by the Bartlett Clark Consultancy, revealed a number of features of probable archaeological origin and significance. Aerial photographs of the route taken by the late Markus Casey on behalf of the National Roads Authority before the commencement of excavation also show earthworks and cropmarks in these fields and the surrounding ones.
Subsequent testing in the form of trial-trenching was carried out by the Irish Archaeological Consultancy Ltd. This revealed ‘a number of features including pits, stone features and linear features spread over the two fields identified prior to testing as A10 and AAP 15. The majority of features were concentrated on the higher and more level southern field, the northern field perhaps representing the periphery of the settlement’.
The site lay on the north-facing slope of a low hill rising to c. 140m above OD. Approximately 1km to the west, a large ringfort, the Rath of Mullamast, lies on the eastern escarpment of the broad flat valley of the River Barrow. This site and the rath are easily visible from each other and the rath was a regional meeting-centre for the local Gaelic tribes in the early medieval period before the Anglo-Norman invasions and it continued to function as such at least until the 16th century.
Bedrock, comprising sandstone above limestone in this part of Kildare, is capped by a thick deposit of glacial till in the form of very mixed deposits of clay, sand and gravel with many erratics. The site sloped down sharply to the north in the northernmost field and a series of springs on this slope fed the modern streams at the bottom of it. These two streams have been straightened in modern times to form field boundaries and they flow downhill to the east towards the River Greese. The more meandering former courses of the streams were clearly visible. They were found to be 4m wide and 1–1.5m deep, containing organic peaty deposits, tufaceous deposits and a small quantity of medieval pottery when sections were cut across them.
Topsoil-stripping at the beginning of the resolution phase of work in April 2007 suggested that archaeological features continued to the south, beyond the modern field boundary originally chosen to define the site’s southern limit. The site was therefore extended to the south into parts of the next two modern fields for a further c. 160m. An area alongside the modern stream which needed future culverting was also included in the site. The total area stripped was 21,660m2, incorporating the extra 5000m2 at the southern end. The main part of the site comprised the full 45m width of the road-take over a distance of c. 470m.
Full excavation began in late April 2007 and was completed with a large site crew in mid-December. Post-excavation assessment and analysis are therefore at a very early stage at the time of writing in January 2008. This report is intended to give a broad picture of what was found of the remains recorded in more than 2000 context descriptions, with more than 5000 associated finds. Many of the observations herein are necessarily provisional.
The majority of the archaeological remains from the site dated from the medieval period but a fulacht fiadh was found in the north-east corner of the site on the southern side of the modern stream. This burnt mound continued beyond the edge to the road-take zone to the south and had been truncated on the eastern side by the modern field boundary marking the end of the site. An c. 12m east to west by 6m width of burnt deposits was recorded. The mound survived as a typical example of this kind of monument, with a deposit of ash, charcoal and burnt stones up to 1m thick. Two troughs were found beneath the burnt spreads. The first of these was shallow (,0.4m deep), circular (,0.7m wide) and unlined, whilst the larger was lined with slabs of green-tinged metamorphosed siltstone and lay immediately under the fence-line marking the edge of the CPO zone of the road scheme and beyond it. This was not excavated and was only recorded as far as practically possible.
An unusual aspect of the mound was that it had apparently been cut through at some point in a straight line running diagonally across the site for c. 8m to a depth of c. 0.2–0.3m into the subsoil level. A series of flat rocks had been laid at the base of this cut resembling stepping-stones. These were subsequently covered over by a redeposition of some of the black burnt deposits. A small sample of animal bone was recovered from the mound, including a large pig or boar tusk, apparently pierced, possibly for use as a kind of ornament. Although they were within the bounds of the site, these features formed a separate group at some distance from the others.
A few other residual prehistoric finds were also recovered from elsewhere on the site, including Bronze Age pottery sherds from an amorphous pit at the southern end, and a few struck flints, including two broken arrowheads. Part of a sandstone quern was also recovered from an undated and isolated pit at the southern end.
The greater part of the site was dated by a large assemblage of over 5000 sherds of medieval pottery. This was found scattered over the whole site in positive features, cut features, horizontal layers and layers disturbed by later agricultural activity. There were substantial quantities of Leinster cooking ware but also a noticeably high proportion of glazed wares, including Dublin wares, south-western English wares, such as Ham Green and Bristol wares, and a few fragments of wares imported from France. The date range of this material appears to lie between the late 12th and early 14th centuries.
Several hundred metal objects were also recovered, including brooch fragments and other parts of various pieces of clothing decoration, knife fragments, nails, a horseshoe, a box key and two metal arrowheads, possibly crossbow bolts. Other household objects such as spindle-whorls and loom weights were found, as well as a moderate-sized assemblage of animal bones.
A silver long-cross penny dating to the reign of King Edward the First was found in the upper fill of a well in the south centre part of the site. This was minted in the city of Canterbury in England between 1294 and 1299 and may suggest a date towards the end of the period of occupation of the site. Another more unusual coin was found on site and appears to be a heavily corroded Roman sestertius, of third-century date. (The letters CON . . . AUG . . . were faintly visible when it was discovered but this will need to be confirmed following its conservation.) This was found in the disturbed stone foundations of a medieval wall in association with some large medieval pottery fragments. Otherwise, coinage was notable by its absence.
The site appears to have been a medieval village. Its basic layout, although modified several times during its existence, seems to have been planned from the outset and to have affected the subsequent topography of the immediate area well after its abandonment.
The bounds of the occupied part of the settlement seem to have been established by two east–west-oriented V-profiled ditches. The first of these lay near the point where the slope down to the north suddenly becomes steeper and continued beyond both edges of the CPO zone. The second lay c. 240m to the south and ran across the CPO zone for most of its width before turning a right angle to the north close to the eastern boundary fence. These were substantial ditches which were 1.5–2.5m wide and 1–1.5m deep as recorded but would probably have been larger when in use taking into account topsoil cover beside them and upcast banking.
Midway between these ditches in the west centre of the site a roughly square open area c. 20–25m across appeared to have lain in the centre of the settlement without ever having been built upon. Clear ploughmarks cut across the area in groups of parallel lines at right angles to each other showing that, although the site was used exclusively for pasture at the time of excavation, it had been ploughed in the past. The resulting topsoil contained a scatter of archaeological finds in the form of pottery, bones and rocks across the whole of the two northern fields, often at the interface between the topsoil and the underlying subsoil and archaeological deposits. The two fields to the south had been bulldozed within living memory and upcast stone had been gathered up and dumped, so that archaeological remains there tended to survive as cut features, with conspicuously less rocks apparent in them.
To the north of the open space, a hard, slightly cambered surface lay on the interface with the glacial till, with side ditches to the east and west. This ran from south-west to north-east, with the side ditches converging to the north towards the point where the modern gate into the next field stood. Overlying layers contained a few loose disturbed rocks and finds and the area had apparently not been built over in medieval times, except for some hard-packed patches of gravel metalling at the southern edges. This feature continued to the north to join with the northern boundary ditch close to the eastern edge of the CPO zone, where subsequent modifications may have represented the remains of a rudimentary bridge or crossing point.
When the settlement was laid out, a similar arrangement seems to have been made on the eastern and southern sides of the central open space. On the eastern side the cambered surface was flanked by ditches c. 8m apart converging slightly to the east and c. 5m apart to the north. As elsewhere on the site, many of these ditches had very definite starting points. The full arrangement appears to represent a village green-type space in the centre of the settlement at a crossroads with roadways with flanking drainage ditches approaching it.
As the ditches of the roadway running downhill from the south would have tended to flood the green, at some point in the occupation of the settlement this access route seems to have been blocked with a dense metalled and cobbled surface covering the ditches. A much larger ditch to the east may have been dug at much the same time, which linked in with the roadway to the east and drained the central area in that direction. The western side of the green appears to have been subsequently developed and any roadway to the west may also have been rerouted, as appears possible from the aerial photographs.
If this settlement dates from the late 12th century, it is likely to have been planted by the Anglo-Norman knight Walter de Riddlesford, whose holding of this part of south Kildare was confirmed by the Treaty of Windsor in 1175. Walter was from southern Wales and had married an illegitimate granddaughter of King Henry I. He was thus related to Anglo-Norman royalty and was one of the first wave of military invaders. If he had followed the typical English or Welsh pattern of the time he might have laid his settlements out in the form of tofts, the places where houses stood or had stood, adjoining crofts, pieces of land enclosed, usually, by boundary ditches and oriented along access routes in the form of roads or pathways. This seems to have been the procedure followed at this settlement.
The roadway in the northern part of the site crosses it obliquely so that there were triangular areas marked out by the road on the east and west sides of it within the CPO zone, with the east–west ditch forming the northern boundary. A complex and evolving series of ditches seems to have marked out the tofts and crofts. There appear to have been ditched enclosures marked out at the northern end of the settlement backing on to the boundary ditch, with building plots between them generally slightly upslope on the flatter areas. On the eastern side of the road at the northern end at least two phases of building took place, whilst on the western side there would have been at least three.
The ditches showed much variation. The initial phases appear to have been straight but the modifications were often curved, especially at the corners of enclosures where they joined others. They were generally 1–2m wide and 1–1.5m deep. Often the secondary ditches within the basic layout were begun with a sloping end, measuring 1–2m long, joining a much longer ditch, sometimes over 20m long, at right angles. These ditches often contained portable finds, such as pottery fragments, metal objects, bones and occasionally charred organic remains. The sloping ends of the ditches are likely to have been close to the actual buildings.
Within the plots, buildings appear in the first instance to have been formed from clay and timber, wattle and daub or possibly sod structures, with redeposited or beaten clay floor surfaces. A few features have been tentatively identified as post-holes, stake-holes and sill-beams but the evidence they supply of structures is fairly scrappy. It is likely that the majority of the early buildings stood on above-ground foundations, possibly timber frames on pad-stones, but, on the east side of the northern road, a beaten clay floor for one phase of building was superimposed over a foundation trench with packing stones for another on a different alignment.
On the western side of the northern road the buildings were extended physically over the first roadside ditch and the 1.2m-wide stone-packed trench foundation running parallel to it at the southern end could have supported a very substantial structure. This foundation was linked with another with an apparently curved foundation and post-dated the ditch. Inside, an occupation surface could be recognised over a redeposited clay floor. This clay floor in turn sealed an earlier phase of occupation represented by a group of post- and stake-holes. At the southern end of the curved structure, rammed metalling from the road ran up to a cobbled gap c. 1.5m wide, possibly representing an entranceway into the main part of the structure, which would have lain to the west of the western edge of the CPO zone. To the south, a further foundation represented by a double line of rocks parallel to the original roadside ditch continued for a further 8m to the edge of the trench. The curved foundation had a similar-sized maximum diameter as seen, c. 8m, and had secondary stone-based subdivisions within it.
The flimsy clay-and-timber buildings may have been rebuilt in stone, possibly in the second half of the 13th century. Large quantities of stone were found in the backfill of some of the ditches but not generally in situ. There were a few places where rocks appeared to form packing stones for posts, usually close to the edges of ditches. It is likely that much of the stone had slumped into position following the abandonment of parts of the settlement, as, where there was evidence of building in situ, it was often very close to the edges of a trench. Green-tinged Ordovician rocks with flat cleavage planes from the southern Wicklow Mountains seem to have been brought in for the more decorative work, around hearths and suchlike features. There was no evidence of mortar being used in any part of the site, suggesting that any buildings may have been made using a drystone walling technique, possibly combined with other materials, such as timber or clay.
Within the ditches themselves, parts of the sides had been revetted with rocks and stones, sometimes along the edges of the original cuts and sometimes as an aspect of a secondary backfilling. Some of these may have functioned as small-scale bridging abutments across the ditches.
South of the central open area, the original roadway was blocked, or at least covered over with stone metalling, above a layer of redeposited clay, which in turn sealed two parallel linear ditches, discussed above. A deep ditch, c. 2m wide and 1–1.5 deep, was cut to the east of, and parallel to, its former line, with two similar-sized ditches running off it to the east, roughly at right angles. The ditch parallel to the roadway had a 3.4m-long deliberate backfill put in soon after its construction creating a crossing point or causeway at the southern end of the excavated part. The north–south ditch was excavated for c. 75m, with the ditches at right angles being excavated for c. 15m and 35m respectively. These latter ditches lay c. 45m apart.
When combined with the southern boundary ditch, two large compounds were marked out by these ditches. The basic initial layout may have been roughly similar, with a metalled access road or path on the west side between the ditch and the buildings and individual plots marked out within them by further ditches; much of the planning seems to have involved marking out plots roughly 5m square, later extended. This simplistic scheme is more applicable as an interpretation of the southern compound than the central one, which was radically realigned on at least one occasion.
The southern compound measured c. 70m from north to south and involved a series of buildings fronting on to the edge of the metalling beside the ditch marking its western edge. These buildings were set at the western ends of a series of ditched enclosures, with an alleyway behind them and a matching series of workshop areas set back across the lane. To the east a series of parallel ditches marking out the crofts crossed the site to join with the eastern arm of the southern boundary ditch near the eastern CPO zone fence. A large hearth or kiln was excavated in one of the rear workshops and another one had been truncated by the modern field boundary ditch and land drain beside it. The north central area of the southern compound was conspicuously empty of large ditches and instead had a series of five parallel shallow linear features that were likely to have been cultivation furrows. The area immediately to the north and north-west of the buildings had numerous intercutting linear features, most of which appeared to be attempts at improving drainage. Three deep pits were excavated in this area to depths of c. 2m, which corresponded to the modern water-table, and may have done so in medieval times; these were probably wells; another two were excavated in the northern part of the site. This part of the site was complicated by further curving ditches marking out other enclosures, across the modern boundary ditch to the south and linked to the medieval boundary ditch for the settlement.
The central compound was similar in some respects to the southern one but was apparently even more concerned with industrial activities, probably with agricultural applications, with a particularly complicated realigned, partially stone-revetted, ditch arrangement involving a central water-filled pool, probably to deal with outbreaks of fire. Some of the more conspicuous features in this area included a keyhole-shaped corn-drying kiln, with wind-breaker and possible clamp; a stone-lined pit in a ditched enclosure with an extensive associated spread of charcoal, ash, post- and stake-holes; a deep enclosure ditch filled with a black deposit of charcoal and ash; and a large hearth arrangement, standing separately in its own enclosure. This latter group of features included a roughly square-shaped cut with rounded corners, which was c. 3m in diameter. It lay directly beside an associated rectangular stone foundation c. 2m long by 1m wide. The cut contained a sequence of ash and charcoal layers c. 0.3m deep overlying the impression of what is thought to have been a metal structure. The evidence for this was a series of six parallel lines impressed down into the natural clay in an area roughly 1.2m square. The stone rectangle appears to have been a flat base for something heavy, possibly a stone water trough. It lay over evidence for more burning associated with a timber structure which it replaced. This group of features is currently being interpreted as a communal forge or smithy for the settlement.
The finds recovered from the medieval settlement suggest that it was abandoned in the first half of the 14th century. During this period, the Scottish army of Edward Bruce ravaged the area and was known to have at least passed very close to the site after sacking Castledermot, when an isolated settlement like this (belonging to the same lord) would have been very vulnerable to hostile forces. Similarly an Anglo-Norman settlement in this part of County Kildare would have been an obvious target for the forces of the Gaelic revival. There was, however, no evidence for widespread deliberate destruction anywhere on the site. Between 1315 and 1320, a series of bad harvests resulted in a famine in Ireland and elsewhere in Europe. This was followed by an outbreak of plague in 1320. The later outbreaks of plague, known as the Black Death, hit Ireland in 1349 and killed perhaps between a third and a half of the entire population. A combination of these factors, military attacks and natural disasters, is likely to have put an end to the occupation of this settlement during the medieval period but one should be wary of too definite an attribution as some of these factors would leave no trace in the current understanding of the archaeological record.
Medieval settlements of this time should be viewed as functioning within a landscape setting rather than as independent islands of activity. There was evidence of such activity beyond both ends of the central occupied part of the village.
To the south a series of north–south ditches may have marked trackways and boundaries between worked fields, possibly linking with the original southern access route into the settlement immediately to the west of the CPO zone.
To the north gravel pathways could be traced coming down the slope from the settlement to the former stream channels at the bottom of it. Between the settlement and the stream lay a pond apparently formed and filled by one of the springs on the slope. The sedimentary sequence included two peat layers and several sherds of medieval pottery. It also contained many deer antlers, apparently more than one would expect in natural circumstances. These have yet to be identified to species but may contain a mixture of antlers of native red deer and fallow deer, as introduced by the Anglo-Normans. These have also yet to be examined for evidence of human working. A large straight stream cut, parallel to some of the nearby croft enclosure ditches, ran down the slope from the medieval settlement across the northern boundary ditch and on to the former stream channels. It passed very close to the pond but narrowly missed it in its current form. It may have been used to maintain the water level of the pond and therefore its potential usage.
Immediately beside the pond on the northern side were a series of eleven shallow circular pits. These were generally ,1m wide and ,0.5m deep. Their fills contained burnt stone, charcoal and ash and were generally dark-grey or black in colour. They resemble the sort of pits associated with a fulacht fiadh and it is possible they are the last remains of such a monument, of which the overlying layers have been destroyed. No dating evidence was recovered from them but it is hoped that the samples taken may be able to fill this gap. The edge of the pond cut through the edge of some of them.
In the southernmost field of the site, the surviving part of a freestanding limekiln was excavated. This took the form of a belowground triangular chamber with an inverted cone from which the lime would have been collected at its apex. Much of the structure would have been aboveground and had been subsequently truncated. Many of the rocks of the side walls were mortared. The boundary ditch between this field and the adjacent one to the north contained many similar mortared stones and had been backfilled within living memory. The rocks may have been reused from the medieval settlement for the kiln. The kiln may have been contemporary with the building of one, possibly the 18th-century version, phase of the nearby Prospect House.
No evidence was found for the dating or interpretation of a visibly upstanding linear earthwork crossing the southernmost field near the limekiln.