The Roscommon Module had two strands of research
- A largescale landscape study, which concentrated on reconstructing the later medieval landscape and settlement fabric of an area coinciding with the diocese of Elphin. This segment also included excavations in the townland of Carns to date relict field boundaries.
- Research on a narrower area, concentrated on sites in the Tulsk, Co. Roscommon.
Tulsk: when a mound becomes a castle
A grass covered mound in the village of Tulsk, Co. Roscommon provided a fascinating insight into life in medieval Roscommon, with many layers of history and some prehistory revealed.
Tulsk, Co. Roscommon was chosen as a study area because it was a central location in the O’Conor lordship. Historical sources mentioned that O’Conor Roe built a castle in Tulsk in 1406 and the castle is mentioned in the Irish annalistic sources until 1501. Local tradition suggested that the former location of the castle was a low grass-covered, stone filled, mound. This mound is located adjacent to the Rathcroghan Visitor centre. Geophysical survey of the mound suggested that it contained a number of archaeological features, but it was only when excavation began that the richness of the site became apparent.
Excavation suggested that there was human activity on the site in prehistory, with both a leaf-shaped arrowhead and a barbed and tanged arrowhead discovery in the earliest layers. A ringfort was later constructed on the site with finds from the 10/11th century. The ringfort was raised at some point, to create a central platform in the interior.
The 15th century castle constructed by O’Conor Roe was a large stone tower, 20m in length and 10m in width, which occupied the eastern side of the site. The platform of the raised ringfort was encastellated and served as a protective bawn.
Finally, there were rich layers dating to the Elizabethan period, at a time when the Queen Elizabeth I’s governor, Sir Richard Bingham visited Tulsk and garrisoned the settlement in the 1590s.
An indepth survey of the priory and graveyard in Tulsk were also conducted to establish how its history intertwined with that of the settlement sites within the mound. The priory at Tulsk was founded in 1448 as a Dominican House. It would have had a community of four to six friars and relied for the most part on the support of its local patrons and the descendants of its founding family. The identity of its founder it disputed , but it was either a member of the MacDowells, a family of Gallowglass origins or Felim son of Felim Clery O’Conor. The monastery was dissolved and the monks left the site in 1584.
The remains of the priory consist of the remains of a church, fragmentary remains of the cloister and ranges, remains of a tower house as well as a number of mausoleums.
Carns: a landscape of lordship
The Medieval Rural Settlement Project carried out an excavation in Carns townland in 2006 as part of a broader investigation of relict field boundaries and deserted settlements in the O’Conor lordship of north Roscommon. Carns was chosen as a case study, based on the presence of extensive earthworks representing abandoned settlements and a field system that was tentatively dated to the later Middle Ages. The excavations set out of establish a date for these settlement sand field systems.
The ridge occupied by Carns townland was the focus of an important prehistoric cemetery. The townland takes its name from a prehistoric burial cairn known as Carnfree, named after Fraoch, the hero of an early medieval tale, Táin Bó Fraích. A tradition that St Patrick chose this location for a church, the Domhnach Mór Maighe Selga, also suggests a concerted effort to integrate this important place into early Irish Christian tradition between the 5th and 7th centuries. The area subsequently became central in the O’Conor kingship of Connacht; they were inaugurated on the prehistoric mound of Carnfree from at least the early 13th century. The presence of a moated site completes a broader ‘landscape of lordship’ comprising church, residence and inauguration site.
A church site
The focus of excavation was a large circular enclosure (RO028–075) classified as ‘ecclesiastical remains’. The enclosure (c. 130m diameter) was incorporated into a later field system that is likely to be of medieval date. The purpose of the excavation was to establish a clear function for the site and to determine if there was a medieval phase. Two areas were investigated: the site of a building located near the centre of the enclosure and an area on the periphery of the enclosure.
Investigation of the building foundations exposed the west gable-end of a medieval church and part of an extension that may have had a residential function. Both phases incorporate punch-dressed stones, indicating a 15th–17th-century date. Burials that pre- and post-dating the construction of the building indicate a that burials took place on the site over a long period, confirming that it was a religious site for an extended period of time. The discovery of an early medieval lignite bracelet and a collapsed ogham stone, of 5th- or 6th-century date, makes it likely that this is, in fact, the site of the Patrician church of Domhnach Mór Maighe Seilge. This assertion is further supported by excavation along the perimeter of the enclosure, which revealed that the site was at one time enclosed by a 2m-thick drystone wall, typical of early church sites.
A settlement site
A field boundary, post-dating the drystone church enclosure and associated with the wider field system, was also excavated. Carbonised cereal remains, found on a possible kiln flue, were analysed by environmental archaeologists. These seeds came from a variety of cereal types rye, barley, wheat and oats. Rotary quern fragments where found on the site, querns are uses to process cereal like those found on site. This evidence, along with the house sites and garden enclosures, suggests there was an agricultural settlement focused on the church.
In 2007 further excavation on settlement sites were carried out in Carns townland. A rectangular building defined by drystone walls was uncovered, its entrance was south-facing and marked by large limestone slabs. Analysis of the exposed remains and the results of geophysical survey suggests that the building measured c. 10m north–south by at least 8m. Internal features included a large hearth and a possible internal partition. Artefacts including a cross-headed bone stick-pin, a perforated bone pin, a segmented tubular blue glass bead and iron knives suggest the building was occupied some time between the 10th and 12th centuries AD. Animal bones and rotary-quern fragments found during the excavation reinforces the impression that the church site incorporated a significant settlement component.
Finds from other periods
Although, the project set out to discover the remains of medieval settlement in also uncovered the remains of post-medieval settlement as well as prehistoric activity
In the townland of Carns evidence for 18th century occupation of a two roomed dwelling was uncovered by excavation. Finds were scarce but included a bone-handled iron knife, a glass-bead necklace, and the stem of a clay pipe. A lead pistol shot was found impacted against the wall beside the northern doorway of the dwelling. When taken in conjunction with the glass bead necklace which had been dropped on the stone threshold of the same doorway, it suggests there was an altercation leading to an eviction or abandonment of the dwelling.
An earlier phase, probably of prehistoric date, was revealed by a geophysical survey (06R0081) conducted by Dr Paul Gibson and Dot George, NUI Maynooth. It consists of a multivallate enclosure surrounding a figure-of-eight structure. Excavation of one of the enclosing elements has established that it was a ditch, 1.6m in depth, which tapered towards a rounded base; animal bone retrieved from the bottom will hopefully establish a date for this earlier phase. Intriguingly, the church precinct wall carefully encloses this earlier phase, which suggests there was a deliberate attempt to incorporate the earlier site into Christian tradition.