Cloonybeirne townland was assigned as the residence and demesne of the Tanaiste of the O’Conor Roe family during the Compossicion of Connacht in 1585. Systematic mapping of relict earthworks has begun to reveal the outline of this important lordship centre.

The team has identified the possible site of the castle which consists of the foundations of a large rectangular building located inside a small circular enclosure. A deserted nucleated settlement which occupies the poor land to the north and west of the castle could have medieval origins. However, the foundations of the vernacular style dwellings and gardens are similar to later deserted settlements nearby which would suggest this site was abandoned in the eighteenth century or early nineteenth centuries. A holy well and a penitential cairn are located in the southeast corner of the townland, although there is no trace of a church. Just across the river, in Ballyglass townland a nineteenth century millrace runs along the side of a large oval enclosure which suggests that the mill site could be considerably older and that the inhabitants of this O’Conor centre may have had access to a mill.

An earthwork which has been identified as a moated site is located at the eastern end of the townland and could be the site of an earlier O’Conor residence. Moated sites, which are the remains of moated enclosures around dwellings or manor houses, are common in various parts of northwest Europe and in Ireland they are traditionally associated with Anglo-Norman settlements. However, in Roscommon it appears that they were generally constructed by the upper levels of the Gaelic Irish nobility, such as the O’Conors, from the fourteenth century onwards. As such they could be seen to represent a redefinition or ‘rebranding’ of the indigenous nobility through their adoption of a widespread European settlement form. The moated site, which occupies the best land in the townland, is located within a relict field system which if contemporary indicates there was an enclosed and farmed landscape around the dwelling. Geophysical survey of the site was undertaken in order to investigate if there were traces of buildings inside the enclosure and if there was any additional evidence for additional enclosing or fortifying elements.

Analysis of the plan of the earthworks suggested that the moated site had been modified in the past. Resistance survey revealed sub-rectangular anomalies within the enclosure which are likely to relate to buildings and associated surfaces. High resistance readings along the sides of the dry moat suggests that it was once defined by stone walls or revetments.

The magnetic gradiometry revealed that this earthwork has a more complicated history which prompts a reconsideration and redefinition of the site. The presence of a circular enclosure in the northwest corner of the rectangular enclosure suggests that the moated site is actually an expanded and remodeled ringfort. There is at least one other similar site nearby at Carns, close to the O’Conor inauguration mound in the same townland. These earthworks sites could therefore reflect a deliberate remodeling of existing sites to legitimise or reiterate Gaelic lordship in the O’Conor territory. However, the plan of the Cloonybeirne earthwork also shares some similarities with an early medieval ringfort/D shaped enclosure at Dowdstown, Co. Meath, which was excavated recently as part of the M3 motorway programme.

Geophysical survey and excavation of Dowdstown 2 revealed that it was a circular enclosure which was enlarged into a D shaped enclosure. Preliminary dates indicate it was occupied between the sixth and ninth centuries AD. This of course raises the shortcomings of attempting to date earthwork sites based on earthwork morphologies and questions our use of site classifications and terminology. If the Cloonybeirne rectangular or D shaped earthwork pre-dates the introduction of ‘moated sites’ by several c

enturies it is more properly defined as a ringfort or D shaped enclosure. However, the rectangular anomalies visible on the resistance survey could be the foundations of the type of large buildings which would be expected in a late medieval moated site. At this stage all that can be deduced is that the site originated as a ringfort and that it was subsequently enlarged at some time between the sixth and fifteenth centuries. The final shape of the earthwork could be due to the organic expansion of a ringfort settlement. Alternatively it could be the result of a later deliberate reshaping of an older ringfort site into a fashionable

moated site. That scenario would represent a bigger conceptual shift as a literal recasting of an iconic lordship centre in broader European terms. Either way, it is tempting, given Cloonybeirne’s significance as an O’Conor lordship centre, to infer long term occupation at this earthwork site, at least until the construction of the castle. Cloonybeirne and other similar earthworks therefore have the potential to inform the debate about dating the abandonment of ringforts and their replacement by alternative settlement forms.