Description
Occupying a position in the centre of the group, Building A stood within a rectangular enclosure. Its foundation trench was dug in two halves with a break at the centre, with steeply sloping sides and a broad flat base 0.7 - 0.8 m deep throughout. Evidence of the positions of timbers in the trench survived in the form of slight depressions in the base and darker patches in the fill, with stones tending to lie vertically in the earth which had slumped into voids left when timbers had been withdrawn.

In the south wall trench the positions of five upright timbers were identified west of the centre and seven to the east, including two corner posts. From the evidence of basal staining it appeared that all were circular in section with a diameter of some 0.3m. On the east side their spacing was regular at approximately 0.95m measured centre to centre. To the west, where recovery was not complete, there may also have been seven timbers. The space in the centre for the entrance was 0.6m wide, and neither the entrance posts nor those at the corners were any larger than the others. In between the posts there was fragmentary evidence of a wall line formed from planks 0.08m wide set to a depth of 0.2m into the trench.

The west end wall was founded in a trench of similar depth and profile to that on the south and with a similar arrangement of posts, seven including the two corner posts, and planking in between. The definition of two of the posts was faint, but clear enough to show that the full set was represented. Towards the centre the spacing between the posts was again about 0.95m, but the distance between the south corner and the next post was greater, and at the north corner two posts were set close together.

The detail recovered in the east end trench was of poor quality. The depth and profile matched the other three trenches, but only towards the north corner were structural features observed. As far as one can tell from the available evidence, the same arrangement of round posts of about 0.3m diameter with planking in between seems to apply to this wall. Its length however was only 6.1m, measured between centres of the corner posts, as compared with 6.9m for the west wall.

Small pits within the building may have held posts, but in no case is the evidence firm. A row of three, all of some 0.2m deep, divided off the west end. The oval pit whose fill spread to some extent over one pit may be a later feature. Towards the centre, two pits were similar in size to the three above. In the north-east one was shallower, slightly more than 0.1m deep. In contrast, the four small pits by the west wall were all between 0.1 and 0.2m deep with steep V profiles.

This text is a quotation (with slight modifications) from:
pp. 61 - 64 of Colm O'Brien and Roger Miket "The Early Medieval Settlement of Thirlings, Northumberland." Durham Archaeological Journal, Vol 7, 1991, 57-91. Users should consult this paper for any further information.

Interpretation
The illustration (fig. 19 from the O'Brien and Miket paper cited above) shows the ground plan of Building A and the excavators' interpretation of the structural type. The internal posts act as the main structural support system holding a ridge beam via a king post assembly and with two side purlins. The bottom ends of the rafters are secured on to a wall plate. In this way gross irregularities in the alignment of the walls and the spacing of timbers along the wall lines can be accommodated. Angled timbers outside the wall line connect to the wall plate. The roof form suggested here is half-hipped for there is no central end wall post to support the ridge beam, while the purlins can be carried through to end wall posts. In the eastern half of the building, with the shorter end wall, the purlins are not parallel but converge towards the end, echoing the alignments of the side walls. An effect of this will be that, if the ridge beam is horizontal, the angle of pitch of the roof will vary across the building. The interpretion of structural type proposed by the excavators is developed from an analysis of this and other buildings in the Thirlings group. It is based on the argument that irregularities in the ground plan of the buildings (opposite walls differing in lengths or not parallel; corners not at right angles) allow for an an analysis which distinguishes two systems, a structural system of roof supports and a walling system, with the one to some extent independent of the other. The O'Brien and Miket paper cited above should be consulted for further detail of this analysis. The analysis also confirms broadly the interpretions proposed by James, Marshall and Millett in a wider review of buildings of this era. (See Simon James, Anne Marshall, Martin Millett "An Early Medieval Building Tradition." Archaelogical Journal Vol 141, 1984, 182-215.)