Domestic Architecture/House Forms
The remains of circular houses are one of the most frequent monument types and their use spans the Bronze and Iron Ages. As outlined in the Introduction, Historic Environment Scotland’s Monuments thesaurus dictates the use of ‘hut circle’ for sites which have above ground remains, and ‘round house’ for those which don’t. In reality the distinction is at best unhelpful, because the former tend to be found in the uplands, the usage restates an upland/lowland divide, which is at least in part an artefact of historic/modern land use and implies a binary division in form that is misleading. As with all vernacular architecture, it is probable that availability of materials would dictate, to some degree, variations in form, and it is likely that these were reflected across the landscape.
The terminology used to describe sites has also varied over the history of research, in part as antiquarians and archaeologists sought to characterise and interpret the emerging archaeological resource. The resulting array of terms has proved a hindrance to understanding, using weighted or inappropriate elements at times. For example, the examination of the historical use of ‘circular fort’, ‘ring-fort’ and ‘homestead’ led to the recognition of the site type as one form of ‘monumental roundhouse’ (Hingley 1992, 14–15; Strachan 2013, 8–10). Equally, Thoms and Halliday (2014) have addressed the regional nuances of hut circle/roundhouse forms; they draw on known forms to suggest ‘post-ring’ and ‘ring-ditch’ as design elements, and ‘tangential pairs’ as possible evidence of sequential construction rather than distinct house types. These approaches are helping to establish a regional baseline for settlement type definition, allowing future research to move on from the ‘self-absorbing’ debates of classification that dominated discourse in the late 20th and early 21st centuries (ScARF National Framework Chapter 5).
The north-eastern uplands of Perth and Kinross include one of the densest concentrations of hut circles known anywhere in Scotland (RCAHMS 1990, 2), which are typically found outwith agriculturally improved areas and at high elevations, often 250m above sea level. They are generally unenclosed and distributed individually in apparent isolation, in small scatters, clusters or rows. They are notable in their diversity of form and are predominantly in single or double-walled forms, and in pairs enclosed by a single concentric wall (Harris 1984, 203–7; RCAHMS 1990, 2–4). A remarkably early interest in these various forms, by the Rev. Stewart in the late 18th century (Thoms and Halliday 2014, 13), was much later first broadly classified by Thorneycroft (1933). While the reasons behind the construction of one type over another remains unclear and merits further research, as a whole they represent a long chronology of roundhouse use between the 2nd millennium BC and 1st millennium AD (Thoms and Halliday 2014, 17). Variation may simply reflect available materials and changes in architectural fashion. As a result, the monument type offers an excellent research opportunity regarding the transition of settlement from the Late Bronze Age to the Early Iron Age and from the Iron Age to early medieval period.
Several sites have been excavated since the mid-20th century, including the group at Tulloch Field, first excavated in the 1970s by Stewart and later by Thoms, which included one roundhouse dated to around 400–200 BC amongst predominantly Middle Bronze Age dated examples (MPK2854; Thoms 1983; Thoms and Halliday 2014). Similarly at the scattered hut circle group on Carn Dubh, an Early Iron Age burnt roundhouse dated to 745–385 BC, was located amidst an otherwise Middle Bronze Age dated group (MPK1752; Rideout 1995).
As a result of this wide chronological span of their use, it is not possible to identify Iron Age hut circles, or groups, without excavation. This inhibits discussion about the known distribution of the site type, which is also seriously restricted by the impact of subsequent phases of agriculture impacting on monument survival (RCAHMS 1990, 1). Furthermore, levelled platforms among some groups have been interpreted as sites of degraded roundhouses (Thoms and Halliday 2014, 15). While excavation of a tangential pair at Pitcarmick confirmed that, at that site at least, the configuration was a result of sequential construction rather than representing a monument type (Carver et al 2012).
Wider discussion of the region’s unenclosed settlement (Halliday 2007) has challenged the common interpretation that these remains represent continuous occupation over hundreds of years. Rather a dynamic process is suggested whereby the hut circles and associated field systems represent shifting cycles of short-term usage at multiple locations over millennia (Thoms and Halliday 2014, 16).
The vast majority of the region’s roundhouses have been identified as circular cropmarks between about 5–15m in diameter, found along river valleys in areas where soils and agriculture promote their detection. Prior to levelling through cultivation, it is likely that many, such as circular post-ring or shallow ring-ditch sites, were architecturally similar to the hut circles in the uplands (Harding 2004, 102; Thoms and Halliday 2014, 15). There were no doubt slight variations in the nature and availability of building and thatch materials between lowland and upland environments, which would have been expressed in vernacular nuances. A key distinction, however, is the common association with souterrains found at lowland roundhouses, such as Newmill East (MPK2308/9; Watkins and Barclay 1981), Loak Farm by Bankfoot, (MPK20126) and Northleys by Luncarty (MPK20138; Wilson and Clarke 2019; Paton and Wilson 2019). All of these sites dated from excavated material to the Late Iron Age (4th–1st century BC).
Roundhouses with ring-ditches are also mostly known as cropmarks, and characterised by a ditch enclosing a circular area, often with an internal ring of postholes. While the interior is usually interpreted as domestic, it has been suggested that the outer areas may have housed animals (Harding 2004, 84). Earthwork ring-ditches are also known in less intensively ploughed areas and feature an external bank to the ring-ditch. While Bronze Age examples are known, many of the 167 records of ring-ditch, and 145 records for roundhouses held in the Historic Environment Records may relate to the Iron Age. While there are a relatively large number of confirmed Iron Age ring-ditch houses, as with hut circles, only further excavation to secure dating will disentangle their chronology. Recently, one of two ring-ditch houses at Easter Croftintygan (MPK11992), near Lawers on Loch Tay, was partially excavated with the aim of comparing site chronologies with nearby crannogs. A series of secure contexts including postholes, a central hearth and the ring-ditch, produced dates in the 1st centuries BC/AD (Stratigos and Hamilton forthcoming).
The excavated examples with an enclosing palisade at Blackford (MPK17955; O’Connell and Anderson 2021) and at Newmill West (MPK2331; Wilson and Clarke 2019; Paton and Wilson 2019) are also notable. Here, two post-ring roundhouses produced Early Iron Age dates (around 726–397 BC) and a single ring-ditch roundhouse from the Late Iron Age (357–61 BC). The palisade would have restricted access to the houses and it is suggested that, in addition to defence, it may also have signified status of the occupants (O’Connell and Anderson 2021, 119). With only three known examples across the region, it remains unclear how common individual roundhouse enclosure was across the lowlands of the area, but does indicate another variation in timber roundhouse construction, which may represent a degree of monumentalism utilising timber instead of stone.
In 2021, excavation at the James Hutton Institute, Invergowrie, revealed a well-preserved roundhouse, with several elongated/oval buildings and a series of probably contemporary outlying postholes and linear features (Nicol pers comm). While as yet undated, both the level of preservation at this site, and the variety of associated features looks set to make a significant contribution to our understanding of roundhouses on the Carse of Gowrie.