4.4.2.1 Early Neolithic

Doon Hill, East Lothian

Excavations by Brian Hope-Taylor between 1964 and 1966 at Doon Hill uncovered the remains of two large, timber-built, roughly rectangular structures, set within a palisaded enclosure enclosing an area of around 60 x 40m (Hope-Taylor 1980). Hope-Taylor assumed that he was excavating early medieval halls and claimed that the earliest (and largest) was built by native Britons around AD 550 and the second, constructed within the footprint of the first, was built by Northumbrian Angles around AD 640. He never produced a definitive report of his excavations. Recent research has overturned this interpretation and has shown that a degree of ‘artistic licence’ was involved in Hope-Taylor’s presentation of the excavation results (Ralston 2019a; 2019b; in press). Both of the structures and the enclosure have been shown, through radiocarbon dating and by the presence of Carinated Bowl pottery and lithic items, including of pitchstone, to be of Early Neolithic date, around the 37^th century BC. They were preceded by Early Neolithic pit-digging activity which was cut by the palisade trench. Ralston’s re-evaluation of the excavation results continues and fresh excavation is required to answer some outstanding questions.

The earlier, and larger, of the two structures – Hall A – measures 23 by 10.4 metres and has slightly bowed-out gable ends featuring massive central posts flanked by smaller posts, all set in a trench. The side walls were constructed using close-set oak posts; the nature of the walling in between these posts is unclear. While Hope-Taylor showed the uprights as being squared off, Ian Ralston has pointed out that such an interpretation is not supported by the photographs and section drawings. There was some internal division of space. Shortly after this structure was destroyed by fire a smaller, rectangular, structure – Hall B – measuring around 15.2 by 8.9 metres and constructed using continuous foundation trenches, was built within its footprint. While this is of less massive construction than Hall A, this was nevertheless a sizeable building. It, too, had internal divisions and was destroyed by fire. The palisade enclosure was built within a near-continuous slot trench and featured upright timbers, although details of its construction remain unclear. Radiocarbon dating and artefactual evidence confirm an Early Neolithic date for it.

While no parallel for the palisade enclosure surrounding a large building is known from Scotland, there are several Early Neolithic parallels for one or other of the buildings themselves, not just in Scotland but further afield in Britain. Several have been excavated in Scotland and shown to be of Early Neolithic date: Balbridie, Aberdeenshire (Fairweather and Ralston 1993) and Crathes Warren Field, Aberdeenshire, just across the river Dee from Balbridie (Murray et al 2009); Claish, Stirling (Barclary et al 2002); Lockerbie Academy, Dumfries and Galloway (Kirby 2011); and Carnoustie, Angus (Smith et al 2025). Others have been claimed from aerial photographs, for example at Sled Hill, East Lothian (Brophy 2007, 80). However, owing to their superficial similarity with early medieval halls, it is usually necessary to ground-truth these through excavation. This is particularly important in south-east Scotland, where there is plentiful evidence for the presence of Northumbrian Angles. Among the aerial photography evidence, that from Whitmuirhaugh, Sprouston, is the most convincing candidate for an Early Neolithic structure (Smith 1991; Millican 2016, 124). It measures 19 by 5 metres and has similar-looking bowed ends to Doon Hill Hall A.

How were these large structures used? The term ‘hall’ is misleading, as it evokes the functions of the early medieval halls such as as royal residences and places for elite feasting. Instead, the term ‘large communal house’ maybe more appropriate. They need to be understood in terms of the process of colonising new areas by incoming farming groups. They are only found during the Early Neolithic and the most plausible explanation is that they were built by groups of farming families when they moved into a new area, so that they could all live together until they felt sufficiently well established to ‘bud off’ into individual farmsteads, with smaller houses. The fact that all the excavated examples had been destroyed by fire could be interpreted as a way of symbolically marking the end to that initial phase of settlement. It can be difficult to prove that a building has been deliberately burnt down, but the fact that only a few artefacts are normally found in these structures points towards a deliberate act of destruction after possessions had been removed.

The fact that a second large house had been built in the footprint of the first does not contradict the model of ‘dispersal following communal living’. It could be that a dominant extended family chose to underline its status by building a substantial house for itself. Positioning it in the footprint of the original communal house would give extra cachet, being associated with the place where the foundation for the whole farming community had been located. A parallel for this practice has recently been found at Carnoustie (Smith et al 2025). The presence of the palisade enclosure, which may well have been built when the first structure was built, suggests that the occupants felt the need to defend themselves. It may well be that there were still groups of Mesolithic hunter-fisher-foragers in the area, for whom the sight of these massive buildings would have been alien and impressive. Whatever was the case, we can say that the second large house had the same fate as the first, being destroyed by fire.

One question arising from these large structures is: to where did the rest of the community go? In theory, there should be several smaller houses in the landscape. Finding them, however, can be difficult.

The Hirsel, Scottish Borders

In contrast to Doon Hill, the Early Neolithic structure at The Hirsel, near the River Tweed at Coldstream is of much lighter construction and is of a different shape (Cramp 2014). It is represented by an arc of stake-holes that could have been associated with a wattle-and-daub wall. The area had been disturbed by later medieval activity, so it is impossible to tell whether this had been a circular or sub-rectangular structure. The arc suggests that an area at least 6m across had been defined. One early Carinated Bowl style pot was found within the area, and another outside it. Geophysical survey indicates that further Early Neolithic settlement evidence, including pits, may remain unexcavated in the area. No radiocarbon dates were obtained for this Neolithic phase of activity, although the style of pottery present suggests a probable date around the 37^th century BC.

It is tempting to interpret this stake-built structure and associated features as an example of an individual farmstead settlement, but many questions remain. Was it used for year-round occupation? Or was it a special-purpose temporary settlement? It may be that fresh excavation, in hitherto-unexcavated areas, could address this question.

Ratho and Maybury Business Park, City of Edinburgh

The structural evidence at Ratho and Maybury is more flimsy than that at The Hirsel. Both of these sites were heavily truncated, and at Ratho, it is not certain that the rectilinear, artefact-free structures are Early Neolithic or later (Smith 1995).

The settlement evidence from Ratho consisted of two pits, plus one or two further pits, at the northern end of the excavated area (Smith 1995). Carinated Bowl pottery, lithics, including a fragment of a pitchstone blade, and plant material were associated with these features. No radiocarbon dates were obtained for these features, but an early fourth millennium date can be assumed on the basis of the artefactual finds.  Further south are the remains of two rectilinear structures, which may have been houses. No artefacts were found in these, and attempts to find reliable radiocarbon-datable material were unsuccessful. It was concluded that while an Early Neolithic date was possible, so was an early medieval date, as material from that period was also present on the site (Smith 1995). Smith also suggested that a Bronze Age date could not be ruled out, but this seems unlikely, since Scottish Bronze Age houses are not rectilinear.

At Maybury Park, the evidence for a possible structure consists of a slightly curving line of five widely-spaced post-holes, an elongated oval pit containing Carinated Bowl pottery which was partly enclosed by the post-holes, and a shallow, H-shaped feature containing burnt stones and clay (Moloney and Lawson 2006). Two radiocarbon dates were obtained from burnt organic material from the oval pit, of which the 3950-3660 cal BC date (GU-11072) is most likely to reflect the date of this activity (Moloney and Lawson 2006, 14). It is hard to resolve these features into a coherent ground plan of a house, although the small set of artefactual and other finds are characteristic of what can be found in a settlement. To the north of these features are the remains of what the excavators described as a Neolithic ’trackway’ made of a band of compacted earth.