There has been a lengthy tradition of studying the landscape settings of long barrows and other Neolithic monuments in the SWHS and environs, driven primarily by the research traditions of field survey (RCHME 1979; McOmish et al. 2002; Bowden et al. 2015, 134–140). In recent decades, phenomenological approaches have also become prominent (e.g. Exon et al. 2000), although these have tended to focus on later monuments in the SWHS (Pollard and Ruggles 2001; Parker Pearson et al. 2006). These surveys have made clear that long barrows in particular, and many monuments more widely in the SWHS, were often, but not always, situated with careful regard to their local topography, although without any strict shared alignment (Bowden et al. 2015, 18).
The longstanding belief that the southern British chalklands were rapidly cleared of woodland by early farmers (e.g. Thurnam 1869, 171–172; Pitt Rivers 1869, 28; Sumner 1913, 1; Piggott 1954, 22, 50; RCHME 1979, xi) has been tempered in recent years by recognition of the complexity of the Wessex landscape in terms of both its pre-Neolithic appearance and the history of woodland clearance (and regrowth) in particular areas (Drewett et al. 1988, 35–36; Allen and Gardiner 2009), as well as considerations of seasonality (Cummings and Whittle 2003, 260). The extent, character and ecology of woodland on the chalk in the Early Neolithic varied from quite open 'park' landscapes of Stonehenge's immediate environs and Cranborne Chase, to dense, closed-canopy woodland in northern Wiltshire and Sussex (Allen and Gardiner 2009; Hazell and Allen 2013). Therefore, in recent years it has become the norm, with good reason, to preface discussions of the visibility of monuments with a disclaimer about the potential effects of woodland. Equally, it is now widely recognised that vision, like hearing and smell, is a sense fundamentally contingent upon cultural values and, therefore, that visibility cannot be understood simply through uncritical use of the 'viewshed' function available in all GIS packages (Gillings and Wheatley 2001, 13). Consequently, current research emphasises the affordances for visibility that may have arisen through movement around the landscape in the course of day-to-day activities, particularly following watercourses with livestock, and suggests that the perception of monuments, visual and otherwise, varied from site to site and region to region (Oswald in prep a; in prep. b).
Current evidence suggests that most of the core of the SWHS landscape was predominantly grassland in the Early Neolithic; in other words conditions that would have allowed near-maximum visibility, although more woodland appears to have been present in the Avon valley (Richards 1990, 256; Hazell and Allen 2013). This interpretation is supported by the molluscan evidence from Historic England's excavation of WS71, with the caveat that this and most of the available data derives from the excavation of monuments, where some degree of woodland clearance must have occurred before the start of construction. There is indirect evidence that would support an alternative view: the widespread use of timbers in various Early Neolithic monuments around the SWHS and environs implies that trees were not entirely absent from the local landscape. Even hazel copses, whose presence must be attested by the regular finds of carbonised nutshells from Early Neolithic sites, can completely obscure views, even in winter, with their dense clusters of stems standing 8–12m high. In any case, during the centuries that the monuments under consideration here retained currency, there is potential for profound change in the vegetative environment, perhaps primarily due to grazing and browsing by domesticates. The problem, then, is not simply one of knowing where trees and bushes stood (Tilley 1994, 73), but also when they stood.
Considerable work has been undertaken to analyse the viewsheds and intervisibility of long barrows in the Stonehenge landscape and the wider Wessex region (e.g. Wheatley 1995; Exon et al. 2000). For the SPTA it appears that the visual impact of most long barrows might be quite local and often confined to the valleys adjoining them (Thurnam 1869, 171–2; McOmish et al. 2002, 27). Likewise, investigation of the Salisbury Plain West Group in the Wylye valley found that views from the barrows focused on maximising views to the valley rather than between each other (Allen and Gardiner 2004, 72). It has also been pointed out that the positioning of a few larger long barrows in prominent upland positions has skewed results, creating a degree of largely coincidental intervisibility (McOmish et al. 2002, 27; Lawson 2007, 52).
A previous study of long barrow intervisibility in the SWHS observed that there was great variation between the viewsheds of different barrows, but that the cumulative visibility suggested that there were small groups of barrows, apparently clustered around the heads of dry valleys, sited so as to maintain intervisibility with each other (Exon et al. 2000, 37–42). It could be inferred from this that people occupied the shelter of the valleys, as Thurnam believed, or, perhaps more likely, that the valleys offered natural routes or droveways along which people regularly passed, and that interfluves, watersheds and springs were significant locations in the Neolithic. For the area around Wilsford and Winterbourne Stoke, it was noted that from a position on the Normanton ridge by WIL30 there was good visibility of all the local long barrows (Exon et al. 2000, 40). WS1 is potentially intervisible (depending upon vegetation cover) with four other long barrows (AM14, WIL30, WIL41 and WIN53) but not with WIL34, only 500m to the south-east. Field observations indicated that WS71 was on a slope down to a shallow dry valley and, when standing on the ground surface, was separated from WS1 by a low crest, preventing intervisibility. Standing on the excavation spoil heap, however, allowed Winterbourne Stoke 1 to come into view. This implied that in an unwooded landscape (as suggested by the molluscan analysis of WS71), the two mounds would have been intervisible to people standing upon the mounds but not to those standing on the natural ground surface nearby. The view of WS71 could, then, potentially be socially restricted, depending on whether access to the top of the barrow mound was permitted. The possibility therefore exists that WS71 was intentionally constructed in a particular location to allow a complex visual interplay between the two monuments, depending on their relative order of construction, which remains unknown.
The current consideration of the examples in the SWHS presents a picture of individual variability that is consistent with the observations in the SPTA (McOmish et al. 2002). WS71 is on a NE–SW bearing, an orientation it broadly shares with four of the barrows in the SWHS and environs (Table 1). Situated on a south-facing slope towards the base of a dry valley, it runs obliquely across rather than with or perpendicular to the gradient. The largest long barrow in the WHS, WS1, is situated on a low ridge c. 500m to the north. Although it runs on the same NE–SW alignment, it is aligned length-wise along the ridge on which it sits. About 400m to the south-east of WS71 is WIL34. This also has a NE–SW orientation and, like WS1, conforms with its local topography along the line of the spur on which it was constructed (Bax et al. 2010, 38). Topographical positioning does vary and DUR24 is situated on a south-west facing slope and runs broadly along the contours. AM14 lies on a south-facing slope, slightly across the gradient. WS86 is on a different, north-south, alignment (Wessex Archaeology 2016b).
The reasons for this variability remain obscure, but the orientation of this newly discovered barrow suggests that it is in some way differently focused from the three neighbouring barrows of WS1, WS71 and WIL34, although these do display significant variation in their alignment relative to the natural topography. McOmish et al. (2002, 22) and Ruggles (1999, 126–127) state that no common astronomical or topographic alignment of long barrows can be clearly detected in the SPTA, and this work demonstrates that the same applies in the SWHS (Figure 14), bearing out Darvill's previous assessment (1997, 178). As discussed above, however, we also suggest that, as in the SPTA, there was some significance attached to the eastern end of long barrows in the SWHS. Many of the long barrows in the SWHS appear to have been enlarged or elaborated at their eastern ends, perhaps indicating where ritual activity was focused following construction of the mound (Field 2006). Field (2006, 69) has hypothesised that WS1 may be aligned with the midsummer sunrise, but notes that this may be a coincidence given its clear topographic alignment with the ridge on which it sits. Local topography and woodland would undoubtedly have had a significant effect on the apparent azimuths of cosmological phenomena, especially the solstices where the angles of declination towards the horizon are acute, but modelling these potential effects in relation to individual monuments is beyond the scope of this article.
Internet Archaeology is an open access journal based in the Department of Archaeology, University of York. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.
Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.