5.3.5 Re-colonisation: was there Mesolithic 'sedentism' on Oronsay?

As from 7000 cal BP the number of activity events increases in number, reaching a peak at between 6200-5800 cal BP, but at only half the level attained prior to 8200 cal BP. The location of these events is almost entirely in the southern area of Argyll, notably on the Isle of Oronsay (Figure 29). This increase in activity events might be explained either by the growth of what had been the residual population by enhancing fertility and/or reduced mortality, or by re-colonisation with people returning to what had been a largely abandoned landscape. As noted, above, however, one must remain aware of the possibility of differential survival of archaeological sites in different phases of the Mesolithic past arising from environmental events.

It is indeed during this temporal phase that the post-glacial transgression reaches it maximum extent. This rise in sea level is likely to have destroyed a number of sites by coastal inundation. An example comes from Fiskary Bay (CANMORE ID 299865), Isle of Coll (Mithen et al. 2007a; Wicks and Mithen, forthcoming). Here a raised beach that contained abraded stone artefacts sealed the in situ Mesolithic deposits: the lower-lying part of a single activity location appears to have been swept over that at a higher level. Mesolithic chipped stone artefacts within a storm beach dating to the maximum transgression also been recorded at Croig, Isle of Mull (Mithen et al., forthcoming). While such coastal inundation might have destroyed some Mesolithic sites and made others harder to discover, Wicks and Mithen (2014) discount this as explaining the temporal pattern of activity events illustrated in Figure 31, primarily because of the chronological evidence was primarily derived from inland Mesolithic sites.

The Oronsay middens are some of the most intensively studied and discussed Mesolithic sites in Britain, Figure 40. These have provided data for the application of numerous scientific methods including the analysis of fish bones for seasonality (Mellars and Wilkinson 1980), human bone for isotopes indicative of diet (Richards and Mellars 1998), and Bayesian statistics for establishing chronology (Wicks et al. 2014). It is nevertheless regrettable that much of the data excavated by Mellars during the 1970s remains unpublished, including the human bones, bird bones, fish bones, stone artefacts, bone artefacts, features and so forth. Moreover the frequent characterisation of Cnoc Coig (CANMORE ID 37818), as a 'shell midden' does little justice to the complexity of its archaeological remains evident from unpublished plans because a range of features indicative of substantial structures are present.

Figure 40: Map of the Oronsay middens © copyright

The studies of the Oronsay material and ensuring debates have contributed to wider discussions about settlement patterns and diet in the Late Mesolithic and potential contrasts with that of the Neolithic (Schulting and Richards 2002; Richards et al. 2003). As such, the key themes of the re-colonisation phase are Mesolithic sedentism and diet.

There are two conflicting interpretations of the Oronsay middens. To quote Mellars (2004, 177);

"That the occupation and use of the different shell middens on Oronsay was essentially the product of a single social group, which remained on the island throughout most if not all of each year, but moving seasonally between the different sites (Richard and Mellars 1998)";

"That the occupation documented in the different middens represent much more sporadic and intermittent visits to the sites by a range of different social groups, who spent the remainder of each year on some of the larger adjacent islands, such as Jura, Islay or Colonsay, or possibly the Scottish mainland, most probably practising economies in these other locations dependent largely on the hunting of red deer and other land mammals (Mithen 2000: 623; Mithen and Finlayson 1991)".

Wicks et al. (2014) provided new evidence to support the second of these interpretations. Their excavations at Storakaig (CANMORE ID 304908), a non-coastal, upland site on Islay, had produced not only a Mesolithic site with a narrow blade assemblage but also, and uniquely for an inland site, preserved animal bones. Although highly fragmented, these represent red deer, roe deer, wild boar, fox and badger. They were preserved by having become heavily calcined from being burned at high temperatures. The location of the Storakaig (CANMORE ID 304908), its artefacts and fauna suggests a small Mesolithic hunting camp, Figure 41. When its eight AMS 14C dates (all on single fragments of charred hazel nut shell) were analysed these indicated a minimum number of four activity events, centered on c. 6320, 6070, 5840 and 5810 cal BP. As such, a strong overlap with the chronology of activity events on Oronsay was demonstrated, which is illustrated most effectively by the overlap in the summed calibrated probability distributions for Cnoc Coig (CANMORE ID 37818) and Storakaig (CANMORE ID 304908) (Wicks et al. 2014), Figure 42.

Figure 41: Storakaig, Isle of Islay © copyright

Figure 42: Overlap of the SCPD for Storakaig and the Oronsay middens © copyright

Although it cannot be demonstrated formally that the same group of Mesolithic foragers were moving between Oronsay, where they were predominately fishing and collecting shell fish, and Islay, to hunt big game, that would appear to be the most reasonable interpretation, lending support to the second of Mellars' hypotheses listed above - although this might reflect the movement of a single social group rather than the 'range' of social groups that Mellars' proposes (whatever 'range' might mean in this context). In this regard the differences between the chipped stone assemblages from Cnoc Coig (CANMORE ID 37818) and Storakaig (CANMORE ID 304908), the former predominately bipolar the latter predominately platform core technology, should be interpreted as responses to available raw materials, types of activities undertaken, anticipation of future needs and the social context of tool manufacture, rather than distinct cultural traditions of separate populations.

If one does indeed favour the second of Mellars' (2004) hypotheses, then the behaviour of the Mesolithic foragers within this period appears similar to that with the residential period: 'thoughtful foragers' moving to different locations to exploit different types of resource at different times of the year, always in a state of uncertainty because on-going change in their environment. While one might perceive an increased focus on coastal foraging, perhaps relating to the marine transgression, shell middens had also been created during the residential period, notably at Sand, Northton and An Corran.