5.3.4 Population decline: the impact of climate change?

Following c. 8200 cal BP the archaeological record shows a dramatic decline in the number of activity events (Figure 31), a decline that appears even more dramatic in the summed calibrated probability distribution (Figure 32). Wicks and Mithen (2014) proposed that this was a consequence of the environmental impacts of the abrupt 8.2 ka cold event, which is synchronous with the decline of activity events and even more so with the summed probability calibration distribution (Figure 32). This event is well documented within the ice cores and a range of terrestrial indicators as a c. 200-year climate anomaly creating cooler, drier and perhaps windier conditions in the North Atlantic regions (Thomas et al. 2007 ). Its impact appears evident in pollen records from the region, such as at Loch an t-Sagairt on the Isle of Coll (Figure 23; Wicks and Mithen 2014).

Wicks and Mithen (2014) argued that the 8.2 ka event reduced the biomass and productivity of the region while increasing the risks associated with sea-travel and coastal foraging. This either resulted in landscape abandonment or a demographic collapse. Wicks and Mithen preferred the latter, arguing that even slight reductions in birth rates and increase in mortality are likely to have significantly reduced the population levels. The continuing presence of activity events and technological continuity throughout the residential, decline and re-colonisation phase suggests the population did not become locally extinct.

One must, of course, be cautious of this interpretation because the decline in the number of activity events might be an artefact of the archaeological record rather than population density. Acknowledging this, Wicks and Mithen (2014) considered a range of alternative explanations including:

  1. Might there have been a change in technology or settlement size immediately after 8.2 Ka that made Mesolithic sites in the period 8200-6600 more difficult to find?
  2. Might there have been a differential rate of destruction or burial of Mesolithic sites resulting in their relative rarity in the period 8200-6600?
  3. Might the 8.2 ka event have caused a change in the spatial distribution of Mesolithic activity within the study region to locations that are no longer archaeologically visible or have received less intensive fieldwork?

None of these possibilities were found to be compelling. As such a demographic collapse arising from the environmental impact of the 8.2 Ka event appears most plausible to explain the decline in activity events. If that is indeed the case, one might reasonably ask why the Mesolithic hunter-gatherers did not adapt their technology, subsistence and social systems to be more appropriate for the new environmental conditions. Many hunter-gatherers have lived, if not flourished, within far harsher environmental conditions than those likely to have been induced in western Scotland by the 8.2Ka event. The most likely explanation for a failure to adapt is a combination of the abruptness of the 8.2 Ka event and low population densities that inhibited the capacity for innovation. A demographic impact would only have been avoided if new patterns of mobility, subsistence and social behaviour could have been required within a generation. That rate of cultural change appears unfeasible within groups of low-density hunter-gatherers (Shennan 2001). A limited capacity for innovation is indeed evident in the c. 4500 years of the Mesolithic record of western Scotland. There is, for instance, a striking similarity in the stone tool technology from the earliest activity events, such as at Kinloch (CANMORE ID 22202) , Creit Dubh and Fiskary (events before 9000 cal BP), to those at the latest, such as Storakaig (CANMORE ID 304908) (events after 6000 cal BP).

In summary, the period between 8200 and 6600 cal BP is interpreted as one of population decline to low, but still viable, population levels throughout the region - towards the c. 260 rather than the 2370 of the potential population range. Activity events are known in the relative south, such as at Bolsay (CANMORE ID 37342 AND CANMORE ID 37347) , Isle of Islay, and in the relative north, such as at An Corran, Isle of Skye. It is also within this period that the first activity event becomes apparent on the Isle of Oronsay.