7. Discussion: Mesolithic Mobility

The early Mesolithic groups undertook a high amount of logistical and low-residential mobility within large annual territories (Barton and Roberts 2004; Preston 2008). Some of these logistical sites were on the Pennines, where we have noted evidence of investment and possible caching behaviour. This seems to be supported by the fact that a certain proportion of tools were used equipotentially (this is consistent with other sites such as Star Carr, where 17% of formal tools were exaptively used) and therefore, as we have seen, may imply that discarded tools were viewed also as resource nodes (though not in every case).

Moreover, the lithics infer long-distance movement. Indeed, other studies (Hind 1998; Jacobi 1978; Conneller 1999; Spikins 1999; 2002) indicate that raw materials were transported to areas where local resources were scarce or of low, unpredictable, quality. One such area is the Central Pennines, where the bedrock is non-karstic and therefore all flint and chert found necessarily was transported in. In these areas there was a need to create resources within the environment. Material brought in to sites that were visited regularly (persistent places) and discarded at the sites could potentially have been caches or resource nodes in the environment. Similarly, amber and shale beads from Star Carr also imply links with the nearby pre-boreal coast (Clark 1954; Mellars and Dark 1998).

Whether these groups that visited the Pennines operated solely inland or inhabited the coastal environments is equivocal at the present time (Clutton-Brock and Noe-Nygaard 1990; Donahue 2000; Evans et al. 2007). However, dietary and stable isotope research (e.g. Richards and Mellars 1998; Richards and Schulting 2000; 2003; Schulting and Richards 2000) imply that the lithic patterns found in the Pennines may represent two possible patterns: 1) that there are two populations concurrently occupying different coastal and highland environments; or 2) one group was occupying both environments. Lithic raw material provenancing studies (Evans et al. 2007) are currently underway to answer the question but results so far are ambiguous.

My preliminary results and other studies (see Hind 1998; Barton and Roberts 2004) seem to suggest that the emphasis of early Mesolithic lithic transport focused on high-quality blade debitage (and smaller numbers of blade cores), which necessitated a highly selective procurement strategy over a large area and thus may actually reflect the range over which people moved during the annual cycle. (Here I use the word 'debitage' as a noun to mean the products of knapping (i.e. the detached pieces), as defined by Inizan, Tixier and Roche (1992, 84). It does not mean the waste as is commonly thought and mistranslated from the French.)

The notion of large annual territories may also be implied by similar tool styles and technology over many sites. And although tool morphology could easily be affected by resharpening or raw material quality, nonetheless similarities in style and morphology seem to be consistent between Star Carr and both Pointed Stone 2 and 3 (40km away), and the Warcock Hill South sites (over 80km away from Star Carr) for example.

Preliminary results of the analyses of material from Central Pennine sites show that some assemblages have a narrow range of materials (often one or two), while others have many, and Figure 1 shows that many of the raw material source regions are over 80km away and may suggest distinct summer and winter ranges. Moreover, contrary to traditionally held views, mobility and transport extended both to the west as well as east of the Pennines, with raw material sources from both sides being exploited. Chert resources from the south and the north were also exploited, as well as possible materials from the glacial erratics in the Trent basin.

This evidence seems to support the idea of a single large territory in the early Mesolithic covering a large proportion of the north of England. Comparable evidence for large territories can be seen in South Wales where flint, chert and mudstone were moved across distances of over 80km (Barton et al. 1995; Barton and Roberts 2004). Also flints from north-eastern England have been provenanced to as far as South Yorkshire (Young 1987).

These 'territories' are broadly consistent with Smith's (1992) palaeoecologically based population density estimates of c. 0.01-0.02 people per km². It is therefore unsurprising, then, that the archaeological evidence is reminiscent of ethnographic accounts of hunter-gatherers in boreal environments (see Smith 1992; Barton et al. 1995; Spikins 1999; 2002), and is also consistent with boreal palaeoecology. The boreal environments have a large number of potential food sources but floral and faunal populations (and hence food) vary seasonally and are unevenly distributed, with game being found only at low densities. For instance, red deer are solitary and only seasonally gather in small groups (Goudie 1984; Larsen 1980; Winterhalder 1981). This palaeoecology causes 1) low human population density and 2) greater investment in effort to obtain resources and reduce risk, thereby resulting in increased logistical and residential mobility (Binford 1980; Smith 1992; Kelly 1995; Barton and Roberts 2004). It is conceivable that equipotential tool use and caches are part of this response.

Ethnographically, hunter-gatherers can be less mobile if resources allow fishing or trapping (Holliday 1998). However, there is no evidence of decreased mobility in the early Mesolithic because there is no reliable evidence of fishing, trapping, or large-scale storage (Barton and Roberts 2004). Only small-scale storage is seen, in the form of small pits containing charred hazelnut shells, which is not enough for prolonged occupation of one location.

We have noted that many have suggested a relationship between early Mesolithic sites (particularly persistent places) and major river catchments. This lends support to the idea that river systems were the main routes of navigation between coastal and inland environments. One suggested reason for this is that it was easier than moving through the increasingly wooded cover (Spikins 1996; Barton and Roberts 2004; Donahue and Lovis 2006).

The later Mesolithic evidence seems to suggest a reduction in the size of social territories, and mobility appears to have been characterised by more frequent residential moves, with a high level of logistical mobility (Barton and Roberts 2004).

These changes resulted from a number of factors including changes to mixed deciduous forest, which had very seasonal biomes (Roberts 1998) thereby necessitating more residential and logistical moves. In addition, changes in technological and economic behaviour (possibly as a response to environmental changes) may also have been a factor.

The changes in technology and economy also precipitated changes in lithic raw material consumption. For example, an overall reduction in size and quality of raw materials is seen, with a wider variety of and more poor-quality materials represented on sites. This may have been due to decreasing visibility in the landscape of material sources, the reduction in social territories and hence restricted access to good-quality sources, or even that the good-quality sources became exhausted. This resulted in smaller blank sizes and narrow blade microliths (Pitts and Jacobi 1979), and preliminary results of my research seem qualitatively to suggest more economising behaviour and exaptive tool use of good-quality materials owing to their relative scarcity during the later Mesolithic.

Traditional Late Mesolithic mobility models (often based on Scandinavian evidence) appear to be misleading because they are based largely on shell midden material, which imply a concentration on coastal resources coinciding with increased sedentism on the coast and depopulation inland (e.g. similar to Scandinavia). This view is too simplistic because 1) the existence of Late Mesolithic Pennine sites shows that groups were active in the interior; 2) isotopic studies show some individuals remained permanently inland while others were semi-permanently on the coast; 3) there is also evidence of artefacts of coastal origin found on inland northern sites, and 4) no agreed evidence for sedentism. Besides, British shell middens are smaller than their Scandinavian counterparts and where they are associated with lithics (e.g. at Culver Well) they are palimpsests (scattered over a wide area) representing short seasonal occupations (Palmer 1976; 1977; Mellars 1978; Jacobi 1979; Larsen and Jeul Jenssen 1983; Rowley-Conwy 1983; Woodman et al. 1999; Zvelebil and Rowley-Conwy 1986; Finlay et al. 1999; Andersen 1987; 2004).


© Internet Archaeology/Author(s) URL:
Last updated: Wed Jul 29 2009