Internet Archaeol 5. Wickham-Jones & Dalland 6.3

6.3 Cultural and chronological comparisons

As has already been noted the majority of mesolithic sites are larger and more complex than Fife Ness, and they have much larger and denser lithic assemblages. This does not mean that it is particularly unique, merely that small sites such as this are rarely found. Fife Ness is, therefore, important in that it provides an insight into an aspect of mesolithic life that is rarely touched upon. The mesolithic is traditionally characterised as having a mobile life-style incorporating a variety of different settlement and activity sites of varying degrees of permanence, size, and complexity: Fife Ness clearly fits into a different slot within this compared to larger sites, such as its neighbour in Fife at Morton (Coles 1971).

The region of Fife lies between two major river estuaries, the Forth and the Tay, and the inhabitants of Fife Ness would have had easy access to both, as well as a rich hinterland. There is, however, little evidence for the mesolithic exploitation of the area. To the north, the site of Morton is well known. It is a much larger site than Fife Ness, and traces of several structures were uncovered, as well as hearths and midden material with organic preservation. The lithic assemblage made use of a variety of stones and included many different tools, though like Fife Ness it was based on the production of blades and included both microliths and scrapers. The microliths included many different types, and crescents were present, but seemingly not important. Detailed analysis of the lithic materials used at Morton suggested that its inhabitants ranged over a wide territory, in the course of which they amassed a variety of stones to be worked (Coles 1971). Nevertheless, little physical evidence has been recorded for mesolithic sites elsewhere in Fife. This may be surprising in view of the available resources and long period over which mesolithic occupation took place, but it is more likely to be an indication of the unsuitability of recent land use for the detection of stone scatters, rather than of the actual lack of mesolithic penetration into Fife. In this respect Fife Ness is rather peripheral, being right on the modern coast, at the most exposed tip of Fife, but it is likely to be an indication of the presence of other larger, undiscovered sites, from which its occupants set out. Its coastal location has clear advantages in terms of the exploitation of marine resources, and there was also local flint. At Fife Ness it was not necessary to exploit the wide range of materials used at Morton.

On a wider basis, Mellars (1976) has combined site size and complexity with the make-up of the lithic assemblages in order to broaden the interpretation of the mesolithic settlement of Britain. Whatever the interpretation of the features at Fife Ness, it is possible to apply his analysis to the site, and this helps to place it in a more general context. In his terms Fife Ness is a Type I settlement, namely a site occupying a restricted area, and it has a Type B, balanced, lithic assemblage, namely one in which microliths are important, but are balanced by other modified tool types, particularly scrapers. This is interesting, because in general he notes that smaller sites tend to have Type A, microlith dominated assemblages, though he does list several "intermediate" sites, many of which are in coastal locations, such as Lealt Bay on Jura (Mercer 1968). The intermediate status of Fife Ness is reinforced by the microlith assemblage which Mellars notes is usually dominated by one specific microlith type in Type A assemblages. Interestingly, he does not list any crescent dominated assemblages, rather sites where scalene triangles, rods, or trapezoids are prevalent.

In some ways, Fife Ness is therefore rather anomalous in terms of Mellars's analysis, being a very small site with a range of other tools to balance the microliths, and with the microlith assemblage dominated by crescents. But he does note in his conclusions the possibility that:

sites occupied on a more ephemeral basis may remain largely if not entirely invisible from the prehistorian's point of view. (Mellars 1976, 397),

and this may be just what is represented here. His work does reinforce the suggestion that there are few parallels to Fife Ness, and this remains the case, even when more recently excavated sites are taken into account. Unfortunately, the evidence for mesolithic settlement in Scotland has generally comprised either larger and more complex sites such as Morton or Kinloch, or largely unstratified scatters of stone tools which often come up during the excavation of later sites such as Eskbank (Hanson, pers comm) and Springwood Park, (Wickham-Jones, forthcoming). In this respect the work undertaken by large scale survey and excavation projects such as the Southern Hebrides Mesolithic Project is particularly interesting as it aims to identify a variety of site types within a certain geographical area, and excavate key locations (Mithen & Lake 1996). Nevertheless, this work is still confined to western Scotland, so that sites like Fife Ness and Morton remain in isolation.

Given that exact parallels to Fife Ness are difficult to find, it is still possible to look at the general nature of the lithic assemblage in the light of lithic material from other mesolithic sites. Mesolithic sites across Scotland and further afield in Britain and northern Europe are often characterised by the use of a particular suite of flaked stone tools, usually including unmodified blades and flakes as well as microliths, scrapers, and edge retouched pieces. All these artefacts occur at Fife Ness, although there is less evidence for the on-site production of tools than is often the case, and this distinction in itself is interesting. Of the modified tools, the microliths are important for scholars of the mesolithic, partly because of their general ubiquity, and partly because they fall into certain well-defined categories of shape (Wickham-Jones 1990; Finlayson et al 1996).

While the meaning of the individual types of microlith is still a matter of great debate (Mellars 1976; Finlayson et al 1996), their presence, or absence, on specific sites is of interest. Crescents are well known on many mesolithic sites (eg Morton, Fife, Coles 1971 fig 11; Lussa River, Jura, Mercer 1971, fig 8), but they are not usually present in any great quantity, much less dominant. Only at the large and complex site of Kinloch on Rhum has an area dominated by crescentic microliths been recorded, and interestingly there was a suggestion that they may have been associated with scrapers on this part of the site (Wickham-Jones 1990, 105-16). To the south of Rhum, at the site of Gleann Mor on Islay, crescents have been reported as common, though scalene triangles were still dominant (212 microliths in total - 50% scalene triangles, 13% crescents; Finlay pers comm). Interestingly, this was a small site on a spur not far from the coast, and it has been interpreted as a short lived hunting camp (Mithen & Lake 1996). Elsewhere, it seems that not only are crescents generally rare, but they are even rarer as sites get further away from the sea. Crescents comprised only 8% of the microlith types recorded by Mulholland (1966) in the Tweed Valley and there was only one, out of twenty eight microliths from the recent excavations at Springwood Park (Wickham-Jones forthcoming). Further south, in the Pennines, Radley et al (1974) recorded very few crescents, and they are not dominant on any site. In contrast to many mesolithic sites Fife Ness has no scalene triangles, while rods, backed bladelets, and fine points, make up only a tiny proportion of the assemblage. It would seem that this assemblage was a different and a very specialised one.

Click on map to access section 7.
Figure 10: Interactive map and database of C14-dated mesolithic settlement sites in Scotland (after Ashmore et al forthcoming)

With regard to date, Fife Ness lies among the earliest mesolithic sites of Scotland (see Figure 10 above). It is therefore to be included among the earliest evidence for the settlement of Scotland. Most of Scotland's early sites lie on the west coast, so Fife Ness is particularly interesting as it confirms that early settlement was not confined to the western seaboard. However its near neighbour in Fife, at Morton, produced dates that are much more recent, thus confirming, not surprisingly, that settlement of Scotland's eastern lowlands continued throughout the mesolithic period.

The lack of other mesolithic sites in the region means that it is difficult to fit Fife Ness into any contemporary context, but there are some general points. It is generally assumed that the sites relating to the earliest settlement of Scotland will reflect in their artefact assemblages the area from which that settlement came. Thus, the lithics from sites on the west coast have been compared with material from the Irish Sea Province (Wickham-Jones 1990), and sites to the east, in the Tweed Valley, have been noted to include broad blade microliths indicative of parallels in northern England (Wickham-Jones, forthcoming). Though it is also in the east, and close to Morton where the lithics have been characterised as broad blade in nature, the assemblage from Fife Ness comprised only narrow blade material, and a restricted range at that. This surely emphasises the specialised nature of the site, rather than casting doubt on the accepted status of broad blade material; though it should be noted that there is a lack of any recently excavated, broad blade sites in eastern Scotland of sound context.

As for the lithics themselves, comments on their date have varied from surprise that crescentic microliths should occur so early, to interest in the confirmation of another early date for crescents. It is clear that a wide range of different microlith types was well established in Scotland by the mid-eighth millennium. The tools in use were no doubt affected by the geographical area, and cultural affinities of their makers, as well as by the specific nature of the site and uses to which they were to be put. Fife Ness has perhaps shown that despite the advanced nature of archaeological theory, especially where hunter gatherers are concerned, it is impossible to predict both the type of site, and nature of the artefact assemblage that "should" be occuring at any one time in any given place.

Returning to the overall nature of the site, small sites are often interpreted as specialised encampments, that may only have been occupied for a matter of days while some particular activity or activities were carried out, possibly without even the need for a shelter of any sort (as in Mithen and Lake 1996). They have been found in a variety of locations from safe harbours on the coast (Bjerck 1989) to along elk migration routes in upland, mountainous regions (Boaz 1996). The lithic assemblage from Fife Ness would clearly support the interpretation of the site as a specialised activity centre, where tool manufacture played a secondary role, and the structural evidence would also be in agreement with this. If the excavated features derive from a shelter, it was not large and appears to have been isolated and used for something that involved high temperatures and in which a general amount of lithic waste got burnt; if they did not derive from a structure then some sort of repeated activity involving heat, pits and the disposal of burnt lithic waste seems to have been undertaken over a short time, in the open. Neither scenario is likely to have involved many people, nor lasted for long.