6.0 Discussion

6.1 The interpretation of the evidence

The interest of the mesolithic site at Fife Ness lies in its unusual make-up and size, as well as in the particularly specialised lithic assemblage. Mesolithic sites comprise a range of types of structural evidence and include cemetery sites as well as habitation sites and more specialised sites, of varying nature and sizes. Initially, the features at Fife Ness suggested the presence of a small shelter, a wind-break or tent formed by an arc of post-holes, together with a cultural layer, an internal hearth, and associated pits. There are other possible explanations, but this is the one that will be considered first.

Individual mesolithic structures are rare, although they are known eg at Broom Hill, Hampshire (O'Malley 1978) and Blubberhouses Moor, Yorkshire (Davies 1963). In Scotland, where structural evidence occurs, it seems usually to point to the presence of more than one shelter (e.g. at Morton, Fife, Coles 1971; or Kinloch, Rhum, Wickham-Jones 1990). This lack of other single structure sites is most likely to relate to the problems of identifying small sites rather than their actual scarcity, but it does make it difficult to find good parallels for Fife Ness. Further afield in northern Europe the picture is the same: few small mesolithic sites have been excavated, and so sites with single structures are rare, but they do exist (eg Svevollen I, Boaz 1996).

However many structures are present on a site, most mesolithic structures are similar to those at Fife Ness in that they combine the use of stake- or post-holes with evidence in the form of a soil discolouration, as at Morton in Fife (Coles 1971). But actual structure size varies greatly and the structure at Fife Ness is at the small end of the range of mesolithic structures. Furthermore, on many but not all sites in both Britain and Europe, the archaeological evidence indicates that mesolithic shelters were slightly sunken, perhaps suggesting more substantial or longer-lived sites than that under consideration (see, for example the sites on Vega, Bjerck 1989, and also Boaz 1996). With regard to the reconstruction of sites like this, ethnographic evidence may be combined with archaeological material, but it is never possible to do more than suggest ways in which a shelter might have been built up from the traces of post-holes and other soil features. The evidence from Fife Ness could certainly have derived from a stable shelter of some form, with an arc of posts to support a rear wall, an internal hearth, and perhaps a single, more substantial, front post.

It is possible, however, that the features do not relate to an upstanding structure at all. The pits within the arc are comparatively wide and shallow: they suggest posts that would have been large for the size of structure, and yet they are not deep. There were no recognisable post-pipes within them, and the "cultural" layer F46 appeared to seal them. Mesolithic sites are notorious for comprising unexplained pits, and it is impossible to imagine all the daily activities that took place and the features that were required for life in a hunter-gatherer community. Nevertheless, pit digging and their subsequent filling with rubbish or other materials was clearly an important aspect of mesolithic life and some sort of open air activity such as this may lie behind the site at Fife Ness.

One problem with the interpretation of pits is that the fills do not necessarily have anything to do with the original reason why the features were cut. Nevertheless, there are some generalisations that may be made. At Fife Ness the pits comprise various types and sizes. Many are steep sided and have flat bases, though some are more curved. Those within the eastern curve of the arc contained fills that included high quantities of heavily burnt lithics, many of which were also very iron stained. There was no sign of burning on the pit walls but the flint within them had clearly been in a hot fire. Nearby lay a hearth site, together with a series of shallow rounded scoops though, in contrast to the pits, these did not contain much flint, and what there was was not heavily burnt. Should activities involving heat and fire have been carried out in the open around the hearth, burnt flint could have been handily disposed of in a nearby pit, that may or may not have been dug as a part of the activity.

If the pits were not related as contiguous parts of an upright structure it is necessary to address the fact that they have a close relationship as an arc around the hearth. This suggests that each was dug within a short time of the others, certainly while they were still visible. For whatever reason they were dug, the individual pits each have the same relationship to the hearth, and so it would seem that the hearth must have been central to the activities on site. Thus, it is possible that the pits were dug one beside another over a short space of time, hence their positions: side by side in an arc, surrounding the central fire. This might be supported by the observation that during excavation it was difficult to distinguish between the fills of the separate pits, though detailed analysis suggested that the lithic component of each was distinct.

Three pits lay outside this complex. F70 (see plan, Figure 2) is small, but very similar to F72 within the arc, and it did contain some burnt material, though not a great quantity. F61 is larger, but not dissimilar to the larger pits within the arc. It contained 120 flints, and 52% of them were burnt, though they were not as iron stained as those to the east of the hearth. Further away lay F41. This was almost twice the size of the other pits, and deeper, with steep sides and a flat bottom. The fill within F41 could be distinguished into various layers, and it contained a large quantity of lithics, 38% of the whole assemblage. The material within F41 was not heavily burnt in comparison with the features of the arc, and it seems likely that these fills had resulted from very different activities to those of the other pits.


© Internet Archaeology URL:
Last updated: Wed Sep 30 1998