Taking Microliths into Account

Article excerpt

I thank Crombe et al. for their comment on an earlier paper (Vanmontfort 2008) and the editor of this journal for providing me with the opportunity to reply to their critique. They use two intensively surveyed and studied areas in the lower Scheldt region to compare the microlith-based method with a site-based approach. Actually this comparison nicely illustrates the potential of the method. My Figure 3 compares the calculated frequencies of sites and microliths--which is not quite the same as the 'Mesolithic use' of the, se regions--over the Mesolithic period for the two regions. The results correspond remarkably well. This agreement is perhaps not surprising, since the presence of microliths is also one of the most determining factors in the attribution of sites to each of the Mesolithic phases. However, the results from the two methods might easily have diverged from each other. Contrary to Crombe et al.'s expectation of correspondence, the distribution of a particular artefact type may offer different and complementary information to site-based data (Vanmonffort 2008: 250).

The microlith-based method claims to provide an image of inter-regional land use variability on a large scale, without claiming to grasp 'the behavioural complexities (...) within particular Mesolithic phases'. This implies a different scale than that Crombe et al. prefer to use. It should be kept in mind that the microlith-based method was developed to tackle the problem of Mesolithic hunter-gatherer activity in the loamy region of the Lower Rhine Area, where detailed information on sites and their ecological context is not available. At this scale, the Somme Valley data approximates to what is expected in the Scheldt and Meuse basins (see also Crombe & Cauwe 2001: 52, Figure 3) and covers the endre Mesolithic period. Possible deviations between the Somme Valley and Scheldt Basin regions are supposed to have had a similar effect on the data from e.g. Hainaut and south-west Brabant and thus do not impede a relevant inter-regional comparison. Evidently, adjusting the method to a new and better chronological framework is part of the continuous improvement in the application of the method. The work of Crombe and collaborators during the last decade has added valuable data and it can only be hoped that it will also yield sufficient data for the later Mesolithic stages in the near future, thus making available a better chronological reference than the one used in the original paper. As for the results of Robinson (2009) (2), I am anxious to learn how the striking differences in the technological trajectories of Somme and Scheldt affect the microlith-based method.

The sampling in the Hainaut area can be regarded as representative for archaeological data that can be collected by fieldwalking, except perhaps for its north-western part, for which I would not want to question Crombe's claim. The rest of the area has been covered by several fieldwalkers whose data has been gathered recently by M. Van Assche (2005: 48). The survey zones together almost attain a full coverage of the area and focus on all landscape positions, just like the archaeological surveys and excavations that accompanied the construction of a gas pipeline, a highway and a railway, all three crossing the area (Van Assche, pers. comm.). These surveys yielded some rare sites in valley position, such as Ormeignies-Autreppe, Masnuy-Saint-Pierre and Rebecq-Le Spinoi (Van Assche 2005: 50).


The survival problem is fundamental and related m the taphonomic processes in this region, which includes plateaux, slopes and valleys (see also Vanmontfort 2008: 150). Such processes, however, are supposed to have played a similar role over the entire loess belt. The differences in pattern between Hainaut and the neighbouring south-west Brabant area thus seem to be real. The Kleine Gete and Hesbaye areas, mentioned by Crombe et al., were not part of the analysis, simply because no reliable datasets were available. …