Pedra Furada in Brazil and Its 'Presumed' Evidence: Limitations and Potential of the Available Data

Article excerpt


Recent prehistoric and palaeoanthropological debate is focusing new interest on the peopling of the Americas - an old and difficult research-case. Proponents of new sites need to submit their putative 'new and revolutionary' discoveries to a wider judgement, mainly contra sceptics and partisans of more conservative theories. Such a review is helpful, and we are grateful to our colleagues for their considered assessment of the evidence proposed in our Brazilian research. Nevertheless their view of the Pedra Furada evidence has gaps, misunderstandings and ambiguities.

The arguments of Meltzer et al. can be summarized as:

* The chrono-straigraphy is not based on lithologically significant units, which allows some arbitrariness in sub-phase division.

* Though on the whole acceptable, the proposed chronology could date a sequence of natural bush-fires, not a succession of hearths and their by-product.

* The lack of fine sedimentary fractions points to a re-working by water, and aeolian material 'may be present'.

* The cultural nature of the artefacts is questioned:

* The purported tools are the result of an artifical selection, displaying a continuum between totally nature-made and 'possible' man-made specimens.

* The control sample of naturally broken pebbles is too small.

* The absence of technological change along 50,000 years points to a non-human origin.

* The absence of macro-wear does not support a human utilization.

* Likewise, the features 'could be' human, but:

* The granulometry of their elements does not differ from that of the filling.

* The topographic definition is unclear with respect of background 'noise'.

* The correlation between sedimentation rate and number of features (and artefacts) would be positive, suggesting natural deposition of the latter.

* The spatial patterning lacks activity areas or any culturally meaningful distribution.

These are the main points on which we focus attention; some minor observations have their complete explanation in Parenti (1993). All details will be available in the future monograph.

Methodological convergences and discordances

Methodologically, we agree that 'Each pre-Clovis claim is independent' and that 'the case for any claim can only be strengthened by exposing the roots of the scepticism' (Meltzer et al. 1994: 696). Evidence and conclusions about each individual site contribute to a more general model.

Like all archaeological studies, our excavations had shortcomings, which we can appreciate only afterwards. Among these are the different criteria adopted when trimming the 'background noise', the sedimentary matrix of the archaeological layers. Another is the reduced amount of detailed stratigraphic work performed during the removing of anthropic units. To a degree, for some problems, the solution has now to be found outside Pedra Furada. Nevertheless, we do not think these difficulties undermine the main conclusions. Several research goals can be achieved with further use of extant data, as suggested: we are considering, for instance, more sophisticated spatial analysis, the use of micromorphology on remnant balks, and a detailed study of use-wear on lithics.

We do not accuse our colleagues of a prejudice against the Early Man model. This prejudice is too often and unhelpfully evoked by Early Man supporters in response to reasoned criticism. Nevertheless the general tone of the paper is balanced between suspicious investigation and fervent suggestions to a scholar held to be wrestling with a problem beyond his grasp. Undoubtedly, the challenge unearthed was - at the time of the first diggings - unsuspected.

The paper has two general characteristics. A constantly sententious and paternalistic tone is revealed, for example, by abundant use of the word discrete which implies that we have observed only coarse, obvious points; lacking the vast experience of the three authors, of which they often remind us, we were held unable to observe correctly discrete facts. …