Page 44 - Kennemerland VOC ship, 1664 - Published Reports

NAUTICAL ARCHAEOLOGY, 4.2
the observed patterns are totally meaningless, or wild speculations on how they were produced. On sites where the ship’s structure has survived in a reasonably coherent form, such as the Mary Rose (McKee, 1973) or the Amsterdam (Marsden, 1974), the distribution of the ship’s contents can be related directly to that structure. However, in situations where the material is extensively scattered, its significance is not immediately apparent. The very fact that the assemblage has been identified as a wreck site indicates that it has a certain significance. On the other hand, it is usually very apparent that the recorded distributions do not reflect directly the distribution of the material within the vessel before it was wrecked; a number of other factors have intervened to cloud the issue. However, it may be that there is some residual patterning hidden behind the effects of those factors, which may be recoverable. Further more, the operation of the factors themselves is also of considerable interest to us as maritime archaeologists; it is our duty to understand such processes as they apply to our sites, just as land archaeologists are concerned with the circumstances of the decay of their sites. The only satisfactory way of tackling such questions is by a rigorous and objective programme of analysis, whose operation is explicated for public observation and criticism, so that the reader can appreciate its limitations, and repeat the processes if necessary.
But such a programme will not only improve the quality of the interpretation of this data; it will also determine the levels of accuracy required in recording it. There is no point in expending effort in achieving a precision beyond that which can be accom- modated within the technique; on the other hand the analysis will carry no authority if its constraints are not satisfied by the character- istics of the material involved. It is thus necessary for the director to decide on the analysis required before excavation and recording begins. It is important that all excavators realize that precision is a relative, not an absolute quality. Total accuracy is literally impossible, since any representation is inevitably a simplification of the original, and requests for the greatest possible accuracy
are equally meaningless, since virtually any level of accuracy is possible, given sufficient time and effort. In the final analysis, the thesis of this paper is that it is more satis- factory that the degree of precision should be relative to the demands of a systematic scheme of analysis, rather than being deter- mined by the whim of the director, the fancy of the divers concerned, or whatever.
Bearing in mind all these points, the programme of analysis being proposed for the investigation of scattered wreck sites runs as follows. During the first season of excava- tion, an area of the site should be excavated totally, to a degree of accuracy consonard with the analysis which will follow. This analysis, whose precise form constitutes the main subject of this paper, is designed to discover whether there are any factors of archaeological importance lying behind the distribution of the artefacts on the sea-bed, and if so, to identify them in general terms. If these tests reveal that no factors of archaeo- logical importance are involved, then the rest of the excavation can proceed with surveying
at a very low level of precision, simply for record purposes. If these tests indicate some archaeological significance in the distribu- tions, then further programmes of analysis must be developed and applied to extract this information, and the level of accuracy in recording will be determined accordingly. More work is needed on the type of approach required in these cases, but some possibilities are outlined in the concluding section of this paper. This programme cannot be taken as a justification for anyone skimping on the surveying of their excavations at all stages in the project, since without the meticulous investigation of a sample area at the start, there is no way of knowing what information is being lost. On the other hand, its applica- tion will enable us to avoid squandering our very limited resources on the recovery of useless information. It will be recalled that it was argued above that any patterning on a site would be of archaeological significance, on the grounds that the archaeologist must understand all stages in the development of his site. However, a conclusion that the only relevant factor on a site was the re-sorting of material subsequent to deposition would not,
174