From the archaeological standpoint, there can be little doubt that Bietak’s excavations at Tell el-Dab’a are of high quality, and that the difficulty here is really a matter of methodological details compounded by the lack of adequate publication. In principle, the system used at Tell el-Dab’a is a combination of the typological and architectural approaches mentioned above. As noted, these represent the major currents of stratigraphic thinking in historical excavations, as the geological-pedological approach is more widely used in prehistoric excavations.
In principle, there is general agreement that the stratigraphy plays a fundamental role in excavations, but the treatment of stratigraphy differs significantly, given the different approaches which are viewed as valid by the members of the respective schools. The principal bones of contention are the importance assigned to objects and architecture. Some excavators (including the author) are convinced that stratigraphic understanding is the precondition for excavation, and that excavation depends upon mastery of the stratigraphy. Others are persuaded that stratigraphic analysis is a time-consuming activity which can ultimately only realistically be handled as part of the ‘post-excavation’ analysis. It should be evident that only the first view can be regarded as scientifically defensible, whereas the second approach corresponds to widely accepted practice. In fact, of course, any excavation is a compromise between these two views, as virtually every excavator will aspire to excavating stratigraphically.
The primary difference between the deposit-oriented approach and the others is that the drawing of sections is viewed as a precondition to excavation, and the understanding of the sections is viewed as a precondition to drawing and analysis. Where difficulties arise, sections of test trenches sunk to expose difficult problems can be drawn. In other approaches, the material is excavated and the stratigraphic sections drawn afterward. In this fashion, any information in the sections cannot be used to guide the excavations, and no effort is made to ascertain the stratigraphic sequence in advance.
Harris actually opposes the analysis of the deposits (assuming that the subjective factors will simply not aid the excavation strategy) and likewise opposes the drawing of sections (as there is no reason to believe that any given trench walls will be representative of the sequence excavated). For Harris, the absolute prerequisite for the creation of the sequence is the recognition of the principal relationships between the units: which are superimposed on which, which are cut by which, which are sealed by which. The key is thus recognizing the interfaces during excavation and separating the units accounting to the interfaces. If this is done faithfully, the final listing of all of the excavation units will provide a virtually objective sequence, without interpretation.
However, as noted, most archaeologists assume that a certain amount of interpretive data should accompany the classification of deposits and excavation units. Yet, there are great differences among those who study the deposits during excavation. One approach assumes that the material can be identified in the field immediately. Thus the analytical description of the layers includes interpretive terms such as ‘fill’, ‘collapse’, ‘floors’, etc. The more rigorous approach (advocated by the author) argues that deposits should be described in detail, so that the descriptions can serve to aid in the interpretation (or more frequently, ‘reinterpretation’) of the stratigraphic sequences, so that classifications such as ‘fill’ do not appear until a second step, after the sections have been described and drawn. This approach also argues that any stratigraphic analysis should implicitly include a conscious reflection about the relationships between the deposits while describing and drawing the material in the field.
Thus, the absolute precondition for an excavation is the separation of the ‘excavation units’ from the ‘stratigraphic units’. This applies to both multi-leveled tells, and ‘one-period sites’ alike, as any archaeological site invariably has a stratigraphic depth, even if that depth only includes one chronologically isolated period. The excavation units (labeled ‘lots’, etc.) should correspond to a unit which can clearly link the objects found in an excavation to the description of the material in which they were found and to a position which can be related to both plans and sections. The stratigraphic units (deposits) should be likewise linked to both sections and plans.
In an ideal case, the excavator will be able to make the arbitrary ‘excavation units’ the same as the ‘stratigraphic units’ (deposits), and thus the sequence of excavation units will correspond to the order of deposition. In general, this is simply impossible for practical reasons, and the excavator should aim at simply separating material as carefully as possible; excavation units can be reassembled into other types of analytical and interpretive units at a later stage. In principal, there are two methods of attempting this: one is to use arbitrary excavation units and specify their dimensions; the other is to attempt to recognize stratigraphic units during the excavation and model the excavation units (which are nevertheless necessarily arbitrary) as closely as possible to the deposits.
In principle, the method based exclusively on arbitrary units has been largely abandoned, and most archaeologists attempt to excavate according to stratigraphic units understood as reflecting the deposits. In principle, the easiest method of recognizing the deposits before excavation is to use a test trench (or sondage) to recognize the deposits in section before excavating them in the horizontal surfaces. This demands that the trench always be extended so that it allows the excavator to look below the lower interface of the unit being excavated, and preferably actually sufficiently deep as to allow the excavator to view the entire thickness of the unit below the one being excavated. This method necessarily entails that the sections be drawn and analyzed during the excavation rather than at the close of the season.
When drawing and analyzing sections, the excavator must move quickly as the fresh section is far easier to read than a section that has been exposed for a substantial period of time (particularly in the Middle East). It is probably not advisable to draw too many lines in the section with a trowel as these can mislead one about the stratigraphy when attempting to draw on paper. One can successfully combine the excavation of horizontal exposures with the deepening of a test trench by leaving detailed analysis and drawing until after the day’s excavation. This has an added benefit since it is always advisable to take advantage of the evening light which is softer than either morning or daylight, allowing faint traces and lines to be recognized. Obviously, ideally sections should be arranged so that the section drawn remains visible during the excavation of the deposits in the relevant ‘square’, and thus the excavation should be to the west of the section. Obviously, it would be equally desirable to excavate only when the soil is slightly moist, and the sun not permitted to ‘burn’ sections. Equally obviously, such ideal conditions cannot possibly determine an excavation strategy, and the excavator must adjust to construction schedules, local customs, academic calendars, local topography, and the specific goals of the campaign, etc.
World History
