COMMUN POTTERY IN
FROM THE END OF THE FIFTH
TO THE MIDDLE OF THE SEVENTH CENTURIES AD
SECTION I - Technological system (Chapters 1-3)
SECTION I contains three chapters: digital morphology (1), the study of capacities (2) and text-based operations (3).
Chapter 1 concerns the theoretical aspects of determining features of the pottery form which could be measured, recorded and the results processed digitally. The methodology used (the “Compass System” – TEODOR E 1996) is that designed by the writer for this kind of study, but most of the terms used are well known (height, partial height, rim diameter, body diameter, etc) or already used by other authors (rim angle, inflection point); although, some terms could look strange (body tangent, body arch, thickness). This summary is not the best place for a lengthy description of the system; suffice to say here only that it relies on work with measurement values (M*, taken from drawings), and calculated values, that are in fact relations between M* values, named primary values (P*) and secondary values (S*). For example, PI/a means the proportion of the total height (I) to the body diameter (a). (p. 12-20)
Since for the period and region studied here, complete pottery vessels are scarce, the author chose to use in his own system measurements which could be applied to fragmentary pottery. The Compass System is also detailed enough for capacities calculations.
The result of processing the data depends not only on the quality of the analytical method applied, or the accuracy of measurement, but also the quality of drawing of the objects studied (p. 20-23). The work contains an illustrated discussion of the most frequent errors in shard representation and the devastating effects these errors can have on calculations (and conclusions!). There is also a repetition of the “well-known” principles of field archaeology, such as a detailed report containing raw data, a complete publication of assemblages, without omissions and “selections”. There are also some recommendations for good object photography (see figures 7, 8, 9). The author also lists the advantages of digital publication of archaeological material which are still low-appreciated in Romania.
In the field of processing of data, our imagination is the only limit. The present work discusses only the procedures used to reach the conclusions, avoiding the general matter (for that see ORTON & 1992). The data stored in “digital” format allow us to get averages for every context, layer, cultural level (phase), site, micro-region, for each functional type. We could get also typological averages for well-developed groups of objects. The averages are abstract entities (not corresponding to anything “real”), but they are resuming the most common “behaviour” of the cultural environment and there are much more comprehensive that the individual “look-like” figures (p. 24-25). The comparison could use drawings of the average figures (p. 25-26), for a better intuitive understanding (for example see the drawings of the average measurements of the old Slav types of pottery, third volume, section IV).
The advanced data processing begins with the scatter graphs. For example, one can use the proportion PI/a (height on body diameter) on x-axis and PIs/I (upper height on total height) on y-axis (example). Expressions like “tall shape” or “medium height pot” became arguments that could be any time recovered, and, more than that, one can study the clustering tendencies, which led to the typology (p. 26-30).
Higher levels of processing data are presented with the techniques of “automatic analogy”, “relative analogy” and serration, which requires advanced skills in computers. Based on numeric data stored in the computer, one can query the database about the objects that fit – more or less closely – one specific subject (pot “A”); that would be the “automatic analogy” (see the “tolerance chart”, i.e. what are the ± values for declaring an analogy; in brackets the usual range values). That is a powerful tool, but in a non-industrial society such a similarity is rare or incidental (p. 30-31).
For a better understanding of the place of one particular morphological group, the relationships with other forms or traditions, in a larger territory, we will use the “relative analogy”. The procedure is a little bit more complicated: the numbers are translated into letters, for predetermined intervals (“A” for 0 to 3, “B” for 3 to 6, etc), which would allow us to gain a seriation (see Appendices, section I, table F, for theory, and Appendices, section V, tables A1 (example) and A2 for the results of the application). For the validation of the “relative analogy” as a “relationship”, we need the match of 7 terms from 10 in the table, and a similarity for the rest of 3 (for example, the shape defined as DEFFCBDCAA is “related” to the shape DEFFCBDCBB, but not with AEFFCBDCBB or DCFDCBDCCC); in the example above, the coloured lines are declared analogies, but the black lines are not . The tables for the “relative analogy” are, in fact, a seriation of the morphological groups (types) of pottery shapes (p. 31-32).
We could know, in this way, the cultural position of each morphological group, in relation to other groups from the same site or with artefacts from far away (possibly related). One particular site (settlement or cemetery) is composed of several morphological groups, which could be differently related. What would be the position of that site in the archaeological landscape? Taking account of the morphological groups and their frequencies, one can make a sites-seriation, which is nothing else than a crosstab query (in database terms), following the rule of the minimum difference (example) and dragging the rows into the appropriate position (p. 33-34).
The sites-seriation gives us a hint about how sites are culturally related (or not), but only a hint, and not a rule. A seriation on one site could give, eventually, an idea about the chronology of contexts. Applied over larger territories, the seriation shows more the cultural specifics, than the chronological evolution, and this is more available when we speak of handmade pottery than anything else (this kind of object defines the most conservative producer).
Chapter 2 is a short introduction to the question of calculation of the capacity of the pots (see also TEODOR E 1998 and 2000 b). The measurements taken for morphological purposes can without effort produce also figures for capacity. The calculation formula (see Appendices, section I, G), like any other way to find out the capacity, is an estimation and one could expect errors up to 2%. Basically, the shape is decomposed into simple shapes - the foot, the lower body, the upper body, the neck (what is above the neck diameter is not considered to be “useful capacity” and is not calculated). These all have the form of a truncated cone, and the formula for calculating the volume of each is used, after that the result is corrected for sherd thickness (or section) and arches (or curvatures). Obviously, that means that we can have capacity figures for partially-preserved pots – an important consideration for dealing with Early medieval pottery, partially-preserved vessels may be as numerous as four times as many as complete pots in most of the settlements, and the conclusions based only on the latter can be amended by this method of study (p. 35-38).
The capacity formula – as the entire Compass System – is designed to deal mainly with pots (the single ceramic form statistical acceptable in early middle age), but it is able to deal with any usual form, made from truncated cones. The special shapes – like cylindrical amphoras – need special treatment (the application already exists).
Chapter 3 concludes the theoretical section with a presentation of the alphanumeric subsystem (TEODOR E 1998 a). The name is explicit for any databases user. The subsystem addresses all ceramic processes that are not compatible with numerical procedures: codified names for shape elements, angularity of shaping (“sill”, sharp angles, nervure), rim and base morphology, handles and pouring spouts, fabric, decoration, modelling and firing. The theoretical approach is again analytical. It would be impossible to give a name for each possible rim shape, but we can try to define all possible descriptive elements. Lets take for example three types of rim: with an exterior thickening (“a”), with grooved end (“b”) and with exterior thickening and grooved (“c”). We will make our research easier if would consider the thickening as an “A” element and the grooved rim as a “B” element. We have than the next picture: the “a” is an “A” type, the “b” rim is a “B” element, and the “c” rim is comprised by two elements: “AB”. If ever somebody would attempt to describe all rims it would take many pages, the descriptive elements that we have used needs only two pages (see Appendices, section I, list J for classification, and Plate VII, Figures 21 and 22 for supplementary elements). The system is “open” (i.e. could be developed further) and can be used for any archaeological culture (p. 41-42).
This analytical approach was used also for description of the bottoms of vessels (see Figure 20), for decoration (Appendices, section I, M.1. to M.5.), manufacture features (Figures 23 and 24), handles and related elements and spouts (Appendices, section I, K.1. and K.2.). The chapter closes with some propositions for recording sheets (for field or laboratory). Some parts of the third chapter follows a relatively recent British manual (ORTON & 1992).
further – Section II
back to the Summary index
the Romanian version general index
back to the National Museum Publications
back to the National Museum index
 See Figure 1 and the list of the “primary values”, with calculation formulas, in Appendices, section I, A (Romanian only; the use of the Figure 1 will be helpful for understanding the system). Take note that in Romanian orthography, in writing numbers, comma and dot signs have inverted significations (you’ll read “24.5” for “24,5”).
 Most of the calculation terms are measurement values (noted M*; for measurements list, see Appendices, section I, C), but some of them (the angles) are primary values (P*). Follow the left column for understand the formula (split for each part of the pot, where Vol_D is the upper body, Vol_E is the neck up to the neck diameter, Vol_C is the lower body, Vol_AB is the foot and Vol_F – optional, not used as a general rule – is the neck upper the neck diameter).