ÇATALHÖYÜK 1995 ARCHIVE REPORT

Report on the 1995 Knapped-Stone

James Conolly

This past season knapped obsidian and flint were collected from three separate activities; from the continuation of the top-scraping program, from new excavations on the northern eminence of the east mound, and from investigations in J. Mellaart's old excavation areas. This report concerns only the obsidian from the new excavations; the flint and obsidian obtained from new work in Mellaart's excavation area has not yet been examined, and this year's material from the top-scraped areas will be discussed within the context of the top-scraping and surface collection assemblage from previous years at a later date.

Method of Analysis

The method of analysis of the obsidian and flint from the new excavation area is similar to the method used for the 1993 and 1994 analysis, although has been expanded in some respects. Debitage is examined in much the same way, but more emphasis has been placed on recording characteristics thought to be diagnostic of the reduction method, be it direct percussion with a hard or soft hammer, or indirect punch, etc. Where possible I have separately noted whether the debitage appears to have been by percussion or pressure. Additionally, I have also attempted to describe the stage in the reduction sequence that individual pieces have come from, by referring to individual pieces as primary, secondary, or thinning elements. This is an experimental addition to my classification system, and may be modified and expanded depending on the results.

Modified pieces were also examined in much the same way as previous years, with a few changes in the format of the recorded data to facilitate computer analysis. A fairly comprehensive attribute analysis has aided in the analysis of the non-formal tools and will (eventually) allow tool variability and distribution patterns to be examined between units.

Quantity of Material

189 pieces of obsidian and flint material were examined from the new excavation area. This represents a total sample of all the knapped lithic material that was collected for analysis, save for obsidian and flint micro-debitage contained in soil samples collected for floatation. The 189 pieces come from 54 separate units and have a total weight of 1675.96 grams. Table One provides a list of individual unit numbers, total weights, counts and average weight of the flint and obsidian from the new excavation area.

Of the 189 pieces, 34 were recorded in three dimensions. Table Two contains details of unit and identification numbers, together with a short description of the individual piece.

Raw Material

181 of the 189 pieces (98.7%) are black/grey obsidian, with the remainder of the pieces different varieties of flint (Graph One). The small proportion of flint in this assemblage is exclusively flake based. A proportion of the flint (4/7 pieces) appears to be tabular flint, the sources of which are unknown although sources may lie both to the west in the Beysehir region, and to the east in Gazientep province. No pieces of flint or obsidian from the new excavation area exhibited evidence of cortical surfaces, indicating that primary preparation of raw material was done elsewhere. As evidence of cortical surfaces is extremely rare at Çatalhöyük, it appears at this point that much, if not all, of the primary preparation of raw material was conducted before the material was imported into the site, possibly at the raw material sources.

Debitage

As noted, the classification of pieces by their debitage class was done in much the same manner as previous years, save for the division of the category ‘blade' into two new categories; ‘regular blades' and ‘non-regular blades' (and their various classes of fragments). Regular blades are defined as pieces that are (or assumed to have been prior to breakage), more than twice as long as wide, and intentionally struck as blades (thus distinguishing them from elongated flakes). Regular blades also display regularity of form in lateral edge morphology and unidirectional scar patterns. Non-regular blades are defined in the same way as regular blades, although they do not exhibit the regularity of form of regular blades and, thus, may have an irregular scar pattern (crossed, multi-directional or bi-directional) and an irregular lateral edge morphology. It was hoped that this distinction, together with other debitage attributes recorded on each piece, may aid in the qualitative differentiation of blades from different contexts (such as early and late contexts where both a qualitative and quantitative change in debitage techniques has already been noted).

Debitage from the new excavation area is dominated by flake and flake-based debitage (Graph Two). Overall, flake debitage constitutes approximately two-thirds of the total assemblage from the new excavation area. Blade debitage, on the other hand, constitutes less than one-sixth of the total assemblage. Chips less than 1cm² are relatively few in number despite all soil being sieved through a minimum of 5mm mesh. Shatter, however, is relatively abundant in quantity, contributing the same proportion to the assemblage as blade debitage.

The size of unretouched obsidian flakes is relatively small, with averages under 2cm in length (Table Three). Unretouched obsidian blades are larger in size than flakes, although show considerably more variation in length, but less in width and thickness.

An examination of proximal ends indicates that the assemblage is divided between flat butts and punctiform butts, with blades mainly (although not exclusively) of the punctiform variety (Table Four). Flakes, while primarily exhibiting flat butts also possess crushed, linear, and punctiform in some numbers. With a greater assemblage, it is anticipated that characteristics of the butt, including butt type, will aid in the differentiation of soft and hard hammer reduction methods. While further data of this type was collected (such as lip morphology) the assemblage is currently too small to make any inferences of this kind. However, it is clear that most of the assemblage is flake-based, and appears to have been flaked by percussion methods (Graph Three). However, a breakdown of reduction method by debitage category demonstrates that there is some variation, with up to 5 pressure-flaked flakes in the assemblage and 9 pressure flaked blades (Table Five). (It should be noted that in this context pressure-flaking means two different things and techniques, as the flakes are removed by pressure as thinning flakes, while the blades are removed by pressure as blanks, not as thinning elements). Debitage was also characterised by reduction stage, using (for the time being) the categories primary, secondary, and thinning (together with an indeterminable category and ‘possible' categories denoted by a question mark proceeding the category name). The category ‘primary' is restricted to those pieces which possess a significant (>50%) amount of cortex on their dorsal surface, ‘secondary' to those pieces with less than 50% or an absence of cortex, and ‘thinning' to those flakes which are thought to be derived from bifacial thinning, blank shaping, or tool retouching activities. While I recognise and accept the criticism that this is a gross system for differentiating ‘stages' of reduction, and there are further categories one could add to refine it, ‘reduction stage' is a provision variable in my classification system. At this point its accuracy is sufficient, as it was anticipated that classifying all debitage in this way would aid in the identification of debitage specific to biface thinning and tool-retouching. If, in due course, the scheme appears to be useful for identifying patterns of lithic debitage, it will be refined.

In the new excavation area, roughly 12% of the debitage are thinning flakes (Graph Four). Once more debitage from different areas of the site have been examined in a similar manner, comparisons can be made and may result in the identification of spatial patterning specifically related to biface manufacture and shaping.

Modified Pieces

Of the 189 pieces of obsidian and flint in the new excavation area, 46 display macroscopic characteristics of intentional retouch or use. Each piece was given a gross functional classification together with an analysis of diagnostic attributes of the modified edge. Some pieces had more than one functional edge, in which case each edge was recorded separately. On these pieces, if the functional classification of the edge modifications are different, the tool was recorded as a combination tool. In total 58 functional edges were recorded on the 46 pieces.

Knife/cutting tools are the most frequent type of modified edge with 14 out of 58 examples (24.1%) (Graph Five). The blank type chosen for knife/cutting implements is almost exclusively blade based, with only 2 non-blade examples. This is in contrast to the 15 scraper/scraping edges (25.9% of the total edge assemblage), where flake based blanks form the majority (8 out of 9 examples). There are a three other functional edge types in the assemblage with fewer representatives; chisel/wedging edges (2/58), notched edges (2/58) and piercer/drilling edges (1/58). There are also 7 modified edges for which I cannot assign a probable function. Pièces esquillées are well represented in the tool assemblage with 6 examples (13.0% of the tool assemblage). Many of the pièces esquillées are well formed and resemble gun flints in shape and size and are manufactured primarily on flake fragments, although there is 1 example of shatter used as a blank. Table Six provides a breakdown of tool functional classification by raw material types, and Table Seven is a breakdown of edge functional classification by modification location. Table Eight is a breakdown of tool functional classification by debitage category.

Discussion

The majority of the debitage from the new excavation area is flake based, with a smaller and unstandardised blade component. Although there are no cores in the sample, it does appear that this assemblage is dominated by a flake based reduction technology. It is possible, however, that blades are under-represented in this assemblage because of their removal, but the evidence from the top-scrape squares suggests that blades are under-represented in the whole northern area of the east mound, supporting the hypothesis that flake debitage is the modal technology in this area. The modified pieces also support this, as the majority are also flake based tools. If the flakes were simply the waste by-products of a blade reduction technology, I would not expect flake based tools to form the majority of the modified pieces. Thus, it does appear that flake production was the primary core reduction behaviour in this area.

This allows something to be said concerning dating of this assemblage, as there is a definite chronological shift in debitage techniques at Çatalhöyük, from an earlier flake dominated industry to a later blade dominated industry, discussed in more detail in last years lithic report. The major change appears to occur between Level V and VI in Mellaart's area, when blades replace flakes as the major type of debitage. By using relative proportions of flakes and blades as a rough chronological marker, it appears that the assemblage from the new excavation area is roughly equivalent to Mellaart's Level V/VI, and is thus earlier in the sequence than the upper levels and surface of the main (southern) portion of the east mound.

The tools from this assemblage are exclusively non-formal types; blanks which have been modified by deliberate retouch, although no significant alterations to the edge profile or plan occur by this modification and edges are chosen for modification by their general shape. This is in contrast to formal tools, where significant changes to blank shape and edge morphology by are performed by retouching (such as seen on a projectile point or a heavy scraper). Given the general distribution of functional types, no specialised activity can be said to be taking place if the tools are taken to be representative of the ‘tool kit' of the area of recovery. This is a small assemblage, however, with an even smaller proportion of tools from any ‘living surface' or floor, so any assumptions made concerning behaviour based solely on the lithic tools must be tentative and subject to a contextual analysis.