The settlement of Haydar includes the area in and around the village of Haydar in the Artuklu district of Mardin province (Fig. 1). The site was visited and surveyed twice in 2020 and 2022 (Kodas et al., 2022). Approximately 2000 lithic artifacts were collected during these surveys. In addition, many artifacts were documented and analyzed in the field. Although the levallois and moustérien industries are dominant in this settlement, fragments belonging to different technologies such as bifaces, blades, blade cores, and discoids were also found (Fig. 2). However, some flakes and their cores, which have important typological similarities with levallois points but are larger, exhibit features that can be defined as a different industry. The aim of this study is the technological and typological presentation of these tools represented by large flakes based on the data obtained from the survey conducted in 2022 (Kodas 2022).
Haydar tool : techno-typology
Chipped stone industries are undoubtedly clear witnesses of the technical characteristics and cognitive infrastructure of prehistoric periods. Through these industries, the technical and typological characteristics of the period can be preserved intact. Not only the typological characteristics of the chipped stone tools but also their technological capacities are analysed. The flakes produced intensively in the Palaeolithic provide important information about the technical characteristics and periodic changes of the period (Inizan et al., 1980; Pelegrin and Roche 2017; Kuhn 2021). The flaking method, which is generally examined through two different technical complexities as simple and complex flaking, enables both cultural and chronological variations to be followed (Leroi-Gourhan 1947; Boëda 2013; Pelegrin and Roche 2017). While there were simpler flaking techniques in the Lower Paleolithic, more complex flaking methods started to emerge in the later phases of this period (Pelegrin and Roche 2017). Some of the chipped stone remains found in the Haydar open-air settlement suggest that we may have encountered a regionally particular techno-typology. The flaking process of the tool, which we named the “Haydar tool” based on the site where it was first found, indicates a preformed flake technology1. In this technology, the main blank to be produced is a large-sized flake, usually triangular (Fig. 3)2. In this context, both the application of the predetermined
concept and the triangular form of the tools in the production of the Haydar tool exhibit similar characteristics with the unidirectional Levallois concept. However, this industry differs from the conventional Levallois concept in some points. The first of these is undoubtedly the size of the tools. The smallest examples of the tools, which we call Haydar tools, have an average length of 13 cm and the largest ones have a length of 21 cm and a width of 6–12 cm (Fig. 4). Tool thicknesses average between 2 and 4 cm. This indicates that these tools are larger than common levallois flakes and even larger than many levallois cores. The second is the technological differences between the two tools. The most prominent point among these is that while the platforms of the blanks of Levallois technology are mostly faceted (Boëda 2013), the platforms in Haydar tool technology are almost completely unprepared (Fig. 5). In other words, the platforms of these tools exhibit a plain- or a polyhedral form. This is also visible on the cores. The striking points exhibit a rounded form and the bulb sizes and radial lines on the negatives on nodules indicate production with hard stone strikers (Inizan et al., 1980).
If we describe the production chain in general, it can be said that the production of these tools is a relatively simple process (Fig. 3–4). Firstly, a striking surface is prepared on a large block, and then a large cortical flake is taken from this striking surface to prepare the core flaking surface. Then two elongated lateral flakes are removed, one on each side of the core surface (sometimes with a cortical surface). Finally, the central flake (Haydar tool), which is located in the middle of the lateral blanks, and which is intended to be produced, appears to be chipped. The production chain in question undoubtedly involves giving triangular forms to many Haydar tool examples. However, the long, angular or rounded examples of the Haydar Tool show that the triangular form is not systematic. At this point, it can be considered that Haydar tool is a tool produced with the Levallois concept, but what distinguishes it from other known occurrences of Levallois point production is both some differences in the production process (no faceting of the platform) and its size. The Haydar tool, which has the same dimensions as many bifacial tools, can therefore be easily differentiated from other Levallois flakes. In other words, the Haydar tool can be defined as a large-sized unifacial blank, sometimes with a pointed distal end. The reason why the production process is similar to the production of levallois points is related to the triangular form that it is given. Here I would like to make clear that the Haydar tool is not always pointed. So, what is interesting here with the Haydar tool is that pointedness was probably not intended but emerged as a side-effect (and this would explain why not every Haydar tool is pointed). This is probably another difference with classic Levallois point production.
Comparison and regionalization
Many Palaeolithic settlements were identified during the surveys in and around Mardin (Fig. 6). Levallois and moustérien industries are abundant in these settlements. In this context, similarities to this tool (techno-typology), which has a production chain similar to the levallois concept and is named as Haydar tool, were also found in open-air settlements such as the Uçurtma Alanı (9 artifacts) and Hırbe Halale (11 artifacts) settlements in the same region (Kodaş 2020; Kodaş 2022). In addition, an example of the mentioned tool was found in the archaeological stratum in the Şıkefta Elobrahimo cave (Uluköy Cave, 1 artifact), where archaeological excavations were carried out for the first time in 2023 (Fig. 7, Kodaş 2023). The tool found in the cave is just above the acheulo-yabrudian layer. At this point, it is thought that this layer dates to the transition phase of the Lower-to-Middle Palaeolithic (Le Tensorer et al., 2005; Muhesen 2012) and it is important for the dating of the tool in question (dating is in progress).