UDK 903.4(56O.3)"634.2":72.05 Documenta Praehistorica XXXVII (2010) Hamzan Tepe in the light of new finds Bahattin Çelik Department of Archaeology, University of Harran, Sanliurfa, TR bcelik@harran.edu.tr ABSTRACT - In this study, findings from the site at Hamzan Tepe will be evaluated, and the similarities and dissimilarities emphasized by a comparison with the nearby PPN sites. In particular, the focus is on the unearthing of two round planned buildings by treasure hunters as a new discovery and the presence of these two building as the first round planned civic architecture elements in the Urfa region. IZVLEČEK - V razpravi predstavljamo in analiziramo podobnosti med najdbami s Hamzan Tepe in sosednjih najdiščpredkeramičnega neolitika (PPN). Posebno pozornost namenjamo odkritju dveh zgradb z okroglimi tlorisi. Odkrili so jih lovci na zaklade in predstavljata prve profane arhitekturne elemente v regiji Urfa. KEY WORDS - Pre-Pottery Neolithic; round plan; Acheulian; Urfa; Göbekli Tepe Introduction The site at Hamzan Tepe - which was first discovered in 2000 during the Sanliurfa Culture Inventory - lies within the city boundaries of Sanliurfa (formerly Urfa and in ancient times, Edessa) in southeastern Turkey as the cornerstone of the Fertile Crescent. Hamzan Tepe is now located 10km south of modern Sanliurfa city center. Like Göbekli Tepe, Karahan Tepe and Sanliurfa-Yeni Mahalle PPN sites, Hamzan Tepe also lies in the high plateaus on the edge of Harran Plain (Map 1). During surveys of Hamzan Tepe in 2000, pools cut into the bed rock and plenty of flint stone tools were determined. By means of these findings it is understood that this site was used as a settlement both in the Lower Palaeolithic and Pre-Pottery Neolithic Periods. Also, a T-shaped pillar which is also familiar through Nevali Çori and Göbekli Tepe was found. As a result of a new survey in 2010, new round plan architectural remains excavated by treasure hunters found out. Location The settlement at Hamzan Tepe is located approximately 600 meters above sea level at 0 482 41 50 North-South (X), 0 410 42 41 East-West (Y). The settlement was constructed on bed-rock, in a mountainous region called the Fatik Mountains (Fig. 1). In this region, without water sources, there is a high number of calcerous rocks formed as a consequence of erosion. The preserved part of the settlement covers an area of approximately 5000m2. On the northern edge is the large city dump of Sanliur-fa. Harran Plain, the most important plain of the region lies about 1km east of the settlement. There are plenty of flint-stone faces on the southeastern edges of Hamzan Tepe. About 400m west of the settlement, there are basalt blocks. Considering the flint finds from the survey, it is estimated that Hamzan Tepe was used as a temporary open air site in the Lower Palaeolithic and as a minor settlement in the Pre-Pottery Neolithic (Çelik 2004. 3-4). The reason for using this site in both periods is probably due to the proximity of the settlement to Harran Plain and the flint bulks covering around an area of 300m2 next to it. The flint stone finds were plentiful all around the site, some 250 per 1m2. As a result, the site where Hamzan Tepe was settled offers an ideal environment for satisfying the need for flint and basalt. Finds Hamzan Tepe seems to have been inhabited in two different periods. Thus, we prefer to separate the small finds according to their period. Accordingly, we will interpret the material in two different groups as Neolithic and Palaeolithic. The Palaeolithic is represented by 29 (17.2%), and the Neolithic by 140 (82.8%) small finds. The total finds having the features of tools from both periods is 169. Map 1. The site at Hamzan Tepe. Pre-Pottery Neolithic assemblage The architectural remains from the settlement are quite poor. The ground level begins from the bed rock changes between 20 and 80cm. This gave occasion to moving away the architectural units in an area with very little soil. Supporting this idea are the holes on the surface with the diameters changing between 40 and 50cm and with the depth of 8 and 10cm. It is assumed that these holes were made in recent times, because the binder dust quarries are located 2km east of the settlement, where the production of gravel used for the construction of motorways takes place. In these quarries, calce-rous rocks are taken from the region as raw material crushed in gravel making machines. This is also why there is very little removable calce-rous rock in the Fatik Mountains where the settlement is located. Next to a wall constituted of a few stone rows on the surface, an in-situ T-shaped pillar which was partly excavated by the treasure hunters was found (Fig. 2). We know of similar T-shaped pillars from Nevali Çori, Göbekli Tepe, Adiyaman-Kilisik (Hauptmann 2000.Abb. 8-10; Verhoeven 2001.9, Fig. 1a-d) Se-fer Tepe (Çelik 2006.23-25) and Karahan Tepe. The Hamzan Tepe pillar, with a width of 50cm and thickness of 20cm, mostly resembles ante T-shaped pillars Garbage dump T Hamzan Tepe Settlement j. V ; j X. »» ■ - - i «te*-:» tš mmZ Fig. 1. Hamzan Tepe settlement and garbage dump. Fig. 2. An in-situ T-shapped pillar. on the walls of Nevali Çori temple (Hauptmann 1993Abb. 7) and to the pillars in the phase II of Göbekli Tepe (Schmidt 2002.24-25, Fig.1). Furthermore, it shares similar features with numerous pillars at Karahan Tepe (Çelik 2000.7). The presence this pillar in the settlement indicates that the custom of building with T-shaped pillars - which can be observed at settlements like Göbekli Tepe, Nevali Çori, Karahan Tepe, Sefer Tepe and Adiyaman-Kilisik - was resumed here. As a result of a second visit to Hamzan Tepe in 2010, two new round plan architectural remains were unearthed by illicit excavations. One of these buildings is destroyed and its presence can only be understood from the wall blocks (Fig. 3). The other building was found in a stable condition, with only its interior part approaching bed rock being destroyed (Fig. 4). The diameter of the stable one is about 4.5m. The wall stones constitute a single row, each of them being about 1m high and 30-40cm thick. There are no similar buildings constructed with this building technique, but the round plan is an architectural tradition that can be seen in settlements of the Early Pre-Pottery Neolithic (Sicker-Akman 2001.389-394). If we set aside the round plan communal buildings in Göbekli Tepe, this building in Ham-zan Tepe settlement is the first round plan civic building in the Urfa region. In the part of the area again unearthed by illicit excavations, where the ground level has been uncovered (Fig. 5), there are small holes constructed with the pool building technique on the bed rock, placed side by side to form a circle (Fig. 6). The diameter of these holes is about 10 cm, and the depths vary between 5 and 8cm. Immediately adjacent to these holes are three circle-like pools cut into the bedrock (Figs. 7 and 8). There are similar examples with a diameter of 1.5-3m and a depth of 40-60cm in pools found in the northern and southwestern parts of Göbekli Tepe (Beile-Bohn et. al. 1998.47-50, Abb.20; Hauptmann 1999.Fig. 32) and eastern and northern parts of Karahan Tepe (Çelik 2003.44-45). Technology and typology Hamzan Tepe was abundant in flint stone finds dated to the Early Pre-Pottery Neolithic, and a few obsidian finds were also found. Most of these flint and obsidian finds relate to the blade industry. There were 140 small finds from this period - Byblos and Nem-rik points, bifacial cores, stone bowl fragments, flat axes made from river pebbles, and pestle fragments of basalt and obsidian. There were 134 lithic artefact finds flint and obsidians, of which 132 (94.4%) are flint, while 2 (1.4%) Fig. 3. One of the building is destroyed and its presence can only be understood from the wall blocks. are obsidian. Sorting by their handlings, the biggest proportion of flint cut stone tools are blades. The length of these blades varies between 14.5 and 4.5cm, the width between 5.7 and 1.4cm and the thickness between 2.2 and 0.4cm. On the other hand, the length of flakes varies between 6.8 and 5cm, while width ranges between 5 and 2.5cm, and thickness between 1.4 and 0.7cm. There are twenty-two flint stone tools (16.6%) in the whole group of flint finds. The flint stone tools comprised fifteen arrowheads (68.2%), two perforators (9%), one end scraper (4.6%), one spearhead (4.6%) and three hammers (13.6%). The proportion of arrowheads in the well-qualified finds is greater the others. Other than the lithic tools made of flint, there are no obsidian tools. The obsidian finds consist of two blade fragments (see Tabs. 1 and 2). The proportion of obsidian in lithic finding group is about 6.1% with respect to flint stone. As raw material well qualified flint stones were used. Surveys have revealed some flint deposits immediately adjacent to the site, with both bipolar and unipolar cores being found. Regarding the range of colours of the flint finds, forty-five are grey (34.1%); thirty are dark grey (22.7%); twelve are brown (9.1%), forty (30.3%) are light brown, and five (3.8%) cream/beige. About 71% of tools that have blades as blanks have bipolar technology and their cross-sections are tra- so 4S 40 BS 30 25 20 15 10 5 0 Hamian Tepe Flint Stone Finds (45) V. 1 (Ï3) «« (ÎSI (22) ^lî 1 (7) H i FlintStone Blade Flakes Cores and core Waste product Tools rejuvenation fragments Tab. 1. Hamzan Tepe flint stone finds (n = 132). Fig. 4. Building was found in a stable condition, with only its interior part approaching bed rock being destroyed. peze. Some 29% were taken from a unipolar core, and are triangular in cross-section. Two of the cores are bipolar, and two are unipolar; all are of light brown. Cores of similar colour were at Göbekli Tepe (Beile-Bohn et. al. 1998.54) and Nevali Çori (Schmidt 1988.162). While the blade debitages of the bipolar technology cores are 1.6 and 1.2cm wide, the blade debitages of the unipolar technology cores are 2.4 and 1.3cm wide (Fig. 9.1-2). The proportions of trapezoid cross-section blades in the tools with blade blanks is 58%, while the proportion of triangular cross-section blades is 42%. In Hamzan Tepe, with the sum of seven (21.2 %) cores and core rejuvenation fragments are very rare in the whole group of flint stone findings. Only four of them (57.2%) are cores. Apart from these, three (% 42.8) strap blade fragments the one of which is concave (Fig. 9.3-4). The group of flint and obsidian finds consisted of fifteen arrowheads, one spearhead, two perforators, one end scraper, three hammers, twenty-five blades, forty-five flakes, thirty-three waste products, and seven cores and core rejuvenation fragments. Examining the flint tools in terms of typology, it is clear that there are tool types from the Pre-Pottery Neolithic, with arrowheads being most abundant. The most significant of these are the Byblos and Nemrik types. The tools other than these comprise spearheads, perforators, end scrapers and a hammer. Of these sixteen heads, one is a spearhead, eight are Byblos arrowheads, two are Nemrik type (Fig. 917-18) and five are unidentified arrowhead Fig. 5. In the part of the area again unearthed by illicit excavations, where the ground level has been uncovered. 41a), Nevali Çori (Schmidt 1988. Abb. 11.5, 12.3, 14.2) and Göbekli Tepe (Beile-Bohn et. al. 1998. Abb. 23.2). The heads with ventrally retouched hafts and points have partly and dorsally retouched haft and point. The handling of this trapezoid cross-sectioned blade head was from a bipolar blade core. A light grey flint was used for the other arrowhead; part of the point is missing. This arrowhead with all sides dorsally retouched, is trapezoid in cross-section. Generally, this type of arrow heads dated to the Early Phase B of Pre-Pot-tery Neolithic, 8500-7500 BC (Koz-lowski 1999.40-42, Fig. 1). fragments (Fig. 9.13-16). There were eight examples of Byblos type arrowheads (Fig. 95-12). Their lengths vary between 8.6 and 4.5cm, with widths between 2.1 and 2cm, and thickness between 0.6 and 0.7cm. The dorsal and ventral surfaces of the haft of one of these samples were retouched alternately. On one side of another sample's haft, one side was dorsally flattened, while the other sides were ventrally retouched alternately. There were six fragmented arrowheads (Fig. 97-12). Of these, three hafts and three haft and body parts were found. Three were retouched ventrally. Retouches on two examples can be seen dorsally and ventrally on haft. The haft of one of them was retouched inverse alternately. One of these arrow heads was formed on a blade which had been taken from a unipolar core and seven on a blade from a bipolar core. Five light brown, two light grey, and one grey flint stone were used as raw material. Four of these have trapezoid and four have triangular cross-sections. Similar arrowheads were found at Çayônû (Co§kunsu 1999. Arrowheads Perforators Spearhead End scraper Hammer Tab. 2. Hamzan Tepe flint stone tools (n = 22). Unidentified points are represented by five samples (Fig. 913-16). Considering their blade width and thickness, they are the upper parts of either arrowheads or spearheads. Nothing is known about their forms. On their retouches, it is observed that two are ventrally, two are dorsally alternately retouched, and on the other hand one of them is reverse alternately retouched. If we consider the colour of the material, for three light brown flint was used, while two are grey. Two are trapezoidal, three triangular in cross-section. Four were formed on blades taken from the unipolar core. One of these points is a partly broken spearhead. As its direction shows at one side, this point with a blank formed as a massive blade was taken from the unipolar core. This light brown flint stone point's dorsal surface has small retouches, while its ventral surface has retouches only on the haft. The length is 14-5cm, width is 5-7cm and thickness is 2-3cm. The finds include two (9%) perforators (Fig. 9.1920). The borer parts are missing; the lower part of one is also missing. The perforator with the original lower part has a flat butt. One of these sample's auger part is formed from lateral retouches on the dorsal surface; the other's auger is formed with reverse alternate retouches. Light brown flint stone was used for both. Their auger parts are long. The perforators were formed on blades. Considering their blade taking directions, it can be understood that they were from both bipolar and unipolar cores. Both are tra-pezoid in cross-section. While perforators from Hamzan Tepe constituted 9% of finds, there were 7.4% in Sefer Tepe, 17.8% in §anliurfa-Yeni Mahalle, 6.82% in Karahan Tepe, and about 10% in tools from Göbekli Tepe Phase II (Schmidt 2001.51, Fig.9; Beile- Bohn et. al. 1998.59; Çelik 2007. 172). End scrapers are represented with example (4.6%), in fragmentary condition. End scrapers with blades as blanks are bipolar. Dark grey flint was used as material. It is 6.8cm long, 2.5cm wide, and 0.7cm thick. While the end scrapers at Hamzan Tepe constituted 4.6% of finds, this ratio is 7.4% in Sefer Tepe, 13.66% in Karahan Tepe, 6% in §anliurfa-Ye-ni Mahalle and 11.2% in Göbekli Tepe (Schmidt 2001.51, Fig. 9). Fig. 6. There technique on There were three hammers (13.6%) in the flint stone tool group. The sides of these flint stone crushers were dull from use. One is dark brown and two are light brown flint knobs. Their dimensions are 10.2-7.1cm by 6.7-5.5cm by 5.4-4.6cm respectively. There were two obsidian finds (1.4%) (Fig. 9.21-22). There are no tools in this group, which consists only of blade fragments. Both are of transparent black obsidian; the dimensions are 2.2-1.7cm by 1.2 and 1.1cm by 0.3cm respectively. A technological and typological analysis of two fragments shows that both of their handles are formed as blades and as their double directions show they are blades from a bipolar core. Furthermore, similar translucent black obsidian blades from bipolar cores were found at Karahan Tepe, Sefer Tepe, §anliurfa-Yeni Mahalle. There were three samples of stone bowl fragments (Fig. 10.2-4) - all of chlorite - in the form of two rims and one bowl. One is 1.8cm high, 4.9cm wide, 0.7cm thick and with a radius of 3cm. It is not possible to establish the radius of the other sample because only a small part of the rim was found; it is 1.9cm high, 2.5cm wide and 0.7cm thick. The bowl fragment is 3.8cm high, 2.3cm wide, with a thickness of 0.7cm. There are two parallel grooves on the outer surface. On account of the fact that all of them were carved out, there are some traces on fragments. Similar examples of chlorite stone are small holes constructed with the pool building the bed rock, placed side by side to form a circle. bowls were found at Hallan Çemi, Demirköy, Göbekli Tepe, Çayônû, Karahan Tepe and Diyarbakir-Körtik Tepe (Çambel 1974.Fig. 14; Özdogan et. al. 1999. Fig. 61; Rosenberg 1994.126; Rosenberg et. al. 1999. Fig. 3). It is assumed that Körtik Tepe, where these types of stone bowls were found in enormous numbers, was a trading centers for these articles (Özka-ya and San 2003.425). Another item among the stone finds is a limestone object (Fig. 10.1), 7.1cm in height, with appearance of a phallus. From this aspect it resembles a Blanchard phallus1. The head is oval, with a circle delineated with a groove. There is a vertical groove beginning from its shoulder and runs through its base. Its rectangular base is 3.8 x 2.7cm, and it sits squa- Fig. 7. Immediately adjacent to these holes are three circle-like pools cut into the bedrock. 1 http://www.nature.com/news/2009/090513/multimedia/news.2009.473.html Fig. 8. Immediately adjacent to these holes are three circle-like pools cut into the bedrock. The most common tool among the Paleolithic lithic assemblage was the triedral pick (Figs. 11.1-9; 13.2). We may assume the picks were used for digging flint stone knobs in the quarries at Hamzan Tepe. Eleven (38%) triedral picks were found. Five are complete, while the upper parts of three are partly missing. The cross-sections of the points of all the triedral pics are triangular. All are formed on flakes, except for one (Fig. 13.2). The point of the one made from a knob is fragmentary and missing. rely on a level surface. There is a circular hole in a metop formed with two vertical grooves in the base. The hole is 0.2cm deep, and the radius is 0.5cm. Palaeolithic assemblage The survey at Hamzan Tepe yielded bifacials from the Upper and Middle Acheulian Phases of the Lower Paleolithic (Ta§kiran 2002.53), triedral picks, Leval-lois cores and Levallois flakes, massive end scrapers and end scrapers. The Middle or Upper Acheulian Phase is marked by the presence of triedral picks (Figs. 11.1-9; 13.2) and bifacials (Fig. 13.1). In this phase, triedral picks are extremely abundant. These kinds of tool were used mainly for digging up roots (Ta§kiran 2003.248). The first phase of Hamzan Tepe settlement should be dated to the Middle or Upper Acheulian Phase of the Lower Palaeolithic, when the settlement was used as an open-air camp (Ta§kiran 2003.248). Similar finds from this period have been made in Northern Syria and Southeastern Anatolia (Ta§kiran 2003. 247, Drawing 4; Hours 1981.Fig. 4.3). The Palaeolithic finds at Hamzan Tepe comprised 29 (17.2%) of the samples collected. One of these is chalcedon flint, while the remainder are of regular flint. These finds, which in terms of technology or typology date to the Palaeolithic, also share a common feature in having a patina. In this assemblage, eleven (38%) are triedral picks; six end scrapers (207%); four Clactonien flakes; three flake fragments (10.4%); two bifacials (6.9%) - one being partly bifacial; one (3.4%) shapeless core; one (3.4%) Levallois core, and one tool with a Clactonien flake retouched surface. The partial points of the triedral picks vary in length between 8.1 and 5cm, in width between 3.9 and 2.4cm, and in thickness between 2.2 and 1.2cm. One is dark grey and two are grey flint; one has an encrusted surface. The retouches of these three were formed by alternate bifacial flaking on the dorsal surface. The handles of the five complete picks are flakes; the points are triangular in cross section. Four are grey flint; one is light brown. The surfaces of four of them are partly crusted. Three have flat butts, while the other butts have been removed. Retouches of all the samples was generally done with big flak- Fig. 10. Limestone phallus (?) object (1), stone bowl fragments (2-4). Fig. 9. Cores (1,2), Strap blade fragments (3, 4), Byblospoints (5-12), unidentified arrowhead fragments (13-16), Nemrik points (17, 18), perforators (19, 20), obsidian blade fragments (21, 22). ing from the ventral surface or dorsal surface. In some case, only the body and point were partly retouched on them ventral surface. The lengths vary between 11.7 and 9.6cm, widths between 7.2 and 4.3cm, and thickness between 3.5 and 2.1cm. Two of three triedral picks which were found incomplete are of grey, one of beige (cream) flint. The point of one of the picks is missing, while the points of the two are partly missing. The handles of both are flakes. Their points are triangular in cross-sections and the lengths range from 13 to 8.7cm, the widths from 8.3 to 5.5cm, and thickness from 4.7 to 2.4cm. Two have partly encrusted dorsal surfaces (Fig. 11.5-6). The retouches on one pick with a missing point were formed with big flaking on two surfaces; and on another, the surface has been totally removed with flaking. The others have been retouched on only two surfaces. There were six tools in this group (Fig. 11.1-4). Four were found complete, but two with parts absent. Three have blade handles, three have flake handles. Three are crested end scrapers, two are end scrapers, and one of them is a rounded end scraper. Fig. 11. Triedral picks. There are three carinated end scrapers, the lower part of two being absent. The butt of the complete sample is flat. One is from a flake, the other two from blades. Two of these carinated end scrapers are grey and one is beige (cream) flint, the length of the complete sample is 10.6cm the width 5.4cm, and the thickness 2.3cm (Fig.12.1-3). The butts of the two convex end scrapers have blade handles removed by flaking, and are trapezoidal in cross-section. Both have partly encrusted surfaces. The grey flint end scrapers have only one side dor-sally retouched. Their lengths are 12.3 and 10.2cm, their widths 3.6 and 3cm, their thickness 1.1 and 0.9cm respectively. Only a single disc-shaped scraper with a flake handle was found; its butt was removed by flaking (Fig. 12.3). This scraper, which is retouched on the dorsal surface with large and small flaking, is grey flint; it is 9cm long, 6.6cm wide and 3.5cm thick. Only two bifacials were found at Hamzan Tepe. One is complete; the point of the other is absent. One is bifacial; the other is a usual (almond shaped) bifacial (Fig. 13.1). The lower parts of both surfaces of this almond shaped bifacial are encrusted. The sides of this bifacial - the whole surface of which was shaped with big flaking - are dull. It is brown flint; the length is 10.9cm, the width 5.8cm, and the thickness 4.3cm. The form dates this example to the Acheulian Period (Ta§kiran 2002.53). One of them, which is of brown flint, with a flat butt, is formed on a Clactonien flake; this partly bifacial is encrusted on one dorsal side. On both sides of the dorsal surface there are retouches formed with large flaking. On the other hand, on its ventral surface there is flaking only on the lower part. It is 8.7cm long 6.3cm wide and 2.5cm thick. A shapeless core was made of grey flint stone, one side of which is missing. Preparation technique can be seen on its striking platform. The dimensions of this partly encrusted core are 12.2 cm long, 8.4cm wide and 5.4cm thick. There is a considerable amount of flaking on the encrusted surface. The recurrent Levallois core of grey flint (Fig. 14.1) is bipolar and well prepared. Its dimensions are 10.4cm long, 97cm wide and 2.9cm thick. There are traces of encrustation on it. It has closer similarities with the Levallois cores of Tigris region than with the usual Levallois cores of the Upper Euphrates. There are four flakes in the assemblage made by Clactonien flaking technique. All have flat butts; three have partly encrusted surfaces. On the surface of one, there are traces of dehydration (Fig. 14.2). The striking platforms and angle of the flaking side on all are wide. One of them the preparation has been made so that it converts into an end scraper (Fig. 14.3). Two are grey, one beige (cream) and one of light brown flint. Their lengths are between 10.1 and 7cm, widths between 8.6 and 4.7cm, and thickness between 3.5 and 1.9cm. The flaked handle was identified in the Clactonien flakes assemblage. Its dorsal (upper) surface was cor- Fig. 12. End scrapers. rected by flaking. The upper part of the ventral surface was corrected as well. The lower side of the upper surface is encrusted. The tool is 4.3cm length, 6.8cm wide and 2.8cm thick. Parallels were recovered at Ilisu, Karkami§ and from the Birecik region2. Concluding remarks The most interesting aspect of Hamzan Tepe is the co-presence of two periods separated by a long period of time, the first being the Acheulian phase of the Palaeolithic, while the second one is Phase B of the Pre-Pottery Neolithic. The closest settlements contemporary with Hamzan Tepe from the Pre-Pottery Neolithic are §anliurfa-Yeni Mahalle, located about 10km north (Çelik 2000.6), and Göbekli Tepe, about 25km east (Schmidt 2002.24). The small finds are similar to those from the §anliurfa-Yeni Mahalle and Göbekli Tepe settlements (Çelik 2003.37,53; Schmidt 1998.Abb. 5.4, 6.4). The pools cut into the bed rock and the holes used for those pools are similar to pools and holes at Göbekli Tepe and Karahan Tepe (Çelik 2000.7; Beile-Bohn et al. 1998.Abb. 20). 2 Personal Communication with Prof. Dr. Harun Tasskiran. Fig. 13. Bifacial axe (1) and triedral pick (2). Fig. 14. Clactonien flakes (1-3). The current section of the Hamzan Tepe settlement resembles the south-western section of Göbekli Tepe. Two different areas have been excavated there, and the ground level ranges from 10 to 40cm from bed rock. In the first area a stele and the base of a kind of game animal stand in situ (Beile-Bohn et al. 1998. 66, Abb. 30). The second was the base for two central piers cut into bed rock, and next to which are pools and adjacent holes cut in the shape of a complete circle and thought to be used in the pool cutting technique (L.c. Abb.20). Moreover, in both the areas excavated flint waste has been found. This amount of waste, as well as the numbers of pools and holes cut into bed rock on the surface at Göbek-li Tepe is similar to Hamzan Tepe. All these similarities strengthen the possibility that these areas were used as workshops or marked the boundaries of the settlement. In a new survey carried out at the settlement in 2010, two round plan spaces, one of which is stable, were found. The dimensions of the rocks used for this building are the most prominent features differentiating it from round plan structures found elsewhere. It would not be wrong to argue that this building is the one and the only example of a civic building in the Urfa region. Furthermore, prospective excavations in Hamzan Tepe area in areas as yet unexplored could date this settlement to Phase A of the Pre-Pottery Neolithic. -ACKNOWLEDGEMENTS- I would like to thank Mr. Nuri Okutan, Governor of Sanliurfa for providing opportunities during the research in 2010 at Hamzan Tepe settlement during studies for the Sanliurfa Culture Inventory. REFERENCES BEILE-BOHN M., GERBER C., MORSCH M. and SCHMIDT K. 1998. Frühneolithische Forschungen in Obermesopotamien. Göbekli Tepe und Gürcütepe. Istanbuler Mitteilungen 48:5-78. ÇAMBEL H. 1974. The Southeast Anatolian Prehistoric Project and its Significance for Culture and History. Belleten 38(151): 361-377. ÇELIK B. 2000. A New Early-Neolithic Settlement: Karahan Tepe. Neo-Lithics 2(3): 6-8. 2003. Sanliurfa Kent Merkezinde Çanak Çômleksiz Bir Neolitik Yerleçim: Yeni Mahalle. Hacettepe Üniver-sitesi Sosyal Bilimler Enstitüsü, Yayinlanmami§ Yüksek Lisans Tezi. Ankara. 2004. A New Early-Neolithic Settlement: Hamzan Tepe. Neo-Lithics 2 (4): 3-5. 2006. A New Early Neolithic Settlement in Southeastern Turkey: Sefer Tepe. Neo-Lithics 6(1): 23-25. 2007. §anliurfa-Yeni Mahalle Balikligöl Höyügü. In M. Özdogan and N. Basgelen (eds.), Anadoluda Uygarli-gin Doguçu ve Avrupa'ya Yayilimi, Türkiye'de Neolitik Dönem, Yeni Kazilar, Yeni Bulgular. Arkeoloji ve Sanat Yayinlari, Istanbul: 165-178. COSKUNSU G. 1999. Çayônû Çanak Çômleksiz Neolitik Dönem Okuçlarinin Biçimsel Özelliklerine Göre Sini-flandirilmasi ve Çakmaktaçi Okuçlarinin Kullanim Izi Analizi ile Içlevlerinin Belirlenmesi. Istanbul Üniversi-tesi, Sosyal Bilimler Enstitüsü, Edebiyat Fakültesi, Prehis-torya Anabilim Dali, Yayinlanmami§ Yüksek Lisans Tezi. Istanbul. HAUPTMANN H. 1993. Ein Kultgebäude in Nevali Çori, Between the Rivers and Over the Mountains. In M. Frangi- pane, H. Hauptmann, M. Liverani, P. Mathhiae and M. Mel-link (eds.), Archaeologia Anatolica Mesopotamica Alba Palmieri dedicata. Dipartimento di Scienze Storiche Ar-cheologiche e Antropologiche dell'Antichita. Universita di Roma "La Sapienza". Rome: 37-69. 1999. The Urfa Region. In M. Özdogan and N. Ba§ge-len (eds.), Neolithic in Turkey, the cradle of civilization. New Discoveries. Arkeoloji ve Sanat Yayinlari, Istanbul: 65-86. 2000. Ein frühneolithisches Kultbild aus Kommogene. In J. Wagner (ed.), Gottkönige am Euphrat. Neue Ausgrabungen und Forschungen in Kommagene. Mainz: Sonderbnde der Antiken Welt; Zaberns Bildbände zur Archaeologie: 5-9. HOURS F. 1981. Le Paléolithique inférieur de la Syrie et du Liban le point de la Question en 1980. La Préhistoire du Levant 598:165-191. KOZLOWSKI S. K. (ed.) 1999. Nemrik 9 - Pre-pottery neolithic site in Iraq, vols. I-V. Wydawnictwa Uniwersytetu Warszawskiego. Warsaw. ÖZDOGAN M., AYHAN A. and DEMIRTAS A. 1999. Mezraa-Teleilat: Firat Havzasinda Bir Neolitik Çag Yerlessmesinin Tanimi. In N. Tuna, J. Öztürk and J. Velibeyoglu (eds.), Ilisu ve Karkamis Baraj Gölleri Altinda Kalacak Arkeo-lojik ve Kültür Varliklarini Kurtarma Projesi 1998 Yili Çalismalari. Orta Dogu Teknik Üniversitesi Tarihsel Çev-re Ara§tirma ve Degerlendirme Merkezi (TAÇDAM). Ankara: 1-7. ÖZKAYA V. and SAN O. 20003. Körtik Tepe 2001 Kazisi. T.C. Kültür Bakanligi Anitlar ve Müzeler Genel Müdürlügü 24. Kazi Sonuçlari Toplantisi, 27-31 Mayis 2002 Izmir. Kültür Bakanligi Milli Kütüphane Basimevi. Ankara: 423-436. ROSENBERG M. 1994. Hallan Çemi Tepesi: Some Further Observations Concerning Stratigraphy and Material Culture. Anatolica 20:121-140. 1999. Hallan Çemi. In M. Özdogan and N. Basgelen (eds.), Neolithic in Turkey, the cradle of civilization. New Discoveries. Arkeoloji ve Sanat Yayinlari. Istanbul: 25-33. SCHMIDT K. 1988. Nevali Çori: Zum Typenspectrum der Silexindustrie und der Übrigen Kleinfunde. Anatolica 15: 161-201. 1998. Frühneolithische Tempel Ein Forschungsbericht zum präkeramischen Neolithikum Obermesopotamiens. Mitteilungen des Deutschen Orient-Gesellschaft zu Berlin 130:17-49. 2001. Göbekli Tepe, Southeastern Turkey. A preliminary report on the 1995-1999 excavations. Paléorient 26(1): 45-54. 2002. Göbekli Tepe-Southeastern Turkey. The Seventh Campaing, 2001. Neo-Lithics 1: 23-25. SICKER-AKMAN M. 2001. Die Rundhütte als Ursprung Zur Entwicklung erster runder Hütten zum geregelten Rechteckbau. In R. M. Boehmer and J. Maran (eds.), Lux ori-entis. Archäologie zwischen Asien und Europa. Festschriftfür Harald Hauptmann zum 65. Geburtstag. Internationale Archäologie: Studia honoraria 12. Rahden: 389-394. TASKIRAN H. 2002. El Baltalari. Arkeoatlas 1:53. 2003 Paleolitik'te Triedrique Pics (Ügyüzlü Kazmalar) ve Karkami§ Baraj Gölü Alanindan Örnekler. In M. Özba§aran, O. Tanindi and A. Boratav (eds.), Achaeo-logical Essays in Honor of Homo Amatus: Güven Arsebük Için Armagan Yazilar. Ege Yayinlari. Istanbul: 245-252. VERHOEVEN M. P. 2001. Person or Penis? Interpreting a 'new' PPNB Anthropomorphic Statue from the Taurus Foothills. Neo-Lithics 1: 8-9.