Introduction
Used for specific purposes, pits generally contain deposits of varied materials that were employed in everyday life (Chapman Reference Chapman2000; Richards & Thomas Reference Richards, Thomas, Bradley and Gardiner1984; Robert et al. Reference Robert, Bhiry and Bain2022; Rogius et al. Reference Rogius, Eriksson and Wennberg2001). Thus, pits are part of an activity in which they were deliberately dug for a specific reason and had materials deposited in them (Richards & Thomas Reference Richards, Thomas, Bradley and Gardiner1984; Thomas Reference Thomas1999). As evidenced by the Early Bronze Age street fills at Liman Tepe and the Middle Bronze Age street fills at Çeşme-Bağlararası in western Anatolia, many objects used in daily life were thrown into the streets (Şahoğlu et al. Reference Şahoğlu, Çayır, Gündoğan, İncirlili and Güler2020). At this point, the question arises as to why special pits were dug to deposit some everyday objects, bones, pottery and organic materials, while others were simply dumped on the streets or in open spaces. Why were these pits given specific symbolic meanings and why were objects, both special and utilitarian, deposited in them? Considering the placement of materials inside the pits and the fact that these activities were influenced by symbolic schemes, contextual analyses can help in understanding the relationship between the objects and the pits as well as the social behaviours, rituals and ideological attitudes of societies and the symbolic meanings they attributed to the objects (Garrow Reference Garrow2012; Hodder Reference Hodder and Hodder1982; Karamurat Reference Karamurat2018; Pollard Reference Pollard2001). Purposefully dug pits may be regarded as an integral part of rituals, but it is almost impossible to categorize these rituals as performance, either spiritual or secular (Chapman Reference Chapman2000; DeMarrais Reference DeMarrais2014; Garrow Reference Garrow2012; Renfrew Reference Renfrew1985).
Rituals generally involve repeatable formal actions and utterances with a specific order and sequence, involving few variable performances (Alexander Reference Alexander2004; Bell Reference Bell1997; Rappaport Reference Rappaport1999; Tambiah Reference Tambiah1979). While rituals can be associated with beliefs (Bell Reference Bell1997; McCauley & Lawson Reference McCauley, Lawson and Kyriakidis2007), they can also occur independently of religion (Rappaport Reference Rappaport1999; Renfrew Reference Renfrew and Kyriakidis2007; Rowan Reference Rowan2011) and serve different purposes in different social contexts (Alexander Reference Alexander2004; Bell Reference Bell1992; McCauley & Lawson Reference McCauley, Lawson and Kyriakidis2007). In summary, rituals carry symbolic meanings of recurring religious or everyday activities and play an important role in shaping social relationships and hierarchies (Bell Reference Bell1992; Reference Bell1997; DeMarrais et al. Reference DeMarrais, Castillo and Earle1996; Firth Reference Firth1951; Rappaport Reference Rappaport1999; Tambiah Reference Tambiah1979).
Rituals, which involve actions and symbols, are often difficult to understand because of their ambiguity and obscurity (Klingbeil Reference Klingbeil2004; Kyriakidis Reference Kyriakidis and Kyriakidis2007a; Mach Reference Mach1993). Therefore, understanding how rituals can be identified and understood in archaeological remains is crucial (Fogelin Reference Fogelin2007; Garrow Reference Garrow2012; Marcus Reference Marcus and Kyriakidis2007; McCauley & Lawson Reference McCauley, Lawson and Kyriakidis2007; Verhoeven Reference Verhoeven2002). This is because in pre-writing periods, it is not clear which behaviours constitute rituals, where these rituals originated, how they developed and where or how they ended (Renfrew Reference Renfrew1985). Rituals involving postures, poses and verbal practices (Insoll Reference Insoll2004) can be studied by reducing them to appropriate social and material contexts (Kyriakidis Reference Kyriakidis and Kyriakidis2007b; Verhoeven Reference Verhoeven2002). They can include everyday social activities unrelated to religion as well as those related to belief systems (Renfrew Reference Renfrew1985; Reference Renfrew and Kyriakidis2007; Rowan Reference Rowan2011).
Although there are no precise definitions for rituals, which can have multiple and diverse social functions, they can be categorized in different ways (Bell Reference Bell1997; Reference Bell and Kyriakidis2007; Verhoeven Reference Verhoeven and Insoll2011). Understanding rituals from a historical and cultural perspective is facilitated by the amount of residue left in the space where the event takes place, which provides insights into social participation (Greenfield & Jongsma-Greenfield Reference Greenfield, Jongsma–Greenfield, Ivanova, Athanassov, Petrova, Takorova and Stockhammer2018; Swenson Reference Swenson2015). In this context, material culture is a crucial factor in archaeological studies, and rituals are studied by archaeologists (Barrett Reference Barrett, Garwood, Jennings, Skeates and Toms1991; Fogelin Reference Fogelin2007; Garrow Reference Garrow2012; Insoll Reference Insoll2004; Kyriakidis Reference Kyriakidis and Kyriakidis2007a; Renfrew Reference Renfrew1985; Reference Renfrew and Kyriakidis2007).
The concentration and characteristics of ritual pits from the third millennium bce vary from region to region. Feasting pits were the most commonly observed pits from this period in both western Anatolia and the western part of the Aegean. Such pits were discovered in southwestern Anatolia in the cemetery areas of Kesikservi and Karataş-Semayük dating back to the early third millennium bce (Aykurt et al. Reference Aykurt, Böyükulusoy, Benli and Deniz2023; Warner Reference Warner1994, fig. 18), and also outside the settlement at Poliochni dating back to the Blue period (Cultraro Reference Cultraro, Graziadio, Guglielmino, Lenuzza and Vitale2013). In the western part of the Aegean, feasting pits that were dated to the Early Helladic I period and associated with ritual practices were unearthed in the cemetery in Tsepi-Marathon (Pantelidou-Gofa Reference Pantelidou-Gofa, Brodie, Doole, Gavalas and Renfrew2008).
In the second half of the third millennium bce, a number of new practices associated with drinking and feasting rituals began to emerge in western Anatolia (Kouka Reference Kouka, Kouka and Karageorghis2011). Numerous votive and feasting pits from this period were discovered in the citadels of Troia II (Bachhuber Reference Bachhuber2009; Blegen et al. Reference Blegen, Caskey and Sperling1950), Liman Tepe (Erkanal et al. Reference Erkanal, Şahoğlu, Tuncel, Kouka, Keskin and Tuğcu2009; Erkanal & Şahoğlu Reference Şahoğlu, Pernicka, Ünlüsoy and Blum2016), Kanlıgeçit (Özdoğan Reference Özdoğan, Pernicka, Ünlüsoy and Blum2016; Özdoğan & Parzinger Reference Özdoğan and Parzinger2012), Küllüoba (Gündem Reference Gündem2020; Türkteki Reference Türkteki2010; Türkteki & Başkurt Reference Türkteki and Başkurt2016; Türkteki et al. Reference Türkteki, Gündem, Balci, Tarhan, Türkteki, Emlük and Özcan2023) and Çeşme-Bağlararası. Feasting pits were also found in the cemeteries of Karataş-Semayük (Eslick Reference Eslick2009; Mellink & Angel Reference Mellink and Angel1968; Warner Reference Warner1994, pl. 81b, 166a-b), Kandilkırı (Oğuzhanoğlu Reference Oğuzhanoğlu, Şimşek, Duman and Konakçı2015; Reference Oğuzhanoğlu and Şimşek2019) and Çeşme-Boyalık (Şahoğlu Reference Şahoğlu2024). However, the two oval pits unearthed at Seyitömer differ from the pits of this period in that their sides were built with stones and their walls were decorated with red and black painted figures (Bilgen Reference Bilgen2015a, fig. 24).
During this period, there was an increase in the number of feasting and ritual pits in the western part of the Aegean and Aegean Islands. Pits were found at Nea Kephisia and Eutresis dating to the Early Helladic II (Georgousopoulou Reference Georgousopoulou, Marthari, Renfrew and Boyd2019; Goldman Reference Goldman1927; Reference Goldman1931), which were used after a feasting event (Pullen Reference Pullen, Gauß, Lindblom, Smith and Wright2011). Feasting pits were also discovered in Kato Akrotiri on the island of Amorgos (Pantelidou-Gofa Reference Pantelidou-Gofa, Brodie, Doole, Gavalas and Renfrew2008). Additionally, figurine and marble vessel fragments discovered at the sanctuary of Kavos on Keros indicate that there were also sacred areas used during this period (Renfrew et al. Reference Renfrew, Boyd and Ramsey2012).
The tradition of votive or feasting pits was also practised in southeastern Anatolia during the third millennium bce. The shallow pits of Gre Virike I yielded cereal grains and animal bones as well as votive objects such as unfired clay figurines and miniature stone axes (Ökse Reference Ökse2006). In Gedikli-Karahöyük, the skeleton of a decapitated sheep or goat was discovered inside a pit with two buff-coloured cups placed around its neck portion as votive offerings. The animal may have been sacred in some way (Alkım & Alkım Reference Alkım and Alkım1966).
From the Early Minoan period onwards, cemeteries and open-air sanctuaries played a significant role in establishing the social dynamics in Crete (Driessen & Letesson Reference Driessen, Letesson, Hitchcock, Laffineur and Crowley2008). Mountain tops and caves were regarded as sacred places during the second millennium bce and were locations of votives offerings and feasts (Davis Reference Davis, Hitchcock, Laffineur and Crowley2008; Hitchcock Reference Hitchcock, Yasur–Landau, Ebeling and Mazow.2011; Nowicki Reference Nowicki, Laffineur and Hagg2001; Peatfield Reference Peatfield, Laffineur and Hagg2001; Reid Reference Reid, Hitchcock, Laffineur and Crowley2008; Tyree Reference Tyree, Laffineur and Hagg2001). In addition to these sacred places, rituals were also performed in cult rooms or feasting places within palace complexes (Driessen & Letesson Reference Driessen and Letesson2023; Hitchcock Reference Hitchcock, Yasur–Landau, Ebeling and Mazow.2011; Letesson Reference Letesson2013) (Fig. 1).
Ritual pits of Liman Tepe in the Middle Bronze Age
Liman Tepe is situated on a small peninsula in the Urla district of İzmir province, western Anatolia. The mound was inhabited from the Chalcolithic until the end of the Late Bronze Age and during the Iron Age the settlement continued to exist under the name of Klazomenai (Erkanal & Şahoğlu Reference Erkanal, Şahoğlu, Bingöl, Öztan and Taşkıran2012). As the settlement with the longest stratigraphy in the region, Liman Tepe played a significant role in establishing the chronology of western Anatolia (Erkanal & Şahoğlu Reference Erkanal, Şahoğlu, Pernicka, Ünlüsoy and Blum2016; Şahoğlu et al. Reference Şahoğlu, Erkanal, Gündoğan, Tuğcu, İncirlili, Mehmet, Fidan, Türker and Yılmaz2022). It was one of the largest centres during the third millennium bce, with a settlement comprising an upper and lower town. However, it lost its power due to the collapse of the Anatolian trade network at the end of the third millennium bce (Şahoğlu Reference Şahoğlu2005). In layer LMT IV 2 of Liman Tepe, dated to the beginning of the Early Bronze Age 3, more than a hundred pits were discovered at the centre of the settlement. These contained numerous animal bones, seashells, carbonized plant remains and, notably, tortoise bones. Additionally, tankards, amphikypellon, wheel-made grey ware, wheel-made plates and shallow bowls were also unearthed from the pits. After a ritual activity, some of the pits were closed and sealed with flat stone slabs (Erkanal & Şahoğlu Reference Erkanal, Şahoğlu, Pernicka, Ünlüsoy and Blum2016; Erkanal et al. Reference Erkanal, Şahoğlu, Tuncel, Kouka, Keskin and Tuğcu2009; Reference Erkanal, Şahoğlu, Aykurt, Kouka and Tuğcu2012; Kouka Reference Kouka, Kouka and Karageorghis2011; Şahoğlu et al. Reference Şahoğlu, Erkanal, Gündoğan, Tuğcu, İncirlili, Mehmet, Fidan, Türker and Yılmaz2022). In the Late Early Bronze Age 3, during the LMT IV 1 phase, the citadel of the settlement was almost completely covered with stone groups and pits (Erkanal & Şahoğlu Reference Erkanal, Şahoğlu, Pernicka, Ünlüsoy and Blum2016; Şahoğlu et al. Reference Şahoğlu, Erkanal, Gündoğan, Tuğcu, İncirlili, Mehmet, Fidan, Türker and Yılmaz2022).
In the early second millennium bce, based on the example of Liman Tepe, it has been determined that the ritual pits opened in the citadel part of the settlement during the Early Bronze Age continued until the Middle Bronze Age 2, when the digging of ritual pits continued in the centre of the settlement. However, a significant change emerged during this period. North of the ritual pits, for the first time in the Middle Bronze Age at Liman Tepe, a megaron structure was constructed. Interestingly, the dwellings at Liman Tepe during this period were oval structures (Tuğcu Reference Tuğcu2019). In addition to possible sacred megaron structures in the Middle Bronze Age 2, new features in the ritual pits of the region have also emerged for the first time. The use of large stones for the floor of the megaron (Erkanal et al. Reference Erkanal, Aykurt, Büyükulusoy, Tuğcu, Tuncel and Şahoğlu2016, fig. 4), its spacious entrance and the presence of hearths inside suggest that the megaron was used as a sacred structure and had a connection with the ritual pits (Fig. 2). In western Anatolia, it is known that there were certain structures that could be considered sacred at Yassıtepe in the first half of the third millennium bce (Derin Reference Derin, Yılmaz, Can and Işıklı2021) and at Seyitömer in the second half (Bilgen Reference Bilgen2015b).
Numerous pits dating from the second half of the third millennium bce and the first half of the second millennium bce have been excavated at Liman Tepe. The ritual pits of Liman Tepe, which are the focus of this article, are also dated to the Middle Bronze Age and these pits show different characteristics from each other.
In the centre of Pit-1, a flat limestone slab measuring 52×55×2 cm was placed perpendicular to the pit floor and was supported by a few stones (Fig. 3). On top of this stone were pottery sherds mixed with other collected stones. The upper part of the pit was partially destroyed by a later pit and the western part by a fourth-century bce well. Therefore, while it was possible to reassemble the pottery found under the stones in the central part of the pit, only some of the pottery in the destroyed areas and on the upper surface could be reassembled. The pit yielded a handled cup, numerous bowls, Minoanizing bridge-spouted jar, and pots (Figs 4–6). The fill of the pit contained carbonized seeds of einkorn wheat, emmer wheat, barley and bitter vetch, as well as mineralized grape seeds and a Lolium seed (T. Maltas, pers. comm. 2023). Many carbonized seeds of various weed species were also found in the fill. In addition to these seeds, the fill also contained a significant amount of carbonized wood fragments. The carbonized seeds and wood fragments indicate the remains of a small fire.
Situated just to the north of this pit and neatly cut into the fills of the Early Bronze Age is Pit-2, which was filled with fine sea sand. The pit, which yielded only a winged flint arrowhead, a miniature axe, a bronze needle, two spindle whorls and no organic material, differs from the other pits of the period (Fig. 7).
Pit-3 was situated northwest of the two aforementioned pits. The pit, which is oval in shape and slightly protrudes towards the northeast, was dug into the Early Bronze Age fill (Fig. 8). At the bottom of the pit, a round clay altar or offering-table held a large number of pottery sherds, while a sheep or goat jaw and bones from cattle, pig and other animals were piled in one corner of the pit (Fig. 9). A highly significant finding was that, in addition to their large number, many animal bones were broken into small fragments. This discovery was the first of its kind in Liman Tepe, since the pits from both the Early Bronze Age 2–3 and the Middle Bronze Age did not yield such small animal bone fragments. It was also the first time that ceramic vessels left on an offering-table were discovered at the site and in the broader region. The pit yielded one jar with an everted rim, one pot with a straight and simple rim and one double-handled jar. A significant amount of carbonized wood fragments were also found alongside a small number of Lolium seeds.
In another pit, Pit-4, located in the third architectural layer of Liman Tepe, a pig's head was unearthed along with the scattered pottery sherds. Beneath these sherds, the ribs, leg bones and back bones of the pig were found. The body, spine and head of the pig were deposited separately inside the pit. The pig's head, which was positioned in the opposite direction to the back bones and ribs, suggests that the animal was cut up before being deposited in the pit and may indicate that it was a sacrificial offering. Additionally, two cups were uncovered, one near the animal's head and the other among the pottery sherds (Erkanal et al. Reference Erkanal, Şahoğlu, Tuncel, Kouka, Keskin and Tuğcu2009, fig. 3) (Fig. 10).
Discussion
Horwitz (Reference Horwitz1987) summarized the remains of animals presented as burial offerings according to seven characteristics. Among these criteria, the close association of animal remains with the grave or human remains, the selection of certain parts of the animal body and articulated body parts are quite common in feast and votive pits in Anatolia in the third and second millennia bce. The animal bones found in the third-millennium bce cemeteries of Alaca Höyük (Bachhuber Reference Bachhuber2015; Koşay Reference Koşay1951), Resuloğlu (Dardeniz & Yıldırım Reference Dardeniz and Yıldırım2022; Yıldırım Reference Yıldırım2006), Sarıket (Massa Reference Massa, D'Alfonso and Rubinson2021; Seeher Reference Seeher2000) and Kesikservi (Aykurt et al. Reference Aykurt, Böyükulusoy, Benli and Deniz2023) as well as the pits found in the cemeteries of Çeşme-Boyalık (Şahoğlu Reference Şahoğlu2024), Kandilkırı (Oğuzhanoğlu Reference Oğuzhanoğlu, Şimşek, Duman and Konakçı2015; Reference Oğuzhanoğlu and Şimşek2019) and Karataş-Semayük (Eslick Reference Eslick2009; Mellink & Angel Reference Mellink and Angel1968) can be associated with burial offerings. The pits of the second millennium bce at Liman Tepe can be considered as votive pits according to Horwitz's criteria due to both their relationship with human remains and the recovery of whole animal bodies (Figs 2, 10).
The erection of stelae and standing stones was practised in western Anatolia during the Early Bronze Age in Troia (Blegen et al. Reference Blegen, Caskey and Sperling1950), Helvacıköy-Höyücek (Doğer Reference Doğer1995), Liman Tepe (Erkanal et al. Reference Erkanal, Aykurt, Büyükulusoy, Tuğcu and Şahoğlu2018), Hacılar Büyük Höyük (Umurtak Reference Umurtak2023) and Bakla Tepe. It has been suggested that during the second millennium bce, standing stones located near the cult room in Gournia on Crete represented the hilltops and had symbolic meanings (Hitchcock Reference Hitchcock, Yasur–Landau, Ebeling and Mazow.2011). Animal and human figurines presented as votive offerings and found alongside numerous bridge-spouted jars, cups, rhyta, cooking vessels and fireplaces at the sanctuaries situated on the mountain summit of the island also indicate that rituals involving the consumption of liquids and feasting activities took place in these areas (Peatfield Reference Peatfield1992; Reference Peatfield1994; Reference Peatfield, Laffineur and Hagg2001; Reid Reference Reid, Hitchcock, Laffineur and Crowley2008).
The standing stone in Liman Tepe was specifically placed in the centre of Pit-1 and was held upright by the stones placed behind it. It is clear that an effort was made to keep the stone standing and that it was intended for a special purpose. Almost all of the broken sherds recovered from the pit can be reassembled (Figs 4–6). The fact that the ritual pits of Liman Tepe also yielded a bridge-spouted jar, examples of which have been encountered in sacred areas in Crete, sheds light on inter-regional relationships. A large number of bones, seeds and pottery sherds were found in the pit. This suggests that the pits may have been used for rituals and feasting.
The unfired clay altar or offering-table in the middle of Pit-3 was deliberately placed at the pit's centre and pottery placed on it. It was observed that the table was cracked and fragmented as a result of exposure to intense burning (Fig. 9). At the base of the table, there is a pedestal of which a very small part is preserved. There is not enough evidence to suggest that this pedestal went all the way around the entire offering-table. There are holes on the surface of the offering-table which were not completely pierced but rather indented by a hard object. Animal bones were also carefully placed in the pit (Fig. 8). An altar and feasting pit, similar to those in Liman Tepe, were also encountered in Eutresis. The disc-shaped altar and the nearby feasting pit in Eutresis indicate that the remaining animal bones were deposited in the pit after the offering (Goldman Reference Goldman1927; Reference Goldman1931). At the peak sanctuary of Juktas in Anemospilia, the Grotto of Psychro and the palace of Phaistos, libations were conducted on altars or portable libation tables and/or carried in shallow bowls. Pottery and animal bones were uncovered together with these artifacts (Davis Reference Davis, Hitchcock, Laffineur and Crowley2008). It is likely that a similar practice took place in Liman Tepe.
Although the large amount of carbonized remains in Pit-1 indicates the use of fire during rituals, the fact that no evidence of fire was found on the stones or the pottery suggests that the fire was used outside the pit and the burned waste was deposited in the pit afterwards. However, the opposite was true for Pit-3. The exposure of the unfired clay offering-table to intense fire and the burned areas and ash remains in and around the pit suggest that the rituals may have been performed in the vicinity of a fire (Fig. 9). The spread of the fire over a significant area indicates that it may have been a large fire built in an open space. Afterwards, the pottery and the bones of the consumed animals were deposited in the pit. Similar to the example in Liman Tepe, large amounts of carbonized wood fragments found in the ritual pits of Küllüoba (Türkteki et al. Reference Türkteki, Gündem, Balci, Tarhan, Türkteki, Emlük and Özcan2023) indicate that fire may have been used during the rituals of the period. In this regard, both explanations are true: a fire was lit inside a pit to be used during rituals and, as was the case in Marathon, Tsoungiza and Lithares (Georgousopoulou Reference Georgousopoulou, Marthari, Renfrew and Boyd2019; Pantelidou-Gofa Reference Pantelidou-Gofa, Brodie, Doole, Gavalas and Renfrew2008), rituals were performed in another area and the remains later deposited in a pit.
Considering the large amount of carbonized remains, including seeds of wheat, vetch, grape, and the pottery found in Pit-1, it can be inferred that some ritual activities took place in this pit. Thirteen mineralized grape seeds were recovered from the pit. Mineralized seeds may be the result of contact with decaying bone-meat and/or shells (Maltas et al. Reference Maltas, Tsirsti, Margaritis, Margaritis, Oikonomou, Nikita and Rehren2023a), which is consistent with the large number of animal bone fragments found in the pits. This could occur in several ways. If it was a ritual feast, the seeds could have been discarded while eating grapes, or they could have come from grape clusters deposited in the pit as votive offerings. As is known, grapes have been cultivated in the Aegean world since the Chalcolithic Age and were important for their use in winemaking (Garnier & Valamoti Reference Garnier and Valamoti2016; Valamoti et al. Reference Valamoti, Mangafa, Koukouli–Chrysanthaki and Malamidou2007).
The discovery of grape seeds and pressed grape skins in Bakla Tepe, dated to the beginning of the third millennium bce (Early Bronze Age 1), provides evidence that winemaking was practised during this period (T. Maltas, pers. comm. 2023). Moreover, grape seeds found alongside wheat grains in the Early Bronze Age 2 cemetery of Bakla Tepe suggest that grapes may also have been included in burial practices. In particular, the transformation of grapes into wine through fermentation, and its use in ritual practices and as votive offerings, had made it an important symbol in social and religious contexts. Thus, the use of wine in cultural, religious and social rituals demonstrates its symbolic significance in the lives of societies. This also shows how grape had become more than just a fruit and had turned into a valued symbol. Biochemical analyses carried out on amphikypellons at Küllüoba show that they contained fermented products and also reveal the presence of certain substances such as salicylic acid (Türkteki et al. Reference Türkteki, Tarhan, Kara and Tuna2022).
The animal bones and the seeds recovered from the pits of Liman Tepe suggest that rituals were followed by feasts. It is well known that for various cultures, feasts were important occasions which served social, political and economic functions (Adams Reference Adams2004; Dietler Reference Dietler, Dietler and Hayden2001; Reference Dietler and Insoll2011). The ritualized sharing of drink and food by means of feasts brought families and communities together, promoting social cohesion and harmony (Arthur Reference Arthur2003; Dietler Reference Dietler and Insoll2011; Hamilakis Reference Hamilakis2013). Feasts also served as political and ideological activities that created bonds among individuals, families and regional political communities (Hamilakis Reference Hamilakis, Vaughan and Coulson1999; Hayden Reference Hayden, Dietler and Hayden2001; Macdonald & Knappett Reference Macdonald and Knappett2007).
As the glorious Early Bronze Age 3 period continues throughout Anatolia, it came to an end around 2200 bce due to severe climatic conditions also known as the 4.2 ka bp event (Dalfes et al. Reference Dalfes, Kukla and Weiss2013; Massa & Şahoğlu Reference Massa, Şahoğlu, Meller, Risch, Jung and Arz2015; Weiss et al. Reference Weiss, Courty, Wellerstrom, Guichard, Senior, Meadow and Currow1993; Wiener Reference Wiener2014). This climatic development changed the daily lives and economic behaviour of the people around the entire Mediterranean and its surroundings (Maltas et al. Reference Maltas, Şahoğlu and Erkanal2023b; Massa Reference Massa, Erciyas and Sökmen2014; Massa & Şahoğlu Reference Massa, Şahoğlu, Meller, Risch, Jung and Arz2015). Eventually, as a result of this climatic event, the western branches of the Anatolian trade network dropped off the big picture and, following that, a period of decline in the settlements of coastal western Anatolia was recorded (Maltas et al. Reference Maltas, Şahoğlu and Erkanal2023b; Massa & Şahoğlu Reference Massa, Şahoğlu, Meller, Risch, Jung and Arz2015; Şahoğlu Reference Şahoğlu2005; Reference Şahoğlu, Şahoğlu, Şevketoğlu and Erbil2019; Şahoğlu et al. Reference Şahoğlu, Erkanal, Gündoğan, Tuğcu, İncirlili, Mehmet, Fidan, Türker and Yılmaz2022). A notable increase in the destruction of settlements, occurring roughly between 2200 and 1950 bce, suggests a swift rise in organized violence (Maltas et al. Reference Maltas, Şahoğlu and Erkanal2023b). The excavation of pits within the citadels of settlements proved to be a widespread practice, not limited to Liman Tepe (Erkanal et al. Reference Erkanal, Şahoğlu, Tuncel, Kouka, Keskin and Tuğcu2009; Erkanal & Şahoğlu 2016) but also evident at contemporary sites such as Troia (Bachhuber Reference Bachhuber2009; Blegen et al. Reference Blegen, Caskey and Sperling1950), Kanlıgeçit (Özdoğan Reference Özdoğan, Pernicka, Ünlüsoy and Blum2016; Özdoğan & Parzinger Reference Özdoğan and Parzinger2012), Küllüoba (Gündem Reference Gündem2020; Türkteki Reference Türkteki2010; Türkteki & Başkurt Reference Türkteki and Başkurt2016; Türkteki et al. Reference Türkteki, Gündem, Balci, Tarhan, Türkteki, Emlük and Özcan2023) and Çeşme-Bağlararası.
Many pits with associated artefacts were found at coastal western Anatolian sites contemporary with this drought period, and no evidence associated with grain storage was found in any of these pits. Therefore, these pits could be remains of various special deposits that can be associated with feasts and rituals held to end droughts. As is also known from the myth of the second-millennium bce Hittite god Telepinu, the gods organized a great feast to end the drought in the Hittite lands (Hoffner Reference Hoffner1998). It is conceivable that feasts were held that were accompanied by rituals in order to discontinue the devastating drought which had affected the whole of Anatolia and beyond. The continuation of feasting activities and animal offerings during the second millennium bce could suggest that order had not yet been fully established in region. This is consistent with the poor architecture of Liman Tepe in the early Middle Bronze Age. The remarkable increase in the number of feast and ritual pits at Liman Tepe during this period seems to have been an attempt to find a solution to the political uncertainty and drought prevailing in the region. The data obtained from the pits of Liman Tepe suggest that participation in the pit-related activities was at a local level, and that these activities could have been organized by small communities or by the gathering of several groups.
Although a considerable number of animal bones were uncovered in the pits dated to the third millennium bce in Liman Tepe, no whole animal heads were found among them. Animal heads only began to appear in the ritual pits during the second millennium bce. The sheep or goat head and the nearly complete pig skeleton deposited in the Liman Tepe pit show that animals were also offered as sacrifices at that time in this site (Fig. 10). The bowls discovered beneath the pig's head could have been filled with the blood from the sacrifice. A comparable discovery to the first example from Liman Tepe, where two cups were found in close proximity to the severed head and skeleton of an animal, was also made in Gedikli-Karahöyük (Alkım & Alkım Reference Alkım and Alkım1966). This custom, which was mostly practised in southeastern Anatolia, may also have been followed in western Anatolia. In relation to the pig skeleton dated to the transitional period to the Middle Bronze Age in Küllüoba, the left half of a cattle skeleton which bore cutting and scraping marks suggests that it was consumed during a ritual feast, and that the rest of the cattle may have been buried together with the pig (Gündem Reference Gündem2020). Written sources from the Hittites a few centuries later associated pits with the underworld. Rituals were performed in addition to digging pits with the purpose of sending evil underground (Beckman Reference Beckman2011; Collins Reference Collins, Mirecki and Meyer2002). The Hittites mostly sacrificed pigs, dogs, and birds to purify evil doings and, after the sacrificed animals were deposited in pits, the rituals were completed with libations (Collins Reference Collins, Mirecki and Meyer2002; Reference Collins2006).
The animal bones discovered in the Liman Tepe pits indicate the consumption of meat. Generally, feasting pits were more commonly found than votive pits in the Aegean world. The deposition of the utilized materials and the remaining animal bones in the pit after the feasting activities could be considered to be a continuation of the ritual. None of the household refuse pits in Liman Tepe yielded bones that had been broken into small pieces. This shows that not only the meat but also the marrow of the sacrifices was consumed during the feasts. In present-day Anatolia, the unconsumed parts of sacrificed animals are thrown in pits, whereas the unconsumed parts of the animals that are slaughtered for meat consumption are thrown in the garbage. This suggests that the remains of the sacrificed animal could also have been considered sacred.
The careful arrangement of various elements such as standing stones, offering-table or altar, almost complete animal skeleton, animal bones, pottery, plant remains and other artefacts within the pits suggests a deliberate and symbolic significance attached to these places.
Pit-1, Square: Y-8/Z-8, Plansquare: Y-8 IX-X/k, Z-8 IX-X/a, Diameter: 1.13 m, depth: 1.85–1.35 m, locus: 6730. Pit-1 has a high carbon content and a soft soil structure. At the centre of the pit is a standing stone surrounded by pottery sherds. One whole bowl was recovered from the pit (cat. no. 33788/3). Sherds of nine different bowls were also recovered (cat. nos 33788/8, 33790/6, 33793/10, 33793/11, 33793/16, 33793/17), three of which were grey ware (cat. nos 33793/4, 33793/18, 33793/19). In addition, two single-handled cups were identified in pit (cat. nos 33788/9, 33793/9). Inside Pit-1, rim, body and base sherds belonging to four different pots, all of which are large sherds, were found (cat. nos 33793/12, 33793/13, 33793/14, 33793/15). Incised decoration was also observed on the body parts of the pots (cat. nos 33793/2, 33790/7). There is a bridge-spouted jar in the pit, more than half of which is preserved (cat. no. 33790/5). Only one piece of bronze (cat. no. 33788/7) was recovered from the pit. Two wheat, one einkorn, one emmer, one barley, three bitter vetch, one Lolium and thirteen mineralized grape seeds (cat. nos 33788/5, 33790/3) were recovered, together with animal bones.
Pit-2, Square: Y-8/Z-8, Plansquare: Y-8 VII-VIII/k, Z-8 VII-VIII/a, Diameter: 1.32 m, depth: 2.54–1.59 m, locus: 6638. The fill of Pit-2 does not contain carbon or animal bones. The pit was filled with special sand. Interestingly, apart from a miniature axe (cat. no. 33649/1), a bronze needle (cat. no. 77773/6) and two spindle whorls (cat. nos 33703/5, 33764/3), no other artefacts were recovered from this pit.
Pit-3, Square: Y-8, Plansquare: IX-X/i-k, Diameter: 1.25 m, depth: 3.03–2.80 m, locus: 6906. The fill of Pit-3 consists of abundant carbonaceous and soft soil structure. Positioned at the centre of this pit is a probable clay altar or offering-table (cat. no. 35599/1). The pit yielded one jar with an everted rim (cat. no. 35429/13), one pot with a straight and simple rim (cat. no. 135429/16) and one double-handled jar (cat. no. 35429/14). One spindle whorl (cat. no. 35429/12), one sickle blade (cat. no. 35429/6) and one lead ring (cat. no. 35429/1) have been discovered. The pit contains dense accumulations of pig, cattle and sheep bones, neatly stacked in the southwestern portion of the pit (cat. no. 35429/5). Despite the presence of abundant carbonized wood fragments within the pit, only one Lolium seed (cat. no. 35429/3) has been identified.
Pit-4, Square: Y-9, Plansquare: III-IV/i-k, Diameter: 0.98 m, depth: 2.96–2.79 m, locus: 3052. Pit-4 stands out from the other samples as it contains a disarticulated complete pig skeleton. Inside this pit, two bowls were found, one of which is located just below the pig's head (cat. no. 34156/1) and the other is grey ware (cat. no. 34156/5). Additionally, alongside the pig skeleton (cat. no. 34156/3), pottery sherds have been found which are fragmented but can be reassembled (cat. no. 34156/4).
The presence of these features, as detailed in Pits 1 to 4, indicates that the careful design of the pits suggests that they may have had sacred or ceremonial significance within the cultural context of the period (Figs 3, 8–10).
In western Anatolia, only a few architectural features can be considered spatially sacred. In the first half of the third millennium bce, the internal arrangement of the sacred structure at Yassıtepe, along with the figurine and special pottery found within it, may indicate its use for sacred purposes (Derin Reference Derin, Yılmaz, Can and Işıklı2021). With the second half of the third millennium bce, religious and administrative structures such as Troy's megaron II A (Blegen et al. Reference Blegen, Caskey and Sperling1950; Bachhuber Reference Bachhuber2009) and the central building at Liman Tepe (Erkanal Reference Erkanal and Sev1996; Erkanal & Şahoğlu Reference Şahoğlu, Pernicka, Ünlüsoy and Blum2016; Şahoğlu Reference Şahoğlu2005) begin to emerge. The presence of phallus examples in the central structure at Liman Tepe (Erkanal & Şahoğlu 2012; Şahoğlu Reference Şahoğlu2005) and an idol found alongside a phallus in Ulucak (Çevik Reference Çevik2013), dated to the same period, may suggest that specific parts of some places were used as cult areas. The sacred structure with a megaron situated in the centre of Seyitömer is also highly significant. Seyitömer's megaron is also a unique example in the region both for its location and for the pottery that it yielded, which were stored mainly for drinking and offerings (Bilgen Reference Bilgen, Şahoğlu and Sotirakopoulou2011; Reference Bilgen2015b) The ceremonial pit opening into the front room of the megaron structure at Kanlıgeçit indicates a connection to the ritual pit, suggesting a link between the megaron and rituals (Özdoğan Reference Özdoğan, Pernicka, Ünlüsoy and Blum2016; Özdoğan & Parzinger Reference Özdoğan and Parzinger2012).
In addition to the sacred buildings, there are some areas that may have been open-air sanctuaries in the second half of the third millennium bce. The idols found together with tankards and cut-away spouted jugs at Troia Ledge (Blegen et al. Reference Blegen, Caskey and Sperling1950), Bakla Tepe (Şahoğlu Reference Şahoğlu, Pernicka, Ünlüsoy and Blum2016; Tuncel & Şahoğlu Reference Tuncel, Şahoğlu, Marthari, Renfrew and Boyd2019) and Miletus (Kouka Reference Kouka, Kouka and Karageorghis2011; Reference Kouka, Marthari, Renfrew and Boyd2019) in western Anatolia indicate the existence of open-air sacred areas independent of architectural structures, suggesting ritual practices in these areas. Similar open-air sanctuaries are also known in the Cyclades. At the sanctuary of Kavos on Keros, fragments of figures and marble vessels were deliberately broken and left in the area (Renfrew et al. Reference Renfrew, Doumas, Marangou, Gavalas, Brodie, Doole, Gavalas and Renfrew2008; Reference Renfrew, Boyd and Ramsey2012; Reference Renfrew, Philaniotou, Brodie, Gavalas and Boyd2018).
Dating from the first half of the second millennium bce, the megaron at Liman Tepe shows a distinct difference in architectural plan and location compared to the oval buildings in the settlement (Fig. 2). The fact that the main room of the building was paved with very large stones suggests that the building was probably used for other than domestic purposes. In particular, the location of the building, its proximity to the settlement square paved with small stones and its connection to the pits in the open area to the southeast of the building strengthen the possibility that this building was a sacred structure. This arrangement supports the idea that the megaron at Liman Tepe may have had a functionality that included various ritual activities beyond its daily use.
Conclusion
The finds in the feast pits, such as carbonized seeds, animal bones and pottery, suggest a strong link between rituals and food and drink. This not only offers fascinating insights into the social interactions of the community, but also provides valuable information about shared ritual food consumption and the economic dynamics of societies. While the artefacts recovered from these pits shed light on the trade relations of the period, the plant remains shed light on the agricultural activities undertaken, and the analysis of animal bones provides an insight into whether the species were reared or hunted. In essence, the study of these ritual practices reveals not only cultural aspects, but also important facets of the economic and subsistence strategies of the community at the time.
Standing stones, altars and meticulous arrangements found in the pits indicate that the rituals have a symbolic and special meaning beyond a simple feast or ceremony. These arrangements provide important insights into the rituals of the community. The changing characteristics of ritual pits from the Early Bronze Age to the Middle Bronze Age are remarkable. Especially the Middle Bronze Age establishes a connection between megaron structures and ritual pits, providing important information on how spatial arrangements and rituals influenced each other. As seen in the example of Liman Tepe, the megaron structure opening to the open area where the pits are located can offer clues about the community's religious organization and spatial utilization. The relationship between megaron and pit can also be observed at Kanlıgeçit in the second half of the third millennium bce. However, all the structures from this period at Kanlıgeçit consist of megarons, and the pit located in front of the megaron indicates a connection with the structure. At Liman Tepe, on the other hand, while the structures in the settlement have an oval plan, a special megaron was built with a clear connection to the pits (Fig. 2). In this context, the relationship between megarons and pits observed in the second half of the third millennium bce appears to be strengthened in the first half of the second millennium bce.
The practice of placing bowls under decapitated animal heads is not observed in western Anatolia during the third millennium bce. However, similar practices were observed at Gedikli-Karahöyük in southeastern Anatolia. Animal bones were generally found scattered in pits dated to the third millennium bce, but by the beginning of the second millennium bce, complete or almost complete animal skeletons were found in some pits. Although such occurrences were common in southeastern and inland Anatolia, they did not appear in western Anatolia until the beginning of the second millennium bce.
In the second half of the third millennium bce, interregional trade was not limited to the transport of exotic goods, but also included the spread of rituals over a wide geographical area. In this respect, it could be argued that a number of different practices typical of Crete and inland Anatolia, as well as some rituals unique to western Anatolia, were carried out in the same area. Since western Anatolia is an ‘in-between region’ due to its geographical location between central Anatolia and the Aegean world, it is only natural that the rituals of this coastal area reflect the ritual practices of these two separate regions. Although a completely hybrid system has not yet been identified, it seems that the standing stones and feasting customs observed in western Anatolia during the third millennium bce continued as previously practised, and in addition, Minoanizing vessels were also used in these feasts. In conclusion, due to its location and importance, western Anatolia shows the influence of both inland Anatolia and Minoan Crete in its various pits. Western Anatolia was able to promote unique cultural interactions in this area by adopting the ritual practices of its neighbouring regions.
Acknowledgements
I would like to thank Professor Vasıf Şahoğlu for his support of my research. Special thanks to my colleagues Mustafa İncirlili, Furkan Yılmaz and Ece Paşalı, who worked with me in the excavation area. I would like to express my gratitude to Dr Tom Maltas for his invaluable insights in evaluating the botanical findings during our fieldwork. I would like to thank Douglas Faulmann (INSTAP) for the drawing of the pottery and Süheyla Değirmenci-Ünal for the drawing of the pig. Special thanks are due to Dr Vasso Choleva for her expertise in evaluating the animal bones, and I am grateful to Elif Yıldırım for her dedication to the restoration of the pit pottery. While it may be impossible to mention every contributor individually, I would like to thank everyone involved in the creation of this article. The Liman Tepe excavation is being carried out as part of the Izmir Region Excavations and Research Project (IRERP) under the framework of the Ankara University Mustafa V. Koç Research Center for Maritime Archaeology (ANKÜSAM), with permits from the Turkish Ministry of Culture and Tourism. The excavations were generously supported by the Ministry of Culture and Tourism, Türkiye; Ankara University Rectorate; Institute for Aegean Prehistory (INSTAP); Ankara University, Faculty of Languages, History – Geography (DTCF); INSTAP-SCEC and General Directorate of State Hydraulic Works (DSİ). For more information on ANKÜSAM and IRERP, see: http://ankusam.ankara.edu.tr