Small carnivores from a Late Neolithic burial chamber at Çatalhöyük, Turkey: pelts, rituals, and rodents

Stone (beech) martens

Distal parts of limbs (n = 37) have been found in five sets of articulated elements (Table 4), representing left forelimbs (Fig. 7), right forelimb, metapodials without proximal ends that preclude accurate identification, and digits (A1–A5, respectively, in Fig. 8). Among these, no selective pattern can be observed since both the left and right paws are present. The minimum number of stone marten individuals is two, based on the complete fourth and fifth right metacarpals. The left paw is the most complete, where, additionally, individual elements show cut marks and evidence of pathology. Cut marks are located on the first and fifth metacarpals (medial and lateral parts of the proximal shaft, respectively), on two of the three indeterminate metapodials (the anterior surface of the middle part of the shaft), and on the first phalange (the posterior surface of the proximal shaft and cut-off proximal end of the phalange) (Table 4). Additionally, the first phalanx also displays cut marks on the posterior surface of its proximal shaft (Table 5), all of which are the result of skinning. Bones from the articulated foot with such marks were dated to 7340 ± 40 BP (OxA-X-2499-7) (Fig. 9). The animal bones at Çatalhöyük have low collagen contents, which is also the case here (the sample gave a slightly lower (3.4 mg) than ideal (5 mg) yield of collagen after ultrafiltration). Pathology can be seen on the distal end of the first phalanx as bone overgrowth that could be associated with exostosis, as well as on the distal end of the right fourth metacarpal as inflammation (Tables 4 and 5).

Open image in new window

Open image in new window

Open image in new window

Although stone marten remains were found in nine flotation samples, they occurred in greater quantities in one sample (8683.s23) (Table 2). This sample is associated with the almost completely preserved skeleton of an infant (unit 17622) located in the southeast part of the burial chamber.

Red fox

Common weasel

The size and morphology of the completely preserved mandibles (NISP = 2), which are also the only elements present in the assemblages, allow them to be identified as belonging to common weasels (Table 9). In both cases, the fourth premolar is erupting, indicating the juvenile age of the individual (MNI = 1)—most likely 8–10 weeks of age (King and Powell 2007).

Taxonomic significance

The carnivore assemblage from the burial chamber in the TP Area at Çatalhöyük consists of stone martens, red foxes, and common weasels—species that have been previously noted at the site. However, common weasels, for example, have been positively identified in a small number of cases (comprising 2% of the total taxa in the microfauna assemblage, according to Jenkins (Jenkins 2005), and less than 0.05% in the faunal analysis by Russell et al. 2013), all exclusively from earlier levels. Thus, our results extend the knowledge regarding the contribution of small carnivores to the faunal composition of the Late Neolithic.

Stone marten remains can be easily misidentified as belonging to pine martens (or even other mustelids). In our results, the morphological features and metrical data for the cranial parts and metapodials—where the completeness of the specimens is striking—allow us to conclude with certainty that they came from stone martens. This is best illustrated by the distance between both oval mental foramens of the mandible (2.1 mm), which clearly falls into the stone marten range (2.0–3.4 mm) and outside the pine marten range (5.9–9.6 mm) (Anderson 1970). This feature is very distinctive and useful in species identification, since the ranges for these two species do not overlap and are thus mutually exclusive. In addition, the metapodials of the stone marten are proportionally shorter and stouter (Anderson 1970), as is the case here, than for the pine marten. In addition, the size of the metacarpals corresponds to that of an average-sized stone marten (Anderson 1970).

The potential of taxonomic distinction manifests itself in the possibility of using the evidence of stone martens as a bioindicator of Neolithic economy. Southwest Asia and Anatolia are considered the ancestral areas for stone martens, which were already present in this area during the Late Pleistocene. It was from here that the species slowly spread into Europe (Vekua 1994). Zooarchaeological and paleontological records indicate that, while the pine marten is a genuine element of European fauna, the stone marten is a recent colonizer (a post-6 ky BC), whose invasion was fostered by the spread of Neolithic economies. The common marten might have benefited from landscape transformations fostered by Neolithic agropastoral communities (Llorente-Rodríguez et al. 2015). As a result, the species is a faunal bioindicator of the existence of a Neolithic economy in southern Europe (the Iberian Peninsula) (Llorente, Montero and Morales 2011). To our knowledge, the radiocarbon date obtained for the Çatalhöyük stone marten (7340 ± 40 BP) is the first for Neolithic Turkey, indicating the presence of this species in Anatolia at that time—much earlier than the earliest known presence in Europe (6255 ± 35 BP). This date can serve as a starting point for research on the timespan of the stone marten in the archeological record of Turkey.

Fur-bearing value of TP small carnivores

The observed skeletal bias—with only the head parts and foot bones being present—is usually associated with the preparation of game skins. Likewise, in this case, it results from the use of martens for their pelts. In such cases, the foot bones are often attached to the pelts—as, for example, with the red fox evidence from Göbekli Tepe (layer III; Peters and Schmidt 2004). The patterns, however, differ among various fur-bearing animals. At a Mesolithic site in Denmark (Tybrind Vig), for example, it seems that pine martens and otters were skinned differently—the pine martens are represented by complete skeletons, but there is evidence of the head and foot bones being removed from the otter (Trolle-Lassen 1987). Thus, the fur of the feet and snout might have been left on the pine marten carcasses, while the head and foot bones were left with the pelt of the otter (Fairnell and Barrett 2007).

The effect of taphonomic processes on the representation of the elements observed here—in the form of the selective preservation of elements—can be excluded given the low rate of weathering of their bones, which were largely deposited intact. The evidence of the articulations may suggest that there were no significant alterations that could have affected the preservation of the stone marten feet. A well-preserved set of stone martens paws suggests that abiotic factors—such as a soil chemistry or the weight of the overlying sediment—were not destructive, allowing the elements to survive in the assemblage.

The processing of the TP stone martens for their pelts is also corroborated by the evidence of cut marks. Val and Mallye (2011) point out that different cut mark patterns on the anatomical elements of small carnivores are produced depending on the elements that remain in the fur. If all the elements are removed, it is likely that cut marks will be seen on potentially all parts of the distal limbs (phalanges, metapods, carpals, and tarsals). However, when the phalanges remain in the fur, they will not show cut marks on these elements, though the rest of the distal limbs will. In the case of the stone martens from the TP burial chamber, it seems that the foot bones were left in the fur, given the presence and nature of the cut marks on the metacarpals. This is especially true for the paws in A1 and A3 (Fig. 10). This usually occurred when the pelt was used to produce a “trophy” or skin (Fairnell and Barrett 2007). The same breakage pattern is observed in the example of the two metapodials (NA in Fig. 10), which could have come from the same paw. Unfortunately, the lack of proximal ends on all of them precludes accurate identification. Given that stone martens have little or no fur on their feet, especially on the soles (Carter 2004), removing it would be extremely difficult and demand great skill. Thus, the cut marks on some phalanges (A4 and lose finds; n = 2; Fig. 10) seem to have been made incidentally, rather than proving that they were removed from the fur. The fur-bearing value of this species is similar to that of the red fox. Examples of the hunting of martens (mainly pine martens) for fur are also known from Mesolithic sites, which is also associated with the lower value of their meat (Degerbøl 1933; Grundbacher 1992; Stubbe 1993a, b; Richter 2005; Aaris-Sørensen 2009).

Open image in new window

Although we noted only a few fox elements in the context examined here, foxes are represented at the Neolithic Çatalhöyük by all body parts, with some bias toward heads, which might indicate that skins were brought to the site in addition to whole carcasses. Foxes might have been killed either for their skins or to protect lambs, suggesting that herds were kept closer to the site (Russell et al. 2013). In addition, recent research on a Late Neolithic assemblage at Çatalhöyük (TP Area) revealed that foxes were not numerous (0.3% NISP; 0.5% DZ), and in one case they appeared in an abandonment deposit, where a fox canine was part of the cluster of items spread throughout the interior of the building (Pawłowska in press; Barański et al. 2015). At sites where foxes are observed as the dominant carnivore, or in particular contexts, it is usually the case that they were hunted for their valuable fur, for their ritual significance, or because of their threat to livestock and game (e.g., Russell et al. 2009b; Maher et al. 2011).

The fur-bearing value of the common weasel is not great due to its very small size. However, Fairnell (2003) noted a factor in favor of the value of weasel fur—namely, its distinctive white color in the winter, which is also when it is at its thickest.

Symbolic significance

To understand the small carnivores in the burial chamber, multiple interpretations must be considered, in terms of both natural and cultural factors. The main focus here is on stone martens, which are predominant among the small carnivores and display distinctive patterns, as discussed above.

If the stone martens accidentally fell into the burial chamber and were trapped there, complete skeletons would be expected. Instead, highly selective patterns in the head and feet were observed. It is also hard to imagine this would have happened to two individuals, as shown in our results. Moreover, stone martens are excellent climbers and can jump—their jumping stride can be up to 1 m (Grzimek 1990; Sidorovich 2009)—which ultimately invalidates this hypothesis.

Another potential factor is rodents. There are some signs of burrowing in this context, which might suggest that the presence of stone martens was an effect of rodent activity. Rodents can displace artifacts and ecofacts—as is well recognized in archeological sites, including Çatalhöyük—and thereby have an effect on archeological deposits (Kelly and Thomas 2012). Although rodents can also be responsible for taphonomic bias by removing some skeletal parts from the specific context or even from the site (Wood and Johnson 1978; Bocek 1986), this is unlikely in our case. The selective nature of the stone marten skeletons is more likely result of human factors, as discussed above. For the same reason, this deposit could not represent a den containing dead stone martens. Moreover, this would be inconsistent with the behavior of stone martens, who do not dig burrows or use those made by other animals, such as rodents (Heptner and Naumov 2002).

In addition, the lack of digestion marks in the form of corrosion or pitting excludes the possibility that they came from scats. Since all these natural factors seem unreasonable in assemblage formations, human activity is most likely the cause.

Certainly, none of these features, such as could be the evidence of plaster, indicate that the stone martens’ feet are remnants of a dismantled installation, as is known to be the case with large carnivores based on the example of a bear’s paw at the site (Russell and Martin 2005).

Since burial fill at Çatalhöyük generally contains everyday refuse items (Bennison–Chapman 2013), it would seem unlikely that pelts—which were carefully processed and required effort and skill to produce—would be regarded and rejected as worthless, significantly weakening the other hypothesis that stone marten paws could be a part of the material used to backfill the chamber. Much of this material clearly differs in both composition (ruminants predominating) and characteristics (no articulation or butchery marks, low completeness of the bone) from the stone marten assemblages, indicating the use of general site refuse as fill material. This is also an argument in favor of their intentional placement here.

The interpretation of the stone marten bones as grave goods seems the most plausible. Our results show, as indicated by the coordinates, that in one case the bones could be related to the infant individual. This is even more likely, considering that the infant skeleton is almost complete (Czerniak and Marciniak 2008; Hager and Boz 2008), indicating a relatively undisturbed burial, clearly standing out among the other human remains. Unfortunately, no data exist from the excavation stage showing a detailed relationship between the stone marten remains and the infant skeleton; the evidence of stone marten paws was exclusively revealed as a result of our analysis. Therefore, it is difficult to say whether the articulated stone marten paws were intentionally placed in relation to this individual—to mark identity, for example—or whether they appeared coincidentally. It is certainly telling that the marten feet occurred in the southeastern part of the chamber, in exactly the same place as the infant—despite the fact that human remains, mostly scattered, were present in the northern part of the structure. It does not help that the rare artifacts—which were part of the infilling material of the chamber and deemed to be grave goods—could not be attributed to specific individuals (Hager and Boz 2008).

Accounting for various scenarios, our study of the stone martens leads us to the conclusion that they were deliberately incorporated into the burial chamber, possibly in connection with infant interment. This could then be interpreted as potential personal items or gifts from the mourners, or a combination of the two, as has been suggested for the burial goods (Hamilton 2005; Nakamura and Meskell 2013a). Animal parts as grave goods are not very common in Çatalhöyük burials, with only a few examples: two babies buried with bird parts and a woman buried with three boar mandibles (Russell et al. 2009a). Russell (2012) provided an extensive overview of animal parts in human burials across sites and time periods, along with the social factors shaping animal bone assemblages. She suggested they might function as food offerings for the afterlife, result from funerary feasts or sacrifices, or have symbolic value. The former two are limited to domestic animals, in which case mostly meaty parts should be present (Russell 2012)—which is inconsistent with our results. Instead, nonmeaty parts were present in our study—that is, those parts usually related to symbolism in burial contexts. We suggest, therefore, that stone marten paws were included for their symbolic value as ritual paraphernalia, most likely in connection with the significance of a particular place—specifically, the multiple burials in a decorative burial chamber. This has not previously been observed at the site, and so this association adds to our knowledge of social practices in late Neolithic Çatalhöyük. In addition, there is no evidence for the utilitarian use of small carnivore skins at Çatalhöyük, even though some animal skins were used (leopard, hare, caprine) (Russell 2012 with further references), which also allows us to assume they played a symbolic role in the burial chamber. This is consistent with other evidence of burials where selective animal body parts are present. A ritual connection is postulated in such cases, regardless of the time periods they represent (e.g., Horwitz and Goring-Morris 2004; Grosman, Munro and Belfer-Cohen 2008).

The link between human and animal remains is attested in two subsequent TP burial contexts, indicative of new burial rites (Hodder 2014). In one, a human skull is abutted by a cattle frontlet; in the other, human postcranial parts coincide with goat horn cores (Pawłowska in press, 2014). The architectural dimension of the changes in burial practice is manifested in the chamber construction. This was a departure from burials usually being placed underneath the floors or platforms of houses (Hodder 2014). In this regard, these TP burials are unlike any Çatalhöyük burial.

Two other small carnivore species differ from stone martens in the patterns of body parts and butchery marks. The presence of isolated fox elements in the burial chamber does not point to any symbolic meaning, which aligns with our current general knowledge of the site. Specifically, there is no indication of the symbolic relevance of foxes at Çatalhöyük, despite the generally rich animal symbolism; this contrasts with their frequent depiction in the Levant and southeast Anatolia (Russell et al. 2013 with further references). An example of fox remains in a burial chamber was found in the Van-Yoncatepe necropolis in eastern Anatolia, where five adult foxes, along with human skeletal remains, were found in chamber M4; this, however, is much more recent, dating from the beginning of the first millennium BCE (Onar, Belli and Owen 2005).

The weasel mandibles in the burial chamber are not unusual in terms of the body parts present at Çatalhöyük—namely, cranial material dominates, despite the fact that all body parts are present at the site (Russell et al. 2013). Although our results are not sufficient to suggest any symbolic significance for weasels in the examined context, this aspect is well known for the site. It is manifested in the deliberately placed scats of small carnivores (possibly weasels) in the burials by human inhabitants as part of a ritualistic practice (Jenkins 2012). In addition, a complete weasel skeleton was evidently placed in a burial with carnivore scat (Jenkins 2005; Russell and Meece 2005; Jenkins and Yeomans 2013). Furthermore, Mellaart (1964) found a weasel skull plastered and protruding from the wall, suggesting a ritualized and symbolic association within the community (Nakamura and Meskell 2013a). In that case, however, a lack of access to the specimen complicated identification. Based on its size and proportions, Russell and Meece (2005) regarded it as a mustelid and a member of the weasel family, most likely a badger.

In summary, our results strongly suggest that stone martens had some symbolic value in the burial chamber of the TP Area at Çatalhöyük East as part of the social practices of the Late Neolithic. The articulated feet of the stone marten, resulting from skinning activities, seem to have been intentionally incorporated in the infilled deposits of the structure. If that is the case, they were likely a part of the rituals involved in the mortuary practice.

Factors affecting the presence of small carnivores at Çatalhöyük

The use of martens in such a context raises questions about their procurement, which can be considered in relation to their environmental habitats and behaviors. Martens as a species are strongly associated with rocky and stone habitats, which are not present around the Çatalhöyük settlement since the site is located on the alluvial sediments of a paleolake. However, martens are also among the most synanthropic of carnivorous mammals. Therefore, their distribution has been strongly correlated with human dispersion and occupation, from the Holocene to the present. It seems that the occurrence of stone martens at Çatalhöyük—like the appearance of two other small mustelids, the polecat and the weasel—can be explained by the fact that might have actually lived on the site or entered it at night to hunt (Russell et al. 2013). The use of traps has been suggested as a way of hunting martens (Richter 2005). However, given the absence of complete skulls (only maxilla) in our assemblage, which might be expected to contain lesions, this scenario could not be confirmed. One factor that might explain the presence of stone martens in human habitations is the occurrence in such places of their potential prey—especially rodents (Stubbe 1993a, b). Some of the recent data (concentrations of microfauna from scat assemblages, which appear to have been in situ accumulations) suggests that there were enough mice in the vicinity of Çatalhöyük to sustain the carnivores (Jenkins and Yeomans 2013). Mice, which comprise 96.2% of the total species among the microfauna found at the site (Jenkins 2005), were attracted to the site and took full advantage of the scavenging opportunities provided by the storage food and refuse, as shown by Jenkins and Yeomans (2013), themselves becoming suitable prey for small carnivores. Mice and other rodents might have lived in the narrow spaces between buildings at Çatalhöyük. Although the stone marten prefers animal prey, it is an omnivorous species whose diet also includes honey (its immunity to bee and wasp stings allowing it to obtain this without injury), and even carrion when food is scarce (Carter 2004; Grzimek 1990; Zhou et al. 2011).

The threat weasels pose to livestock is generally exaggerated, but all “weasel-like” animals were (and are) strongly persecuted for this reason (Reichstein 1993a, b; Yalden 1999). The main reason for the presence of weasels in human settlements is rodents; the distribution and abundance of weasels is clearly related to the local population of rodents, since weasel populations are capable of enormous sudden increases, usually associated with “mouse years” (King 1980; King and Powell 2007). Since the weasel is a highly specialized rodent killer (Heidt 1972), it is possible that weasels were attracted to the site by the abundance of house mice (Jenkins 2005). It is also possible that weasels were tolerated at Çatalhöyük because they helped control the mouse population (Hodder 2013c), which is widely known to be the main benefit of this species (e.g., Erlinge 1975; Sullivan and Sullivan 1980). Recently, some evidence for considering small mustelids—such as weasels and polecats—as rodent predators at Çatalhöyük was proposed by Jenkins and Yeomans (Jenkins and Yeomans 2013), who noted the existence of puncture marks on mouse bones.