Archaeological interpretations of prehistoric humanoid figurines, made and used by the hunter-gatherers of the Stone Ages, have traditionally relied on Western concepts of hylomorphism and iconology. Consequently, these figurines are depicted as finished and static objects of art, often separated from their archaeological contexts. Analysis of these figurines has been focused on identifying what they represent, rather than considering what they do or how people used them. This paper draws on new animism and Indigenous knowledge, combined with visual ethnographic analysis to create a visual ethnoarchaeology of northwestern Siberian humanoid figurines, here viewed as material spirits, within their animistic contexts. We argue that archaeologists’ interpretative focus on representation should be abandoned in favour of attending to animation and the material and immaterial ecological relations these figurines shared with their prehistoric makers.

Prehistoric humanoid figurines, made and used by hunter-gatherers, hold a special place among the archaeological materials from the Palaeolithic, Mesolithic, and sub-Neolithic. Humanoid figurines are first known from the beginning of the Upper Palaeolithic (Conard 2009; Farbstein 2017; Pettitt 2017) and continue to be made and used throughout the world by hunter-gatherers and hunter-herders up until the present-day (Golovnev & Osherenko 1999:114; Haakanson Jr & Jordan 2011:172–174; Ivanov 1970; Willerslev 2007:129). Contemporary or recent past humanoid figurines, often termed ‘idols’ in the ethnographic literature, and the figurines made and used by prehistoric hunter-gatherers are viewed very differently in the academic literature. The former are viewed in their religious contexts as products of animism, shamanism and totemism, and are interpreted as being household gods, deities, ancestors, or totems (Bolinder 1927:3–4; Castrén 1853:197; Czaplicka 1914:200; Donner 1922:130–134; Jordan 2003; Vallikivi 2011; Willerslev 2007:129), while the latter are viewed as ‘portable art’, representing everything from fertility gods to toys (Farbstein 2017; Iršėnas 2007; Jonuks 2021; Pettitt 2017; Rice 1981; Ucko 1962; see also Lindström 2024:27 and references therein). Additionally, in popular scientific contexts prehistoric figurines, especially those deriving from Stone Age Europe, are frequently presented, together with cave art, as evidence for the origins of art.

In both cases, but especially in archaeological discussions, it is argued that the figurines and ‘idols’ represent something beyond themselves that can be deciphered (see Mitchell 1986:8). Examples of this include Palaeolithic Venus figurines as representations of fertility and womanhood (Berenguer 1973:51) and ‘idols’ as presentations of false gods or the idolisation of false gods (see Vallikivi 2024:132–133). This representationalist perspective is the result of Western notions of hylomorphism and iconology.

Hylomorphism derives from the works of Aristotle (Ingold 2013:20; see also Simondon 2020). It describes the process of making artefacts which, according to Aristotle, begins with an idea of form in the mind of the maker. This idea is then imposed on raw material – a lump of clay becomes a ceramic vessel, a bundle of reeds is plaited into a basket, and so on. The active mind of the maker (culture) leads and the passive raw material (nature) must follow (Ingold 2013:20–24).

The related concept of iconology is often used in combination with hylomorphism. Iconology ‘[…] is the branch of the history of art which concerns itself with the subject matter or meaning of works of art […]’ (Panofsky 1972:3). Archaeologists Ing-Marie Back Danielsson, Fredrik Fahlander and Ylva Sjöstrand (2012) have critiqued archaeologists’ tendency to default to iconology and the iconographic method when analysing visual materials. They argue that:

Iconology and iconography are an extension of hylomorphism, in which the forms and visuals hold a meaning, representation, or symbolism that the maker wishes to communicate to the observer (cf. Panofsky 1972:7–15). Images are viewed as signs representing something beyond their making and as forms which can be deciphered (Mitchell 1986:8–9). An artefact and its visual elements are then reduced to a representation of an idea in the mind of the maker (c.f. Ingold 2013:20–21).

Is it then reasonable to view both the figurines, made and used by huntergatherers in prehistory, and the contemporary or recent past ‘idols’, made and used by hunter-herders, as presentations of something beyond themselves, when such interpretations separate them from their own ontological contexts and force them into an explanation grounded in Western representationalist thought? This is not to say that figurines cannot be representational in their making and use but images, in this case figurines, are multiple; they can be and do more than what they might or might not represent (Back-Danielsson & Jones 2020:2). The questions then follow, what do these ‘idols’ represent, if they represent something? And what do they do within their own ontological context? And how can the answers to these questions regarding the ‘idols’ assist in the reasoning and theoretical discussion of the humanoid figurines made and used by prehistoric hunter-gatherers?

In this paper, we aim to answer the questions stated above by studying northwestern Siberian humanoid figurines (‘idols’), that is material spirits and ancestral images, in their own animistic ontological contexts. This is done in order to understand the material spirits rather than explaining them based on a Western representationalist perspective. This case study is followed by a critical discussion, regarding the representational interpretations commonly employed by Stone Age archaeologists studying the prehistoric figurines made and used by the hunter-gatherers of the Upper Palaeolithic, Mesolithic and sub-Neolithic (Farbstein 2017; Iršėnas 2007; Jonuks 2021; Mcdermott 1996; Meskell 2017; e.g. Nuñéz 1986). It is not our aim to offer a specific interpretation of the prehistoric figurines but rather to discuss the theoretical premises and preconceptions of the representational interpretations and argue that animistic ontological premises are the most likely point of departure when dealing with the figurines made by prehistoric hunter-gatherers.

In 1983 Lewis R. Binford predicted that future archaeologists would make use of the visual ethnographic materials, i.e. drawings, photographs and video (Pink 2021:6), that ethnographers have produced since the beginning of the 1900s. Binford, asked ‘How can we take such pictures and convert them into usable archaeological information?’ (Binford 1988:26).

In the cases where material spirits are present within the visual ethnographic materials, so too are their contexts. It is this aspect of visual ethnographic materials that makes them of significance to archaeologists (Binford 1988:26). The material spirits are not separated from their contexts as they are in the archaeological record (in terms of the taphonomic factors affecting their preservation) or in museum collections. Recent studies by visual ethnographers and anthropologists have drawn on non-representational theoretical approaches (see Ingold 2011b; Stafford 2006) to reconceptualise visual ethnography. These approaches involve a shift away from visual research methods being viewed as recording methods which produce ethnographic data, towards an understanding of the visual as cultural experience, ways of knowing and things to learn from (Pink 2021:6, 31). To analyse the visual, then, becomes a practise of theorising how and why the visual was made, and what can be learned from the ethnographic contexts and the things they portray.

The material analysed in this study is formed of visual ethnographic materials from northwestern Siberia (c. 1600s–present). The study focuses on the tundra Nenets and how they relate to their material spirits (N. Khekhe/Syadei). Visual materials and ethnographic literature of neighbouring peoples, such as forest Nenets, Enets (tundra and forest), Nganasan (western and eastern), Selkup (northern and southern), and Kets is also considered. Drawing on Matthew Desmond’s (2014) concept of relational ethnography, which emphasizes the study of ‘[…] fields rather than places, boundaries rather than bounded groups, processes rather than processed peoples, and cultural conflict rather than group culture’ (Desmond 2014:574), the focus is on these Indigenous hunter-herders and their relations with each other, as well as their contact with the colonising governments and Christian missionaries.

The visual material consists of illustrations and photographs from northwestern Siberia, in combination with a short, 3-minute video (Sharing a meal with the ancestors in Gyda tundra, unpublished), containing footage of a Nenets herder feeding his family Khekhe (ancestral images) within his chum (Nenets conical tent, N. Mya).

The analogies drawn between the ethnographic recent past/present and the archaeological distant past, that is the prehistoric hunter-gatherers of the Palaeolithic, Mesolithic, and sub-Neolithic, are not meant to be directly analogical. Instead, they serve as a means to think beyond the confines of Western thought, which in this case rest upon a representational bias (see Günther 2022:29 for a similar analogical reasoning). By drawing on new animist theory, Indigenous knowledge, and the results from the visual ethnographic analysis presented below, analogies are made between the present and the past in order to create premises through which animist perspectives might take part in theoretical discussions relating to the interpretation of prehistoric figurines. Since prehistoric hunter-gatherers made their living based on hunting, fishing, and gathering, it is more likely that they perceived their environments and their non-human co-inhabitants (material and immaterial) in a manner similar to the animists of northwestern Siberia, rather than aligning with Western, modern perceptions of reality. Thus, the ethnographic analogies made in this paper should be regarded as deconstructive and relational analogies:deconstructive in terms of criticising Western representationalism as being universal in both time and space, and relational in terms of building arguments based on environmental and subsistence similarities (Wylie 1985).

The Nenets, Enets, Nganasan, Selkup, and Kets are all Indigenous to northwestern Siberia. The tundra Nenets live in an area that stretches from the Ural Mountains in the west, along the northern Siberian coast to the Gydan tundra east of the gulf of Ob (Nikolaeva 2014:1–2). The lands of the Taymyr peninsula further east are the herding grounds of the tundra Enets and the Nganasan (Eidlitz Kuoljok 1993:22–23). To the south, in the taiga zone along the rivers of Ob, Nadym, Pur, and Taz, and their tributaries, live the forest Nenets and Selkup (Maloney 2016:117; Sammallahti 1974:12). Along the rivers of Yenisei and its tributaries are the lands of the forest Enets and the Kets (Eidlitz Kuoljok 1993:22; Vajda 2016:298).

Reindeer herding (large-scale since the beginning of the nineteenth century), fishing and hunting are the main subsistence of the tundra peoples (Eidlitz Kuoljok 1993:51–52; Golovnev & Osherenko 1999:7; Istomin 2011:237). Before the nineteenth century the tundra peoples were fishers and sea mammal and wild reindeer hunters (Golovnev 1992). In the taiga zone hunting, fishing and gathering are still of great importance, often in combination with small scale reindeer herding (Eidlitz Kuoljok 1993:47– 50; Vajda 2016:298). All of the communities discussed here are nomads, or semi-nomads, and, since the domestication of the wild reindeer, they have used reindeer and sleighs for transportation and migration to varying degrees. The reindeers are also used for food, clothing, tent canvases, craft materials, and ritual offerings (Jordan 2016a).

Indigenous oral history and ethnographic accounts establish the peoples of northwestern Siberian as animists who are spiritually guided by shamans. Their animism(s) centre on the animation, that is the subjectification and life-giving, of the non-human. Spirits and beings of various kind, as shall be seen below, affect hunting, fishing and herding practises and determine the fate of all individuals (Haakanson & Jordan 2011:164; Maloney 2016; Vallikivi 2024:33). It is important to note that the people of northwestern Siberia have, over the last two centuries, practised syncretic mixes between animism(s) and other religions as a result of contact with missionaries (c.f. Eidlitz Kuoljok 1993:76–85; Haakanson Jr & Jordan 2011:164; Maloney 2016:119; Toulouze et al. 2022; Vajda 2016:299–307; Vallikivi 2024). These different syncretisms are mostly mixes between animism(s) and Christianity (Russian Orthodoxy and Evangelicalism), which have been affected by Tsar Russian, Soviet and contemporary Russian politics, as well as secular reasoning, global capitalism and neoliberalism (Forsyth 1992:155; Vallikivi 2011, 2009, 2024). While many consider themselves fully converted to the new religions, or new ideologies, animistic practises are still present in this region (see Toulouze et al. 2022; Vallikivi 2024). It should also be added that syncretic mixes between different kinds of animism(s) exist as a result of cultural encounters through prehistoric and historic times (see Chernetsov & Moszynska 1974; Forsyth 1992). While ontological and cultural similarities exist across this region, differences in animism and animation can still be found (Vajda 2010:126). Therefore, these animistic syncretisms need to be pluralised – animisms and syncretisms – rather than being described as a unified animism or a syncretism of two binary positions.

Social anthropologist Tim Ingold defines animism as ‘A complex network of reciprocal interdependence, based on the give and take of substance, care and vital force […] [that] extends throughout cosmos, linking human, animal and all other forms of life’ (Ingold 2011a:133).

Compared to the first theory of animism proposed by anthropologist Edward B. Tylor in 1869, ‘[…] the doctrine of all men who believe in active spiritual beings […]’ (Tylor 1869:566), Ingold’s definition can be considered a contribution to the so-called ‘Anthropologists’ revisitation [of animism]’ (Harvey 2017:20), or new animism. Whereas Tylor’s focus lies on explaining animism as a cognitive undeveloped, or failed, epistemology (Bird-David 1999), Ingold’s focus, shared by others contributing to the revisitation, such as Bird-David (1999), Descola (2014), Viveiros de Castro (1998), Willerslev (2007) and Pedersen (2001), instead views animisms as relational perceptions of reality, or as ontologies which differ from the Western, structuralist and dualistic, modern thought and perception of reality. From this perspective, what the Western researcher believes and argues, on scientific basis, to be true or not true is of no importance. The importance lies in the acknowledgement of other ways of perceiving reality – in this case the perception of the material spirits and their ontological contexts. Animisms set the ontological premises for perceiving reality and the non-human co-inhabitants – being both of material and immaterial kind. We argue that this is the reason why the animisms of northwestern Siberia have persisted, in spite of the quest of the missionaries and colonising assimilations, since the new religions and ideologies have been built upon animistic ontological premises (c.f. Toulouze et al. 2022; Vagramenko 2017; Vallikivi 2022, 2024). Ingold’s definition of animisms, although generalising and slightly Rousseauian in nature (the interdependence in northwestern Siberian animisms often entails a demand rather than reciprocity), offers a focus on the relational correspondences between the material and immaterial, and between human vital forces and non-human vital forces, which make the animation visible. Such a material-immaterial continuance, as seen below, does not only apply to Western categories of living beings (e.g. humans. animals, plants) but also to other materials and non-organic things.

One of the first images portraying material spirits (N. Syadei) from northwestern Siberia is published in German historian Gotthard Artusius’s publication from 1614 (Figure 1). The image portrays a sacred hill with twelve standing spirits surrounded by reindeer antlers stuck into the ground (Artusius 1614). Around the hill are six tundra Nenets, or Samoyeds as he calls them (an earlier Russian and Western name for Nenets), all worshiping the spirits. In front of them is another person on a sleigh pulled by three reindeers. It is evident that the image is made through a Christian lens, as the material spirits’ appearance is a combination of Christian icons, Greco-Roman statues and Nenets material spirits. Furthermore, the clothing of the Nenets resembles medieval European clothing rather than Nenets traditional clothing and the reindeer are portrayed as deer-like. The act of worshiping takes the form of kneeling in front of the spirits – a practice common in Christian worshiping contexts and not in Nenets rituals (c.f. Islavin 1847:117). It is likely that the artist was not witness to the event, and that the image is based on a description of a tundra Nenets’ ritual.Nonetheless, the image has value as one of the earliest depictions of the tundra Nenets material spirits in a sacred site context.

Another early image of tundra Nenets material spirits is published in arctic explorer Frederick G. Jackson’s travelogue The Great Frozen Land from 1895 (Figure 2). It is a drawing of a sacred site situated on the southwest point of the island Vaygach (Jackson & Montefiore 1895:34). The island is sacred to the Nenets and is referred to in the tundra Nenets language as khekhe ngo – sacred and/or spirit island (Kharyuchi 2004:174). The image portrays a pile of reindeer skulls along with other bone remains. Some of the skulls are attached to wooden sticks while others are lying on the ground. Standing around the pile in a semicircle are ten small wooden material spirits which are facing the sun in the background.

Nenets ethnographer Galina Kharyuchi (2013:73) writes that the piles of skulls and bones are the chum (N. mya) of the master spirit that guards the sacred site and its vicinities, up to five kilometres in diameter. It is forbidden to conduct economic actives within this area (Kharyuchi 2018:132).

Figure 3 shows a contemporary, actively-used sacred site, photographed in 2001 at Yamal peninsula. No material spirits are to be seen in the pile but a photograph, taken by Russian ethnographer Leonid V. Kostikov, in 1927 at Gydan tundra, shows a material spirit, a master spirit, guarding a sacred site, possibly prior to the addition of reindeer offerings (Figure 4). This evidence dates this form of sacred site back at least four hundred years (c.f. Ivanov 1970:73). Russian archaeologist Sergey V. Ivanov (1970:97–98) argues that the material spirits portrayed in the image by Arthus (1614) and in figure 4 date to a pre-Nenets tradition, that is, from before 500 A.D. (Leontieva & Bugaeva 2013:126).

Figure 3 and 4 also show wooden poles. They are called Symzy in tundra Nenets and are sacred – coming from larch trees in the taiga zone which have been transported to the tundra on sacred sleighs (N. khekhe khan, Figure 5) (Kharyuchi 2013:45).

Offerings of reindeer skulls, pelts, ribbons of coloured cloth, rope, and belts, are attached to the poles as gifts to the master spirit of the sacred site (Islavin 1847:118; Kharyuchi 2013:18–25). Depending on the type of sacred site, for example family, clan, or situationally created sacred site (see Kharyuchi 2013:65), the sites can be guarded by more than one master spirit. Originally the sacred island of Vaygach, mentioned above, was the home of hundreds of material spirits before Archimandrite Veniamin destroyed the sacred sites and burned over four hundred spirits in 1827 (Islavin 1847:118–119).

As part of the offering ritual, the master spirit demands to be fed with reindeer blood and fat from a sacrificed reindeer, along with tobacco and vodka (vodka bottles can be seen among the antlers in Figure 3). This forms a ritual meal which is shared with the humans who themselves are eating the sacrificed reindeer and drinking its fresh blood (Islavin 1847:117; Khomich 1966:207).

The master spirits of the sacred sites are most often made of wood, but stone spirits occur too. The master spirits are about half a metre in length, often missing both arms and legs, with pointy heads and a carved face in the form of a mouth and eyes. Some deviate from the norm by having arms and legs, rounded heads, or being more log-like by lacking a head (Ivanov 1970:73–89; Khomich 1966:202–208).

Smaller wooden spirits, such as those portrayed in Frederick G. Jackson’s (1895) travelogue described above (Figure 2), are also brought to the sacred sites to accompany the master spirit. These are sticks (c. 5–15 cm) which have been chosen because of their appearance, for example having a face or a head (Kharyuchi 2013:66). After some rough additional carvings, such as adding details in the form of a neck, a mouth, nose, or eyes, they are brought to the sacred sleighs and transported to the sacred sites. They become the assistants to the master spirit (Kharyuchi 2013:40) and were, in the past, potentially related to deceased relatives of their makers (Ivanov 1970:73).

All things that have spent a long time at a sacred site become animated – even non-spiritual things, such as stones and twigs, which can be taken from the sacred site and kept on the sacred sleighs (Kharyuchi 2018:126). The sites are a source of vitality and cause of animation but can, at the same time, be a source of sickness and death. If a sacred site is abandoned or not visited in a long time, it results in a harmful master spirit that needs to be dealt with or avoided (Kharyuchi 2013:24–25).

According to Ingold’s definition of animism(s), the offering involves the giving of vital forces, including the soul(s) of the sacrificed reindeer and the materials of a living sacred larch tree (the material used to create the poles and the material spirits). What the tundra Nenets get back from the gods and the spirits is aid in hunting, fishing, herding and maintaining good relations with the co-inhabitant spirits (Lar 1998:10–13). This is the demanded reciprocal interdependence and the correspondence between the human and the non-human.

In the animistic ontologies of northwestern Siberia, as in most animistic ontologies, an animated being has more than one soul. The Cartesian division between a material body and an immaterial soul does not apply (Ingold 2022:56–57; Pedersen & Willerslev 2012:467–468), hence the addition of (s) when soul was mentioned above. There are no words in the Samoyedic languages (Nenets, Enets, Nganasan, and Selkup) that correspond to the Western notion of an immaterial soul housed in a material body (Gračeva 1983 in Eidlitz Kuoljok 1993:77). The souls, vitalities, or life-forces, can be material or immaterial continuances of materials, making it difficult to say exactly what counts as a soul and how many souls there are or might be (Gračeva 1983 in Eidlitz Kuoljok 1993:78; c.f. Kim 2000:461–465).

Among the Nenets, Enets, Nganasan, and Selkup there are at least two souls: the breath soul (N. Yindq, E. Beddu, NG. Bachu, S. Qwej) and the shadow soul (N. Sidyangg, E. Ki (?), NG. Sidanka, S. Ella/Ilsat) (Dyekiss 2019:12; Kim 2000:462–465; Vallikivi 2024:164). These two, together with the material body, make up a living human (Vallikivi 2024:164). At first it might seem like a hylomorphic dualism between the immaterial souls (morphe) and the material body (hyle) but among the Enets and the Nganasan the breath soul is part of the heart or the chest and takes the form of strings, similar to sunbeams, that are exhaled and inhaled (Kim 2000:462). This results in the breath soul being part of the air shared with other living beings, including plants. Among the Nenets, when a being with a breath soul dies the soul lingers in the air and is able to hear and see what the living are doing around the dead body (Lar 1998:20–25). An upper world type of shaman, a Tadibya in the case of tundra Nenets, is needed to guide the spirit of the dead to the upper worlds of the sky and to the high god Num and prevent it from going to the underworlds of the god Nga, through which it can return to the middle world and cause harm to the living (Lar 1998:18–20).

Among the Nenets, the shadow soul relates to a person’s hair. Whenever a person’s hair is cut, it is saved and never left behind as it is believed that evil bird-spirits can collect the hair cuttings, resulting in the owner of the hair falling ill and not being able to work (Vallikivi 2024:91). Among the Selkup, the shadow soul is a humanoid version of the owner which can move freely and come and go as it like, potentially being the actual shadow of the material body (see detailed discussion in Kim 2000:463–465).

Theses souls are then an abstraction, lying somewhere in-between the immaterial and material. They do not assume binary positions, as in the hylomorphic and Cartesian models, but exist as relations or correspondences or, more specifically, as immaterial continuances of materials.

This notion of immaterial continuances of materials can be exemplified by drawing on one of the author’s childhood memories from visiting her nomad Nenets family in the Gydan tundra.

At the age of ten, while playing with her cousins, Naglaya stumbled upon a duckling who could not fly due to a wing injury. She decided to take care of the duck and protect it from their dogs and potential predators. A little tent was constructed and the area was surrounded with a fence made from fishing nets. The adults were telling her to release the duck back into the wild as they strongly believed it might be a reincarnation of someone’s dead relative’s soul. As Naglaya was quite determined as a child, she decided to keep the duckling until it could regain the ability to fly again. Over the next few days she would check up on the duckling’s progress and one day it was gone. The adults told her that the duckling had likely recovered and flown away but she knew they had released it as a way of maintaining good relations with the spirits and the ancestors.

The duckling is the immaterial continuance of past materials within the form the material duckling. The deceased human person, most likely its breath soul (air), was transferred, via a process of reincarnation, to the new material, in this case the body of the living duckling. The duckling is material in its bird form yet animated as an immaterial continuance of the deceased human. It is an animal but also a spirit-subject relating to the souls of the deceased human. This spirit-subject is known as Khekhe in the tundra Nenets language (Kharyuchi 2018:123–124) and it is the animation, the life-giving, the subjectification of the non-human.

Estonian ethnographer Laur Vallikivi (2011:88, 2024:164) reports that the tundra Nenets material spirits have both a breath soul and a shadow soul. Furthermore, Kharyuchi (2013:18–25, 40, 2018:121) writes that the master spirits of the sacred sites are related to the gods by being their shadows and they are regarded as Khekhe.

Sacred sites with master spirits, similar to those of the tundra Nenets, are also found in taiga and the on the Taymyr peninsula tundra. Apart from their environmental settings, there are no differences between the creation and use of sacred sites among the tundra and forest Nenets. Similarly, the sacred sites of the tundra Enets and Nganasan do not differ much from those of the tundra Nenets (Kharyuchi, pers. comm.).

In the taiga, among the Selkups, forest Enets, and Kets, the sacred sites are sacred groves and small islands and promontories in rivers and lakes hidden from hunting and migration paths (Maloney 2016; Vajda 2016:302). Their material spirits are also subjects but instead of being housed in piles of skulls and reindeer antlers they are living in sacred huts (Rus. ambarchiks) located within the sites, or they are standing, supported by a tree, in the open-air. The spirits also demand ritual meals to be shared at sacred sites, along with offerings of colourful ribbons hung from sacred trees (birch or cedar). In return, the spirits provide aid in hunting, fishing, gathering, herding, and overall well-being (Donner 1922:131; Prokofyeva 1963:124; Maloney 2016; Vajda 2016).

Finnish ethnographer Kai Donner (1922:128) describes a sacred site close to the river of Ket and most likely among the Selkup, though possibly with the Kets – as early linguistic classifications complicate distinctions between these peoples. While at this site, Donner removed a stone from what he described as its temple. The photograph of this stone-spirit (Figure 6) shows a stone with the shape that looks like a neck which creates a head. A face, eyes and a mouth, have been carved onto the head and pieces of cloth are tied around it, forming its clothes. Donner had to give an offering of 20 kopeck in order to move the stone-spirit. These coins were added to previous coin offerings lying within the temple (Donner 1922:128). Donner gives no description of the temple but it is probable, based on the location, that it was a sacred ambarchik or a wooden chest, likely made of a birch or a cedar tree, which are regarded as sacred among both the Selkup and the Kets (Maloney 2016:121, 130; Vajda 2016:302).

The Selkups and the Kets also carve faces into the trunks of the living sacred trees (S. Kåssyl, K. Holai). These are the master spirits of the sacred sites, and the other material spirits leaning against the trunk are the master spirits’ children (Ozheredov et al. 2015; Vajda 2016:302). A photo of such a tree can be found in Donner’s publication from 1922 (Figure 7). The caption says ‘Skogsguden’ (Donner 1922:134), meaning ‘the forest god’ in Swedish.

The sacred living and growing trees hold a fundamental place in the animisms of northwestern Siberia (Avdeeva et al. 2019; Lar 1998:28–29; Maloney 2016:130; Vajda 2016:302). It is from these sacred trees that the material spirits, who relate to the gods in the form of being an immaterial continuance of the gods and being materialised through the sacred tree, are made. As with all gods and spirits, they are not good or evil, they have their own agency. To live a good life, especially in past times, one had to stay on good terms with the material spirits in order to avoid harm, sickness and death caused by the spirits.

Among all peoples of northwestern Siberia there are also material spirits, or ancestral images, that are kept at the camp sites. Among the tundra Nenets, these smaller humanoid ancestral images (referred to as Khekhe in Nenets, E. Kekho (?) NG. Koika, S. Porge, K. Allel) are created by duck bills dressed in pieces of cloth or fur similar to the traditional Nenets clothing. These ancestral images help with various tasks, such as dealing with bad spirits and providing aid in hunting, fishing and herding, as long as they are respected and cared for properly. They are, like the other material spirits of the sacred sites, neither good nor bad but follow their own will.

Not to be confused with the ancestral images are the metal pendants (humanoid and animaloid) attached to the Nganasan, Selkup, and Kets shamans’ regalia which function as their spirit helpers. These relate solely to the shaman to whom the regalia belongs, as well as to subsequent generations of shamans who inherit them and use them in rituals for specific purposes (Djarvoskin 2003; Dolgikh 1996; Gračeva 1996; Joki 1996; Prokofyeva 1963; Vajda 2010:135). Unlike the ancestral images, these pendants are not used by entire families or communities. The question of how an understanding of these spirits might contribute to discussions of shamanism in prehistory must be addressed separately, given the challenges of identifying prehistoric shamans from archaeological materials (Solfeldt, forthcoming).

Among the tundra Nenets there is also a type of material spirit made solely of cloth – both the body and its clothing (N. Myad pukhutsya). This material spirit aids women’s work with the chum, childbirth and sickness, and it is created when a woman is newly married and moves into a new chum with her husband (Khomich 1966:204). The ancestral images, Khekhe, are inherited on the paternal side of the family.

During Naglaya’s last visit to the Gydan tundra she made a short video (Sharing a meal with the ancestors in Gyda tundra, unpublished) depicting a typical encounter with ancestral images. In this video, a Nenets reindeer herder sits on the floor in his chum and in front of him is a table on which hot tea and bread has been served. In his lap he has three ancestral images lying on a reindeer pelt (Figure 8). His wife brings an empty glass, into which he pours vodka from a newly opened bottle. He then pours a drop of vodka from the glass onto the mouths of the ancestral images and says: ‘the rest is for me’. After that he drinks the rest of the vodka, takes a piece of bread from the table, eats it, and continues to smoke his cigarette.

This video is of a ritual meal, a feeding of the ancestral images, similar to the feeding and the ritual meals at the sacred sites, described above. By caring for the ancestors and answering to their demands, the ancestors in return will care for the herder and his family.

The herder told Naglaya that these ancestral images are his dead relatives and that the reason for giving them vodka was a ‘celebration, special day’ of an unexpected visit from a friend. To understand the concept of dead relatives becoming material spiritual ancestors, which exist somewhere in-between the immaterial and the material, we return to the reasoning of souls among the Nenets.

When an ancestral image is made, the body – wood, stone, or an animal part – is taken from a sacred site. In return the master spirit demands an offering of a reindeer and additional cloth ribbons and coins. Clothing is then sewn and applied to the ancestral image’s body. Over time, as the ancestral image is cared for, additional layers of clothing are added, and the ancestral image gradually grows in size (Vallikivi 2024:141).

Part the hair of the deceased relative, its shadow soul, is then applied to the ancestral image – transferring the shadow soul to the ancestral image. The ancestral images are kept for as long as the memory of the person becoming an ancestral image is preserved.

Finnish pioneer ethnographer and linguist Matthias A. Castrén pointed out, back in 1853, that the Indigenous Altaians of south Siberia do not regard their material spirits as representations. He writes:

We argue, based on the analysis above, that the same applies to the ancestral images of northwestern Siberia. What can be concluded, based on this analysis, is that the material spirits, the master spirits of the sacred sites, and the ancestral images, made and used by hunter-herders in the northwestern Siberian tundra and taiga, are not art. They are not representations of an idea in the mind of the maker, which the maker wants to communicate to an observer of the material spirits. Hylomorphism and iconology do not apply to the animistic ontologies of northwestern Siberia. These material spirits are regarded as living subjects rather than static objects. Their origin relates to the gods, immaterial spirits or ancestors which are materialised via the living and the sacred trees or animals and are shaped as humanoid beings based on the forms of unworked sticks, stones, and logs – a form that affects the human. ‘[…] a khekhe is thus not a mere representation of a deity – it is a deity, or rather, it is a relational being tied to the sacred site of its origin, being a person in person.’ (Vallikivi 2014:142).

To explain these material spirits in terms of hylomorphism and iconology is reductive. To understand them, we must view them as ecological material and immaterial relations which affect the peoples (past and present) of northwestern Siberia, especially in relation to hunting, fishing, herding, gathering and well-being.

In this article we have argued that material spirits, made and used among animist hunter-herders, must be viewed as fluid correspondences between the material and the immaterial. They are part of the on-going ecological relations between humans and non-humans sharing the same environment. In this environment, all things have the potential to be alive – not only humans and animals. Prehistoric humanoid figurines, we suggest, are best understood in relation to the animistic ontologies found in northwestern Siberia.

This research has identified six types of material spirits among the animists of northwestern Siberia: 1) the master spirits of the sacred sites, found among all peoples discussed here, in different sizes and contexts within the sacred sites, mainly aiding in hunter, fishing and herding, in relation to the gods and other spirits; 2) the smaller wooden stick spirits, found at sacred sites among Nenets, which assist the master spirit of the sacred site; 3) the ancestral images, found among all peoples discussed here, which are kept in the camps and are the ancestors which aid in various everyday tasks, in keeping sickness away, and ensuring well-being; 4) the cloth spirit, found among the Nenets, which has particular functions regarding women’s work in the chum; 5) the carved tree spirits found among the Selkups and Kets, the masters of the sacred sites, mainly aiding in hunter, fishing, and herding; 6) the non-worked material spirits, such as stones and twigs and other things that have spent a long time at a sacred site, which are often kept on the sacred sleighs. The sacred poles and the sacred sleighs found among the tundra Nenets should also be considered in relation to this typology of material spirits. They are not humanoid material spirits in the same ways as the categories described above, yet they are sacred, being regarded as khekhe among the tundra Nenets. They are a fundamental part of the sacred sites and are derived from sacred trees. The Nganasan, Selkup, and Kets shamans’ material spirits and various animaloid material spirits (attached to the Nenets’ sacred sleigh, Figure 5), not dealt with in this paper, could potentially also be added to this typology.

Hylomorphism and iconology are often opaquely embedded in the interpretations of figurines made and used by prehistoric hunter-gatherers – creating overarching representational interpretations in which prehistoric figurines are viewed as static objects of art.

Based on this analysis of the material spirits of northwestern Siberia, such representational interpretations result in an unlikely characterization of the ontologies of the Palaeolithic, Mesolithic, and sub-Neolithic huntergatherers. Archaeologists should not solely focus on studying the forms of figurines and what they are supposed to portray. As archaeologists Lynn Meskell (2017) and Rebecca Farbstein (2017), in their overviews of this field of study, have both stated, we can no longer ignore the figurines’ contexts. Here, we argue that this context extends beyond the typical archaeological context of find locations, structures and layers at archaeological sites. Context also includes the material the figurines are made of and the potential immaterial contexts they evoke, including immaterial aspects of archaeological sites and structures. For example, we argue that humanoid figurines found in burial contexts do not necessarily have the same function as those found in camp site contexts. And when it comes to loose finds, where the context is unknown, archaeologists must ask themselves ‘what traces does a sacred site leave?’, especially considering that sacred sites may have been seen as dangerous places, as well as locations in which economic actives were forbidden.

Regarding the dualism of ‘portable art’, to which the prehistoric figurines belong, and ‘parietal art’ (cave art), we argue for a category in-between. This category includes the carved tree spirits in the ethnographic material of the Selkup and the Kets and, from the archaeological record, the Shigir idol from the Urals mountains (Terberger et al. 2021) and the figurines from Gorbunovo peatbog in the trans-Urals (Chairkina 2014). We argue that the division between ‘portable art’ and ‘parietal art’ is not useful when it comes to integrating animist reasoning into the interpretation of the prehistoric images. It is evident that prehistoric figurines could travel (Lucenius & Brorsson 2021), while the ancestral images of northwestern Siberia travel at least as far as a nomadic human, since they are kept and cared for as long as they are remembered – these memories spanning across the generations. Such a generational time-perspective, along with the sacred site perspective, in which the master spirits are placed and stay until the sites are forgotten and abandoned, needs to be considered in relation to the chronology and chorology of prehistoric figurines.

Lastly, regarding the notion of prehistoric figurines as toys (see Iršėnas 2007), traditional Nenets toys are not separated from the animistic ontology, since they are a mentor and teacher to the child, having their own voices, and can become harmful if not cared for properly (Sázelová et al. 2014:57–59). The binary positioning of toy versus material spirits should not be taken for granted. Polish ethnographer Marie A. Czaplicka (1914:200) describes how Russian modern made toy-dolls became material spirits in northwestern Siberia in the beginning of the 1900s. Such a dualistic reasoning partly derives from the notion of hylomorphism, in which nature stands against culture, body versus mind/spirit/soul, and the material versus the immaterial. Instead, we argue, that material spirits in animist huntergatherer and hunter-herder societies are fluid connections between material and immaterial worlds. These spirits are embedded in ecological relations wherein all elements of the environment, not only humans and animals, are potentially alive. We suggest that prehistoric humanoid figurines reflect ontologies similar to those found in northwestern Siberia. Exactly how these relations work and what the figures do is a question for further research.