Anthropomorphic clay figurines comprise an enigmatic category of finds associated with Pitted Ware culture sites during the latter part of the middle Neolithic period (c. 2900– 2300 BC) in the Baltic Sea region. As with most figurative objects, previous research has often been preoccupied with questions of representation, for example focusing on what the figurines might depict. In this paper, the anthropomorphic figurines are instead explored through their physical properties, primarily their ability to look back at their human makers, handlers and onlookers. Considering these figurines as clay beings that have the ability to look back at their viewers shifts the perspective from representation to presentation. This conceptual shift results in a more dynamic picture of human-figurine interactions at Pitted Ware culture sites.

Beginning in around 2900 BC, anthropomorphic clay figurines start to appear at several sites in eastern Sweden and Åland (Finland) associated with the marine hunter-gatherer groups known as the Pitted Ware culture (PWC). The eyes of these figurines are prominently featured in the middle of their upwards-tilted faces, seemingly probing and seeking the human gaze. These artefacts have almost always been viewed as enigmatic and, accordingly, they have been the subject of many different interpretations over the years. The figurines in question are undoubtedly viewed as expressive, charismatic objects, and their ability to stimulate the imaginations of both archaeologists as well as the public is well attested. When new finds are publicized by news media, there is a propensity to describe them as ‘curious’ or ‘mysterious’. As more and more of these anthropomorphic figurines have been unearthed, patterns of similarity and variability have emerged within the assemblage. Despite the growing number of these artefacts, our knowledge of their significance and use remains limited. I argue that this is the result of a research tradition too preoccupied with questions surrounding representation.

This article instead explores the relationships between humans and anthropomorphic figurines at PWC sites. To accomplish this, the anthropomorphic figurines are examined from a perspective that avoids purely representational or symbolic analysis (see e.g. Harris 2018; Eriksen 2022; Lindström 2024:85–86). The figurines have previously been interpreted as gods, humans, seals or human-seal hybrids (Björck et al. 2020:16; Cederhvarf 1912; Stenbäck 2003:86–87; Storå 2001:48–51). In this paper, instead of treating them as representations of beings, they are treated as beings themselves. Following this, I explore the ways in which their physical characteristics facilitated certain types of human-figurine engagements, particularly eye contact or ‘reciprocal gazing’. Reframing them as beings, rather than representations, challenges preconceived notions of their meaning and use, and has implications for our understanding of the human-figurine interactions at the PWC sites.

The study of prehistoric figurines has, for a long time, been moored in symbolism and representationalism. This approach is founded on the idea that successful identification of what an image depicts will provide the key to understanding why it was depicted (see e.g. Back Danielsson et al. 2012; Nanoglou 2009; Wengrow 2003). Prehistoric figurines are thus reduced to a single aspect: what they are believed to depict. They are often viewed as material stand-ins for beings, whether physical or spiritual, that are not actually present, with many discussions treating them as representations of individuals (Bailey 1994; Hamilton 1996) or mother goddesses (e.g. Gimbutas 1974). The research into the PWC anthropomorphs is no exception to this, with interpretations that describe them as depictions of gods, spirits, ancestors, shamans, shamanic spirit helpers, seal-human hybrids, prominent people, cult heroes or dead people in wrappings (see Björck et al. 2020:15–16; Cederhvarf 1912; Nilsson Stutz 2006; Núñez 1986; Storå 2001:48–51; Wyszomirska 1984:119).

The focus on representation and categorization in the study of PWC figurines, and figurine studies in general, has resulted in an interpretative roundabout where a myriad of equally plausible (or implausible) interpretations are stacked on top of each other. This inclination towards representation and categorization can cause us to look past the objects themselves, leaving the figurines absorbed by the categories in which they have been placed (cf. Weismantel & Meskell 2014:234). Although these artefacts might very well depict something, it is not universally the case that representation (and the separation of the depiction from that which is depicted) is the key concern in the creation of figurines. Indeed, there are many figurative traditions in which images are treated differently, as alive and potent, possessing their own wills and wants (e.g. Freedberg 1989:11, 32–33; Gell 1998:129; Graeber 2005; Howey 2020; Jones 2017:171; Mitchell 2005:5–6, 29–31). Images often defy the subject-object divide; this is made evident by the way in which the image may also be given the same name as the spiritual entity that it materializes or for which it acts as a vessel (Alberti 2013; Graeber 2005, 2015:29; MacGaffey 1990, 1994, 1998; Pedersen 2011:158).

If we allow ourselves to momentarily bypass questions of representation, we can simultaneously allow for other aspects to enter the discussion (Eriksen 2022:67; Gosden 2020). There are avenues of investigation that do not depend upon taxonomical determinations, with one such avenue being characterized by a greater focus on the materiality of figurative imagery.

The specific properties of three-dimensional imagery in general, and figurines in particular, are important to note. While my focus here is chiefly centred upon what can be termed the visual properties of figurines, I am mindful of the critique aimed at the so-called ‘ocularcentrism’ in the field of figurine studies. The privileging of vision has, according to some researchers, diminished the importance of other senses in relation to figurines and related three-dimensional depictions (Bailey 2014; Eriksen 2022:68; Papadopoulos et al. 2019). Doug Bailey (2014) has, for example, argued persuasively for the importance of knowing by touching, while Mary Weismantel (2013, 2015) has argued for a kind of embodied ‘slow seeing’ in relation to carved stelae, stressing that visual apprehension can sometimes demand an active and engaged human viewer.

The above approaches bring the human body, its stature, abilities, and movements into consideration and thereby avoid treating carvings and figurines as two-dimensional pictures or presenting humans as disembodied eyes. By focusing on their materiality, we can ask what forms of interaction they facilitated with humans (e.g. Bailey 2014; Fahlander 2021; Freedberg 1989; Mitchell 2005; Morgan 2014:87; Weismantel 2013). What can figurines do in their engagements with humans? What sort of interactions with a maker, handler or onlooker did they demand, incite or invite? With these questions in mind, we shall now turn to the anthropomorphic clay figurines found at the PWC sites and explore how their physical properties encouraged reciprocal gazing between humans and figurines.

In the Neolithic of Northern Europe, figurines are primarily associated with foraging communities, such as the PWC and the neighbouring Comb Ceramic culture. They are virtually absent at Funnel Beaker culture associated sites in this area (Iversen et al. 2024). PWC sites generally appear in clusters, close to the shoreline (Björck 2003; Österholm 1989:168). Osteological finds indicate a subsistence based on sealing and fishing, something which is also supported by stable isotope analysis of human bone collagen (see e.g. Eriksson 2004; Fornander et al. 2008; Lidén & Eriksson 2007). Bones from domesticated animals and small amounts of carbonized grains (Ahlfont et al. 1995; Vanhanen et al. 2019) are present on many PWC sites, but these domesticates did not constitute staple foods.

The PWC anthropomorphic figurines are coil-shaped and consist of a head, an armless torso and a foot piece. Heads are generally round or oval with slightly dish-shaped faces, a centrally placed nose or snout-like protuberance, marked eyes and, in some cases, also mouths and nostrils (Figure 1). The torsos, commonly featuring clearly marked shoulders, taper into foot pieces that often feature pronounced toes and heels. Some of the figurines with unusually large, protruding foot pieces have been interpreted as depicting sitting or kneeling humans (Björck et al. 2020:130–132). The figurines vary considerably in height from a couple of centimetres up to more than 20, although accurate size determinations are problematic due to the fragmentary state of most artefacts. The shapes and postures are similar regardless of size. A variety of decorations consisting of impressed pits and drawn lines in different patterns are present on most of the figurines. Depending on their placement, these decorations have been described as depictions of hair, facial hair, clothing or tattoos. Figurines sporting markings on the chin and sides of the face have been interpreted as depictions of male individuals, although it is also possible to interpret these ‘beard-like’ markings as tattoos. The markings resemble chin tattoos worn by women among circumpolar indigenous groups such as Iñupiaq, Yup’ik, Qilakitsoq and Chukchi (Bogoras 1904:254, pl. XXIX, fig. 2 & 3, fig. 186; 1907:342–343, 360; Krutak 2019). While rounded applications resembling female breasts are found on a handful of the figurines (see e.g. Björck et al. 2020:fig. 79; Cederhvarf 1912:311–312, taf. V 3a), most figurines do not show any clear indication of biological sex. Another feature found on some figurines are ‘head pits’ at the crown of the heads. These pits have been interpreted as holes used for fastening decorative and perishable objects such as straw, sticks or feathers (Cederhvarf 1912). Micro computed tomography (µCT) scans of one large figurine from Tråsättra lends support to this theory (see Björck et al. 2020:198–199; Ericson 2019).

Examining the sites at which these anthropomorphic figurines were found, we can determine that they belong to the latter part of the middle Neolithic period (MN B) and that they are especially common on Åland (see for example Lindström 2024:98–104; Núñez 1986). While clay figurines in general, including zoomorphic or indeterminate figurines, are regularly found on PWC sites, anthropomorphic figurines are rarer and most examples have been found at only a handful of sites dated to MN B. Until quite recently, few anthropomorphic figurines had been found at the Swedish PWC sites (Broström 1978; Segerberg 1978). Although many examples have now been found at sites located on the Swedish mainland, particularly from excavations connected with large infrastructure projects since the early 2000s, Åland is still the core area for PWC associated anthropomorphic figurines. Contacts between Åland and several of the Swedish PWC sites with anthropomorphic figurines are also evident in the form of finds of flaked rhyolite or similar igneous rocks (often labelled ‘quartz keratophyre’ in Nordic archaeological publications) typically found in abundance at the Ålandic sites (Björck et al. 2020:157; Darmark 2006; Kihlstedt et al. 2023; Runeson 2023:27).

Parallels have often been drawn between the PWC figurines and the figurines of the so-called Comb Ceramic culture (CCC), particularly the standing anthropomorphic figurines associated with the late CCC phase which is coeval with the younger PWC phase (Núñez 1986; Wyszomirska 1984; cf. Loze 2005, 2008). Given the almost 1000-year hiatus between the so-called bentor foetus shaped anthropomorphic figurines from the Early Comb ceramic period and the standing anthropomorphic figurines from the middle Neolithic, any direct links between them seem improbable (see also Núñez 1986).

Regarding the specific, on-site find contexts of the figurines, it is notable that most have been found alongside pottery sherds, flaked stone and animal bones in the deep cultural layers at the PWC sites (Björck et al. 2020:164; Edenmo et al. 1997; Lindström 2024:113–117). For a long time, archaeologists connected these figurines with ritual activities and ritual structures such as burials (see Wyszomirska 1984). Since their representation in ritual find contexts is modest at best, the jury is still out regarding their function. Their potential role in what can be described as ‘mundane ritual’ will be discussed below.

For this paper, 62 figurines and figurine fragments with facial markings for eyes, mouths and nostrils, and/or tilted heads, have been compiled (table 1). The figurines in question have been found at the Swedish sites Fräkenrönningen, Masmo, Norvik, Torslunda and Tråsättra, and the Ålandic sites Geta, Jettböle II and Svinvallen (Figure 2). Most of the figurines and figurine fragments are included here with their respective inventory numbers, the exception being the finds from Geta. A full report on these finds is expected during 2026 (Fast & Soisalo 2023).

The identification of eyes, mouths, nostrils and tilted heads can often be problematic due to the fragmented state of the figurines. The porous clayor ceramic ware can make it difficult to separate eyes from naturally occurring holes, particularly when only one of the possible ‘eyes’ is preserved. Some faces are so dotted with holes that it is virtually impossible to accurately identify facial features. However, without much doubt, the eye markings are the most consistently marked facial feature among the anthropomorphic figurines found at the PWC sites. Eye markings (46) are more than twice as common as markings for mouths (18), and almost fifteen times more common than nostrils (3). Only three figurines are completely eyeless. The eyes are often marked with two parallel, small impressions surrounded by stamped circle. Some anthropomorphs deviate from this and exhibit what will be referred to here as ‘atypical’ eyes, either due to being diminutively marked or marked in an unusual manner. These atypical eyes include those marked with miniscule, incised dots (Figure 3a), and eyes seemingly marked by negative spaces surrounded by decoration (Figure 3b). These atypical eye markings, as well as the lack of eye markings altogether (Figure 3c), will receive special attention later. Mouths are commonly horizontal slits below the nose, but there are exceptions such as the large pit impression on the figurine from Masmo (108675_HST), as well as the row of small pits on one of the figurines from Geta. Nostrils are few and far between and are always made in the form of two pits in, or just below, the protruding noses.

In total, 45 of the 62 figurines and figurine fragments were identified as having upwards-tilted heads, although due to the fragmented nature of the figurines, a definitive determination is often problematic. None of the figurines are obviously forward-facing. As previously mentioned, the tilted heads can indicate that the figurines were oriented towards towering onlookers, for example humans. This interpretation has been suggested, albeit not explored further, in a recent paper (Kashina 2023). In line with the prevalent view of these artefacts as symbolic and/or religious, it is hardly surprising that earlier explanations have tended to gravitate towards interpretations that describe the figurines as depictions of human shamans in a submissive, adoring position in relation to ‘higher powers’ residing in the sky (Artursson et al. 2023; Björck et al. 2020:16).

What is significant regarding the PWC associated anthropomorphic figurines is that the facial markings are quite unevenly distributed. It does not seem that a full set of facial markings was essential for all figurines. The complete lack of eyes among the zoomorphic figurines found at PWC sites is also an indication that realistic depiction was not the point in figurine making (Lindström 2024:94–97). In comparison to bodies of flesh and blood, the PWC anthropomorphs are clearly incomplete. Interestingly they are also incomplete in different ways, with some lacking mouths and nostrils, while others lack eyes. The eyes, as well as the head tilt as a director of the gaze, are singled out in this paper. It should however be mentioned that the other markings could conceivably be approached in a similar fashion. Indeed, the facial features of figurines are often a factor in human-figurine relations among circumpolar foragers. Regarding the household spirit figurines of the Siberian Khanty, Peter Jordan has, in passing, described them as ‘having eyes and ears to see and hear all’ (2003:fig. 6.13). A marked mouth can also be of great importance, since it enables a figurine to receive food offerings. Among the Siberian Chukchi, such food offerings are given in the form of tallow or bone marrow that is smeared on the mouths of figurines (Bogoras 1907:350, 364). The conclusion that stems from such insights into how figurines have been used in the ethnographic present is both simple and profound: the physical features of figurines can endow them with certain sensory capacities that they otherwise would not possess. In turn, these sensory capacities influence human-figurine interactions.

The prevalence of eyes among the anthropomorphic figurines, as well as their tilted faces, suggest that they could reciprocate the gaze of the onlooker. But what are we to make of the, albeit few, figurines that are ‘eyeless’? Most humans have eyes, and indeed most anthropomorphic figurines from other regions and periods are given eyes. The reason why some eyes are entirely omitted or, alternatively, marked in a manner distinguishing them from other figurines, could of course be attributed to the ‘artistic license’ of the figurine makers. Possibly, eyes could have been painted onto the figurines. The PWC associated anthropomorphs might, from a representational perspective, be viewed as being accurate depictions, with the eyes missing because they are depicting dead humans with their eyes closed (cf. Jonuks 2021). In a similar vein, the lack of limbs could be explained by describing the figurines as depictions of dead humans wrapped for burial (Nilsson Stutz 2006). Ethnographic accounts suggest that the mere presence of eyes might be what differentiates images of humans from images of powerful spiritual beings (Minkin 2020). Another aspect to consider is that the appearance of eyes and their capabilities are often not considered to be identical in humans and spirits. In several Inuit traditions, spirits are said to possess a different way of seeing and can sometimes be recognized by their unusual looking eyes. The eyes might be oriented differently, for example vertically rather than horizontally, and the eyesight of spirits is generally considered to be stronger than that of humans or animals (Laugrand & Oosten 2016:254, 259).

Another explanation, following Doug Bailey (2007), would be that the figurines are deliberately incomplete, and that the incompleteness has psychological effects that helped Neolithic people negotiate personal and communal identities. When we, as humans, are confronted with such incomplete figurines, we focus our attention on the glaring omissions, and our brains fill in the gaps. The absences thus offer room for contemplation in a way that the presences do not and force us to make inferences regarding that which is missing (Bailey 2007:121). It should be noted that while the Balkan Neolithic Hamangia figurines discussed by Bailey always lack facial features, most PWC anthropomorphs exhibit at least some of them.

Along with Bailey, I believe that the absence of certain features in the PWC associated figurines is neither representational nor accidental. Instead of this absence being related to aspects of personal identity, it could be seen as a way of limiting the agency of the figurine-beings. Perhaps the eyes could make some figurine-beings dangerous? Among indigenous peoples of Siberia, figurines used as charms can be purposely left unfinished. This is because the more detailed a figurine is, the higher the risk is that it might catch the attention of evil spirits, or that the spirit that it embodies may take control and emancipate itself (Sázelová 2014:128). Omitting certain features such as extremities, eyes, nostrils and mouths, or marking them differently, could be understood as a means of reducing the capabilities of certain potentially erratic and harmful beings (cf. Fahlander 2021) or clearly marking them out as such.

Although the PWC associated anthropomorphic figurines are quite small, as the diminutive suffix suggests, the plentiful decorations and the possible feather applications suggest that they were nonetheless visually prominent. For humans, decorated and marked objects in general are easier to distinguish than undecorated ones (Mühlenbeck et al. 2017). Alfred Gell (1998:74–76) has argued that even non-figurative decoration is inherently functional, making objects ‘come alive’, and that it is attention drawing; decoration has an allure that captures the viewer (Gell 1998:82).

Returning to the gaze and attempting to further explore the proposition that the PWC anthropomorphs were designed to look back at their beholders, we might ask ourselves what this exchange accomplished. What happens when one gazes into the eyes of such non-human, albeit humanlike, beings? Eye markings are the most prominent feature in prehistoric anthropomorphic imagery (Watson 2011). Putting eyes on something is generally all that it takes needed to make inanimate objects ‘come alive’ (Watson 2011) and eye markings can thus function as basic ‘building blocks’ of animation. Faces, whether real, artificial or incidental, produce powerful responses in our brains, and we identify emotions such as joy and sadness in the ‘faces’ we see in clouds, the moon or electrical outlets (see Alais et al. 2021; Guthrie 1993; Hadjikhani et al. 2009; Hodgson & Galvenston 2006). Further underscoring the importance of eyes, they are also the most important features in human facial recognition (Keil 2009).

In addition to the mere presence of eyes, how they look and how they manage to be the most salient part of a face, the direction of the gaze also bears importance. Humans seem to prefer a direct gaze from a young age (Senju & Johnson 2009). Furthermore, a direct gaze signals interest in the observer and can function as a ‘preliminary to interindividual interactions’ (George & Conty 2008). In other words, a direct gaze precedes, and sets the stage for, sociality between the two who lock eyes. Crucially for this paper, the positive connotations and social aspects of a direct gaze are valid also when it comes to figurative images. In an experiment with different portraits of Jesus Christ, the ones in which the gaze is directed straight-on towards the viewer were repeatedly associated with more positive values than the ones in which the gaze is averted (Folgerø et al. 2016). Although tested on two-dimensional portraits, the results may, to a certain extent, be transferable also to three-dimensional objects such as figurines, with the caveat that the three-dimensional objects can be manipulated and the direction of their gaze somewhat controlled.

What happens when one looks into the eyes of the supposedly inanimate, only for it to reciprocate the gaze? For a believer, seeing a sculpture or similar image of a spiritual entity, such as a deity, and feeling it reciprocating their gaze, can be a profound and enchanting experience. Certain images have active, probing eyes and function as focal objects or interactive visual devices, that meet the eye of the viewer as a ‘corresponding other’ (Morgan 2014, 2018; see also Mitchell 2004:352). Gell (1998:118–120) describes this form of reciprocal, relational gazing as ‘imagistic devotion’ and sees it as opposed to devotional practices such as prayer or scripture reading. Through imagistic devotion, the union between the devotee and the divinity comes about through the profoundly important eye contact, which trumps other representational details that might mark the identity of the specific divinity. The reciprocation of the gaze is for Gell (1998:120) an ‘ocular exchange’ which creates intersubjectivity and animation through the realization that one is being literally looked at by a divine being. Through this process of ocular exchange, a sort of mutual apprehension is achieved whereby we can begin to see ourselves through the eyes of the other (Weismantel 2013:30).

The PWC anthropomorphs, apparently by design, drew the gaze of humans and promoted an ocular exchange between people and figurines. This has implications for our understanding of the use of anthropomorphic figurines at the younger PWC sites. How the figurines might have functioned will be discussed momentarily.

The PWC anthropomorphs appear suddenly, in a distinct and recurring form suggesting adherence to some sort of established convention. They are ubiquitous at a handful of sites but are rare or non-existent at most others. Looking at, and being looked back at by, anthropomorphic figurines was not practiced across the entire PWC area. For the subset of the PWC that made and used these figurines, they might have had a unifying function that strengthened the local communities. Were they materializing spiritual entities in order to facilitate social interaction with them?

Their use notwithstanding, they likely inspired strong feelings in the people who shared their gaze. The age-old question of exactly how the figurines were used remains unanswered and might even be impossible to answer. At Tråsättra, the site with the most figurine fragments to date, figurines were found virtually everywhere, leading the excavators to conclude that they must have been a concern for everybody at the site (Björck et al. 2020:164). Of course, there is not a straightforward correlation between the find contexts of the figurines and the contexts in which they were used: they could have been used for rituals, while broken or otherwise unusable figurines could have been unceremoniously discarded after completion (e.g. Larsson & Lindberg 2008:310).

Although this paper has been more concerned of the ‘hows’ rather than the ‘whys’ of human-figurine engagements, it is still important to discuss the function (for lack of a more suitable word) of figurines. Ethnographically documented figurine use suggests that figurines are multi-purpose, not easily classified as either ritual or quotidian by modern parameters. Figurines are used in both play and ritual, and those used as playthings and those used as ritual objects may be visually indistinguishable from one another (Laugrand & Oosten 2008; Ucko 1962:47). There are numerous anthropological examples of children making and using clay figurines resembling humans or animals as toys (e.g. Fewkes 1923, 1938: 30, 34, 55). The same can be said for the use of figurines to summon and command spirits, for example among Yoruba sorcerers (Morton-Williams 1960:36, fig. 2; Witte 1987) and circumpolar shamans, who use them as spirit-helpers (Fitzhugh & Driscoll Engelstad 2017:375).

Archaeologists have often viewed figurines as religious objects, and as material representations of spirits or divine entities. This is also true for the anthropomorphic figurines found at the PWC sites, which have been described as ‘idols’ (Cederhvarf 1912). While obviously being interpretatively loaded, ‘idol’ is also a somewhat derogatory term. Although archaeologists today may not intend to convey negative connotations with the use of this word, the term ‘idol’ denotes a graven image of a false god. In monotheistic religions, worship is supposed to be practiced in relation to the one ‘true’ God, while worship directed towards other entities, or ‘false gods’, is considered idolatry (see e.g. MacGaffey 1994:126). Terms such as ‘idol’ and ‘worship’, therefore, seem at best insufficient and at worst outright misleading in relation to the PWC anthropomorphic figurines, and perhaps in relation to many other types of figurines from other geographical areas and time periods. Indeed, the religiosity of small-scale groups is often horizontal rather than vertical, meaning that their surroundings are populated by an array of spirits or godlings who can be interacted with directly, rather than through prayer (Boyer 2001).

What is often seen in the use of figurines and similar objects is not vertically oriented worship, directed towards powerful beings residing in an otherworldly realm, where the figurine functions as a symbol of one’s devotion. Figurines and similar objects are instead pragmatic, functional objects designed to allow the users to influence spirits and access spiritual powers (cf. Witte 1987:130). In general, then, vernacular religious rituals are not so much centred around praying and hoping as they are aimed at influencing and actively bringing about change.

The PWC figurines might have been ritual or religious objects, but we should try to divorce their use from modern, doctrinal religiosity. In smallscale societies, ritual is often integrated into daily life and is part of supposedly ‘mundane’ tasks (e.g. Bradley 2005; Brück 1999; Hofmann 2020). The overall distribution of figurines on the sites, as well as their ability to look back at people, lends support to the interpretation that the figurines were involved in interactions with the PWC people on a fairly regular basis. There is little to suggest that their use was limited to a certain class of people, for example a priesthood, or that they were treated as precious relics. The figurines could very well have been used in public gatherings. Some examples are quite large and would have functioned in communal settings where there was a demand for a certain degree of visibility (see Hofmann 2014), while others are too small to have been effective in that regard.

There are many good ethnographic examples of figurative imagery, providing refreshing perspectives that challenge the conventional views of prehistoric figurines as ‘idols’. Even where figurative images function as powerful spiritual beings, they are not necessarily approached in the sort of god-fearing humility one might expect (e.g. MacGaffey 1990, 1994). They are interacted with as beings who, while perhaps possessing great powers, are not omnipotent residents of some abstract, otherworldly realm. Instead, they inhabit the same world as humans. The term ‘fetish’ has been used for these, but they are more appropriately described as power objects or medicine figurines. Here, I will restrict myself to the example of the minkisi minkondi (sing. Nkisi nkondi) or ‘spike fetishes’ from West Africa (MacGaffey 1988, 1990, 1994), although the anthropomorphic figurines (sigidi) of the Yoruba people, Nigeria, function in a similar fashion (Wolff 2000).

Minkisi minkondi are often made in the form of anthropomorphic wooden sculptures in which spirits are given a local habitation, a process that, in turn, makes spiritual powers available for human use (MacGaffey 1988:190). The minkisi minkondi are used in a form of defensive magic by medicine men to rectify misfortunes or illnesses commonly caused by witches. By driving iron nails or wedges into these wooden sculpturebeings, or by taunting them, the spirit is awakened and sent off in pursuit of witches and other wrongdoers. The case of the minkisi minkondi also exemplify that PWC figurines were not necessarily designed with the purpose of accurately depicting a certain spiritual entity. Although mistakenly associated with a westernized and sensationalized pop-culture idea of the ‘voodoo doll’ (see Armitage 2015:92), minkisi minkondi are neither made to resemble an intended victim nor the spirit that is to take possession of it. The physical characteristics are instead a means of ensuring that they can act in the desired, effective manner. Nkisi nkondi kozo is the name of a nkisi in the form of a sculpture of a two-headed dog, with one head at each end. Nkisi nkondi kozo is not however a depiction of a two-headed dog spirit. This particular physical form is intended to endow the sculpture with a dog’s ability to track witches, both in the world of humans and the world of the dead (hence the two heads looking in opposite directions).

As far as ethnographic examples go, the idea of the static ‘idol’ is decidedly less inspiring than the multifaceted power object. Viewing the PWC anthropomorphic figurines as interactable beings, engaging with their makers and handlers, and divorcing them from traditional ideas of worship can help create a more dynamic understanding of how they might have functioned at the PWC sites. They were probably not only symbols or conduits for beings that were not present but, as figurine-beings, demanded human attention. After all, they seem to be probing for it. Viewing them in this way is to treat them as part of the lives of humans. Approaching the PWC figurines as active in the day-to-day lives of humans unlocks a potentially productive future avenue for further exploration of their role in PWC society.

In this article, I have argued that the anthropomorphic clay figurines found at several younger Pitted Ware culture sites invite a shift from representational to relational interpretations, redefining their role within middle Neolithic contexts. Instead of solely focusing on deciphering what these figurines might depict – be it gods, humans, or human-animal hybrids – they can be understood as beings with the capacity to ”look back”, promoting dynamic engagements with their human counterparts. This re-evaluation places their physical attributes front and centre and highlights their potential for reciprocal gazing, thereby challenging traditional symbolic interpretations.

The PWC associated anthropomorphic figurines are virtually always amply decorated, and often have prominent eyes, upwards-tilted heads and, in some cases, pits that might have been used to fasten feathers or plant material. All these things contributed to their striking visual qualities, and their eyes and upturned gazes made them visual objects that could also look back at their beholders. The variation in facial features, and their associated abilities, hint that not all figurines needed (or were allowed to?) see, smell, eat or speak. Other examples were imbued with these abilities, which allowed for certain interactions with humans.

Sharing gazes with figurative images, particularly those recognized as spiritual beings, can create profound feelings of intimacy and reassurance in a viewer. That is not to say that they were necessarily omnipotent, otherworldly beings. If they functioned as materialized spiritual entities, they were interactable participants in the world of humans. Perhaps they could be asked, coerced or tricked into helping people.

By emphasizing presentation over representation, this approach deepens our understanding of human-figurine interactions. It not only redefines the significance of the figurines within the Pitted Ware culture but also expands our understanding of the lived experiences of Neolithic foraging communities in the Baltic Sea region.