‘ Sheep are your mother ’ > rhyta and the interspecies politics in the Neolithic of the eastern Adriatic

Recent theories have suggested that artefacts are much more than just objects: they are active subjects in a web of relationships between persons and things (Gell 1998; Latour 2005). I develop the argument that ‘persons’ in the Neolithic of east Adriatic included not only humans, but also animals. Thus relationships between animals and humans must be understood as social relations. Animals also had the power to ‘act back’ and influence human lives, therefore making relations between species political.


Introduction
The four-legged 'cult' vessel or rhyton is a vessel of specific and distinctive shape, found in middle and late Neolithic contexts of the western Balkans.Its curious shape and decoration has puzzled many researches; most of them agree that their distinctive shape had a specific purpose.
Recent theories have suggested that artefacts are much more than just objects: they are active subjects in a web of relationships between persons and things (Gell 1998;Latour 2005).I develop the argument that 'persons' in the Neolithic of east Adriatic included not only humans, but also animals.Thus relationships between animals and humans must be understood as social relations.Animals also had the power to 'act back' and influence human lives, therefore making relations between species political.I explore the role of rhyta in the politics of humananimal relations in the Neolithic of the western Balkans.The main issues I tackle are: what makes rhyta agents, and how they can transform and modify social relations between people and animals?

Traditional studies of rhyta
The origins of rhyta are blurred, and it is not my intention in this paper to make them any clearer (for more recent detailed discussion of the chronology and origins of rhyta see Biagi 2003;Marijanovi≤ 2007).
Rhyta are probably connected with the zoomorphic vessels and altars which are a common feature of the early Neolithic contexts of the southern and central Balkans, together with monochrome or painted pottery, figurines, zoomorphic amulets, clay tripods and stamp seals (Nea Nikomedeia in Greece Macedonia, Rakitovo in Bulgaria and Donja Branjevina in Serbia).This particular form of zoomorphic vessel, the rhyton, seems to appear after 6000 calBC somewhere in the southern or western Balkans.
Possibly the earliest dated fragment comes from Achileion, middle Neolithic phase IIIb, dated to around 6000 calBC (Gimbutas et. al 1989.Fig. 7.69).A few fragments of rhyta were found in eastern Albanian early Neolithic sites such as Vashtëmi (Korkuti 1982.145;Korkuti 1995.Taf. 15) and Barç (Lera 1993.31).Albanian rhyta are often found in contexts with Impressed Ware (suggesting Mediterranean influences) on the one hand, and painted pottery, clay tripods or cult tables, anthropomorphic and zoomorphic figurines, split-leg figurines, clay pins and stamp seals (assemblages common in the central and eastern Balkans; see for example Budja 2003) on the other.Some recent finds of rhyta from early Neolithic Impressed Ware contexts from Dalmatia (Crno Vrilo; Marijanovi≤ 2007) and Italian Apulia (Caverna Elia, Le Macchie; Biagi 2003) are keeping the discussion of their origins alive.
In a few hundred years after 6000 calBC the rhyta became a characteristic element of the Middle Neolithic cultures of the western Balkans, such as Danilo and Kakanj on the eastern Adriatic coast and Bosnia, and Cakran, Dunavec and Topoljan in Albania.After 5600 calBC they can be found at a wide range of sites from the Peleponnese in the south to the Triestine Karst in the north, from Kosovo and central Bosnia in the east and the Italian mainland and the Lipari Islands to the west (Fig. 1).
Their curious shape has provoked many interpretations of their function.Almost every commentator on these vessels agrees that they have some specific purpose.They have been called scoops (Ihde 1995), salt-pots (Chapman 1989) or coal-scuttle vases (Weinberg 1965).However, most authors have granted them cult status, therefore calling them 'cult' rhyta or four-footed vessels.
Most authors put them in the context of fertility cults.Josip Koro∏ec (1958.55-59;1964.73-74)saw them as vessels used in libation-like rituals used in the worship of water, and later see them as stylized representation of the (human) female lower torso, and as female pendants to a male phallus cult.πime Ba-tovi≤ interpreted them as lamps used in the worship of ancestors.In his later works he points to the zoomorphic qualities of rhyta and connects them with pastoralists and animal fertility (Batovi≤ 1958;1979. 560).Alojz Benac (1964.65-66;1973.38;1979.403-405)saw them as symbols of life and fertility of animals and fields.He interprets ring handles as representations of curved animal horns.In the same way, Borivoj ∞ovi≤ (1976.22-24)explained them as instruments used in a cult of fertility of women, animals and land.The oval receptacle, often painted red, represents the uterus, where the legs stand for the teats of the udder.He believes that they are part of 'Near Eastern' mother goddess worship.
Slavi∏a Peri≤ (1996), in his detailed interpretation of rhyta, follows the same line of thought.He sees them as a representation of the womb, udder and teats of different species of animals, especially sheep and goats, pigs and cattle.He uses the shape of the legs (which he interprets as teats) to differentiate the animals which were represented by the rhyta, such as cows (common in Kakanj culture) or sows (in Thessaly).
We might argue that the fact that researches have seen many things in them testify that they are not simple representation, but ambiguous depictions involving powerful imagery composed of elements composed of attributes animal/female/fertility/nurture attributes.But can this abundance of female/animal-fertility imagery condensed in rhyta be under- stood in simplistic terms of fertility rites and mother goddess worship?Their ambiguity suggests that are polyvalent and multi-referential; they mean not one thing, but can condense a whole range of different meanings (see Thomas 2005).
Researchers have sought to create trans-regional cultural connections based on the similarities of rhyta from a wide area.Their wide distribution is often seen as an effect of the seasonal or nomadic migrations of pastoralists (for example Peri≤ 1996; Montagnari Kokelj 2003), an idea deriving from rich ethnographic data from the Balkans (such as Vlachs or Sarakatsani).
However, we might question the assumption that resemblance over a wide area necessarily means that they were used in the same way.Modern ethnographically documented pastoral practices, used to explain the distribution of rhyta, are not remnants from the deep past, or timeless adaptations to the Mediterranean landscape.Modern pastoralism is a complex result of adaptations to different natural and historical rhythms, economic conjectures, political processes and events, and above all, the emergence of capitalism.The projection of modern pastoral practices into the past to explain Neolithic social processes is therefore utter anachronism (see Mleku∫ 2003;2005).

East Adriatic Neolithic communities
Rhyta can be found in a variety of contexts along the eastern Adriatic coast.The Neolithic on the eastern Adriatic coast is not a homogenous and totalizing entity.It has different forms, the results of different historical processes which accompanied and modulated the adoption of novel resources and lifestyles.A useful heuristic device for distinguishing different historical processes along the east Adriatic coast is the division of sites into two groups: camps and villages (Mleku∫ 2005).
Cave sites are located in mountainous hinterland, away from lowlands suitable for cultivation.They are marked by low densities of pottery and animal bone, the majority of which are ovicaprine.Cave sites are usually 'deep', with long occupational histories, often extending back into the Paleolithic.Caves can be interpreted as seasonal hunting or herding camps.Archaeological, geoarchaeological and archaeozoological data suggest that they were used as sheep pens and shelters for small, autarchic and very mobile (nomadic) groups, which relied heavily on large flocks of ovicaprines (Mleku∫ 2005).This archaeological record is in striking contrast to the lowland settlements located near water sources and land suitable for agriculture (Müller 1994).These sites usually yield evidence of architecture, large quantities of pottery, and domesticated plants and animals.They can be identified as villages, practicing a mixed farming subsistence economy.Lowlands are settled by predominately small-scale, dispersed settlement units, often abandoned or relocated (Chapman et al. 1996.335-343).Most Neolithic stock keeping in villages seems to have been small scale, involving localized movements around settlements, predominately in the lowlands.Although some faunal assemblages are dominated by smallstock, most display a mixture of livestock species more reminiscent of small-scale mixed farmers.

Interspecies politics
In contemporary western societies, the functions of raising, slaughtering, and consuming is usually subject to a division of labour, where the consumer of meat does not meet the slaughterer or the raiser of the livestock.But among Neolithic societies every person was a herder, sacrificer, slaughterer and consumer of the animal at the same time: "Neolithic lives and worlds were undoubtedly different from our own.Their relations with animals were closer, part of everyday life, a substantial part of economic/social relations.Relations with animals may have been mediated through relations with other persons (and vice versa), but the connections between one person and another, one person and their domestic animals or prey, may have been extremely important for the identity of that person (e.g. through food taboos, or through shared aspects of identity between people and certain animals).Neolithic personal experience may have been greatly shaped by the animals they bred, exchanged, hunted with, hunted, ate (Fowler 2001. 160).
Our relations with animals are specific to our historical context, and rooted in Western ontological assumptions which distinguish between people (agents, subjects), animals (non-intentional, prey, predators, pets), and objects (non-sentient things) (Nadasdy 2007.26).

Human and non-human persons
Irving Hallowell observed that the Ojibwe, with whom he worked, thought of animals (not to men-tion inanimate objects) as sentient and intelligent persons.For the Ojibwe (and other circumpolar peoples) the concept of the person transcends human beings: "All animate beings of the person class are unified conceptually in Ojibwa thinking because they have a similar structure -an inner vital part that is enduring and an outward form (e.g. human, animal, stone, etc.) which can change.Vital personal attributes such as sentience, volition, memory, speech are not dependent on outward appearance, but upon the inner vital essence of being (Hallowell 1960.21, my emphasis).
Hunting in these societies is a long-term relationship of reciprocal exchange between animals and humans.Hunter-gatherers believe that they can only catch animals when the animal gives itself to them voluntary.Hunting is understood as a rite of regeneration: consumption follows the killing of an animal, just as birth follows intercourse, and both acts are integral to the reproductive cycles of animals and humans.However, animals can be offended.They will not return to a hunter who has treated them badly in the past, and they can be offended if their meat is not properly shared among all those in the community who need it.
Regeneration of lifeworld depends upon a maintenance of balance in the reciprocal give-and-take of vital forces.Animals give life to humans, but humans should receive only what is offered, rather than seek to extract vitality by force (Ingold 2000.123).

Trust and domination
Robert Brightman (1993) argues that there exists a tension between two distinct and mutually contradictory principles governing human-animal relations among the Rock Cree.These are the principles of 'reciprocity' on one hand, and 'domination' on the other.He argues that although Cree hunters do subscribe to the notion that animals surrender themselves to hunters, at other times these same hunters think of themselves as locked in an advisory relationship with animals, who are conceived as powerful beings that must be overcome and dominated if the hunters are to survive (Nadasdy 2007.27).
Tim Ingold (2000.61-76)developed this idea in the mutually exclusive principles of 'trust' and 'domination'.
On the other hand, Paul Nadsady (2007.26-28)argues that there is no theoretical need to make such as distinction; moreover, to do so is to artificially separate aspects which form a coherent whole: they must be understood within a general theory of gift and exchange (for example Mauss 1954;Godelier 1999).
The notion that the principle of domination is somehow opposed to the principle of reciprocity is inconsistent with anthropological understandings of exchange.Altruistic giving is rare: gifts are neither spontaneous, nor freely given.There are many examples of reciprocal exchange systems that embroil their participants in unequal, competitive, and even adversary relations: "There is a tension, but it is a tension inherent in the gift relationship itself, rather than arising from a contradiction between two distinct principles of 'trust' and 'domination'.… animals must be viewed as powerful and dangerous trading partners" (Nadasdy 2007.28).

Pastoralism
When hunters became pastoralists they began to relate to animals, and to one another, in different ways.The incorporation of tame animals in a human household, where animals gain the status of quasi-persons, is the first pre-condition for pastoralism.Tame animals are ubiquitous in hunter-gatherer societies, where they have the role of hunting assistants, transport animals, or decoys (Ingold 1980. 95-112).Pastoral property relations become explicit when the status of animals changes from agents of production to sources of food.It is also a change in the animals' status from quasi-persons to resources.Animals in the pastoral mode of production become means of reproducing the social relations of pastoral production.Reproduction and the multiplication of domestic animals make possible the accumulation of wealth (Ingold 1980.144).The slaughter of domestic animals frees people from the obligations of sharing that apply in the case of hunted animals.Social fragmentation into autonomous, selfsufficient domestic units is therefore not the cause, but the effect of drawing on domestic herds for subsistence.Tim Ingold describes the new relations between humans and animals in pastoralism as 'domination': "It is the herdsman who takes life-or-death decisions concerning what are now 'his' animals, and who controls every other aspect of their welfare, acting as he does as both protector, guardian and executioneer.He sacrifies them; they do not sacrifice themselves to him" (Ingold 2000.72).
But pastoral animals are not only good to eat or think, they are also -to follow Donna Harraway (2003) -good to live with.The social relations between domestic animals and people are different, but also much closer and intimate than relations with wild animals.Animals are not only socialized in human societies, but they also socialize people in their herds.This sociability affects both species, human and non-human, and new social forms emerge, with animal and human peoples' societies acting upon each other and creating new social relations.
Thus in the upland caves of the eastern Adriatic hinterland, humans and sheep shared living spaces, smells and sounds, and herds dictated the movement from cave to cave in search of pasture.However, live animals were probably less involved in the exchange and social relations between isolated and autarkic households.
This sociability might have been different in permanent lowland settlements, where animals were probably physically separated from humans, but more involved in relations between families through gifts or bridewealth.
Pastoralists, when asked about their animals, talk about them in the same way as hunters do.Navajo use the metaphor of mother, generalized reciprocity, when they speak about their sheep: "those called sheep are your mother, sheep are life" (Witherspoon 1973(Witherspoon .1442)).And even if we accept that there are elements of domination in human-animal relations under pastoralism, they are not prevalent.Animals and people are involved in tight social relations, and in order to maintain long term relationships, politics of reciprocity between humans and animals must be maintained.
Pastoralists depend on herds, their growth and accumulation.They live in constant fear of epidemics and natural disasters that might exterminate the herd (see Ingold 1980.163-176).To protect against future catastrophes and to maintain the reproduction and growth of herds, humans must rely on maintaining a balance in the reciprocal give-and-take of life-force (see Witherspoon 1973).Animals (or their spiritual masters) can be offended and they will not return herders who have treated them badly or wasted the life-force.This means that animals have the power to 'act back' and alter human life.Relations between animals and humans are therefore political.The politics between animals and humans is not based on human 'domination', but rather on balanced reciprocal relations derived from an animistic ontology of life-force flow.
The slaughter of an animal incurs a spiritual debt that must be repaid through the observance of a whole series of different ritual attitudes and practices.Thus among pastoralist, animals are never slaughtered, but sacrificed.Sacrifice, like hunting, is connected with many taboos and must be performed in the proper way, otherwise it might disrupt relations between humans and animals (see Evans-Pritchard 1956.196-230;Abbink 2003).
Animals are therefore simultaneously sacrificial victims (as cooked meat, consumed by the humans) and, among the sacrificial recipients, as the 'life-force' or 'inner vital essence of being' of sacrificed animal is returned to animals.This reciprocity often includes 'middle persons', such as the spirits and spiritual owners of animals; however, the principle of exchange of life-force is the same as in hunting.
This might have very deep implications for our understanding of Neolithic herd management strategies, as derived from faunal records and kill-off curves.Neolithic herd management might not have been aimed at 'optimizing' the production of meat or milk.Instead, it might have been based on the principle of the reciprocal flow of the 'life-force' and the idea that more (proper) slaughtering (e.g.sacrifice) produces more animals.
Hunting and pastoralism must be understood as a set of social relations not only among humans, but also between human and animal persons (Nadasdy 2007.29).These relations have profound political dimensions.In the following chapters I am going to develop the argument that rhyta were intertwined in this network of relations between people and animals, held it in place and modified it through their own agency.

Iconicity of rhyta
Main point of departure of this paper is to treat rhyta not as mere representations.They might have some -often ambigous -iconic resemblance to animals, humans, or body parts, but this resemblance is not in the function of representing them, but rather in revealing the hidden properties of animals and humans and their relations.

Depictions of animals
Tim Ingold, in his essay 'Totemism, Animism and the Depiction of Animals' (Ingold 2000.111-131)shows that depictions of animals by hunter-gatherers from Australia and the circumpolar North explain and can be explained by their respective totemic and animistic ontologies.
In the Australian totemic ontology, the life force is concentrated in the land, set down in the era of the 'Dreaming' by powerful ancestral beings who moved across landscape and deposited their life-force in the processes of creation and transformation of places.The land embodies the creative powers of the ancestors, and humans and animals relate primarily to the land.
By contrast, Ingold argues, the 'vital force' in animistic ontology is distributed among beings that inhabit the world.These beings are mutually engaged with one another, creating a complex network of reciprocal interdependence based on the exchange of substances and 'vital force'.The life of one being is predicated on the mortality of another, making the transformation and becoming essential for the maintenance of the flow of life.
The difference is also expressed in ceremonial practices and depictions of animals.In Australia, animals are portrayed as static and associated with the morphology of the landscape, "a world which is already made, not in the making" (Ingold 2000.120).Songs, dances and storytelling re-enact ancestral activity, stressing their co-substantiality with the ancestors.
In the circumpolar north, carved wooden mask reveal the agencies of animals.They are not realistic representations of animals, but use distorted human features to capture the underlying character of the animal (Ingold 2000.124).Masks have the power to invoke the spirits of animals during rituals.

Rhyta as depictions
Rhyta are not aniconic.Rhyta imagery is composed of attributes and features of animal and human bodies.There is obviously a high degree of ambiguity present.Some rhyta are explicitly anthropomorphic.Most have zoomorphic qualities.Some are humananimal hybrids, and some resemble animals from one viewpoint, but can also be like part of an animal, perhaps the udder, from another (Fig. 2).
The most distinctive and recognizable features are the four legs which support the torso and ring handle.Legs can appear in many shapes, from thin and long to short and swollen.Some even have elaborated tips, shaped as animal hoofs or (rarely) as human feet.Wide voluminous legs are often pointed, ambiguously suggesting that they can also represent full teats.
There is a class of rhyta which is obviously anthropomorphic (Fig. 3).They represent part of the human torso in a kneeling or crouching position, with legs widely apart.An enormous rounded belly rests on thighs.The abdomen is opened.There is no upper torso; instead, a large ring is fixed on the back of the rhyta.
When a rhyton is positioned on a level surface in front of us, the legs are hidden from view, and only two features become prominent: the large ring and the empty volume of the interior of the abdomen.
They clearly show what rhyta primarily display: the interior of the body or abdomen.
Alfred Gell (1998.141-142)states that 'idols' or sculptural works intended for cult use rather than mere representations are often merely hollow envelopes, possessing 'significant interiors'.He cites the examples of hollow mauri stones, hollow sorceryimages, or a class of medieval sculptures called vierges ouvrantes, an image of the Virgin which can be opened to reveal a representation of the Trinity in her abdomen.
Also appropriate examples are the masks of circumpolar peoples (Ingold 2000.111-131,see above), with hinged doors that open to reveal a face, effecting a transformation which reveals the true face of the animal.Other masks achieve the same effect by exploiting ambiguous imagery using figure-ground reversal illusions.A mask may look realistic from one angle, but from another can show entirely different features and a different face (Ingold 2000.124).Again, it is the interior of the idol which hides the true face and the mask is merely a surface which keeps it contained until the dramatic event of transformation: "[t]he metamorphosis is not covering up, but an opening up, of the person to the world (Ingold 2000.94)." Artefacts that resemble the body have the particular power to be attributed with agency.They achieve their effectiveness in invoking a projection of the mind through an iconicity of internality/externality. Rhyta can be seen as containers, skins around 'significant interiors'.Their agency is attributed through the invocation of an interior beyond the beyond the surface of the artefact.This correspond to the animistic view of personhood, where vital personal attributes such as sentience, volition, memory and speech are not dependent on outward appearances, but upon the inner vital essence of being.
The most important modes to communicate the property internal to the artefact are orifices and enclosures.Especially agentive are artefacts with eyes, a particularly effective icon of mind.Rhyta do not have eyes.Instead, they have a large ring, a circular opening that might resemble a vagina or wound, especially as the ring is often covered with red pigment (see below).Rings thus communicate an internal property of rhyta which can be connected to the inner vital essence of being.
However, containment is of little value if substances contained inside the container can not pass to the outside and vice versa.This opening might have a temporal dimension, as an act of opening, perhaps through the deliberate breaking of a rhyton.It can also be communicated through the iconicity of a rhyton.Anthropomorphic rhyta which depict women giving birth in a kneeling or crouching position there-fore communicate the process of transformation or flow of life-force contained in a rhyton (Fig. 3).

Skins and envelopes
Rhyta can be therefore understood as 'skins' or enveleopes around 'significant interiors'.Some of them have bare skin, with smooth, polished surfaces.However, large numbers of rhyta are ornamented with incised decoration covering the whole surface (skin) of the vessel (Figs 4,5).Some parts are omitted: inside parts of the legs, the interior of the receptacle, the interior part of the handle.The repertoire of motifs is shared with contemporary pottery, such as hatched triangles, spirals, chess-boards, zigzags, and running spirals etc.However, they are used in a different way than on pots, as they cover the whole surface and not only the perimeter of the pot.The ring receives special attention; it is emphasized with different decoration from the rest of the rhyta (Fig. 4).
According to Gell (1996;1998.73-94),ornamental art is much more than a function of aesthetic pleasure.Objects decorated with skilfully executed geometrical and decorative patterns appear animated in a 'non-representational way', as parts of the pattern relate to neighbouring motifs, thus testifying to the agency of the pattern and object as a whole.This 'dance' of complex patterns, their multiplicity and the difficulty we have in grasping the logic of interplay between them, draws our attention and catches it in the 'unfinished business' of deciphering its logic.
However, this 'unfinished business' of unravelling the geometric pattern causes us to relate to an artefact in a special way.The inexhaustible pattern, always in a state of being deciphered, creates a long-term, biographical relation between the decorated artefact and the person looking at it, in the same way as watching and contemplating animals create bond of domestication and close social relations.Evans Pritchard describes how the bond between the Nuer and their cattle is created and maintained through acts of looking: "The men wake about dawn at camp in the midst of their cattle and sit contentedly watching them till milking is finished.Then they … take them to pasture and spend the day watching them graze … When the cattle return in the evening they tether each beast … and sit in the windscreens to contemplate them and watch them being milked" (Evans-Pritchard 1940.36-37, cited in Ingold 1980.181).
Complex geometric patterns can serve as 'mind-traps' and can be used in a apotropeic role as 'demon-traps' (Gell 1996;1998.86-90).

Colour
The color of rhyta -made of fired clay -ranges from buff through brown to grey or almost black.However, it looks as if most, if not all, rhyta were painted with red pigment.In some cases the whole fragment (or perhaps the whole vessel) was painted.However, in most cases, only parts of the vessels were coloured.These include mainly the ring (or more precisely, the interior part of the ring), the receptacle, and the interior part of the legs (Fig. 5).
Again, colours are more than mere representation or decoration.They are agentive; capable of transforming things.Victor Turner, in his highly influential essay on red, white and black (1967) proposed that, based on a universal human organic experience, red has a universal significance related to blood.The association of red with blood has been much discussed in the literature.There are many examples which do not support Turner's universalist theory (see Young 2006).For example, red may also be as-sociated with ambiguity, magical powers and transformation.
Instead of seeing colour as a static propriety of the vessel, we may try to consider its temporal aspects.Rhyta were covered with pigment after firing, as pigment rubs off easily.Painting the vessel may have been a part of its lifecycle and associated with special limnic events such as use in rituals or end of lifedeliberate breaking.
A colour change might also be thought as transformation in itself, not only symbolically.Colour, or its material form, pigment, has an agency, transforming objects.Anthony Forge wrote of the yam cult, where all magical substances are classed as paint, and paint is a ceremonial medium through which initiates are turned into men (Forge 1962).By the act of applying the dye to the surface and inside of the rhyton, it entered into a process of transformation.Here, the agency of red can create a whole network of analogies which place the rhyton in a web of relations between people and animals: "Killing, hunting, eating, and menstruation, sex, conception and birth are the constituents of a fundamental metaphor likening the provisioning of society to its reproduction.The explicit sex (women) and hunting (animals) likeness in dreams and verbal polysemy are the potent expression of this metaphor.Menstrual blood possesses multiple values in this scheme.A symbol of female fertility, because coextensive with it in the life cycle, menstrual blood is simultaneously the material cause of fertility as the substance out of which foetal flesh is formed.Some Cree men say that menstruating women are especially likely to conceive.Women biologically reproduce the human community, their ability to do so evidenced by the flow of blood that, however, in the event of conception, they begin to retain within their bodies.Human life is also visibly reproduced by killing and eating animals.
Hunting and trapping are paradigmatically male occupations through which men enact a reproductive role complementing that of women.The animal blood spilled at kill sites and trap sets corresponds to menstrual blood, which is a precondition of female fertility."(Brightman 1993.128-129)Red can serve as a marker and agent of transformation of a rhyton, but it can also have an apotropeic role.Red pigment can act as an additional layer of protection of vital body parts in the process of transformation; thus the interior of the recipient, 'the meaningful emptiness', the opening of the ring and interior part of the legs, and area around the genitalia.Bori≤ (2002.28)cites many examples of the use of red in folk costumes associated with fertility and protection.
Thus red dye contains a powerful agency transforming the rhyta.It is a marker, agent and protector of transformation of the rhyta.

Fractal depictions
There are a few rhyta from lowland villages which are decorated with small animal heads, added to the body or ring of the rhyta.Those animal heads look like small animals sprouting from the rhyton's surface (Fig. 6).
Alfred Gell (1998.137-141)discusses the carving of the deity A'a by the Rurutu in the Austral Isles, where the features of the god are represented by small figures which repeat in miniature the overall form of the god as a whole.It is a 'fractal image', which displays self-similarity at different scales.Its body is depicted as composed of other bodies.A'a is not an individual in the sense that there is a clear boundary between it and others.Instead, it is a 'fractal person', represented as an aggregate of external relations instantiated in the 'inside' persons.There is no border between internal and external.
This notion of 'fractal person' is based on the anthropological analyses of Melanesian societies by Mary Strathern (1988) andRoy Wagner (1991).People, objects and non-human beings are therefore indexes of relations -they are made up of, or constituted by, their relations or connections.They are not so much individuals as 'dividuals': who they are and what they do is generated by their transactions with each other, with animals and with artefacts (see Fowler 2001;2004;Jones 2005).
Animals are involved in social relations between humans through gifts or bridewealth.Animals circulate among households.Wealthy owners whose holdings exceed the maximum manageable size will find it mutually advantageous to loan or give some animals to other households.Conversely, if someone is short of animals, they may seek gifts or loans from the better-off (Dahl and Hjort 1976.136-137;Ingold 1980).Animals produce milk for the household where they are situated, irrespective of who owns a particular animal; however, the owner retains control over the slaughter of an animal and over its offspring.Alternatively, complementary types of animals allow poorer households to exploit the high reproductive potential of small stock to build their herds and then exchange them for larger stock (Dahl and Hjort 1976.230-234).Households spread their interests by distributing animals as gifts and loans to a range of stock-associates.Such herds typically consist of animals from a number of separate owners under the management of a single household.This establishes a network of social relations between households, which are reflected in herds.(Ingold 1980.17;Evans-Pritchard 1940.66-67, 153-154).
Thus someone's personhood is separable into relations with other persons, humans and animals.These are constitutive of people and animals, who are therefore composed of the sum of the relations between them.But it is not only the flow of animals and hu-mans between households which constitutes personhood.
When an animal is sacrificed at a wedding, the meat is distributed on the same pattern as a bridewealth herd.Like the bridewealth herd, the sacrificial victim is the focus of multiple claims which derive from the cumulative history of prior transactions spanning generations of both humans and animals (Evans-Pritchard 1951.66-67, 153-154).Humans and animals are therefore mutually engaged with one another, creating a complex network of reciprocal interdependence based on the exchange of substances and 'vital force'.Thus when an animal is sacrificed, the object of sacrifice is not the 'individual' animal, but a fractal person who 'stands for' the whole network of animals and humans.
Rhyta are persons, too just like animals and humans -not individuals, but 'fractal persons' composed through many relations between people and animals, or mediated through animals.
Paradoxically, I have described rhyta as if they were whole, although no unbroken rhyta has been found yet.Only fragments, parts of rhyta are found in archaeological contexts.One can only agree with John Chapman's statement (2000.67)that the process of fragmentation is vital to their understanding.

Transformation, breaking and deposition
John Chapman's innovative study (2000) focused on the practices of fragmentation and accumulation in the Neolithic of Balkans as processes which link people to object through production, exchange and consumption.
Chapman has pointed out that rhyta are highly fragmented.He states that only about 8 % of published rhyta are complete.But even this is an overestimate, as most published 'whole' rhyta are reconstructions based on a few fragments.It is safe to state that most, if not all, rhyta are broken.
Open-air sites have yielded much higher number of rhyta; for example, in Smil≠i≤, around 200 fragments of rhyta were excavated, while in the Caput Adriae caves only 22 fragments were found (Montagnari Kokelj and Crismani 1993).
Chapman explains the absence of whole rthyta and their low number in peripheral areas of rhyta distribution such as the Caput Adriae caves and Lipari, as a result of the down-the-line movement of fragments creating 'enchained relations' between people along the path.
However, the fragmentation analysis of rhyta from Smil≠i≤ and Caput Adriae caves shows that they are comparable (Fig. 7).Rhyta from the Caput Adriae caves are not more fragmented than those from open-air settlements at Smil≠i≤.There are pieces that can be fitted together, and most fragments are broken at the 'weak point' at the junction of the body and leg.While in Smil≠i≤ there are more legs which were broken below the weak point, indicating that they were broken at least twice or took more effort to break.
Two provenance studies from Triestine karst caves (Edera; Spataro 1999 and Mala Triglavca;Ωibrat Ga∏pari≠ 2004) have clearly demonstrated that rhyta were produced locally in the Karst plateau and not traded or brought from elsewhere, not even from coastal villages only a few kilometres away.On the other hand, based on one typological criteria, one fragment of a ring handle from the Bosnian site Obre I was identified as an "import from the Adriatic coast" (Benac 1973.84, Fig. 16).The deliberate breaking of rhyta is an efective ritual, where it is transformed into something else.In the process of breaking, the 'significant interior', the 'vital force' of a rhyton, is revealed and released, while the rhyton is fragmented into parts.The resulting fragments may now have different quality as a whole rhyton, and may be circulated among the participants, perhaps in the same pattern as meat or a bridewealth herd.Births, marriages, deaths and epidemics may have formed the context for the deliberate breaking of rhyta.
Fragments of rhyta were then moved around the settlement (or camp) and deposited in the same deposits as the body parts of slaughtered animals and pottery fragments.Rhyta fragments in lowland villages are often associated with anthropomorphic and zoomorphic figurines and decorated 'lids' or phalluses (see Batovi≤ 1968).
Rhyta fragments from upland caves were most probably found in contexts which were the result of periodic cleaning of the cave interior (after or before visit of a herd), with pottery, animal and human bones mixed together in the sediment and deposited at the cave wall.
The metonymic qualities of many deposited objects (such as parts of rhyta, human and animal bones) suggest that they were concerned mainly with the maintenance of the flow of vital force and not with 'practicalities' of living floor maintenance (Brück 1999;Chapman and Gaydarska 2006.71-79).These deposits are not generalized, de facto refuse, but rather special, structured deposits, which maintain or established relations between humans and animal herds, place and previous occupations.

Conclusions
Rhyta are highly ambiguous depictions composed from many attributes derived from animals and humans, especially the female body.However, they are not mere depictions as such, or some kind of ritual or cult paraphernalia, but also powerful agents.They do not represent, they do, act, and change relations between persons.
Rhyta have a life history.They are modelled from clay, incised with decoration, fired.Their potency is crea-ted through the effective arrangement of iconic elements.In the process of being made, a rhyton is transformed from raw materials into an object of power.
But the rhyta are not only powerful by themselves: their potency is acted out and enhanced through ritual and performance.We have already noted the role of red pigment, which was applied after the firing, effectively starting and marking the process of transformation of rhyta.But the most powerful transformation of rhyta was their deliberate breaking when their 'vital force', contained in a 'significant interior' was released, and the resulting fragments were exchanged and deposited.
Animals in the Neolithic societies of eastern Adriatic are not just passive 'meat packages', waiting to be slaughtered and consumed.They are persons, involved in a complex network of exchange of substances and 'vital force' with other animals and humans.As such, they have power to 'act back' and alter human life.The politics between animals and humans is therefore not based on human 'domination', but rather on balanced reciprocal relations derived from the an animistic ontology of life-force flow.Doing good politics involves the maintenance of social relations.The making and deliberate breaking of rhyta might have been a way of reciprocating the flow of vital force.They invoke the presence of non-human sources of power, animal spirits, with which humans must perforce transact in order to keep vitality in circulation.Births, marriages, deaths, epidemics, points in the animal reproduction cycle or other events where the balance of vital force was disturbed, may have formed the context for the deliberate breaking of rhyta.There might be subtle differences in the context of breaking and deposition of rhyta, based on different social relations between humans and animals in the Neolithic communities of the east- ern Adriatic.Thus rhyta from the lowlands emphasize the fractal nature of persons, humans and animals, created through the exchange of animals and persons, while upland pastoral communities might be focused more on the flow and exchange of vital force.
Rhyta are made by humans, and imbued with life through a variety of iconic properties and external activities.They embody complex belief about the connections between animals and humans.Rhyta are the human/animal politics made durable.They literally render the social negotiation between humans and non-humans visible and tangible.

Fig
Fig. 5. Traces of red pigment on the rhyton leg from Smil≠i≤ (Archaeological Museum Zadar).