Archaeological data on symbolic thinking in the European neolithic

Eszter Bánffy
(Archaeological Institute of the HAS, Budapest)

In this short paper I would like to add to the questions about the formulation and use of symbols, of symbolic thinking in the Early and more developed Neolithic era of the Carpathian Basin and its surrounding area, the Balkans and the inner parts of Central Europe. While excavating, field archaeologists often do not have the time or opportunity to meditate on the symbolic meaning of artifacts. It is often the archaeological context that might be a help in the interpretation -- a phenomenon that can be observed or neglected by the excavator, and that perishes immediately afterwards. In periods before writing appeared, it is exclusively the excavator's responsibility, whether any hints for the thoughts lying behind artifacts can be registered. I should like to draw attention to some prehistoric finds in which the thought behind them may well be a symbol, as part of a whole symbolic system.

The Greek word SYMBOLON not only means a sign for something, but the affix SYN- (con- in Latin) means it is something which goes together with something else: the archaeological object or the custom practiced in the given society has got a practical, everyday meaning But it is not the only key for interpreting the find or phenomenon. Something goes together with something else, which is of importance to a group of people.

Specialists of these problems agree that there was a period - an early stage in the development of humankind - in which we most probably cannot speak about symbols being used. And they also agree that there is a period - beginning with the late Palaeolithic - where they are definitely present.

According to Mithen, symbolic thought, the use of symbols, appeared in the Upper Palaeolithic when the human mind integrated the knowledge and information stored in separate 'chapels' of the brain and transformed it into a 'cathedral', and thus became capable of drawing conclusions based on knowledge acquired in other wakes of life (Mithen 1996a: 151-153). In his view this marked the birth of 'cognitive fluidity' and of the 'modern mind', of flexible and creative thought (Mithen 1991; Mithen 1996a: 153, 165-166). I. Wunn noted that two types of realities are in constant flux in the human mind: information on the surrounding world of lesser importance and information provided by the nervous system of greater significance is combined to create an evaluation of the world and of the ego (Wunn 2001: 10-11). Symbolic thought, called to life by the need for communication and the general necessity for co-operation, can be demonstrated for hunter-gatherer societies. Although the symbols themselves are arbitrary and the meaning attributed to a specific symbol may vary in space and time, symbols were necessary for the organization of hunting and the distribution of the booty and other resources, as well as for transmitting knowledge to younger generations. St. Mithen argued that a receptiveness to symbols can also be traced in the manufacture of purpose oriented tools instead of the earlier, uniform implements - by removing the superfluous sections of an antler, a special tool suited only to fishing was created (Mithen 1996a: 185). In this sense, transformation itself is a symbolic event.

Rituals, an early form of religion, were most likely practiced by pre-neolithic hunter-gatherer communities. In an earlier study I have discussed the concept of numinosum introduced by R. Otto that I consider to have been fundamental in this respect (Bánffy 1990-91: 196; Otto 1963). The concept of sacré as an expression of a universally present conscious structure was first proposed by M. Eliade, based on R. Otto's thesis (Eliade 1969; Eliade 1976). The idea of the "awesome" and of the "sacred", the practice of conferring qualities transcending biological and everyday experience on various objects and phenomena developed in all mobile communities. The core of any religion is made up of two components: (a) the belief that inanimate objects of nature (water, rocks, the Moon) possess the qualities of animate beings (humans, animals, plants): they too are born, they live and they eventually die; and (b) the belief that all actors of the world may possess qualities that contradict the laws of biology: these include man himself after his death, all beings that belong to a supernatural power and, later, deities. Irrespective of whether one attributes a sacral or a practical function to Upper Palaeolithic cave art, the archaeological evidence suggests that the mental structures described above emerged well before the Neolithic and that they had an impact on the mobile communities' perception of the world. From which it follows that rites, a series of repetitive acts ensuring the correct functioning and order of the world too made their appearance (Mithen 1991; Mithen 1996b: 86-95; In his new study, St. Mithen lists five traits common to all religions: Mithen 1998: 98-99).

If inanimate objects can behave as animate beings, animals too can have the same range of feelings and behave as humans or, conversely, humans can behave like animals. One of the basic qualities of the world is that all of its actors function similarly - plants and especially animals can think and act according to a human mentality. The opposite can also happen. The antrhopomorphization of animals can be traced in folk tradition to this very day, while its opposite, totemism was a salient feature of all hunter-gatherer societies. St. Mithen has argued that transitional creatures, beings that are half animal and half human, certainly existed in the human imagination since the Upper Palaeolithic. To which we may add that these creatures survived into the Neolithic and the mythical beasts of classical mythologies are most certainly their descendants.

Animals were especially important elements of nature in the life of the hunter-gatherer communities: they were a major source of food, their behaviour often forewarned of some danger and by the Mesolithic some became man's companion: dog was a domesticate by this time and we know that pig and, in some areas, cattle was domesticated in temperate Europe during the Mesolithic (Zvelebil 1995: 86; Rowley-Conwy 1986: 23).

No matter whether all elements of the above inferences are acceptable or not, symbolic thinking is there by the life of early food productive communities, in a ready, developed form. In the following, I try to argue for the truth of this statements by two examples. One is a custom, a phenomenon that can be observed in the 6th Millennium BC architecture of the Carpathian Basin and of Central Europe. The other example is an object type within its archaeological context, occurring in the Neolithic of the Carpathian Basin and of South east Europe, where beneath the every-day meaning a symbolic interpretation is well provable.

The symbolic significance in the orientation of Linear Pottery Culture houses

The siting and orientation of a settlement is generally based on two main considerations: climatic conditions (this being true of both larger regions and smaller, local areas) and cultural traditions. The houses on most early Linear Pottery settlements were strictly north to south oriented. My newly excavated 6th Millennium BC houses in Western Hungary (Transdanubia), (Szentgyörgyvölgy)-Pityerdomb are no exception. Let us briefly review the possible factors influencing this orientation.

As regards the larger region, the prevailing wind direction was northwesterly, as in most parts of the Carpathian Basin. In the case of a north oriented house, only its narrower part would be exposed to the stormy winds. It has been suggested that slightly stronger westerly-northwesterly winds can be assumed in the 6th millennium BC (Mattheußer 1991: 35). The wind perhaps blew through the roof structure and cleared the smoke from the house (Coudart 1998: 86-89 and Fig. 102); the smoke from the hearth in the southern part of house II probably left through the roof structure or the southern entrance. It must also be borne in mind that forestation is highest in western Transdanubia and that the region was probably even more heavily forested in the Early Neolithic than it is today. It is my assumption that there was a wooded tract between the two houses, explaining the total lack of finds and settlement features between the two buildings, and a similarly wooded area can be assumed around house I. It is possible that the forest was cleared to some extent around the houses and that the area, constantly trodden by domesticates, prevented the forest from encroaching on the settlement. E. Mattheußer localized the earliest Linear Pottery settlement among 40 m high trees in a densely forested area where the need for protection against the violent winds was hardly a consideration (Mattheußer 1991: 39). In other words, the strict northward orientation of the buildings could hardly have been motivated by practical considerations, such as the prevailing wind direction characterizing the broader region.

In view of the fairly large distance between the two houses at Pityerdomb, it seems rather unlikely that their siting and orientation was influenced by each other's position. The settlers chose the western slope of the roughly east-west ridge. The only practical reason for this choice appears to be the fact that the nearest source of water, the Szentgyörgy stream, flowed by this side of the ridge. Since we did not find any wells during the investigation of the settlement, it seems likely that the occupants obtained their water from this stream. The practical reason for choosing the unfavourable northern side may have been the proximity of water. However, this can hardly be seen as a reason for the northern orientation of the houses in the case of a hilltop settlement. The practice of erecting houses perpendicular to the source of water has been observed on a number of settlements lying on river terraces.

The above would suggest that the northward orientation of early neolithic Central European typed houses had been influenced by some cultural tradition, rather than climatic considerations or local conditions. The extremely conservative nature of this custom can also reveal something about its co-meaning, i.e. symbolic character.

When considering a larger region, we have to face with the fact that the orientation of the Early Neolithic houses from South-East Europe is rarely specified in the publications and very often the illustrations lack a north arrow. C. Lichter's monograph, covering the entire Neolithic and the Early Copper Age (and thus with only a few data on early Neolithic houses), contains a wealth of data on the orientation of prehistoric buildings; his book, however, contains little information on Early Neolithic houses (Lichter 1993). A collation of his catalogue with my own data collection offers a number of interesting points concerning the orientation of Linear Pottery houses.

The first group is made up of settlements where there is no indication of a uniform orientation, suggesting that the houses were sited according to some other principle. Most of the southern, stratified tell settlements can be assigned here. One good example of this type is provided by the lower levels of the Karanovo tell: the buildings of phase I and II were more or less aligned to each other. The same phenomenon can be observed on a number of other settlements, where the space enclosed by the houses was more important than the orientation of individual buildings (Cf. J. C. Chapman's theory "from space to place": Chapman 1989a; Steadman 2000: 194). It has also been suggested that one of the main principles of tell architecture was that new houses were erected over earlier ones and that the continuous use of the areas between the newly built houses was a form of remembering the ancestors and expressed "the continuity of the house and the household" (Kotsakis 1999; Nanoglou 2001). In the case of closely packed houses, the position of neighbouring buildings was also important. At Servia, for example, it is fairly obvious that the houses had been built relatively close to each other, without a uniform orientation. The same holds true for the Veluška Tumba settlement, where the first house was east-west oriented, while the second was north-south oriented. Two west-east and two northwest oriented houses were found at Cavdar. It would appear that the orientation of individual houses did not play a role on stratified tell-settlements and that the position of houses or house clusters relative to each other was more important. These settlements can mostly be found in the southern Balkans, where the strict orientation characterizing Linear Pottery houses is lacking.

The second group is made up of the Early Neolithic houses that have a specific orientation, but this differs from the strict northern or northwestern orientation of early Linear Pottery buildings. Examples for this can mostly be quoted from northeastern Bulgaria and the Lower Danube region in Romania. At Ovcarovo-Platoto, for example, the orientation of the buildings of the preceding phase took primacy over an orientation according to the cardinal points. The houses of the Suplacu de Barcau (Berettyószéplak) settlement in Romania shared a common orientation, usually to the east or northeast. The same orientation characterized the buildings of the Cîrcea-Viadukt settlement. A northeast oriented building of the Alföld Linear Pottery has been reported from Krasznokvajda, lying on the fringes of the culture's distribution, in the Cserehát Mountains (Losits 1980). Three other early settlements are known from eastern Hungary where the houses were east-west or east-southeast-west-northwest oriented: a Körös house from Tiszajeno and from Szolnok-Szanda, as well as the houses of the early Alföld Linear Pottery culture from Füzesabony-Gubakút, probably contemporaneous with the late Körös culture (Domboróczki 1997: 20-21; Domboróczki 1999; Domboróczki 2001a: 84). A late Linear Pottery site was investigated during the rescue excavations preceding the motorway construction leading northeast from Budapest; all the settlement features were aligned towards the northwest (Bánffy 1999: 140-142). Although very few Alföld Linear Pottery houses are known, it seems likely the buildings of the Kompolt 14 settlement followed the traditions of the early Szatmár II period.

The tradition of aligning houses to the north irrespective of the narrower or broader environment appears to have originated in western Bulgaria, northwestern Bulgaria and the Morava valley. The southernmost such houses have been reported from Nea Nikomedeia, where twenty-four buildings were aligned to the north or the northwest. A similarly north-oriented house is known from Porodin. The Jaša Tepe, Muldava, Slatina and Gradešnitza-Malo Pole sites too featured north-oriented houses; a total of sixteen houses were aligned to the north at the latter settlement.

Only one single piece of information for orientation is known from the Starcevo settlements in the Srem (Croatia): the three rectangular buildings uncovered at Zlatara were north-south oriented (Lekovic 1988a: 109). Unfortunately, information on the orientation of the buildings are lacking exactly from the northern and northwestern distribution of the Starcevo culture, from the very region that may have been a contact zone. In the lack of data and in view of the other archaeological evidence for cultural diffusion, we may at the most only assume that the above-ground houses of the Starcevo culture - that must have existed - in Slavonia and southern Transdanubia were oriented to the north.

We know that a few oval buildings of the Central European Mesolithic were aligned to the north: even a hearth in one such building was oriented to the north (the houses at Oerlinghausen and Sarching. Cf. Luley 1992: 6, 147, 150). We may therefore assume that the northern orientation of the buildings had some kind of significance. Even though there is no evidence for Mesolithic buildings from Transdanubia, the Mesolithic finds suggest that these communities maintained contact with each other and were perhaps also related.

As it has been said before, the two houses uncovered at (Szentgyörgyvölgy)-Pityerdomb were exactly north-oriented. This is the southernmost site from where there is information on the orientation of early Central European buildings. The list can be complemented with a number of Austrian and German sites, such as house 4, the earliest Linear Pottery building at Strögen, also oriented to the north (Lenneis 1995: 17). The early Linear Pottery houses found at Schwanfeld, Mintraching, Wang/Freising, Enkingen and Kleinsorheim in Bavaria were also strictly aligned to the north (Lüning 1984; Lüning 1987), as were all the earliest buildings at Bylany in Bohemia (Pavlu 1982: 197). Later, the orientation of Central European Linear Pottery houses diverged a little to the west; an excellent discussion of the causes leading to this shift in orientation has recently been published (Coudart 1998: 78-91).

In sum we may say that the Central European house developed from the house construction traditions of the local Mesolithic communities and the Neolithic houses of the western Balkanic region. Both traditions included above-ground buildings. The massive roof structure, the "Längsgrube" (i. e. long ditches along both longer sides) and the strict northern orientation can either be attributed to Mesolithic influences or was a joint tradition resulting from the contact between the two populations.

The house-builders apparently took some features as stricktly canonized rules. Climatic factors undoubtedly played a role (roof structure, the increased importance of timber as a building material, the partial roofing of the work-pits), but perhaps even to a larger extent some other, cultural, social and symbolic traditions (orientation).

Although the above outline of the possible origins of the Linear Pottery longhouse contains a number of hypothetical elements, one point emerges quite clearly. It has been suggested in relation with the mesolithic buildings in the Iron Gates region that a change in architectural elements reflects a change in the cultural identity of the given community (Radovanovic 1996: 41). Although I regard this observation highly relevant, it seems to me that in the case of the canonized orientation of Linear Pottery houses we are witnessing the exact opposite. No matter to what extent the above suggestion on the origins of Linear Pottery longhouse will be confirmed or rejected, irrespective of where and when this house type first appeared, there can be no doubt that its emergence marked the birth of a community's symbolic identity. Clearly, this symbolic and cultural identity was to a large extent reflected in the conservative architectural elements of the Linear Pottery communities from the Vistula to the Rhine, from Magdeburg to Hódmezovásárhely for several centuries. Moreover, the late neolithic groups of Central Europe and the Carpathian Basin, developing on a Linear Pottery basis and no doubt indicating the survival of the Linear Pottery population, continued this tradition. This can be ascribed to the cross-cultural and extremely conservative nature of this -probably important - symbol in the system of neolithic symbolic thinking.

The cult corner within Neolithic houses and its find assemblage, with a special reference to house models

The Neolithic buildings in South East and Central Europe have yielded cultic objects as well. However, their number was insignificant compared to those countless idols, anthropomorphic vessels and other such finds which have come to light at settlements. The questions of where and, more importantly, how this great amount of objects was used has not been answered satisfactorily as yet. My attempt below to find an appropriate answer starts out from the archaeological context of these objects.

Idols and similar cult objects which have come to light from findspots other than burials or their proximity, or from sacrificial pits, are associable with the sacred part within an every-day-house, a "cult corner". The number of sufficiently researched and recorded cult corners is still too small to enable us to reliably reconstruct or typify them in details. In the single-room houses, the cultic part was most probably situated in a corner near the wall. In the late neolithic three-room houses, the cultic part should be sought in one of the outside rooms. The proximity of fire must also have been important. These cultic parts often included clay benches or (wooden) stands, and occasionally pits dug in the floor. Other common inventory items were the omphalos, clay ornaments applied on the floor or on the wall of the cult corner (e.g. bull's horns), coarsely and finely executed idols, and occasionally the "altarpieces" (miniature or life-sized pieces of furniture). In those houses where there was no separate store-room, this cult corner must have been used for storing the anthropomorphic vessels filled with the life-giving corn. (The vessels with facial lid discovered at Vinca are believed to have served a similar - chthonic - function, and the same should apply to those assemblages where there were idols hidden in the vessels filled with corn: cf. The Cucuteni-Tripolje culture and the Selevac site of the Vinca culture (Gimbutas 1982, Chapman 1981: Pl. 26). However, the proximity of some other vessels, stone objects and the fireplaces all indicate that the "cult corner" was also the venue for other religious activities and offerings.

The researcher attempting to interpret the cultic objects discovered in the sacral part of the buildings is bound to walk on real thin ice. And yet, there is no disregarding in the following questions: why were there ornamented and finely executed idols at one place and coarse and roughly executed ones at another, seemingly very similar spot, and what was the purpose of the other, non-cultic objects which accompanied the idols? To carry on with the questions: Is it possible to interpret the whole assemblage as symbolic? Or only the finely elaborated pieces belong to this category? Do these hypothetic symbols formulate a system? Since we know of a number of assemblages which contained both finely executed and coarse pieces, it appears to be justified to leave the talents of the craftsmen who produced them, out of consideration here. The significance of this qualitative difference, which applied not only to the idols but also to the house models, should thus be sought somewhere else.

Researchers have come out with countless hypotheses and theories to date on what the actual use of these "cult objects," especially of figurines could be. M. Gimbutas' reconstruction of an eastern Mediterranean and South-East European pantheon populated chiefly by female goddesses is rather arbitrary to say the least. Her books paint an idyllic, almost utopian world of peace preceding the world dominion of men (Gimbutas 1982; Gimbutas 1989a; Gimbutas 1991). Certainly, her ideas were mostly rejected by the processual archaeologists but also strongly criticized by the followers of "gender" and "postmodern" archaeologists. The reaction to the positivism of Gimbutas' critics led to the rise of post-processual 'reflexive' archaeology, as well as to several new studies presenting and analyzing neolithic finds and religious beliefs, enriching the already prolific works in this field. The most outstanding representatives of this approach regard the archaeological heritage, and especially the cult assemblages containing little data and allowing a wide berth for interpretation, as the fossils of a set of symbols, the integrated part of a bygone system of communication, and the reflection of spiritual contents in the material culture.

Having analysed a large collection of finds from the Mediterranean, P. Ucko and L. Talalay concluded that their function of the figurines was far from uniform (Ucko 1968, Talalay 1983). While Ucko never stepped beyond this conclusion, Talalay went further by stating that the idols should be seen as a form of non-verbal communication. According to Talalay, the human body as a social symbol has always had a prime significance throughout history. Accordingly, it appears to be possible to create an artificial language whereby this system of signals could be deciphered, Talalay would like research to concentrate on the following issues: the subject of the depiction, anomalies dictated by experience, ornamentation and deliberate alterations made subsequently (e.g. fragmentation). Her conclusion is definite on two points: on the one hand, she considers the idols symbols which are approachable through the social aspects of religion, and on the other hand she agrees with Ucko that the idols should be regarded as the manifestations of more than one religious concept.

The main line of reasoning in the interpretations challenging M. Gimbutas' views was that not all idols can be regarded as depictions of the Magna Mater since they probably had a variety of other functions. P. Ucko's monograph categorized the female depictions of the Eastern Mediterranean in this spirit, similarly to the more recent studies written by L. Talalay, Chr. Marangou and P. Biehl (Ucko 1968; Ucko 1996; Talalay 1987; Marangou 1992; Biehl 1996; Biehl 1997). D. Bailey, K. Gallis and L. Orphanidis went as far as to suggest that the Neolithic idols from Bulgaria and Thessaly were in fact individual portraits (Bailey 1994; Gallis-Orphanidis 1996). I. Hodder believes that the figural representations of the Neolithic were an expression of power created by the depiction itself - to which D. Bailey added that idols were means of transmitting ritual knowledge and important cultural values. B. Bender noted that control of ritual knowledge was a sign of the existence of a social hierarchy (Bender 1985).

If we accept these views, Neolithic idols should be interpreted along entirely different lines than previously. We should forget about "Frog Goddesses" and "Resting Peasants" alike. A different meaning can be construed for idols recovered from different contexts: for example, the high number of more or less identical idols used for initiation rites in a communal building or a statuette from a house used before giving birth vested the same idol with a different meaning, ranging from teaching devices to a chthonic function. Still, it is unlikely that these figurines depicted priestesses. J.-P- Démoule and C. Perlès' analysis revealed that there is nothing in the archaeological record to suggest that such an institutionalized function could be attributed to these idols (Démoule-Perlès 1993). If this is indeed the case, we can hardly speak of goddesses or deities. In their study on the figural representations of the Aegean, D. Kokkinidou and M. Nikolaidou arrived at a similar conclusion: instead of a pantheon, the cult life of farmers was characterized by some form of ancestor worship, a belief in higher, impersonal natural powers and rituals performed inside the house (Kokkinidou-Nikolaidou 1997: 101).

A new approach like this must definitely take the archaeological context as its point of departure. Since this research has yielded only initial results to date, it would clearly be premature to try and summarize it now. I agree with those who say that the idols differed according to their functions. However, we cannot specify as yet the function of the finely executed pieces as contrasted to theat of the coarse ones. But there are two questions we may attempt to answer right now: what dates can we assign to the former and the latter types, and how were they used in the cult corner?

Since these questions would be difficult to answer exclusively on the strength of the available idols, I decide to approach the problem through another group of finds, which are most specific, whose form is easier to interpret, and whose quantity is easier to cope with. These are the house models. I believe that the research into the function of the house models can be likened to the point at issue here in terms of both the methods and the projected results. Perhaps the only significant difference is that the house models are much easier to analyse.

The number of house models coming from Central and South east Europe and dating from the period between the Early Neolithic and the Chalcolithic varies widely by their age. For example, the house models discovered in Hungary basically date from three periods: the early neolithic Körös culture, the transition period between the Middle and Late Neolithic and the late neolithic Lengyel culture. The fluctuation cannot be accounted for by the varying intensity of archaeological research, since all the periods have been studies steadily and uniformly. The highly conservative tradition, that continues to survive over many centuries in the 6th-4th Millennium BC, can also be an argument for their symbolic character (Schwarzberg 2003: 81).

From a typological point of view, the house models can be classed into five groups.

The first group includes the early neolithic house model from Röszke (Trogmayer 1966: 235-240, Figs 1-2) and the somewhat younger pieces associated with the Thessalian Sesklo culture (Khaironeia: Theokharis 1981: Fig. 6, Krannon: Theokharis 1973: Fig. 29, Stephanovikaios region: Papathanassopoulos 1981: Fig. 19). They can even depict two-storey houses, proving the existence of this kind of architecture in neolithic (Baku, Larissa: Toufexis 1996: 161 and Cat. 264). Characteristic of these models is their naturalistic execution, to the extent that some of them even exhibit architectural motifs.

The second group is represented by the simple model applied on lids. Thes models are often termed by literature as "ovens". I have to add here that this term I consider fairly obscure, and for the lack of the exact definition it remains a matter of taste whether we identify a model as a "house" or an "oven". Beside five such pieces from Aszód, and the Slovakian finds which undoubtedly depict houses, finds of this kind have come to light in large numbers in the 5th Millenium BC, at the sites of the Balkan Early Chalcolithic Gumelnita culture (Ovcarovo: Todorova 1979: Figs 63, 23/1-5; Kodžadermen: Müller-Karpe 1968: Figs 157/C/1-3, 6; Perniceva 1978: 2, Fig. 1; Ruse: Kostov 1926: Fig. 117; Gaul 1948: 49; Vinica: Radunceva 1976: Figs 5/4, 10/13, 29/6, 42/5, Radunceva 1976: Fig. 7/9; Azmaška Mogila: Georgiev 1962: Figs 2/a, b; Nevski: Perniceva 1978: Figs 3/3, 6/7, 6/9; Deve Bargan: Popov 1926: Fig. 157), Sitagroi: Toufexis 1996: Cat. 267). Since these latter finds postdate the Aszód pieces and were discovered at a remote distance as well, we have no ground to prove their relationship with the Hungarian Late Neolithic as yet.

The models in the next two groups depict only certain parts of the houses. Quite often this part is the floor, and the interior is depicted in these models includes the fireplace, the furnace, and occasionally the bed and also some pithoi. Perhaps the best known of these models are the ones recovered at Ovcarovo and Popudnia (Todorova 1979: Fig. 63; Müller-Karpe 1974: Figs 677/1/1-2; Gimbutas 1982: Fig. 23).

As to the fourth type, I know of only one model which depicts the walls and roof of a house, without a floor. This model has been brought to light recently from the early Tisza period layer of the Öcsöd-Kováshalom settlement (Bánffy 1985: Pl. 26/1: Bánffy 1990/91: 213). The structure of this model resembles closely that of the real houses of the period. Its surface is decorated with incised Tisza culture patterns and with white and yellow painting. The fragmented model has no door, window or roof-hole. This model is considered unique on the grounds that it has no floor or footing, as contrasted to the majority of te Chalcolithic or even Bronze Age house models that have come to light anywhere between the Middle East and Central Europe, which were box-shaped and were thus fit for storing things in them. Not improbably this find could be identified as the upper part or cover of an Ovcarovo-typed house model. Reasonable as this presumption may seem, the lack of parallels prevents us from jumping to conclusions here.

The main peculiarity of the last group is that they depict "unusual" houses. On the strengh of this apparently symbolic depiction, the majority of these models have been identified as "sanctuary models". By way of example, we can cite here the models at Trusesti and Porodin (Petrescu-Dîmbovita 1950: 172-186; Grbic 1960: Fig. 34/1). The "unusual" character of these models is exemplified by the animal and human heads applied on their lids. The Cascioarele model depicts an unusually arranged group of houses and a stand. The incised patterns on it, which M. Gimbutas identified as a water motif, emphasize further the ritual character of the find (Dumitrescu 1965a: 215-218, Figs 2a-b; Dumitrescu 1965b). There is another peculiar type of these so-called "sanctuary models": there the house was fit together with a dish, which was most probably used for sacrificial purposes (Öcsöd: Bánffy 1985, Raczky 1987). These signs include the stylised "holy horns" on the thresholds of the four doors, the omphalos-like double ring in the middle of the floor, the bowl painted red and yellow on both sides and applied on the top of the house, and finally, the four figurines fit on the rim of the model. Adding to the significance of the Öcsöd model is the fact that its quality of execution is beyond compare in the Middle and Late Neolithic of Hungary.

Let us now try to approach the problem of the house models from a peculiar angle, which I believe will lead us to some valuable conclusions on cult objects in general as well as perhaps more on the cultic life and symbolic thinking of the Neolithic and chalcolithic groups. I must begin by stating that I am fully aware of the limited scope of such an approach, and that my exposition is bound to remain sketchy for the lack of reliable data and research background.

On the strength of the associations between the archaeological assemblages we can establish that the house models are markedly homogeneous in terms of their provenance: all of them come from settlements, and the circumstances of their discovery indicate that they were all used in houses. Several of these models were found inside buildings, on the floor or in the debris. At the Aszód site, the models were discovered in the refuse pits next to the houses. To mention some examples: Troy (Blegen 1963: 53-55; Öcsöd: Bánffy 1985: Fig. 26; Aszód: Kalicz 1985: Fig 3, 4a-b, 27/5a-b; Popudnia: Müller-Karpe 1974: Figs 677/1/1-2; Vadastra II: Mateescu 1962: Fig. 189/2; Trusesti: Petrescu-Dimbovita 1950: 172-186; Izvoarele. Vulpe 1957; Gimbutas 1982. Figs 69, 22; Cascioarele: Dumitrescu 1965a: 215-218, Figs 2-5; Idem. 1965b; Porodin: Grbic 1960: Fig. 34/1, Bereketska Mogila: Jungsteinzeit in Bulgarien: Fig. 146a; Gradesnitsa: Nikolov 1974: Figs 18, 65; Ruse: Gaul 1948: 119, 140; Vinica: Radunceva 1976: Figs 5/4, 7/9, 10/13, 29/6: Platia Magula Zarkou: Gallis 1985: 20-24). Consequently, the function of the house models must be associable with the houses themselves.

To the best of my knowledge none of the house models recovered to date have come to light as grave-goods in the Neolithic and Chalcolithic. In the subsequent periods, the occurrence of house models can be likened to that of idols in the point that both find types were missing at the Early and Middle Bronze Age sites. They reappeared at sites in Italy and the East Mediterranean only after the 12th Century BC, but then these objects already served as "house-urns" (to contain human ashes) (Oelman 1959; Trianti 1984, Staccioli 1968). I wish to emphasize here that I could establish no link between these latter objects and the pre-Bronze Age ritual models. (Any reference to this stage to the presumed relationship between these two types of objects is bound to remain hypothetical. However, the facts remain that the above-mentioned hiatus applies to both types, and also that while the majority of the neolithic and chalcolithic idols comes from settlements, the "re-vival"-period idols were exclusively burial offerings - Letica 1973.)

I am inclined to subscribe to the theory which maintains that the houses with erect walls derive from the roofless abodes or the wind-breaks erected around the fires. Accordingly, the original notion behind the term "abode" must have been the fireplace and not the "enclosed space". Ever since structures have been built around fireplaces, they have been considered the focal part of the buildings. In each house the fireplace signalled the link between the ancestors and the descendants, and it als symbolized continuity and survival. Consequently, the rites as well as the symbolic objects connected to these notions must have taken place at the fireside.

A recently discovered find, mentioned already, from Thessaly proves sufficiently that it is far from accidental that the house models regularly come to light in the proximity of fireplaces. At the Platia Magula Zarkou site a house model was buried in a pit dug next to the fireplace of an early Dimini-culture building (Gallis 1985). The inventory of this building bears a close resemblance to that of houses dating to the Tripolje culture in Ukraina, and also to that of the Ovcarovo site (Bulgaria). Besides a fireplace and some pieces of furniture, this house model also includes figurines representing the members of a three-generation family: grandparents: adult and younger children and even a young couple with a baby. The excavator interpreted this find as a construction offering, quite common in the south east European Neolithic and Chalcolithic. However, the arrangement of the find can clearly be considered unusual: as we will se below, the construction offerings were normally placed in pits which the builders had dug prior to the commencement of the building operations. In some instances these offerings were placed straight to the foundation pit. Undoubtedly, the aim of these offerings was to ensure the success of the constructions. But the pit at Magula Zarkou was dug next to the fireplace, and this could only be done after the building of the house. Accordingly, the purpose of this model must have served the well-being of the dwellers of the house.

The Magula Zarkou find appears to support the assumption that the house models were meant as a symbol to protect life and well-being of the dwellers, i.e. that their ritual function was primarily protective.

In reference to the conclusions reached above, I now make an attempt to analyse the function of the five house model types.

The function of the naturalistic types must have been to symbolize the building itself, and the same should apply to the models representing the top of the bottom parts of the houses. Although the ritual role of these models has been treated as a commonplace since the publication of Frazer's work (Frazer 1965), we have no ground to doubt that these models were used for signifying whatever the dweller wanted to happen to their real house. Under this parts pro toto (synekdokhe) principle, the dwellers kept the model in a well-protected spot near the fireplace in order to ensure the safety of the building itself. This in other words means the people of the day attempted to control their fate by performing the impending series of acts beforehand. This is otherwise a clear example for Fraser's transforming magic, too.

Let me refer back at this point to the figurines again. In my opinion, the key element of the rite described above is that the objects may have played an active role. Moreover, we have every right to believe that one type of the house models were roofless just to enable the owners to furnish them with chairs, tables and anthropomorphic and zoomorphic figurines. And there are three more facts to be considered here:
1. Quite a number of these cultic objects are discernably worn;
2. If we presume that the objects at issue were treated as "active participants" in a rite and not as images of super-human and statically reigning gods imbued with the mysterium tremendum, we get a step closer to understanding the marked abundance of them at certain sites;
3. The regular production, use and discarding of these ritual tools may well account for the fact that the idols often exhibit traces of intentional fragmenting. The people of the day could well associate their cult and the related objects to limited periods only. The ritual accessoires associated with the sowing season they could not - or perhaps were not allowed - to use on other occasions, e.g. during harvesting. Similarly, the ritual accessoires used during burials or at the bedside of women in labour could well be considered disposable.

The next type of the house models is the one applied on lids. Their role must have been similar to that of the odels described above, although they were related more "passive". I would liken the function of these models to that of the anthropomorphic vessels. A vessel with a lid which had a house model as a handle must have contained something to which the family attached special importance - sowing seed, for example.

I deliberately left the issue of the so-caled "shrine model" last in this assassment. The reason is that at this point we have to decide whether these objects should be termed "house models"or "sanctuary models". In other words, the question is whether these models served sacral or profane purposes. On the one hand, as we have seen earlier, they could be naturalistic, i.e. they could be patterned after (parts of ) real structural elements. On the other hand, however, there are countless characteristics of the models with archaeological context which lead us to conclude that thex must have been associated with the notion of the "sacré". Now, is this a contradiction here?

I believe that the same questions could be put in connection with real dwelling houses as well. After all, while the houses obviously served a host of profane purposes (the dwellers lived, worked, slept and ate in them), there are indications that they also had sacral significance. In proof of the latter point we could cite the practice of burying the dead under the house floor, the construction offerings, the existence of the "cult corners", the animal heads applied on the gable or the numerous objects with presumed cultic significance found inside the houses.

On the strength of these facts, and approaching the problem from a purely archaeological viewpoint (i.e. disregarding the religiohistorical considerations), we are bound to conclude that the function of the dwelling houses in the Neolithic and Chalcolithic periods was neither exclusively sacral, nor exclusively profane, but instead, it was both. Ritual acts also had their every-day use and meanwhile, profane activities and objects too had a bit of symbolic meaning. In all probability this distinction between these two qualities was far less marked at that time than it is today.

In short, we can establish that the house models were images of the real houses, and were thus also profane and sacral at the same time. To all aparences, the same conclusion applies to the other objects typical of the "cult corners" - the figurines, the anthropomorphic vessels and the small altarpieces. Just as the human- or animal-headed "sanctuary models", the ornamented figurines or the miniature pieces of furniture with incised symbolic patterns were paraphernalia used for certain festive rituals, so were the poorly executed and coarse "oven models," idols and the "furnace models" the reflections of symbolic practices done while having profane periods. There must have been more sacral and more profane periods and the cult objects prepared in each period exactly reflected this symbolicity. So can the same house be depicted in the sacral period of life interpreted by the excavators as a "sanctuary model", while a depiction of the same dwelling house in a profane, every-day period could be down-graded by considering it as an "oven model". This, I think, is what the archaeological data clearly suggest.

If we wish to go beyond this point, i.e. if we want to find out why did the attractive and ugly, finely and poorly executed objects occur regularly at each site, we should approach the problem from a religiohistorical viewpoint. One possible (as yet unproven but at least logical) answer was provided by M. Eliade. Notwithstanding the fact that in this branch of science verification is considered far less compulsory than in natural sciences in general, on this specific point Eliade's theory is still reliably buttressed by the available archaeological data. This way archaeology could prove to be a reliable source for the researcher of the history of early religion.

At this point it is to be confessed that the recent knowledge on neolithic cult activities is still rather scanty. It is possible to try to reconstruct some elements of the neolithic ritual. Firstly, that these objects were in use. Further on, where, what objects were used - and perhaps even some hints of when and how these objects were used in ritual activities - but we have hardly any knowledge about what exactly the essence of these activities were. We can try to reconstruct some elements of this system of symbols - but as yet, we should remain at this very level and avoid to make guesses about their exact religious content.

E. Bánffy

References

Bailey, D. (1994) Reading prehistoric figurines as individuals. WA 25 (1994): 321-331
Bánffy, E. (1985) Cultic finds of the 1983 campaign. In: Raczky-Seleanu-Rózsa et al.: Öcsöd-Kováshalom. The intensive topographical and archaeological investigation of a late neolithic site. Preliminary report. MittArchInst 14. 268-270, 368-370
Bánffy, E. (1990-91): Cult and archaeological context in Central and South East Europe in the Neolithic and Chalcolithic. Antaeus 19-20: 183-250
Bánffy, E. (2002) Notes on the connection between human and zoomorphic representations in the Neolithic. In: P. Biehl - F. Bertemes (eds): The Archaeology of Cult. Symbols of the other world. Archaeolingua 13. Budapest 2001: 53-71
Vaday, A. - Bánffy, E. - Bartosiewicz, L. - T. Biró K. - Gogaltan, F. - Horváth, F. - Nagy, A. (1999): Kompolt-Kistér: újkokori telep. Újkokori, bronzkori, szarmata és avar lelohely. Leletmento ásatás az M3 nyomvonalán (A Neolithic, Bronze age, Sarmatian and Avar site. Rescue excavation at the M3 motorway). Eger. pp. 13-170
Bender, B. (1985) Prehistoric developments in the American Mid-Continent and in Brittany, northwest France. In: T. D. Price - J. A. Brown (eds): Prehistoric Hunter-Gatherers: the emergence of cultural complexity. New York 1985: 21-57
Biehl, P. F. (1996) Symbolic communication systems: symbols on anthropomorphic figurines of the Neolithic and Chalcolithic from South-East Europe. JEA 1996/4: 153-176.
Biehl, P. F. (1997) Overcoming the "Mother-Goddess-movement": a new approach to the study of human representations. Latvijas zinatum Akademijas Vestis 1997: 59-67
Blegen, C. W. 1963: Troy and the Troyans. London
Bökönyi, S. (1969) A Lepenski Vir-i ostelep gerinces faunája (The vertebrate fauna of the prehistoric settlement at Lepenski Vir). ArchÉrt 96 (1969): 157-160.
Bökönyi, S. (1970) Animal remains from Lepenski Vir. Science 167 (1970): 1702-1704
Chapman, J. C. (1981) The Vinca Culture of South East Europe. Studies in chronology, economy and society. BAR IntSer 117 I-II. Oxford 1981
Chapman, J. C. (1989a) The early Balkan village. In: S. Bökönyi (ed.): Neolithic of Southeastern Europe and its Near Eastern Connections. VAH 2. Budapest 1989: 33-53
Chapman, J. C. (2000) Fragmentation in Archaeology. People, places and broken objects in the prehistory of Southern Europe. London - New York 2000
Coudart, A. (1998) Architecture et société néolithique. L`unité et la variance de la maison danubienne. DAF 67. Paris 1998
Démoule, J.-P. - Perlès, C. (1993) The Greek Neolithic: a new review. JWP 7/4 (1993): 355-416
Domboróczki, L. (1997) Füzesabony-Gubakút. Újkokori falu a Kr. e. VI. évezredbol = Neolithic village from the 6th Millennium B.C. In: Raczky - Kovács - Anders (1997): 19-25.
Domboróczki, L. (1999) Az újkokor idosebb szakasza ÉK-Magyarországon, a Heves megyei régészeti leletek fényében (The older phase of the Neolithic in North-Eastern Hungary in the light of archaeological finds from Heves County). In: T. Petercsák - L. Domboróczki (eds): Early Farmers in Europe. A korai földmuvelok Európában. Valencia 1999: 15-47.
Domboróczki, L. (2001b) The excavation at Füzesabony-Gubakút. A preliminary report. In: R. Kertész - J. Makkay (eds): From the Mesolithic to the Neolithic. Proceedings of the International Archaeological Conference held in Szolnok 1996. Archaeolingua 11. Budapest 2001: 193-214
Dumitrescu, V. 1965a: Les principaux resultats des deux campagnes de fouilles dans la station néolithique récente de Cascioarele. SCIV 16, 215-238
Dumitrescu, V. 1965b: Cascioarele. Archaeology 18
Eliade, M. (1969) The Quest. History and meaning in religion. Chicago 1969.
Eliade, M. (1976) Patterns in Comparative Religion. London 1976
Frazer, J. G. (1965) Az aranyág (The Golden Bough). Budapest 1965
Gallis, K. (1985): A late neolithic foundation offering from Thessaly. Antiquity 65: 20-24
Gallis, K. - Orphanidis, L. (1996) Figurines of Neolithic Thessaly I. Academy of Athens, Research Centre for Antiquity, Monograph 3. Athens 1996
Gaul, J. H. (1948) Gaul, J. H.: The neolithic period in Bulgaria. BASPR
Gimbutas, M. (1982) The Goddesses and Gods of Old Europe. Myths and cult images. London 1982.
Gimbutas, M. (1989a) The Language of the Goddess. London 1989.
Gimbutas, M. (1991) The Civilization of the Goddess: the world of Old Europe. San Francisco 1991
Grbic, M. et al. (1960) Porodin, kasnoneolitsko naselje kod Bitolja. Bitolj
Kalicz, N. (1985) Kokori falu Aszódon (Neolithisches Dorf in Aszód). Aszód
Kokkinidou, D. - Nikolaidou, M. (1997) Body imagery in the Aegean Neolithic: ideological implications of anthropomorphic figurines. In: J. Moore - E. Scott (eds): Invisible People and Processes. Writing gender and childhood into European archaeology. London - New York 1997: 88-112
Kostov, L. (1926) Predistoriceskata mogila da Ruse. Godisnik..Sofia 1926
Kotsakis, K. (1999) What tells can tell: social space and settlement in the Greek Neolithic. In: P. Halstead (ed.): Neolithic Society in Greece. Sheffield 1999: 66-76
Lekovic, V. (1988a) Zlatara - Ruma. In: D. Srejovic (ed.): The Neolithic of Serbia. Belgrade 1988: 108-109
Lenneis, E. (1995) Altneolithikum: Die Bandkeramik. In: Jungsteinzeit im Osten Österreichs. St. Pölten 1995: 11-56
Letica, Z. (1973) Antropomorfne figurine bronzanog doba u Jugoslavii. Beograd
Lichter, C. (1993) Untersuchungen zu den Bauten der südosteuropäischen Neolithikums. Internationale Archäologie 18. München 1993
Losits, F. (1980) Neolithische Siedlung in Krasznokvajda. FolArch 31 (1980): 7-34
Luley, H. (1992) Urgeschichtlicher Hausbau in Mitteleuropa. Grundlagenforschung, Umweltbedingungen und bautechnische Rekonstruktion. Bonn 1992
Lüning, J. (1984) Neolithische Hausgrundrisse in Schwanfeld. Das archaologische Jahr in Bayern 1983. Stuttgart 1984: 31-32
Lüning, J. (1987) Ausgrabungen zur ältesten Bandkeramik. AJB 1986. Stuttgart 1987: 33-34
Marangou, Chr. (1992) . Figurines et miniatures du Néolithique Récent et du Bronze Ancien en Grèce. BAR IntSer 576. Oxford 1992
Markevic, V. I. (1965) Issledovanja neolita na Sredniem Dniestre. KSIAKiev 1965: 85-90
Mateescu, C. (1962) Sapaturi arheologice la Vadastra. Mat. Si Cerc. Arh. 8
Mattheußer, E. (1991) Die geographische Ausrichtung bandkeramischer Häuser. In: Studien zur Siedlungsarchäologie I. UPA 6. Bonn 1991
Mithen, St. (1991) Thoughtful Foragers. A study of prehistoric decision making. Cambridge 1991.
Mithen, St. (1996a) The Prehistory of the Mind. A search for the origins of art, religion and science. London 1996.
Mithen, St. (1996b) Ecological interpretations of palaeolithic art. In: R. W. Preucel - I. Hodder (eds): Contemporary Archaeology in Theory. A reader. Oxford 1996: 79-96.
Mithen, St. (1998) The supernatural beings of prehistory and the external storage of religious ideas. In: C. Renfrew - Chr. Scarre (eds): Cognition and Material Culture: the archaeology of symbolic storage. Cambridge 1998: 97-106
Handbuch der Vorgeschichte II. Kupferzeit. München 1968
Müller-Karpe, H. (1974) Handbuch der Vorgeschichte III. Kupferzeit. München 1974
Nanoglou, Str. (2001) Social and monumental space in Neolithic Thessaly, Greece. EJA 3-4 (2001): 303-322
Nikolov, B. (1974) Gradešnica. Sofia
Oleman, F. (1959) Pfahlhausurnen. Germania 37
Otto, R. (1963) Das Heilige. Über das Irrationale in der Idee des Göttlichen und sein Verhältnis zum Rationalen. München 1963
Papathanassopoulos, G. (1981) Neolithic and Cycladic Civilisation. Athens
Pavlu, I. (1982) Die Entwicklung des Siedlungsareals Bylany 1. In: J. Pavúk (ed.): Siedlungen der Kultur mit Linearkeramik in Europa. Internationales Symp. Nové Vozokaný 1981. Nitra 1982: 193-206
Perniceva, L. (1978) Glineni modeli na zilista ot kalkolita v bulgarskite zemi. Arheologia (Sofia) 1978/2
Petrescu-Dimbovita, M. (1950) Trusesti. Iasi
Popov, R. (1926) Mogilata Deve Bargan. Godisnik...Sofia
Radovanovic, I. (1996) Mesolithic/neolithic contacts: a case of the Iron Gates region. Porocilo 23 (1996): 39-48
Raczky, P. (1987) Öcsöd-Kováshalom. In: The Late Neolithic of the Tisza region, Budapest-Szolnok, 61-83
Radunceva, A. (1976): Vinica. Eneolitno seliste i nekropol. Razkopki i proucvania VI. Sofia
Rowley-Conwy, P. (1986) Between the painters and crop planters: aspects of the temperate European Mesolithic. In: M. Zvelebil (ed.): Hunters in Transition. Cambridge 1986: 17-32
Steadman, Sh. R. (2000) Spatial patterning and social complexity on prehistoric Anatolian Tell sites: models for mounds. JAA 19 (2000): 164-199
Schwarzberg, H. (2003) On problems in identifying ritual pottery: the example of the so-called "cult-tables". In: L. Nikolova (ed.): Early symbolic systems of communication in Southeast Europe. BAR S1139, Oxford: 79-84
Staccioli, R. A. (1968) Modelli di edifici etrusco-italici. I modelli votivi. Roma
Talalay, L. E. (1983) Neolithic Figurines of Southern Greece: their form and function. PhD dissertation, Indiana University 1983.
Talalay, L. E. (1987) Rethinking the function of clay figurine legs from Neolithic Greece. An argument by analogy. AJA 91 (1987): 161-169
Theokharis, D. (1973) The neolithic civilisation. A brief survey. In: D. Theokharis (ed.): Neolithic Greece. Athens 1973: 17-128.
Theokharis, D. R. (1981) Neolithikos politismos. Athina 1981
Todorova, H. (1979) Eneolit Bolgarii. Sofia
Toufexis, G. (1996) House models. In: Neolithic cultures in Greece (ed.: G. A. Papathanassopoulos) Athens, pp. 161-162
Trianti, I. (1984) Hausmodelle aus Masi. Mitteilunges des DAI, 113-119
Tringham, R. (1973) The Mesolithic of Southeastern Europe. In: St. K. Kozlowski (ed.): The Mesolithic in Europe. Warszawa 1973: 551-572
Trogmayer, O. (1966) A Körös-csoport lakóházáról. Újkokori házmodell-töredék Röszkérol (On the dwelling of the Körös group). ArchÉrt 93 (1966): 235-240
Ucko, P. (1968) Anthropomorphic Figurines of Predynastic Egypt and Neolithic Crete with comparative material from the Prehistoric Near East and Mainland Greece. London 1968
Ucko, P. J. (1996) Mother, are you there? In: Can we interpret figurines? Cambridge Archaeological Journal 6 (1996): 300-304
Vulpe, R. (1957) Izvoare. Sapaturile din 1936-48. Bucuresti
Wunn, I. (2001) Götter, Mütter, Ahnenkult. Religionsentwicklung in der Jungsteinzeit. Beiheft der Archäologischen Mitteilungen aus Nordwest-Deutschland 36. Rahden/Westf. 2001
Zvelebil, M. (1995a) Hunting, gathering or husbandry? Management of food resources by the late mesolithic communities of temperate Europe. In: D. V. Campana (ed.): Before Farming: a hunter-gatherer society and subsistence. MASCA research papers in science and archaeology, Supplement 12. Philadelphia 1995: 79-104

Abbreviations

AJA = American Journal of Archaeology (New York)
ArchÉrt = Archaeologiai Értesíto (Budapest)
BAR IntSer = British archaeological reports. International series, supplementary (Oxford)
BASPR = Bulletin of the American School of Prehistoric Research (Cambridge, Mass.)
DAF = Documents d'`archéologie Française (Paris)
EJA = Journal of European Archaeology (Aldershot)
FolArch = Folia Archaeologica (Budapest)
JAA = Journal of the Anthropological Society (Washington)
JEA = Journal of European Archaeology (later: EJA)(Aldershot)
JWP = Journal of World Prehistory (New York)
KSIAKiev = Kratkie Soobšceniâ Instituta Arheologii AN SSSR (Kiev)
Mat. si Cerc. Arh. = Materiale si Cercetari Arheologice. Bucuresti
MittArchInst = Mitteilungen des Archäologischen Instituts der Ungarischen Akademie der Wissenschaften (later: Antaeus) (Budapest)
Porocilo = Porocilo o Raziskovanju Paleolitika, Neolitika in Eneolitika v Sloveniji later: Documenta Praehistorica) (Ljubljana)
SCIV = Studii si Cercetari de Istorie Veche (Bucuresti)
UPA = Universitätsforschungen zur prähistorischen Archäologie (Bonn)
VAH = Varia archaeologica Hungarica (Budapest)
WA = World Archaeology (London)