Archaeology of religion--possibilities and prospects/Religiooniarheoloogia--voimalikkus ja voimalused.

Jonuks, Tonno

On uldteada nali, et kui arheoloog leiab eseme, mida on raske interpreteerida voi mille funktsioon ei ole uheselt teada, siis klassifitseeritakse see kultuslikuks. Nalja teine pool on aga pigem kurb--nimelt kaotatakse seejarel enamasti eseme vastu teaduslik huvi ning eset eksponeeritakse koikvoimalike publikatsioonide aukohal, umbritsedes seda "kultusliku" auraga, millest labipaasu ega isegi selle voimalust ei nahta ning interpretatsioonid lopevadki todemusega "kultuslikust esemest" vms.

Kaesolevas puuan vaadelda moningaid erinevaid lahenemisviise, kuidas ja millistel alustel oleks Eesti usundiliste esemete ja enti Eesti usundi enda uurimine arheoloogiliste meetoditega edasiviiv ning mis voimaldaks astuda jargmist sammu--sammu, mis "kultuslikest esemetest" viiks edasi oletusteni, millised olid uskumused ajal, mil neid esemeid kasutati. Kindlasti viivad nii monedki sellised mottekaigud spekulatsioonidele, kuid spekuleerida tasub usundi teemadel kindlasti. Ehk isegi rohkem kui teistes arheoloogia valdkondades. On ju ka spekulatsioon uks arvamuste vorme ning vaarate spekulatsioonide umberlukkamine voiks viia hoopis toenaolisemate jareldusteni. Vahemasti valistab nende umberlukkamine mingigi osa arvukatest voimalustest.

Kogu artikli arutlus puudutab ennekoike Eesti materjali, selle arengulugu ja uurimisvoimalusi. Usun, et koikeholmavate, universaalsete teooriate koostamisel on suurem oht sattuda kritiseeritud fenomenoloogide teele, kus teooriad on kehtivad vaid vaga uldises mastaabis, andes voimaluse analuusida ainult inimese uldist religioosset kaitumist. Loomulikult on sellised laiapohjalised arutluskaigud aluseks koikidele kitsamatele uurimustele. Kuid konkreetse piirkonna nagu Eesti usundi arengulugu on varga otseselt seotud selle piirkonna allikalise materjaliga ja teiste piirkondade pohjal koostatud teoreetilisi mottekaike on voimalik kasutada vaid vaga uldiselt.

Eesti muinasusundi vastu on uurijad huvi tundnud juba 18.-19. sajandi rahvusromantilistest liikumistest peale. Romantilistest kasitlustest ja antiikmaailma laenudest vurtsitatuna on sellised paganliku Eesti usundi jumalapanteonid praeguseks huljatud. Kull aga on mitmed teised selle ajastu romantilised kasitlused jatnud oma tugeva jalje nii rahvalikku kui ka akadeemilisse usundilukku.

Marksa teaduslikuma mootme sai usundiuurimine 20. scrjandi esimesel poolel, mil tootasid mitmed tanapaevalgi pohiautoreiks peetavad folkloristi taustaga uurijad nagu Matthias Johann Eisen, Oskar Loorits ja Uku Masing. M. J. Eiseni materjalikasitlusi kasutatakse usundiuurimises praegugi. Teaduslikus mottes vaartuslikumad on aga O. Looritsa allikapublikatsioonid. Kuid tema rahvuspsuhholoogiast ning romantilisest soomeugri urgdemokraatiast kantud teooriad on spekulatiivsed ja seetottu tanapaevases teaduses ettevaatusega kasitletavad. Kogu 20. sajandi I poole usundiuurijate toid labiva fenomenina voib jalgida teravat vastandamist saksa ja skandinaavia kultuuriruumile ning nende usundilisele sumbolile--kristlusele. Arvestades selleks ajaks valjakujunenud noort Eesti intelligentsi ning vastset Eesti Vabariiki, on sellised tendentsid ka moistetavad: on ju usund uks olulisemaid meie-tunde loojaid ning varskelt formeerunud rahvusriigi oluline ideoloogiline komponent.

Parast Teist maailmasoda jaidki senised usundikasitlused kestma, kuid usundi-uurimine ise haabus. Ilmus vaid uksikuid uurimusi, kus varasemad tendentsid olid endiselt jalgitavad. Eestlasi peeti jatkuvalt egalitaarse urgsoomeugriliku kultuuri kandjateks, kelle lahimad, nii keelelised, kultuurilised kui maailmavaatelised naabrid on Venemaa soome-ugri keeli konelevad rahvad. Kindlasti sobis selline kasitlus ka kaibel olnud poliitilise situatsiooniga. Probleemseka kasutati usundiuurimises aga endiselt rahvaparimuslikke allikaid kui pohiallikat, mille pohjal koostati susteem ning mida illustreeriti arheoloogilise materjaliga.

Uus ja tosine muutus saabus alles 1990. aastatega, mil avanes juurdepaas Laanes tehtud teoreetilistele kasitlustele ning mil selliseid kasitlusi hakati kasutama ka Eesti materjali interpreteerimisel. Selle perioodi uued sotsiaalsed teooriad kummutasid ka varasema usundikasitluse. Probleemseks on aga jaanud eesti arheoloogide suundumus pigem uhiskonna sotsiaalsete probleemide suunas, millega usundiuurimine on korvale jaanud.

Eesti muinasusundi kohta on labi erinevaie uurimisperioodide ja rohuasetuste kasutatud erinevaid termineid. Neist levinuim on Rahvausund, mida on kasutanud juba esimesed uurijad. Rahvausundi uurimisel on rohuasetus enamasti kull selle mittekristlikule osale pandud ning tihtipeale on puutud selles naha ka otsest jarglast muinasusundile, eeldades, et vcrhepealsetest ajaloolistest ja usundilistest muutustest hoolimata on pohiosa sailinud muinasaegsena. Mitmed kaasaegsed uurimused on aga naidanud, et 13. sajandil toimus maailmapildis tugev murrang ja kesk- ning uusaegset rahvausundit ei saa muinasusundi uurimise allikana kasutada, hoolimata seal leiduvast eelkristlikust parandist.

Akadeemilises kirjanduses on kasutatud ka terminit eelkristlik, mis on markeerinud just muinasaja lopusajandeid. Vast korrektseim termin, kasitledes muinasaegset usundit alates inimasustuse algusest Eestis ning lopetades 13. sajandi ristisojaga, voiks olla muinasusund, mis on uhelt poolt ajaliselt piiritletud ja kattub kogu muinasajaga, teisalt ei sea see ka liiga pretensioonikaid piiranguid.

Usundiuurimine on koikjal Euroopas olnud suhteliselt meetodivaene, kuid siiski ennekoike interdistsiplinaarne valdkond. Valdavaks metoodiliseks lahenemiseks on seni olnud fenomenoloogiline, mille pohjal on tehtud ka enamik Eesti materjali puudutavatest uurimustest. Viimasel aastakumnel on see meetod saanud aga mitmesuguse kriitika osaliseks: uhelt poolt just oma piiratuse, teisalt aga ebamaarasuse tottu.

Sellest lahtuvalt tahaks toonitada moningaid lahtepunkte, mis minu arvates on Eesti muinasusundit uurides olulised endale teadvustada.

Usund on dunaamiline ja pidevas muutumises, seetottu oma olemuselt ka tihedasti seotud sotsiaalsete struktuuridega, mille muutumisega, mida arheoloogiakirjandus viimastel aastatel jarjest rohkem rohutab, peab jarelikult muutuma ka usund. Seega ei ole voimalik raakida mingist abstraktsest "muistsest Eesti usundist", kuna igal konkreetsel perioodil on see usund olnud erinevaae nuanssidega.

Oluline on, et uuemad usundifenomenid sobitataks varasemale pohjale. Nagu eespool rohutatud, on usund pidevas arengus. Samas ei muusu koik usundit moodustavad fenomenid vordselt, vaid soltuvalt paljudest teguritest voib mone areng olla kiirem voi aeglasem. Selline ebauhtlane, kuid siiski pidev protsess nouab, et uued fenomenid, mida laenatakse, voi fenomenid, mis teevad labi mingi muutuse, sobiksid olemasolevasse susteemi. See aga tahendab, et ei ole voimalik laenata mingit fenomeni, mille pohialused oleksid uldkehtivast susteemist erinevad, ning samuti ei scra uks fenomen labi teha jarsku ja vaga pohjalikku muutust.

Rituaalsus on arheoloogilise perioodi usundiuurimise puhul votmetahtsusega. Jarjest enam on toonitatud, et arheoloogid ei kaeva valja muute, vaid jalgi rituaalidest. Arvestades neid jalgi, mille kaigus on maha jaanud esemed, ehitatud kalmestruktuurid, pandud kalmesse surnu ja tema panused ning surnukehadega voi kremeeritud jaanustega uhel voi teisel moel kaitutud, saame hakata tegema oletusi selle kohta, millised olid rituaalid, millest jaid maha sellised jaljed. Neid rituaale teades ja arvestades saame hakata omakorda pustitama oletusi, millised olid need usundilised arusaamad, mis neis rituaalides valjendusid. Seega ei saa kalmest voi ukskoik millisest muust kontekstist uht eset valja vottes jargmise sammuna otse pustitada oletusi usundi, mentaliteedi vms kohta. Ka rituaali ja eriti rituaaliteooriate kasutamine arheoloogias on keerukam, kui esmapilgul tundub. Nagu Liv Nilsson Stutz on valja toonud, on teiste distsipliinide, peamiselt antropoloogia poolt koostatud teooriate otse kasutamine arheoloogias ohtlik, ja nende kujunemiskaiku ja tausta mitte tundes voib sattuda ummikteele. Tema poolt valja pakutud tee oleks arheoloogide suurem suunatus rituaalile kui tegevusele, mitte kui ideele (thought) (Nilsson Stutz 2003, 51). Samas on rituaali taga siiski alati ka idee, uskumuslik taust, mida rituaal valjendab, ja rituaali kui tegevuse jalgede interpreteerimisel tuleb arvestada ka seda, et rituaal ja selle kaude hoomatav idee oleksid kooskolas.

Viimase aastakumne usunditeoreetikute uks olulisemaid kriitikapunkte fenomenoloogilise meetodi vastu puudutab moistete liiga vaba ja ebamaarast kasutamist. Selliste moistete defineerimine ning lahtimotestamine aitaks kindlasti kaasa, et lahendada kohati absurdseid olukordi, kus esivanemate kultust ning pohimotteliselt sarnast hauatagust elu on nahtud koikide perioodide puhul ja nii terminid kui kirjeldusviis on pohimotteliselt sarnased. Iseloomulik on ka, et arheoloogid on pigem valmis nagema piirkondlikke erinevusi ning vahem on tahelepanu pooratud ajas muutunud kujutelmadele (vt Jaanits et al. 1982, 99, 414). Ometigi on enamik uurijaid uhel arvamusel, et usund on muutunud ajas koos teiste nahtuste muutumisega ja seega ei saa ka hauataguse elu kontseptsioon pusida samasugusena pikka aega.

Kindlasti vajaksid iga autori poolt eraldi lahtimotestamist ka sellised arheoloogide meelisfenomenid nagu viljakuskultus, animism, totemism; laiemalt vottes ka nii kultus kui rituaal. On ju selge, et sarnase terminiga saab iseloomustada kullaltki erinevad nahtusi, mis erinevad oma sisus lahtuvalt kontekstist ja materjalist. Seega erinevad selliste terminite tahendused igas uurimuses.

Usundit peab vaatama kui tervikpilti ja alles seda arvestades saame hakata uksikuurimusi tegema. Tervikpilt on seotud ka lahtekohaga, mille jargi koik usundis uhel hetkel eksisteerivad fenomenid peavad olema omavahel seotud ning uksteisega sobima. Seega moodustub laiem raamistik, milles iga fenomen omavahel suhtleb ning uksteist taiendab. See lisab aga arheoloogilise materjali pohjal tehtavatele oletustele voimaluse, et on voimalik teha oletusi toenaoliste fenomenide ja nende iseloomu ule ka juhul, kui need ise voi materiaalsed jaljed neist ei ole sailinud. Tervikpildi arvestamine aitaks valistada ka ohtu sattuda klassikalisele fenomenoloogilisele rajale, kus, keskendudes liigselt uhele detailile (fenomenile) ning kaotades silmist uldpildi, naeme uht detaili usundist tugevasti voimendatuna, aga kui me ei pane seda laiemasse konteksti, ei suuda me siiski adekvaatselt jalgida selle detaili kujunemist ja seoseid teistega. Arvestades tervikpilti kogu usundi arengus laiemalt, on ka adekvaatsemaid uksikfenomeniuurimusi kergem koostada.

Loomulikult on ka selge, et tervikpilt muinasaegse usundi kohta on ja jaabki uurijatele vaid idealiseeritud eesmargiks, mille loplik saavutamine on voimatu. Samas peaks see olema aga siiski uurimuste laiem eesmark, isegi kui see viib monikord ebatoenaoliste spekulatsioonideni.

It has become a common joke that if an archaeologist discovers an object he or she finds hard to interpret or assign a precise function to, it is categorised as a cultic object. Behind the joke, however, lies the bitter fact that this often results in losing scientific interest in the object. Furthermore, the object is displayed on the covers of various publications, thus surrounded with a "cultic" aura, rendering it impenetrable and even ruling out the possibility that it may be understood; so, any interpretation leads to defining the object as a "cultic" or "ritual" object.

In the following I will attempt to analyse some approaches, and examine how and on which basis is the study of religious objects and prehistoric religion in Estonia productive and enables progress in the area of research--namely, progress from "cultic objects" to speculations about religious beliefs at the time the objects were used. Some lines of thought will, no doubt, lead to speculations, but speculations on religion will be certainly gratifying. Perhaps even more than in other areas of archaeology. Speculation is, after all, a form of argumentation and refuting wrong speculations may lead to more promising conclusions and will rule out at least some of the numerous possibilities.

The entire discussion that follows will deal with the Estonian material, the history of studying prehistoric religion and research possibilities. I believe that in building comprehensive universal theories there is more risk to cross paths with phenomenologists, where theories are applicable only on a very general scale and enable to analyse only the general human religious behaviour. Similar extensive lines of argumentation will naturally form the foundation for narrower studies. But the history of religion in a specific region, like Estonia, is directly linked to the source material of the region, and theoretical argumentations based on the material of other regions can be applied only on a very general scale.

On the history of study

Since the systematic study of religion emerged already in the 19th century, simultaneously with the awakening of National Romanticism, literature and scholars on the topic abound. True, the folk religion which remains outside mainstream Christianity and therefore has been mostly considered a superstition, has attracted constant scholarly interest since the mediaeval period and particularly by the clergy, who considered "recognising and rooting out the Satan" imperative. Analogous pieces of writing, which have been mostly affected by the classical antiquities and Romanticist approaches, have proved effective in studying the 18th-19th centuries mentality, but provide a rather subjective view of prehistoric religious conceptions.

In relation to the all-European national Romanticist movement in the 19th century, the focus of interest shifted from the contemporary superstition characteristic of the mediaeval period to the pure and innocent nature religion, untouched by the influences of Christianity, of the ancient heroicised period of independence. This was, in various aspects, a remarkable period and has exerted its indirect influence on conceptualising religion until today. This period saw the compiling of the national epic, which the non-academic audience still regard as authentic folklore, which has been orally transmitted from one generation to another from prehistoric times. Also, folklore collection got a head start around the same period with one of its main foci on belief reports. At the same time, in 1881, the Chronicle of Henry of Livonia was translated into Estonian (Tarvel 1982, 14) and some time prior to that the connection between the Ebavere hill and the birth place of the ancient god Tharapita, the only pagan god we know, first mentioned in the Chronicle of Henry, was established (Knupffer 1836).

Thus, in the 19th century the foundation was laid for many conceptions that were quite recently still acceptable. Considering this tendency against the European tradition it seems a relatively natural one. After all, the majority of the first museums were founded and the first collections of prehistoric findings were compiled around the same time, and people were actively involved in search of their roots. Also, literacy began to spread more widely, folklore collections were established in different parts of Europe, and the humanities became the focus of scholarly interest.

While in the relatively stable European countries the study of religion soon became a matter of academic study, in Estonia it largely remained a political tool, oscillating within a wide range. To counterbalance the 19th century theory of Goths in the Eastern Baltic and the cultural invasion of Germanic tribes (see further in Tvauri 2003), which was definitely evident in at least some authors' views on Estonia and its inhabitants, the study of the eastern kinsfolk of the Estonians was initiated. The introduction of the language tree and the theory arguing that the distant predecessors of the Estonians arrived from the area near the Urals certainly played a role in this. This national approach promoting freedom from German cultural influences was, no doubt, more fitting for the historical consciousness of the nation in the period of national awakening. Through this, religion as the central concept in public mentality was adopted as an ideological tool for the young Estonian intelligentsia and later also for the Republic of Estonia, and was used to emphasise the uniqueness of the Estonians, and the role of ethnic culture and its various phenomena even among other cultures (see e.g. Masing 1939). The ancient Estonian folk religion, reconstructed in Romantic form was manifest on various levels, assuming a more concrete form among the followers of Taara faith, built on similar Romantic notions (see Deemant 1988). The Romantic approach of the ancient religion also had an impact on folklore, which earlier authors have treated as the main source of folk and prehistoric religion. One of the most illustrative examples of the intersection of the study of history, historical consciousness and folklore is perhaps the following story, which was recorded in 1930:

The Sacred Stone of Kunda village

The sacrificial stone is located in the orchard of the Parijogi farm, village of Kunda, where allegedly there used to be a sacred grove of ancient Estonians. This is what people say about the stone. In the old days, when there was still a lake in Kunda, the Estonians lived in pole huts built on the lakeshore and in the lake, catching fish and hunting in the woods. A sacred grove, which was situated on an elevated site at the lakeshore, was their sanctuary. There was a sacred stone in the middle of the grove, where people brought offerings to seek protection against wars, illnesses and other ailments, as people prayed there solemnly. When the Germans and Danish reached the northern regions of Estonia with their Christianisation, the Danish cut down the grove and forced the locals to attend sermons in the churches built by the Danish. After the grove trees were cut down, people still secretly prayed at the sacrificial stone, but soon the rulers of the land forbid it and people were Christianised. The stone had cup-marks, symbolising the dead, because when somebody died, his or her close relatives had an obligation to carve a cup-mark into the stone. The marks are clearly visible even now. As is the fire pit in the centre of the stone. The stone is slightly idented from weathering, which somewhat ruins its appearance (ERA 11221, 340/3 (24)).

Here we can notice several features characteristic of the 19th century, which largely originate in the National Romantic treatments. In addition to the mythological perception of time, where discrepancies in the course of time and the course of events are of no major consequence (the Lammasmagi settlement in Kunda is dated to 8,700-4,950 BC and Denmark's crusade in northern Estonia was launched only in 1219 AD), the account emphasises that each grove must have a sacrificial stone (in reality, however, the distance between the stone described above and the Kunda Hiiemagi (Kunda Grove) is some kilometres in a bird's-eye view) and also the fact that fire is made on the sacrificial stone, and that it has to look nice. The two latter aspects of the story have probably been influenced by classical mythology and religions in the classical antiquities, where sacrificial fire was made on altar rocks (Fig. 1).

Yet, the late 19th century and early 20th century treatments of religious history cannot be altogether ruled out from the viewpoint of scholarly research. Most important aspects here are terminology and exposition of problems. What is the theme of research? What are the sources? How reliable are the used sources in providing answers? Largely from the context of the 19th century traditions emerged Matthias Johann Eisen, self-learner and the first who started systematically collecting belief reports. In the 1920s another scholar, Oskar Loorits further developed the research. Loorits had an academic education and he brought the study to the academic level. Both Eisen and Loorits, who identified themselves as folklorists (Loorits 1998, XIV), did not set any clearly formulated problems in their works, causing misinterpretations, which led to misconceptions about the topic of their works. Relying on recently recorded folkloric belief reports as main sources, both authors wrote about the Estonian folk belief, whereas their work is first and foremost referential, and as such highly noteworthy. Unfortunately, Eisen's works remain only overview of sources. Even though Loorits considered the presentation of sources important, "aspiring exhaustive comprehensiveness" in some areas (Loorits 1998, XVI), he also emphasised the importance of analysis, and used extensive linguistic material in addition to the folkloric. Loorits himself did not elaborate on his definition of the concept 'folk belief', but the context suggests that differentiating between the folk belief of the pre-Christian period and that of the Christian period has proved no problem for him (or for other contemporary scholars). Differentiation between the Christian religion and pagan beliefs, however, has been crucial. This exposition of problems, relatively vague in terms of sources, has led to the situation where assumptions have been made across hundreds and thousands years on the basis of folklore at Loorits' disposal, recorded only a few decades before, the folklore collected by him personally and the very early stages of the study of Finno-Ugric linguistics. On top of that, Loorits presented a relatively chaotic view of the material, for example, narrating on Christian and non-Christian cultural phenomena in a single story (see e.g. Loorits 1998, 11, 14). Thus it seems most appropriate to denote the research topic of Loorits, Eisen and other authors of the period with the same term they have used--namely, folk belief. But before moving on to interpreting, the terms need to be defined, and I will return to that below.

[FIGURE 1 OMITTED]

The study of religion in Estonia underwent dramatic changes in the period following World War II. Oskar Loorits, former leader in the field, went into exile, where he published his voluminous thesis about Estonian prehistoric religion, having previously fallen into serious conflict with his colleagues (see Moora & Annist 2002). Among other collections of essays that Loorits published in exile, he also issued the book Eestluse eluj6ud ('Viability of Estonian Identity'; Loorits 1951), founded on Romantic notions, where folk belief played an important role, but the treatment of history had been discarded as unreliable a while ago, which is why the conclusions presented there have no more consequence than as descriptive of Loorits' personal disposition. There is no doubt, however, that Loorits' main work "Grundzuge des estnischen Volksglaubens " volumes 1-3, volume 3 (1957) in particular, has left an impressive mark in the Estonian folk belief historiography. This was, after all, a precedent in attempting to view folk belief as a complex system, which evolves in time. Considerable attention has been paid to the assumption of soul and power. Most of Loorits' conclusions, however, are considered unreliable and seen as subjective speculations, where the key words are "ethnic psychology", an idealised conception of Estonians as primitive Finno-Ugric tribes, where the prevailing social order is "a primitive democratic equality and parity in rights" (Loorits 1990, 78), etc. (for the critique of Loorits' work see e.g. Moora & Annist 2002, 247-263).

One of the few Estonian-born scholars of folk belief, who received the most progressive education of his time, was Ivar Paulson, who worked in Sweden and was a disciple of Ernst Arbman, professor of religious history at the University of Uppsala. Paulson's Ph.D. thesis Die primitiven Seelenvorstellungen der nordeurasischen Volker (1958) focuses on conceptions of soul in northern Eurasia. During the last years of his academic career Paulson turned his attention specifically towards Estonian folk belief. As was considered proper in this period, Paulson concentrated on issues surrounding the origin of religion; he also considered religion a constantly evolving phenomenon, emphasising the clear and relevant distinction between hunter-fisher-gatherers and farmer-herders.

Under the Soviet regime, the study of religion was somewhat more complicated in Estonia. Since the Department of Theology at the University of Tartu had been closed down, and the only institution providing education in religious matters, the Institute of Theology, mainly focused on training Lutheran ministers, no systematic theological education could be pursued in Estonia. However, since the study of religion could not be avoided, some studies were published. The mitigating factor here might have been the east-oriented conceptualisation of ancient Estonian religion, which circulated already in the pre-war period. Aliise Moora's article on the ancient religion of the Estonians, Eestlaste muistsest usundist (Moora 1956) was a follow-up to the pre-war tradition, where the main source of the history of religion was folklore. Moreover, she began to introduce archaeological findings, though fitting these into a folklore-based system and using archaeological material only for illustrative purposes. The very first study into prehistoric religion by an Estonian scholar, which was mostly based on archaeological material and which clearly formulates the research topic as prehistoric folk belief, is an article by Lembit Jaanits--Jooni kiviaja uskumustest ('Characteristics of Beliefs in the Stone Age'; Jaanits 1961). Unfortunately, this remained the only study on the topic by archaeologists for a long period of time. In this article Jaanits studies Neolithic findings, which include the largest number of figurative pendants--i.e. objects easiest to interpret from the aspect of religion. The treatment of religion in Eesti esiajalugu ('Prehistory of Estonia', Jaanits et al. 1982), where religion is discussed only in relation with (seemingly) easily interpretable objects, is analogous in that sense. Among other archaeologists, next to Jaanits, the issue of religion has been perhaps most comprehensively studied by Vello Lougas, whose central topic of research was sun worship and its manifestations in stone-cist burials and the Kaali meteorite crater (Lougas 1996). Lougas also published some minor studies on the history of Estonian ancient religion (Lougas 1972). In terms of more recent studies, I cannot overlook Juri Selirand's research on Late Iron Age mortuary traditions (Selirand 1974). The focus of his research is on object analysis and description of burial types and less on assumptions on religion, which is understandable given the lack of social studies at the time.

The 1990s saw a new beginning in research, when theoretical studies into archaeology as well as religious history, conducted in the meantime in western countries, became available for Estonian scholars. Still, Estonian archaeologists have mostly specialised in the social sphere and have published few studies on religion. The most consequential of these are Tarmo Kulmar's Ph.D. thesis on soul phenomenology of prehistoric Estonian religion (Kulmar 1994) and a series of articles under the same title (Kulmar 1992), which represent a novel viewpoint in the study of prehistoric religion. Relying mainly on archaeological studies and the works of (mostly German) theoreticians of religious history, Kulmar compiled quite an intricate system of Stone Age soul phenomena, at the same time demonstrating their interrelations and evolvement in time. Among the thesis subtopics were soul conceptualisation, as well as fear for the dead and beliefs about the living dead.

Other archaeologists have studied prehistoric religion, but to a far lesser extent. Here I must mention Valter Lang's attempt to conceptualise cultural landscape, which is rendered meaningful through religion (Lang 1999a). Likewise, Andres Tvauri in his study of cup-marked stones has introduced the religious principle, although linking it only to fertility cult (Tvauri 1997). Other archaeologists have touched upon the topic, but the main focus of their studies lies in the social aspect, and religion is used only for the purpose of interpreting social behaviour. An important subject in studying religion among Estonian archaeologists has been the Christianizing of the country (Magi 2002; Valk 2001; 2003).

I have consciously excluded the studies of folklorists and ethnographers of the second half of the 20th century from the above brief historiographical overview, mainly because these studies, especially the more recent ones, largely deal with folk belief of the modern period, i.e. the living present-day folk belief.

Sources reveal that the majority of conclusions in research history so far rely on folkloric heritage, linguistic etymologies and dating, and to a great extent on anthropological parallels. The latter applies mostly to Finno-Ugric tribes in Russia and Siberia. Archaeological material has been used first and foremost for illustrative purposes, sometimes even referring to it as "silent findings from earth", and their importance has started to grow only after the 1990s. Around this time there was a breakthrough in the general treatment of folk belief and henceforth archaeological material has been preferred over other sources.

Terminology

As the above overview suggested, several misconceptions have been occasioned by the confusion in terminology, which stresses the need to define, both phenomenologically and chronologically, what is being studied. The degree of precision in defining folk belief and its various phenomena is, of course, an altogether different question. Religion with its different manifestations often appearing in other fields is difficult to delineate or define in detail. In religious research it even seems practical to avoid establishing too strict boundaries or construe generally applicable detailed models, which will later start inhibiting research. After all, living religions, except for the canonical major world religions, are not committed to defining concepts, and the different concepts and phenomena interrelate and interact with each other rather than are subject to differentiation or definition.

For further discussion, however, at least some terms need to be explicated. Various terms have been used for marking Estonian folk beliefs at different periods. The earliest and perhaps the most common of these is rahvausk/rahvausund, or folk religion/belief. The terms were adopted by the very first scholars, Eisen and Loorits, to distinguish between the belief of native Estonians and the official Christianity. Unfortunately, none of the earlier scholars have attempted to explicate the terms, and it appears that their use of the term served the purpose of distinguishing between the Christian and non-Christian material, which they referred to as folk religion. This is how the term has also been interpreted in academic treatments (Viires 1986; Valk 1998). The most recent definition of the term folk belief has been proposed by Aado Lintrop, who defines it as "a popular interpretation of opinions and concepts of the dominant religion established in scriptures and comments to it on the basis of (in the Estonian tradition also preChristian) religious convictions at various times" (Lintrop 2003, 9). A different question is whether it is even possible to establish the system of folk religion upon a religion introduced later (in this case, Christianity). After all, it is generally known that folk religion includes many non-Christian elements, which cannot be regarded as interpretations of the scripture or its comments.

Even though the emphasis in the study of folk religion is laid on its nonChristian part, it is still a set of beliefs where Christian elements intersect with pagan ones. Depending on the sphere the proportion of Christian and non-Christian elements varies and I doubt that it is possible to find a single criterion, which would enable to determine the religious affiliation of a sphere or a phenomenon. This, in turn, will make the definition of the elements more difficult as each of these needs to be approached individually.

Another term used alternately with folk religion, is paganlus, or paganism. This term, however, has a strong qualitative nuance, which renders its use in academic writing somewhat problematic.

Another alternately employed term, which is considerably more specific than the temporally vague folk religion, is eelkristlik usund, or pre-Christian religion (Valk 2001). The term itself as well as its context of use clearly point to what it means--the term is most appropriate to mark the religion followed in the final centuries of the prehistoric period, or the period prior to the Christianisation of the country, in 13th century. On the other hand, the term cannot be used to mark far too distant periods, as, by doing that, it distances itself from its meaning - its opposition to Christianity, the official religion.

While generally discussing the religion on the Estonian territory since the beginning of human settlement up to the official Christianisation of the country, and hence the adoption of the term folk religion, the most appropriate term would be muinasusund, or prehistoric religion. The term has become increasingly used by archaeologists and religious historians (Kulmar 1992; 1994) and ethnographers (Viires 2001). Overlapping the concept of prehistory, the term prehistoric religion signifies a period which is not very narrowly defined, but still within certain limits. Compared to pre-Christian religion, prehistoric religion is a more neutral term and does not give preference to any other religion.

In addition to chronological terminology misinterpretation has been generated by various other, mainly religious concepts, which are employed relatively loosely and without further explication: for example, scholars often neglect defining terms like totemism, shamanism, ancestral cult, etc. This issue will be addressed below.

Sources

In the following I will primarily discuss sources connected with the Estonian prehistoric religion. As to the origin, the sources may be divided in three major groups: folkloric, written and archaeological. Certainly, the sources of religion are not limited solely to those that will be discussed below, but these have been most common in the Estonian tradition and therefore deserve greater attention. In addition linguistic sources have been used, but mostly by earlier scholars, Loorits, Masing and Paulson, but not so much in recent studies. In the following, in any case, we cannot overlook anthropological sources and those of other disciplines that have so far been used less systematically.

Folkloric sources mainly consist of folk tales and folk songs recorded in the late 19th and during the 20th centuries. Since these sources contain a great deal of religious material, this type has often been considered primary in the study of the Estonian prehistoric religion and the basis of the conclusions dates back to the 19th century stretching even further back to the Stone Age (Loorits 1932; Moora 1956). Authors presenting such conclusions usually tend to overlook the temporal space distancing the 19th 20th centuries from the prehistoric era, as well as various other changes in the religious context of Estonia (further on this see Valk 1998, 81-86). Also, they often fail to consider the history of folkloric interpretation, which is still largely influenced by the context of national awakening. Around this time the social need for free ancestors and the heroic past arose, and folkloric material was used for studying prehistoric period, on which relatively scanty information was available (further on this see Honko 1998). Reformation has also played an important role in the formation and development of folklore. In the period following the triumph of Reformation and Lutheranism, motifs rooted in Catholicism interrelated with conceptions of the pre-Christian religion and the Catholic elements became a part of the so-called paganism or folk religion. It has been argued that one reason why Catholic beliefs were retained in non-Christian folk religion was the general political situation of the time (Valk 1998, 76).

The study of one particular folklore genre, namely runo songs, might prove most effective. After all, runo songs, because of their stable and strict form, have often been considered thousands of years old (Kunnap 2001, 14). Thus, the various motifs of the pre-Christian period might be retained particularly in runo songs. Unfortunately no uniform method has been worked out to determine these, and opinions on this topic are widely varied even about a single motive (cf. Valk, U. 2000; Lintrop 2001).

While discussing the use of folklore in studying prehistoric religion and world view, we cannot overlook the 18th-19th century Moravian Brotherhood and the heaven-goers' movements, which played a critical role in introducing Christian motifs among the wider general public. After all, the Moravian brethren and the heaven-goers were the ones who, by emphasising the personal experience in perceiving god and reading the Bible, managed to do away with folk religion based on non-Christian principles and convert the majority of the population into Christianity (Plaat 2001, 32-60).

Folkloric material can thus hardly be the main type of source in the study of prehistoric religion. Among the reasons is the nature of the material, as well as problems in associating certain motifs with the pre-Christian religion (see Lang 1999b, 172). Obviously, motifs of prehistoric beliefs have survived in folklore, but their recognition and further and more precise dating solely on the basis of folklore is hardly possible. In addition, as Lauri Honko has pointed out, the seemingly original starting point may prove to be an end result, or a result of some complex process (Honko 1998), which will render the analysis of folkloric material all the more complicated. And moreover, according to modern archaeological approach it is advisable to avoid the use of folkloric and written sources as primary in prehistoric religion research and archaeological data should be preferred in studying earlier periods. Folklore, however, should not be discarded altogether, since it does contain material from the pre-Christian period, although caution should be applied when constructing a religious system on the basis of random material (under the religious system I mean a system formed of different phenomena and their interplay, but which cannot be called a religion, since it does not include all the phenomena of a religion, but only a selection, either based on source materials or a scholar's preference).

Under written sources I have grouped contemporary chronicles, sagas, and other sources. In the context of contemporary prehistoric religion we might distinguish between two types of sources. The former cover the period up to the mid-13th century, and describe belief reports, which are considered "living" and are applicable to and practised by the majority of a society. The most classical and important chronicle here is, no doubt, the Chronicle of Henry of Livonia. It is hardly the only one, since allusions to (prehistoric) religion are also present in other Livonian chronicles (see Tamm 2001) as well as in several major European chronicles like the Chronicle of Adam of Bremen.

The latter type of source is mediaeval, which described non-Christian beliefs, but which are concerned with a religion that is no longer dominant and which practices are followed secretly from the chronicler (and other members of the clergy), or with a religion that is only known from second-hand sources (Bartholomaeus Anglicus). This type includes the chronicles of Balthasar Russow, Johann Renner, and other mediaeval Livonian chronicles.

Interpretation of belief reports of the chronicles is far more complicated than the way it has been used to date. The aim of chronicles has not been to describe a certain sphere of life, but was far more specific. Accounts of earlier chronicles usually present descriptions by a Christian author for readers of Christian countries about a foreign, and therefore dangerous or at least strange country. On the other hand, the purpose of belief reports in mediaeval chronicles has been mainly presentation and introduction of the heresies of the local rural population and to point out the need of Christianization of the country. These chronicles therefore describe what the chronicler has seen as deviant and what he has considered worth recording. This, in turn, renders the use of most of the chronicles in the study of prehistoric religion relatively problematic. Also, the chroniclers may have misunderstood some customs, or have included phenomena that they have considered complementary to the chronicle, but which they themselves had never witnessed, or which perhaps did not even exist. All in all, chronicles are unavoidable in the study of prehistoric religion, but should be approached with certain caution and considering the risks of interpreting chronicle accounts. Misinterpretation is most often caused by different points of emphasis--for example, while discussing the early 13th century burials, mediaeval chronicles only mention cremation, whereas archaeological material indicates that by the 13th century, inhumation had already acquired an important position. As cremation was more characteristic pagan religion and therefore of more consequence for the chronicler, a greater emphasis was laid on this type of burial.

Another versatile feature of pagan religions in the accounts of mediaeval chronicles is their authors' wish to embed biblical quotations of miracle tales and magic stories into their accounts (Tarvel 1982). In addition, loans from authors of the antiquities and stereotypic stories have widely been used in mediaeval chronicles and not recognising these may result in serious misunderstandings (see Metssalu 2004, 51).

The material and interpretation of belief reports in post-mediaeval chronicles relies on completely different sources. Even though the chroniclers are also members of the clergy, enough time has passed from the official Christianisation of the country, so that elements of Christianity have started to influence the folk religion described in these. But here, too, the problem in interpreting the chronicles is that instead of the entire information on the religion, they only tend to include incidents that the clergy has considered disturbing. Thus, most of the chronicle reports inform of mortuary traditions and some more noteworthy sacrificial rituals, overlooking the more common and ordinary religious practices. Also, the dating of the sources and the material included, and new Christian influences have proved problematic for later chroniclers (see further in Valk 1998, 75-77).

As for the 18th-19th centuries chronicles, it is important to consider the context of time, as the description of reports was then influenced by the National Romantic approach. The influence of such National Romantic visions is very probably present also in modern religious perception. And this is exactly where the 20th century religious history stems from. Last but not least, I would like to point out the modern chroniclers' disposition to antique mythology and European prehistory, as various phenomena of the Estonian prehistoric religion have been borrowed from these sources. Thus modern chronicles and the belief reports included in them deserve further and wider-scale analysis.

The most significant archaeological sources are definitely graves. Since graves are usually the most important and often the only ancient relics, they have been thoroughly studied. Burials have also been used to reconstruct everyday life, or life outside the sacral sphere, which is the primary purpose burials serve.

The large number of graves compared to other types of relics in general, and more specifically on the Estonian territory, and also the active excavation of burials, are the main reasons why they can be considered the most examined type of prehistoric relics. Throughout times burials have been interpreted in different ways by different authors; the most common interpretation is perhaps their being burying sites, a view consistently held by Estonian archaeologists up to the 1990s ([TEXT NOT REPRODUCIBLE IN ASCII.] 1955; Selirand 1974; Jaanits et al. 1982). The spread of and access to the theoretical studies by West-European archaeologists has brought along a shift in interpreting burials from the social aspect also among the Estonian archaeologists (Ligi 1995; Lang 1996; 2000; Magi 2002); this view is mainly built on Ian Hodder's theoretical approach to graves as manifestation sites of the social elite (Hodder 1991). In addition to social interpretations, other approaches which emphasise the importance of rituals over social manifestations (Lang 1999a; Konsa 2003) and the need to consider the religious context while interpreting burial material, have been recently introduced in the Estonian archaeology.

A novel topic in research of religion is also the study of landscape (Lang 1999a; Vedru 2002), which has become particularly popular in the neighbouring countries. According to many authors, folk religion is a worldview, through which people discover and conceptualise their place in the world. This view connects religion with the physical world. Therefore, the nature of religion largely depends on the symbolic values that a particular group of people has attributed to the surrounding environment. In other words--environment is one of the factors that shape our worldview and religion (Meyer-Dietrich 1999, 165). Environment and landscape appear to be the main reasons why most burials and other religious structures are located on hillocks or hills. Michael J. Moore, for example, has argued that in Great Britain for someone participating in a ritual both were important--monuments around the ritual place but also monuments, which were visible from the ritual place (Moore 1995, 234). Every single object created a so-called religious space around it, but every one had an important role in relation to others.

While discussing religious sources we definitely cannot overlook find material, of which the most distinctive are pendants. Symbolic value, no doubt, may be attributed to any object regardless of the context of its discovery and its relative value for its owner. Such objects are, for instance, weapons, personal items (knives, combs, etc.) and jewellery. Certainly, symbolic value can also be attributed to tools and pottery. At the same time, the symbolic value of such objects is often secondary, being of consequence in a particular context, but is less expressive of general religious beliefs. This is why pendants are so important--next to the decorative function they strongly reflect religious conceptions, which have determined their shape. Among the largely geometrical shapes, some figural pendants stand out, inspiring interpretations with their different shapes. There are certain risks behind these seemingly easily interpretable objects, and I will address this issue below.

Attempts have been made to link various other archaeological monuments to religion. In Scandinavia some types of strongholds or buildings have been interpreted as cultic buildings, the same has been speculated about some temples in the settlements in the Baltic and Slavonic area. Unfortunately, Estonia so far lacks comparative archaeological material. This void could be filled with systematic archaeological study of sacred groves and sacrificial sites that have so far been neglected in research. Even though the cultural layer of the area is non-existent and object findings have been scanty, the application of natural-scientific methods in addition to the traditional archaeological methods may prove effective in the study of sacred groves.

Methods

It has been emphasised that compared to other fields of research, the studies into religious history have paid little attention to the methodological aspect (see e.g. Ahlberg 1999, 9). There are several reasons for that, but the most important of these is that religious history is an interdisciplinary field of study, and the application of a uniform method on widely different disciplines has proved relatively difficult.

Like elsewhere, the prehistoric religion of Estonia has so far been studied by the means of the phenomenological method (Loorits 1932; 1959; Jaanits 1961; Selirand 1974; Paulson 1997; Masing 1995; Kulmar 1994; Viires 2001, 198-214). By applying this method the focus of the studies is religious phenomena and their versatility, considering also their development. According to G. van der Leeuw, one of the founders of the phenomenological method, a religious phenomenon is something which appears or which exists and which the phenomenological method attempts to describe and systematically study (Leeuw 1986, 671; further on the phenomenological method see Hedin 1997). However, the phenomenological approach fails to observe the development of religion as a system of phenomena in general. Also, individual approach to single phenomena will not be able to provide a homogeneous view of religion.

The phenomenological approach is used not only by Estonian scholars, but has been characteristic of the European religious historical discourse in general and especially until the second half of the 20th century (Vries 1970; Dumezil 2001; Leeuw 1986). Although attempting to provide homogeneous views of religion, these studies have been structured according to phenomena, and the coexistence and interaction of different phenomena are difficult to follow.

In recent years the phenomenological method has prompted increasing criticism. Dag Hedin, who represents the critique of the traditional phenomenological method, argues that phenomenological method is justified only in examining single issues (Hedin 1997, 122). Hedin also suggests that phenomenology should concentrate not so much on compiling ideal typologies of particular phenomena (sacrifice, prayers, etc.), but should attempt to understand the real religious conception through dialogue (Hedin 1997, 128). The theory of Hedin, who is a historian of religion, relies on materials of traditional history of religion and is based on texts and hermeneutic methods, which intercept with interdisciplinary discourse and construct the context necessary for interpretation. Unfortunately, these methods cannot be applied to archaeological "texts", constituting the basis of voliqious historical research. Since Hedin takes as his source the "living" culture and religion, he also emphasises the dialogue between a scholar and transmitter of culture (Hedin 1997, 128), another aspect that cannot be applied in studying the archaeological past. Jeppe Sinding Jensen, a Danish religious historian, agrees with Hedin and suggests that instead of the current phenomenological approach reliqion should be studied and described narratively (Jensen 2003). Jensen also points to the need for defining, emphasising that one of the main problems of the phenomenological method is creating confusion by speaking about some phenomena which exist on the theoretical level, but which are not present in actual religious practice. This is a far-flung problem, which I referred to at the beginning of this article, as many scholars use terms without explicating them and often the meanings seem to be lost to the scholar himself. One way to solve the problem is perhaps paying more attention to terminology and the explication of general terms. This would also preclude the situation where a vague definition is used in describing different phenomena by different scholars.

Some recurring phenomena in prehistoric religion

Recently, numerous studies have been published on the relationship of rituals and archaeology (e.g. Bruck 1999; Nilsson Stutz 2003). Also, discussions abound on the purpose and function of rituals. Perhaps the most concise definition here is that ritual integrated an individual into a group, strengthened the solidarity and sense of identity within the group, simultaneously determining social boundaries (Sundgvist 2003, 32).

Many anthropologists have studied ritual and its role in a living society (see Ahlbdck 1993; 2003), focussing on the psychological aspect of ritual on members of the society. The most popular trends in archaeological research in recent time have been separating ritual and myth, claiming that archaeologists dig up traces of rituals, not of myths. From this follows a line of argumentation that by recognising and understanding traces of rituals we will be able to understand them, and only then will we be able to understand myth, i.e. the ideological context that triggered the ritual.

Rituals occur on two different, but closely interrelated levels: (i) primary level, which is oriented to the ritual object, or a god, a deceased ancestor, etc., and (ii) secondary level, which is oriented to the living, and through which the community strengthens its unity and social strategies. Ritual is used to pass on important messages to the community and in addition to strengthening in-group relationships, it also strengthens ties with ancestors and deities, creating thus a homogeneous community (Boyer 2001, 232). The latter view, the role of ritual in a social system, is generally acknowledged in religious historical and archaeological discourse (Sundgvist 2002; Kaliff 1997; Lang 1999a), while the former, the primary level oriented to gods, the dead, etc. has often been overlooked. From the viewpoint of studying religion, however, this level is of greater significance, as it enables to understand religious concepts through rituals.

While analysing ritual in the Estonian archaeological context, I would first and foremost like to discuss stone graves--relics that provide more information than any others among prehistoric rituals. Assumptions have been made about various regions in Europe that places of cultic worship in settlements (Turcan 2001), buildings erected in settlements, and special structures constructed for ritual purposes elsewhere (Parker Pearson 1999) enable to interpret ritual behaviour. In Estonia and the surrounding area, corresponding analyses have not been conducted. It is possible, though, that this is only a matter of research and the corresponding hypotheses will be formulated. One example could be the supposable cultic site beside the early tarand-grave in T6nija, in Saaremaa (Magi 2001).

Stone graves with constructions are cultic places rather than burial sites. The generally acknowledged interpretation until the 1990s was that the primary function of stone graves was burial sites. In recent years, however, the function of stone graves has been revaluated, mostly owing to the spread of the views of theoretical archaeology in Estonia. Although in the 1990s burials were largely interpreted from the social aspect, it is associated with the idea of a burial as a ritual place. Studies published thus far have interpreted burials as manifestation sites of the society's elite, where the latter performed certain rituals to display their position (Ligi 1995; Lang 1996; 2000; Magi 2002). Related to it is the interpretation of stone grave as a symbol of land ownership (Ligi 1995; Lang 1996, 492) or a landmark (Tuovinen 2002). The treatment of ritual stone constructions as symbols of land ownership is quite popular and widely recognised in modern archaeological and anthropological research (Wallin 1993, 115; Kaliff 1997; 1998; Widholm 1998; Tuovinen 2002). In the archaeological theories spread in Scandinavia, the role of religious rituals in structural stone graves has been pointed out (Kaliff 1997; Widholm 1998; Victor 2002). It is true that religious rituals interrelate closely with social rituals which society's elite exploited to secure their status, but for the purposes of the present article I will attempt to accentuate religious rituals over social ones. Furthermore, societal factors that are manifest in mortuary rituals, as well as the burial and the religious ritual itself are mainly influenced by belief systems (see David & Kramer 2001, 379). Several theoretical studies into archaeology and religious history express the view that a ritual (and a sacred) place is where humans encounter supernatural forces, where primordial myths are reconstructed and through that the relationship of humans and supernatural forces, and indirectly also the relationship of societal forces, are established (Eliade 1958), which is manifest mostly in how a part of a society gains access to objects required for sacrifice (Wallin 1993, 129). Many archaeologists also proceed from the view that the symbolism of ritual communication reflects power relationships in the society, even though this view has prompted criticism, and ethnographic parallels have been drawn to prove that this reflects how it should be in an ideal situation (Parker Pearson 1982, 112), sharing similarities with the ideal culture and myth conceptualisation formulated by Lauri Honko (see Honko 1998). Therefore, no uniform interpretation claiming that burials with lavish grave objects were used by a "wealthy" family can be provided on the basis of grave material.

While analysing stone constructions used at rituals it is important to consider their chronology and temporal overlapping. In Estonia, both Vello L6ugas and Valter Lang have noted that stone graves were still in use long after burying had ended (Lang 2000, 104). Maintaining a grave construction for 1,000 years with no burying in the meantime (Lang 2000, 104) indicates that the site was not merely a burial place, but an object of broader ideological significance for the society.

Unlike several views discussed above, I hardly think that attributing excessive symbolism to burials and other similar relics (Kulmar 1999, 163; Lang 1999a; cf. Kaliff 1997; 1998) is rational, and have regarded graves primarily as ritual structures. Certainly, the symbolic meaning of graves cannot be entirely ruled out, but these are still first and foremost ritual constructions, which symbolic context is revealed only in relation of its rituals, the ritual function of the grave and the general religious context.

Having a clear understanding of burial and its ritual significance, we can proceed to speculate about symbolism underlying graves and related objects and structures. The objects and structures, no doubt, are of greater consequence in connection with the grave than taken separately (Renfrew 1996). Of various objects used in the study of prehistoric religion, pendants are most common by associated with religion. The interpretation of pendants, on the other hand, is often limited to stating that these are magical objects and amulets of mainly protective magic. Owing to the narrow limits of the phenomenological method, this may lead to a dead end. A good example here is pendants of mostly beaver but also marten astragalus (Fig. 2) and beaver figures which were used over a seemingly long period of time from the Neolithic to the Late Iron Age. Then again, pendants were used only in the Neolithic period and in the Late Iron Age, and not in the intermediary periods. Hence, their observation as a single phenomenon is not justified, because they represent two independent traditions. Also, the Late Iron Age pendants may not be representative of beaver cult in its religious sense (Tvauri 2001, 161), but were rather objects signifying social status. Pendants carved of beaver and marten astragalus (Luik 2003) may have been symbols of fur hunters or traders, instead (see Leimus & Kiudsoo 2004). In inhumation burials in Estonia, claw pendants have been discovered in male burials, whereas astralagi have been found in female burials (Luik 2003, 166). Beaver and marten were highly valued for their fur, and the claw of a furbearer may have been a symbol of a successful (and, consequently, wealthy) hunter and his wife. Thus, the claws of furbearers may not be connected to religion on the primary level (talking about cult of beaver or beaver totemism in the 13th century), but were first and foremost symbols of social status and were connected with religion only through this function. Animal claws, which retain the bone, thus indicating the existence of the object, were certainly not the only symbols - strips of fur and tails were used for the same purposes. All these animals whose paws or claws have been discovered could be grouped under a common name 'furbearers'. Analysing the material from this aspect may attribute altogether different meanings to objects that have been so far unquestionably linked to religion.

[FIGURE 2 OMITTED]

Some important aspects in the study of prehistoric religion

To conclude, I would like to present some starting-points, which may prove consequential in the view of the research history of Estonian prehistoric religion, and which may perhaps concretise and animate the current static image of "cultic objects" and "ritual places".

Religion is dynamic and undergoes constant changes. Religion is by nature closely related to social structures, and changes in social structures inevitably cause changes in religion. Consequently, we cannot talk about an abstract notion called "Estonian prehistoric religion", as in different periods it has displayed different nuances.

Categorisation of different periods is also problematic. In real life such boundaries are never established, and those that we draw ourselves, mostly in order to systematise and present material, always remain arbitrary. Various transitional stages tend to form next to the already established periods, which may make the general situation all the more confusing. Division into specific periods is also disputable. If a periodisation is constructed on the basis of certain relics, phenomena, objects, etc., it means that the rest of the phenomena will have to be fitted into forced boundaries. Not all phenomena have undergone similar changes: some last longer and in a more stable form over many periods or belief systems. Perhaps the most stable indicators might be graves, which are known from every archaeological period and which I myself have used in constructing periodisation (Jonuks 2003). On the other hand, burials provide an uneven view of prehistoric religion, as the focus is mainly on particular phenomena (such as beliefs connected to soul and the otherworld) and not specifically on prehistoric religion. Also, burials that have been discovered so far represent only a part of prehistoric burials. Still, of all the available phenomena, burials are the most stable ones to base a periodisation on.

Newer phenomena are fitted to earlier material. As already mentioned, religion undergoes constant changes. At the same time, not all phenomena constituting a religious system evolve equally: depending on various factors the development may vary. This constant, though uneven process requires that new phenomena, either borrowed or undergoing a transformation, should fit into the existing system. This means that it is impossible to borrow a phenomenon the principles of which would be different from the established system, and also that no phenomenon can undergo a sudden or drastic change. The best example here would be the bauta grave cemetery in Valkla, North Estonia, where remains of a single cremation burial had been inhumed within a stone circle laid in front of a bauta stone (Fig. 3). This is quite irregular from the widely practised custom of the period to spread the cremated remains of several individuals between stones in the grave. The ideological explanation for the Valkla example might be the conception of individual soul (which explains the individual burial of the dead); this, however, did not fit into the generally established conception of collective soul, and has therefore remained a unique phenomenon in Estonia.

[FIGURE 3 OMITTED]

Such process also implies that in order to understand the phenomena of a certain period and their formation one must consider the period preceding these--that is, the broader context where these phenomena stem from.

Interdisciplinarity is of critical importance in the study of prehistoric religion. My emphasis on the importance of archaeological source material above does not necessarily mean ruling out methodologies of other disciplines. Archaeological source material has been given prominence only in consideration of the period, as no adequate written sources are available for this period, and transference of folkloric motifs is questionable. Nevertheless, other disciplines and methods must definitely be applied in interpreting archaeological sources. In interpreting religious material anthropological parallels have often been used, but while on the theoretical level authors agree that the parallel must be drawn with a society as close in the economical and technological advancement as possible, this principle has often been overlooked in practice, and the religion of Siberian hunter-fishers has been applied in interpreting the Estonian Iron Age, and comparisons are based on language affinity and the speculated similar worldview based on that (cf. Loorits 1959).

Rituality. The importance of rituality and the role of its study have been discussed above. On the basis of various traces of rituals, in the course of which objects have been left behind, burial structures have been constructed, bodies of the dead and grave objects have been inhumed in stone graves, and corpses or cremated remains have been handled in one way or another, we can speculate on the nature of these rituals. Having an understanding of and considering these rituals we can pose hypotheses about the underlying religious concepts. Consequently, it is impossible to form hypotheses about religion, mentality, or anything else on the basis of a single object, separately from the burial it belongs to, or any other context (see e.g. Antanaitis 1996). The use of rituals and especially theories about rituals in archaeology are far more complicated than they may seem. Liv Nilsson Stutz has pointed out that the use of theories from other disciplines, especially those posed in anthropology, is a risky business in the field of archaeology, and ignorance of their formation and context may lead to a dead end. Stutz suggests that a solution to this problem for archaeologists might be orientation to ritual as an action, rather than thought (Nilsson Stutz 2003, 51). However, behind a ritual there is always a thought, a religious context represented by the ritual, and while interpreting the traces of ritual as an action, it is important to consider that the ritual and the thought behind it would be in conformity.

Defining key terms. One of the main arguments against the phenomenological method of theoreticians of religion of the past few decades concerns the loose and vague use of terminology. Providing definitions for and explicating such terms would definitely facilitate the solution of sometimes absurd situations, where ancestral cult and basically analogous afterlife have been assigned to every possible period, using basically similar terminology and descriptive style. Typically, archaeologists are more likely to notice regional differences and have paid less attention to concepts that have transformed in time (see Jaanits et al. 1982, 99, 414). Regardless of that, most scholars agree that beliefs have changed in time in accordance with changes in other phenomena, and the conception of afterlife is bound to change at some point.

Also, there are certain key concepts favoured by archaeologists, such as, for example, fertility cult, animism, totemism, also the broader terms cult and ritual, which need to be defined by each author individually. Clearly, a similar term can be used to characterise quite different phenomena, the distinction of which depends on their context or material. Therefore, definitions of such terms widely vary in different studies.

Prehistoric religion has to be viewed as a general framework, and studies into narrower topics should proceed from this view. Several authors of recent studies have pointed out that the study of prehistoric religion is possible only if it is considered in its entirety (see Nilsson Stutz 2003, 53). The general context is associated with the view according to which all phenomena existing in religion at a certain point of time have to be linked and in concordance. Thus forms a general framework, where all phenomena communicate and complement each other. In addition to speculations relying on archaeological material, this approach suggests that hypotheses can be made about probable phenomena and their nature even if none of these phenomena or no material trace of them has been preserved. For example is quite likely that independent and clear-cut beliefs in god emerged in the Late Neolithic or Early Bronze Age, especially if we consider linguistic sources, etymologies and dating (Kulmar 1994; Sutrop 2002, 31), the distribution of some stone axes, which reportedly served ritual purposes (Salo 1990), and sacred grove hills that were taken into use towards the end of the Early Bronze Age--Pre-Roman Iron Age (Jonuks 2003).

Consideration of the prehistoric religion in its entirety would also enable to avoid the risk of treading the same path as traditional phenomenology, where focussing too much on single details (phenomena) and loosing the general view from sight leads to seeing a single emphasised detail of prehistoric religion. However, unless it is set in a broader context, it is impossible to adequately observe the formation of the given detail and its interrelation with others. Consideration of the general framework in the development of the entire religious system also facilitates the compilation of more adequate studies into individual phenomena. The fragmentariness of archaeological material, which does not provide us with a comprehensive view, can be overcome with the application of a long-term perspective, which may compensate the incompleteness of material on a specific moment or a relic (Nilsson Stutz 2003, 53). Naturally, a comprehensive view of prehistoric religion has been, is, and always will be an idealistic goal that cannot be achieved. Yet, I believe that this is what all research should aim for, even if it may sometimes lead to far-fetched speculations.

Acknowledgements

I highly appreciate the comments of my good colleagues and reviewers of this article, which made it much easier for me to formulate my ideas and to present the fmal version of this paper. I would also like to express my gratitude to Kait Tamm, translator of this text.

References

Ahlberg, N. 1999. Methodological choice and the study of sensitive issues.--Approaching Religion. Based on Papers Read at the Symposium on Methodology in the Study of Religions held at Abo, Finland, on the 4th-7th August 1997. Part II. Ed. T. Ahlbdck. (Scripta Instituti Donneriani Aboensis, 17, 2.) Abo; Stockholm, 9-31.

Ahlback, T. 1993. The problem of ritual. Based on papers read at the Symposium on Religious Rites held at Abo, Finland, on the 13th-16th of August 1991. Ed. T. Ahlbdck. (Scripta Instituti Donneriani Aboensis, 15.) Abo.

Ahlberg, T. 2003. Ritualistics. Based on papers read at the Symposium on Ritualistics held at Abo, Finland, on the July 31st-August 2nd, 2002. Ed. T. Ahlback; editorial assistant B. Dahla. (Scripta Instituti Donneriani Aboensis, 18.) Abo.

Antanaitis, I. R. 1996. Interpreting the meaning of East Baltic Neolithic symbols.--Cambridge Archaeological Journal, 8: 1, 55-68.

Boyer, P. 2001. Religion Explained: The Evolutionary Origins of Religious Thought. New York.

Briick, J. 1999. Ritual and rationality: some problems of interpretation in European archaeology.--European Journal of Archaeology, 2: 3, 313-344.

David, N. & Kramer, C. 2001. Ethnoarchaeology in Action. (Cambridge World Archaeology.) Cambridge.

Deemant, K. 1988. Taarausulistest.--Edasi, 30.04 and 7.05.

Dumezil, G. 2001. Indoeurooplaste miuidid ja jumalad. (Ajalugu. Sotsiaalteadused.) Tallinn.

Eliade, M. 1958. Sacred and Profane. New York.

Hedin, D. 1997. Phenomenology and the Making of the World. (Studia Philosophiae Religionis, 19.) Uppsala.

Hodder, I. 1991. Reading the Past: Current Approaches to Interpretation in Archaeology. Cambridge.

Honko, L. 1998. Folklooriprotsess.--Maetagused, 6. Tartu, 56-84.

Jaanits, L. 1961. Jooni kiviaja uskumustest. Ed. E. Jansen. Religiooni ja ateismi ajaloost Eestis, 2. Artiklite kogumik. Tallinn, 5-70.

Jaanits, L., Laul, S., Lougas, V. & Tonisson, E. 1982. Eesti esiajalugu. Tallinn.

Jensen, J. S. 2003. The study of religion in a new key: theoretical and philosophical soundings in the comparative and general study of religion. Studies in religion 3. Arhus.

Jonuks, T. 2003. Eesti metalliaja usundi pohijooni. Magistritoo. Tartu. (Kasikiri Tartu Ulikooli raamatukogus.)

Kaliff, A. 1997. Grav och kultplats: Eskatologiska forestallningar under yngre bronsalder och aldre jarnalder i Ostergotland. (Aun, 24.) Uppsala.

Kaliff, A. 1998. Grave structures and altars: Archaeological traces of Bronze Age eschatological conceptions.--European Journal of Archaeology, 1: 2, 177-198.

Knupffer, G. M. 1836. Der Berg des Thorapilla: Ein historicher Besuch.--Das Inland: Eine Wochenschrift fur Liv-, Esth- und Curlandische Geschichte, Geographie, Statistik und Litteratur, 22. Ed. F. G. v. Bunge. Dorpat, 361-366.

Konsa, M. 2003. Eesti hilisrauaaja matmiskommete ning uhiskonna kajastusi Madi kivivarekahnistus.--Arheoloogiga Laanemeremaades. Uurimusi Juri Seliranna auks. (Muinasaja teadus, 13.) Tallinn, 119-142.

Kuhnar, T. 1992. Eesti muinasusundi hingefenomenoloogiast, III. Hingekujutlused Eesti kiviaja arheoloogiaaineses.--Akadeemia, 9. Tartu, 1870-1887.

Kulmar, T. 1994. Eesti muinasusundi vanima kihistuse vae-, jumala- ja hingekasitluste teoloogia. Doktorivaitekiri. Tartu. (Kasikiri Tartu Ulikooli raamatukogus.)

Kulmar, T. 1999. Kultuurmaastikku luues ehk kuidas esiaja inimene ikkagi motles.--EAA, 3: 2, 162-164.

Kunnap, A. 2001. Kas eesti regivarss pakub puuduvaid andmeid eesti keelest enne aastat 1500?--Regilaul--keel, muusika, poeetika. Eds. T. Jaago & M. Sarv. Tartu, 9-15.

Lang, V. 1996. Muistne Ravala. Muistised, kronoloogia ja maaviljelusliku asustuse kujunemine Loode-Eestis, eriti Pirita joe alamjooksu piirkonnas, 1-2. (Muinasaja teadus, 4.) Tallinn.

Lang, V. 1999a. Kultuurmaastikku luues. Essee maastiku religioossest ja sumboliseeritud korraldusest.--EAA, 3: 1, 63-85.

Lang, V. 1999b. Kultuurmaastik ja arheoloogia: vastus kommentaaridele.--EAA, 3: 2, 170-174.

Lang, V. 2000. Keskusest aaremaaks. Viljelusmajandusliku asustuse kujunemine ja areng Vihasoo-Palmse piirkonnas Virumaal. (Muinasaja teadus, 7.) Tallinn.

Leeuw, G. van der. 1986. Religion in Essence and Manifestation. New Jersey.

Leimus, I. & Kiudsoo, M. 2004. Koprad ja hobe.--Tuna, 4. Tallinn, 31-47.

Ligi, P. 1995. Uhiskondlikest oludest Eesti alal hilispronksi--ja rauaajal.--Eesti arheoloogia historiograafilisi, teoreetilisi ja kultuuriajaloolisi aspekte. Ed. V. Lang. (Muinasaja teadus, 3.) Tallinn, 182-270.

Lintrop, A. 2001. "Ema haual" lego ja lauluna.--Regilaul--keel, muusika, poeetika. Eds. T. Jaago & M. Sarv. Tartu, 299-313.

Lintrop, A. 2003. Udmurdi usund. (Eesti Rahva Muuseumi sari, 5.) Tartu.

Loorits, O. 1932. Eesti rahvausundi maailmavaade. (Elav teadus, 12.) Tartu.

Loorits, O. 1951. Eestluse elujoud. (Iseseisvuslaste kirjavara, 5.) Stockholm.

Loorits, O. 1959. Grundzuge des Estnischen Volksglaubens, III. Lund.

Loorits, O. 1998. Liivi rahva usund, I-III. Mit einem Referat: der Volksglaube der Liven. Tartu.

Loorits, O. 1990. Eesti rahvausundi maailmavaade. Tartu.

Luik, H. 2003. Luuesemed hauapanustena rauaaja Eestis.--Arheoloogiga Laanemeremaades. Uurimusi Juri Seliranna auks. (Muinasaja teadus, 13.) Tallinn, 153-172.

Lougas, V. 1972. Vaikeste lohkudega kultusekivid.--Eesti Loodus, 12, 729-732.

Lougas, V. 1996. Kaali kraatrivaljal Phaetonit otsimas. Tallinn.

Masing, U. 1939. Taara paritolust.--Usuteaduslik Ajakiri, XI: 1, 1-16.

Masing, U. 1995. Eesti usund. Tartu.

Merkel, G. 1798. Die Vorzeit Lieflands. Ein Denkmahl des Pfaffen- und Rittergeistes. Erstes Band. Mit Kupfern und eine Karte. Berlin.

Metssalu, 12004. Rahvausust varauusaegsetes kroonikates. Puud, ussid ja pikne.--Pro Folkloristica. Tartu, 50-73.

Meyer-Dietrich, E. 1999. Ecology of religion. A hermeneutical model.--Approaching Religion: Based on Papers Read at the Symposium on Methodology in the Study of Religions held at Abo, Finland, on the 4th-7th August 1997. Part II. Ed. T. Ahlback. (Scripta Instituti Donneriani Aboensis, 17, 2.) Abo; Stockholm, 155-167.

Moora, A. 1956. Eestlaste muistsest usundist.--Religiooni ja ateismi ajaloost Eestis. Tallinn, 7-41.

Moora, H. & Annist, A. 2002. Teos rahvausundi uurimise alalt.--Meie rahvuskultuuri kusimusi. Eds. H. Runnel & A. Marksoo. (Eesti mottelugu, 47.) Tartu, 247-263.

Moore, M. 1995. A Bronze Age settlement and ritual center in the Monavullagh Mountains, County Waterford, Ireland.--Proceedings of Prehistoric Society, 61, 191-243.

Magi, M. 2001. Probable cult site beside the Tonija tarand-grave on the island of Saaremaa.--AVE, 2000,48-55.

Magi, M. 2002. At the Crossroads of Space and Time. Graves, Changing Society and Ideology on Saaremaa (6se1), 9th-13th centuries AD. (CCC papers, 6.) Tallinn.

Nilsson Stutz, L. 2003. Embodied Rituals and Ritualized Bodies. Tracing Ritual Practices in Late Mesolithic Burials. (Acts Archaeologica Lundensia, Series in 80, No. 46.) Lund.

Parker Pearson, M. 1982. Mortuary practices, society and ideology: an ethnoarchaeological study.--Symbolic and Structural Archaeology. Ed. I. Hodder. Cambridge, 99-114.

Parker Pearson, M. 1999. Food, sex and death. Cosmologies in the British Iron Age with particular reference to East Yorkshire.--Cambridge Archaeological Journal, 9: 1, 43-69.

Paulson, I. 1997. Vana Eesti rahvausk. (Eesti mottelugu.) Tartu.

Plaat, J. 2001. Usuliikumised, kirikud ja vabakogudused Laane-ja Hiiumaal. Usuuhenduste muutumis-protsessid 18. sajandi keskpaigast kuni 20. sajandi lopuni. (Eesti Rahva Muuseumi sari, 2.) Tartu.

Renfrew, C. 1996. The archaeology of religion.--The Ancient Mind. Elements of Cognitive Archaeology. Eds. C. Renfrew & E. V. Zubrow. Cambridge, 47-54.

Salo, U. 1990. Agricola's Ukko in the light of archaeology. A chronological and interpretative study of ancient Finnish religion.--Old Norse and Finnish Religions and Cultic Place-Names. Based on Papers Read at the Symposium on Encounters between Religions in old Nordic Times and on Cultic Place-Names held at Abo, Finland, on the 19th-21st of August 1987. (Scripta Instituti Donneriani Aboensis, 13.) Abo, 92-190.

Selirand, J. 1974. Eestlaste matmiskombed varafeodaalsete subete tarkamise perioodil (11.-13. sajand). Tallinn.

Sundqvist, O. 2002. Freyr's Offspring. Rulers and Religion in Ancient Svea Society. (Historia Religionum, 21. Acta Universitatis Upsaliensis.) Uppsala.

Sundqvist, O. 2003. Rituale.--Reallexikon der Germanischen Altertumskunde, 25. Herausgegeben von H. Beck, Bonn, D. Geuenich, Duisburg, H. Steuer, Freiburg. Berlin; New York, 32-47.

Sutrop, U. 2002. Taarapita-saarlaste suur jumal.--Maetagused, 16. Tartu, 7-38.

Tamm, M. 2001. Uus allikas Liivimaa ristiusustamisest. Ida-Baltikumi kirjeldus Descriptiones terrarum'is (u 1255).--Keel ja Kirjandus, 12, 872-884.

Tarvel, E. 1982. Sissejuhatus.--Henriku Liivimaa kroonika. Tallinn, 5-21.

Tuovinen, T. 2002. The Burial Cairns and the Landscape in the Archipelago of Aboland, SW Finland, in the Bronze Age and the Iron Age. (Acta Universitatis Ouluensis. B: Humaniora, 46.) Oulu.

Turcan, V. 2001. Old-Slavonic sanctuaries in Czechia and Slovakia.--Studia Mythologica Slavica, IV. Ljubljana, 97-114.

Tvauri, A. 1997. Eesti lohukivid.--Arheoloogilisi uurimusi, 1. Ed. H. Valk. (TUAKT, 9.) Tartu, 11-53.

Tvauri, A. 2001. Muinas-Tartu: Uurimus Tartu muinaslinnuse ja asula asustusloost. (Muinasaja teadus, 10.) Tallinn; Tartu.

Tvauri, A. 2003. Balti arheoloogia maailmaajaloo poorises ehk gooti teooria saatus.--EAA, 7: 1, 38-71.

Valk, H. 1998. Eesti 13.-17. sajandi rahvausundi allikatest, uurimisseisust ja probleemidest.--Eestimaa, Liivimaa ja laane kristlus. Eesti-Saksa uurimusi Baltimaade kirikuloost. Estland, Lettland and westliches Christentum. Estnisch-deutsche Beitrage zur baltischen Kirchengeschichte. Eds. S. Rutiku & R. Staats. Kiel, 75-88.

Valk, H. 2001. Rural Cemetries of Southern Estonia 1225-1800 AD. (CCC papers, 3.) Visby; Tartu.

Valk, H. 2003. Christianisation in Estonia. A process of dual-faith and syncretism.--The Cross Goes North. Process of Conversion in Northern Europe, AD 300-1300. Ed. M. Carver. York, 571-579.

Valk, U. 2000. Regilaul kui kommunikatsioon teispoolsusega: dialoogist nagemusteni.--"Kust tulid lood minule ..." Artikleid regilaulu uurimise alalt 1990. aastatel. Eds. T. Jaago & U. Valk. Tartu, 245-276.

Vedru, G. 2002. Maastik, aeg ja inimesed.--Keskus--tagamaa--aareala. Uurimusi asustushierarhia ja voimukeskuste kujunemisest Eestis. Ed. V. Lang. (Muinasaja teadus, 11.) Tallinn; Tartu, 101-122.

Victor, H. 2002. Med graven som granne om bronsalderns kulthus. (Aun, 30.) Uppsala.

Viires, A. 1986. Paar pilguheitekatset eesti muinasusku.--Looming, 12, 1666-1675.

Viires, A. 2001. Kultuur ja traditsioon. (Eesti mottelugu, 39.) Tartu.

Vries, J. de. 1970. Altgermanische Religionsgeschichte, I-II. Berlin.

Wallin, P. 1993. Ceremonial Stone Structures. The Archaeology and Ethnohistory of the Marae Complex in the Society Islands, French Polynesia. (Aun, 18.) Uppsala.

Widholm, D. 1998. Rosen, ristningar och riter. (Acta Archaeologica Lundensia. Series prima in 40: 23.) Stockholm.

[TEXT NOT REPRODUCIBLE IN ASCII.] M. 1955. [TEXT NOT REPRODUCIBLE IN ASCII.]

Tonno Jonuks, Estonian Literary Museum (Eesti Kirjandusmuuseum), Vanemuise 42, 51003 Tartu, Estonia; tonno@folklore.ee