Programme: Pre-Christian Baltic Religion & Religious Transformation
Status: First research cycle complete with limitations; corrective primary-source review completed; public release package ready
Project lead: BAFES Chair / Project Lead
Research period: Spring-Summer 2026
First research cycle completed: August 2026
Working language: Russian
Public language: English; Lithuanian summary where appropriate
Research dossier: INTERNAL
Public outputs: Project Page, Research Note 01, Selected Bibliography and curated Open Questions ready for public release; substantial research paper requires external specialist review
Public Outputs
- Research Note 01: What the Early Sources Actually Allow Us to Say About Pre-Christian Baltic Religion
- Selected Bibliography
- Open Questions
The complete working dossier, Claim Matrix, internal confidence decisions and research log remain INTERNAL. Public release is selective and does not imply publication of the complete research infrastructure.
Research Question
What can the surviving evidence actually establish about pre-Christian Baltic religious traditions, and where does the evidence end and later scholarly, folkloric or modern reconstruction begin?
BAFES-R-001 examines early written testimony, archaeology and modern scholarship without assuming in advance that the Baltic peoples shared a single fixed pantheon, a centralised priesthood or an unchanged religious system.
Scope
The project focuses on medieval and early modern evidence concerning Lithuanian, Prussian and other Baltic religious traditions, with comparative material used only where there is a specific historical or methodological reason.
The main written-source map extends from the earliest surviving medieval evidence to the seventeenth century. Later folklore and ethnography are used as evidence for later tradition and as possible comparative material, but they are not automatically projected back into the pre-Christian period.
Topics include:
- named divine or cult figures;
- divination and ritual decision-making;
- ritual specialists;
- sacred landscapes;
- funerary practices and ideas about the dead;
- animal and human sacrifice in written sources;
- the Romow / Criwe tradition;
- the relationship between early evidence and later reconstructed pantheons.
Method
The project works from source to claim rather than from a reconstructed religious system back to selected evidence.
For major claims, BAFES distinguishes between:
- what a source explicitly says;
- what can reasonably be inferred from several forms of evidence;
- what remains a scholarly hypothesis;
- what belongs to later tradition or modern reconstruction;
- what cannot currently be determined.
Repeated statements are not automatically treated as independent confirmation. Where chronicles, compilations or manuscript witnesses may be related, that relationship is considered before the evidence is counted.
Archaeology is used as an independent material control where appropriate, but material evidence is not treated as if it could by itself reveal the exact beliefs or explanations of historical participants.
First-Cycle Findings
1. Polytheistic traditions are strongly supported, but a single fixed Baltic pantheon is not
Early sources provide strong grounds for describing several Baltic religious traditions of the thirteenth and fourteenth centuries as polytheistic in a minimal analytical sense: they refer to multiple gods or named divine and cult figures.
This does not establish one common Baltic pantheon, a uniform theology or a single state religious system.
2. The early Lithuanian lists are evidence, not a ready-made ranked pantheon
The Galician-Volhynian Chronicle and the manuscript tradition of the Chronograph of 1262 preserve important early divine names.
Direct control of the critical Chronograph tradition confirms the sequence Andaj, Perkūnas, Žvorūna and Teliavelis, with explanatory material attached to several figures. The existence of this list is therefore not merely a modern reconstruction.
However, a unified ranked state pantheon emerges only after several additional scholarly identifications and assumptions are made. Proposed equivalences such as Diviriks = Perkūnas, Nunadievis = Andaj and Medeina = Žvorūna remain hypotheses of different strength rather than direct statements of the medieval texts.
3. The two early Lithuanian corpora should not automatically be counted as fully independent confirmation
The Galician-Volhynian Chronicle and the Chronograph of 1262 are distinct thirteenth-century textual corpora. Textual scholarship has not demonstrated direct borrowing of one Lithuanian religious list from the other, and their wording and sets of figures differ substantially.
At the same time, both works belong to a closely connected literary environment. Their overlapping names therefore should not automatically be counted as two fully independent historical observations.
This distinction matters especially when evaluating attempts to reconstruct a single early Lithuanian pantheon from both texts.
4. Perkūnas has strong early cross-source attestation
The Livonian Rhymed Chronicle independently names Perkune in a Lithuanian context, while the directly checked Chronograph tradition contains Perkūnas with a thunder gloss.
This gives strong early support for Perkūnas as a named Lithuanian divine or cult figure. It does not by itself establish his precise rank within a state pantheon or prove that every later tradition about Perkūnas was already present in the thirteenth century.
5. Divination by lot is one of the stronger cross-source patterns
Several medieval sources describe the use of lots in decisions connected with gods, warfare or future action among more than one Baltic group.
The evidence supports a widespread practice among several groups, but not one standardised pan-Baltic divinatory institution.
6. Specialised ritual performers are probable; a centralised priesthood is not demonstrated
The Treaty of Christburg refers to Tulissones and Ligaschones in a funerary context, while the Livonian Rhymed Chronicle describes a specialised ritual performer usually transmitted as blûtekirl.
Together these sources strongly support ritual specialisation. They do not establish a single hereditary priestly caste or a centralised Baltic hierarchy.
7. Sacred natural places have substantial historical support
Medieval written sources and modern interdisciplinary research support the historical importance of sacred groves, fields, waters and other places in parts of the Baltic region.
A specific modern location cannot, however, be identified as a pre-Christian sanctuary from a place-name, a late legend or a single archaeological find alone. Individual sites require separate historical, linguistic, geographical and archaeological assessment.
8. The Prussian and Sambian funerary material has an independent archaeological basis
Archaeological research documents recurring combinations of human burials or cremations with horses, weapons and horse equipment in the Prussian and Sambian material of the tenth to thirteenth centuries.
This independently supports important material elements found in medieval written descriptions.
The religious meaning is a separate question. Written sources provide grounds for linking possessions or social status with continued existence after death, but archaeology alone cannot establish that interpretation.
9. Ritual use of horses is well supported; a single Baltic sacrifice system is not
Written and archaeological evidence support ritual and sacrificial uses of horses in parts of the Baltic region.
The evidence does not establish one common ceremony with the same recipient, purpose or procedure across all Baltic groups.
10. Medieval chronicles report executed ritual killings of captives, but frequency remains unknown
The first research cycle identified several medieval reports in which captives were actually killed in a ritual context, including cases described as offerings to gods.
This supports the historical plausibility of human sacrifice in some contexts. It does not establish that such sacrifice was frequent, universal or governed by one standard Baltic ritual.
Findings That Remain More Uncertain
Curche
The Treaty of Christburg securely places Curche in an annual post-harvest cultic context. A targeted comparison of the published facsimile of Latin copy A with the revised transcription confirms the relevant reading for the present project.
The common description of Curche simply as a “god of harvest” or “fertility god” is nevertheless an interpretation rather than the wording of the treaty itself. The exact function and etymology remain open.
Teliavelis
Teliavelis is named in two distinct thirteenth-century textual corpora. In the Chronograph of 1262 he is described as a smith associated with forging the Sun and placing it in the sky.
The origin of this smith-solar characterisation remains debated. Ilja Lemeškin has drawn attention to possible literary influence from Tubal-Cain and related material within the same chronographic environment. This does not prove that Teliavelis was invented by the compiler, but it means that the source’s description should not automatically be treated as an unmodified record of a living Baltic myth.
Equations such as Teliavelis = Kalvelis or Teliavelis = Velnias remain scholarly reconstructions rather than direct statements of the thirteenth-century sources.
Romow / Criwe
Peter of Dusburg’s account of Romow and Criwe is historically important, but the project does not treat it as proof of a pan-Baltic institution comparable to a church headed by a Baltic equivalent of the pope.
A more limited historical core, involving a significant sacred or assembly place and a religious authority, remains possible but unproven.
The “one third of the booty” motif
The Livonian Rhymed Chronicle contains a promise to dedicate a third part to the gods, and Peter of Dusburg gives a related account involving Criwe. The first cycle found insufficient grounds to treat the two passages as independent proof of a fixed rule requiring one third of military booty.
Corrective Primary-Source Review
The first cycle was originally completed with two explicitly disclosed technical source limitations. Both were later revisited when the relevant critical publications became directly available.
Chronograph of 1262
BAFES directly checked Ilja Lemeškin’s critical reconstruction and the relevant passages of the surviving manuscript tradition, including both editorial branches, and also checked the corresponding passage in Baltų religijos ir mitologijos šaltiniai, vol. I.
The relevant religious sequence is confirmed in the critical tradition. The four manuscript witnesses are not treated as four independent historical testimonies, because part of the manuscript tradition is genealogically dependent.
Treaty of Christburg
The original treaty has not survived. BAFES therefore does not claim to have inspected a lost original.
For the present research claims, BAFES performed a targeted visual comparison between the published facsimile of Latin copy A and the revised printed transcription. The relevant forms Curche, Tulissones, Ligaschones and nisum were confirmed at the level required for the first-cycle claims.
This is not presented as a new complete diplomatic or palaeographic edition of the charter.
Continuing Limitations
The first cycle remains complete with limitations, but these are now primarily substantive rather than technical access problems.
The main continuing limitations are:
- regional diversity among Prussian, Lithuanian, Samogitian, Curonian, Semigallian and other evidence;
- uncertain informational independence between some written corpora;
- unresolved historical equivalences between several divine names;
- uncertain exact functions and etymologies of figures such as Curche and Teliavelis;
- the unresolved historical core of Romow / Criwe;
- uncertainty about the frequency and normative status of human sacrifice;
- the methodological gap between archaeological material and the exact beliefs attached to it;
- the chronological gap between medieval evidence and much later folklore.
Research Record
The first internal research cycle contains:
- 50 registered sources and scholarly controls;
- 60 structured research claims;
- 28 tracked open questions;
- source dossiers and textual-control files;
- historiographic comparison of competing models;
- source-dependency audits;
- citation and formulation audits;
- an internal Russian-language research synthesis, currently at corrective DRAFT 0.4;
- a corrective internal review after direct acquisition of previously unavailable critical editions.
The complete working dossier remains INTERNAL. Completion of the research cycle does not imply open publication of all notes, matrices or working translations.
Selected Sources and Scholarship
A separate Selected Bibliography gives the public reading list and short notes on the evidential role of key works.
The full internal Source Register remains INTERNAL.
Open Questions
A curated public set of unresolved questions is available separately as Open Questions.
These questions are published to make the limits of the evidence visible. They are not presented as defects that need to be hidden before publication.
Access and Review Status
Research dossier: INTERNAL
Project page: READY FOR PUBLIC RELEASE
Research Note 01: Version 1.0, READY FOR PUBLIC RELEASE; internally reviewed; not externally peer reviewed
Selected Bibliography: READY FOR PUBLIC RELEASE
Open Questions: READY FOR PUBLIC RELEASE
Substantial research paper: External specialist review required before publication
The requirement for external specialist review applies to a future substantial academic paper because of the complexity of historical Baltic linguistics, source criticism and disputed identifications. It is not a prerequisite for releasing the present Project Page, Research Note, Selected Bibliography or curated Open Questions.