Myth Index

A cross-cultural index of mythological motifs, one row per attestation: a named primary text, a locator into it, a two-sentence summary, and Thompson Motif-Index cross-references — spanning nine motif families from the flood to the immortality quest, across the ancient Near East, the Mediterranean, India, East Asia, Scandinavia, and the Americas.

Motif families covered

Flood (11), divine council (17), theomachy (8) — imported from the three specialized Wheel of Heaven concordances — plus six new families: creation of humans (9), sky-descent (6), tower of Babel (5), giants (7), garden-paradise (6), and the immortality quest (6).

How the index relates to the specialized datasets

The Myth Index is the spine. For the three families that already have dedicated concordances, each row's see_dataset column points at the specialized table — the Flood-Myth Concordance, the Divine-Council Index, and the Theomachy Cross-References — which carry the full per-family column depth (survivor and vessel, council term and presiding figure, champion and weapon). The index row gives every family the same comparable shape.

The scholarly crosswalk

Each row carries thompson_motifs — motif numbers from Stith Thompson's Motif-Index of Folk-Literature (A1010 Deluge, A167.1 Council of the gods, F531.6.1.1 Sons of God and daughters of men, and so on). Motif numbers are the standard scholarly address system for exactly this material, so the index joins cleanly against any motif-indexed corpus.

The atu_type and aarne_1910_type columns are present but sparse — and that sparseness is deliberate honesty: the Aarne–Thompson–Uther system catalogues folktales, while myth material is addressed by the Motif-Index's A-section. Rows only carry a tale-type number where a genuine type exists. The companion dataset, the Aarne 1910 Tale-Type Index, publishes the founding tale-type catalogue itself in structured form.

How the rows were built

Every row requires a named primary text and a locator — the same "primary text required" rule as the specialized concordances. Summaries describe what the source says; they are deliberately separated from the woh_reading column, which carries the Wheel of Heaven canon reading where the canon states one (the ark as genetic repository, Eden as the creators' laboratory-garden, the tower as a rocket project). Blank woh_reading means the corpus offers no canon-specific claim for that attestation — the comparative layer stands on its own.

How Wheel of Heaven reads it

Read across the families and the corpus's larger claim comes into view: the same small set of administrative scenes — an assembly that decrees, a laboratory that creates, a garden that encloses, a reset that spares one lineage, a technology withheld — recurring across traditions that had no contact. The corpus reads that convergence as memory of a shared history rather than a shared imagination; the argument is developed across the wiki and the library of digitized sources this index links into.

Also available on

This dataset is also published, CC0, on two open-data platforms — the same files, mirrored for reach and preservation:

• Hugging Face — huggingface.co/datasets/wheelofheaven/myth-index
• Kaggle — kaggle.com/datasets/zarazinsfuss/myth-index