Japanese Mythology A To Z

How I Got Around Having to Sort Through Hundreds of Shinto Deities Last Year

My company needed a complete reference document covering every named entity in Japanese mythology for a game localization project. I spent three weeks mapping it out and ended up with what we called the Japanese Mythology A To Z because the client wanted everything alphabetized by English transliteration. Here is how I actually did it, the problems I ran into, and what I would do differently now. The core problem is that Japanese mythology doesn't have a single canonical source. You have the Kojiki and the Nihon Shoki as primary texts, both compiled in the early 8th century, but they contradict each other on basic facts like who begat whom. Then there are regional folklore collections, shrine traditions, and later Buddhist syncretic additions that get lumped into "mythology" by people who haven't read the original texts. My first move was to establish a source hierarchy: Kojiki and Nihon Shoki entries get primary citations, regional folklore gets tagged as secondary, and anything from Edo-period compilations or modern pop culture gets flagged separately. Without that discipline, your A to Z becomes a mess of contradictory genealogies. I built the document in a spreadsheet first, not a word processor. Each row was a single deity or mythological figure. Columns tracked: English name, Japanese name in kanji and kana, primary source text, alternative names, family relationships, domains or roles, major myths they appear in, and notes on regional or sectarian variations. The spreadsheet format forced me to make decisions instead of writing vague paragraphs. If I couldn't fill in a cell, I knew there was a gap I needed to research.

The alphabetical ordering itself caused immediate friction. Amaterasu omikami appears under "A" in English but the Japanese order would place her under "Ama" in gojūon sequence. Some entities have no standard English rendering — a local mountain spirit might only appear in a single village shrine's records with a name that resists clean transliteration. I solved this by adding a Japanese alphabetical cross-reference column and using a secondary sort by kana, which let me catch entries that wouldn't surface in a strict English alphabetical scan.

A Specific Problem That Almost Ruined the Project

About halfway through, I realized I had double-counted roughly twelve entities because the same figure appeared under different names in different source texts. Susanoo appears in the Kojiki as the storm god who descends to Izumo, but he is also referenced in later shrine traditions under epithets that made him look like a completely separate deity. I had entries for "Susanoo no Mikoto" and "Kuninushi no Mikoto's rival" sitting side by side as if they were unrelated. The fix was to create a master disambiguation table where each unique entity got a single ID number, and every entry in the main spreadsheet pointed to that ID. It added maybe two hours of work but saved me from having to explain to the client why their glossary listed the same god twice under different letters. Another issue was the Kami versus Tsukuyomi versus other celestial beings boundary. Some entries I wrote for "night gods" or "moon deities" turned out to be later interpretive labels rather than actual named figures from the primary sources. The Kojiki doesn't give Tsukuyomi much narrative presence beyond the scene where he kills Ukemochi and Amaterasu turns away in disgust. Everything I found about Tsukuyomi beyond those passages was either Buddhist-influenced elaboration or modern retelling. I flagged those entries clearly so readers wouldn't mistake Edo-period commentary for ancient myth.

What the Document Actually Covers

A complete run-through would include the creation deities like Izanagi and Izanami, the major celestial kami such as Amaterasu, Tsukuyomi, and Susanoo, the (yaoyorozu no kami) that appear during the stirring of the ocean, the land deities of Izumo like Okuninushi, and hundreds of localized spirits tied to specific mountains, rivers, and shrines. The Nihon Shoki has five different versions of the same myths, so some entries needed to present all variants rather than picking one "correct" version. I noted which version came from which text so anyone reading could see where the traditions diverged. The tricky section involves the jinchōkoku and other spirit classifications that don't map cleanly onto Western categories of "god" or "demon." Entities like tengu, kitsune, and onryo occupy a gray area between folklore and religious belief. Whether they belong in a mythology reference depends entirely on your definition of mythology. I included them but separated them into a distinct section with clear sourcing notes, because mixing them in with the Kojiki pantheon without that distinction misleads people about what the ancient texts actually contain.

Where This Approach Breaks Down

The biggest limitation is that any A to Z format inevitably flattens the hierarchical and relational nature of Shinto cosmology. In the actual tradition, these entities aren't alphabetical entries — they exist in genealogies, shrine lineages, and regional networks of obligation. Alphabetizing them makes it easy to miss that a minor river kami in one province is considered a daughter of a major deity in another. The format forces you to treat each entity as standalone when the sources consistently present them as interconnected. Another failure point is the transliteration problem. Using Kunrei-shiki versus Hepburn romanization changes how entries sort and how they look to someone who knows Japanese. I chose Hepburn because the English-language gaming industry uses it, but I noted the Kunrei equivalents for entries where the difference matters for disambiguation. Some names simply don't have a stable English form, and forcing one creates false precision. If you are building something like this for professional use, I would recommend starting with the scholarly apparatus rather than the primary texts directly. Works like William George Aston's translation of the Nihon Shoki, Basil Hall Chamberlain's Kojiki translation, and more recent scholarship from authors like Helen McCullough and Mark Teeuwen will save you from repeating interpretive errors that have been corrected decades ago. The document took me roughly forty hours of research and writing, and I still left out entire categories of regional deities because there simply isn't room. If you need comprehensive coverage of local shrine traditions, you are looking at a multi-volume project, not a single reference file.