bazi-nayin / README.md
Shann5's picture
v2: original-language data only (English readings live on the site)
ba742fd verified
|
Raw History Blame Contribute Delete
5.59 kB
---
license: cc-by-4.0
language:
- en
- zh
pretty_name: The 30 Nayin (纳音) — English Translations and Classical Relations
size_categories:
- n<1K
task_categories:
- translation
tags:
- chinese
- glossary
- terminology
- chinese-metaphysics
- bazi
- four-pillars
- nayin
configs:
- config_name: default
data_files: nayin.csv
- config_name: pillars
data_files: v2/pillars.csv
- config_name: meetings
data_files: v2/matrix.long.csv
---
# The Thirty Nayin (纳音) — English Translations
The **nayin** (纳音, *nà yīn*, "received tones") are thirty poetic sound-images from
Chinese metaphysics. Each covers two adjacent pairs of the sixty Jiazi (六十甲子)
stem-branch cycle and attaches an elemental image to them — not just *Metal*, but
*Gold in the Sea*; not just *Fire*, but *Thunderbolt Fire*.
English sources render these names inconsistently: the same nayin turns up as three or
four different English strings across sites, which makes the terms impossible to join
or cross-reference. This dataset is one carefully considered rendering, published with
the reasoning behind each choice so it can be argued with.
## v2 — classical readings (configs `pillars` and `meetings`)
Structured from volume 1 of the *Sānmìng Tōnghuì* (三命通會, Ming dynasty, Siku Quanshu
edition on Wikisource), in the original Chinese. Every relation carries a verbatim quote;
`classical/verify.py` on GitHub checks all 750 against the source text. English readings
of these pillars and meetings are on [auspiceoracle.com](https://auspiceoracle.com/en/content/nayin).
```python
from datasets import load_dataset
pillars = load_dataset("Shann5/bazi-nayin", "pillars") # 60 rows: arc stage, image, likes, welcome/avoid/unharmed_by
meetings = load_dataset("Shann5/bazi-nayin", "meetings") # 663 rows: subject nayin meets another → + / - / ~ / 0, with quotes
```
- **`pillars`**: all sixty stem-branch pillars. Arc stage (the thirty images read as one
life of qi, six stages), the per-pillar image and what it likes, and the stem-branch
pairs or elements it welcomes, avoids, or is unharmed by.
- **`meetings`**: from the thirty chapters on how each nayin fares meeting the others.
`verdict` is `+` welcomes, `-` avoids, `~` conditional (the condition is in the quote),
`0` no effect; `scope` is `element` when filled from a blanket statement
("other metals are of no use"). 663 of 900 pairings; empty pairings are absent.
Relations are directional: in 41 pairs the two chapters disagree.
This records what one Ming-dynasty book says, not a claim that any of it comes true.
The verdict for each quote is a curation call; the source of truth, JSON versions and
the source text are on GitHub: **[Shann5/bazi-open-data](https://github.com/Shann5/bazi-open-data)** — corrections belong there as issues.
## Fields
| Field | Type | Description |
|---|---|---|
| `hanzi` | string | Nayin name in Chinese characters (e.g. 海中金) |
| `pinyin` | string | Tone-marked pinyin (e.g. *hǎi zhōng jīn*) |
| `english` | string | English translation of the image (e.g. Gold in the Sea) |
| `element` | string | Five-Element category in Chinese (金木水火土) |
| `element_en` | string | Five-Element category in English |
| `ganzhi` | list[string] | The two stem-branch pairs this nayin covers |
| `ganzhi_pinyin` | list[string] | Pinyin of those two pairs |
30 rows, covering all 60 stem-branch pairs. `nayin.json` keeps `ganzhi` and
`ganzhi_pinyin` as two-element arrays; `nayin.csv` splits them into the columns
`ganzhi_1`, `ganzhi_2`, `ganzhi_1_pinyin`, `ganzhi_2_pinyin`.
## Translation principles
1. **Keep the image, not the word order.** 海中金 is "Gold in the Sea", not "Sea Middle Gold".
2. **Say *Gold* when the image is refined treasure, *Metal* when it is raw or forged.**
The same character 金 splits into two English words: 海中金 / 沙中金 are gold;
剑锋金 (sword-edge) stays metal.
3. **Use the established English term where one exists** (Milky Way for 天河), and plain
concrete English elsewhere.
## Provenance
The ganzhi→nayin pairing follows the traditional sixty Jiazi order (甲子乙丑海中金 …
壬戌癸亥大海水), cross-checked against
[lunar-typescript](https://github.com/6tail/lunar-typescript). The English translations
are original work.
One caveat worth stating: a nayin lookup keys off the stem-branch pillars, and those key
off birth time in *solar* time, not clock time. Clock time can be off by more than an
hour once historical DST, longitude offset from the zone's standard meridian, and the
equation of time are stacked — near a two-hour branch boundary that is a different
pillar and therefore a different nayin.
[How the correction works](https://auspiceoracle.com/en/content/true-solar-time).
## Source & canonical copy
- GitHub (canonical): <https://github.com/Shann5/bazi-open-data>
- Reference page with a prose gloss for each image:
<https://auspiceoracle.com/en/content/nayin>
This Hugging Face repo is a **mirror**. Corrections go to the GitHub repo as issues.
## License & citation
**CC BY 4.0.** Share and adapt freely, including commercially, with credit and a link.
> Nayin translations by [Auspice Oracle](https://auspiceoracle.com), used under
> [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).
```bibtex
@misc{auspice_bazi_nayin,
title = {The Thirty Nayin: English Translations},
author = {Auspice Oracle},
year = {2026},
url = {https://github.com/Shann5/bazi-open-data},
note = {CC BY 4.0}
}
```