The 384 line texts of the Zhouyi, read in the classical Chinese, hold a handful of verdict words: 吉 (auspicious), 凶 (ominous), 悔 (regret), 吝 (humiliation), 咎 (blame, mostly 無咎, no blame) and 厲 (danger). Each cell below is the share of the texts in that cell that hold the word, with the count beside it. Colour runs from warm (more 凶 than 吉) through paper (even) to cool (more 吉 than 凶), and every cell prints its numbers, so nothing depends on colour. Tap a hexagram to read its six lines.
What stands out, computed
The clearest pattern is the place: the central lines (2 and 5) carry 吉 in 44% of their 128 texts against 24% of the other 256, and 凶 in 7% against 17% (p = 0.00016 and p = 0.0073, exact Fisher).
Solid or broken matters much less. In the fifth place the broken (yin) line carries 吉 in 17 of 32 texts (53%) and the solid (yang) line in 12 of 32 (38%), yet with 32 texts a side that gap does not separate them (p = 0.32).
Lines in a correct place (yang in 1, 3, 5; yin in 2, 4, 6) carry 吉 in 30% of 192 texts, no more than lines in an incorrect place (32%).
These are counts of words, a pattern to test and not a rule. 貞吉 ("firmness brings good fortune") is a condition, 悔亡 ("regret vanishes") holds 悔, and one text can hold both 吉 and 凶. The p-values are not adjusted for the many cuts on this page, and the 384 texts are not independent draws.
Loading…
Place, yin and yang
The six places (top line first, as a hexagram is drawn) against the solid (yang, a 9 when it moves) and broken (yin, a 6) lines. ✓ marks a line in its correct place; the central places are 2 and 5.
Correct places and central lines
A line is in a correct place when it is yang in an odd place or yin in an even one. The table crosses that with the central places (2 and 5).
The eight trigrams
Each trigram as the lower trigram (lines 1 to 3) and as the upper (lines 4 to 6), in Fu Xi order.
Count
The 64 hexagrams
Each hexagram's six line texts. Tap one to read them.
Order
Every word, every cut
The same counts for all seven words. Each column is shaded on its own scale (cool for 吉, warm for 凶, ink for the rest).
How it is counted
The texts. The 384 line texts (64 hexagrams, 6 lines) in the classical Chinese of grammars/i-ching-chinese-original (Project Gutenberg #25501, public domain). The Judgments and the Wings are not counted.
The words. A text is counted once for each word it holds. 無咎 is also counted on its own, as a part of 咎 (85 of its 90 texts). The text uses 無 and 厲, never 无 or 厉.
Yang and yin. A line is solid (yang, a 9 when it moves) or broken (yin, a 6) in the hexagram as it stands, from scripts/hexagram-binary.json.
The colour. In the default view a cell is shaded by its share of 吉 minus its share of 凶, fully at 50 points apart. Choose a single word to shade by its own share instead. The pale cells are the near-even ones.
The p-values. Exact two-sided Fisher tests on the pairs named above, as a guide to how much a gap could be chance. They are not corrected for the many cuts here and the texts are not independent, so none of them settles anything.
What this does and does not show
It shows where the words fall in the book as written. It does not say what a line means, how a reader should weigh a verdict word against the rest of its text, or what a cast will bring. The same counts for the fifth place are the table in The Changing-Lines Lens.
Data: data/verdict-words.json, built by scripts/verdict_words.py, which scripts/changing_lines.py --check runs. Line texts on tap: Legge's English (grammars/zhouyi-core, public domain) beside the Chinese.