5D Library › Almanac Auspicious Boolean Lattices, Jianchu Twelve-Officer Congruences and Runic Combinatorics · 中文
Chapter 6. What a Reversal Can Say, and How Much the Rune Set Matters
Three exact consequences
If reversals carry meaning in a reading, their frequencies should be known before they are interpreted. Three consequences follow directly from the 15 reversible runes.
Calculation
# Chapter 6. p11: What a reversal can and cannot say. Exact results for
# the Elder Futhark (15 of 24 runes reversible), drawing without
# replacement and turning each with probability 1/2: (1) the chance that
# a Norns reading of 3 shows no reversal, against 1/8 under the shortcut;
# (2) the chance that a given drawn rune is reversed, by aett (family of
# 8); (3) where reversed runes come from: no reversed rune can ever be
# one of the 9 symmetric runes, and Hagal's aett supplies only 2 of 15.
from fractions import Fraction
from math import comb
FUTHARK = ("Fehu Uruz Thurisaz Ansuz Raidho Kenaz Gebo Wunjo "
"Hagalaz Naudhiz Isa Jera Eihwaz Perthro Algiz Sowilo "
"Tiwaz Berkano Ehwaz Mannaz Laguz Ingwaz Dagaz Othala").split()
SAME_TURNED = {"Gebo", "Hagalaz", "Naudhiz", "Isa", "Jera", "Eihwaz",
"Sowilo", "Ingwaz", "Dagaz"}
if __name__ == "__main__":
p0 = sum(Fraction(comb(15, k) * comb(9, 3 - k), comb(24, 3)) *
Fraction(1, 2 ** k) for k in range(4))
print(" Norns reading with no reversal: {:.4f} (shortcut 0.125, "
"ratio {:.2f})".format(float(p0), float(p0 / Fraction(1, 8))))
for a in range(3):
aett = FUTHARK[8 * a:8 * a + 8]
rev = [r for r in aett if r not in SAME_TURNED]
print(" aett {} ({:7s}): {} reversible, P(reversed) {:.4f}, "
"share {:.3f}".format(a + 1, aett[0], len(rev),
len(rev) / 16, len(rev) / 15))
sym = [r for r in FUTHARK if r in SAME_TURNED]
print(" symmetric runes: {}".format(", ".join(sym[:5])))
print(" {}".format(", ".join(sym[5:])))
# Norns reading with no reversal: 0.3197 (shortcut 0.125, ratio 2.56)
# aett 1 (Fehu ): 7 reversible, P(reversed) 0.4375, share 0.467
# aett 2 (Hagalaz): 2 reversible, P(reversed) 0.1250, share 0.133
# aett 3 (Tiwaz ): 6 reversible, P(reversed) 0.3750, share 0.400
# symmetric runes: Gebo, Hagalaz, Naudhiz, Isa, Jera
# Eihwaz, Sowilo, Ingwaz, DagazFirst, a Norns reading with no reversal, often read as a “clear path”, has probability 0.3197, 2.56 times what the shortcut assumes; nearly a third of all three-rune readings look “clear”. Second, reversals are not spread evenly over the three aettir. In the first aett (Fehu’s) 7 of 8 runes are reversible and a drawn rune is reversed with probability 0.4375; in the second (Hagalaz’s) only 2 of 8, probability 0.1250; in the third (Tiwaz’s) 6, probability 0.3750. Third, of all reversed runes, 0.467 come from the first aett and only 0.133 from the second, and the nine symmetric runes never appear reversed at all. A reader who notices that “difficult” second-aett runes rarely come up reversed is noticing their shapes.
How much depends on the set
Rune sets are not standard. Some add a blank stone, popularised in modern divination, which cannot be reversed; makers draw some runes so that 14 or 16 of them, rather than 15, look different upside down.
Calculation
# Chapter 6. p12: How much the answer depends on the rune set. Modern
# sets differ: some add a blank rune (25 stones, the blank never
# reversed), and makers draw some runes so that 14 or 16 of them look
# different turned over. For each variant: the mean number reversed in a
# Norns reading of 3, the chance of none, and the readings a sceptic
# would need to tell the exact law from the shortcut binomial (3, 1/2)
# by the mean, two-sided alpha 0.05, power 0.80.
import math
from fractions import Fraction
from math import comb
from statistics import NormalDist
def law(n, N, rev):
p = [Fraction(0)] * (n + 1)
for k in range(max(0, n - (N - rev)), min(n, rev) + 1):
pk = Fraction(comb(rev, k) * comb(N - rev, n - k), comb(N, n))
for r in range(k + 1):
p[r] += pk * Fraction(comb(k, r), 2 ** k)
return p
if __name__ == "__main__":
z = NormalDist().inv_cdf(0.975) + NormalDist().inv_cdf(0.80)
print(" stones reversible mean P(none) readings to tell")
for N, rev in ((24, 15), (24, 14), (24, 16), (25, 15)):
p = law(3, N, rev)
mean = float(sum(r * x for r, x in enumerate(p)))
var = float(sum(r * r * x for r, x in enumerate(p))) - mean ** 2
n = math.ceil((z * math.sqrt(var) / (1.5 - mean)) ** 2)
print(" {:6d} {:9d} {:.4f} {:.4f} {:8d}".format(
N, rev, mean, float(p[0]), n))
# stones reversible mean P(none) readings to tell
# 24 15 0.9375 0.3197 16
# 24 14 0.8750 0.3498 13
# 24 16 1.0000 0.2915 21
# 25 15 0.9000 0.3378 14With 15 reversible runes of 24, the Norns mean is 0.9375 and the chance of no reversal 0.3197. With 14 the mean drops to 0.8750 and no reversal rises to 0.3498; with 16, 1.0000 and 0.2915; with a blank stone added to 15 reversible runes, 0.9000 and 0.3378. Every variant is far from the shortcut’s 1.5 and 0.125. A sceptic comparing the mean number of reversals with the shortcut needs only 13 to 21 recorded Norns readings to tell them apart: the shortcut is not a subtle error.
Hypothesis · not established
The author’s interpretation (not yet tested): readers who interpret reversals report more “blocked” or “difficult” themes for spreads drawn from the first aett than from the second, simply because first-aett runes are reversed more often. What would show it wrong: in recorded readings, the share of “difficult” interpretations does not differ between spreads dominated by first-aett and by second-aett runes.
Summary
A Norns reading shows no reversal with probability 0.3197, 2.56 times the shortcut. Reversals come mostly from the first and third aettir, because most second-aett runes are symmetric. Variants of the rune set move the numbers a little (mean 0.8750 to 1.0000), and none of them comes close to the shortcut.
Fan, Y. (2026). Almanac Auspicious Boolean Lattices, Jianchu Twelve-Officer Congruences and Runic Combinatorics: Constraint Satisfaction Paradoxes, Solar-Term Phase Jumps and Elder Futhark Inversion Distributions [黄历择吉布尔格、建除十二直同余与卢恩符文组合学] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090239 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).