5D Library › 玛雅卓尔金双模齿轮、神谕十字对合变换与六十四闸门网络渗流 · English
第七章 《德累斯顿抄本》金星表与先天圆图
为什么在这里读这些原典
第一章的历法轮、第四章的闸门轮都有古老的来源。《德累斯顿抄本》是现存四部玛雅抄本之一,其中的金星表与许多 260 日历表把第一章的同余算术推到了更大的周期 (Thompson 1972; Bricker and Bricker 2011)。闸门轮所依据的先天圆图出自北宋邵雍,十八世纪初经由在华耶稣会士白晋传到莱布尼茨手中,被他读成二进制数 (Swetz 2003; Leibniz 1703)。本章核对这两份材料中可以写成计数的部分。
金星表的算术
证据
已核实: 《德累斯顿抄本》第 24 页与第 46 至 50 页的金星表以 584 天为一个会合周期,并把它分成 236、90、250、8 天四段(晨星、隐没、昏星、隐没);五个周期排成一行,表共 13 行,回到出发的卓尔金日 (Thompson 1972; Aveni 2001)。
推导
# Chapter 7. p13: The arithmetic of the Dresden Codex Venus table.
# The table runs 5 Venus synodic periods of 584 days (canonical 236, 90,
# 250 and 8 days: morning star, invisible, evening star, invisible), and
# is laid out with 13 lines so that it closes on the tzolkin. We check
# the coincidences it rests on, and the error of 584 against the mean
# synodic period of Venus (583.92 days).
from math import gcd
def lcm(*xs):
out = 1
for x in xs:
out = out * x // gcd(out, x)
return out
if __name__ == "__main__":
parts = (236, 90, 250, 8)
print(" phases {} sum to {}".format(parts, sum(parts)))
print(" 5 x 584 = {} = 8 x 365 = {}".format(5 * 584, 8 * 365))
print(" lcm(584, 260) = {} = {} Venus periods = {} tzolkin".format(
lcm(584, 260), lcm(584, 260) // 584, lcm(584, 260) // 260))
print(" 13 x 2920 = {} = 146 x 260 = {}".format(13 * 2920, 146 * 260))
big = lcm(584, 365, 260)
print(" lcm(584, 365, 260) = {} = {} Calendar Rounds = {} haab".format(
big, big // 18980, big // 365))
print(" = {} Venus periods = {} tzolkin".format(big // 584,
big // 260))
drift = 584 - 583.92
print(" error of 584 days: {:.2f} day per period, {:.1f} days over"
" 65 periods".format(drift, 65 * drift))
# phases (236, 90, 250, 8) sum to 584
# 5 x 584 = 2920 = 8 x 365 = 2920
# lcm(584, 260) = 37960 = 65 Venus periods = 146 tzolkin
# 13 x 2920 = 37960 = 146 x 260 = 37960
# lcm(584, 365, 260) = 37960 = 2 Calendar Rounds = 104 haab
# = 65 Venus periods = 146 tzolkin
# error of 584 days: 0.08 day per period, 5.2 days over 65 periods四段之和为 584;五个会合周期 天,恰好等于八个哈布年 。表排成 13 行,是为了让总长 天同时是 146 个卓尔金周期:37960 正是 584 与 260 的最小公倍数,即 65 个金星周期。它同时还是 365 的倍数:37960 是 584、365、260 三者的最小公倍数,等于两个历法轮、104 个哈布年。三个周期在这里同时闭合。
584 天比金星平均会合周期 583.92 天长 0.08 天,65 个周期累积 5.2 天。抄本中另有修正方案处理这一累积误差 (Aveni 2001);本书不涉及其细节,只指出它的存在说明编表者清楚地知道 584 是一个取整值。
先天圆图与二进制
原典
莱布尼茨 1703 年论文的标题
Explication de l’Arithmétique Binaire, qui se sert des seuls caractères 0 & 1 … et sur ce qu’elle donne le sens des anciennes figures Chinoises de Fohy
标题所说的「伏羲的古老图形」,就是邵雍的先天方圆图。第四章已经说明,人类图的闸门轮就是这个圆图从第 41 卦起的读法。
推导
# Chapter 7. p14: Shao Yong's circle, Leibniz's binary numbers and the
# "mandala" read as positions. Leibniz (1703) saw that Shao Yong's square
# and circle list the hexagrams as the binary numbers 0..63. We list the
# circle's two half-circles, show that each step changes the number by 1
# (except at the two joins), and count how many neighbouring gates on the
# 64-gate wheel differ in exactly one line.
KW = ((1, 34, 5, 26, 11, 9, 14, 43), (25, 51, 3, 27, 24, 42, 21, 17),
(6, 40, 29, 4, 7, 59, 64, 47), (33, 62, 39, 52, 15, 53, 56, 31),
(12, 16, 8, 23, 2, 20, 35, 45), (44, 32, 48, 18, 46, 57, 50, 28),
(13, 55, 63, 22, 36, 37, 30, 49), (10, 54, 60, 41, 19, 61, 38, 58))
ORDER = (7, 4, 2, 1, 0, 3, 5, 6)
def king_wen(v):
return KW[ORDER.index(v >> 3)][ORDER.index(v & 7)]
if __name__ == "__main__":
values = list(range(63, -1, -1))
circle = values[:32] + values[32:][::-1]
print(" first half (63 down to 32): King Wen " + " ".join(
str(king_wen(v)) for v in circle[:8]) + " ...")
print(" second half (0 up to 31): King Wen " + " ".join(
str(king_wen(v)) for v in circle[32:40]) + " ...")
steps = [abs(circle[i] - circle[(i + 1) % 64]) for i in range(64)]
print(" steps of size 1 round the circle: {} of 64; others: {}".format(
steps.count(1), sorted(s for s in steps if s != 1)))
flips = [bin(circle[i] ^ circle[(i + 1) % 64]).count("1")
for i in range(64)]
print(" neighbours differing in exactly one line: {} of 64".format(
flips.count(1)))
print(" lines changed between neighbours: {}".format(
{k: flips.count(k) for k in sorted(set(flips))}))
gray = [v ^ (v >> 1) for v in range(64)]
print(" a Gray-code wheel would give 64 of 64: {}".format(
all(bin(gray[i] ^ gray[(i + 1) % 64]).count("1") == 1
for i in range(64))))
# first half (63 down to 32): King Wen 1 43 14 34 9 5 26 11 ...
# second half (0 up to 31): King Wen 2 23 8 20 16 35 45 12 ...
# steps of size 1 round the circle: 62 of 64; others: [32, 32]
# neighbours differing in exactly one line: 34 of 64
# lines changed between neighbours: {1: 34, 2: 16, 3: 8, 4: 4, 5: 2}
# a Gray-code wheel would give 64 of 64: True沿先天圆图走一周,相邻两卦的二进制数在 64 步中有 62 步相差 1,只有在两个半圆相接处相差 32。但相邻两卦只改变一爻的情形只有 34 处;改变两爻的有 16 处,三爻 8 处,四爻 4 处,五爻 2 处,因为二进制加一常常引起进位。若把 64 卦排成格雷码,每一步都只改变一爻。所以先天圆图的「连续」是数值上的连续,而不是爻象上的连续;闸门轮继承的是前者。
假说 · 尚未验证
作者的诠释(尚未验证): 人类图选用先天圆图而非文王序或格雷码,是因为先天圆图在黄道上把「数值相邻」与「空间相邻」统一了起来。什么会证明这一读法不当: 若该体系的原始文献明确说明闸门次序另有来源,并与先天圆图的相合只是巧合,这一读法就不成立。
小结
《德累斯顿抄本》的金星表把 584、365、260 三个周期在 37960 天上同时闭合:65 个金星周期、104 个哈布年、146 个卓尔金周期、两个历法轮。先天圆图沿圆周是二进制数的连续计数,相邻两卦只改变一爻的只有 34 处;人类图的闸门轮继承的就是这种数值上的连续。
参考文献
Fan, Y. (2026). Mayan Tzolkin Modular Gears, Oracle-Cross Involutions and Sixty-Four-Gate Network Percolation: Coprime Cycle Synchronization, Antipode-Analog Permutations and Nine-Center Connected Components [玛雅卓尔金双模齿轮、神谕十字对合变换与六十四闸门网络渗流] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090193 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).