关于「字辈诗是同余码:漏代、多代与轮回撞名」,《族谱合龙与义庄永续》给出的定量结论是什么?

结论:第四章把字辈诗当同余码,不重字之诗单边漏代全部检出,有重字时降到 0.9500;八字诗用完从头再用,跨轮同名占 0.517。

数据口径:按字辈起名,同代人只剩一字可变,示意参数下一代 11 人就有一半机会撞名;两本 60 人对 40 人之支谱按全名硬接,接对者只占 0.208。 书中数据全为合成或示意,不含任何真实在世者资料;全部计算由书中程序生成;本书是非个人化分析,不构成法律、理财或经营建议,也不做任何血统高下或族群比较。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.9500

A. share of single-edge copying errors caught
   error           D20      R20      D8
   dup (s=-1)     1.0000   0.9500   1.0000
   skip 1 (s=+1)  1.0000   0.9500   1.0000
   skip 2 (s=+2)  1.0000   1.0000   1.0000
   any |s|<=3     1.0000   0.9833   1.0000
B. offsets between two island books that agree mod L
   within +-40 generations
   D20: 5 candidate offsets [-40, -20, 0, 20, 40]
   R20: 5 candidate offsets [-40, -20, 0, 20, 40]
   D8: 11 candidate offsets [-40, -32, -24, -16, -8, 0, 8, 16, 24, 32, 40]
C. offsets left by a dated anchor (25 years per generation)
   +-30 years: D20 [0]; R20 [0]; D8 [0]
   +-250 years: D20 [0]; R20 [0]; D8 [-8, 0, 8]
查看完整验算程序
# Chapter 4. p41: the generation poem as a congruence check.
# Model: a poem of L characters is used in order; generation g takes
# the character at position g mod L, so a son's position must equal his
# father's plus 1 mod L. A copying error that drops k generations, or
# repeats one, shifts the son's apparent step from 1 to 1 + s. It is
# caught iff poem[(g + 1 + s) mod L] != poem[(g + 1) mod L]. Poems are
# coded as integers (synthetic): D20 has 20 distinct characters, R20
# repeats four of them, D8 is a short poem of 8 distinct characters.
# Detection is averaged over every position g and every shift s.
from fractions import Fraction

POEMS = {"D20": list(range(20)),
         "R20": list(range(16)) + [3, 7, 11, 0],
         "D8": list(range(8))}
SHIFTS = {"dup (s=-1)": [-1], "skip 1 (s=+1)": [1], "skip 2 (s=+2)": [2],
          "any |s|<=3": [-3, -2, -1, 1, 2, 3]}


def rate(poem, shifts):
    n = len(poem)
    hit, tot = 0, 0
    for g in range(n):
        for s in shifts:
            tot += 1
            if poem[(g + 1 + s) % n] != poem[(g + 1) % n]:
                hit += 1
    return Fraction(hit, tot)


def offsets(n, span):
    return [d for d in range(-span, span + 1) if d % n == 0]


if __name__ == "__main__":
    print("A. share of single-edge copying errors caught")
    print("   error           D20      R20      D8")
    for label, sh in SHIFTS.items():
        print("   {:<14} {:>6.4f}   {:>6.4f}   {:>6.4f}".format(
            label, *[float(rate(POEMS[k], sh)) for k in POEMS]))
    print("B. offsets between two island books that agree mod L")
    print("   within +-40 generations")
    for k in POEMS:
        n = len(POEMS[k])
        print("   {}: {} candidate offsets {}".format(
            k, len(offsets(n, 40)), offsets(n, 40)))
    per_gen = 25
    print("C. offsets left by a dated anchor ({} years per generation)"
          .format(per_gen))
    for tol in (30, 250):
        row = []
        for k in POEMS:
            n = len(POEMS[k])
            row.append("{} {}".format(k, [d for d in offsets(n, 40)
                                         if abs(d * per_gen) <= tol]))
        print("   +-{} years: {}".format(tol, "; ".join(row)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《宗族世系有向无环图合龙算法、字辈同余编码与范氏义庄跨代治理母典》第 4 章「字辈诗是同余码:漏代、多代与轮回撞名」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-334 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026334,
  author    = {Fan, Yafu},
  title     = {Genealogical DAG Merge Algorithms, Generation-Poem Congruence Coding and the Institutional Economics of the Fan Clan Charitable Estate},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论