关于「东西方原典考辨:《宋史·选举志》的初考、覆考与详定,博尔达的等差计分」,《不存在完美裁决器》给出的定量结论是什么?

结论:A 段用博尔达本人的二十一名选举人。\(s\) 为 1 时,相当于每人只投「最不想要谁」的反对票,C 以 20 分胜出;\(s\) 为 2 就是博尔达计分,C 以 26 分胜出;\(s\) 为 3 与 6 时 C 仍然领先,但优势逐步缩小,\(s\) 为 6 时三人分别是 48、49、50 分;\(s\) 恰好为 7 时三人都是 56 分,完全打平;\(s\) 为 8 时 A 以 64 分反超。也就是说,只要首位的分量足够大,按名次计分就会退化成相对多数,重新选出那个两两对决全输的 A。

数据口径:多道检查意见不一时,经营者最常见的想法是:只要合成规则设计得足够聪明,总能得到一个公正、一致、无可挑剔的裁决。 阿罗 1950 年与 1951 年证明,选项不少于三个时,不存在同时满足无限制域、弱帕累托、无关选项独立与非独裁的排序合成规则。

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: 20, 26, 48, 49, 50, 56

A. Borda's 21 electors under points (0, 1, s)
s1: A 8 B 14 C 20 winner C
s2: A 16 B 21 C 26 winner C
s3: A 24 B 28 C 32 winner C
s6: A 48 B 49 C 50 winner C
s7: A 56 B 56 C 56 winner ABC
s8: A 64 B 63 C 62 winner A
s20: A 160 B 147 C 134 winner A
B. 216 profiles: winner set differs from Borda's
s1: 96 of 216 (0.4444)
s3/2: 36 of 216 (0.1667)
s3: 18 of 216 (0.0833)
s7: 18 of 216 (0.0833)
s100: 18 of 216 (0.0833)
查看完整验算程序
# Chapter 7. p72: Borda's equal steps versus other point scales.
# Borda gave last, middle and first place a, a+b, a+2b: equal steps.
# Any a and any b > 0 rank options alike, so the scale is (0, 1, 2) up
# to an affine map. Here we let the top step grow: points (0, 1, s).
# s = 1 is "vote against one" (anti-plurality), s = 2 is Borda, large s
# approaches plurality. Part A: the 21-elector profile of p31. Part B:
# over all 216 profiles of 3 reviewers, how often the winner set under
# (0, 1, s) differs from Borda's. Exact Fraction arithmetic.
import itertools
from fractions import Fraction

PROFILE = [(1, "ABC"), (7, "ACB"), (7, "BCA"), (6, "CBA")]
OPTS = "ABC"


def score(prof, s):
    pts = (s, Fraction(1), Fraction(0))
    out = {c: Fraction(0) for c in OPTS}
    for k, o in prof:
        for pos, c in enumerate(o):
            out[c] += k * pts[pos]
    return out


def winners(prof, s):
    sc = score(prof, s)
    top = max(sc.values())
    return "".join(c for c in OPTS if sc[c] == top)


if __name__ == "__main__":
    print("A. Borda's 21 electors under points (0, 1, s)")
    for s in (1, 2, 3, 6, 7, 8, 20):
        sc = score(PROFILE, Fraction(s))
        print("   s = {:>2}: A {:>3}  B {:>3}  C {:>3}  winner {}".format(
            s, str(sc["A"]), str(sc["B"]), str(sc["C"]),
            winners(PROFILE, Fraction(s))))
    print("B. 216 profiles: winner set differs from Borda's")
    orders = ["".join(p) for p in itertools.permutations(OPTS)]
    profs = [[(1, o) for o in pr]
             for pr in itertools.product(orders, repeat=3)]
    base = [winners(p, Fraction(2)) for p in profs]
    for s in (1, Fraction(3, 2), 3, 7, 100):
        k = len([1 for p, b in zip(profs, base)
                 if winners(p, Fraction(s)) != b])
        print("   s = {:>4}: {:>3} of 216  ({:.4f})".format(
            str(s), k, k / 216))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《阿罗不可能定理与多评审规则聚合、五件套拒绝码冲突时为何不存在完美裁决器母典》第 7 章「东西方原典考辨:《宋史·选举志》的初考、覆考与详定,博尔达的等差计分」(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-310 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026310,
  author    = {Fan, Yafu},
  title     = {Arrow's Impossibility Theorem and the Aggregation of Multiple Reviewers: Why No Perfect Arbiter Exists When Kit Refusal Codes Conflict},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论