关于「承诺可信度:观众成本、守约记录与重复关系」,《施为言语与聚点定框》给出的定量结论是什么?
结论:第五章讨论观众成本与守约记录,一次失约让二十次记录的下界损失 0.074。
数据口径:书中对话全部合成,参数为示意,全部计算由书中程序生成;本书是非个人化经营建议,不是法律意见;分润只写单层、只给直接推荐人。 第一章用奥斯汀与塞尔的施为句理论,把报价拆成六个适切条件:示意模型里结构化报价把适切概率从 0.305 提到 0.734,一份完全可信的报价让对方期望熵下降 0.918 比特。
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: 0.074
| kept / promised | mean | 5% lower bound |
| 5 / 5 0.857 0.607 | ||
| 10 / 10 | 0.917 | 0.762 |
| 9 / 10 | 0.833 | 0.636 |
| 20 / 20 | 0.955 | 0.867 |
| 19 / 20 | 0.909 | 0.793 |
| 18 / 20 | 0.864 | 0.729 |
| 40 / 40 | 0.976 | 0.930 |
| 38 / 40 | 0.929 | 0.854 |
| one break after 20 kept costs 0.074 of the lower bound | ||
查看完整验算程序
# Chapter 5. p52: a record of kept promises as a Beta posterior.
# Prior Beta(1, 1) on the keep rate. After k kept out of n promises the
# posterior is Beta(1 + k, 1 + n - k). For integer parameters the CDF
# is a binomial tail, I_x(a, b) = P(Bin(a + b - 1, x) >= a), summed with
# math.fsum; the 5 % lower bound is found by bisection. It shows what a
# counterpart can safely rely on. The asymmetry: one broken promise
# moves the lower bound more than several extra kept ones. Synthetic.
import math
def cdf(x, a, b):
n = a + b - 1
return math.fsum(math.comb(n, j) * x ** j * (1 - x) ** (n - j)
for j in range(a, n + 1))
def lower5(a, b):
lo, hi = 0.0, 1.0
for _ in range(60):
mid = (lo + hi) / 2
if cdf(mid, a, b) < 0.05:
lo = mid
else:
hi = mid
return lo
print("kept / promised mean 5% lower bound")
for k, n in ((5, 5), (10, 10), (9, 10), (20, 20), (19, 20), (18, 20),
(40, 40), (38, 40)):
a, b = 1 + k, 1 + n - k
print(" %2d / %2d %.3f %.3f" % (k, n, a / (a + b),
lower5(a, b)))
print("one break after 20 kept costs %.3f of the lower bound" % (
lower5(21, 1) - lower5(20, 2)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《施为性言语行为、谢林聚点定框与高维语言坍缩商业现实母典》第 5 章「承诺可信度:观众成本、守约记录与重复关系」(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-428 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026428,
author = {Fan, Yafu},
title = {Performative Speech Acts, Schelling Focal-Point Framing and Verbal Reality-Collapse in High-Stakes Negotiation},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}