关于「东西方原典考辨:孔子的益者三友与西塞罗的几斗盐」,《邓巴分层与私域半衰期》给出的定量结论是什么?
结论:孔子的「益者三友,损者三友」可以落到一个可测的量上,即获客方式不同的两批客户的半衰期;示意参数下,诚实批次在第 40.7 天反超活跃人数,长期累计是奉承批次的 1.875 倍。「不可则止」与「朋友数,斯疏矣」对应停止即停与频率上限。西塞罗把友谊收束到「少数人之间」,与分层结构的最内层一致;他的「好几斗盐」换算下来是 1746 个共餐小时,远多于现代估计的 200 小时,但方向相同。最后一章把全书的计算变成一份可以执行、也可以被推翻的方案。
数据口径:书中的层级下限、半衰期、权重与停止概率均为示意参数;全部计算由书中程序生成;本书是非个人化经营建议,合规规则以各地现行法规为准。 书中以标注框区分已发表的证据、可运行的 Python 计算(19 段程序)、原典与作者的假说。
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: 1.875, 40.7
| day | active honest | active flattery | cum honest | cum flat |
|---|---|---|---|---|
| 0 | 1000.0 | 1600.0 | 0 | 0 |
| 30 | 840.9 | 951.4 | 27545 | 37431 |
| 90 | 594.6 | 336.4 | 70184 | 72922 |
| 180 | 353.6 | 70.7 | 111915 | 88252 |
| 365 | 121.4 | 2.9 | 152099 | 92167 |
| 730 | 14.7 | 0.0 | 170570 | 92332 |
| honest overtakes in active count at day 40.7 | ||||
| honest overtakes in customer-days at day 100.2 | ||||
| long-run customer-days ratio honest/flattery 1.875 |
查看完整验算程序
# Chapter 7. p71: the Analects' three helpful and three harmful friends
# as two ways of recruiting private-channel customers (illustrative).
# "Straight, sincere, well-informed" recruitment (honest advice) brings
# 1000 customers with a half-life of 120 days. "Ingratiating, soft,
# glib" recruitment (flattery and pressure) brings 1600 customers with
# a half-life of 40 days. Active count N0 * exp(-lam * t); cumulative
# active customer-days N0 / lam * (1 - exp(-lam * t)). The program
# finds when the honest cohort overtakes in active count and in
# cumulative customer-days (bisection).
import math
COHORTS = [("honest", 1000.0, 120.0), ("flattery", 1600.0, 40.0)]
def active(n0, hl, t):
return n0 * math.exp(-math.log(2.0) / hl * t)
def cum(n0, hl, t):
lam = math.log(2.0) / hl
return n0 / lam * (1.0 - math.exp(-lam * t))
def cross(f, lo, hi):
for _ in range(100):
mid = 0.5 * (lo + hi)
if f(mid) < 0:
lo = mid
else:
hi = mid
return hi
def main():
(_, n1, h1), (_, n2, h2) = COHORTS
print("day active honest active flattery cum honest cum flat")
for t in (0, 30, 90, 180, 365, 730):
print("{:3d} {:14.1f} {:16.1f} {:11.0f} {:9.0f}".format(
t, active(n1, h1, t), active(n2, h2, t),
cum(n1, h1, t), cum(n2, h2, t)))
t_a = cross(lambda t: active(n1, h1, t) - active(n2, h2, t),
0.0, 730.0)
t_c = cross(lambda t: cum(n1, h1, t) - cum(n2, h2, t), 1.0, 730.0)
print("honest overtakes in active count at day {:.1f}".format(t_a))
print("honest overtakes in customer-days at day {:.1f}".format(t_c))
print("long-run customer-days ratio honest/flattery {:.3f}".format(
(n1 * h1) / (n2 * h2)))
main()唯一数据源与引用
范雅福(Fan, Y.). (2026). 《邓巴数分层、关系维护时间预算与私域客户衰减半衰期母典》第 7 章「东西方原典考辨:孔子的益者三友与西塞罗的几斗盐」(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-231 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026231,
author = {Fan, Yafu},
title = {Dunbar's Layered Numbers, Relationship-Maintenance Time Budgets and the Decay Half-Life of Private-Channel Customers},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}