关于「转换点的信息量:惊异度、词长与信息均匀」,《嵌词阈值与单语等效》给出的定量结论是什么?

结论:迂回表达总长 16 个音节,峰值 2.67 比特,方差只有 0.10,没有任何音节超出容量。嵌入词把总长压到 12 个音节,总信息量 35.00 比特,略低于迂回表达的 36.00 比特,可峰值升到 7.50,方差升到 4.20,超出容量的部分合计 5.00 比特。平方代价(信息均匀理论惩罚峰值的一种写法)从 82.67 升到 152.50。

数据口径:嵌词收益只在读者共享双语词库时成立:把单语表达、嵌入与嵌入加注释写成覆盖率的三条成本线,示意参数下 kiasu 在书面渠道的覆盖率阈值为 0.844;读者不发问时,失败会变成沉默的误读;各地区分页面值得嵌的词,合成混合页面后可能全部跌破阈值。 书中参数为示意,全部计算由书中程序生成,不对应任何真实客户、对话或成交;马来语、闽南语、粤语、英语、华语一律平等对待。

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: 0.10, 12, 152.50, 16, 2.67, 35.00

renderingsylbitsmeanvarpeaksq-costover
circumlocution1636.002.2500.102.6782.670.00
insert1235.002.9174.207.50152.505.00
insert+gloss1637.002.3124.257.50153.505.00
查看完整验算程序
# Chapter 2. p22: how three renderings spread the same meaning in time.
# Ten context syllables carry 2 bits each. The gap concept needs 14 bits.
# Circumlocution spreads 14 bits plus 2 bits of function syllables over
# six syllables. A known inserted word packs 14 bits plus a 1-bit switch
# cue into two syllables. Insertion plus a four-syllable gloss repeats
# the meaning, so the gloss syllables carry 0.5 bits each for a listener
# who knew the word. Capacity 5 bits per syllable is illustrative; the
# squared cost follows the uniform-information-density idea (Jaeger 2010).
import math

CONTEXT = [2.0] * 10
CAP = 5.0


def profile(kind):
    if kind == "circumlocution":
        gap = [16.0 / 6.0] * 6
    elif kind == "insert":
        gap = [7.5, 7.5]
    else:
        gap = [7.5, 7.5] + [0.5] * 4
    return CONTEXT[:5] + gap + CONTEXT[5:]


if __name__ == "__main__":
    print("rendering        syl   bits   mean  var    peak  sq-cost  over")
    for kind in ("circumlocution", "insert", "insert+gloss"):
        x = profile(kind)
        n = len(x)
        tot = math.fsum(x)
        mean = tot / n
        var = math.fsum((v - mean) ** 2 for v in x) / n
        sq = math.fsum(v * v for v in x)
        over = math.fsum(max(0.0, v - CAP) for v in x)
        print("{:<15} {:>4} {:>6.2f} {:>6.3f} {:>5.2f} {:>6.2f} {:>8.2f}"
              " {:>5.2f}".format(kind, n, tot, mean, var, max(x), sq, over))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《多语参杂语码转换信息论、词汇空缺补偿与单语种等效精准表达母典》第 2 章「转换点的信息量:惊异度、词长与信息均匀」(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-381 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026381,
  author    = {Fan, Yafu},
  title     = {Information Theory of Code-Switching, Lexical-Gap Compensation and Monolingual Precision Equivalence in Multilingual Markets},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论