关于「句长从哪里切:切分规则、均值与中位数」,《中文可读性与字频句长》给出的定量结论是什么?

结论:三种规则得到的单位数分别是 5、6 与 16 个。规则甲的平均句长是 26.0 字,中位数 20.0 字,最长一句 69 字;规则乙把那句六十九字的长句在分号处一分为二,平均句长降到 21.7 字,中位数反而升到 22.5 字,最长一句变为 42 字;规则丙按小句切,平均长度只有 8.1 字,中位数 9.0 字,最长的小句 16 字,没有一个单位超过三十字。

数据口径:作者自拟的四篇小样本中,四字短句堆僻字的文本平均句长最短,按本书的示意难度公式却最难,得分 19.15;同样短句、全用常用字的文本只有 0.45。 书中参数为示意;全部计算由书中程序生成;本书属于非个人化经营建议。

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: 16, 20.0, 21.7, 22.5, 26.0, 42

paragraph: 130 hanzi
ruleunitsmeanmedianmaxlong>30
A stops526.020.0691
B +semicolon621.722.5421
C +comma168.19.0160
rule B units: 12 4 25 42 27 20
rule A units: 12 4 25 69 20
查看完整验算程序
# Chapter 3. p31: the sentence length you get depends on where you cut.
# Rule A cuts only at the full stop, question and exclamation marks;
# rule B also at the semicolon (the book's rule); rule C also at the
# comma and the enumeration comma (clauses). Sample paragraph is the
# author's own. Length = hanzi per unit; long = more than 30 hanzi.
import re
import statistics

HAN = re.compile(r"[一-鿿]")
RULES = (("A stops", re.compile(r"[。!?!?\n]+")),
         ("B +semicolon", re.compile(r"[。!?;!?;\n]+")),
         ("C +comma", re.compile(r"[。!?;,、!?;,\n]+")))
PARA = ("很多人以为句子越短越好读。其实不然。读者卡住的地"
        "方,往往不是句子的长度,而是一个不认识的字。一句"
        "话哪怕有四十个字,只要每个字都眼熟、每个词都顺口"
        "、意思一层一层往下走,读者也能一口气读完;反过来"
        ",四个字的短句里只要藏着一个生僻字,读者就得停下"
        "来猜。所以,判断一段文字难不难,要同时看句长和字"
        "频。")


def lengths(text, stop):
    out = []
    for part in stop.split(text):
        n = len(HAN.findall(part))
        if n:
            out.append(n)
    return out


if __name__ == "__main__":
    print("paragraph: {} hanzi".format(len(HAN.findall(PARA))))
    print("rule           units   mean  median  max  long>30")
    for name, stop in RULES:
        ls = lengths(PARA, stop)
        print("{:<13s}  {:5d}  {:5.1f}  {:6.1f}  {:3d}  {:7d}".format(
            name, len(ls), len(HAN.findall(PARA)) / len(ls),
            statistics.median(ls), max(ls),
            len([x for x in ls if x > 30])))
    print("rule B units: " + " ".join(str(x) for x in
                                      lengths(PARA, RULES[1][1])))
    print("rule A units: " + " ".join(str(x) for x in
                                      lengths(PARA, RULES[0][1])))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《中文可读性公式与字频句长、落地页与专著段落的阅读难度母典》第 3 章「句长从哪里切:切分规则、均值与中位数」(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-323 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026323,
  author    = {Fan, Yafu},
  title     = {Chinese Readability from Character Frequency and Sentence Length: Reading Difficulty of Landing Pages and Monograph Paragraphs},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论