What does Spirit Writing, Ideomotor Action and Text Attribution conclude, in numbers, about “Classics East and West: Xu Dishan's Study of Spirit Writing and Faraday's Table-Turning Experiments”?

Answer: Chapter 7 examines Xu Dishan's 1941 study of spirit writing and Faraday's reports and turns them into an out-of-vocabulary test and a sealed-question test.

Scope of the numbers: Its parameters are illustrative and its revealed texts are synthetic demonstrations written for it; all calculations are generated by its programs. Chapter 5 demonstrates three attribution methods on a synthetic corpus, character-bigram cosine, Burrows's Delta and naive Bayes; all six revealed texts are attributed correctly, while the naive-Bayes log odds of 8.92 to 21.37 are systematically too large.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python only

medium knows K ranks of 6000; slip rate 0.002
Koutside-rate qchars Ncut cq / slip
15000.149450274.7
25000.094470247.2
35000.0581150329.0
45000.0310330415.5
Show the full program
# Chapter 7. p71: Xu Dishan's thesis as an out-of-vocabulary test
# Xu Dishan (1941) argued, in substance, that what the pen writes does
# not go beyond what the people holding it know. As a test: character
# use follows a Zipf law over 6000 ranks (exponent 1). The medium knows
# the top K ranks. If the text came from the medium, characters outside
# those K appear only through slips, at rate 0.002 per character; if it
# came from a writer who knows all 6000, they appear at the Zipf mass
# beyond rank K. We find the shortest text (N characters) and the cut c
# such that P(X >= c | medium) <= 0.01 and P(X >= c | outside) >= 0.99.
# K, the slip rate and the exponent are illustrative.
# 模型:许地山命题写成超出词表率的二项检验;参数为示意。
import math

V, SLIP = 6000, 0.002


def tail_beyond(k):
    w = [1.0 / r for r in range(1, V + 1)]
    return math.fsum(w[k:]) / math.fsum(w)


def upper(n, p, c):
    # P(X >= c) for X ~ Binomial(n, p)
    terms = [math.comb(n, x) * p ** x * (1.0 - p) ** (n - x)
             for x in range(c)]
    return max(0.0, 1.0 - math.fsum(terms))


def shortest(q):
    n = 10
    while True:
        for c in range(1, n + 1):
            if upper(n, SLIP, c) <= 0.01:
                if upper(n, q, c) >= 0.99:
                    return n, c
                break
        n += 10


if __name__ == "__main__":
    print("medium knows K ranks of {}; slip rate {}".format(V, SLIP))
    print("     K   outside-rate q   chars N   cut c   q / slip")
    for k in (1500, 2500, 3500, 4500):
        q = tail_beyond(k)
        n, c = shortest(q)
        print("  {:4d}   {:14.4f}   {:7d}   {:5d}   {:8.1f}".format(
            k, q, n, c, q / SLIP))

Single source and citation

Fan, Y. (2026). Ideomotor Micro-Movements in Spirit Writing, Random-Walk Stylus Trajectories and Stylometric Attribution of Revealed Texts (Version 1.0.0), Chapter 7: Classics East and West: Xu Dishan's Study of Spirit Writing and Faraday's Table-Turning Experiments. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-226 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026226,
  author    = {Fan, Yafu},
  title     = {Ideomotor Micro-Movements in Spirit Writing, Random-Walk Stylus Trajectories and Stylometric Attribution of Revealed Texts},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book