What does Error Bars in Dating conclude, in numbers, about “Why Porcelain Carries No Carbon Clock: Materials, Sampling and the Limits of Method”?

Answer: Chapter 5 matches materials to clocks and shows that thermoluminescence cleanly separates new copies from old pieces but can hardly separate two periods a century apart: at a relative error of 0.25, the ninety-five per cent range excludes the adjacent period with probability only 0.109.

Scope of the numbers: Objects, prices and parameters are illustrative, the calibration curve is a hand-set illustrative line rather than IntCal data, and all calculations are generated by the book's programs. Chapter 6 writes testing as a problem of expected loss; under illustrative parameters a buyer should not buy untested below a prior of 0.655, and a test is worth most at middling priors.

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 onlyprinted values that match the answer: 0.109, 0.25

rP(30 a reads >200)P(500 a reads <200)P(600 excl 500)
0.0894.04e-1013.70e-250.535
0.155.78e-375.03e-100.228
0.251.62e-141.24e-040.109
0.41.05e-061.10e-020.066
composite: P(at least one drill hits new part)
new share fk=1k=2k=3
0.100.1000.1900.271
0.250.2500.4380.578
0.500.5000.7500.875
Show the full program
# Chapter 5. p52: what a TL test on porcelain can and cannot separate.
# A measured age is modelled as true age times exp(e), e ~ N(0, r),
# so r is the relative 1-sigma error (0.089 is the excavated budget
# of p42; 0.25 stands for a low-sensitivity porcelain sample with no
# burial context; both illustrative). Part 1: chance that a modern
# piece (30 a) reads older than a 200 a cut-off, and that an old
# piece reads younger. Part 2: chance that the 95% interval of a
# 600 a piece excludes 500 a (telling two adjacent periods apart).
# Part 3: a composite of old and new material is caught only if a
# drill point lands on new material: P = 1 - (1 - f)^k.
import math


def phi(z):
    return 0.5 * (1.0 + math.erf(z / math.sqrt(2.0)))


def p_above(true, cut, r):
    z = math.log(cut / true) / r
    return 0.5 * math.erfc(z / math.sqrt(2.0))


if __name__ == "__main__":
    cut = 200.0
    print("  r      P(30 a reads >200)  P(500 a reads <200)  "
          "P(600 excl 500)")
    for r in (0.089, 0.15, 0.25, 0.40):
        fa = p_above(30.0, cut, r)
        miss = 0.5 * math.erfc(-math.log(cut / 500.0) / r / math.sqrt(2.0))
        sep = 1.0 - phi((math.log(500.0 / 600.0) + 1.96 * r) / r)
        print("  {:<5}  {:>18.2e}  {:>19.2e}  {:>15.3f}".format(
            r, fa, miss, sep))
    print("composite: P(at least one drill hits new part)")
    print("  new share f   k=1     k=2     k=3")
    for f in (0.10, 0.25, 0.50):
        ps = [1.0 - (1.0 - f) ** k for k in (1, 2, 3)]
        print("  {:<11.2f}   {:.3f}   {:.3f}   {:.3f}".format(f, *ps))

Single source and citation

Fan, Y. (2026). Error Bars in Thermoluminescence and Radiometric Dating: Multiple Solutions on Calibration Plateaus and When to Send Antiques for Testing (Version 1.0.0), Chapter 5: Why Porcelain Carries No Carbon Clock: Materials, Sampling and the Limits of Method. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-331 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026331,
  author    = {Fan, Yafu},
  title     = {Error Bars in Thermoluminescence and Radiometric Dating: Multiple Solutions on Calibration Plateaus and When to Send Antiques for Testing},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book