Section 3: On the Possibility of Solving Problems by Straight Edge and Compasses — Notes

Historical, Mathematical & Cross-Repository Commentary (pp. 104–112)

Historical, Mathematical & GEOMETOR Commentary

1. Petersen’s Algebra of Constructibility & Hilbert’s Chapter VII

Petersen was equally distinguished in algebra (Theorie der algebraischen Gleichungen (1878)) and graph theory (where the Petersen graph, 1898, and Petersen’s theorem on 3-regular bridgeless graphs bear his name). His Appendix III (pp. 97–102 / 104–112) bridges field theory and synthetic geometry:

  • Building on Pierre Laurent Wantzel's 1837 theorem (Liouville's Journal) and Petersen’s own 1871 dissertation (Om Ligninger, der løses ved Kvadratrod), Petersen translates tower degree \([\mathbb{F} : \mathbb{Q}] = 2^k\) into curve-order criteria (Theorems 1–4, pp. 98–99), allowing students to prove geometric impossibility directly from locus degree without setting up cubic polynomials!

  • Compare Hilbert’s Grundlagen, Chapter VII (§§ 36–39, 1899) (geometor/hilbert): where Petersen characterizes full straightedge + compass via arbitrary real quadratic radical towers, Hilbert refines the criterion to straightedge + segment-transferer (Eichmass), isolating totally real Pythagorean radical towers \(\sqrt{\alpha^2 + \beta^2}\)!

2. The 1879 Pre-Lindemann Witness on \(\pi\) (p. 101)

Petersen’s remark on page 101 of our 1879 edition is a jewel of nineteenth-century mathematical history:

  • In 1873, Charles Hermite had proved the transcendence of \(e\).

  • In 1879, when Petersen penned and Haagensen translated “which, as yet, nobody has been able to prove” for \(\pi\), squaring the circle remained the single surviving unsolved classical Greek construction problem!

  • Just three years later (June 1882), Ferdinand von Lindemann (Über die Zahl π, Math. Ann. 20) proved \(e^{i\pi} + 1 = 0\) impossible for algebraic \(\pi \neq 0\), establishing that \(\pi\) is transcendental—and a fortiori outside Petersen’s quadratic radical towers.