She wrote under a dead student’s name so the mathematics would be read, and gave the corpus the equation under its own Chladni figures — where a coincidence between sums of two squares becomes a family of shapes in sand.
Sophie Germain taught herself mathematics from her father's library during the Terror, when she was not permitted to leave the house, and later from lecture notes borrowed under a dead student's name — Antoine-Auguste Le Blanc, who had left the École Polytechnique. She submitted work under that name to Lagrange, who asked to meet the author. She wrote to Gauss under it too, for three years, and revealed herself only when Napoleon's army reached Braunschweig and she feared for his safety.
Both stories are usually told for their pathos. They are worth telling for something else: she chose the pseudonym because the mathematics would be read, and dropped it when a life was at stake. The instrument and the judgement were the same faculty. Gauss wrote back that a woman who overcomes the obstacles to understanding number theory “must have the noblest courage, quite extraordinary talents and superior genius” — and then, when Göttingen offered her an honorary degree in 1831, she had been dead a month. CITED
She belongs in this gallery twice over, and the two bricks are unrelated to each other, which is itself unusual. One is in number theory. The other is in the theory of vibrating surfaces, and it is the one this corpus stands on.
Call a prime $p$ a Germain prime when $2p+1$ is also prime. The first twenty are
2, 3, 5, 11, 23, 29, 41, 53, 83, 89, 113, 131, 173, 179, 191, 233, 239, 251, 281, 293
and there are 190 below $10^4$, 1 171 below $10^5$, 7 746 below $10^6$. COMPUTED Whether there are infinitely many is open.
Her theorem, communicated to Legendre and printed in his 1808 supplement, is the first general result on Fermat's Last Theorem that is not a single exponent:
If $p$ is an odd prime and $2p+1$ is prime, then $x^p + y^p = z^p$ has no integer solution with $p \nmid xyz$.
One sentence, and it disposes of the first case for infinitely many exponents at once if the Germain primes are infinite — and unconditionally for every one of them known. Legendre extended the auxiliary-prime method she invented; the method, not the particular primes, is the contribution. Her fuller programme, recovered from the manuscripts in the 1990s and 2000s, was more ambitious than the corollary she is remembered for. CITED
In 1808 Chladni came to Paris and scattered sand on brass plates. Bowed at the edge, the sand fled the moving regions and piled on the still ones, and the plate showed its nodal lines as a figure. Napoleon set a prize: give the mathematical theory of elastic surfaces and account for Chladni's figures. Nobody entered but Germain. She was the sole entrant three times, in 1811, 1813 and 1816, and won on the third attempt.
Her first two attempts were rejected because the variational derivation was wrong in its handling of the fourth-order terms — Lagrange corrected it, and the corrected equation is the one that carries her name. It is a biharmonic:
$$D\,\nabla^4 w \;+\; \rho h\,\frac{\partial^2 w}{\partial t^2} \;=\; 0,\qquad \nabla^4 = \frac{\partial^4}{\partial x^4} + 2\frac{\partial^4}{\partial x^2 \partial y^2} + \frac{\partial^4}{\partial y^4}$$
The prize committee noted that the derivation still rested on assumptions it could not verify. The corpus's own standard requires saying so: the equation is right and her 1816 derivation of it was not complete, and the modern derivation from three-dimensional elasticity is Kirchhoff's, in 1850. CITED OPEN
Here is the brick, and it is the reason this chapter exists rather than being a biography. Take the simply supported square plate. Its modes are $w_{mn} = \sin(m\pi x/a)\,\sin(n\pi y/a)$ and its frequencies go as $m^2 + n^2$. Two different modes can therefore ring at exactly the same pitch:
m²+n² = 50 ← (1,7) and (5,5)
m²+n² = 65 ← (1,8) and (4,7)
m²+n² = 85 ← (2,9) and (6,7)
m²+n² = 125 ← (2,11) and (5,10)
m²+n² = 130 ← (3,11) and (7,9) COMPUTED
These are not the trivial pairs $(m,n)$ and $(n,m)$, which the square's symmetry makes inevitable. $50 = 1^2+7^2 = 5^2+5^2$ is a genuine coincidence of sums of two squares, and it is a fact about the integers that leaks into a piece of brass.
When an eigenvalue is degenerate, any combination $w = \cos t\,\phi_1 + \sin t\,\phi_2$ is a mode at the same frequency. So the plate does not have a figure at that pitch. It has a one-parameter family of them, and which one the sand shows depends on where the plate is clamped and where it is bowed.
The boundary does not draw the figure. The boundary removes options, and what is left is the family the constraint could not exclude. This is the corpus's own claim about form — ortogênese, form generated under constraint in the directions the constraint leaves open — stated on a square of brass in 1816 with a number-theoretic coincidence doing the work at the degenerate pitches.
And it is a threshold, not a scale: nothing about the figure changes gradually as the bowing point moves. It holds, and then at a degenerate frequency the whole family becomes available at once. SHOWN
The corpus reaches this from the other side. dm³ Soundworks works the Chladni figures as resonance; Strang and Faraday hold the neighbouring rungs. None of them names the person who wrote the equation. That is the gap this chapter closes.
| Operator | In this chapter | In dm³ |
|---|---|---|
| C | the boundary condition — a square, clamped | compression: the constraint that issues the family |
| K | the driving frequency raised toward a degenerate eigenvalue | approach to $\kappa^*$ |
| F | degeneracy: one pitch, a one-parameter family of nodal curves | the fold — threshold, not scale SHOWN |
| U | the figure the sand actually settles into, selected by where the bow touches | unfolding onto a branch |
Every number on this page is produced by book7/ch-sophie-germain-verify.py. It records in its own closing block what it establishes and what it does not.
S. Germain, Recherches sur la théorie des surfaces élastiques, Paris, 1821.
A.-M. Legendre, Théorie des nombres, 2nd ed., 1808 — supplement, where Germain’s theorem first appears in print.
E. F. F. Chladni, Die Akustik, Leipzig, 1802.
G. Kirchhoff, “Über das Gleichgewicht und die Bewegung einer elastischen Scheibe”, Crelle 40, 1850.
R. Laubenbacher and D. Pengelley, “‘Voici ce que j’ai trouvé’: Sophie Germain’s grand plan to prove Fermat’s Last Theorem’”, Historia Mathematica 37, 2010 — the manuscripts, and how much larger her programme was.
L. Bucciarelli and N. Dworsky, Sophie Germain: An Essay in the History of the Theory of Elasticity, Reidel, 1980.