Bernhard Riemann

⚡ Key Takeaways

Bernhard Riemann (1826–1866) was a German mathematician whose ideas reshaped analysis, geometry, and number theory. One line: he introduced rigorous integration, founded modern differential geometry through curved manifolds, and posed the Riemann Hypothesis — arguably mathematics' deepest open problem. Applies in: complex analysis, geometry, general relativity, analytic number theory. Who/When: Georg Friedrich Bernhard Riemann, active at Göttingen, ~1851–1866.

💡 The Core Insight

Riemann's unifying genius was geometric thinking applied to analysis. Where his predecessors manipulated formulas, Riemann asked what space is this object living on? This shift — from algebraic expression to underlying manifold or surface — turned multivalued functions into single-valued maps on surfaces, turned the parallel postulate into a question about curvature, and turned prime counting into the geometry of zeros in the complex plane.

DefinitionThe Riemann Integral

For a bounded function $f:[a,b]\to\mathbb{R}$ and a partition $P=\{a=x_0<x_1<\cdots<x_n=b\}$ with sample points $\xi_i\in[x_{i-1},x_i]$ , the Riemann sum is $S(P,f)=\sum_{i=1}^n f(\xi_i)\,\Delta x_i$ where $\Delta x_i = x_i-x_{i-1}$ . The function is Riemann integrable if the limit below exists independent of choices, as the mesh $\|P\|=\max_i \Delta x_i \to 0$ .

1

\quad \int_a^b f(x)\,dx = \lim_{\|P\|\to 0}\sum_{i=1}^n f(\xi_i)\,\Delta x_i

Derivation hint: bound upper and lower sums $U(P,f)=\sum \sup_{[x_{i-1},x_i]}f\,\Delta x_i$ and $L(P,f)=\sum\inf f\,\Delta x_i$ ; integrability is $\inf_P U = \sup_P L$ .

DefinitionThe Riemann Zeta Function

For complex $s$ with $\operatorname{Re}(s)>1$ , the zeta function is defined by a series equal to an Euler product over primes $p$ :

2

\quad \zeta(s)=\sum_{n=1}^\infty \frac{1}{n^s}=\prod_{p\text{ prime}}\frac{1}{1-p^{-s}}

Derivation hint: expand each factor as a geometric series $\sum_k p^{-ks}$ and multiply; unique factorization of integers reproduces every $n^{-s}$ exactly once. Riemann analytically continued $\zeta$ to all of $\mathbb{C}\setminus\{1\}$ and proved the functional equation relating $s$ to $1-s$ :

3

\quad \zeta(s)=2^s\pi^{s-1}\sin\!\Big(\frac{\pi s}{2}\Big)\,\Gamma(1-s)\,\zeta(1-s)

Derivation hint: use the integral representation $\Gamma(s/2)\pi^{-s/2}\zeta(s)=\int_0^\infty \psi(x)x^{s/2-1}dx$ where $\psi$ is a theta sum, then apply the Jacobi theta transformation $\theta(1/x)=\sqrt{x}\,\theta(x)$ .

DefinitionRiemannian Metric and Curvature

A Riemannian manifold is a smooth manifold $M$ equipped with a smoothly varying inner product (metric tensor) $g_{ij}$ on each tangent space. Lengths and angles are computed by the line element:

4

\quad ds^2 = \sum_{i,j} g_{ij}(x)\,dx^i\,dx^j

Derivation hint: generalize the Pythagorean $ds^2=dx^2+dy^2$ by allowing position-dependent coefficients that encode how the space is curved. The intrinsic curvature is captured by the Riemann curvature tensor $R^l_{\ ijk}$ , built from the metric's Christoffel symbols $\Gamma$ :

5

\quad R^l_{\ ijk}=\partial_j\Gamma^l_{ik}-\partial_k\Gamma^l_{ij}+\Gamma^m_{ik}\Gamma^l_{mj}-\Gamma^m_{ij}\Gamma^l_{mk}

Derivation hint: measure the failure of second covariant derivatives to commute, $[\nabla_j,\nabla_k]V^l = R^l_{\ ijk}V^i$ — parallel transport around a small loop detects curvature.

TheoremCauchy–Riemann Equations and Conformality

A complex function $f=u+iv$ is holomorphic on an open set iff $u,v$ have continuous partials satisfying

6

\quad \frac{\partial u}{\partial x}=\frac{\partial v}{\partial y},\qquad \frac{\partial u}{\partial y}=-\frac{\partial v}{\partial x}

Derivation hint: require the difference quotient $\lim_{h\to0}(f(z+h)-f(z))/h$ to be independent of the direction of $h\in\mathbb{C}$ ; equating real and along-imaginary limits yields both equations. Holomorphic maps with nonzero derivative are conformal (angle-preserving), the analytic foundation of Riemann surface theory.

ProofSketch: The Riemann Mapping Theorem

Statement. Any simply connected open proper subset $U\subsetneq\mathbb{C}$ is biholomorphic to the open unit disk $\mathbb{D}$ . 1. Reduce to a bounded family. Fix $z_0\in U$ . Consider the family $\mathcal{F}$ of injective holomorphic maps $g:U\to\mathbb{D}$ with $g(z_0)=0$ and $g'(z_0)>0$ . Show $\mathcal{F}$ is nonempty: since $U\ne\mathbb{C}$ , pick $a\notin U$ , take a branch of $\sqrt{z-a}$ (possible since $U$ is simply connected), and compose with a Möbius map into $\mathbb{D}$ . 2. Maximize the derivative. Let $M=\sup_{g\in\mathcal F}g'(z_0)$ . The family is uniformly bounded (maps into $\mathbb{D}$ ), hence normal by Montel's theorem: every sequence has a locally uniformly convergent subsequence. 3. Extract the extremal map. Take $g_n$ with $g_n'(z_0)\to M$ ; a subsequence converges to $f$ with $f'(z_0)=M$ . Hurwitz's theorem keeps $f$ injective; $f$ maps into $\overline{\mathbb D}$ and the maximum modulus principle forces it into $\mathbb D$ . 4. Surjectivity by contradiction. If $f$ missed some $w\in\mathbb{D}$ , an explicit square-root-and-Möbius construction produces $\tilde f\in\mathcal F$ with $\tilde f'(z_0)>M$ , contradicting maximality. Hence $f$ is the desired biholomorphism. $\blacksquare$

∎

⚠️ Common mistake

Riemann's original 1851 proof used the Dirichlet principle — asserting that an energy-minimizing function exists. Weierstrass exposed a gap: minimization over an infinite-dimensional space need not attain its infimum. The theorem is true, but Riemann's justification was incomplete; Hilbert later rehabilitated the Dirichlet principle. This is a textbook case of correct intuition outrunning available rigor.

mathematician · Georg Friedrich Bernhard Riemann | 1826–1866

Born in Breselenz, Kingdom of Hanover, to a Lutheran pastor, Riemann was shy, deeply religious, and chronically anxious about money and health. He entered Göttingen to study theology at his father's wish but switched to mathematics with permission. He studied under Carl Friedrich Gauss and Dirichlet, absorbing the highest standards of the era. His 1851 doctoral thesis on complex functions, his 1854 habilitation lecture on geometry (delivered to a committee including an aged Gauss, who was reportedly astonished), and his 1859 prime-counting memoir form the spine of modern mathematics. Plagued by tuberculosis, he sought relief in Italy's climate and died at Lake Maggiore aged 39. His housekeeper reportedly discarded unpublished papers — a permanent loss to mathematics.

🧠 Intuition

Picture three of Riemann's leaps as the same move applied in different rooms:

Integration: instead of asking for a formula, chop the domain finely and add up rectangles — the area is whatever the sums converge to. Geometry over algebra.
Riemann surfaces: a function like $\sqrt z$ is "two-valued" only because we forced it onto a flat plane. Build the right surface — two sheets stitched along a branch cut — and $\sqrt z$ becomes a clean single-valued map. The multivaluedness was the plane's fault, not the function's.
Curvature: you don't need to step outside a surface to know it's curved. Walk a triangle and measure whether the angles sum to more or less than $180°$ . Riemann made this intrinsic — curvature lives inside the metric, no embedding required.

The connecting idea: the right space makes hard things easy.

Figure

A two-sheeted Riemann surface for √z: walking around the branch point at the origin moves you between sheets.

```svg

```

ExampleComputing a Riemann integral from the definition

Given: $f(x)=x^2$ on $[0,3]$ . Use a uniform partition into $n$ equal subintervals with right endpoints.

Steps:

Width: $\Delta x = 3/n$ ; sample point $\xi_i = 3i/n$ .
Riemann sum: $S_n=\sum_{i=1}^n (3i/n)^2\cdot(3/n)=\dfrac{27}{n^3}\sum_{i=1}^n i^2$ .
Use $\sum_{i=1}^n i^2=\dfrac{n(n+1)(2n+1)}{6}$ :

S_n=\frac{27}{n^3}\cdot\frac{n(n+1)(2n+1)}{6}=\frac{27(n+1)(2n+1)}{6n^2}.

Take $n\to\infty$ : $\dfrac{27\cdot 2n^2}{6n^2}\to \dfrac{54}{6}=9$ .

Answer: $\boxed{\displaystyle\int_0^3 x^2\,dx = 9}$ (square units). Matches $\frac{x^3}{3}\big|_0^3=9$ .

ExampleCurvature of a sphere from the metric

Given: the unit-radius round sphere has metric $ds^2 = d\theta^2 + \sin^2\theta\,d\phi^2$ , so $g_{\theta\theta}=1,\ g_{\phi\phi}=\sin^2\theta$ . Find the Gaussian (sectional) curvature.

Steps:

Nonzero Christoffel symbols: $\Gamma^\theta_{\phi\phi}=-\sin\theta\cos\theta$ , $\Gamma^\phi_{\theta\phi}=\Gamma^\phi_{\phi\theta}=\cot\theta$ .
The relevant curvature component: $R^\theta_{\ \phi\theta\phi}=\partial_\theta\Gamma^\theta_{\phi\phi}-\Gamma^\phi_{\phi\theta}\Gamma^\theta_{\phi\phi}$

= -\cos2\theta\;-\;\cot\theta\,(-\sin\theta\cos\theta)= -(\cos^2\theta-\sin^2\theta)+\cos^2\theta=\sin^2\theta.

Gaussian curvature $K = R_{\theta\phi\theta\phi}/(g_{\theta\theta}g_{\phi\phi}) = \sin^2\theta/\sin^2\theta = 1$ .

Answer: $\boxed{K=1}$ everywhere (dimensionless / curvature units $1/\text{radius}^2$ ). A unit sphere has constant positive curvature — independent of position, as Gauss's Theorema Egregium anticipated and Riemann generalized.

⚠️ Common mistake

"Riemann proved the Riemann Hypothesis." He did not. In his 1859 paper he conjectured that all nontrivial zeros of $\zeta(s)$ have real part $1/2$ , remarking he could not prove it and setting it aside as not essential to his immediate goal. It remains unproven. Likewise, Riemann did not invent the integral concept from nothing — he made rigorous and general what Cauchy and others had handled for continuous functions.

Counterexample

Not every bounded function is Riemann integrable. The Dirichlet function

D(x)=\begin{cases}1 & x\in\mathbb{Q}\\ 0 & x\notin\mathbb{Q}\end{cases}

on $[0,1]$ has upper sum $U(P,D)=1$ and lower sum $L(P,D)=0$ for every partition, since each subinterval contains both rationals and irrationals. So $\inf U=1\neq 0=\sup L$ and the Riemann integral does not exist. (It is Lebesgue integrable, with integral $0$ — a key motivation for the later Lebesgue Integral.)

connection

Riemann's work is a web binding much of modern mathematics:

The Riemann Integral is generalized and repaired by the Lebesgue Integral, which handles the pathological functions Riemann's theory rejects.
Riemann Surfaces fuse Complex Analysis with Topology and Algebraic Geometry; the genus of a surface controls the dimension of holomorphic forms (Riemann–Roch).
Riemannian Geometry (metric + curvature tensor) is the exact mathematical language of General Relativity, where $g_{ij}$ becomes the gravitational field.
The Riemann Zeta Function and Riemann Hypothesis tie the Prime Number Theorem to complex analysis via the explicit formula linking primes to zeta zeros.
The Cauchy–Riemann Equations connect holomorphicity to Harmonic Functions and potential theory.

🔬 Where this appears

⏳ History

- 1826 — Born September 17 in Breselenz, Hanover, second of six children of a Lutheran minister.

- 1846–47 — Enters University of Göttingen (theology, then mathematics); attends Gauss's lectures, moves to Berlin to study under Dirichlet, Jacobi, Eisenstein.

- 1851 — Doctoral dissertation under Gauss: Grundlagen für eine allgemeine Theorie der Functionen — introduces Riemann surfaces and the mapping theorem.

- 1854 — Habilitation lecture Über die Hypothesen, welche der Geometrie zu Grunde liegen ("On the Hypotheses which lie at the Foundations of Geometry"); founds Riemannian geometry. Gauss is deeply impressed.

- 1857 — Major paper on Abelian functions; appointed extraordinary professor at Göttingen.

- 1859 — Elected to the Berlin Academy; submits Über die Anzahl der Primzahlen unter einer gegebenen Größe — the prime-counting memoir containing the functional equation and the Hypothesis.

- 1859 — Succeeds Dirichlet as full professor at Göttingen.

- 1862 — Marries Elise Koch; falls seriously ill with tuberculosis shortly after.

- 1866 — Dies July 20 at Selasca, Lake Maggiore, Italy, aged 39. His collected works, edited posthumously, sustain decades of research.

open-problem

The Riemann Hypothesis — that every nontrivial zero of $\zeta(s)$ satisfies $\operatorname{Re}(s)=\tfrac12$ — is the most consequential unsolved problem in mathematics, a Clay Millennium Prize Problem ( $1{,}000{,}000$ ). Billions of zeros have been verified numerically on the critical line, and deep partial results exist (a positive proportion of zeros lie on the line; zero-free regions near $\operatorname{Re}(s)=1$ give the Prime Number Theorem). Yet no proof or disproof is known. A resolution would sharpen our understanding of the distribution of primes from "approximately" to "essentially optimally." Adjacent open frontiers include the generalized and grand Riemann hypotheses, the Hilbert–Pólya conjecture (seeking a self-adjoint operator whose eigenvalues are the zeros), and connections to random matrix theory (Montgomery–Odlyzko pair-correlation statistics).

Formal Statement of Riemann's Principal Contributions

Why It Holds — The Riemann Mapping Idea

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also