MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

There are certain things in mathematics that have caused me a pleasant surprise -- when some part of mathematics is brought to bear in a fundamental way on another, where the connection between the two is unexpected. The first example that comes to my mind is the proof by Furstenberg and Katznelson of Szemeredi's theorem on the existence of arbitrarily long arithmetic progressions in a set of integers which has positive upper Banach density, but using ergodic theory. Of course in the years since then, this idea has now become enshrined and may no longer be viewed as surprising, but it certainly was when it was first devised.

Another unexpected connection was when Kolmogorov used Shannon's notion of probabilistic entropy as an important invariant in dynamical systems.

So, what other surprising connections are there out there?

share|cite|improve this question
It should be mentioned that the connection you refer to is due to Furstenberg ( Later Furstenberg and Katznelson together used this connection to derive other combinatorial results, including a multidimensional extension of Szemeredi's theorem and a density version of the Hales-Jewett's theorem. – Joel Moreira Aug 23 '15 at 17:46

77 Answers 77

Although it is not all that spectacular, since it does not really relate two different fields of mathematics, it has always been surprising to me that the gradient flow equation for the Chern-Simons functional on a (closed, oriented) 3-manifold $Y$ turns out to be the ASD(=Yang-Mills) equation on the cylinder $Y\times\mathbb{R}$.

The next thing is not really a connection, but definitely one of my favorite surprises in mathematics. By the work of Michael Freedman, the classification of closed, oriented, simply-connected topological 4-manifolds is basically equivalent to the classification of unimodular, symmetric bilinear forms (uSBFs) over the integers. As nice as this is, it comes with the grain of salt that the classification of uSBFs is not an easy task. Specifically, the classification of definite uSBFs is a hard problem and far from being solved.

And now comes the surprise: Simon Donaldson tells us that if we look at smooth 4-manifolds, then the only definite uSBFs that can occur are the trivial ones!

share|cite|improve this answer

I am always impressed how countability conditions and topological properties interact, like in the following cases.

Assume there is a topological group $P$ which is, as an abstract group, isomorphic to a direct product of groups $G$ and $H$. Assume all groups to be Hausdorff and locally compact. Then $P$ is isomorphic as a topological group to $G\times H$ in the product topology if $G$ and $H$ are sigmacompact.

And another example: Every non-discrete locally compact totally disconnected group has uncountable cardinality.

share|cite|improve this answer
Abel, you'd better rule out discrete groups in your last sentence. – Todd Trimble Jul 25 '11 at 17:08
That's right of course. Thank you. The usual mistake of neglecting the trivial case... – Abel Stolz Aug 17 '11 at 12:28

Definitely not a pure interplay between two subfields of math, but the omnipresence of quaternions in physics is stunning. I even figured out a few years ago that writing down the equivalent of Cauchy-Riemann equations for functions of a quaternionic variable in a matricial form and multiplying this matrix on the left by the metric tensor of special relativity directly gives rise to continuity equation and wave equation, which is truly intriguing and amazingly beautiful.

share|cite|improve this answer

Classification of symmetric spaces by using classification of simple lie algebras due to cartan is my favourite one!

share|cite|improve this answer

The homotopy hypothesis, namely that two concepts, one that of a (weak) $n$-groupoid arising in higher category theory, and the other that of (the homotopy type of) a topological space $X$ with $\pi_i\left(X\right)=0$ for all $i > n,$ are the same thing.

share|cite|improve this answer
Weak homotopy type. – Todd Trimble Mar 26 '15 at 14:16
Or homotopy type of a CW-complex... – David Roberts Apr 12 '15 at 9:45

A recent connection between macroeconomics and electrodynamics by Maldacena: see Appendix here that derives Maxwell equations from the condition that currency exchange banks don't go bust (in a certain made-up but not too unreasonable currency exchange setup).

share|cite|improve this answer


  1. The Langrange polynomial interpolation formula;
  2. The Chinese Remainder Theorem.


  1. Let $\ K\ $ be a field of characteristic $\ 0.\ $ Let $\ \phi:A\rightarrow K\ $ be an arbitrary function, where $\ A\ $ is a non-empty finite subset of $\ K.\ $ Then there exists an exactly polynomial $\ f:K\rightarrow K\ $ of degree $\ n < |A|,\ $ such that $\ \forall_{x\in A}\ f(x)=\phi(x)$.
  2. Let $\ A\ $ be a nonempty finite set of positive integers such that $\ \gcd(a\ b)=1\ $ for every two different $\ a\ b\in A.\ $ Let $\ \phi:A\rightarrow\mathbb Z\ $ be arbitrary. Then there exists $\ f\in\mathbb Z\ $ such that $\ \forall_{a\in A}\ f\equiv \phi(a) \mod a.\ $ The integer $\ f\ $ is unique in the following sense: $$\ \forall_{a\in A}\ g\equiv\phi(a)\!\!\!\mod a\quad \Rightarrow\quad g\equiv f\!\!\!\mod \prod A$$

Crucial special (basic) cases:

  1. There exists exactly one polynomial $\ f_b:K\rightarrow K\ $ of degree $\ n < |A|,\ $ such that $\ f_b(b)=1,\ $ and $\ \forall_{x\in A\setminus\{b\}}\ f_b(x)=0\,\ $ for every $\ b\in A\ $ (actually $\ \deg(f)=|A|-1$).
  2. There exista exactly one integer $\ f_b\!\!\mod\prod A\ $ such that $\ f_b\equiv 1\!\!\mod b,\ $ and $\ \forall_{a\in B\setminus\{b\}}\ f_b\equiv 0\!\!\mod a\ $ for every $\ b\in A$.

Once we get the basic elements $\ f_b,\ $ then $\ f\ $ is uniquely obtained as the respective linear combination of elements $\ f_b\ $ both in the Lagrange and in the Chinese cases.

Construction (of the basis elements):

  1. $\ \forall_{t\in K}\ f_b(t):=\frac{L_b(t)}{L_b(b)},\ $ where $\ L_b(t):=\prod_{a\in A\setminus\{b\}}\ (t-a)$
  2. $\ f_b\ := C_b\cdot d_b,\ $ where $\ C_b:=\prod_{a\in A\setminus\{b\}} a,\ $ and $\ d_b\cdot C_b\equiv 1\mod b$.

We see that $\ C_b\ $ corresponds to $\ L_b,\ $ and $\ d_b\ $ to $\ \frac 1{L_b(b)}$.

This connection, when generalized, unites the algebraic number theory and the theory of algebraic functions.

share|cite|improve this answer
I don't see why this is surprising. – Todd Trimble Jan 13 '15 at 14:46
@ToddTrimble -- what about the topological dimension and the fixed point property? (I am just curious, see above--and you're welcome to down-vote it too, fair is fair, it should be a two-way street). – Włodzimierz Holsztyński Jan 13 '15 at 20:03
I don't have a real opinion on that since I don't know what you are alluding to there. Offhand, it sounded interesting. – Todd Trimble Jan 13 '15 at 20:18
@ToddTrimble -- (me and sophisticated alluding? :-); there is my answer in the same surprising connection thread just a couple places above this answer on which we are commenting right now (well, I am more diverting than commenting). – Włodzimierz Holsztyński Jan 13 '15 at 20:34
You mean here: Yeah, I knew which answer you meant, but I still don't know what that connection is about. Sorry I can't respond more intelligently. Perhaps you could explain just a bit more there? – Todd Trimble Jan 13 '15 at 20:42

Using the Chinese remainder theorem for proving Gödel's incompleteness theorems.

share|cite|improve this answer
Why the downvotes? – Timothy Chow Nov 8 '13 at 16:13
@TimothyChow -- I didn't down-vote. Nevertheless I feel that the Chinese here is quite arbitrary in the Gödel's context, and relatively very trivial. One could use other schemes as well with the same success. – Włodzimierz Holsztyński Jan 13 '15 at 2:18
@WłodzimierzHolsztyński : What you say is true, but when exponentiation is not directly available, the options for a simple encoding are limited. – Timothy Chow Jan 13 '15 at 18:09

The connection between 'Electric-magnetic duality in string theory' and Langlands Program. See e.g. Witten-Kapustin. Not exactly a connection between two mathematics areas, but I think it nevertheless partially qualifies.

share|cite|improve this answer

The existence of Nash equilibria is an example that connects elementary aspects of game theory, probability, geometry, and algebraic topology.

share|cite|improve this answer

In topology, the connection between the fixed point property and the topological dimension (the covering dimension)--the following two theorems are equivalent:

  • the cube $\ I^n\ $ has the fixed point property;
  • there exists a normal topological space $\ X\ $ such that $\ \dim X\ \ge\ n$.

for every $\ n=0\ 1\ ...$

The connection is my notion of the universal function (or universal morphism in general). The beginning of the story is:

THEOREM $\ \dim X \ge n\ \Leftrightarrow\ \exists\ universal\,\ f:X\rightarrow I^n\ $ (for every completely regular space $\ X$).

share|cite|improve this answer
@ToddTrimble -- done! :-) – Włodzimierz Holsztyński Jan 13 '15 at 23:01
Thanks. I have to keep reminding myself what it means for $f$ to be "universal" (for every $g: X \to I^n$ there exists $x \in X$ such that $f(x) = g(x)$). By the way, just so you know: even if you type @name, the intended recipient won't receive notification unless he/she commented before (or was the one who posted). I just happened to notice you made an edit, so I took a look and saw you tried to reach me. – Todd Trimble Jan 13 '15 at 23:28
@ToddTrimble -- yes, about the system of notifications, thank you for the info. And this time indeed, the system reacted to at-T immediately, expanded it to your full name. – Włodzimierz Holsztyński Jan 14 '15 at 0:14

Another point that hasn't been mentioned yet: To prove the nonexistence of scissors congruences, one typically uses algebraic $K$-theory.

share|cite|improve this answer
Can you more explain or give some references?Thank you. – Ali Taghavi Feb 3 at 11:05
There is a nice book by Johan Dupont. The connection goes via group cohomology, as far as I recall. – Sebastian Goette Feb 3 at 16:13

I have been studying a paper recently called "Pointwise Fourier Inversion: a Wave Equation Approach" by Mark Pinsky and Michael Taylor. Even though Fourier analysis and PDEs have close connections, a particular connection I like is that the solution to the standard wave equation \begin{equation*} (\partial^2_t-\Delta)u=0 \end{equation*} with initial velocity $0$ and initial position $f(x)$ is used to establish pointwise convergence of the partial Fourier Integrals \begin{equation*} S_Rf(x)=\frac{1}{(2\pi)^n}\int_{|\xi|\leq R}\hat{f}(\xi)e^{ix\cdot\xi} \end{equation*} in $L^2$ as $R\to\infty$ so that we have "concrete" statements to deal with rather than arbitrary information and distributions.

share|cite|improve this answer

I also guess the links between differential geometry and geometric analysis on one hand and algebraic topology on the other were rather surprising when they were found.

  1. The Pontryagin-Thom construction. Smooth cobordism is described by homotopy groups of the Thom spectrum (which seems to have forgotten the smooth structures entirely). On the way, it gives one one of the most geometric motivations to study homotopy theory and spectra.

  2. The Atiyah-Singer index theorem allows one to guess the dimension of solution/moduli spaces of (sometimes even nonlinear) partial differential equations using characteristic classes that do not involve any hard analysis at all. Because the topological formula for the index also has a geometric interpretation, one gets applications to curvature questions in Riemannian geometry as a bonus. The surprise continues when one compares the different proofs of this theorem using either abstract $K$-theory (Atiyah-Singer) or the heat equation (Atiyah-Bott-Patodi, Getzler, Bismut, and others) or the geometry and representation theory of Lie groups (Berline-Vergne). The combination of these methods is still leading to new insights not only in differential topology.

share|cite|improve this answer

The Selberg trace formula, relating chaotic geodesic motion on a compact space of negative curvature with the eigenvalues of the Laplacian operator on that space. (someone mentioned trace formulas already, but from a different perspective)

share|cite|improve this answer

I'd like to share the very elementary fact (so elementary that I found surprising only after I taught Calculus course) that all the elementary functions are analytic in the global way. Of course, that's no surprise for polynomials. But I found no intuition why the trigonometric functions and the exponential functions, in the way they are originally considered by human, turn out to be equal to their Taylor expansions everywhere. Consider again the fact that Taylor expansion uses only the information on an infinitesimal neighborhood at a point, a function which is not originally defined by power series should be of extremely little chance to equal its Taylor expansion. I don't know if I'm right, but I finally told my students this is really a miracle.

share|cite|improve this answer
You are completely correct, that this is in some way an astonishing thing. The downvotes are an expression of the absence of this astonishment from the official account of things... So, by accident, it is not surprising that you'd get downvotes. But I think you are perfectly correct... – paul garrett Feb 4 at 0:32

That the Jeffreys' prior in the pole-zero parameterization of a transfer function is the hyperbolic transfinite diameter of the support of the poles and zeros.

It's my favorite because I just discovered it last month. I like laughing at my own jokes.

share|cite|improve this answer

protected by François G. Dorais Sep 30 '13 at 0:52

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.