Euler's Identity and the Grammar of Mathematical Beauty

Euler's Identity and the Grammar of Mathematical Beauty

1.1 The Equation at the Center of Mathematics

There is one equation that appears, with suspicious unanimity, on the shortlists of every poll asking mathematicians to name the most beautiful result in their discipline: $e^{i\pi} + 1 = 0$. The Physics World poll of 1990 placed it first; readers of *The Mathematical Intelligenzer* rank it consistently at the top; essayists from Bertrand Russell to Richard Feynman have singled it out; a 1990 episode of *The Simpsons* embedded it as a punchline that an entire demographic was expected to recognize. Whatever else divides mathematicians, this equation is a shared inheritance. The first task of this book is therefore not to celebrate the identity but to dissect it. We want to know what work the word "beautiful" is doing when we apply it to this particular line of algebra, and whether that work can be made technical, operational, and transferable to other results. If we can answer those questions for Euler's identity, we will have a grammar of mathematical beauty we can apply throughout the rest of this volume.

The identity is, on its surface, the simplest possible kind of statement: an equation, with a single equals sign, asserting that two arithmetic expressions evaluate to the same number. The right-hand side is $0$. The left-hand side is $e^{i\pi} + 1$. The drama lives entirely in the constants that appear inside that left-hand side. Five of them gather there: $0$ and $1$, the additive and multiplicative identities; $\pi$, the ratio of circumference to diameter in Euclidean geometry; $e$, the base of the natural logarithm and the limit of $(1 + 1/n)^n$; and $i$, the imaginary unit, a number whose square is $-1$. Three of these constants — $0$, $1$, and $\pi$ — were known to the ancient Greeks. The other two — $e$ and $i$ — are products of the seventeenth and eighteenth centuries. The claim of the identity is that these five numbers, drawn from arithmetic, geometry, analysis, and the algebra of complex numbers, are not independent. They satisfy a single linear relation in which each appears with coefficient $\pm 1$. The economy of the relation is striking. There are no free parameters, no summations, no approximations. Just five symbols and one equality.

The pedagogical problem we need to confront, however, is that the very features that make the identity famous — its brevity, its compression, its air of revelation — also make it pedagogically opaque. A reader encountering $e^{i\pi} + 1 = 0$ for the first time has no internal mechanism by which to verify it, no obvious intuition for why it should be true, and no natural bridge from any prior mathematics they know. The equation looks like a magic trick whose method has been hidden. Our first technical task is to remove the magic by exposing the derivation. Once the derivation is in hand, we will be in a position to ask the harder question: granted that the derivation is straightforward, why does the result feel like more than the sum of its derivational steps? What is the additional content that makes the conclusion feel inevitable, deep, and beautiful rather than merely correct?

1.2 Deriving the Identity from First Principles

The cleanest derivation of Euler's identity proceeds through Taylor series, and we will follow it in full technical detail. The reason we go through every step rather than simply asserting the result is that the derivation itself contains the structural facts that ground the aesthetic experience. Anyone who has internalized the derivation can never again be surprised by the identity in quite the same way; the surprise migrates from the conclusion to the premises. The premises are where the real content lives.

We begin with the exponential function $e^x$, defined as the sum of its Taylor series. For real $x$, we have

$$e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!} = 1 + x + \frac{x^2}{2!} + \frac{x^3}{3!} + \frac{x^4}{4!} + \frac{x^5}{5!} + \cdots.$$

This series converges absolutely for every real $x$, and the function it defines is smooth, strictly positive, and equal to its own derivative. The radius of convergence is infinite.

The first move toward Euler's identity is to observe that the same Taylor series formula, viewed as a formal power series, makes sense whenever $x$ is replaced by any quantity for which powers and addition are defined. In particular, if we let $x = i\theta$ for a real number $\theta$, we obtain

$$e^{i\theta} = \sum_{n=0}^{\infty} \frac{(i\theta)^n}{n!} = 1 + i\theta + \frac{(i\theta)^2}{2!} + \frac{(i\theta)^3}{3!} + \frac{(i\theta)^4}{4!} + \frac{(i\theta)^5}{5!} + \cdots.$$

Now we use the periodicity of $i^n$. Since $i^2 = -1$, we have $i^3 = -i$, $i^4 = 1$, $i^5 = i$, and in general $i^n$ cycles through the four values $1, i, -1, -i$ as $n$ runs through $0, 1, 2, 3, \ldots$. Grouping even-indexed and odd-indexed terms separately, the series becomes

$$e^{i\theta} = \sum_{k=0}^{\infty} \frac{(i\theta)^{2k}}{(2k)!} + \sum_{k=0}^{\infty} \frac{(i\theta)^{2k+1}}{(2k+1)!}.$$

The even-indexed terms contain $(i\theta)^{2k} = (i^2)^k \theta^{2k} = (-1)^k \theta^{2k}$, while the odd-indexed terms contain $(i\theta)^{2k+1} = i \cdot (-1)^k \theta^{2k+1}$. Pulling the factor of $i$ out of the second sum gives

$$e^{i\theta} = \sum_{k=0}^{\infty} \frac{(-1)^k \theta^{2k}}{(2k)!} + i \sum_{k=0}^{\infty} \frac{(-1)^k \theta^{2k+1}}{(2k+1)!}.$$

These two series are precisely the Taylor series for $\cos\theta$ and $\sin\theta$, respectively:

$$\cos\theta = \sum_{k=0}^{\infty} \frac{(-1)^k \theta^{2k}}{(2k)!} = 1 - \frac{\theta^2}{2!} + \frac{\theta^4}{4!} - \cdots,$$

$$\sin\theta = \sum_{k=0}^{\infty} \frac{(-1)^k \theta^{2k+1}}{(2k+1)!} = \theta - \frac{\theta^3}{3!} + \frac{\theta^5}{5!} - \cdots.$$

Therefore

$$e^{i\theta} = \cos\theta + i \sin\theta.$$

This is Euler's formula, the parent identity from which $e^{i\pi} + 1 = 0$ descends by a single substitution. Setting $\theta = \pi$, we have $\sin\pi = 0$ and $\cos\pi = -1$, so

$$e^{i\pi} = \cos\pi + i \sin\pi = -1 + i \cdot 0 = -1.$$

Adding $1$ to both sides yields $e^{i\pi} + 1 = 0$.

The derivation, fully unpacked, takes perhaps three lines of technical content. The reader who has never seen it before may need a paragraph to absorb the trick of substituting $i\theta$ for $x$ and separating even and odd terms. The reader who has seen it before can reconstruct it in under a minute. The derivation is short. Yet something in the experience of working through it feels structurally disproportionate to its length, and this disproportion is the first datum we must accommodate in any theory of mathematical beauty.

For the geometric series cousin of this identity, the story is similarly short and similarly informative. The geometric series

$$\frac{1}{1 + x} = \sum_{n=0}^{\infty} (-1)^n x^n = 1 - x + x^2 - x^3 + \cdots$$

converges for $|x| < 1$. Substituting $x = i$, which has $|i| = 1$, takes us outside the radius of convergence and the series no longer converges in any ordinary sense. But the right-hand side function $1/(1+x)$ is still defined and is analytic except at the pole $x = -1$, and one can verify by formal rearrangement that the same series, viewed as an asymptotic or Borel-summed object, equals $1/(1+i) = (1-i)/2$. This observation is the seed of analytic continuation: a function defined by a convergent series on one region can be extended, often uniquely, to a much larger domain. Euler's identity is in some sense a particular instance of analytic continuation, in which the function $e^x$ — defined as a real power series — extends naturally to a complex function with the property that $e^{i\pi} = -1$. The fact that the extension agrees with the value $-1$ at the point $i\pi$ is what gives the identity its bite.

The Mermaid flowchart below summarizes the logical chain that connects the real exponential function to Euler's identity, with the path of analytic continuation made explici