Maths SolverThe story you almost always hear about imaginary numbers goes like this: mathematicians wanted to solve $x^2 + 1 = 0$. There was no real solution, so they made one up — called it $i$, declared $i^2 = -1$, and built a whole new number system on top of it.
This story is satisfying. It is also wrong. Mathematicians did not invent $i$ to solve quadratic equations. The reason mathematicians needed $i$ is much more interesting, and it makes complex numbers feel much less arbitrary once you know the story.
The actual history
For most of the history of mathematics, “$x^2 = -1$ has no solution” was not seen as a problem. It was seen as a fact about the world, like “you can’t have a negative number of sheep.” A quadratic equation with a negative discriminant simply had no real-world meaning, and that was that. No one was lying awake worrying about it.
The thing that forced mathematicians to take imaginary numbers seriously was not quadratics. It was cubic equations — specifically, the formula for solving them, derived in the 16th century by Italian algebraists (Cardano, Tartaglia, del Ferro, in a famous and bitter dispute). The cubic formula is roughly the quadratic formula’s much uglier cousin, and it has the same overall shape: an expression involving square roots that gives you the roots of any cubic equation $ax^3 + bx^2 + cx + d = 0$.
Here is the awkward part. The cubic formula sometimes contains square roots of negative numbers even when all three roots of the equation are perfectly ordinary real numbers. This was not what anyone wanted. You could write down a cubic equation with three nice real solutions, plug it into the formula, and the formula would say “to find the roots, take the square root of $-121$ and...” At which point you would normally throw the formula away and say it doesn’t apply.
But Cardano noticed that if you held your nose, treated $\sqrt{-121}$ as if it were a real number, did the algebra anyway, and trusted the imaginary parts to cancel at the end... they did. The cubic formula worked. It just had to pass through imaginary territory to get from the real-number question to the real-number answer.
The pressure that comes from this
This is the moment imaginary numbers became unavoidable. As long as the cubic formula required them on the way to producing real answers to real questions, mathematicians could no longer dismiss them. They might be philosophically uncomfortable, but they had to be at least provisionally accepted, because the formula obviously worked.
This is also why, for a couple of centuries, mathematicians used imaginary numbers without quite knowing what they meant. Euler in the 1700s was already proving identities like $e^{i\pi} = -1$, which by modern standards is a deep statement about the geometry of the complex plane — but Euler did not yet have the geometry. He had formal symbol-pushing rules and the conviction, born of doing several decades of computation, that the rules were consistent.
What changed: the geometric picture
The reason “imaginary” sounds bad is that it suggests “not real” in the everyday sense. Calling them imaginary was Descartes’s idea, and he meant it dismissively. The name stuck even after mathematicians figured out that imaginary numbers are perfectly real in the “these things have a clear meaning and you can compute with them” sense; they just are not real numbers in the technical sense of being on the number line.
The shift came in the early 1800s, when several mathematicians (Wessel, Argand, Gauss) independently realised that complex numbers have a natural geometric interpretation: they are points in a plane. The number $a + bi$ is the point with horizontal coordinate $a$ and vertical coordinate $b$. Adding two complex numbers corresponds to adding the two vectors in the plane. Multiplying by $i$ corresponds to rotating by $90^\circ$ counterclockwise. Multiplying by a general complex number combines a rotation with a stretch.
This single picture transforms imaginary numbers from a piece of symbolic bookkeeping into a tool for doing two-dimensional geometry algebraically. Suddenly $i^2 = -1$ stops being mysterious: rotating $1$ by $90^\circ$ takes you to $i$; rotating again by $90^\circ$ takes you to $-1$. So $i \cdot i$ is just “rotate twice,” which lands at $-1$.
$e^{i\pi} = -1$ becomes equally clean. The function $e^{i\theta}$ takes the real number $\theta$ and gives you a point on the unit circle, $\theta$ radians around from the positive real axis. So $e^{i\pi}$ is “the point on the unit circle $\pi$ radians around,” which is $-1$. The famously deep identity is just “rotate $180^\circ$.”
What complex numbers are useful for
Once you have the geometric picture, complex numbers are the natural language for anything involving rotation, oscillation, or wave behaviour. A few real applications:
- AC electricity. Alternating current oscillates sinusoidally; the “phase” of the oscillation is naturally an angle. Electrical engineers represent voltages and currents as complex numbers (called “phasors”) and use complex arithmetic to handle circuits with capacitors and inductors. Without complex numbers, the same calculations require trigonometric identities so painful that nobody does them.
- Signal processing. The Fourier transform — which decomposes a signal into its constituent frequencies — is naturally written in terms of $e^{i\omega t}$. Almost all audio compression, image compression, and radio communication runs on Fourier-transform algorithms that work on complex numbers throughout.
- Quantum mechanics. The state of a quantum system is described by a complex-valued function. The probabilities of measurement outcomes come from squaring the absolute value of these complex numbers. The famous “wave function” of an electron is a complex object, not a real one.
- Fluid dynamics in 2D. Two-dimensional incompressible flow has a complex-valued representation that makes the analysis of flow patterns much cleaner. This is how aerofoil shapes are studied analytically.
None of these uses were anywhere on the Italian algebraists’ mind in the 1500s. The applications followed the mathematics by several centuries. This is a recurring pattern in mathematics: a piece of formalism gets developed for one reason, and turns out to be the right language for something completely different that nobody had thought about yet.
The thing about “imaginary”
There is a small culture-war about whether the names “imaginary” and “real” should be replaced with something less misleading. Some textbooks use “lateral” for “imaginary,” emphasising the geometric direction rather than any judgement about reality. Most working mathematicians do not bother; the names are historical, everyone knows what they mean, and renaming them would just produce confusion.
The thing to internalise is that the names are nothing more than labels. Imaginary numbers are not less real than negative numbers were when negative numbers were first invented. (Mathematicians spent the 16th century complaining that negative numbers were “absurd” and “fictitious,” for similar reasons.) Both are perfectly meaningful mathematical objects; both have clean geometric interpretations; both are necessary for huge parts of physics and engineering.
What to take away
The right one-line summary of imaginary numbers is something like: “Complex numbers are the natural extension of the real numbers to two dimensions; they were forced on us by the cubic formula in the 1500s and turned out, two centuries later, to be the right language for everything that rotates or oscillates.”
That summary doesn’t mention $x^2 + 1 = 0$ once, and it shouldn’t. The quadratic motivation is a pedagogical convenience — it’s the simplest equation where you can see that the real numbers are not enough. But it is not the historical or the conceptual reason. The reason is that we needed a number system in which rotation worked, and once we built it, the cubic formula worked too, and so did electricity, and so did Fourier analysis, and so did quantum mechanics. That is more than enough justification for any number system.