What Ruler and Compass Can and Cannot Build — The Story of the Three Classical Construction Problems

An unmarked straightedge and a single compass. With nothing more than these tools, the mathematicians of ancient Greece managed to construct a surprising number of figures. Bisect a segment, bisect an angle, build an equilateral triangle or a regular hexagon—each construction is elegant, with nothing wasted.

And yet, in front of a few questions, they came to a halt. "Trisect an angle into three equal parts." Bisection is so easy that trisection ought to be within reach too. But no one succeeded. Not for two thousand years. Then, in the nineteenth century, the matter was finally settled. And the resolution was not "a clever method was found" but the utterly unexpected verdict that "it is simply impossible." In this article, let us take a look at this mathematical adventure of pinning down the limits of a tool.

The Rules of the Construction Game

First, let us make the rules crystal clear. Only two moves are permitted: drawing the line through two points with a straightedge, and drawing a circle with a given center and radius using a compass. The straightedge has no markings, so you cannot measure lengths. The only points allowed as new points are the intersections of the lines you have just drawn—where a line meets a line, a line meets a circle, or a circle meets a circle.

How far can you draw within this severe restriction? That was the game called "construction," handed down since ancient Greece. The Elements of Euclidean geometry builds its arguments on exactly this kind of construction from the very opening of Book I. Not merely asserting that a figure "exists," but showing that, following the rules, it can actually "be built"—that was their intellectual honesty.

The Three Fortresses That Would Not Fall

Once you get used to the rules, most figures turn out to be buildable. But three problems stubbornly refused to fall. They are the famous puzzles later known as the "three classical construction problems of Greece."

The first is trisecting an angle: given an angle, divide it into three equal parts using only ruler and compass. The second is doubling the cube (the Delian problem): construct the edge of a cube whose volume is exactly twice that of a given cube. The third is squaring the circle: construct a square with exactly the same area as a given circle.

Put into words, any of them is simple enough for a child to understand. And yet, for two thousand years, not a single person solved them correctly. Challengers were beyond counting, and each time a claim of "solved it!" appeared, an error was found. It was as if the problems themselves, at some fundamental level, were refusing to let anyone in.

Translating a Problem of Figures into a Problem of Numbers

The breakthrough was to stop staring at the figures as figures and to think in coordinates instead. Place a segment of length $1$ at the starting point of the construction and set up coordinates on the plane. The question "what points can a construction produce?" then turns into the question "what numbers (coordinates) can it produce?"

Here a decisive observation takes hold. Drawing a line with a straightedge corresponds to solving a linear equation; drawing a circle with a compass corresponds to solving a quadratic equation. If you trace the calculations for the intersection of a line with a line, a line with a circle, or a circle with a circle, you find that every newly obtained length is limited to what you get by applying addition, subtraction, multiplication, division, and square roots a finite number of times to the lengths you already have.

The numbers obtained this way are called constructible numbers. In other words, all the construction game can reach is the world of numbers that can be written starting from $1$ using only the four arithmetic operations and $\sqrt{\phantom{x}}$ (and the square roots may be nested to any depth, as in $\sqrt{2+\sqrt{3}}$). This is the moment when "buildable or not" for figures was translated into "expressible in this form or not" for numbers. It is an idea continuous with the thinking about transformations and coordinates that you meet at the basic level.

What It Means to Prove "Impossible"

Once the translation is done, the contest moves onto the turf of algebra. The problem of trisecting a general angle, by way of the trigonometric identities, ends up as the task of solving a certain cubic equation (for example, trisecting $60°$ amounts to constructing $\cos 20°$, and this $\cos 20°$ is a root of the cubic equation $8x^3 - 6x - 1 = 0$). Doubling the cube is likewise the cubic equation $x^3 = 2$. But the numbers you can build by combining only square roots spread out, so to speak, only along a "ladder of powers of two." The solution of a cubic equation, in general, spills off this ladder—it can never be written with a finite combination of square roots.

The man who proved this rigorously, in 1837, was Pierre Wantzel. He showed algebraically that trisecting an angle and doubling the cube do not fit inside the world of constructible numbers, and put an end to a two-thousand-year-old question in the form of "it cannot be done." Squaring the circle was tougher still: Lindemann showed that the circle constant $\pi$ is a special kind of number (a transcendental) that is not the solution of any algebraic equation at all—let alone one built from the four operations and square roots—and so this problem too was confirmed impossible.

"No method has been found" and "no method exists" are worlds apart. To assert the latter, you need a piece of logic that seals off every possible procedure at once. The resolution of the three classical construction problems was an event that showed the world the sheer power of mathematics—the ability to prove impossibility itself.

In Closing — Knowing the Limits of a Tool

What can and cannot be built with ruler and compass. That boundary was not about being good or bad at handling a compass, but about the very outline of the algebraic world of "numbers that can be written with the four arithmetic operations and square roots." The moment the question of figures was translated into the language of numbers, the fortresses that had not budged for two thousand years suddenly revealed their shape.

What this resolution left behind was more than the answers to three puzzles. From this point on, mathematicians began to ask not "which figures can be drawn?" but "which numbers can be built?" The questions of geometry were translated into the language of algebra, and here was born the prototype of an idea that runs through all of modern mathematics—bridging distant fields to crack a hard problem. The three classical construction problems were not a mere ancient riddle; they were an event that pushed the very way of thinking in mathematics one step wider.

Knowing the limits of a tool is not a defeat. On the contrary, by making clear "what cannot be done," figures, numbers, and equations all connect into a single thread, and a wide-open vista comes into view. In that it opens a new world when you question the premises and the framework, it is a story that resonates with the other ending of the parallel lines. The theme of reinterpreting figures within a different framework will deepen further in the intermediate level that lies ahead.