The Shape of the Primes

Here are the first ten thousand whole numbers, each dropped onto the page by one simple rule. The primes are lit in gold. I did not arrange them into spirals. Nobody did.

Let it run, or press Show All. For the first few hundred numbers it looks like nothing, a scatter of dots. Then, somewhere in the low thousands, arms appear. Curved spokes wheeling out from the centre, and finer rays threaded inside them. It looks designed. It looks like the numbers are trying to tell you something, and for a moment you feel you have caught mathematics in the act of keeping a secret.

I have spent a long time around numbers, and that feeling still gets me every time. So let me do two things at once: explain exactly where the spiral comes from, and then admit what it does, and does not, mean. Both halves are the point.

The mechanism

Why the arms appear

The rule is the whole trick. Take each number n, turn by n radians from the centre, and step outward by the square root of n. The square root matters: a circle of radius r has room proportional to r squared, so spacing the points by √n spreads them at a roughly even density instead of bunching them near the middle. That choice alone produces a smooth, uniform field of dots. No arms yet.

The arms come from a coincidence about the number that runs the whole show: 2π, roughly 6.283, the number of radians in a full turn. It happens to sit very close to the simple fraction 44 over 7. Step forward by 44 in n and you have turned through almost exactly seven complete circles, landing at very nearly the same angle you started from. So numbers spaced 44 apart stack up along the same direction, over and over, and the eye reads those stacks as spokes. Zoom further out and an even better fraction takes over, 710 over 113, which is closer still to 2π, and the spokes resolve into the finer rays. The spirals are real, but they are a portrait of how well 2π can be approximated by fractions. They are a fact about geometry, wearing the costume of a fact about primes.

So where do the primes come in? Once the arms exist, the primes simply refuse to stand in certain ones. Every prime past 3 is one more or one less than a multiple of 6, because anything else is divisible by 2 or by 3. Filter the field down to primes and whole arms go dark, the ones reserved for the even numbers and the multiples of three and five, while the surviving arms glow gold. The striped, selective spiral you are looking at is two things layered together: the geometry of 2π drawing the arms, and the primes choosing which arms to haunt. If you want the deeper version of this, the mathematician Grant Sanderson laid it out beautifully in a film called Why do prime numbers make these spirals?

The mystery

The order that will not be predicted

Here is the honest part, the part I think is more beautiful than the spiral. The pattern on the screen is mostly a trick of the plotting. The genuinely strange thing about the primes is not visible there at all.

The primes thin out as you climb. Near a number n, the chance that it is prime is about one in the natural logarithm of n, so they grow steadily rarer without ever stopping. That much we can predict. What we cannot do is say where the next one will fall. They are not random; each is fixed for all time by the definition of division, as rigid and determined as anything in mathematics. And yet their fine spacing dances just out of reach of every formula we have written. The single greatest unsolved problem in mathematics, the Riemann Hypothesis, is in the end a question about how evenly the primes are distributed, and it has stood unanswered for more than a century and a half.

I find that combination almost unbearably lovely: completely determined, and still unpredictable. It is the same shape of truth I keep meeting elsewhere. A weather system obeys fixed physical law and still cannot be forecast two weeks out. I wrote about that gap once, through Edward Lorenz and the butterfly effect, and the primes are its purest example. The lesson is not that the world is chaos. It is that something can be perfectly lawful and still keep its own counsel.

The stakes

What the primes were quietly guarding

There is a sentence my father said to me when I was young, and I have carried it ever since. He told me I was good at mathematics, and that he hoped I would do good in life. He did not say it as pressure. He said it the way you hand someone a small, true thing and ask them to keep it safe. I have kept it for decades.

I came to software through mathematics. I studied applied and classical mathematics and physics before I ever shipped a line of production code, and the first book I bought with my own money was about Alan Turing, which is how a boy who liked numbers fell into cryptography and, eventually, into a career. So the primes were never an abstraction to me. For years afterwards I built payment systems for a global payments company, the machinery that moves other people's money, and the security underneath all of it rests on one stubborn fact about primes.

Multiplying two enormous prime numbers together is easy. A computer does it instantly. Taking the result and pulling it back into the two primes it came from is, as far as anyone knows, brutally hard, hard enough that the entire planet is willing to bet its secrets on the difficulty. That asymmetry is the foundation of the encryption guarding card numbers, bank transfers, and private messages. Every time a payment clears, somewhere a wall is standing that is made of nothing but the unpredictability we just watched on the screen. The same property that frustrates mathematicians is the property that protects you.

There is a quiet justice in that. The thing about the primes we cannot tame is the exact thing that makes them useful. We did not defeat their unpredictability; we hired it.

The belonging

Why we are a little like the primes

There is one more thing the primes have taught me, and it is the part I love most. A prime is, by its very definition, indivisible. You cannot break it into smaller whole pieces. It does not slot into the tidy families that other numbers form, and if you pick any two different primes they share no common factor at all. In the plain arithmetic of it, primes do not belong to one another. Each one stands alone.

And yet they are neither lonely nor minor. They are the opposite. Every other number, all the composite ones that fit together so comfortably, is built out of primes and nothing else. The solitary, indivisible numbers are the exact material the whole system is made from. They belong to one another not by fitting together, but by sharing one stubborn quality: each is whole, and each refuses to be divided.

I have come to think people are a little like that. Some of us never quite slot into the families and factors the world expects. We do not divide cleanly into the group, and for years it can feel like not belonging. The primes make me suspect the opposite is true. The ones who stand a little apart, who cannot be reduced to anyone else's terms, are often the ones the rest is quietly built on. You do not have to fit with everyone to belong. You only have to be, like a prime, fully and unmistakably yourself. You do not belong with each other, and you are each a prime number, and that is the belonging.

The reward

The reward for looking

You do not need to solve the Riemann Hypothesis to be moved by that spiral. That is what I would say to anyone earlier in their career than me, worried that the deep mathematics is for other people. The wonder is free. It is available to anyone willing to plot ten thousand dots and lean in.

More than two decades into building systems for a living, the thing I am most quietly protective of is not a skill or a title. It is the capacity to still be astonished by something as ordinary as a list of numbers. The day a spiral made of primes stops stopping me in my tracks is the day I should worry. Curiosity is the renewable resource of this whole craft, and it costs nothing to keep topped up.

If I could press a single recommendation on you, with more conviction than I bring to almost anything else, it would be this: let yourself fall in love with mathematics. It is, to my mind, the most beautiful thing our species has ever made. Whether we discovered it, sitting there waiting and already perfect, or invented it ourselves, I have never managed to decide, and I have happily stopped trying. Either way, everything else we have built, every machine and market and line of code, is standing on top of it.

So let it run one more time. Watch the gold dots find their arms. Remember that most of the shape is the geometry of a turn, that the real secret is hiding somewhere the eye cannot follow, and that this small, stubborn unpredictability is out there right now, doing honest work, keeping the world's money and its secrets safe.

The hope

Did I do good?

My father hoped I would do good in life. I have thought about that sentence more than any equation. I am not sure I have met it, and I am not even sure I could tell you what the words are meant to contain. After more than two decades around brilliant people and large systems, I still could not give you a definition of "good," or of "success," that would survive five honest minutes of questioning. If anything, the primes have taught me to distrust any pattern that looks too tidy, and a tidy definition of a whole life is the tidiest pattern of all.

But there is one thing I am sure of. I am at peace, and I am happy. Somewhere along the way I stopped measuring myself by how high I had climbed and began asking only whether I had stayed whole, and grounded, and humble, and it turned out that was the thing I had needed the entire time. Maybe that is what my father meant. Maybe doing good in life was never about the height at all. Maybe it is about refusing, the way a prime refuses, to be broken into smaller pieces to fit someone else's arithmetic.

If that is the measure, then quietly, with no need to announce it, I believe I did do good. I hope he believes it too.