Gaussian Multiplicative Chaos: Hidden Order Inside Mathematical Randomness

When Chaos Follows Rules

Chaos, by definition, seems to mean the absence of any rules. But in 2023, two French mathematicians, Vincent Vargas and Christophe Garban, noticed a striking pattern: inside the wildest randomness, a strict order can be hidden — and it only reveals itself as long as the chaos does not grow too strong.

To grasp this idea, imagine a swirling cloud of infinite “eddies within eddies.” This kind of multi-scale turbulence is modeled by Gaussian multiplicative chaos — a framework invented by Jean-Pierre Kahane back in 1985. Today it appears everywhere: from Brownian motion to quantum chaos and even the distribution of prime numbers.

Two Dimensions of Gaussian Multiplicative Chaos

Garban and Vargas observed that this chaos has two entirely different “sizes.”
One describes how energy is distributed in a fractal manner.
The other captures the “frequencies” of hidden oscillations — as if you were listening to chaos as a complex sound.

And they dared to ask the impossible: what if these two sizes are equal?
That question gave birth to the Garban–Vargas conjecture — an elegant formula linking the correlations and harmonics of chaos.

The Proof: Martingales Across All Scales

For a year, no one could prove it. Then in 2024, three young mathematicians from China — Zhaofeng Lin, Yanqi Qiu, and Mingjie Tang — found the key. It turned out that chaos behaves like a chain of fair “mini-games”: no matter how far you zoom in, the expected payoff stays the same. In probability theory, this structure is called a martingale.

Thousands of these “fair games,” assembled across all scales, produced exactly the formula Garban and Vargas had envisioned.

The Critical Threshold: Where Order Breaks Down

There is a catch: the moment chaos crosses a critical threshold, the entire hidden structure collapses. That boundary is precisely where the new frontier of research lies today.

Why does order emerge inside disorder?
And why does it vanish once randomness grows too strong?
These are the questions mathematics is only beginning to answer.