Get AI summaries of any video or article — Sign up free
The Closest We’ve Come to a Theory of Everything thumbnail

The Closest We’ve Come to a Theory of Everything

Veritasium·
6 min read

Based on Veritasium's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.

TL;DR

The brachistochrone problem’s fastest path is a cycloid, derived by mapping mechanics to an optics-style refraction setup using Snell’s law.

Briefing

A single “stationary action” principle links the motion of falling objects, the bending of light, and the equations of mechanics—turning what once looked like separate laws into one unifying rule. The core claim is that nature doesn’t merely follow a path; it selects a trajectory for which a particular quantity, called action, changes as little as possible under small variations. That requirement—formalized by Euler and Lagrange—reproduces Newton’s second law and, in special cases, Fermat’s least-time principle for light.

The story begins with the fastest-descent problem: sliding a mass from point A to B, what ramp shape minimizes travel time? Common sense suggests a straight line, but Galileo showed that bending the ramp early lets the object accelerate sooner, beating the straight path. The question then sharpened into a challenge. In 1696, Johann Bernoulli posed the problem to mathematicians worldwide; no one submitted solutions. When Gottfried Leibniz persuaded Bernoulli to extend the deadline, Isaac Newton responded quickly—yet Bernoulli’s own solution outshone Newton’s.

Bernoulli’s key move was to borrow an idea from optics. Light traveling through a single medium follows the shortest path, but when it crosses between media it refracts, obeying Snell’s law. Pierre Fermat provided the missing “why” by proposing that refraction follows the path that minimizes total travel time. Bernoulli then translated the mechanics problem into an optics-like setup: treat the speeding-up of a falling particle as if light moves through layers whose effective speed changes with depth, so that Snell’s law holds at each interface. In the limit of infinitely thin layers, the resulting continuous curve becomes a cycloid. The fastest descent from A to B is therefore an arc of a cycloid—also known as a brachistochrone curve.

That optical-mechanical bridge set the stage for a deeper generalization. About forty years later, Pierre Louis de Maupertuis noticed that light and particles can behave similarly and proposed a new quantity—action—roughly tied to mass, velocity, and distance. He claimed that among all possible trajectories, the one nature chooses minimizes action. The idea drew ridicule and accusations of plagiarism, but Euler defended and strengthened it mathematically. Euler showed that the principle works cleanly when total energy is conserved across the compared paths, and he developed methods to handle continuous variations.

The final leap came with Joseph-Louis Lagrange, who delivered a general proof. The modern formulation reframes “least action” as “stationary action”: the first-order change in action vanishes for the true path. When the action is written using the Lagrangian (kinetic energy minus potential energy), the resulting Euler–Lagrange equations reproduce Newton’s second law for mechanics. The payoff is practical as well as philosophical: once kinetic and potential energy are known, the same variational machinery can generate equations of motion in multiple dimensions and even in awkward coordinate systems—making complex systems like a double pendulum far more tractable than force-based approaches.

In short, the unifying principle is not just a clever trick for one problem. It’s a framework where light refraction, particle trajectories, and classical mechanics all emerge from the same requirement: the action is stationary under small changes to the path.

Cornell Notes

The fastest way to move from A to B and the way light bends at boundaries both follow an optimization rule. Bernoulli solved the fastest-descent problem by translating it into an optics scenario using Fermat’s least-time principle, leading to a cycloid (the brachistochrone curve). Maupertuis then proposed a broader quantity—action—that should be stationary (not necessarily minimized) for the actual path, extending the idea beyond light. Euler and Lagrange supplied the mathematical rigor: with conserved energy and the modern Lagrangian (T − V), the stationary-action condition yields the Euler–Lagrange equations, which reproduce Newton’s second law. This matters because one variational framework can generate equations of motion across many mechanical systems, not just special cases.

Why does the fastest-descent path turn out to be a cycloid rather than a circle or a straight line?

The problem asks for the ramp shape that minimizes travel time. Galileo showed that a bent ramp can beat a straight path because it accelerates the object earlier. Bernoulli then reframed the mechanics problem as an optics problem: treat the falling particle like light moving through layers where the effective speed changes with depth, so Snell’s law applies at each interface. As the layers become infinitely thin, the resulting continuous curve matches the cycloid equation. That curve is the brachistochrone: it gives the shortest travel time between two points under gravity.

How did Fermat’s principle of least time explain Snell’s law?

Snell’s law states that the ratio of sines of incidence and refraction angles is constant for two media. Fermat’s contribution was to treat refraction as a time-minimization problem: among all possible light paths crossing the boundary, the actual path is the one with the smallest total travel time, given different light speeds in different media. When the time along candidate paths is minimized, Snell’s law emerges as the condition for the minimizing trajectory.

What did Maupertuis mean by “action,” and why was it controversial?

Maupertuis proposed that nature minimizes a quantity called action, described as mass times velocity times distance, summed over segments of a trajectory. In a simple frictionless example (a ball rolling and then bouncing), action adds across segments, and the chosen trajectory is the one with the smallest total action among alternatives. The idea was attacked and mocked, including accusations from Samuel Konig and Voltaire that Maupertuis was wrong or plagiarizing Leibniz, and critics also noted the lack of a clear reason why that specific quantity should be minimized.

What mathematical conditions did Euler add to make the principle of least/stationary action work reliably?

Euler replaced discrete sums with integrals to handle continuous changes in paths. He found that the principle works under specific constraints—most importantly, that total energy is conserved across the compared trajectories and that the relevant energy is the same for all paths in the comparison. With those conditions, the variational calculation becomes rigorous and leads to correct equations of motion.

How does Lagrange’s proof connect stationary action to Newton’s second law?

Lagrange’s general proof uses the calculus-of-variations idea: at the true path, the first-order change in action is zero (stationary), meaning small “bumps” to the path don’t change action to linear order. In modern form, the action is written as an integral of the Lagrangian L = T − V over time. Applying the stationary condition yields the Euler–Lagrange equations, which reduce to F = ma for standard mechanics. Fermat’s least-time principle also becomes a special case within this broader framework.

Why is the phrase “least action” misleading?

The principle is more accurately “stationary action.” In calculus, setting a derivative to zero can correspond to a minimum, maximum, or a saddle point; similarly, the true trajectory makes the action stationary, not always strictly minimal. The first-order variation vanishes, while the second-order behavior determines whether it’s a minimum in a given situation.

Review Questions

  1. In Bernoulli’s translation of fastest descent into an optics problem, what role does Snell’s law play as the layers become thinner and thinner?
  2. What specific change did Euler introduce to move from Maupertuis’s original action idea to a continuous, mathematically workable principle?
  3. How does the stationary-action condition using the Lagrangian (T − V) lead to the Euler–Lagrange equations and connect back to Newton’s second law?

Key Points

  1. 1

    The brachistochrone problem’s fastest path is a cycloid, derived by mapping mechanics to an optics-style refraction setup using Snell’s law.

  2. 2

    Fermat’s least-time principle explains refraction by selecting the light path that minimizes total travel time across media with different light speeds.

  3. 3

    Maupertuis generalized the optimization idea by introducing action as a trajectory-dependent quantity, but the proposal initially lacked clear justification and faced strong criticism.

  4. 4

    Euler strengthened the principle mathematically by converting sums to integrals and showing it requires conserved total energy across compared paths.

  5. 5

    Lagrange provided a general proof that the true trajectory makes the action stationary (first-order variation zero), not necessarily strictly minimal.

  6. 6

    In modern mechanics, writing action as the time integral of the Lagrangian L = T − V leads to Euler–Lagrange equations that reproduce Newton’s second law.

  7. 7

    The variational framework scales beyond simple 1D motion, enabling systematic equation-building in multiple dimensions and convenient coordinate systems.

Highlights

The fastest descent from A to B is not a circle or a straight line but an arc of a cycloid—an outcome reached by treating gravity like an optics refraction problem.
Fermat’s least-time principle turns Snell’s law from an empirical rule into a consequence of minimizing travel time.
Maupertuis’s “action” idea—later made rigorous by Euler and Lagrange—unifies light, particles, and classical mechanics under one variational rule.
The modern principle is “stationary action”: the action’s first-order change vanishes for the real path, and that condition yields the Euler–Lagrange equations.

Topics

Mentioned