English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

What Really Superposes Is Amplitude: Probability Amplitudes and the Magic of Negative Signs

Forum topic · 小凯 · 2026-11-07

Summary

This tutorial-style post from a Chinese quantum computing series explains why quantum superposition operates on amplitudes rather than probabilities. Classical probability multiplies and adds probabilities directly, but quantum mechanics keeps a finer ledger: each path contributes a complex amplitude (a directed arrow), amplitudes add first, and only then is the squared length of the combined amplitude taken as the probability. The order matters because directions can cancel: two individually possible paths with opposite phases can sum to zero, making an outcome's probability exactly zero—an effect demonstrated by the double-slit experiment, where opening a second slit makes some regions darker. The author ties this to earlier installments: superposition means 0 and 1 amplitudes coexist; complex numbers provide the freedom to rotate phases; interference in quantum algorithms works by canceling wrong-answer amplitudes and reinforcing right ones; and decoherence destroys directional information, reducing quantum bookkeeping to classical probability. The post also carefully limits the water-wave analogy: amplitudes live in abstract space, are never directly observable, and quantum cancellation can zero an outcome entirely.

What Really Superposes Is Amplitude: Probability Amplitudes and the Magic of Negative Signs

> Prerequisites: You need lessons 011 (complex arrows), 012 (the state sphere), and 004 (interference). > Next episode preview: 014, "Gates Are Matrices."

Start with a kindergarten probability question—please answer with "common sense."

Toss an ordinary coin: heads comes up with probability one-half. Now imagine a device: the coin first passes through machine A, which changes the heads probability to one-third; then through machine B, which on top of that changes it to one-half. What is the final probability of heads?

The common-sense algorithm: one-third × one-half = one-sixth. Probabilities multiply along the way and add when combining—this rule governs lotteries, weather, and insurance, and has never failed in three hundred years.

Before the quantum world rolls its dice, put this rule on hold. It is only half right: probability does appear at the end, but it is not the first-hand ledger. Before probability there is a finer layer of accounts—the amplitude.

Bookkeep First, Square Later

Quantum mechanics' bookkeeping rule: for every "path," first record a directed arrow (the complex arrow from lesson 011, formally called the amplitude). When multiple paths lead to the same outcome, add the arrows first (the relay rule: same direction reinforces, opposite directions cancel), and only then square the length of the combined arrow to get the probability.

Note the order: add arrows first, square later. It is not the case that you convert each path's arrow into a probability and then add. Change the order and you change worlds—this is the dividing line between quantum mechanics and classical probability.

Why does the order matter so much? Because arrows have direction. Direction can fight it out during the "add arrows" step (opposite phases cancel), but once you square, direction is erased—squared probabilities are always non-negative, and there is no more "canceling" available. Classical probability is quantum bookkeeping that has lost its direction.

The Magic of the Negative Sign: Turning "Possible" into "Never"

Now see what this bookkeeping can do.

An outcome reachable by two paths. The first path contributes an arrow; the second contributes an arrow pointing in exactly the opposite direction (a 180-degree phase difference—lesson 011's "two left turns make a U-turn"). The two arrows add in relay—the sum is zero. Zero squared is zero: the probability of this outcome is zero.

Read that conclusion again and let it sink in: two paths that are each individually possible combine into something that never happens. Open either path alone and the event appears; open both paths simultaneously and the event disappears. In the language of classical probability this is pure madness—adding two non-negative probabilities to get zero is illegal by the rules themselves.

But it is not a logic game. Throw two pebbles into water: where the two ripples meet, some spots have crest meeting trough and the surface stays perfectly still—the energy has not vanished; an "upward motion" and a "downward motion" exactly cancel. The double-slit experiment staged the same show for photons: one slit open gives bright fringes; with both slits open, certain spots become dark. Open one more route and some places get no visitors at all—that is the negative sign (anti-phase) magic documented by experiment.

Collecting Three Old Plot Threads

Here the threads planted in earlier lessons can be gathered at once:

  • Lesson 001's "propensity": what superposes in a superposition state is not probability but amplitude—each propensity is a directed arrow, so "being 0 and 1 at the same time" precisely means "the amplitude of 0 and the amplitude of 1 are both on the books."
  • Lesson 011's complex numbers: amplitudes are complex numbers, because cancellation requires the freedom to "rotate to any angle." The negative sign is merely the simplest case of anti-phase (a 180-degree turn); the general case is any angle.
  • Lesson 004's interference: the computing power of a quantum algorithm = designing operations so that the wrong answers' amplitude arrows cancel each other while the right answer's amplitudes add in phase. Today you know why it works: probability = squared length of the combined amplitude, and the combined amplitude can be shorter than any single path's—short enough to be zero.
  • Lesson 012's sphere: latitude determines probability—the precise statement is that the polar angle determines how the lengths of the two amplitudes (for |0⟩ and |1⟩) are distributed; the longitude is the phase angle between them, which does not change their lengths but decides who helps whom in a future interference.
  • Where Classical Probability Went Wrong

    To avoid misunderstanding, restate the position: classical probability is not "wrong"—it is simply good enough in a world without interference. At everyday scales, countless environmental disturbances scramble the directions of amplitudes (lesson 006's decoherence). Once direction information is scrambled, adding arrows degenerates into adding probabilities—your dice and your insurance all live in this "already decohered" approximation.

    The entire business of a quantum computer is doing deals in the microscopic world where this approximation fails: preserving arrow directions and choreographing how arrows meet. This is also why decoherence is catastrophic for quantum computing—it is not "a bit more noise"; it confiscates the very foundation of the business: direction.

    Dismantling the Analogy

    Where the water-wave analogy falls short:

  • Difference one: a water wave is a real displacement of the water surface; energy genuinely resides in the medium. Amplitude arrows do not live in any medium—they are entries in an abstract space, and the "cancellation" cancels probability, not energy or matter.
  • Difference two: water waves can be observed point by point (you can watch the ripples meet); amplitudes can never be directly observed—only the squared probability is visible. The very act of observing an amplitude does not exist; this is not an instrument problem but a matter of principle (lesson 002: "observing is disturbing," taken all the way down).
  • Difference three: water-wave cancellation is local (quiet here, louder elsewhere); quantum cancellation can be exact—an outcome's total probability is zero—as if the particle's behavior were "collectively negotiated," with no single path having independent standing.

Luggage to Take Home

1. Probability is not the first-hand ledger: each path records an amplitude (a directed complex arrow) → arrows add → square the combined arrow's length = probability. The order cannot be swapped. 2. Anti-phase cancellation: two individually possible paths can combine to probability zero—open one more slit and some places get darker. This is the heart of the double-slit experiment and the entire principle by which quantum algorithms "erase" wrong answers. 3. Classical probability = amplitude bookkeeping with direction lost. The everyday world's directions are already scrambled (decoherence); quantum computing is the business of preserving and manipulating direction.

Hook for the next stop: you now have arrows on the books—so what about "operations"? How does a quantum gate change an arrow? Spoiler: every gate is a small table of numbers, and humanity long ago gave a name to "using a table to process arrows"—the matrix. Next up: "Gates Are Matrices" (014).

*Tags from the original: quantum computing, from coins to error correction, amplitude, double slit.*

Tags

#quantum-computing#probability-amplitude#superposition#interference#double-slit-experiment#decoherence#complex-numbers#quantum-algorithms

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178636847