⚛️ quantum physics

Causal Consistency Selects the Born Rule: A Derivation from Steering in Generalized Probabilistic Theories

This paper demonstrates that within generalized probabilistic theories satisfying purification, the requirement of relativistic causality (no-signaling) uniquely forces the predictive probability to be linearly related to the geometric transition probability, thereby deriving the Born rule as the only causally consistent probability assignment.

Original authors: Enso O. Torres Alegre

Published 2026-02-05

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Enso O. Torres Alegre

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Question: Why is the "Born Rule" the way it is?

In quantum mechanics, there is a famous rule called the Born Rule. It tells us how to calculate the probability of finding a particle in a certain state. Mathematically, it says the probability is the square of a specific number (the "overlap" between two states).

For a long time, physicists have treated this rule as a fundamental fact of nature, like a law written in the stars. But this paper asks a deeper question: Is the "squaring" part just an arbitrary choice, or is it the only way things can work without breaking the laws of physics?

The author, Enso O. Torres Alegre, argues that it's the latter. He proves that if you want to keep the universe from allowing "faster-than-light" communication (which would break causality), the probability rule must be a straight line. In quantum mechanics, a straight line relationship between geometry and probability happens to look like a square.

The Two Types of "Closeness"

To understand the proof, we need to distinguish between two things that usually look the same but are actually different:

Geometric Closeness ( $\tau$ ): Imagine two arrows pointing in slightly different directions. You can measure how "close" they are just by looking at their shapes. This is a structural fact.
Predictive Probability ( $P$ ): This is the actual chance that an experiment will give a specific result.

In standard quantum mechanics, these two are identical. If the arrows are 50% "close" geometrically, the chance of the experiment working is 50%.

The Paper's Idea: What if they weren't identical? What if the chance of an experiment working was a non-linear function of that closeness?

Maybe if the arrows are 50% close, the chance is actually 25% (a curve).
Or maybe if they are 50% close, the chance is 75% (a different curve).

The paper asks: Can the universe allow these curved, non-linear rules?

The "Magic Remote Control" (Steering)

The key to the answer is a quantum phenomenon called Steering.

Imagine Alice and Bob share a pair of "magic coins" that are entangled. They are far apart.

If Alice flips her coin and gets "Heads," Bob's coin instantly becomes "Heads."
If Alice flips and gets "Tails," Bob's coin becomes "Tails."

Here is the weird part: Alice can choose how she flips her coin.

Protocol A: She flips in a way that makes Bob's coin a 50/50 mix of Heads/Tails.
Protocol B: She flips in a different way that also makes Bob's coin a 50/50 mix of Heads/Tails.

To Bob, looking only at his coin, Protocol A and Protocol B look exactly the same. His coin is in the exact same "average state" in both cases.

The Trap: How Non-Linearity Breaks the Universe

Now, imagine the probability rule is curved (non-linear) instead of straight.

In Protocol A: Alice forces Bob's coin to be a mix of two specific states. Because the rule is curved, the average of the probabilities of these two states might be, say, 60%.
In Protocol B: Alice forces Bob's coin to be a mix of two different states. Even though the average state is still 50/50, the average of the probabilities might be 40% because of the curve.

The Result: Bob looks at his coin. He sees that in Protocol A, he gets "Heads" 60% of the time. In Protocol B, he gets "Heads" 40% of the time.

Even though Alice is far away and Bob can't see her, Bob can tell exactly which protocol Alice chose just by counting his results.

This means Alice can send a message to Bob instantly (faster than light) just by choosing which way to flip her coin. This violates Relativistic Causality (the rule that nothing travels faster than light).

The Conclusion: The Universe Must Be "Straight"

The paper proves that the only way to prevent this "magic remote control" from sending faster-than-light messages is if the probability rule is a straight line (linear).

If the rule is a straight line, the "average of the probabilities" is exactly the same as the "probability of the average."
In that case, Bob sees the exact same statistics no matter what Alice does. No message is sent. Causality is safe.

In the specific geometry of quantum mechanics, a "straight line" relationship between the geometric closeness and the probability turns out to be the squared relationship we know as the Born Rule ( $P = |\text{overlap}|^2$ ).

Summary in a Nutshell

The Problem: Why is the quantum probability rule a square?
The Test: What if the rule was a curve instead?
The Mechanism: Using "Steering" (entanglement), a curved rule would let Alice send secret messages to Bob instantly by changing how she prepares his state.
The Verdict: Since faster-than-light communication is impossible, the rule cannot be a curve. It must be a straight line.
The Result: The only probability rule that fits this "straight line" requirement in our universe is the Born Rule.

The paper concludes that the Born Rule isn't just a random mathematical choice; it is a safety mechanism required to keep the universe from allowing time-traveling messages.

Technical Summary: Causal Consistency Selects the Born Rule

Problem Statement
The Born rule, $P(i) = |\langle \phi_i | \psi \rangle|^2$ , occupies a unique position in quantum mechanics: it is empirically indispensable yet theoretically underived from the other postulates. While various derivation strategies exist (e.g., Gleason's theorem, decision-theoretic approaches, envariance), they often rely on specific structural assumptions like Hilbert space geometry or noncontextuality. This paper addresses the foundational question of whether the quadratic form of quantum probabilities is a fundamental postulate or a necessary consequence of deeper operational principles, specifically relativistic causality. The authors investigate this within the framework of Generalized Probabilistic Theories (GPTs), which encompass classical probability, quantum mechanics, and hypothetical alternatives.

Methodology
The authors employ a theory-neutral GPT framework to distinguish between two quantities often conflated in standard quantum mechanics:

Geometric Transition Probability ( $\tau$ ): A structural quantity defined as the maximum probability of accepting a state $\phi$ when the system is in state $\psi$ . In quantum mechanics, $\tau(\psi, \phi) = |\langle \phi | \psi \rangle|^2$ .
Predictive Probability ( $P$ ): The probability assigned to a measurement outcome, which is compared against experimental frequencies.

The core hypothesis is that predictive probabilities may be a nonlinear function of geometric transition probabilities: $P(\phi|\psi) = \Phi(\tau(\psi, \phi))$ , where $\Phi: [0, 1] \to [0, 1]$ . The standard Born rule corresponds to $\Phi$ being the identity function.

The derivation proceeds by analyzing the constraints imposed on $\Phi$ by no-signaling in the presence of steering. The authors assume standard GPT axioms:

No-signaling: Local marginal statistics are independent of distant measurement choices.
Purification: Every mixed state is the marginal of a pure bipartite state, with essentially unique purifications.
Continuous Reversibility: Reversible transformations act transitively on pure states.
Spectrality: States admit decompositions into perfectly distinguishable pure states.

The argument relies on the physical mechanism of steering: in theories with purification, a distant party (Alice) can choose a measurement that remotely prepares different ensembles for a second party (Bob) that share the same average state but have different decompositions.

Key Contributions and Results

Formalization of the Distinction: The paper rigorously separates geometric structure ( $\tau$ ) from probabilistic prediction ( $P$ ) within GPTs, allowing for the general functional relationship $P = \Phi(\tau)$ .
Derivation of Linearity: The authors prove that if a GPT satisfies purification (and thus admits steering), the requirement of no-signaling forces $\Phi$ $Φ$ to be the identity function ( $\Phi(p) = p$ $Φ (p) = p$ ).
- Mechanism: If $\Phi$ were strictly convex or concave (nonlinear), Alice could choose between two measurement protocols that steer Bob's system to different ensembles with the same average state. Due to the nonlinearity of $\Phi$ , the expected predictive probabilities for Bob would differ between the two protocols ( $P_1 \neq P_2$ ). Since Bob's statistics would depend on Alice's distant choice, this would enable superluminal signaling, violating the no-signaling axiom.
- Conclusion: To prevent signaling, $\Phi$ must be affine. Combined with boundary conditions ( $\Phi(0)=0, \Phi(1)=1$ ), the unique solution is $\Phi(p) = p$ .
Explicit Qubit Example: The paper provides a concrete calculation using a qubit system and a maximally entangled state. It demonstrates that a nonlinear rule (e.g., $\Phi(p) = p^{3/2}$ ) creates a detectable gap in outcome statistics (approx. 14.6% in the example) between different steering protocols, despite the average state remaining identical.
Reconstruction of the Born Rule: By combining the result $P = \tau$ with established GPT reconstruction theorems (which identify the geometric structure of complex quantum mechanics), the paper derives the standard Born rule $P(\phi|\psi) = |\langle \phi | \psi \rangle|^2$ as the unique causally consistent probability assignment.

Significance
The paper claims that the Born rule is not an arbitrary postulate but a "causal fixed point." The quadratic form of quantum probabilities emerges as the unique assignment compatible with relativistic causality in theories that possess entanglement and steering.

Key distinctions from prior work include:

Framework: Unlike Gleason's theorem, which assumes Hilbert space structure and noncontextuality for dimensions $d \ge 3$ , this work operates in the general GPT framework without presupposing the specific geometry of quantum theory.
Mechanism: The derivation identifies steering as the specific physical mechanism that converts mathematical nonlinearity into physical signaling.
Scope: The result applies to composite systems and relies on the interplay between probability, entanglement, and causality.

The authors conclude that for any physical theory exhibiting quantum-like features (specifically purification and steering), the Born rule is inevitable, as any deviation would violate the principle of no-signaling.