⚛️ quantum physics

What Kind of World Supports Darwinian Evolution? Quantum Foundational Options

This paper argues that Darwinian evolution necessitates a world with heritable records and irreversible copying—conditions only met in specific classical sectors of quantum mechanics—and evaluates various ontological frameworks, including decoherence, agent-relative facticity, and stochastic mechanics, to explain how such classical structures emerge from a quantum foundation while satisfying the constraints of agency and measurement.

Original authors: Partha Ghose

Published 2026-02-19

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Partha Ghose

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: Can Life Evolve in a Quantum World?

Imagine you are trying to run a bakery. To make bread, you need three things:

A Recipe: A stable record of what to do (the "genotype").
Copying: The ability to photocopy the recipe so you can make more loaves, but sometimes make a tiny mistake (a mutation) so you can invent new flavors.
Throwing Away: The ability to toss out the failed loaves so you don't keep baking the same bad bread forever. This is "irreversibility."

This is how Darwinian Evolution works. It needs a stable "record" to copy and a way to delete the failures.

The Problem: The paper asks, "What if the universe is fundamentally made of Quantum Mechanics?"
In the quantum world, things are fuzzy. A particle can be in two places at once (a superposition). In this fuzzy world, you cannot simply "copy" or "delete" information the way you do in a computer or a bakery. The laws of physics say:

No-Cloning: You can't perfectly copy a quantum state you don't know.
No-Deleting: You can't perfectly erase a quantum state without a trace.

If the whole universe is just a giant, fuzzy quantum soup, how can evolution ever get started? You can't copy a recipe if the paper keeps turning into smoke and back again!

The Solution: We Need a "Classical" Corner

The author uses a branch of math called Categorical Quantum Mechanics to explain that "copying" and "deleting" aren't universal tools. They are special tools that only work in a specific "classical" zone.

The Analogy: The Translator
Imagine the universe is a chaotic, noisy room where everyone is speaking a different, shifting language (Quantum).

Copying is like taking a photo. You can only take a clear photo if the subject is standing still and wearing a specific color (a Preferred Basis).
If the subject is vibrating and changing colors (a quantum superposition), your photo comes out blurry. You can't copy it.

For evolution to happen, the universe must have a way to "freeze" the chaos into a stable, clear picture. This is called a Classical Data Sector. It's like a translator that turns the fuzzy quantum noise into clear, stable text (like the letters A, B, C) that can be copied and pasted.

How Do We Get This "Classical" Zone? (Decoherence)

The paper discusses Decoherence. This is the process where a quantum system interacts with its environment (like air molecules or light).

The Analogy: Imagine a spinning coin on a table. As long as it spins, it's both Heads and Tails (Quantum). But as soon as it hits the table and wobbles, it settles into either Heads or Tails (Classical).
The Catch: Decoherence explains how the coin stops spinning and becomes a clear "Heads" or "Tails." But it doesn't explain which one it becomes. The universe might still be holding both possibilities in a giant, invisible entangled web.

So, Decoherence gives us the stable records (the coin is now flat), but it doesn't tell us if there is only one single reality or a million parallel realities.

The "Agency" Problem: Can You Be an Agent in a Quantum World?

The paper brings up a clever argument by Adlam, McQueen, and Waegell called the Agency Constraint.

The Idea: To be an "agent" (a being that thinks, plans, and acts), you need to remember things and make copies of information.
The Problem: If you are in a "purely quantum" world where nothing is ever copied or deleted (because everything is a fuzzy superposition), you can't have a stable memory. You can't plan for tomorrow if your memory of today keeps changing.
The Conclusion: You cannot be a thinking, acting agent in a purely quantum world. You need a classical, copyable record to exist as an agent. This means evolution (which relies on agents copying genes) is impossible without a "classical" foundation.

Four Ways to Fix the Universe (The Options)

Since the standard quantum rules make evolution tricky, the author outlines four different ways physicists try to fix the story. Think of these as four different "operating systems" for reality:

Option A: The Single-Story Universe (Unique History)
- The Idea: There is only one real timeline. When a choice happens, the universe physically "collapses" or picks a path (like a hidden variable guiding a particle).
- Evolution: Easy. There is one clear record, one clear copy, and one clear history.
- Analogy: A movie reel. Only one frame is playing at a time.
Option B: The Multiverse (Many Worlds)
- The Idea: The universe never collapses. Every possible outcome happens, but they split off into different, non-communicating branches.
- Evolution: Inside each branch, the records are stable and evolution works fine. But there are infinite versions of you.
- Analogy: A choose-your-own-adventure book where every time you make a choice, the book splits into two new books. You only read one, but all exist.
Option C: The "It Depends on Who You Are" Universe (Agent-Relative)
- The Idea: Facts aren't absolute. A fact is only real relative to the person observing it.
- Evolution: Evolution works for you because you have your own stable records. But your friend might have a different set of records.
- Analogy: Two people looking at a Rorschach inkblot. One sees a bird, the other sees a butterfly. Both are "right" relative to their own perspective.
Option D: The Stochastic Bridge (Random Walks)
- The Idea: The universe is actually made of tiny, random walks (like a drunk person stumbling). Sometimes the stumbling is very wild (Quantum), and sometimes it's very straight (Classical).
- Evolution: This offers a smooth bridge. The universe isn't "Quantum" or "Classical"; it's a sliding scale. When the "stumbling" slows down, you get classical records and evolution.
- Analogy: A river. Upstream, the water is turbulent and chaotic (Quantum). Downstream, it flows smoothly and predictably (Classical). Evolution happens downstream.

The "Stress Test": The Wigner's Friend

The paper uses a famous thought experiment called Extended Wigner's Friend to test these ideas.

The Setup: Imagine a friend is in a lab measuring a particle. To the friend, the particle has a definite result (a record). But to an outside observer (Wigner), the whole lab (including the friend) is still in a fuzzy quantum superposition.
The Conflict: How can the friend be a "thinking agent" with a memory (requiring a classical record) while simultaneously being a "fuzzy quantum object" (requiring no record)?
The Point: This experiment forces us to admit that we must assume some classical resources (like stable records) exist for the friend to be an agent. If we don't, the whole story breaks down.

The Bottom Line

For Darwinian evolution to work, the universe must provide a stable stage where information can be copied and deleted.

Pure quantum mechanics is too fuzzy for this.
We need a mechanism (like Decoherence or Stochastic Mechanics) to create a "classical corner" where records become solid.
Once that corner exists, evolution can run its course, copying genes and discarding the failures, just like a baker making bread.

The paper concludes that whether you believe in one timeline, many timelines, or random walks, you must accept that somewhere, somehow, the universe creates stable, copyable facts. Without that, life and evolution are impossible.

1. Problem Statement

The paper addresses a foundational tension between Darwinian evolution and fundamental quantum mechanics.

The Conflict: Darwinian evolution relies on a cycle of stable records (genotypes), copying with variation, and irreversible selection (dissipation). However, standard quantum mechanics (QM) is governed by unitary dynamics, lacks a preferred basis, and is constrained by the No-Cloning and No-Deleting theorems.
The Core Question: In a world that is fundamentally quantum (with no preferred basis and only unitary evolution), how can the physical preconditions for evolution—specifically the existence of copyable records and irreversible deletion—be realized?
The Gap: While decoherence explains the emergence of stable "pointer states," it does not inherently select a unique outcome (history) or guarantee the existence of a "classical data sector" required for copying operations.

2. Methodology

The author employs a multi-faceted theoretical approach:

Categorical Quantum Mechanics (CQM): Used to formally define the structural requirements for "copying" and "deleting." CQM demonstrates that these operations are not generic quantum transformations but require a specific commutative structure (a classical data sector).
Ontological Analysis: The paper systematically separates the problem into three logically distinct questions:
1. Record Question: What guarantees a robust, copyable record sector (preferred basis)?
2. Outcome Question: Do interactions yield a single unique history or a multiplicity of decohered histories?
3. Agency Question: What physical resources allow for record-based processes (copying, filtering, agency)?
Stress-Testing via Extended Wigner's Friend (EWF): The author uses EWF scenarios to test various ontological frameworks. These scenarios force a simultaneous treatment of observers as (i) coherent quantum systems and (ii) agents with stable records, exposing contradictions in frameworks that cannot reconcile these views.
Stochastic Mechanics: The paper explores Nelson-type stochastic mechanics as a foundational alternative, utilizing variable diffusion parameters to bridge quantum and classical regimes.

3. Key Contributions and Theoretical Framework

A. The Necessity of a "Realized Classical Sector"

The paper argues that Darwinian evolution is only well-posed if a realized classical data sector exists.

CQM Insight: In CQM, copying ( $\Delta$ ) and deleting ( $\epsilon$ ) are defined only relative to a specific orthonormal basis (a "classical structure"). For a superposition $\alpha|0\rangle + \beta|1\rangle$ , copying yields an entangled state ( $\alpha|00\rangle + \beta|11\rangle$ ), not a cloned product state.
Implication: Without a physically selected preferred basis, "heredity" and "copying the same symbol" are ill-defined operations. Evolution requires a commutative sub-algebra within the quantum structure.

B. The Agency Constraint

Drawing on Adlam, McQueen, and Waegell, the paper introduces the Agency Constraint:

In a strictly coherent, basis-unselected "purely quantum" regime, minimal agency fails.
Agency requires world-model construction, deliberation, and reliable action selection. These processes fail without the ability to copy and store records (due to No-Cloning/Linearity).
Conclusion: Any world supporting evolution (or agency) must possess substantial classical resources (stable records) that are not merely emergent approximations but structurally necessary.

C. Four Ontological Options

The paper categorizes the main ontological packages based on how they answer the Record, Outcome, and Agency questions:

Option	Description	Mechanism for Records/Outcomes
A: Unique-History Realism	Single actual macroscopic history.	Objective Collapse (e.g., GRW) or Hidden Variables (e.g., de Broglie–Bohm). Outcomes are definite; records are grounded in beables/trajectories.
B: Unitary + Decohered Multiplicity	Many effectively classical histories.	Many-Worlds/Decoherence. A record basis emerges dynamically within each decohered sector, but no single outcome is selected globally.
C: Agent-Relative Facticity	Facts are relative to agents.	Relational/QBist. Outcomes are facts relative to specific interactions/agents, not global absolute facts. Records are what is stably communicable.
D: Stochastic Foundation	Classical stochastic trajectories as fundamental.	Nelson-type Mechanics. The Hilbert space is an emergent ensemble description. A single realized trajectory exists per run; quantum behavior emerges at specific diffusion strengths.

D. Stochastic Mechanics as a Unifying Bridge

The paper highlights Stochastic Mechanics (Option D) as a distinct foundational option:

Variable Diffusion: It posits a fundamental ontology of classical stochastic trajectories $x(t, \omega)$ $x (t, ω)$ with a diffusion parameter $\nu$ $ν$ .
- $\nu \to 0$ : Classical Hamilton-Jacobi behavior.
- $\nu \sim \hbar/2m$ : Standard quantum behavior.
Measurement Update: Collapse is not a primitive axiom but a consequence of Bayesian conditioning on the outcome, combined with a time-symmetric minimal-change principle (Pavon's rule). This selects the post-measurement diffusion path that minimally disturbs the pre-measurement dynamics.
Significance: This approach treats the quantum-classical divide as a continuous regime change rather than a fundamental split, while retaining an ontology of individual events.

4. Results and Analysis

Decoherence is Insufficient: While decoherence explains the selection of a pointer basis (solving part of the Record Question), it does not solve the Outcome Question (unique history) nor the Agency Question (providing the resources for copying) on its own.
The EWF Stress Test: Extended Wigner's Friend scenarios reveal that any consistent ontology must explicitly define:
1. Where the record basis comes from.
2. How (or if) a unique outcome is selected.
3. What physical resources underwrite the "facts" used by agents.
- Purely unitary descriptions (Option B) struggle to accommodate the "friend" as an agent with stable records without invoking additional structure or relativity of facts.
Structural Requirement: Evolution is not just a biological phenomenon but a structural requirement on physics. A world without a realized classical sector (a preferred basis) cannot support Darwinian evolution.

5. Significance

Foundational Clarity: The paper provides a rigorous structural argument for why "classicality" (in the form of a copyable record sector) is a prerequisite for complex processes like evolution and agency, moving beyond vague appeals to "macroscopic" scales.
Re-evaluating Interpretations: It forces interpretations of QM to explicitly address the "Agency Constraint." If an interpretation denies a unique history or a global preferred basis, it must explain how agents within that framework possess stable memories and perform record-based selection.
Stochastic Mechanics Revival: By framing stochastic mechanics as a continuous bridge between regimes where measurement update is a principled conditioning step, the paper offers a potential resolution to the measurement problem that avoids ad-hoc collapse postulates while maintaining an ontology of individual events.
Interdisciplinary Bridge: It connects biological evolution, information theory (CQM), and quantum foundations, suggesting that the laws of life are deeply constrained by the algebraic structure of the underlying physical theory.

In summary, Ghose argues that for Darwinian evolution to exist, the universe must possess a realized classical data sector that permits copying and deletion. This requirement acts as a filter for quantum interpretations, favoring those that can account for stable records and unique outcomes (or agent-relative facts) without violating the structural constraints of quantum mechanics.