pith. sign in

arxiv: 2512.18257 · v1 · submitted 2025-12-20 · ⚛️ physics.hist-ph · quant-ph

What is Stochastic Supervenience?

Youheng Zhang This is my paper

Pith reviewed 2026-05-16 21:12 UTC · model grok-4.3

classification ⚛️ physics.hist-ph quant-ph
keywords stochastic supervenienceMarkov kernelsdeterministic superveniencedependence relationsphysicalisminformation theorymultiple realizationstatistical mechanics
0
0 comments X

The pith

Higher-level properties supervene on base states through Markov kernels mapping to probability distributions, recovering deterministic supervenience as the Dirac limiting case.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a general framework for stochastic supervenience using Markov kernels to represent how base states determine distributions over higher-level configurations rather than fixed points. Axioms are formulated to ensure law-like fixation, nondegeneracy, and directional asymmetry in these dependence relations. This setup places classical deterministic supervenience as a limiting case within a topological space of possible dependence relations. Information-theoretic diagnostics are incorporated to connect the framework to empirical distinctions between structural stochasticity and epistemic uncertainty in domains like statistical mechanics.

Core claim

Stochastic supervenience is formalized via Markov kernels that map base states to probability distributions over macro-level configurations, satisfying axioms for law-like fixation, nondegeneracy, and directional asymmetry, such that deterministic supervenience appears as the limiting Dirac case in the topological space of dependence relations.

What carries the argument

Markov kernels equipped with axioms for law-like fixation, nondegeneracy, and directional asymmetry that define the space of stochastic dependence relations.

Load-bearing premise

Markov kernels together with the axioms for law-like fixation, nondegeneracy, and directional asymmetry adequately represent stochastic dependence relations in a non-circular way.

What would settle it

Discovery of a stochastic dependence relation that cannot be represented by any Markov kernel satisfying the three axioms, or a failure of deterministic supervenience to correspond to the Dirac limit in the defined topology.

read the original abstract

Standard formulations of supervenience typically treat higher level properties as point valued facts strictly fixed by underlying base states. However, in many scientific domains, from statistical mechanics to machine learning, basal structures more naturally determine families of probability measures than single outcomes. This paper develops a general framework for stochastic supervenience, in which the dependence of higher level structures on a physical base is represented by Markov kernels that map base states to distributions over macro level configurations. I formulate axioms that secure law like fixation, nondegeneracy, and directional asymmetry, and show that classical deterministic supervenience appears as a limiting Dirac case within the resulting topological space of dependence relations. To connect these metaphysical claims with empirical practice, the framework incorporates information theoretic diagnostics, including normalized mutual information, divergence based spectra, and measures of tail sensitivity. These indices are used to distinguish genuine structural stochasticity from merely epistemic uncertainty, to articulate degrees of distributional multiple realization, and to identify macro level organizations that are salient for intervention. The overall project offers a conservative extension of physicalist dependence that accommodates pervasive structured uncertainty in the special sciences without abandoning the priority of the base level.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The manuscript develops a framework for stochastic supervenience in which higher-level structures depend on a physical base via Markov kernels mapping base states to probability distributions over macro configurations. Axioms secure law-like fixation, nondegeneracy, and directional asymmetry; classical deterministic supervenience is recovered as a Dirac delta limit in the topological space of dependence relations. Information-theoretic diagnostics (normalized mutual information, divergence spectra, tail sensitivity) distinguish structural stochasticity from epistemic uncertainty and identify intervention-salient macro organizations.

Significance. If the axioms and limiting-case embedding hold without circularity, the work supplies a conservative extension of physicalist dependence that accommodates structured uncertainty in statistical mechanics, machine learning, and the special sciences while preserving base priority. The unified topological treatment of deterministic and stochastic cases, together with the empirical diagnostics, is a clear strength.

major comments (3)
  1. [Axioms for law-like fixation, nondegeneracy, and directional asymmetry] The law-like fixation axiom (formulated for the Markov kernel dependence relation) is load-bearing for the non-circularity claim yet appears to be satisfied precisely by kernels chosen to encode the target higher-level probabilities; no independent criterion is supplied for when a kernel counts as 'fixed by the base' rather than stipulated to match macro statistics. This risks reducing the construction to a description of structured uncertainty rather than a metaphysical definition of supervenience.
  2. [Limiting Dirac case within the topological space of dependence relations] The central claim that deterministic supervenience appears as the Dirac limit requires an explicit proof of the limiting behavior in the stated topology on the space of dependence relations; the abstract asserts the result but the manuscript must supply the convergence argument and confirm that the axioms are preserved in the limit.
  3. [Introduction and related-work discussion] The framework must include explicit comparisons to prior stochastic supervenience accounts (e.g., those using probability measures or conditional probabilities directly) to demonstrate that the Markov-kernel representation plus the three axioms adds non-redundant structure rather than re-describing existing proposals.
minor comments (3)
  1. Define the precise topology on the space of Markov kernels and state the continuity or convergence notion used for the Dirac limit.
  2. Clarify the normalization and interpretation of the information-theoretic indices when applied to finite versus continuous state spaces.
  3. Add a short table or diagram illustrating a simple base-to-macro kernel that satisfies the three axioms versus one that violates directional asymmetry.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed report. The comments identify key areas where the manuscript can be strengthened through clarification, added proofs, and expanded discussion. We respond point by point below and indicate planned revisions.

read point-by-point responses

  1. Referee: [Axioms for law-like fixation, nondegeneracy, and directional asymmetry] The law-like fixation axiom (formulated for the Markov kernel dependence relation) is load-bearing for the non-circularity claim yet appears to be satisfied precisely by kernels chosen to encode the target higher-level probabilities; no independent criterion is supplied for when a kernel counts as 'fixed by the base' rather than stipulated to match macro statistics. This risks reducing the construction to a description of structured uncertainty rather than a metaphysical definition of supervenience.

    Authors: We appreciate the referee's concern about potential circularity in the law-like fixation axiom. The axiom requires the Markov kernel to be a measurable map from the base sigma-algebra to the space of probability measures on macro configurations, with measurability taken in the weak topology; this criterion depends only on the base space's measurable structure and does not presuppose or fit to any particular macro-level probabilities. It is thus independent of the target statistics and encodes base priority directly. To make this distinction sharper, we will add a clarifying paragraph and a short example in Section 3 contrasting a kernel satisfying the axiom (derived from base dynamics) with one that merely matches macro frequencies without base measurability. revision: partial

  2. Referee: [Limiting Dirac case within the topological space of dependence relations] The central claim that deterministic supervenience appears as the Dirac limit requires an explicit proof of the limiting behavior in the stated topology on the space of dependence relations; the abstract asserts the result but the manuscript must supply the convergence argument and confirm that the axioms are preserved in the limit.

    Authors: We agree that an explicit convergence argument is required. The manuscript introduces the topology on the space of dependence relations (Markov kernels under weak convergence) but omits the full limiting proof for brevity. We will insert a dedicated subsection in Section 4 that proves Dirac kernels are the limit points of sequences of stochastic kernels and verifies that the three axioms (law-like fixation, nondegeneracy, directional asymmetry) are preserved under the limit operation. revision: yes

  3. Referee: [Introduction and related-work discussion] The framework must include explicit comparisons to prior stochastic supervenience accounts (e.g., those using probability measures or conditional probabilities directly) to demonstrate that the Markov-kernel representation plus the three axioms adds non-redundant structure rather than re-describing existing proposals.

    Authors: We accept that the introduction and related-work section require more explicit engagement with prior accounts. The current text mentions stochastic dependence ideas only briefly. We will expand the discussion to compare the Markov-kernel approach with accounts that employ conditional probabilities or direct probability measures on macro states (citing relevant work in philosophy of science and statistical mechanics). The revision will highlight the non-redundant contributions of the kernel formalism together with the three axioms and the topological embedding, which enable the unified treatment of deterministic and stochastic cases plus the information-theoretic diagnostics. revision: yes

Circularity Check

1 steps flagged

Stochastic supervenience defined via Markov kernels and axioms that encode the target dependence properties by construction

specific steps
  1. self definitional [Abstract]
    "This paper develops a general framework for stochastic supervenience, in which the dependence of higher level structures on a physical base is represented by Markov kernels that map base states to distributions over macro level configurations. I formulate axioms that secure law like fixation, nondegeneracy, and directional asymmetry, and show that classical deterministic supervenience appears as a limiting Dirac case within the resulting topological space of dependence relations."

    The dependence relation is introduced as Markov kernels that are then required to satisfy axioms for law-like fixation and directional asymmetry; the representation is therefore stipulated to possess exactly the metaphysical features it is meant to capture, rendering the framework equivalent to its input desiderata by construction rather than deriving fixation from any prior independent notion of base priority or physical law.

full rationale

The paper constructs its framework by representing dependence relations directly as Markov kernels and then formulating axioms (law-like fixation, nondegeneracy, directional asymmetry) that the kernels are required to satisfy in order to count as stochastic supervenience. This makes the central definition self-contained within the chosen mathematical objects and their stipulated properties, with deterministic supervenience recovered as a Dirac limit inside the same space. No external derivation, data, or independent criterion for 'fixation by the base' is supplied beyond the axioms themselves; the construction therefore matches the pre-specified metaphysical desiderata by design. The result is moderate definitional circularity without self-citation or fitted predictions.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the introduction of Markov kernels as the dependence relation and three custom axioms for law-like fixation, nondegeneracy, and directional asymmetry; no free parameters are fitted to data and no new physical entities are postulated.

axioms (1)
  • ad hoc to paper Axioms securing law-like fixation, nondegeneracy, and directional asymmetry for the Markov kernel dependence relation
    Formulated in the paper to ensure the stochastic dependence satisfies the intended metaphysical properties.
invented entities (1)
  • Markov kernels as the representation of stochastic dependence no independent evidence
    purpose: To map base states to probability distributions over macro-level configurations
    Core mathematical device introduced to generalize deterministic supervenience.

pith-pipeline@v0.9.0 · 5482 in / 1346 out tokens · 72053 ms · 2026-05-16T21:12:03.569284+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

  1. [1]

    structured uncertainty

    Discussion 3.1 Philosophical Implications The framework of stochastic supervenience reorients debates about inter-level dependence by identifying the proper relata of law-like constraint: not point-values, but stable families of probability distributions. This shift does more than merely accommodate the “structured uncertainty” found in quantum mechanics,...

    work page doi:10.1093/bjps/51.1.115 2009

  2. [2]

    https://doi.org/10.2307/2302572 Efron, B., & Tibshirani, R. J. (1994). An Introduction to the Bootstrap. Chapman and Hall/CRC. https://doi.org/10.1201/9780429246593 Fodor, J. A. (1974). Special Sciences (Or: The Disunity of Science as a Working Hypothesis). Synthese, 28(2), 97～115. 30 Frankish, K. (2007). The Anti‐Zombie Argument. The Philosophical Quarte...

    work page doi:10.2307/2302572 1994

  3. [3]

    https://doi.org/10.1007/s10670-007-9073-y Hubbell, S. P. (2001). The unified neutral theory of biodiversity and biogeography. Princeton University Press. http://archive.org/details/unifiedneutralth0000hubb Kass, R. E., & Raftery, A. E. (1995). Bayes Factors. Journal of the American Statistical Association, 90(430), 773～795. https://doi.org/10.1080/0162145...

    work page doi:10.1007/s10670-007-9073-y 2001

  4. [4]

    Supervenient and yet not deducible

    https://doi.org/10.2307/2107423 Kim, J. (1987). Strong and Global Supervenience Revisited. Philosophy and Phenomenological Research, 48(2), 315～326. https://doi.org/10.2307/2107631 Kim, J. (1990). Supervenience as a Philosophical Concept. Metaphilosophy, 21(1/2), 1～27. Kim, J. (2006). Emergence: Core ideas and issues. Synthese, 151(3), 547～559. https://do...

    work page doi:10.2307/2107423 1987

  5. [5]

    Almost-Objective

    https://doi.org/10.1007/s11098-006-9002-y Myrvold, W. C. (2021). Epistemic Chances, or “Almost-Objective” Probabilities. In W. C. Myrvold (eds.), Beyond Chance and Credence: A Theory of Hybrid Probabilities (pp. 106～121). 33 Oxford University Press. https://doi.org/10.1093/oso/9780198865094.003.0005 Napoletano, T. (2015). Compositionality as weak superven...

    work page doi:10.1007/s11098-006-9002-y 2021