To Throw a Stone with Six Birds: On Agents and Agenthood

Ioannis Tsiokos

arxiv: 2604.03239 · v1 · submitted 2026-02-03 · 💻 cs.AI

To Throw a Stone with Six Birds: On Agents and Agenthood

Ioannis Tsiokos This is my paper

Pith reviewed 2026-05-16 08:17 UTC · model grok-4.3

classification 💻 cs.AI

keywords agencySix Birds Theoryviability kernelempowermentobjecthoodpersistencecontrolinduced closures

0 comments

The pith

Six Birds Theory defines an agent as a maintained theory object whose feasible policies steer outside futures while remaining viable.

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

The paper gives a type-correct account of agency inside Six Birds Theory by treating macroscopic objects as induced closures rather than primitives. This move separates the persistence of an object from its capacity to make a counterfactual difference through policy choices. A sympathetic reader would care because the separation supplies concrete, checkable criteria that can be audited without invoking goals, consciousness, or biological details. The account is made operational in finite controlled systems through four components whose behavior is demonstrated by matched-control ablations in a minimal ring-world.

Core claim

A theory induces a layer with an explicit interface and ledgered constraints. An agent is a maintained theory object whose feasible interface policies can steer outside futures while remaining viable. This contract is operationalized with four components: ledger-gated feasibility, a robust viability kernel computed as a greatest fixed point under successor-support semantics, feasible empowerment measured as channel capacity, and an empirical packaging map whose idempotence defect quantifies objecthood under coarse observation. Ablations on repair, protocols, and operator rewriting produce four separations that supply hash-traceable tests distinguishing agenthood from agency.

What carries the argument

The four operational components—ledger-gated feasibility, robust viability kernel as greatest fixed point, feasible empowerment as channel capacity, and empirical packaging map—together enforce the contract that a maintained theory object steers futures while staying viable.

If this is right

Enabling repair collapses the idempotence defect of the packaging map.
Protocols raise empowerment only at horizons of two or more steps.
Learning to rewrite operators increases median empowerment from 0.73 to 1.34 bits.
Calibrated null regimes with single actions exhibit zero empowerment and block model-misspecification false positives.
The four separations yield hash-traceable tests that distinguish agenthood from agency.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same ledger and viability checks could be applied to synthetic systems to test for agency without assuming internal goals or rewards.
The packaging map offers a route to quantify when coarse observation turns a collection of parts into a single persisting object.
Extending the successor-support semantics to continuous or stochastic dynamics would test whether the viability kernel remains computable outside finite rings.
The framework supplies a way to compare different agent architectures by measuring how their interface policies affect both empowerment and idempotence defect.

Load-bearing premise

Treating macroscopic objects as induced closures supplies a sound basis for separating persistence from control, and the four components fully capture the intended notion of agency.

What would settle it

A run of the ring-world ablations in which enabling repair fails to reduce the idempotence defect, or in which operator rewriting produces no monotonic rise in median empowerment, would falsify the claim that the four components capture agency.

Figures

Figures reproduced from arXiv: 2604.03239 by Ioannis Tsiokos.

**Figure 2.** Figure 2: Protocol holonomy yields horizon-dependent control. Median feasible empowerment (bits) [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗

**Figure 3.** Figure 3: Noise–maintenance sweep (phase diagram). [PITH_FULL_IMAGE:figures/full_fig_p024_3.png] view at source ↗

**Figure 4.** Figure 4: Operator rewriting (P1) increases difference-making. Median feasible empowerment (bits) at horizon H = 2 using output lens f(s) = y (outside position), grouped by discrete skill θ that reduces effective slip/noise in the ring-world kernel. Empowerment increases monotonically with θ, consistent with P1 as an induced law change rather than merely an internal record. 25 [PITH_FULL_IMAGE:figures/full_fig_p025… view at source ↗

read the original abstract

Six Birds Theory (SBT) treats macroscopic objects as induced closures rather than primitives. Empirical discussions of agency often conflate persistence (being an object) with control (making a counterfactual difference), which makes agency claims difficult to test and easy to spoof. We give a type-correct account of agency within SBT: a theory induces a layer with an explicit interface and ledgered constraints; an agent is a maintained theory object whose feasible interface policies can steer outside futures while remaining viable. We operationalize this contract in finite controlled systems using four checkable components: ledger-gated feasibility, a robust viability kernel computed as a greatest fixed point under successor-support semantics, feasible empowerment (channel capacity) as a proxy for difference-making, and an empirical packaging map whose idempotence defect quantifies objecthood under coarse observation. In a minimal ring-world with toggles for repair, protocol holonomy, identity staging, and operator rewriting, matched-control ablations yield four separations: calibrated null regimes with single actions show zero empowerment and block model-misspecification false positives; enabling repair collapses the idempotence defect; protocols increase empowerment only at horizons of two or more steps; and learning to rewrite operators monotonically increases median empowerment (0.73 to 1.34 bits). These results provide hash-traceable tests that separate agenthood from agency without making claims about goals, consciousness, or biological organisms, and they are accompanied by reproducible, audited artifacts.

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.

Desk Editor's Note private letter to a colleague

The paper introduces Six Birds Theory to define agency as a viable maintained theory object with four checkable components, tested via ring-world ablations that separate persistence from control.

read the letter

The main thing to know is that this paper builds Six Birds Theory around induced closures for macroscopic objects, then defines an agent as a maintained theory object whose interface policies steer futures while preserving viability. It breaks that into four explicit pieces—ledger-gated feasibility, viability kernel as greatest fixed point, empowerment via channel capacity, and packaging map idempotence defect—and runs matched ablations in a minimal ring-world to show separations like repair collapsing the defect and operator rewriting lifting median empowerment from 0.73 to 1.34 bits. The artifacts are reproducible and the tests stay inside the framework without invoking goals or biology. That is the actual new piece: a scoped, type-correct operationalization with hash-traceable results. It does a clean job of making the criteria checkable and avoiding the usual conflations between being an object and making a difference. The simulations are controlled enough to isolate effects like horizon-dependent protocol gains and null-regime false positives. The stress-test finds no load-bearing circularity or missing derivation step in the supplied text. The softer spots are that everything remains internal to SBT, so the results confirm consistency within the ledger rather than external applicability. The ring-world is deliberately minimal, which limits how far the separations generalize, and the empowerment gains are modest. The foundational move of treating objects as induced closures is presented as primitive, which is fine if you accept the setup but leaves the usual questions about scaling to real systems. This is for readers who work on formal agency models in AI or robotics and want non-anthropomorphic, testable criteria. Someone interested in conceptual modeling and simulation validation will find the operationalization useful. It shows clear thinking and honest scoping, so it deserves a serious referee even if revisions will be needed on generalization.

Referee Report

2 major / 2 minor

Summary. The paper introduces Six Birds Theory (SBT), which models macroscopic objects as induced closures rather than primitives. It offers a type-correct definition of agency inside SBT: a theory induces a layer with an explicit interface and ledgered constraints, and an agent is a maintained theory object whose feasible interface policies steer outside futures while remaining viable. This is operationalized via four checkable components (ledger-gated feasibility, viability kernel as greatest fixed point under successor-support semantics, feasible empowerment as channel capacity, and packaging map whose idempotence defect quantifies objecthood). The framework is demonstrated in a minimal ring-world simulation with toggles for repair, protocol holonomy, identity staging, and operator rewriting; matched-control ablations produce four separations, including zero empowerment in null regimes, repair collapsing the idempotence defect, horizon-dependent protocol effects, and monotonic empowerment increase (0.73 to 1.34 bits) under rewriting. The results are presented as hash-traceable tests that separate agenthood from agency without invoking goals, consciousness, or biology, accompanied by reproducible artifacts.

Significance. If the central construction holds, the work supplies a scoped, operational, and internally consistent account of agency that cleanly separates persistence from counterfactual control inside a ledgered theory framework. The use of explicit, checkable components (viability kernel as fixed point, empowerment as channel capacity, idempotence defect) together with controlled ablations and reproducible artifacts constitutes a genuine methodological contribution to formal AI and agency research. The absence of external premises about goals or biology makes the separations falsifiable within the constructed semantics and potentially useful as a benchmark for future agent modeling.

major comments (2)

The viability kernel is defined as the greatest fixed point under successor-support semantics, yet the manuscript does not supply the explicit inductive step or the precise successor relation used in the ring-world; without this derivation the claim that the kernel robustly captures viability (and thereby separates persistence from control) cannot be independently verified from the reported numbers alone.
Empowerment is reported to rise monotonically from 0.73 to 1.34 bits under operator rewriting, but the channel-capacity calculation (including the precise input/output alphabets and the horizon at which capacity is evaluated) is not shown; this step is load-bearing for the fourth separation and for the assertion that rewriting increases difference-making capacity.

minor comments (2)

The abstract and main text use the phrase 'hash-traceable tests' without defining what is hashed or how traceability is implemented; a short example or pseudocode would clarify the reproducibility claim.
Notation for the packaging map and its idempotence defect is introduced without an explicit equation or small worked example; adding one would improve readability for readers outside the immediate subfield.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading, positive assessment, and recommendation for minor revision. We address each major comment below and will incorporate the requested clarifications into the revised manuscript.

read point-by-point responses

Referee: The viability kernel is defined as the greatest fixed point under successor-support semantics, yet the manuscript does not supply the explicit inductive step or the precise successor relation used in the ring-world; without this derivation the claim that the kernel robustly captures viability (and thereby separates persistence from control) cannot be independently verified from the reported numbers alone.

Authors: We agree that the explicit inductive step and precise successor relation are required for independent verification. In the revised manuscript we will add the full derivation of the viability kernel as the greatest fixed point, including the successor-support semantics and the exact successor relation implemented for the ring-world simulation. This addition will allow readers to reproduce the kernel computation directly from the reported parameters and confirm the separation between persistence and control. revision: yes
Referee: Empowerment is reported to rise monotonically from 0.73 to 1.34 bits under operator rewriting, but the channel-capacity calculation (including the precise input/output alphabets and the horizon at which capacity is evaluated) is not shown; this step is load-bearing for the fourth separation and for the assertion that rewriting increases difference-making capacity.

Authors: We acknowledge that the channel-capacity details are essential for substantiating the reported monotonic increase. In the revision we will explicitly state the input and output alphabets, the horizon used for capacity evaluation, and the precise computational procedure (including any assumptions) that produced the values 0.73 to 1.34 bits. This will make the fourth separation and the effect of operator rewriting fully verifiable. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper introduces Six Birds Theory (SBT) as a novel framework in which macroscopic objects are treated as induced closures. It then defines agency operationally inside that framework as a maintained theory object whose interface policies steer futures while preserving viability. This definition is immediately operationalized via four explicit, checkable components (ledger-gated feasibility, viability kernel as greatest fixed point, empowerment as channel capacity, and packaging-map idempotence defect) and tested in a controlled ring-world simulation with ablations. No load-bearing step reduces by construction to a prior fitted parameter, a self-citation chain, or an ansatz smuggled from the authors' own prior work. The reported separations (null regimes, repair effect, horizon-dependent protocol effects, monotonic empowerment gain) are generated inside the constructed semantics and are presented as internal consistency checks rather than external predictions. The derivation chain is therefore self-contained within the newly introduced theory and does not exhibit any of the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 2 invented entities

The central claim rests on the introduction of Six Birds Theory as the background ontology; no explicit numerical free parameters are mentioned, but the framework itself functions as the primary invented structure.

axioms (1)

domain assumption Macroscopic objects are induced closures rather than primitives
Stated in the opening of the abstract as the foundational premise of SBT.

invented entities (2)

Six Birds Theory (SBT) no independent evidence
purpose: Provides the ontological layer in which agency is defined
New framework introduced to treat objects as induced closures
Maintained theory object as agent no independent evidence
purpose: Core definition of agency
Postulated entity whose policies steer futures while remaining viable

pith-pipeline@v0.9.0 · 5552 in / 1355 out tokens · 22890 ms · 2026-05-16T08:17:56.784765+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean and Foundation modules on fixed-point/closure forcing reality_from_one_distinction and iterate_top_greatest_fixpoint-style closure theorems echoes

?

echoes
ECHOES: this paper passage has the same mathematical shape or conceptual pattern as the Recognition theorem, but is not a direct formal dependency.

viability kernel K is the greatest fixed point of V ... computed by iterating V from the top safe set until convergence ... Lean lemma ... iterate_top_greatest_fixpoint (F : Finset α → Finset α) (hmono : Monotone F) (hsub : ∀ s, F s ⊆ s)

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

7 extracted references · 7 canonical work pages

[1]

Jean-Pierre Aubin.Viability Theory

doi: 10.1109/TIT.1972.1054753. Jean-Pierre Aubin.Viability Theory. Birkh¨ auser, Boston,

work page doi:10.1109/tit.1972.1054753 1972
[2]

Leonardo de Moura and Sebastian Ullrich

doi: 10.1109/TIT.1972.1054855. Leonardo de Moura and Sebastian Ullrich. The lean 4 theorem prover and programming language. In Andr´ e Platzer and Geoff Sutcliffe, editors,Automated Deduction – CADE 28, pages 625–635, Cham,

work page doi:10.1109/tit.1972.1054855 1972
[3]

doi: 10.1007/978-3-030-79876-5

Springer International Publishing. doi: 10.1007/978-3-030-79876-5

work page doi:10.1007/978-3-030-79876-5
[4]

Claude E

doi: 10.1109/CEC.2005.1554676. Claude E. Shannon. A mathematical theory of communication.Bell System Technical Journal, 27 (3):379–423,

work page doi:10.1109/cec.2005.1554676 2005
[5]

The Bell System Tech- nical Journal27(3), 379–423 (1948) https://doi.org/10.1002/j.1538-7305.1948

doi: 10.1002/j.1538-7305.1948.tb01338.x. Part II: vol. 27, no. 4, pp. 623–656, DOI: 10.1002/j.1538-7305.1948.tb00917.x. Alfred Tarski. A lattice-theoretical fixpoint theorem and its applications.Pacific Journal of Mathematics, 5(2):285–309,

work page doi:10.1002/j.1538-7305.1948.tb01338.x 1948
[6]

A lattice-theoretical ﬁxpoint theorem and its applications,

doi: 10.2140/pjm.1955.5.285. The mathlib Community. The lean mathematical library

work page doi:10.2140/pjm.1955.5.285 1955
[7]

Ioannis Tsiokos

doi: 10.48550/arXiv.1910.09336. Ioannis Tsiokos. To wake a stone with six birds: A life is a theory, Jan 2026a. Preprint. Ioannis Tsiokos. Six birds: Foundations of emergence calculus, Jan 2026b. Preprint. 26

work page doi:10.48550/arxiv.1910.09336 1910

[1] [1]

Jean-Pierre Aubin.Viability Theory

doi: 10.1109/TIT.1972.1054753. Jean-Pierre Aubin.Viability Theory. Birkh¨ auser, Boston,

work page doi:10.1109/tit.1972.1054753 1972

[2] [2]

Leonardo de Moura and Sebastian Ullrich

doi: 10.1109/TIT.1972.1054855. Leonardo de Moura and Sebastian Ullrich. The lean 4 theorem prover and programming language. In Andr´ e Platzer and Geoff Sutcliffe, editors,Automated Deduction – CADE 28, pages 625–635, Cham,

work page doi:10.1109/tit.1972.1054855 1972

[3] [3]

doi: 10.1007/978-3-030-79876-5

Springer International Publishing. doi: 10.1007/978-3-030-79876-5

work page doi:10.1007/978-3-030-79876-5

[4] [4]

Claude E

doi: 10.1109/CEC.2005.1554676. Claude E. Shannon. A mathematical theory of communication.Bell System Technical Journal, 27 (3):379–423,

work page doi:10.1109/cec.2005.1554676 2005

[5] [5]

The Bell System Tech- nical Journal27(3), 379–423 (1948) https://doi.org/10.1002/j.1538-7305.1948

doi: 10.1002/j.1538-7305.1948.tb01338.x. Part II: vol. 27, no. 4, pp. 623–656, DOI: 10.1002/j.1538-7305.1948.tb00917.x. Alfred Tarski. A lattice-theoretical fixpoint theorem and its applications.Pacific Journal of Mathematics, 5(2):285–309,

work page doi:10.1002/j.1538-7305.1948.tb01338.x 1948

[6] [6]

A lattice-theoretical ﬁxpoint theorem and its applications,

doi: 10.2140/pjm.1955.5.285. The mathlib Community. The lean mathematical library

work page doi:10.2140/pjm.1955.5.285 1955

[7] [7]

Ioannis Tsiokos

doi: 10.48550/arXiv.1910.09336. Ioannis Tsiokos. To wake a stone with six birds: A life is a theory, Jan 2026a. Preprint. Ioannis Tsiokos. Six birds: Foundations of emergence calculus, Jan 2026b. Preprint. 26

work page doi:10.48550/arxiv.1910.09336 1910