O0-MATH-008

Nested Boundary Algebra

**Version:** 1.0

**Research status:** ACTIVE

CLAIM STATUS: EXPLORATORY FORMALIZATION
EVIDENCE TYPE: MATHEMATICAL FORMALIZATION
PHYSICAL VALIDATION: NONE
INDEPENDENT REPLICATION: N/A
PHILOSOPHICAL PROVENANCE: O/0 ARCHIVE
ARCHIVE ENDORSEMENT: LIMITED TO REPORTED RESULT

Abstract

This document defines algebraic operations on boundaries in the O/0 substrate: composition, nesting, merger, and fission. We investigate the closure properties of these operations and determine what algebraic structure the set of boundaries carries. The main results are: (1) the set of boundaries under composition forms a monoid; (2) boundaries can be organized into a category where morphisms are boundary interactions; (3) nesting defines a partial order on boundaries; and (4) merger and fission satisfy partial group-like axioms but lack full inverses. Several algebraic axioms fail, and we characterize the failures precisely.

Source proposition

The O/0 archive describes boundaries as dynamic entities that can nest within each other, merge, split, and interact. This document asks whether these operations satisfy any systematic algebraic laws.

Scientific audit

Algebraic formalization of dynamical structures is standard mathematical practice (e.g., the algebra of topological spaces, the category of dynamical systems). The specific algebraic structures we discover will constrain what kinds of theories are consistent with the simulation data. This is pure mathematics motivated by the substrate framework; no empirical claims are made.

Research question

What algebraic structure does the set of boundaries in the O/0 substrate possess? Specifically: Is the set of boundaries closed under composition, merger, and fission? Does it form a group, monoid, semiring, or category?

Operational definitions

1. **Boundary**: A connected subset B of the substrate that satisfies the Markov blanket criterion (O0-MATH-002) for some interior region.

2. **Interior**: For boundary B, the connected component of M \ B that is "enclosed" (in oriented manifolds) or designated by the PEM dynamics.

3. **Exterior**: The complement of interior union B.

4. **Boundary space**: The set B = {B_1, B_2, ...} of all boundaries present in the substrate at a given time.

Hypothesis

The set of boundaries under appropriate operations forms a structured algebraic object (at minimum a monoid, possibly a category) with composition as the primary operation and nesting as a partial order.

Null hypothesis

Boundary operations are chaotic and satisfy no systematic algebraic laws; the set of boundaries has no useful algebraic structure beyond being a set.

Competing explanations

1. The algebraic structure may be trivial (e.g., a trivial monoid with only one element in practice).

2. The structure may depend on the specific substrate parameters rather than being universal.

3. Multiple incompatible algebraic structures may be simultaneously applicable, with no single "correct" one.

Formal model

**Definition 1 (Boundary Object).** A *boundary object* is a quadruple b = (B, I, E, q) where:

• B subset M is the boundary set (connected subset of the substrate manifold),
• I subset M is the interior (connected component of M \ B enclosed by B),
• E = M \ (I union B) is the exterior,
• q in R_+ is the boundary quality (from O0-MATH-002).

Two boundary objects b_1 = (B_1, I_1, E_1, q_1) and b_2 = (B_2, I_2, E_2, q_2) are *disjoint* if (I_1 union B_1) intersection (I_2 union B_2) = empty set.

**Definition 2 (Composition).** For two disjoint boundary objects b_1 and b_2, their *composition* is:

b_1 * b_2 = (B_1 union B_2, I_1 union I_2, E_1 intersection E_2, min(q_1, q_2))

This represents two independent boundaries coexisting in the substrate.

**Proposition 1 (Monoid Under Composition).** Let (B_disjoint, *) be the set of all collections of pairwise disjoint boundary objects with composition as above. Then:

• Closure: the composition of disjoint boundaries is a valid boundary collection.
• Associativity: (b_1 * b_2) * b_3 = b_1 * (b_2 * b_3) when all pairs are disjoint.
• Identity: the *null boundary* e = (empty, empty, M, infinity) satisfies e * b = b * e = b for all b.
• NOT commutative in general when boundary quality affects dynamics.

Therefore (B_disjoint, *) is a monoid.

*Proof sketch.* Closure: the union of disjoint Markov blankets is a Markov blanket for the union of their interiors (conditional independence is preserved under union of disjoint conditionally independent sets). Associativity: set union is associative. Identity: the empty boundary encloses nothing and does not affect any other boundary. QED

**Definition 3 (Nesting).** Boundary b_1 is *nested inside* boundary b_2, written b_1 <= b_2, if I_1 union B_1 subset I_2 (the entirety of b_1 is contained in the interior of b_2).

**Proposition 2 (Partial Order).** (B, <=) is a partially ordered set:

• Reflexivity: b <= b (trivially, since I union B subset I by convention when considering degenerate nesting).
• Antisymmetry: b_1 <= b_2 and b_2 <= b_1 implies I_1 union B_1 subset I_2 and I_2 union B_2 subset I_1, which for non-degenerate boundaries (non-empty interiors and boundaries) is impossible. Hence b_1 = b_2.
• Transitivity: b_1 <= b_2 and b_2 <= b_3 implies I_1 union B_1 subset I_2 subset I_3, so b_1 <= b_3.

**Definition 4 (Nesting Depth).** The *nesting depth* of boundary b is:

depth(b) = max{n : there exist b_1 < b_2 < ... < b_n = b}

The global nesting depth of the substrate is:

D = max_{b in B} depth(b)

**Definition 5 (Merger).** Two boundary objects b_1 and b_2 with adjacent or overlapping boundaries (B_1 intersection B_2 != empty or dist(B_1, B_2) < delta for some threshold delta) can *merge* into:

b_1 + b_2 = (B_merged, I_1 union I_2 union (B_1 intersection I_2) union (B_2 intersection I_1), E_merged, q_merged)

where B_merged is the outer boundary of the connected region I_1 union B_1 union I_2 union B_2, and q_merged is determined by the PEM dynamics on the new configuration.

**Proposition 3 (Merger Properties).**

• Merger is commutative: b_1 + b_2 = b_2 + b_1 (set union is commutative).
• Merger is associative: (b_1 + b_2) + b_3 = b_1 + (b_2 + b_3) when all mergers are well-defined.
• Merger has NO identity element: there is no boundary e such that b + e = b for all b (merger always changes the boundary set).
• Merger has NO inverses: there is no b' such that b + b' = e for any non-trivial b.

Therefore, (B_adjacent, +) is a commutative semigroup but NOT a monoid or group.

**Definition 6 (Fission).** A boundary object b = (B, I, E, q) can undergo *fission* into b_1 and b_2 if:

• I can be partitioned into I_1, I_2, B_new with I = I_1 union B_new union I_2 (disjoint),
• B_1 = B_new union (B intersection closure(I_1)) and B_2 = B_new union (B intersection closure(I_2)) are valid Markov blankets.

The fission operator is written: b -> (b_1, b_2) = fission(b).

**Proposition 4 (Fission Properties).**

• Fission is NOT unique: a given b can fission in multiple ways (different partitions of I).
• Fission is NOT the inverse of merger in general: if b_1 + b_2 = b, then fission(b) may yield (b_1', b_2') != (b_1, b_2).
• Fission preserves total interior area: |I_1| + |I_2| + |B_new| = |I|.
• Fission generically increases total boundary perimeter: |B_1| + |B_2| > |B|.

**Theorem 1 (No Group Structure).** The set of boundaries B does not form a group under any of the defined operations (composition, merger, fission). Specifically:

• Under composition: no inverses (you cannot "undo" a boundary by composing with another).
• Under merger: no identity and no inverses.
• Fission is not a binary operation (it maps one boundary to two).

*Proof.* For composition: suppose b * b' = e. Then I union I' = empty and B union B' = empty, requiring b = b' = e. So only the identity has an inverse. For merger: suppose b + b' = e. Then B_merged = empty, requiring both B_1 and B_2 to be empty, so b = b' = trivial. No non-trivial inverse exists. QED

**Definition 7 (Boundary Category).** Define the category Bound with:

• Objects: boundary objects b = (B, I, E, q).
• Morphisms: a morphism f: b_1 -> b_2 is a continuous, quality-non-decreasing transformation of the substrate that maps b_1's configuration to b_2's. This includes:
• Nesting: if b_1 <= b_2, the inclusion is a morphism.
• Merger: the merger operation b_1 + b_2 -> b_merged defines morphisms from b_1 and b_2 to b_merged.
• Fission: fission(b) -> (b_1, b_2) defines a morphism from b to the product b_1 x b_2 (if the category has products).
• Composition of morphisms: sequential transformation.
• Identity morphism: the trivial transformation (no change).

**Proposition 5 (Categorical Properties).**

• Bound has an initial object: the null boundary e.
• Bound has a terminal object: the "total boundary" B_max = (M, empty, empty, 0) (the entire manifold is boundary, nothing is interior).
• Bound has finite products: b_1 x b_2 = b_1 * b_2 (composition of disjoint boundaries) when they are disjoint.
• Bound does NOT have all coproducts (merger is not always defined).
• Bound is NOT cartesian closed (the internal hom [b_1, b_2] does not always exist).

**Open Question 1.** Does the boundary category admit a monoidal structure with a braiding (symmetry)? If so, boundary interactions may be modeled using string diagrams.

**Open Question 2.** Is there a natural functor from Bound to the category of finite-dimensional Hilbert spaces, connecting boundary algebra to quantum information theory?

**Open Question 3.** Can the merger and fission operations be unified into a single algebraic structure (e.g., a Frobenius algebra or a bialgebra)?

Methods

1. Enumerate all boundaries in simulation data at convergence.

2. Test closure under composition by artificially combining boundary configurations.

3. Test merger by initializing two adjacent agents and observing whether their boundaries merge.

4. Test fission by initializing a large agent and observing whether its boundary splits.

5. Verify monoid axioms (associativity, identity) computationally.

Controls

1. Test with exactly two disjoint boundaries to verify composition closure.

2. Test with exactly two adjacent boundaries to verify merger semigroup properties.

3. Use a known algebraic structure (e.g., the boundary of convex sets in R^2, which is a well-understood monoid) as a positive control.

Predictions

1. Composition will satisfy monoid axioms in all tested cases.

2. Merger will be commutative and associative when defined, but the set of "mergeable" pairs will be a proper subset of all pairs.

3. Fission will be non-unique, with the number of possible fission modes growing with interior area.

4. No group structure will be observed (consistent with Theorem 1).

Falsification criteria

1. If composition fails associativity, the monoid claim is wrong.

2. If merger fails commutativity, the symmetric property of set union is not preserved (suggesting a dependence on PEM dynamics ordering).

3. If an inverse operation is found for any non-trivial boundary, Theorem 1 has an error.

Results / Expected Outcomes

No simulation results reported. The algebraic framework is self-contained. Preliminary analysis suggests the monoid structure is robust. The categorical structure is speculative and requires further development.

Uncertainty

1. The boundary quality q_merged after merger is determined dynamically (by PEM convergence), not algebraically, introducing dependence on simulation specifics.

2. The categorical framework may be overly abstract for the current state of the research program.

3. The definition of "adjacency" for merger involves a threshold delta that affects which mergers are permitted.

Limitations

1. The algebra is defined for boundaries at a fixed time. Time evolution introduces additional structure (a dynamical category) not treated here.

2. The categorical framework is sketched, not fully developed; verifying all axioms requires substantial further work.

3. The operations are defined for 2D substrates; extension to higher dimensions may require modification.

4. The algebra does not capture continuous deformation of boundaries, only discrete operations (merge, split, nest).

Replication status

Algebraic definitions and proofs are self-contained and independently verifiable.

Data and code

No data or code generated. Algebraic verification to be implemented as symbolic computation in Python/SymPy.

Relationship to philosophical archive

The archive speaks of boundaries as dynamic, interacting entities — nesting, merging, dissolving. The algebraic framework provides a precise language for these processes. The fact that boundaries form a monoid (not a group) means that boundary formation is irreversible in a precise algebraic sense: you can create and combine boundaries, but you cannot algebraically "undo" them. Whether this has philosophical significance is a matter of interpretation, not mathematics.

References

1. Mac Lane, S. (1998). *Categories for the Working Mathematician* (2nd ed.). Springer.

2. Leinster, T. (2014). *Basic Category Theory*. Cambridge University Press.

3. Coecke, B., & Kissinger, A. (2017). *Picturing Quantum Processes*. Cambridge University Press.

4. Baez, J. C., & Stay, M. (2011). "Physics, topology, logic, and computation: a Rosetta Stone." *New Structures for Physics*, Lecture Notes in Physics, 813, 95–172.

5. Kock, J. (2004). *Frobenius Algebras and 2D Topological Quantum Field Theories*. Cambridge University Press.

Revision history

• v1.0: Initial formalization. Composition monoid established. Nesting partial order defined. Merger semigroup and fission operator characterized. Boundary category sketched. Three open questions raised.