Controlled Experiment on Minimality

• The paper demonstrates that minimal objects in recursively enumerable theories, set forcing, and CNFs typically lack absolute minimality, revealing dense and non-minimal degree structures.
• It employs rigorous parameter control and constructive techniques—such as fusion in forcing and partition-based CNF construction—to explore minimality across diverse logical settings.
• Empirical findings indicate that as problem parameters scale, minimal unsatisfiability and absence of true minimal elements become prevalent, guiding future refinements in theory.

A controlled experiment on minimality investigates, with rigorous parameter control, the existence and structure of minimal objects within a given mathematical or logical context, such as theories exhibiting Gödel incompleteness, forcings in set theory, or minimally unsatisfiable structures in propositional logic. Three distinct strands from recent research—Cheng's study of first incompleteness and recursively enumerable (RE) theories, set-theoretic perfect tree forcing, and algorithmic experiments for minimal unsatisfiability in CNFs—provide comprehensive models for such experimental approaches to minimality, combining precise definitions, construction techniques, structural analysis, and empirical verification.

1. Theoretical Frameworks for Minimality

Minimality takes varied precise forms across logical and combinatorial settings. In the context of Gödel incompleteness, minimality can be formulated via Turing degree (reducibility of theorem sets) or interpretability (logical strength via mutual interpretation of theories) (Cheng, 2021). For forcing notions in set theory, minimality refers to the property that a generic extension introducing a new object (e.g., an $\omega$-sequence) contains no generically added objects of "intermediate" complexity (Levine et al., 2021). In the field of Boolean satisfiability, minimal unsatisfiability means that removal of any clause from a CNF formula results in satisfaction, ensuring irredundancy relative to unsatisfiability (Cowen, 2012).

2. Formalization and Definitions

For recursively enumerable theories:

• $G_1$ holds for an RE theory $T$ if every consistent RE theory $S$ interpreting $T$ is incomplete; equivalently, $T$ is essentially undecidable.
• Turing-degree minimality: A theory $U$ is minimal (among RE theories) w.r.t. Turing reducibility if there is no RE theory $W$ with $W <_T U$.
• Interpretation-degree minimality: A theory $U$ is minimal w.r.t. interpretability if no $G_1$0 exists with $G_1$1.

For perfect tree forcing $G_1$2:

• $G_1$3 consists of perfect trees (both $G_1$4-closed and ever-branching) in $G_1$5, where $G_1$6 is a sequence of regulars up to singular $G_1$7 with $G_1$8.
• Minimality for $G_1$9-sequences: Forcing with $T$0 is minimal if any new sequence $T$1 in the generic extension already generates the extension, $T$2.

For minimally unsatisfiable CNFs:

• A formula $T$3 is minimally unsatisfiable (MU) if $T$4 is unsatisfiable and for every clause $T$5, $T$6 is satisfiable.

3. Negative Results and Absence of Minimal Elements

Extensive controlled investigations reveal broad nonexistence of genuinely minimal structures:

• In RE theory: There is no minimal essentially undecidable theory under interpretation; for every such theory, a strictly weaker essentially undecidable theory exists. The general schema (if class $T$7 is closed under $T$8, all $T$9-theories are essentially undecidable, and all effectively inseparable theories are $S$0) entails that no $S$1-minimal exists under interpretability (Cheng, 2021).
• In perfect-tree forcing: When $S$2 and $S$3 is a $S$4-strong limit, no minimality for $S$5-sequences is achievable; the forcing fails $S$6-distributivity and collapses $S$7 to $S$8 (Levine et al., 2021).
• For minimally unsatisfiable CNFs: While individual instances can be MU, these are only "minimal" relative to clause deletion and not to broader notions of redundancy or independence.

4. Constructive Techniques and Exemplary Constructions

Several highly controlled construction paradigms underpin the experimental exploration of minimality:

Construction Component	Domain	Purpose
Janiczak’s theory $S$9	RE-theories	Decidable base with independent formulas $T$0
Shoenfield’s recursion transfer	RE-theories	Transfers RE degree into theory properties
Bernstein labelling	Forcing	Produces names witnessing collapse/lack of minimality
Partition-based CNF construction	SAT/Combinatorics	Generates MU-CNFs by block-wise clause generation

• Janiczak + Shoenfield: Given a non-recursive $T$1, $T$2 assembles essentially undecidable RE theories at prescribed Turing degrees, producing rich degree structures and demonstrating density and absence of minimal elements both in Turing and interpretation degree orderings (Cheng, 2021).
• Fusion/finiteness in forcing: $T$3 employs fusion and thinning at splitting nodes to ensure $T$4-distributivity and minimality for countable cofinality, but analogous arguments break down in the uncountable case (Levine et al., 2021).
• Random permutation/partition for CNFs: For clause-size $T$5 and block parameter $T$6, variables are partitioned, and all $T$7-clauses from each block (and permutations thereof) are used to build MU-CNFs with high probability as $T$8 increases (Cowen, 2012).

5. Quantitative and Empirical Results

Empirical studies complement structural theorems with concrete data:

Minimally Unsatisfiable CNF Construction (for $T$9):

$T$0	Clause Count	MU Percentage	Mean Sat. Number	Std Dev
5	52	65.3%	51.3	0.71
6	60	75.6%	59.6	0.89
7	68	91.6%	67.6	0.89
8	76	95.8%	75.7	0.71
9	84	98.4%	83.4	0.66
10	92	99.8%	91.7	0.93
11	100	100%	99.8	0.82

The fraction of MU instances increases rapidly with $T$1, approaching certainty for larger clause sets. Even non-MU instances typically require removal of only one additional clause to be satisfiable, indicating highly "compressed" unsatisfiability (Cowen, 2012).

This suggests that the combinatorial construction method efficiently saturates the space of minimal unsatisfiable formulas as the block parameter grows.

6. Lattice Structures and Complexity Implications

• The set of all RE theories under interpretability forms a dense distributive lattice, with explicit $T$2 and $T$3 operations providing infima and suprema, bottom given by theories interpreting Robinson’s $T$4, and top by the inconsistent theory (Cheng, 2021).
• Arithmetical complexity: the set $T$5 is $T$6-complete, reflecting the intricate recursion-theoretic landscape underlying minimality phenomena.
• In forcing, generalizations and variants involving closure and splitting yield a range of distributivity behaviors, with minimality phenomena tightly controlled by fusion properties, closure (vertical/splitting), and the cofinality of the relevant cardinals (Levine et al., 2021).

7. Broader Impact and Future Directions

The experimental paradigm elucidates that minimality, in many rigorous senses, is unattainable for a broad spectrum of nontrivial mathematical properties—be it incompleteness, undecidability, or sequence-generation in extensions—except under sharply delineated parameters. Every nontrivial object (e.g., essentially undecidable RE theory, nontrivial perfect-tree forcing, MU-CNF) can typically be "weakened" further without loss of the focal property (Cheng, 2021, Levine et al., 2021, Cowen, 2012). Practical implications include the recognition that degree structures (Turing, interpretability) are densely ordered with no minimal elements beyond certain canonical bounds (e.g., Robinson’s theory $T$7), and that algorithmic generation of minimal objects tends to saturate as problem size increases.

A plausible implication is that future research may need to either seek alternative minimality criteria, study relative instead of absolute minimality, or focus on finer invariants capturing "irreducibility" within specific subclasses. The lattice-theoretic and probabilistic methods developed in these controlled experiments provide systematic frameworks for further exploration across logic, combinatorics, and set theory.