Commitment Density Scan (CDS) — Version 0.2
Commitment Density Scan (CDS) — Version 0.2
Instrument architecture: Dave Kelly, 2026. Theoretical foundations: Grant C. Sterling, Eastern Illinois University. Prose rendering: Claude (Anthropic).
Purpose
The Commitment Density Scan tests whether a given clause of a passage requires one or more of Sterling’s six philosophical commitments (C1–C6) as a condition of possibility for its claim to be coherent. The standard is strict: not consistency with the commitment, not thematic relevance to it, but structural necessity — the commitment must be load-bearing at that specific clause or it is not marked Operative.
CDS is applicable to any primary text whose relationship to the six commitments is implicit rather than named — classical sources, field literature, thinker corpora. It is not the appropriate instrument for passages that already cite theorem numbers or name their own commitments explicitly; those require a Dependency Trace, not a density scan.
The Six Commitments
C1 — Substance Dualism. The rational faculty is ontologically distinct from the body and from all externals. The self that judges is not identical with what it judges.
C2 — Libertarian Free Will. The agent genuinely originates his assent. The act of giving or withholding assent is not determined by prior causes running through him.
C3 — Ethical Intuitionism. Foundational moral truths are directly accessible to the rational faculty. They are apprehended, not inferred from prior premises.
C4 — Foundationalism. Truths are organized in a stable dependency structure. Some truths are foundational — not revisable by the coherence of what is stacked on them.
C5 — Correspondence Theory of Truth. A judgement or impression is true if and only if it corresponds to reality. Truth is alignment with how things actually are, independent of preference or construction.
C6 — Moral Realism. Moral facts are objective. What is genuinely good or evil holds independently of what any agent, culture, or system believes, prefers, or constructs.
Procedure
Step A — Clause Segmentation. Divide the passage into clauses by independent assertion. Each clause must contain exactly one claim capable of being true or false, instructed, or causally asserted. Compound sentences are split at coordinating conjunctions where each half makes a distinct assertion. The segmentation is recorded before any testing begins.
Step B — Exhaustive Pairing. All six commitments are tested against every segmented clause without exception. No commitment is omitted on grounds that it “obviously” does not apply. This is the correction that distinguishes v0.2 from v0.1: selective testing is not permitted. The negative space — what is not Operative — is as important a finding as what is.
Step C — Operative Gate. A commitment is Operative at a clause only if removing it produces one of two named failures:
Subject failure — the clause requires an agent, faculty, or standard that does not exist without the commitment. Example: an instruction addressed to an agent who must genuinely originate his response fails as an instruction if C2 is removed, because there is no agent in the relevant sense.
Content failure — a key term in the clause changes meaning when the commitment is removed. Example: “falsely” in “falsely believing externals are good” softens to “unhelpfully” if C6 is removed, because without moral realism there is no objective fact the belief fails to match.
If removal of the commitment produces neither a Subject nor a Content failure, the verdict is Not Operative. Thematic relevance, background presence, and general consistency with the passage do not pass the gate.
Step D — Report All Verdicts. Every clause-commitment pairing receives a recorded verdict: Operative (with failure type named) or Not Operative (with brief justification). Not Operative verdicts are not omitted. The full table is the output, not a summary of what was found to be Operative.
Output Format
Each clause-commitment pairing is reported in the following locked format:
CLAUSE: [exact quoted text, under 20 words where possible] COMMITMENT TESTED: [C1 / C2 / C3 / C4 / C5 / C6] FAILURE TYPE IF REMOVED: [Subject / Content / Neither] VERDICT: [Operative / Not Operative] JUSTIFICATION: [one sentence naming the specific failure or its absence]
For full passage runs, verdicts for all six commitments against each clause may be presented in tabular form, with failure type noted inline for Operative verdicts.
Contamination Guard
Before finalizing any Operative verdict, the justification is scanned for the following failure patterns. Presence of any one requires rewrite or reclassification as Not Operative:
Associative justification. The justification invokes the commitment because it “fits the Stoic worldview” or is “consistent with the framework” rather than naming a specific failure produced by its removal.
Interchangeable justification. The justification could be applied to a different commitment without change. If the same sentence would work as justification for C1 and for C2, it has not located the specific clause-level dependency and must be rewritten.
Summary language. The justification uses phrases like “captures the spirit of,” “reflects the underlying commitment to,” or “is deeply connected with” rather than naming a determinate failure. These are contamination markers indicating that the analysis has not reached the level of specificity the gate requires.
Self-Audit
A mandatory self-audit is appended to every CDS run before output is finalized. The self-audit confirms:
(1) Step B was exhaustive: six verdicts appear for every clause, no commitment was tested against some clauses but not others.
(2) No commitment was marked Operative on associative grounds: every Operative verdict traces to a named Subject or Content failure.
(3) Justifications are clause-specific: no justification was reused verbatim across two different clauses without restating the distinct failure at each.
(4) Not Operative verdicts are present in the output: if every verdict across the full run is Operative, this is a signal of contamination and the run must be reviewed before finalizing.
Version Note
CDS v0.1 used selective testing — commitments were chosen for testing based on inspector judgment about which were likely relevant, and only Operative verdicts were reported. This produced a structurally incomplete output that could not be checked for exhaustiveness and was prone to selection bias toward expected findings. v0.1 is superseded and must not be used. v0.2 requires exhaustive pairing (Step B) and full reporting of all verdicts (Step D) as non-negotiable procedural requirements.
Relationship to Other Instruments
CDS operates at the clause level on primary texts whose commitment relationships are implicit. It is the appropriate instrument for Epictetan passages, field literature, thinker corpora under CPA, and any text where the presence of the six commitments is to be demonstrated rather than assumed.
The Dependency Trace is the appropriate instrument for passages that already cite theorem numbers or name their own derivations. The Dependency Trace tests whether a stated derivation actually holds without gaps — a different question from whether commitments are load-bearing in text that does not name them.
CDS findings feed into CPA and CFA work by supplying clause-level evidence for commitment verdicts that would otherwise rest on passage-level summary judgment. A CDS finding that C1 is Operative at a specific clause is harder to dispute than a CFA verdict that a field is Contrary on C1 without textual demonstration.
First Full Density Map
The first complete CDS v0.2 run covered the full text of The Little Enchiridion — Enchiridion Sections 1–5 and 30 — producing forty-eight clause-commitment pairings across eight clauses in Section 1 alone, and a complete exhaustive map across all six sections. The density pattern finding from that run is a ratified corpus result:
C2 (Libertarian Free Will) is the most frequently Operative commitment across the full text — present at nearly every instructional clause, confirming that every genuine instruction Epictetus issues requires an agent who can actually follow it.
C5 (Correspondence Theory) is the most evenly distributed, threading throughout all six sections wherever the text asserts that a judgement or classification can be correct or incorrect.
C1 (Substance Dualism) clusters wherever the text distinguishes the judging agent from what is judged — spiking at prohairesis clauses and at the harm-guarantee.
C4 (Foundationalism) clusters at structural and hierarchical claims: the foundational binary of Section 1, the rule-priority of the examination instruction, the unconditional guarantee, and the reserve clause.
C3 (Ethical Intuitionism) is the narrowest by frequency — Operative only at direct-recognition appeals and examination-instruction clauses. Four appearances across the full text.
C6 (Moral Realism) is the narrowest by domain — appearing only at clauses making unconditional claims about harm, dread, or moral obligation. Six appearances, each at a point of maximal moral stakes.
Instrument architecture: Dave Kelly, 2026. Theoretical foundations: Grant C. Sterling, Eastern Illinois University. Prose rendering: Claude (Anthropic).
0 Comments:
Post a Comment
<< Home