The Classical Judgment Recovery Instrument (CJRI) — Version 1.3
The Classical Judgment Recovery Instrument (CJRI) — Version 1.3
Theoretical foundations: Grant C. Sterling (Eastern Illinois University). Analysis and synthesis: Dave Kelly. Prose rendering: Claude (Anthropic). 2026.
Corpus in use: Core Stoicism, SLE v4.3, Nine Excerpts, THE SIX PHILOSOPHICAL COMMITMENTS THAT GROUND STOIC PRACTICE, Seddon’s Glossary, Sterling’s Framework for Personal Decision Making v3.3.
Version Note
v1.3. Corrects the commitment letters in Step Three. Versions 1.0 through 1.2 assigned grounding by file-order numbering rather than canonical numbering. The substance of all five detection criteria is unaffected — the descriptions were correct and only the letters were wrong. A seventh self-audit item is added to prevent recurrence. No registered runs inherited the error. The attribution line is normalized to the invariant tripartite form, and the corpus-in-use line is updated from SLE v4.0 to v4.3.
The canonical numbering, authoritative over any file-order numbering still in circulation: C1 Substance Dualism, C2 Libertarian Free Will, C3 Ethical Intuitionism, C4 Foundationalism, C5 Correspondence Theory of Truth, C6 Moral Realism.
Purpose
The Classical Judgment Recovery Instrument (CJRI) locates and characterizes instances of genuine rational agency in any text, regardless of subject matter. Its output is designed for curation: each confirmed instance is portable, characterized by judgment type, and grounded in one or more of the six classical philosophical commitments. Over time, confirmed instances accumulate into a standing collection — a recovery of genuine rational agency from across the full range of human discourse.
The CJRI is a harvesting instrument. Every other instrument in the corpus audits for failure, demonstrates grounding, or governs engagement. The CJRI collects what is genuinely present. Subject matter is not the target. Reasoning moves are the target.
The Detection Problem
Texts on any subject contain four kinds of apparent judgment: genuine rational agency, rule-following, expertise-pattern-matching, and ideologically-scripted response. The first is the instrument’s target. The other three are look-alikes, joined by two further failures identified through initial instrument runs: tradition-transmission and assertion-without-grounding. None of these six are distinguished by subject matter, by the confidence of the author’s tone, or by the sophistication of the vocabulary deployed. They are distinguished by whether the reasoning move satisfies the detection criteria derived from the six classical commitments.
The detection criteria do not ask what the text is about. They ask whether a rational faculty is genuinely operative in a specific reasoning move — whether the pause between impression and assent is structurally present in what the text records.
Procedure
Step One: Topic Suspension
The core question is: Have I set aside the subject matter of the text as a detection criterion?
Read the text without topic-sensitivity. A text about economics, politics, science, craft, law, medicine, or any other domain is equally eligible. The CJRI does not favor texts whose subject matter resembles philosophy. It looks for reasoning moves, wherever they occur, in whatever domain the text inhabits.
Proceed only when the subject matter has been suspended as a filter.
Step Two: Candidate Identification
The core question is: Which passages contain a reasoning move rather than a bare assertion or a reported outcome?
A candidate passage is one in which an author or an agent described by the author makes a move that could in principle be wrong — where something is being weighed, distinguished, refused, or arrived at rather than merely stated or executed. Bare assertions (“X is true”), reported outcomes (“he decided to leave”), and protocol executions (“she followed the procedure”) are not candidates unless the reasoning move that produced them is also present in the text.
Flag all candidate passages before proceeding to Step Three. Do not apply detection criteria during candidate identification; the two operations are kept separate to prevent confirmation bias.
Step Three: Detection
The core question is: Which of the five judgment types is present in this candidate passage, and which commitment grounds the identification?
Apply each detection criterion in turn to each candidate passage. A passage may satisfy more than one criterion; all satisfied types are recorded. A passage that satisfies none is not a confirmed instance.
Agency-Judgment (grounded in C1 and C2): The reasoning move presupposes a genuine rational faculty making a real choice. Deliberation is recorded, not merely outcome. The agent could have chosen otherwise; the text gives evidence that the alternative was genuinely available and was declined. The pause between impression and assent is structurally present: something was weighed before something was affirmed. Passages in which an agent simply executes a role, follows a script, or responds automatically do not satisfy this criterion even when the outcome is correct.
Moral-Apprehension-Judgment (grounded in C6): The reasoning move treats a practical or moral fact as real and independent of the agent’s preference. The agent is not constructing a value; he is recognizing one. The judgment would not change if the agent’s desires changed, because the fact being apprehended is not a function of those desires. Passages in which an agent arrives at a value conclusion by aggregating preferences — his own or others’ — do not satisfy this criterion.
Correspondence-Judgment (grounded in C5): The conclusion is tethered to how things actually are. The agent is tracking reality rather than constructing a preferred version of it. The move is falsifiable in principle: if things were otherwise, the conclusion would be otherwise. Passages in which a conclusion is rationalized after the fact, or in which the reasoning would reach the same conclusion regardless of the evidence, do not satisfy this criterion.
Intuitive-Judgment (grounded in C3): Direct apprehension is present. Something is seen that calculation, data, or protocol alone could not supply. The move is not the output of an algorithm applied to inputs; it is the recognition of something that requires a rational faculty capable of direct contact with what is the case. Passages in which a conclusion follows mechanically from premises, however correctly, do not satisfy this criterion unless a further move — the selection of the right premises, the recognition of what is relevant — exhibits the direct apprehension that calculation cannot perform.
Foundational-Judgment (grounded in C4): The agent correctly identifies what is load-bearing rather than incidental. The judgment orders dependent and independent claims correctly: it does not treat a derived conclusion as a foundation, and it does not treat a foundational commitment as merely one consideration among others. Passages in which an agent recognizes that a particular claim must be true for other claims to hold — or that a particular move is doing the weight-bearing work in an argument — satisfy this criterion.
Step Four: Look-Alike Audit
The core question is: Does this candidate passage fail a detection criterion in a way that makes it a look-alike rather than a confirmed instance?
Six named look-alike failures must be checked before any instance is confirmed.
Protocol-Execution: The agent follows a procedure correctly and arrives at a correct result, but no genuine deliberation is recorded. The reasoning move is the execution of a rule, not the exercise of a faculty. This is the most common false positive for Agency-Judgment. Correct protocol-execution is not judgment; it is competence.
Expertise-Pattern-Matching: The agent draws on deep experience to reach a conclusion quickly and correctly. This is intellectus — immediate pattern-recognition — not ratio — deliberate reasoning through genuine alternatives. Expertise-pattern-matching produces reliable outputs and is not a defect; it is simply not what the CJRI is harvesting. A passage satisfies Intuitive-Judgment only if what is seen exceeds what the pattern alone could supply.
Ideological-Scripting: The conclusion was predetermined by the agent’s governing dogmata; the reasoning visible in the text is post-hoc rationalization. This failure is most relevant to Correspondence-Judgment and Moral-Apprehension-Judgment. The diagnostic question: would the agent have reached a different conclusion if the evidence had gone the other way? If not, the passage records ideology, not judgment.
Positive Indicator — Counter-Preference Move: The complementary marker to Ideological-Scripting. Where an agent follows the argument to a conclusion that goes against his own stated preference, interest, or prior commitment, this is the clearest available evidence that genuine judgment is operating rather than ideological scripting. The diagnostic question inverts: does the conclusion cost the agent something? If it does — if the agent acknowledges a fact, concedes a point, or reaches a finding that he would prefer to avoid — the case for confirmed judgment is strengthened across all applicable types. The Counter-Preference Move does not itself confirm an instance; it raises the confidence of a confirmation already supported by the detection criteria. It is noted explicitly in the characterization when present.
Preference-Anchoring: The conclusion tracks what the agent wants rather than how things are. The reasoning is genuine in form but corrupted at the root: it begins from a premise that is itself a value judgment dressed as a fact. This failure is most relevant to Correspondence-Judgment and Moral-Apprehension-Judgment.
Tradition-Transmission: The reasoning move belongs to a received tradition being faithfully reported rather than to the author exercising his own judgment in the text. The author is the vehicle for the tradition’s reasoning, not its source. This failure is distinct from Ideological-Scripting: the tradition may be correct, and the transmission may be accurate; neither fact makes the transmission a confirmed instance of judgment. The diagnostic question: is the author arriving at the move in the text, or delivering a move already arrived at elsewhere? Where the tradition’s reasoning is being reported, attribution in the curation output goes to the tradition and its named sources, not to the transmitting author. Tradition-Transmission does not automatically disqualify a passage; it governs attribution and flags the boundary between reporting and reasoning.
Assertion-Without-Grounding: A strong claim is made with confidence but no reasoning move is visible. There is nothing to detect. The passage is a candidate only if the move that produced the assertion is recoverable from the text.
Step Five: Curation Output
The core question is: Is this instance recorded in a form that is portable, characterized, and grounded?
Each confirmed instance is recorded in four parts.
Source and Passage: Full citation of the text and precise location of the passage. If the passage is quoted, it is quoted accurately and in full.
Judgment Type(s): All satisfied judgment types are listed by name. Where more than one type is present, the primary type — the one most central to the reasoning move — is identified first.
Grounding Commitment(s): The commitment or commitments that ground the identification are named explicitly, by canonical letter and by name. The connection between commitment and judgment type is stated, not assumed.
Characterization: One to three sentences stating what the judgment is doing in the passage: what is being weighed, what is being recognized, what alternative is being declined, and why the move satisfies the detection criteria rather than a look-alike.
Self-Audit
Mandatory before any confirmed instance is recorded. Each item must be checked in sequence.
- Topic-Suspension Verified: The subject matter of the text did not influence candidate selection. ✓ or Failure noted.
- Reasoning-Move Present: The passage contains a move, not merely an assertion or a reported outcome. ✓ or Failure noted.
- Detection Applied in Order: All five judgment types were tested against the candidate passage before any type was confirmed. ✓ or Failure noted.
- Look-Alike Audit Completed: All six named look-alike failures were checked. ✓ or Failure noted.
- Training-Data Contamination Absent: No judgment-vocabulary imported from adjacent literatures — management, therapy, behavioral economics, academic philosophy — without derivation from the corpus. ✓ or Failure noted.
- Characterization Grounded: The curation output states the commitment, names the type, and characterizes the move without substituting praise for analysis. ✓ or Failure noted.
- Canonical Numbering Verified: Commitment letters in the curation output follow the canonical numbering — C1 Substance Dualism, C2 Libertarian Free Will, C3 Ethical Intuitionism, C4 Foundationalism, C5 Correspondence Theory of Truth, C6 Moral Realism — and not file-order numbering. ✓ or Failure noted.
Self-Audit Complete. Proceed to confirmed output only when all items pass.
A Note on Curation
The CJRI produces individual confirmed instances. Curation — the assembly of confirmed instances into an organized collection — is a separate act governed by Dave Kelly as ratifying authority. The instrument does not determine how the collection is organized, what threshold of instances constitutes a judgment type’s representation, or how the collection is published. It supplies the characterized, grounded raw material from which curation proceeds.
The collection the CJRI builds over time is a recovery project in the same sense as the broader corpus: genuine rational agency has always been present in human discourse across all subjects; the instrument’s function is to find it, name it, and preserve it against the displacement that reduces all reasoning to preference-expression, ideology, or pattern-execution.
CJRI Run — Sterling, ISF, 10 February 2020: The Bathing-Speed Distinction
Theoretical foundations: Grant C. Sterling (Eastern Illinois University). Analysis and synthesis: Dave Kelly. Prose rendering: Claude (Anthropic). 2026.
0 Comments:
Post a Comment
<< Home