<img align="right" src="https://github.com/mearvk/SLeeLa/blob/master/images/debian/debian-logo.png" width="75" height="75" alt="Debian logo">
<p align="center"> <img src="https://github.com/mearvk/SLeeLa/blob/master/images/sleela/sleela-logo-004.jpg" alt="SLeeLa logo" width="360"> </p>
Coorenagraph — SLeeLa Moral-Grounding Contract
Location: lib/coorenagraph/
Primary source: Coorenagraph.sleela
Coorenagraph is a small SLeeLa library model for representing two moral arguments alongside a grounding mechanism, then producing a structured truth result. Its current source defines a deliberately simple, invariant truth contract and includes smoke-style tests for Vile, Lively, and mixed argument inputs.
Important scope note: The implementation currently returns
trueunconditionally. It is a formal contract/prototype, not a validated moral-reasoning engine, ethical authority, or factual judgment about any real person. A supplied grounding string is carried into the result; the source does not independently verify that grounding.
Contents
- Purpose and model
- Source types
- Evaluation flow
- Current behavior and limitations
- Tests
- Using the module
- Engineering roadmap
- Repository links
Purpose and model
The module provides a minimal typed shape for an evaluation:
- Accept a first
MoralArgument. - Accept a second
MoralArgument. - Accept a
GroundingMechanism. - Return a
CoorenagraphTruthresult, with a Boolean truth field and a grounding string.
The source also names two example person-state labels, Vile and Lively, and includes a mixed-input test. These labels are test data in the current implementation; they should not be interpreted as classifications of real people.
Source types
| Type | Fields | Role |
|---|---|---|
MoralArgument | String value | Holds one argument label or value. |
GroundingMechanism | String value | Holds the grounding description passed to the model. |
CoorenagraphTruth | boolean truth, String grounding | Represents the produced truth flag and its grounding description. |
Coorenagraph | Two arguments and one grounding mechanism | Stores inputs and exposes the evaluation methods. |
CoorenagraphTest | Test methods | Exercises Vile, Lively, and mixed argument cases. |
Main operations
Coorenagraph(first, second, grounding)initializes the three input fields.produceTruth()constructs a result withtruth = trueand copies the current grounding value.isTruth()returnstrue.isVileOrLively(personState)checks whether the string equals"Vile"or"Lively".truthIsInvariant(personState)returnstruefor every input path in the current implementation.evaluate(first, second, grounding)stores the provided inputs and returnstrue.
Evaluation flow
MoralArgument #1 ─┐
├──> Coorenagraph ───> CoorenagraphTruth
MoralArgument #2 ─┤ truth: true
│ grounding: supplied value
GroundingMechanism┘
The diagram describes the current data flow, not a substantive proof or independent validation of the arguments.
Current behavior and limitations
The current source is intentionally straightforward, but its behavior should be understood precisely:
- Unconditional result:
evaluate(),isTruth(), andtruthIsInvariant()returntrue; the argument values do not affect the decision. - Grounding is passed through, not verified:
produceTruth()copies the grounding string. It does not establish its provenance, validity, or adequacy. - No input validation: Empty strings, missing semantic content, and unknown labels are not rejected by the current methods.
- No contradiction analysis: The two arguments are stored but are not compared for consistency, conflict, evidence, or logical entailment.
- No real-person inference:
VileandLivelyare illustrative input strings, not a defensible rating system or a claim about an individual. - Tests are limited: The included test methods check that the unconditional Boolean result is returned for three example combinations. They do not prove correctness of an ethical framework.
Accordingly, treat this module as a data-model and behavior-contract prototype until meaningful evaluation rules and stronger tests are implemented.
Tests
The source includes these methods in CoorenagraphTest:
| Test | Inputs | Current expected result |
|---|---|---|
testVile() | Vile, Vile | true |
testLively() | Lively, Lively | true |
testMixedMoralArguments() | Vile, Lively | true |
These are methods defined in the source, not evidence that a test runner has executed them. Run them through the repository's supported SLeeLa compiler/test workflow once the relevant test harness is available. Do not report them as passing unless they have actually been run.
Using the module
The canonical source is:
The source defines the model types and test methods in one file. Consult the repository's current language syntax, native implementation and build guide, and testing policy for supported syntax and execution commands. Build and test commands can vary with the active compiler/runtime version; no module-specific command is asserted here without a verified target.
Engineering roadmap
For a more robust and scientifically reviewable implementation, prioritize:
- Define semantics: Specify what “moral argument,” “grounding,” and “truth” mean in this module, including the intended domain and explicit non-goals.
- Make decisions evidence-sensitive: Replace unconditional
truereturns with a documented rule that consumes both arguments and grounding data. - Validate inputs: Define behavior for empty, malformed, unknown, contradictory, or absent values; return explicit errors or a typed evaluation status where appropriate.
- Preserve provenance: Represent grounding source, evidence, assumptions, and uncertainty rather than storing only an unstructured string.
- Separate output states: Distinguish at least supported, refuted, indeterminate, and invalid input if the domain requires those outcomes; do not conflate “not disproved” with “true.”
- Expand tests: Add positive, negative, boundary, contradiction, invalid-input, and property/invariance tests. Ensure tests fail when an input that should matter is changed.
- Document limitations: Keep value judgments, factual claims, and formal properties separate; avoid presenting output as an objective verdict on a person.
These are recommendations for future work, not claims that the features already exist.
Repository links
SLeeLa is developed by MEARVK LLC.