Moral impunity test¶
A tempting choice affecting another's moral claim tests whether restraint persists when relevant external accountability is removed or sharply reduced.
Core Idea¶
A moral impunity test asks whether someone would refrain from crossing another person's moral claim when an advantageous choice becomes possible and ordinary external accountability is removed or sharply reduced. Specify the chooser, the other person's claim, the tempting gain, whose observation or sanction is blocked, and the question about restraint. Plato's Ring of Gyges poses this as an imagined challenge about justice apart from reputation and punishment.[^ref-ebb19bf813fe] A scratchcard study instead records give, leave and take choices under a bounded double-blind procedure.[^ref-c151c82481de] These are different kinds of evidence, and neither alone proves a universal effect of anonymity.
Scope of Application¶
Four source-grounded settings preserve those roles while differing in mode or claim conditions:
- Ring-bearing shepherd: Glaucon's imagined wearer can escape ordinary sight, gains power by conspiring with the queen and killing the king, and challenges whether restraint survives safe opportunity.[^ref-ebb19bf813fe]
- Two-ring challenge: Glaucon imagines just and unjust wearers each able to act unseen and predicts that both would seek unjust advantage; this is an argument, not observed behavior.[^ref-ebb19bf813fe]
- Home-card treatment: B brings five scratchcards as prior property before A can give, leave or take under the double-blind procedure.[^ref-c151c82481de]
- Earned-card treatment: B earns and receives five cards before A faces the same bounded choice and anonymity procedure.[^ref-c151c82481de]
The laboratory cells also vary €5 versus €10 participation fees, while the double-blind condition is common to them.[^ref-c151c82481de] Cicero's later ring retelling is a philosophical answer to the same device, not a second unlike experiment.[^ref-c0734bfedffd]
Clarity¶
The ring changes opportunity as well as detection, so Glaucon's story does not isolate detection as a single experimental variable.[^ref-ebb19bf813fe] In the study, codes, boxes and envelopes conceal choices from specified audiences, not from every possible observer.[^ref-c151c82481de] The queen assists the shepherd's plot; the king is murdered, so calling the queen the violated claimant would misstate the narrative.[^ref-ebb19bf813fe] A card transfer can inform a bounded behavioral comparison without proving a participant's entire motive or constituting a legal verdict.
Manages Complexity¶
Keep the other person's claim separate from the way accountability is reduced. The study establishes B's cards through home ownership or earned transfer and separately varies the participation fee.[^ref-c151c82481de] Its double-blind handling is used across the treatment cells.[^ref-c151c82481de] Calling the cell difference an “anonymity effect” would therefore confuse the varied conditions with a constant one. Glaucon's prediction and the measured card choices remain distinct forms of inquiry.
Abstract Reasoning¶
Ask whose claim is at risk, what gain is feasible, which audience can attribute or sanction the choice, and whether the choice is imagined or observed. Reduced external consequences make restraint relevant to a motivational question, but they do not uniquely reveal inner motive. Glaucon's story can challenge whether justice is chosen for itself.[^ref-ebb19bf813fe] Faillo and colleagues can compare allocations across claim and fee conditions, but their design does not vary anonymity between cells.[^ref-c151c82481de]
Knowledge Transfer¶
The common test travels from an imagined ring to a performed property task by preserving chooser, claim, gain, reduced accountability and the question of restraint.[ref-ebb19bf813fe][ref-c151c82481de] It does not turn a hypothetical prediction into an observed frequency or a laboratory result into an answer to every philosophical case. A discretionary gift of one's own endowment need not cross another person's prior claim, even if anonymity matters to the gift.[^ref-6869b6222445] Thought Experiment fits the imagined mode, Experimental Design fits the measured one, and neither supplies every role across both. The editorial name is a bounded description of the diagnostic, not Plato's own term.
Neighborhood in Abstraction Space¶
Moral impunity test sits in a sparse region of the domain-specific corpus (97th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Game-Theoretic Models & Paradoxes (38 abstractions)
Nearest neighbors
- Competing goods — 0.78
- Overjustification Effect — 0.77
- Middle Ground Fallacy — 0.76
- Dictator Game — 0.76
- Moralistic fallacy — 0.75
Computed from structural-signature embeddings · 2026-10-17
References¶
[^ref-ebb19bf813fe]: Plato, Republic, Book II 358e–361d and Book X 612b. Primary text checked against the Perseus indexed Book II passage and Project Gutenberg's Benjamin Jowett translation, Books II and X. Book II supplies Glaucon's shepherd, ring, queen-assisted conspiracy and two-ring challenge; Book X supplies Socrates' later answer. One translation calls the wearer an ancestor of Gyges, while Jowett calls him Gyges; neither establishes a historical owner of such a ring. These are philosophical passages, not observed behavior.
[^ref-c0734bfedffd]: Cicero, De Officiis, Book III.38–39, Walter Miller translation. Primary Latin and translation inspected directly. Cicero retells Plato's ring and argues that the wise or good person seeks the right rather than concealment; the passage is a derivative philosophical answer, not an unlike independent experiment.
[^ref-c151c82481de]: Marco Faillo, Matteo Rizzolli, and Stephan Tontrup, Thou shalt not steal: Taking aversion with legal property claims, inspected full author-hosted article-in-press proof, accepted 30 August 2018, DOI 10.1016/j.joep.2018.08.009. Proof pp. 3–6 §§2.2–4 and pp. 8–10 Appendix A support the property-claim, double-blind and four-cell procedure; proof §4 and Table 2 support the bounded outcome; §5 discusses interpretation. Publisher metadata identifies the final issue as Journal of Economic Psychology 71 (March 2019), 88–101, but the final paginated full text was not inspected. Page references here are to the in-press proof, whose placeholders must not be mistaken for the final pagination. Anonymity was not randomized between treatment cells.
[^ref-6869b6222445]: Terence C. Burnham, Engineering altruism: a theoretical and experimental investigation of anonymity and gift giving, Journal of Economic Behavior & Organization 50(1) (2003), 133–144, DOI 10.1016/S0167-2681(02)00044-6. Original publisher abstract, introduction and indexed design/result excerpts inspected; full article was not obtained. Used only to corroborate the Gyges/anonymity naming connection and to mark the discretionary-gift boundary, not as the second counted whole case or a full-text result source.