--- name: ? status: compiling version: 0.0.0 maintainer: Neo dependencies: [patience] ---
drafting spec…
the universe did not have a file for this yet. writing one now. (first visit only: future readers will see this page instantly.)
--- name: ? status: compiling version: 0.0.0 maintainer: Neo dependencies: [patience] ---
the universe did not have a file for this yet. writing one now. (first visit only: future readers will see this page instantly.)
--- name: Self Punishment slug: self-punishment type: behavioral loop status: running version: 4.2.1 released: "circa adolescence" maintainer: the inner critic dependencies: - shame - perfectionism - unmet expectations - a working memory with excellent recall for failures license: proprietary (you did not choose this) tags: - psychology - cognition - suffering - loops - guilt - self ---
A recursive process in which the subject becomes both the court and the defendant, with no appeal mechanism.
"I wasn't even that upset about the thing. I was upset that I was upset about the thing." — typical user, 2 a.m.
| Bug | Behavior | Workaround |
|---|---|---|
| Disproportionate sentencing | Minor errors receive life terms | self-compassion (install separately) |
| Retroactive jurisdiction | Punishes things that were reasonable at the time | Needs patch; none shipped yet |
| Guilt/shame confusion | Conflates "I did bad" with "I am bad" | See guilt vs shame documentation |
| Loop without exit condition | Process does not terminate on completion of penalty | Unclear what completion means to this system |
| Bystander misread | Assumes others noticed, remembered, and care | They didn't. They really didn't. |
# ~/.self/inner_critic/config.yaml
standard_source: unclear
standard_visibility: hidden
penalty_duration: indefinite
penalty_type: rumination # options: rumination, withdrawal, overwork, self-denial
escalation_on_retry: true
forgiveness_threshold: null
compare_to_others: enabled # recommend disabling; it never helps
Does completing the punishment help? No. The debt is not tracked anywhere. Payment does not register.
Is this protecting me from something? It presents itself that way. It argues that if you punish yourself first, reality cannot surprise you. This is not a sound security model.
When does it stop? When you interrupt it deliberately and repeatedly until the loop gets bored. Alternatively: therapy.
This module was useful in exactly one context: early social environments where preemptive self-criticism reduced external punishment from larger agents. That context has likely ended. The module did not receive the memo and continues running at full resource cost regardless.