| name | blind-inversion |
| description | Use when a problem is stuck when approached directly, when defining success criteria, when designing for reliability/safety, or when a goal is vague ("make it good"). Applies Jacobi's inversion as practiced by Charlie Munger — solve problems backward: avoid the failure instead of chasing the success. |
Inversion
Overview
"Man muss immer umkehren" (invert, always invert) — attributed to Carl Gustav Jacobi (earliest print appearance is Van Vleck, 1916, already as "is said to have"; plausible folklore, fitting for the man who inverted elliptic integrals) and operationalized by Charlie Munger: many problems are easier solved backward. Instead of "how do I achieve X?", ask "what would guarantee NOT-X — and how do I avoid all of that?" Munger: "All I want to know is where I'm going to die, so I'll never go there." Forward thinking samples from paths to success (huge, fuzzy space); inversion enumerates paths to failure (smaller, sharper space) — and failure paths are where blind spots hide, because nobody was looking there.
When to Use
- Vague goals: "make this fast/clean/reliable/good" → invert to the concrete badness to eliminate
- Stuck on a direct approach
- Designing guardrails, reviews, checklists, on-call runbooks
- Evaluating a decision: before asking "why will this work?", ask "what would make this a disaster?"
The Procedure
- State the goal, then flip it: write the anti-goal explicitly ("users churn immediately", "the migration corrupts data", "nobody can read this code").
- Enumerate reliable ways to achieve the anti-goal. Be a competent saboteur: how would you ensure churn / corruption / unreadability? Concrete mechanisms, not categories.
- Audit the current plan against the sabotage list. Every mechanism the plan doesn't block is a finding.
- Convert to avoidance rules. Often the entire deliverable: a short "never do these" list beats a long "try to do these" list.
Worked Example
Goal: "onboarding should convert well" (fuzzy) → Invert: "how would I guarantee signups abandon?" → require credit card before value, 12-field form, confirmation email that lands in spam, empty state with no next action → check the actual flow for each → three concrete fixes, found in minutes.
Common Mistakes
- Inverting the goal but not the thinking — listing generic risks instead of asking "how would I deliberately cause this?" The saboteur stance is the tool.
- Stopping at avoidance when the direct path was fine — inversion complements forward planning, it doesn't replace it.
- Abstract anti-goals ("bad UX"). Invert to mechanisms you could actually execute; if you couldn't do it on purpose, you can't check for it.