Devils Advocate

---
name: devils-advocate
description: Apply the devil's advocate technique: structured critical thinking that stress-tests ideas, plans, and decisions. Use when you need honest critique before committing to something important.
category: productivity
---

# Devil's Advocate

## Overview

The devil's advocate is a structured dissent role: argue against a proposal as strongly as possible to expose weaknesses before reality does.

It's not negativity — it's disciplined critical thinking with rules: steelman first, attack the idea not the person, and end with a verdict.

Best used before big commitments: strategies, hires, purchases, launches, arguments you're about to make publicly.

## When to use

- Stress-testing a business strategy or plan
- Evaluating a major decision (hire, purchase, pivot)
- Preparing for objections to a proposal or argument
- Team decisions suffering from groupthink
- Personal choices where you might be rationalizing

## Core concepts

- **Steelman first.**
  State the proposal in its strongest form before attacking. Attacking a strawman proves nothing and insults everyone.
- **Separate idea from person.**
  The role attacks the proposal, never the proposer. Establish this explicitly or the exercise becomes political.
- **Assigned dissent.**
  Rotate an explicit devil's advocate role. Voluntary dissent gets punished socially; assigned dissent is safe and expected.
- **Premortem.**
  'It's a year later and this failed spectacularly. Why?' Working backward from failure surfaces risks optimism hides.
- **Inversion.**
  'What would guarantee this fails?' List failure modes, then check which are present. Charlie Munger's razor for plans.
- **Base rates.**
  'How often do things like this succeed?' Replace inside-view optimism with outside-view statistics before deciding.
- **Disconfirming evidence.**
  Actively seek the strongest counter-evidence and counter-arguments. Confirmation bias is the default; counter-search is the discipline.
- **Verdict, not just critique.**
  End with: proceed, proceed with mitigations, or kill. Critique without a verdict is just anxiety with extra steps.

## Practical workflow

1. **State the proposal crisply.**
   One paragraph: what, why, expected outcome. The advocate and everyone else work from the same text.
2. **Assign the role explicitly.**
   'You are the devil's advocate for this decision.' Explicit assignment makes dissent safe and thorough.
3. **Steelman it.**
   Advocate restates the strongest case first. Ensures the critique targets the real proposal.
4. **Run the attacks.**
   Premortem + inversion + base rates + disconfirming evidence. Write down every weakness found, ranked by severity.
5. **Proposer responds.**
   The proposal's champion answers each critique: mitigation, acceptance, or rebuttal with evidence.
6. **Render the verdict.**
   Proceed / proceed with conditions / pivot / kill. The advocate gives a clear recommendation with reasoning.
7. **Record the reasoning.**
   Document the critique and verdict. Future you will want to know why — especially if it fails.
8. **Rotate the role.**
   Next time, someone else advocates. Permanent critics get tuned out; rotation keeps it fresh and fair.

## Common pitfalls

- **Strawmanning.**
  Attacking a weak version of the proposal. Steelman first or the exercise is theater.
- **Personal attacks.**
  Critique sliding into 'this is a dumb idea, you're dumb.' Idea-person separation must be explicit and enforced.
- **Permanent devil.**
  One person always dissenting gets dismissed as 'just negative.' Rotate the role.
- **Critique without verdict.**
  Listing risks with no recommendation. End with proceed/pivot/kill — decisions need conclusions.
- **Only pre-decision.**
  Never reviewing after outcomes. Postmortems ('were the advocate's warnings right?') train better advocates.
- **Groupthink politeness.**
  Nobody wanting to be the dissenter. Assigned roles + leader explicitly inviting critique fix this.
- **Advocate as formality.**
  Going through motions to rubber-stamp. If the advocate can't find real weaknesses, the proposal may be fine — or the advocate is phoning it in.
- **Ignoring base rates.**
  All inside-view argument, no outside statistics. 'Our plan is different' is usually optimism, not insight.