JH
Jonathan Haber

Ask the three restorative questions

Start every harm conversation with: who was hurt, what do they need, and who is responsible for meeting those needs?

Key takeaways

  • What it is: Start every harm conversation with: who was hurt, what do they need, and who is responsible for meeting those needs?
  • Why it works: Conventional responses to harm focus on the rule violated and the punishment warranted — a backward-looking, perpetrator-centered frame. The three restorative questions shift focus forward and outward: to the person harmed, their actual needs, and the community responsible for meeting them. This reframe changes who has voice in the process and what counts as resolution — from "punishment delivered" to "harm repaired."
  • Evidence: Backed by observational / correlational evidence.
  • Avoid: Asking what happened and immediately jumping to who was at fault, rather than first establishing who was harmed and what they need — which determines what resolution even means.

Why it works

Conventional responses to harm focus on the rule violated and the punishment warranted — a backward-looking, perpetrator-centered frame. The three restorative questions shift focus forward and outward: to the person harmed, their actual needs, and the community responsible for meeting them. This reframe changes who has voice in the process and what counts as resolution — from "punishment delivered" to "harm repaired."

How to do it

  1. 1When harm occurs — an offense, a conflict, a breach of trust — pause and explicitly ask: "Who was hurt by what happened?"
  2. 2Ask the harmed party directly: "What do you need in order for this to be right?" Listen without proposing solutions.
  3. 3Ask the person who caused harm: "What needs to happen for you to take responsibility for this?" — not as an accusation but as a genuine accountability question.

What the evidence says

Observational

Restorative justice meta-analyses find consistently higher victim satisfaction with restorative processes compared to conventional criminal justice, along with moderate evidence for reduced reoffending in some juvenile and adult contexts.

Honest caveat: Most restorative justice research is in criminal contexts; generalization to workplace or interpersonal conflicts involves contextual differences. Outcomes vary by type of offense and voluntary participation.

References
  • — Sherman, L. W. & Strang, H. (2007). Restorative Justice: The Evidence. The Smith Institute, London.

Common mistake

Asking what happened and immediately jumping to who was at fault, rather than first establishing who was harmed and what they need — which determines what resolution even means.

IX Coach coaches you through the three restorative questions in the order they matter: harm first, needs second, responsibility third — so the conversation stays oriented toward repair.

Practice this with IX Coach →

Related practices