Leadership

Double-Loop Learning

Double-loop learning is the practice of questioning and modifying the underlying assumptions, goals, and norms that shape how a team operates, rather than simply correcting errors within existing rules. It distinguishes organizations that adapt from those that merely react.

Also known as: reflective learning, generative learning, assumption questioning

Why It Matters

Most organizational learning is single-loop: something goes wrong, the team fixes it, and moves on. The thermostat analogy applies. A thermostat detects the room is too cold and turns on the heat. That is single-loop. Double-loop learning asks: is the target temperature correct? Should we be heating this room at all? Organizations that only practice single-loop learning keep solving the same categories of problems because they never examine the assumptions that produce those problems in the first place.

The Original Framework

Chris Argyris (Harvard) and Donald Schon (MIT) introduced double-loop learning in their 1977 Harvard Business Review article and expanded it in "Organizational Learning: A Theory of Action Perspective" (1978). Their central finding: most organizations get stuck in single-loop because questioning underlying assumptions feels threatening. It requires admitting that the rules, strategies, or goals themselves may be flawed, which challenges authority and identity.

How to Practice It

Double-loop learning requires creating space where questioning assumptions is safe and expected. This means building retrospective practices that go beyond "what went wrong" to ask "what assumptions led us here." It means treating repeated failures as signals that the system needs redesign, not just better execution. And it means leaders modeling the willingness to question their own decisions and frameworks.

  • When a problem recurs, ask what assumption or rule is producing it
  • Retrospectives should include "what should we stop believing" alongside "what should we do differently"
  • Leaders model assumption-questioning by revising their own frameworks publicly
  • Treat persistent friction as a signal that the system needs redesign, not just better compliance

Source

Chris Argyris and Donald Schon. "Double Loop Learning in Organizations," Harvard Business Review, 1977.