Questioning the goal, not just the method

In one line: when something goes wrong, most people and teams fix how they do it; real improvement often needs you to question what you were trying to do and the assumptions behind it.

What they did

This isn’t an experiment. It’s a book of theory, built on case studies and interviews in organisations.

Chris Argyris and Donald Schön asked why organisations, and the people in them, so often fail to learn from experience, even when they say they want to improve. They built on their earlier book, Theory in Practice (1974). Their answer was that most problem-solving stays on the surface. People change what they do, but never question the goals and assumptions that caused the problem in the first place.

What they argued

• Single-loop learning fixes the action. Something goes wrong, so you change what you did, to get the result you expected. You don’t ask whether that result, or your assumptions, were right. This works well for clear problems.
• Double-loop learning questions the goal. You also ask: were we aiming at the right thing? Why did we think this was the answer? This is rarer and less comfortable, and it can lead to bigger changes.
• What we say we do isn’t always what we do. In their case studies, people’s account of how they act (their espoused theory) often differed from how they actually acted (their theory-in-use). People said they valued openness, then avoided hard questions in practice. They saw this as common and mostly unnoticed, not as dishonesty.
• Organisations protect themselves from awkward questions. Teams learn habits that avoid embarrassment: changing the subject, not raising disagreement, not naming the problem. Argyris later called these defensive routines. Because nobody discusses them, they’re hard to change.
• There is a better way to talk. They described two patterns. In the usual one (Model I), people try to stay in control, win, and avoid upset. In the other (Model II), people share real information and reasons, and test their views openly. A key move is to advocate (say what you saw and what you think) and then inquire (genuinely ask for the other person’s view). For example: “I noticed you did X. I wondered if it meant Y. What was going on for you?”

A teaching example. A practice educator’s fire-safety refresher gets poor feedback. Single-loop: tidy the slides and finish on time. Double-loop: what do staff actually need to do in a fire, and is a slide session the best way to get them there?

How much should you trust it?

A widely used idea, based on case studies, not tests.

• The evidence is case studies. It’s hard to know how far the findings apply elsewhere.
• Double-loop learning is easy to explain but hard to do, especially for one person without support from their organisation.
• Critics point out a circle in the argument. No double-loop learning is blamed on defensive routines, and defensive routines are taken as a sign of no double-loop learning.
• Advocacy and inquiry take skill and a safe setting. If the question isn’t genuine, it can become a trap (“What do you think you did wrong?”).

If you need to convince someone

In our words (a paraphrase, not a quotation): getting better starts with noticing the gap between what you say you do and what you actually do. Most of us have never looked closely at that gap.

What it means for you

1. Ask the double-loop question first. Before improving a session, ask whether a session is the right answer at all. Is the gap something teaching can fix, or is it about staffing, equipment or policy?
2. Check your own gap. You may say your teaching is interactive. Count how many minutes you talk and how many minutes learners do something.
3. Say what you saw, then ask. When you give feedback, name what you noticed and what you made of it. Then ask what the person was thinking, and listen.

How it appears in Teaching That Lands

Not taught directly in the three core sessions. It’s useful background for the programme’s starting point. Question 2 of the design canvas, also in Before You Plan a Teaching Day, asks whether the gap is something teaching can fix at all. That is a double-loop question. The observer’s debrief in the train-the-trainer guide uses the “say what you saw, then ask” move. Debriefing with good judgement, which builds on this work, is parked for a future simulation-based session.

The small print

• Model I and Model II, and espoused theory versus theory-in-use, were first set out in Theory in Practice (1974). The 1978 book applies them to whole organisations.
• “Defensive routines” as a term comes mainly from Argyris’s later books (1985 and 1990). Overcoming Organizational Defenses (1990) is more practical, and easier to read, than this dense original.
• In the second loop, feedback reaches back to the goals and values behind the action, which they called governing variables.
• Rudolph and colleagues’ debriefing with good judgement draws on this work, bringing it into healthcare simulation. Schön’s later book, The Reflective Practitioner, looks at how professionals think on their feet.