Model I vs Model II Communication
Why otherwise-honest people stop telling the truth at work — and what changes that.
Most of the people in your business are honest. They want the work to go well. They want you to succeed. If you asked them in a quiet moment, away from the office, they could probably tell you exactly what’s wrong with the team — clearly, accurately, and with examples.
You will rarely, if ever, hear that version inside the building.
What you’ll get instead is a version that’s been smoothed at the edges. Concerns mentioned but not pressed. Disagreements raised then withdrawn when the room goes quiet. The hard sentence almost said, then replaced with something easier. Nobody is lying. Everyone is editing.
Chris Argyris spent forty years trying to understand why this happens — and why it happens even in companies full of intelligent, well-meaning people who know honesty would serve them better. He gave the pattern a name: Model I.
Who was Argyris
Chris Argyris taught at Yale and Harvard and is one of the most cited thinkers in organisational behaviour. Working with Donald Schön, he developed a body of work on how organisations actually behave — as opposed to how they say they behave. The Model I and Model II distinction is part of that work, and it remains one of the most useful tools available to anyone trying to fix the way a team communicates.
Espoused theory and theory-in-use
To understand Model I and Model II, you have to start with one of Argyris’s other key ideas — and arguably the most important one in his whole body of work.
There is a difference between what people say they believe and what their behaviour actually shows they believe. Argyris called these espoused theory (what you say) and theory-in-use (what you do).
Most people in most organisations are operating with a gap between the two. They espouse openness, candour, and learning. Their actual behaviour, under any kind of pressure, is to protect themselves, smooth conflict, control the conversation, and avoid being wrong in public.
The interesting and uncomfortable finding from Argyris’s research is that people are usually unaware of this gap. When you record a meeting and play it back, the participants are often genuinely surprised by what they hear themselves doing. The behaviour is automatic. Nobody chose it. It’s just the water everyone is swimming in.
Model I is the name Argyris gave to that water.
What Model I looks like
Model I is governed by four implicit goals. People rarely state these out loud. They drive behaviour anyway.
Stay in unilateral control
Define the situation, manage how it unfolds, and avoid letting anyone else set the terms. In a meeting, this looks like steering the conversation, not exploring it. In a conflict, it looks like winning rather than understanding.
Maximise winning, minimise losing
Treat disagreement as a contest rather than an enquiry. The point is to come out of the conversation with your position intact, not to come out of it with a better understanding than you went in with.
Suppress negative feelings
Don’t surface frustration, doubt, or concern directly. If they have to come out, deliver them in code — through tone, through what doesn’t get said, through a question that’s actually a statement.
Be rational
Stay calm. Stay measured. Frame everything as analytic and impersonal. Anything that looks emotional is treated as a failure of professionalism, including, often, the perception of an honest reaction.
These four governing values produce a predictable set of behaviours. People speak in qualifications. They protect their own position and, courteously, the positions of others. They avoid the topic that actually matters by spending most of the meeting on the topic that doesn’t. They leave saying “good meeting” when nothing has changed.
Argyris called these defensive routines — the patterns of behaviour a team develops to keep everyone safe from the embarrassment of being wrong, exposed, or in conflict. Defensive routines are, in his view, the single biggest reason intelligent organisations make poor decisions.
The hard part is that defensive routines feel like maturity. They feel like good professional behaviour. People are trained, often from childhood, to operate this way. It’s only when you measure the cost — what doesn’t get said, what doesn’t get fixed, what walks out the door with the resignation letter — that the size of the problem becomes visible.
What Model II looks like
Model II is the alternative Argyris and Schön proposed. It isn’t softer or more emotional. In some ways it’s harder, because it asks for a kind of honesty most workplaces aren’t built for.
Model II has three governing values.
Use valid information
Say what you actually see. Bring the evidence. If you have a hunch, name it as a hunch. If you have a worry, put the worry on the table rather than translating it into a procedural objection. The aim is for everyone in the conversation to be working from the same picture of reality.
Make free and informed choices
Give other people the information and reasoning they need to make their own judgement, rather than steering them toward yours. Don’t manage the conversation toward a predetermined conclusion. Argue your view, and let the other person argue theirs, and let the better argument win — including when the better argument isn’t yours.
Build internal commitment
Decisions taken under Model II hold up because everyone involved actually understands why the decision was made and genuinely agrees with it. Decisions taken under Model I tend to need constant re-enforcement because people complied without committing.
In practice, Model II conversations sound different from Model I conversations. People say what they think. They show their reasoning out loud, including the parts that aren’t certain. They invite challenge instead of deflecting it. They treat being wrong as new information rather than as a status loss.
This sounds simple. It is genuinely hard, because the entire surrounding culture — schools, professional norms, most workplaces — has trained people in Model I behaviour for decades. Moving to Model II usually requires a deliberate, sustained, conscious effort, and someone in a position of authority modelling it before anyone else will take the risk.
Why this matters in a small business
In a large organisation, Model I behaviour is expensive but absorbable. There are enough layers, enough redundancy, enough people, that the truth eventually finds a way through.
In a small business, Model I behaviour is the difference between a business that adapts and one that doesn’t.
If your team is operating in Model I — and most are — three things are happening, every week, that you cannot see.
You’re not getting the information you need. The most useful thing about a small team is that the people closest to the work can tell you what’s actually happening. In a Model I culture, they don’t. You make decisions on the smoothed version.
Your good people are exhausted by the editing. The cost of working under Model I is borne mostly by the people doing the editing — the ones holding back what they really think, managing your reactions, protecting their own position. Over time, this is how you lose them.
The same problems keep recurring. This is the link to Single-Loop vs Double-Loop Learning. The assumptions that produce recurring problems stay invisible because nobody can name them out loud without breaking Model I norms.
The owner sets the ceiling. If the owner operates in Model I — controls the conversation, treats disagreement as personal challenge, smooths over their own discomfort with rationality — the team will operate one notch below that. If the owner takes the risk of Model II — says what they actually think, shows their reasoning, lets themselves be visibly wrong — the team will gradually do the same.
Three questions to start with
These questions are designed to make Model I behaviour visible in your own business, including in yourself.
When was the last time someone in your team disagreed with you, in front of others, and the disagreement was treated as useful? If you can’t remember, that’s the data. The team has read the room and concluded that disagreement isn’t safe. They aren’t wrong about that until you change something.
What’s the topic that everyone is carefully not raising in meetings? There is almost always one. It might be a person, a strategic decision, a financial reality, or a relationship. The fact that everyone is editing around it is the signal that Model I is in charge.
When you’re about to deliver feedback or concern, what version of it do you actually deliver? The honest version, or the one you’ve smoothed for delivery? The gap between those two versions is the size of your own theory-in-use problem. Argyris would say nobody’s gap is zero, including his own.
Where this connects
Model I behaviour is what makes problems recur. The framework that explains why is Single-Loop vs Double-Loop Learning — when the assumptions producing a problem can’t be named out loud, the problem keeps regenerating.
The deeper version of the same conversation is in Joy at Work. Deming’s framework names fear as the first thing that destroys good work — and Model I is, at its core, a sophisticated and socially-acceptable form of fear management.
Frequently asked questions
What is Model I in Argyris's framework?
Model I is a theory of action governed by four implicit goals: stay in unilateral control, maximise winning and minimise losing, suppress negative feelings, and stay rational. It produces defensive routines — patterns of communication that protect everyone from embarrassment but quietly stop honest information from moving. Most workplaces operate predominantly in Model I.
What is Model II?
Model II is the alternative Argyris and Schön proposed. Its governing values are: use valid information, support free and informed choice, and build internal commitment to decisions. In practice, Model II looks like saying what you actually think, showing your reasoning, inviting challenge, and treating being wrong as useful rather than threatening.
What is the difference between espoused theory and theory-in-use?
Espoused theory is what people say they believe — the values they would describe if asked. Theory-in-use is what their behaviour actually shows they believe, especially under pressure. Argyris’s central finding was that most people, and most organisations, have a significant gap between the two and are unaware of it.
Why do otherwise-honest people stop telling the truth at work?
Because Model I behaviour is the social default in most professional environments. People are trained to manage how things look, smooth conflict, and protect their own position — and these behaviours feel like maturity rather than dishonesty. The result is that valuable information stops moving even though everyone involved is acting in good faith.
How do I shift my team toward Model II?
The owner sets the ceiling. The first move is to model it yourself: say what you think directly, show your reasoning out loud, treat disagreement as useful rather than personal, and let yourself be visibly wrong on something significant. The team will move when they have evidence that doing so is safe. This usually takes months, not weeks.