Agency vs. Coercion
Or, Why Telling People What to Do Almost Never Works (Including When You’re Right)
I have spent a good portion of my adult life sitting across from people who desperately wanted their lives to change. Less anxiety.Less chaos.Less pain.Better relationships.Better boundaries.Better choices. Sometimes just to feel better.
And—this is important—they often wanted someone else to tell them what to do and how to do it.
This is where things get tricky.
Because one of my favorite coping skills is being right. And it turns out that being right is not the same thing as being helpful.
My Dirty Secret
Here’s something they don’t tell you in graduate school: If I could fix people by giving good advice, I would have retired early and bought a boat. Maybe two.
I thought I gave excellent advice. Thoughtful advice. Beautiful, well-structured advice. Advice supported by research, experience, and a deep desire to be helpful. And much of it didn’t work. Or it worked briefly.
Or people thanked me, agreed with me, and then went right back to doing whatever they had been doing before.
For years I thought the problem was that they weren’t following the advice. Eventually it occurred to me that maybe the problem was also me. Advice itself isn’t coercion. Neither is persuasion. There is nothing inherently wrong with saying: Here is what I think. Here is what I have seen. Here is what the research suggests. Here is what I would consider if I were standing where you are.
The trouble starts when I become invested in making you choose my answer. That is a very different thing.
Care can quietly become control. And you can wrap control in a greeting card, but eventually it opens to the same thing.
Agency Is Annoying That Way
Agency means people get to choose. Even when they choose badly.No matter how smart I think I’m being.
No matter how strong I think the evidence is.No matter how obvious the answer looks from my chair. They make the choice.
That’s not a bug in the system. That is the system. Which is rude, because I brought charts. Every therapist learns some version of this the hard way. You can’t make someone quit drinking. I tried. You can’t make someone leave a bad relationship. I tried. You can’t make someone stop sabotaging themselves, no matter how elegant your explanation is. You can’t even make someone want to want change.
Sheesh.
Even when they tell you they want it. You can get agreement. You can get promises. You can get compliments. You can get that special look that says: This time it’s going to stick. But agreement is not ownership. And compliance is a temporary hallucination. It’s the emotional equivalent of: Sure. I’ll start Monday.
Advice, Persuasion, Structure, and Coercion
I wish I had understood earlier that these are not the same thing. Advice says: Here is what I think you should consider. Persuasion says: Here are my reasons. Structure says: Here are the boundaries, expectations, and consequences that already exist. Coercion begins when enough leverage is applied that refusing my preferred choice becomes substantially harder, more costly, or impossible.
And agency says: The decision is still yours. So are the consequences. That distinction matters. Because sometimes people need advice. Sometimes they need information. Sometimes they need structure. Sometimes legitimate restraint is necessary.
But helping becomes something else when I decide that your freedom to choose is an obstacle to the outcome I want. Even when I am convinced the outcome is good. Especially then.
Coercion Feels Efficient
Coercion is seductive because it feels like progress. Just do this. You need to stop that. Trust me. I know where this leads. Or the infamous use of the declarative " should". You know the one I think Karen Horney called "Shoulding on yourself," or worse, Shoulding on someone else.
Sometimes people nod. Sometimes they comply. Sometimes things even get better. For a while. Then they rebel. Or disappear. Or return to the old behavior. Or come back angry that somebody “made” them do something.
I eventually became suspicious of the word relapse, too. A person has to change before they can truly relapse. Actually change. That is different than just complying.
Otherwise, sometimes it is simply the same movie with different snacks and a slightly stronger opinion. What looks like change may only have been compliance under pressure. And compliance creates movement without ownership. Movement can happen quickly. Ownership usually can’t.
The Difference in Plain English
Agency tends to look like: Choice, even when the choice is uncomfortable. Or even costly.
It becomes responsibility without humiliation. Insight that arrives later than I would prefer. Progress that is slow, uneven, and occasionally irritating. People surprising you. Sometimes pleasantly. Sometimes not.
And one of my favorites: Someone “quotes me” by explaining what they understood rather than repeating what I said. That means it has become theirs.
Coercion looks different. Compliance followed by resentment. Short-term improvement that collapses when the pressure disappears. Moral certainty replacing curiosity. Someone else being blamed when the plan fails. And a growing investment by the helper in getting the other person to behave correctly. I could tell you about a kid that I challenged about his commitment to change. I got a message back that started with a expletive that started with F, and then a pointed "you" at me. He had proved me wrong. He didn't know that I had won.
As a therapist, I eventually learned a rule: If I am working harder than the client, something is wrong. Usually I am working too hard. And sometimes I am trying to steal their agency for their own good.
I’m not particularly proud of that.
I have tried to save people with the verbal equivalent of a cattle prod. Eventually I became less interested in telling people what to do. I made a friend of Socrates and started asking questions. Thousands of hours of questions. This has annoyed more than one person in my life.
Apparently asking someone what they think when they would prefer that you simply tell them the answer is not universally appreciated. Who knew? But I began to understand something. Advice can become a way of talking myself into believing I’m in charge.
Questions remind me that I’m not.
The Moment Helping Becomes Controlling
I think this may be the harder problem. Most people who become controlling don’t wake up in the morning and say: Today I would like to deprive someone of agency. They say: I’m trying to help. And sometimes they really are. Sometimes they are even right.
That is what makes this difficult.
The temptation to control is strongest when I am convinced I know what will happen if you make the wrong decision. I can see the wreck coming. You can’t. Or at least I don’t think you can. So I push. Then I persuade harder. Then I increase the consequences. Then I explain that I am only doing this because I care. Somewhere along the way, helping can become controlling without anybody announcing the transition. That threshold matters.
This Is Where Public Life Gets Weird
The same pattern shows up outside therapy rooms. Families do it. Institutions do it. Governments do it.
Communities do it. When people feel unsafe, control becomes attractive. When people feel unheard, force becomes easier to justify. When people become morally certain, coercion can start looking like virtue. Then language changes. Force becomes care. Control becomes compassion. Compliance becomes progress.
That does not require evil people. It requires ordinary people who are certain enough of the desired outcome that another person’s resistance starts looking like a defect. Agency is slow. Messy. Unpredictable. Emotionally unsatisfying. Coercion promises relief. Immediate relief. And usually someone to blame.
Blame is the comfort food of uncertainty. I have eaten it directly from the carton more than once. I apologize to all those I blamed for what eventually turned out to include my choices.
The Curmudgeon Observation
After enough years in therapy rooms, you begin noticing patterns. People certainly fail because they lack information sometimes. They also fail because they lack resources, skills, opportunity, support, health, or plain old luck. But there is another possibility I think we miss. Sometimes what looks like resistance to change is actually change being pushed faster than ownership can develop. That matters in: families,
relationships, institutions, organizations, governments, revolutions, and therapists who talk too much.
I learned to say some version of this: I can tell you what I think you should do. But if you do it only because I told you to, your commitment is mostly to me. That sentence ended some conversations. It started better ones.
The Cost of Agency
Nobody likes this part. Choice isn’t free. Agency costs something. Time. Frustration. Failure. Regret.
Ambiguity. Discomfort. And responsibility.
That is why control keeps returning. It looks cheaper. Faster. Cleaner. More efficient. But coercion often moves the cost downstream, where it becomes harder to see and more expensive to repair.
In therapy, it may produce compliance without change. In families, rebellion. In institutions, quiet resistance. In politics, backlash. In systems, brittle stability. I used to tell families something they did not particularly enjoy hearing.
If you want to increase the chances that your son, daughter, husband, wife—or whoever you love—returns to drinking, drugs, fighting, running, or whatever destructive pattern they have been using, one effective strategy is to try to control them.
Weird, eh?
Agency Does Not Mean No Constraints
This is where I need to be careful. Agency is not the absence of constraint. That would be nonsense. A drunk driver does not have an unlimited right to continue driving. A parent stops a toddler from walking into traffic. Professional licenses come with boundaries. Communities have rules. Actions have consequences.
So the alternative to coercion is not: Everybody does whatever they want.
The better question is: How much meaningful choice can remain inside necessary constraint? That is where restraint becomes important. Not merely restraining somebody else. Restraining myself. My certainty. My urgency. My desire to fix. My willingness to use every bit of leverage available simply because I can. Sometimes the person with the most power is the person who most needs restraint.
I will come back to that.
Why I Still Choose Agency
Even when I don’t feel like it. Even when it bites me in the rear-end. Change becomes more durable when the person owns it.
If I do not own the costs of my choices, I can always assign responsibility somewhere else. you know, how it goes, My therapist made me do it. My spouse made me. My parents made me. The court made me. The treatment center made me. The government made me.
Sometimes those forces really do constrain choices. But ownership begins with asking: Given the choices that actually remain, which one is mine? People do not need me to control them. They need enough space, structure, information, feedback, and restraint to exercise whatever agency remains available. And every time I forget that—every time I become too invested in pushing, persuading, or pressuring—I eventually discover that I have created another problem to clean up.
Agency is slower. It is also more real.
The Remainder
Viktor Frankl pushed this idea toward an extreme. His circumstances in the concentration camps were not chosen, fair, or negotiable. Enormous portions of ordinary human agency had been violently stripped away. Yet he later argued that even under profound constraint, some remainder of choice could persist—particularly in how a person responded to circumstances that could not themselves be chosen. I don’t think that means people are infinitely free.
Clearly they aren’t.
It means agency may not disappear simply because constraint increases. Sometimes what remains is very small. But the remainder matters. Maybe that is one reason I eventually left therapy.
I became less interested in fixing people and more interested in understanding what choices remained available to them.
That question still interests me.
The Closing Thought
I am suspicious of systems—personal, professional, or political—that promise improvement primarily by overriding choice. Not because choice always produces good outcomes. It doesn’t. People make terrible decisions. They make terrible decisions often. I have made enough of my own to establish the point.
But I have also watched coercion turn against the people who believed they were using it for good.
And I don’t say that from a distance. I have watched myself do it. I have pushed someone toward what I was convinced was best because uncertainty made me itchy and control looked like relief. I have confused being right with being responsible for somebody else’s decision. I have mistaken compliance for change.
After a few decades as a therapist, the pattern stopped being theoretical. I started noticing my own fingerprints on the glass.
Shoot.
Where did I put the glass cleaner?