What Marshall Rosenberg Taught Me About the Need Underneath the Need

Most resistance isn't disagreement — it's a need nobody has asked about yet.

I recently finished Marshall Rosenberg's Nonviolent Communication, and I have a feeling I'm not actually finished with it.

I found the book indirectly.

I was listening to Satya Nadella speak at Stanford about AI, innovation, and leadership. He got onto empathy — developing a sense of where the other person is coming from, as he put it, "not having your amygdala always triggered" — and he mentioned a book: Marshall Rosenberg's Nonviolent Communication.

That was it. He didn't tell me to read it. He just planted a seed.

Eventually I picked it up. I found out later that he'd handed the book to Microsoft's senior leadership team when he became CEO in 2014, which made more sense of why it came up at all.

I've read quite a few books about leadership, communication, and understanding people. This one affected me differently.

01A way of seeing, not a script

Listening to Rosenberg himself mattered. I listened to the audio version — five hours of him teaching it in his own voice — and what got me wasn't the four-step framework. It was his voice, his cadence, the way he moved through a conversation.

In telling his own story, he thanks a sociologist, Michael Hakeem, for helping him see "the scientific limitations and the social and political dangers of practicing psychology in the way that I had been trained: with a pathology-based understanding of human beings." Seeing those limits, he wrote, "stimulated me to search for ways of practicing a different psychology."

He was not looking for a better way to categorize people. He was looking for a way to be with them.

That reframed the four familiar elements of NVC — observations, feelings, needs, and requests. Rosenberg was insistent they were not a script. "NVC is not a set formula," he wrote, and more pointedly:

But he didn't treat the form as optional either. He gives a chapter to each element. He describes applying them mechanically, without awareness of the underlying purpose, as a stage beginners pass through rather than a misuse of the method.

The form is scaffolding. The attention it holds in place is the point.

That distinction matters to me. Because if I simply learn the right sequence of words, I could technically "do NVC" without becoming any more empathetic.

What seems much harder — and much more valuable — is developing the instinct to become curious about what's alive in another person, and in myself.

02What's underneath

That idea kept pulling me down to the root.

We argue about what someone said. We resist a change at work. We get frustrated with a spouse, a friend, a coworker.

And underneath all of it, Rosenberg keeps asking about something more fundamental:

What's alive in this person right now? And what's alive in me?

That last part has stayed with me.

If someone resists something I'm trying to accomplish as a leader, I can try harder to persuade them. Or I can get curious.

What might they be trying to protect? And what's happening in me that makes their resistance hard to sit with?

Maybe I'm frustrated because I need to see progress. Maybe I'm worried I'm not being effective. And maybe what looks like stubbornness from where I'm standing is someone trying to hold onto a say in something that's happening to them.

I don't actually know until we have the conversation.

That's an important distinction. NVC isn't a better method for diagnosing people. It's almost the opposite. It's an invitation to stay curious enough to ask.

03Resistance is information

That has also changed how I think about technology.

I've described my approach as people-centered technology for a while now. This book gave me a sharper way to say what I mean by it.

People-centered technology doesn't start by defending the solution. It starts with the need.

Not "how do we get people to use this," but:

• Where are people hitting friction?
• What are they trying to accomplish?
• What are they afraid of losing?
• What isn't working for them today?

Technology becomes one possible strategy for meeting a need, rather than the objective itself.

And resistance becomes information.

When someone is worried about AI, I don't have to begin by explaining why they're wrong. There's usually a need underneath the worry — security, competence, autonomy, trust, contribution.

Notice that "a job I'm good at" isn't on that list. That's a strategy for meeting several things that are.

I don't know which need it is. That's the whole point. But I can ask.

In practice this has changed the small stuff more than the big stuff. A ticket. A complaint about a tool. Someone flagging a concern about a rollout.

Someone tells me they don't have time for it. That's usually true. It's also usually not the whole thing.

So instead of solving for the stated objection, I've been trying to listen for what's underneath it — and then say it back to them as a need rather than as an obstacle. Sometimes that surfaces a strategy nobody had considered, including not doing the rollout that way, or not on that timeline.

I'm not consistently good at it. But naming what I think I'm hearing, and being wrong out loud often enough to get corrected, has changed those conversations more than I expected.

And once you understand what someone actually needs, you can often find something that works for the person and the organization, instead of assuming one side has to lose for the other to win.

I want to be careful here, because there's a version of this that's just persuasion with better manners. Asking about someone's need in order to route around it isn't curiosity. It's a technique.

Sometimes the organization's need and the person's need genuinely collide, and part of my job is making a call that costs someone something. Curiosity doesn't dissolve that. It just means I can't tell myself I didn't know.

04The part I'm not sure I can do

There are parts of NVC I still find intimidating.

Rosenberg could stay present in extraordinarily difficult conversations and keep searching beneath someone's words for the feelings and needs underneath them.

Sometimes his conversations reminded me of Columbo. The apparently simple question. Then another. Then starting to walk away, turning around, and saying, "Oh, just one more thing."

Not confrontational, but working gently beneath the surface until something previously hidden becomes visible.

The difference is that Columbo already knew what he was looking for. Rosenberg didn't. He was asking without a theory of the case. That's the much harder version, and it's the part I'm not sure I can do yet.

Some of the examples in the book are extraordinary — anger, hostility, situations that were genuinely unsafe. I finished a few of them thinking, I understand what you're doing. I'm not sure I could do it.

That skepticism feels important to hold onto. This isn't magic. Empathy doesn't guarantee another person's response, and there are situations where safety has to come first.

But one story in the book helped me see how someone actually begins.

Rosenberg describes a friend, Sam Williams, who jotted the basic components of the process on a three-by-five card and used it as a cheat sheet at work. When his boss confronted him, Sam would stop, look at the card in his hand, and take the time to remember how he wanted to respond. He used it at home too, pulling it out whenever there was an argument in the family.

After about a month, he felt comfortable enough to put it away.

Then one evening he and Scottie, his four-year-old, got into a conflict over the television, and it wasn't going well.

I loved that. The card wasn't the destination. It was scaffolding.

Maybe getting better at this isn't about reading a book and suddenly communicating perfectly. Maybe it's about practicing a different way of paying attention until some of that attention becomes instinctive.

05Where practice starts

I've already caught myself doing it in small ways.

A few weeks ago I was in the sauna at the gym after a lift, talking with two guys I'd never met. One of those unexpectedly good conversations with people who had been strangers twenty minutes earlier.

I noticed myself asking slightly different questions. I was more curious about what was underneath what they were telling me.

On my way out I said something like, "Thanks. Good conversation."

Walking to my car, I realized that wasn't quite what I meant.

What I actually wanted to say was what the conversation had given me. It had met something in me — a need for connection and camaraderie.

Now, I'm not going to walk around the gym announcing that strangers have met my need for human connection. That isn't me.

But I could have said, "I really enjoyed talking with you guys. It's nice actually connecting with people here."

It's a tiny difference. But that's probably where practice starts.

Not learning to sound like Marshall Rosenberg. Learning to notice a little more of what's actually happening in myself and another person, and then finding language that's genuinely my own.

The need underneath the need

Rosenberg kept bringing me back underneath all the complexity.

Human beings have needs. Our strategies for meeting them collide. Our language obscures them. Our judgments obscure them. And sometimes what we're certain we need isn't the need at all — it's just the strategy we've attached ourselves to for meeting something deeper.

That's probably why I don't feel finished with this book. I expect I'll read it again. Not to memorize the four steps, but to get better at seeing what Rosenberg was trying to see.

Satya Nadella never told me to read this book. He said it was a great read if I hadn't read it, in the middle of a talk about something else, and the thread stayed with me until I followed it.

He planted a seed. So I'll leave one here.