Library / Understanding Is Leverage
Chapter 10 — Safety and Disclosure
In the mid-1990s a young researcher named Amy Edmondson went looking, inside several hospitals, for a sensible result: that better nursing teams make fewer mistakes. She had the teams' quality ratings and she had their rates of reported medication errors, and she expected the two to line up the obvious way — strong teams, few errors. They lined up the opposite way. The better the team, by every independent measure, the more errors it reported. For a moment the data seemed to be saying that good teams are sloppier, which was absurd.
The resolution, when it came, reframed an entire field. The good teams weren't making more errors. They were reporting more — because on a good team it was safe to admit a mistake, and on a bad team it wasn't. Where a nurse who flagged her own error got support and a fix, errors surfaced and got caught. Where flagging an error got you humiliated or blamed, the errors didn't disappear; they just went silent, buried to avoid the punishment, festering. Edmondson had stumbled onto the thing that separates teams that learn from teams that quietly rot, and she gave it a name: psychological safety — the shared belief that you can take an interpersonal risk, admit a mistake, ask a naive question, or voice a concern, without being shamed or punished for it.
Two decades later, Google went looking for the same thing without knowing it. In a multi-year study of 180 of its own teams called Project Aristotle, the company tried to find what made some teams excel and others stall. It expected the answer to be about talent — the right mix of stars. It wasn't. The single largest factor, ahead of who was on the team, was psychological safety: whether members felt safe enough to speak up, disagree, and be wrong out loud. The same invisible quality Edmondson found in the hospital wards turned out to be the best predictor of performance in one of the most data-saturated companies on earth.
This chapter is about why that quality is also the precondition for every good negotiation, and how you create it across a table — including the surprisingly specific moves that get another human being to tell you the truth.
Why safety comes before substance
Recall the central claim of Part I: the whole game is to get past stated positions to real interests, because that's where the value is. Now add the uncomfortable fact that the other side's real interests, constraints, fears, and priorities are exactly the things people hide when they don't feel safe. A counterpart who suspects you'll exploit any admission will never tell you that the deadline is what's really killing them, that their boss has tied their hands on price, that the relationship matters more to them than the last five percent. They'll posture instead. They'll defend a position. And you'll negotiate blind, fighting over surface demands because the information that would have unlocked a better deal stayed locked inside a person who didn't trust you with it.
Safety is what unlocks it. When the other side believes that opening up won't be turned into a weapon, they disclose — and disclosure is the raw material of every creative agreement. This is the bridge from the last chapter made concrete: trust is the medium, and psychological safety is the specific, buildable condition that lets trust do its work in the room. The negotiator's job is not just to be trustworthy but to actively make it safe for the other person to be honest.
The reciprocity of disclosure
So how do you get someone to open up? The deepest answer comes from one of the most charming experiments in social psychology. In 1997, the psychologist Arthur Aron brought strangers into his lab in pairs and had them work through a list of thirty-six questions together. The questions weren't random; they were arranged in three escalating sets, starting light ("Would you like to be famous? In what way?") and deepening, step by step, into genuinely intimate territory ("When did you last cry in front of another person?", "Tell your partner what you like about them"). Pairs who did this — total strangers — reported dramatically more closeness afterward than pairs who'd spent the same time on small talk. The procedure became famous as a way to "make anyone fall in love"; one early pair of participants later married.
The mechanism Aron identified is the engine you can use at the table: escalating, reciprocal self-disclosure. Closeness is built when one person reveals something real, the other reciprocates with something of similar depth, and the exchange ratchets gently downward into more honest territory, each disclosure signaling I trust you with this, and it's safe for you to trust me back. The key word is reciprocal. Disclosure that isn't returned doesn't build closeness; it just leaves the discloser exposed. But disclosure that is met with matching disclosure creates a spiral of growing trust.
The applied lesson is that to get the other side to open up, you usually have to go first. Waiting for a guarded counterpart to volunteer their real constraints is waiting forever; someone has to take the first interpersonal risk, and the more skilled, more secure negotiator is the one who does it — offering a genuine piece of context ("I'll be honest, the timeline is the part that worries my team") that invites a matching honesty in return. You prime the pump. You model the behavior you want. And then you make it safe to reciprocate by doing the one thing the next section is about.
Making it safe — and not blowing it
Creating safety is mostly about what you don't do at the moment of disclosure. When the other side admits something — a constraint, a worry, a mistake, a soft spot — the entire future of the negotiation can turn on your next three seconds. The instinct, especially for an aggressive negotiator, is to pounce: they just admitted the deadline is killing them, so squeeze. That instinct is a catastrophe. The moment you exploit a disclosure, you teach the other person, permanently, that honesty with you is dangerous — and the flow of information you depend on shuts off, not just for this admission but for every one after it. The pounce wins the point and loses the game.
The discipline instead is to acknowledge rather than exploit. When someone hands you something real, you label it and honor it ("that timeline pressure sounds genuinely hard — let's see what we can do about it") before you ever think about leverage. You reward disclosure with safety, which produces more disclosure, which is worth far more over the arc of a negotiation than any single admission you could have squeezed. This is the practical face of the low self-orientation from Chapter 9: the negotiator who is visibly trying to solve the other person's problem is the negotiator people tell the truth to.
Two honest cautions keep this from becoming naïve. First, disclosure is calibrated, not reckless. Going first does not mean handing over your weak BATNA or your real walk-away; it means offering genuine context that builds trust without surrendering leverage. There's a craft to revealing things that are true and trust-building but not self-destructive, and the skill is in the calibration. Second, reciprocity assumes good faith, and not everyone has it. A bad-faith counterpart will take your disclosure and give nothing back, or worse, weaponize it — and the tell is exactly that: you go first, and the spiral doesn't form, because they pocket your openness and stay closed. When that happens, stop disclosing, fall back on structure and your BATNA, and recognize that you're in a different kind of negotiation than the one this chapter describes. Safety is how good-faith deals get made; it is not a strategy for dealing with someone who's decided to exploit you.
Trust, safety, disclosure — these are the relational foundation. But all of them depend on something inside you: the capacity to stay regulated enough to create safety instead of destroying it, to read the other person accurately, and not to let your own emotions blow up the very conditions you're trying to build. That self-mastery is the subject of the next chapter, and it begins with a surprising question: does it ever pay to get angry?
Try this. In your next real negotiation, go first with one genuine, calibrated disclosure — a true constraint or concern that's safe to share — and then watch whether it's reciprocated. If it is, you've found a good-faith partner and a spiral you can build on. If it isn't, you've learned something just as valuable about who you're dealing with.
Sources & notes
Amy Edmondson's finding that higher-performing hospital teams reported more errors — because they felt safe to do so — and her development of the concept of psychological safety appear in "Psychological Safety and Learning Behavior in Work Teams," Administrative Science Quarterly (1999) and in her book The Fearless Organization (2018). Google's Project Aristotle, which identified psychological safety as the top factor in team effectiveness across 180 teams, was reported by the company's re:Work initiative and popularized in The New York Times Magazine (Charles Duhigg, 2016). Arthur Aron's 36-question "closeness" procedure is from Aron et al., "The Experimental Generation of Interpersonal Closeness," Personality and Social Psychology Bulletin (1997). The principle of reciprocal self-disclosure is well established in social psychology; its application to negotiation, and the warning that it can be exploited by bad-faith counterparts, is this book's synthesis.