Tactical Empathy: Listening as Influence

The Objection You Never Heard

We closed the last lesson with an asset and a spending plan. Credibility is built by costly truth, drained by exposed gaps, and spent on the one or two asks that actually matter. And we said the account most leaders neglect is Care — the listener's estimate of whether you are for them or whether they are a means to something you want. Here is the question that leaves open: how do you fund that account when the person across from you has already decided you are not for them?

You do not fund it by saying so. Nothing is cheaper than a leader announcing that he cares. You fund it by demonstrating that you have understood the objection — and understanding is demonstrable in a way that caring is not. This is the move almost no one makes under pressure, because under pressure every instinct pushes the other direction. Somebody attacks your position and your body prepares a defense. The defense is the mistake. Not because defending is wrong, but because you are about to defend against an objection you have not actually heard.

Watch how it usually goes. A council member says your housing plan will destroy his district. You have anticipated this, so you deploy the numbers: units produced, tenants protected, the affordability tiers. He does not move. You conclude he is not arguing in good faith. But his objection was never about unit counts. His objection was that the last three administrations promised his neighborhood something and delivered a construction site, and he will be the one at the community board meeting explaining why he believed you. You answered a question about arithmetic. He asked a question about humiliation. You will never win that exchange with better numbers, and you could have discovered the real objection in ninety seconds by asking.

This is why listening belongs in an influence module rather than a manners module. Listening is not the courteous pause before your argument. It is reconnaissance, and it is also the argument. When a person feels accurately understood, two things happen that nothing else produces. The defensive posture drops, because a defense is only necessary against a misunderstanding. And a debt opens — a mild, reliable obligation to extend the same courtesy back. You have made a deposit in the Care account that the other person watched you make.

The tradition that has pressed this furthest is not corporate communication training. It is hostage negotiation, where the negotiator has no leverage, no authority, no ability to compel, and a person on the other end of the phone who is frightened, armed, and convinced that everyone is lying to him. Strip a negotiator of every instrument except words and what remains is the pure craft. That is where we are going to get our tools, and then we are going to be honest about their evidentiary status.

The Ladder to 'That's Right'

Chris Voss spent years as the FBI's lead international kidnapping negotiator and set his method down in Never Split the Difference. He calls the core of it tactical empathy — the deliberate, disciplined demonstration that you understand another person's position and the feelings driving it, without necessarily agreeing with either. I want to organize his tools as a ladder, because in practice they are sequential and each rung makes the next one possible. Silence, Mirror, Label, Calibrated Question, Summary. Call it the Ladder to 'That's Right.'

The first rung is silence, and it is the one leaders skip. When a person finishes a sentence, wait. Most people have not finished; they have paused, and the second thing they say is usually truer than the first, because the first was the prepared version. A leader who fills every gap will spend a career hearing only prepared versions. Silence costs nothing and it is the single highest-yield behavior on this list.

The second rung is the mirror. Repeat the last few words the person said, as a question, and stop. That is the entire technique. It sounds absurdly simple and it works because it does two things at once: it demonstrates that you were listening at the level of their actual words, and it invites elaboration without asking a question that might steer. A commissioner says the timeline is unrealistic given what we are dealing with, and you say, given what you are dealing with? And then you find out what he is actually dealing with, which is a vacancy rate he has not told you about.

The third rung is the label, and this is the tool with the highest leverage in hostile settings. You name the emotion or the position you are observing, in a tentative form, and then you stop talking. It sounds like you are frustrated that this is the third time you have raised this. Or, it seems like you think I am here for the photograph. Voss recommends starting with it sounds like, it seems like, or it looks like rather than I hear that you or I think you feel — because the first form describes what you observe and leaves the other person free to correct you, while the second inserts you into their experience. Two things about labels. They work on negative emotions by naming them, which reliably reduces their charge, and they work best when you label the hardest thing rather than the safest thing. Labeling a mild irritation is useless. Labeling the accusation the person has not yet said out loud is what changes a room.

The fourth rung is the calibrated question — an open question, usually beginning with what or how, that cannot be answered with a yes or a no and that hands the other person the problem. How am I supposed to do that? What about this matters most to you? What would it take? Voss makes a point of avoiding why, which in English reads as an accusation almost regardless of tone. A calibrated question converts an attack into a shared engineering problem, and it does so without conceding anything.

The fifth rung is the summary, and it is the one that closes the loop. You state the other person's position, including its emotional core, so accurately that they say: that's right. Voss draws a sharp and useful line between that's right and you're right. That's right means they heard their own position come out of your mouth intact and confirmed it — that is agreement about the facts of their world, and it is the breakthrough. You're right usually means the opposite: it is what people say to end a conversation they have decided is not going anywhere. If you are getting a lot of you're rights from your team, you are not persuading them. You are exhausting them.

Slow to Speak, and the Man Who Answered Questions With Questions

Now let me put this on its actual foundation, because none of it is new and none of it originates in law enforcement. James 1:19, NASB 1995: 'This you know, my beloved brethren. But everyone must be quick to hear, slow to speak and slow to anger.' The Greek is worth seeing. Quick to hear is tachys eis to akousai, and slow to speak is bradys eis to lalēsai — bradys is the ordinary word for slow, sluggish, dull. James is not recommending a pause for effect. He is prescribing a deliberate reduction in speed on the output side and an increase on the intake side, and then he attaches anger to the same list, which tells you what he thinks the failure mode is. Leaders who speak fast get angry fast, and the second follows from the first more often than we admit.

Then there is the pattern in how Jesus actually conducted conversations, and it is the most striking thing about His method once you start counting. He asks constantly, and He very often answers a question with a question. Martin Copenhaver counted three hundred and seven questions asked by Jesus in the Gospels, one hundred and eighty-three questions put to Him, and by his count only three that He answers directly. I want to be careful with that statistic, because counts like this depend entirely on how you delimit a question and how you handle parallel accounts across the Gospels — other scholars working the same texts arrive at different numbers, and Copenhaver's precision should be taken as an argument rather than a measurement. But the pattern the count is pointing at is unmistakable and anyone can verify it by reading.

Look at three. A blind man is shouting at Him from the roadside, and the need could not be more obvious. Jesus asks anyway: 'What do you want Me to do for you?' — Mark 10:51, NASB 1995. That is a calibrated question asked of a man whose need was visible from fifty feet, and it hands him the dignity of stating it himself. A lawyer asks what he must do to inherit eternal life, and Jesus answers: 'What is written in the Law? How does it read to you?' — Luke 10:26. He does not supply the answer the man already possesses; he makes the man produce it, which is the difference between being told and knowing. And when the chief priests demand to know by what authority He is acting, He says: 'I will ask you one question, and you answer Me, and then I will tell you by what authority I do these things. Was the baptism of John from heaven, or from men? Answer Me.' — Mark 11:29-30. That is a man declining to accept a frame, and doing it by asking rather than refusing.

On the Emmaus road He walks with two disciples who are grieving and confused about events He was personally at the center of, and He opens with: 'What are these words that you are exchanging with one another as you are walking?' — Luke 24:17. He knew. He asked anyway, and He let them tell Him the story wrong, at length, before He corrected anything. If you want a model for how to enter a conversation with someone whose account of the facts is mistaken, that is it, and notice how much later the correction comes.

I am not claiming Jesus was practicing tactical empathy. The direction of the claim runs the other way: the negotiation craft is a partial, secularized rediscovery of something that was always there in how He dealt with people. And there is a difference the parallel makes visible. A negotiator asks in order to move someone toward an outcome he has already selected. Jesus asks in order to bring a person to see something true about himself — which is a form of questioning that can afford to receive an answer the questioner did not want.

What the Evidence Actually Supports

I am going to be direct about the status of this material, because the leadership market is full of people who cite Voss as though he had run randomized trials. He did not, and he does not claim to have. Never Split the Difference is practitioner methodology — craft knowledge distilled from a career in federal hostage negotiation, refined against real cases with real stakes, and reported by the practitioner himself. That is a legitimate and valuable kind of knowledge. It is not peer-reviewed psychology, it carries the selection and memory biases every practitioner account carries, and the specific effect sizes are simply unknown. Label it Practitioner consensus, use it, and do not oversell it. One specific caution: Voss repeats the familiar claim that communication is seven percent words, thirty-eight percent tone, and fifty-five percent body language. That figure traces to Albert Mehrabian's studies from the nineteen-sixties, which concerned inconsistent messages about feelings and attitudes under narrow laboratory conditions. Mehrabian himself has said publicly that the generalization to communication at large is unwarranted. Drop that number. Keep the rest.

There is controlled research on listening, and it is encouraging and modest in roughly equal measure. Harry Weger and colleagues, publishing in the International Journal of Listening, compared active-listening responses against simple acknowledgment and against advice in initial conversations and found active listening produced greater felt understanding — though the advantage over other supportive responses was not enormous. Work by Guy Itzchakov and Avraham Kluger has found that speakers who experience high-quality listening become less defensive and more willing to hold complexity in their own attitudes, including recognizing inconsistencies in views they had been defending. That finding is directly relevant to leadership: good listening does not merely make people feel better, it appears to loosen the grip of a position enough for the person to look at it.

The most important finding for our purposes is a corrective, and it should change your practice. Tal Eyal, Mary Steffel, and Nicholas Epley ran a large series of experiments testing whether perspective-taking — deliberately imagining what another person is thinking and feeling — improves accuracy in reading them. Across the set, it did not. In several conditions it made accuracy worse, presumably because a confident simulation of another mind is still your own mind. What did improve accuracy was what they call getting perspective: actually asking the person and listening to the answer. Sit with the implication. The empathetic leader who imagines his way into his employees' experience is often less accurate than the blunt one who asks. The label, the mirror, and the calibrated question are all forms of asking. That is why they work, and it is the strongest empirical support this lesson has.

One more line of evidence, with a caveat I want to state plainly. Field experiments by David Broockman and Joshua Kalla, published in Science and later in the American Political Science Review, tested a canvassing method built on nonjudgmental listening and the exchange of personal narratives, and found durable attitude change from a single ten-minute conversation on contested social issues. I cite the mechanism, not the target attitudes; nothing in this academy's commitments is settled by a canvassing study, and I am not asking you to endorse any position those campaigns advanced. What the work establishes is that a brief, genuinely non-adversarial conversation can move an attitude further and longer than a great deal of expensive messaging. The backstory is also worth your time as a lesson in verification: this line of research began when Broockman and Kalla investigated an earlier, celebrated study on the same question, found it fabricated, and forced its retraction — and then went and did the work honestly. That is the standard.

So the honest summary is this. The controlled literature supports the underlying claim that being accurately understood reduces defensiveness and opens attitudes, and it supports asking over assuming. The specific verbal moves — the mirror, the label, the calibrated question — rest on practitioner evidence rather than experimental evidence. That is enough to use them. It is not enough to promise anyone a result.

Thou, Not It — and the Reporter in the Third Row

Everything in this lesson can be run in two entirely different modes, and the techniques are identical in both. Martin Buber gave us the language for the difference in 1923. He argued that we relate to the world through two basic word-pairs. I-It is the stance of experience and use: the other is an object I observe, categorize, and employ for my purposes. I-Thou is the stance of encounter: the other is met whole, addressed rather than described, and I am changed by the meeting. Buber's claim, in Ronald Gregor Smith's rendering, is that all real living is meeting. Walter Kaufmann's later translation puts it as encounter, which is closer to the German Begegnung.

Now apply that to what we just learned. You can label a person's fear in order to meet them or in order to open them. You can mirror in order to understand or in order to extract. The words coming out of your mouth are the same words. The difference is whether the person in front of you is a Thou or an It, and here is what makes this more than a philosophical nicety: people detect the difference, unreliably in the moment and almost infallibly over time. A leader who runs tactical empathy in I-It mode builds a reputation for being very good at making people feel heard, which is one of the more corrosive reputations available. And this is where the academy's commitment lands with full weight. Every person you will ever negotiate with is an image bearer — not because they are reasonable, not because they are your kind, and not because they deserve it. That is not a reason to listen well. It is the reason, and it is the only one that holds when listening well stops producing what you want.

Practically, in public office, this is the highest-value tool you have for hostile press and open town halls. Understand what a hostile question actually is. It is an instrument built to produce a specific product, and the product is your reaction. A reporter who asks whether you are hiding something is not primarily seeking information; he is seeking footage of a man who looks like he is hiding something. Argue with the question and you manufacture it for him. Label it and the instrument stops working. It sounds like you think I have not been straight with you about this. Then stop. Let the silence sit. Then answer the actual question, plainly and with a number in it. What you have done is name the accusation before he could imply it, which converts a segment engineered around your defensiveness into a segment about a man who was not defensive.

The same move handles the town hall, where the emotional stakes are higher and the person shouting at you usually has a legitimate grievance buried under an illegitimate delivery. Do not correct the delivery. Climb the ladder. Mirror the last few words, label the thing under the anger — it seems like you have raised this before and nothing happened — and ask a calibrated question about what would actually constitute a fix. You will not satisfy everyone in that room and you should stop trying. But the two hundred people watching are not evaluating whether you won the exchange. They are evaluating whether you are the kind of person who hears them, and that evaluation is the one that follows you to the polls.

One boundary, because tactical empathy is not a substitute for candor and this is where nice leaders get into trouble. Understanding a position is not agreeing with it, and a leader who climbs the ladder and then never states his own view has simply found a sophisticated way to avoid the hard part. The order is what matters: understand first, fully and demonstrably, then say the true thing plainly. Charitable first, then unflinching. You have not earned the right to be unflinching until the other person would agree you stated their case fairly. In the next lesson we take up what happens after you have listened — how the same true thing, said two different ways, produces two different outcomes, and exactly where that stops being stewardship and starts being manipulation.