Framing and the Power of the Ask

Six Hundred People and Two Sentences

Last lesson ended with an instruction and a boundary. Understand first, demonstrably, and then say the true thing without flinching. The boundary was that understanding is not agreement and that a leader who climbs the listening ladder and never states his position has found an elegant way to avoid his job. So now we are at the saying. And the saying turns out to be less simple than it looks, because the same true thing, said two ways, produces two different answers from the same person.

In 1981, Amos Tversky and Daniel Kahneman published a paper in Science that has been replicated more thoroughly than most findings in social science. They gave people a public health problem. An unusual disease is expected to kill six hundred people. Two programs are proposed. One group was told that Program A will save two hundred people, and Program B has a one-third probability of saving all six hundred and a two-thirds probability of saving no one. Roughly seventy-two percent chose the certain option, Program A. A second group got the identical arithmetic described the other way: Program C means four hundred people will die, and Program D carries a one-third probability that nobody will die and a two-thirds probability that six hundred will die. Roughly seventy-eight percent chose the gamble, Program D.

Sit with that. Two hundred saved out of six hundred is four hundred dead out of six hundred. The programs are numerically identical. The preference reverses almost completely. And the people whose preferences reversed were not fools; the effect shows up in physicians, in statisticians, and in people who have already been told about it. Frame the outcome as a gain and people protect the sure thing. Frame it as a loss and people gamble to avoid the certain loss.

The honest scientific status matters here, so let me give it. This is not one of the shaky findings. The framing effect was included in the Many Labs replication project, which ran it across dozens of samples in different countries, and it replicated robustly with a substantial effect size. If you want to know which behavioral findings survived the replication crisis, this is one of them. Some of the theoretical apparatus built on top of it is shakier, and I will get to that.

Here is what a leader is supposed to take from it, and it is not a technique. It is a fact about the world that removes an option you thought you had. There is no way to state that public health problem neutrally. You must choose lives saved or lives lost; the language has no third setting. There is no frameless sentence. Which means the leader who says he simply presents the facts and lets people decide is not describing a practice, he is describing a self-image — he is framing, unreflectively, and taking no responsibility for the frame he chose. Once you understand that, the ethical question changes shape entirely. It is not whether to frame. It is which frame is truest to the thing, and how you would defend that choice if someone asked.

The Objection, at Full Strength

Before I teach you a single move, I owe you the strongest case against all of it, stated the way its best advocates would state it. This is the hardest ethical question in this module and I am not going to give you a soft version of it in order to knock it over.

The objection runs like this. Rational choice has an axiom called description invariance: if two descriptions refer to the same state of affairs, a rational agent's preference between them should not change. The framing effect is defined as a violation of that axiom. So framing does not work by giving a person better information or a better argument. It works precisely in the gap between what a person would choose on reflection and what they choose when the words are arranged a particular way. It is a technique whose whole efficacy lies in a defect of human reasoning, and using it means routing around a person's rational agency rather than engaging it.

Kant sharpens the same point in the Groundwork: act so that you treat humanity, in yourself and in others, never merely as a means but always also as an end. When I engineer sentences to exploit a known bias in your cognition, I am using your machinery as an instrument to produce an output I selected. Robert Noggle's analysis of manipulation makes it more precise still: manipulation is leading a person's beliefs, desires, or emotions away from the ideals that ought to govern them — and framing does exactly that by design.

And then the test that finishes the argument. A manipulation is any influence whose disclosure would defeat it. If I say, I am going to describe this as four hundred deaths rather than two hundred lives saved because that phrasing will push you toward the option I prefer — the effect evaporates and you are annoyed. Persuasion survives daylight. Framing does not. Therefore framing is manipulation, and the fact that everyone does it makes it universal, not innocent.

That is a serious argument and parts of it are simply correct. Here is where it fails, and it fails at one specific joint. The disclosure test assumes there was a non-manipulative alternative — that you could have described the problem without a frame and chose not to. There was no such alternative. Every description is a description. If I flip to the gain frame instead, that is also a frame, also chosen, also pushing. Neutrality is not the baseline that framing departs from; it is a position that does not exist. So the disclosure test has to be applied to a real choice between real options, and stated properly it becomes this: would you be willing to show your audience the choice you made among available frames, and your reason for it? That version is answerable. I chose to describe the shelter budget in terms of people who will sleep outside rather than dollars per bed, because dollars per bed is technically accurate and systematically hides who we are talking about. Say that out loud and the frame does not collapse. It strengthens, because the reason for it is the argument.

Which gives us the test, and it has three questions. Is the underlying claim true? Does the frame make the reality more visible or less? And would it survive full disclosure — could you show the audience how and why you built it and still have it work? Call it the Glass Frame. A frame you would let people look through and see the joinery. A frame that makes a true thing land is stewardship of a listener's limited attention. A frame that makes a false thing land, or one that would break if the listener saw how it was made, is manipulation regardless of how good the cause is.

The Rich Man's Lamb, and Two Words in Hebrew

Scripture does not treat framing as a modern discovery or a moral problem. It treats it as the ordinary craft of anyone who wants a truth to actually arrive, and the clearest case in the Old Testament is a prophet who had one shot at a king.

Nathan has to confront David over Bathsheba and Uriah. Consider what a direct accusation would have produced. David has already had a man killed to cover this; a prophet walking in and naming the crime is a prophet who gets a defensive king, and possibly a dead prophet. So Nathan opens with a case at law. Two men in a city, one rich with many flocks, one poor with a single ewe lamb he had raised with his children — 'it would eat of his bread and drink of his cup and lie in his bosom, and was like a daughter to him.' A traveler comes to the rich man, and rather than take from his own flocks the rich man takes the poor man's lamb. David does not hear a parable. He hears a docket, and he responds as a judge: 'Then David's anger burned greatly against the man, and he said to Nathan, As the LORD lives, surely the man who has done this deserves to die.' Then the frame closes. 'Nathan then said to David, You are the man.'

The Hebrew is two words: attah ha-ish. That is the entire reversal. And notice exactly what the frame accomplished, because this is the ethical heart of the whole lesson. It did not deceive David. It delivered him to a true judgment he had already made, and then removed the distance he had been using to avoid applying it to himself. The disclosure was not a failure of the technique. The disclosure was the technique. A frame whose whole purpose is to be revealed is the opposite of a manipulation, and that is the model.

Jesus works the same way at industrial scale, and the Gospel writers say so plainly. Matthew 13:34, NASB 1995: 'All these things Jesus spoke to the crowds in parables, and He did not speak to them without a parable.' Mark adds the crucial qualifier at 4:33: 'With many such parables He was speaking the word to them, so far as they were able to hear it.' So far as they were able to hear it. That is a statement about calibrating delivery to a listener's capacity, and it is offered as a description of how the Son of God chose to communicate.

The Greek word is parabolē, from para, alongside, and ballō, to throw. Something thrown alongside the thing you are trying to understand so you can see it by comparison. That is exactly what a frame is. And it is worth noticing that the parables are not simplifications; several of them are harder than a direct statement would have been, and Jesus says elsewhere that they function to reveal and to conceal at once, which is a genuinely difficult theological matter I am not going to pretend is settled. What is not in dispute is the method. A kingdom truth about persistent grace is delivered as a father running down a road. A truth about who counts as a neighbor is delivered as a Samaritan on a highway with a wounded man. Told as propositions, those are agreed with and forgotten. Told as stories, they get inside people who came to disagree.

What the Research Establishes, and Where It Is Contested

Anchoring is the second workhorse, and it is older than most people realize. In their 1974 Science paper, Tversky and Kahneman spun a wheel of fortune in front of subjects, then asked them to estimate the percentage of African nations in the United Nations. Subjects who saw a high number on the wheel gave higher estimates than those who saw a low one, even though the wheel was visibly random. The mechanism is that a number entering the conversation first becomes the reference point everything after it is adjusted from, and the adjustment is systematically insufficient.

Replication status, honestly: the core anchoring effect is among the sturdier findings in the field. The Many Labs projects tested anchoring across large numbers of samples and it held up well. What has not held up as well are the more exotic demonstrations — anchoring on numbers that are transparently unrelated to the judgment, such as the last digits of a subject's Social Security number, has produced considerably weaker and less consistent results in large replication attempts than the original studies suggested. So: a first number shapes the range of the negotiation. A first number that is obviously arbitrary may not do much at all. Plan accordingly.

Loss aversion deserves a flag too, because leadership writing treats it as settled physics. The standard claim is that losses loom roughly twice as large as equivalent gains. David Gal and Derek Rucker published a substantial challenge in the Journal of Consumer Psychology arguing that the evidence for loss aversion as a general psychological principle is far weaker than its status suggests, and that many demonstrations are better explained by other mechanisms. The debate is live and serious researchers are on both sides. My advice is to use loss framing where the loss is real and specific to the listener, and to stop asserting a two-to-one ratio as though it were a constant.

One more, because intellectual honesty about your sources is part of what this academy is teaching. Kahneman's Thinking, Fast and Slow is on the reading list and belongs there. It also contains a chapter on social priming that did not survive the replication crisis, and Kahneman has acknowledged publicly that he placed too much confidence in that literature. That does not damage the framing or anchoring work, which came from his own experiments and has replicated. It is a useful reminder that a great book by a Nobel laureate can contain a section that is wrong, and that citing an author is not the same as verifying a claim.

Now the practical architecture, which is where Renvoise and Morin earn their place. Their message rules are sound and heavily field-tested: lead and close hard because the middle of any message is the least retained part; use contrast, because a position is far more legible against what it is not; make it concrete, because a street name and a number are remembered where an abstraction is not; and end with something specific enough to act on. Take those rules. Leave the mechanism. Their neuroscience framing — the appeal to an old or reptilian brain making the decisions — descends from MacLean's triune brain model, which comparative neuroanatomists have largely abandoned as an account of how brains evolved. The practical rules do not depend on the mechanism story being right, and you should not repeat the mechanism story just because the rules work.

The Ask, and a Street With a Number On It

Everything so far is preparation for the part most leaders leave out entirely: the ask. You can frame beautifully and still get nothing, because the audience walked away moved and with no idea what you wanted them to do. Four moves, in order. Anchor, contrast, concrete, ask.

The anchor is the first number or reference point you put in the room, and you should choose it deliberately rather than letting your opponent choose it. If a negotiation opens at their figure, every subsequent move is an adjustment from their figure. The ethical constraint is straightforward: your anchor has to be a number you can actually defend, because an indefensible anchor is a bluff and bluffs are withdrawals from the Character account we built two lessons ago. The contrast is what makes your position legible. Nobody can locate what you are for until they can see what you are against. Not a caricature of an opponent — the honest alternative, stated fairly, which is the steelman discipline applied to your own message. The concrete is where most leaders fail. Do not say we are addressing street homelessness. Say there were people sleeping on the sidewalk outside a specific station this winter, and here is the number, and here is what it will be by a specific month. A street and a number beat an adjective every time, and they also make you accountable, which is exactly why leaders prefer adjectives.

And then the ask, which has to name a person, an action, and a date. Not support this initiative. Vote yes on this at the meeting on the fourteenth. Not we need community engagement. Come to the school gym on Thursday at seven and bring one neighbor. The specific ask is the difference between a message and an outcome, and the reason to state it at the open and again at the close is the same reason to lead and close hard on everything else: the middle is where retention goes to die.

The New York cases are instructive in both directions. In 2013 Bill de Blasio ran a mayoral campaign built on a single borrowed frame, a tale of two cities, and it did what a good frame does — it made a mass of separate grievances about inequality, policing, and schools legible as one thing. He was in fourth place in the summer polling and won the Democratic primary outright in September with enough of the vote to avoid a runoff. That is the power of the instrument. It is also a warning, because the frame set an expectation that the administration then had to redeem with management, and the eventual gap between the two became the durable story about him. A frame is a promise about reality. If reality cannot pay it, you have taken out a loan against your credibility ledger.

The other direction is the New York Daily News front page of October 30, 1975: FORD TO CITY: DROP DEAD. President Ford had spoken to the National Press Club the day before, saying he would veto any bill providing a federal bailout to a city he described as having mismanaged itself. He never said those words. The headline was a framing act by a headline writer, it was devastatingly effective, and Ford himself said afterward that it hurt him in a state he then lost the following November. Run the Glass Frame test on it. Was the underlying claim true? Substantially — he had refused the aid. Did the frame make the reality more visible or less? That is where it fails. It made an economic and constitutional judgment about municipal finance read as a personal insult from a president to eight million people, which is a different thing than what happened. It was brilliant and it was not honest, and the distinction between those two words is the entire subject of this lesson. Take the craft. Keep the test. In the next lesson we ask what any of this is worth when you hold no position at all — influence without authority, which is the purest examination this module gives.