One Big Idea

The Speech Everyone Praised and Nobody Remembered

Lesson 6.1 left you with a bill. The Fog Tax is what you charge other people when you will not commit in language they can act on, and the test was whether a reasonable person a year from now could prove you wrong. If you did the Fog Audit honestly, you found at least one sentence you had been rewording for weeks in order not to say it. Good. That was a courage problem, and courage problems have the mercy of being obvious once you name them.

This lesson is about a failure that survives all of that. A leader can be entirely clear, entirely honest, entirely specific, and still communicate nothing, because he was clear about eleven separate things in ninety minutes and the listener walked out holding none of them. Every sentence was defensible. The whole was inert. Nobody in the room could have told you what mattered most, which means the leader made the decision about priority by accident and handed the result to people who had no basis for making it themselves.

You have seen the artifact. The strategy deck with six pillars, each with three initiatives, each with owners and metrics, all of it true. The state-of-the-organization address that covered everything the organization did that year and functioned as a receipt rather than a message. The candidate with a comprehensive platform on twenty-two issues who cannot tell you in a sentence what he is for. In each case somebody worked hard, and the hard work is what did the damage, because comprehensiveness feels like diligence and reads like noise.

Here is the diagnosis. Adding material is easy and it flatters everyone involved — the leader looks thorough, the contributors see their section survive, and nobody has to tell anybody no. Cutting material is hard, and it is hard in a specific way that this academy has been preparing you for since Module 1: to cut good material you must be willing to disappoint people whose work you are cutting, and you must be willing to be wrong in public about what mattered most. Both of those cost something the leader pays personally. So the default is to keep everything and let the audience sort it out, which is the Fog Tax again in a different currency.

The alternative is a discipline with a name. Every communication that has ever moved anyone had one governing claim underneath it, and everything else in the piece was either advancing that claim, proving it, or making it memorable. I call it the Governing Sentence. Finding it is the work. Everything after that is subordination.

The Governing Sentence

A Governing Sentence is not a topic. Affordability is a topic. Public safety is a topic. Discipleship is a topic. A topic is a container into which any amount of material can be poured without resistance, which is precisely why leaders reach for one. A Governing Sentence is a complete, arguable claim — something a reasonable person could hear and disagree with. New York does not have a housing shortage, it has a construction shortage, and every other affordability fight is downstream of that. You can argue with that. That is the point.

Test any candidate sentence three ways. First, is it arguable? If nobody could dissent from it, it is a platitude and it will produce agreement without movement. Second, is it complete? Subject, verb, claim — a fragment is a topic in disguise. Third, and this is the one that eliminates most drafts, does it commit you to something you would have to defend? If the sentence costs you nothing to say, it will cost your listener nothing to ignore.

Then the subordination rule, which is where the pain lives. Once the Governing Sentence is fixed, every remaining element must do one of three jobs: advance the claim, prove the claim, or make the claim memorable. Anything that does none of the three is cut. Not shortened, not moved to an appendix, not saved for the Q and A — cut. And notice what actually gets cut when you apply this rule honestly. It is almost never the weak material; you removed that on the first pass without effort. It is the excellent paragraph that belongs to a different argument, the story you are proud of, the data point you spent a week finding. That is why most leaders will not do this. It feels like destroying value, and in a narrow sense it is. You are spending good material to buy the one thing your listener cannot manufacture for himself: a sense of what matters most.

Distinguish it from a slogan, because the two look alike on a page and behave nothing alike in a room. A slogan is built to be repeated; a Governing Sentence is built to be reasoned from. The test is what happens when a subordinate hits a case you never anticipated. If your sentence is a slogan, he has nothing — slogans do not generate decisions, they only signal loyalty to them. If it is a real claim, he can derive an answer you would have given, without calling you. That is the executive function of the thing, and it is why the sentence has to be arguable: only a claim with content can settle a question. A leader who mistakes a slogan for a governing sentence will discover the difference when he is out of the office and forty judgment calls get made without him.

The classical tradition had a name for the discipline and put it second among the five canons of rhetoric — inventio, dispositio, elocutio, memoria, pronuntiatio — as the Rhetorica ad Herennium and Quintilian's Institutio Oratoria lay them out. Dispositio, arrangement, is routinely taught today as organization, meaning the order things go in. That is a shrunken reading. Arrangement in the classical sense is the decision about what is principal and what is subordinate, and order is merely its visible consequence. Aristotle pushes it to the bone in Book III of the Rhetoric, where he says that strictly speaking a speech has two necessary parts: you state your case, and you prove it. Everything else is contingent. A leader who genuinely believed that would delete half of most presentations he has ever given.

Before you go on, do this rather than nod at it. Take the thing you have to communicate next. Write its Governing Sentence in under twenty words, arguable and complete. Then write down the single best piece of material that does not serve it. That second item is the real assignment, and you already know what it is.

I Determined to Know Nothing

Scripture gives you the discipline at two scales, and the larger one is easy to miss because it runs across an entire public ministry rather than a single speech. Mark introduces the ministry with a summary: 'The time is fulfilled, and the kingdom of God is at hand; repent and believe in the gospel' (Mark 1:15, NASB 1995). From there the teaching returns to that center relentlessly. The parables of Matthew 13 are a sequence of angles on one subject — the kingdom of heaven is like a sower, a mustard seed, leaven, a treasure, a pearl, a net. Six images, one governing claim, each approach illuminating a facet the others could not.

The argument here leans on a word, so take it directly. The Greek is basileia, and while it can denote a realm, the dominant emphasis among New Testament scholars is on the dynamic sense — reign, kingly rule, the exercise of royal authority — rather than a territory on a map. That is a majority reading, not a unanimous one, and it matters for our purposes because a reign is a claim about who is in charge, which is arguable, while a realm is a location, which is a topic. Matthew's preference for kingdom of heaven where Mark and Luke have kingdom of God is usually explained as a reverential circumlocution for the divine name, though not every scholar accepts that account. Say what the evidence supports and no more.

Then the extreme case. 'For I determined to know nothing among you except Jesus Christ, and Him crucified' (1 Cor. 2:2, NASB 1995). Read that as a working writer and it is startling. Paul is addressing a congregation with factions, a lawsuit problem, sexual scandal, disorder at the Lord's table, confusion about marriage, meat offered to idols, and chaos in worship. He addresses every one of those in the letter. So the verse is not a claim that he restricted his content; that reading collapses against the epistle it sits inside. It is a claim about subordination. One thing governs, and every other subject in that letter is argued from it rather than alongside it.

Jesus performs the same move on the entire Torah, and it is worth watching because of what he does with the material he is not discarding. Asked which commandment is greatest, he names love of God first and love of neighbor second, and then makes the structural claim: 'On these two commandments depend the whole Law and the Prophets' (Matt. 22:40, NASB 1995). The verb rendered depend is krematai, from kremannymi, to hang or be suspended. That is an architectural image, not a dismissive one. Nothing in the Law and the Prophets is deleted; all of it is hung from two load-bearing commands. That is subordination stated as precisely as it can be stated, and it is the answer to every person who hears one big idea and assumes you mean throw the rest away.

He names the discipline explicitly a few chapters later: 'For I delivered to you as of first importance what I also received' (1 Cor. 15:3, NASB 1995). The Greek phrase is en protois, in the first things — which carries both an ordering sense and a ranking sense, and commentators have taken it either way. Either reading gives a leader the same instruction. Some material is first. Everything else is not, and pretending otherwise is not humility, it is abdication.

Now apply it where it costs you. Most leaders in ministry and business alike are functionally unwilling to say that one thing governs, because saying so means the people whose work is subordinate will know that it is. That is a real cost and I am not going to pretend it away. But consider the alternative you are choosing when you refuse: you are not sparing them a hierarchy, you are hiding one. The hierarchy exists either way. The only question is whether your people can see it and plan around it, or whether they have to guess at it and be quietly punished later for guessing wrong.

Four, Not Seven

There is a number in every communication seminar you have ever attended, and it is wrong. In 1956 George Miller published 'The Magical Number Seven, Plus or Minus Two' in Psychological Review — an essay he opened by joking that he had been 'persecuted by an integer.' The paper is a classic and it is more careful than its reputation. Miller's point was about chunks rather than items, and about the striking convergence of several unrelated capacity measures, not about a universal law that listeners can hold seven things.

The modern estimate is lower and better controlled. Nelson Cowan's 2001 review in Behavioral and Brain Sciences argued that when you prevent chunking and rehearsal — the two strategies that inflated earlier estimates — capacity for independent items comes out at roughly four, with individual variation. Label this accurately: Cowan's roughly-four is the mainstream current estimate, the exact figure is still discussed in the literature, and nobody serious is defending seven as a capacity for unrehearsed, unrelated items. The practical consequence is severe. A message built on five or six coequal points does not deliver five or six points at reduced strength. The points compete, and what survives is whichever one happened to be vivid, which means the audience made your priority decision for you on grounds you did not choose.

Two further findings sharpen the design work, and both deserve conservative statement. Slamecka and Graf reported in 1978 what became known as the generation effect: material a person generates himself is retained better than the same material read. It is a well-replicated laboratory phenomenon with known boundary conditions, and it is why a question that makes the listener produce your conclusion outperforms a slide that hands it to him. And serial position — the primacy and recency advantages documented in Murdock's classic free-recall work — is robust in list learning. Extending it to speeches and meetings is a reasonable inference rather than a demonstrated finding, and I will label it that way. What it suggests is unglamorous and probably true: whatever you put in the middle is the material most likely to disappear.

Put the three together and the design rule falls out. One governing claim, a small number of supports well under the capacity limit, the strongest material at the edges rather than buried, and at least one moment where the listener generates the conclusion instead of receiving it. That is not a template for charisma. It is an accommodation to the actual architecture of the mind you are speaking to, which is the most basic form of respect a communicator can offer.

And be honest about the direction this cuts. Every one of these findings tells you to say less. None of them tells you to say less that is true. The reduction is in coequal claims competing for one slot, not in evidence, not in nuance, and not in the hard parts of the argument — which is precisely the distinction that a leader looking for permission to be shallow will fail to make.

Message Discipline in a Five-Borough City

New York in 2013 ran the experiment in public. Bill de Blasio entered the Democratic mayoral primary as Public Advocate, running fourth, with less money and less name recognition than the front-runners. What he had was discipline. The campaign said a very small number of things over and over: end stop-and-frisk as it was then practiced, deliver universal pre-kindergarten, and pay for it with a tax on high earners. The television advertisement featuring his son Dante, which ran in August, is widely credited with the surge that followed. He won the primary outright, clearing the forty percent threshold that would have forced a runoff he could not have afforded, and won the general decisively.

The instructive contrast is Christine Quinn, the City Council Speaker, who led much of the early polling and finished third. Her message was competence and continuity — genuinely her record, and a defensible thing to run on. But competence is a topic, not a governing sentence, and it is especially weak as one when a specific vulnerability sits underneath it: as Speaker she had shepherded the 2008 extension of term limits that allowed a third Bloomberg term, and that fact organized her opponents' story about her more effectively than her own material organized hers. When you do not supply the governing sentence, somebody else supplies it for you.

Then the second half of the case, which Lesson 5.6 already made from the framing side and which looks different from here. Universal pre-K was in fact delivered, at a scale of tens of thousands of seats within two years — a real administrative achievement of exactly the sort message discipline makes possible, because an organization that knows the one thing can be pointed at it. The tax on high earners, however, required Albany's approval and never got it; the money came from the state budget by another route. Note the lesson precisely. Message discipline is what allowed the pre-K promise to be redeemed. It could not manufacture authority the city never had. Those are two different failures and a candidate must not confuse them.

In office, the same discipline scales. PlaNYC, released on Earth Day 2007, contained one hundred and twenty-seven initiatives spanning land, air, water, energy, transportation, and climate — a list that should have been unintelligible. It was not, because it hung from a single governing claim: roughly a million more people will live in this city by 2030, and the infrastructure has to be built for them now. That sentence made a hundred and twenty-seven items legible as one program, including a thirty percent greenhouse gas reduction target. It also could not overcome the authority problem either: congestion pricing, one of its signature elements, died in Albany in 2008. A governing sentence organizes what you control. It does not enlarge it.

There is one more scale to this, and it is the one candidates discover too late. A governing sentence is not adopted when it is published; it is adopted when other people can say it without a document in front of them. A mayor speaks through roughly forty agencies and a few hundred senior appointees, and every one of them will be asked, in some borough hall or radio interview you will never hear, what this administration is actually about. If you get four different answers, you do not have a message with an implementation problem. You have no message, and you have a communications shop generating paper. The practical test is unglamorous: pick three people at three levels, call them without warning, and ask.

So here is the executive standard, and I will state it flatly. A leader with five priorities has none, because the word priority means the thing that comes before other things and you cannot have five of those. What you have instead is a list that will be resolved by whoever is loudest in the room on a given Tuesday. Jim Collins put the discipline in an unusually usable form when he argued that a stop-doing list matters at least as much as a to-do list, and most executives have never written one. Write one. Then go back to your Governing Sentence and cut the best thing that does not serve it, and understand that the discomfort you feel doing that is the whole skill. Lesson 6.3 takes up what carries the governing sentence into a human being — story — along with the reason story is the most dangerous instrument in this entire module.