Two senior engineering leads have been at a company for years. Call them Dana and Marcus. When the director role opens up, Marcus gets it. Dana has been there longer. She knows why the billing service is held together the way it is, which customer escalations go to which person, and which of the original architects will still take her call.
Dana posts congratulations in Slack within the hour. Then, over the next few weeks, she gets a little quieter.
Marcus notices and decides to give it time. That feels like the kind thing to do. Bringing it up feels like rubbing it in, and he assumes the disappointment will cool on its own.
Dana does her job and answers questions when asked. What changes is everything around the edges.
Her answers in one-on-ones get shorter. She stops showing up to the architecture review she used to run in practice, if not in title. When someone asks her how a design decision should go, she says, “That’s Marcus’s call now.” The history she used to hand out freely, like why the queue was built that way or what happened the last time someone tried to split the monolith, now comes out only when someone asks exactly the right question.
You could not write any of it up as a complaint. All of it costs Marcus, because what Dana knows is what he needs most in his first year: the history of the system, the people who built it, and the trust of the engineers who have followed her for years.
People in Dana’s position rarely quit loudly. They pull back a little at a time and wait for the new leader to miss. When Marcus makes his first bad call, and he will, Dana will not say “I told you so.” She will simply not have warned him.
Pulling back is Dana’s choice, even if the disappointment behind it was not, and it costs her too: the trust she spent years earning starts to look like a grudge. The first move still belongs to Marcus, because he holds the power.
From Marcus’s side, giving it time feels like respect. Dana sees a new boss who got the job she wanted and has not said one word about it in two months. Every week of silence confirms the story she is already telling herself: he does not need me, and he does not care what this cost me.
Linda Hill followed new managers through their first year for her research in Becoming a Manager. Most expected the hard part to be the work itself. The harder lessons turned out to be about people: a new title brings much less authority than it seems to, and cooperation has to be earned again, especially from former peers. Dana is the hardest version of that lesson Marcus will get.
The best-documented version of this story is more than 160 years old.
In 1860, William Seward was the best-known Republican in the country and the favorite for his party’s presidential nomination. He lost it to Abraham Lincoln on the third ballot at the Chicago convention. Seward was older, far more experienced, and widely considered the more qualified man. He knew it, and so did everyone else.
Lincoln made him Secretary of State, the most visible job in the cabinet. He did the same with other men who had run against him, which is the story Doris Kearns Goodwin tells in Team of Rivals. He did not wait for their disappointment to cool somewhere far away. He brought them into the room and gave them real work.
That did not settle it. Seward assumed he would be the real power in the administration. On April 1, 1861, a month into the job, he sent Lincoln a memo titled “Some Thoughts for the President’s Consideration.” It said the administration had no policy, at home or abroad, and offered, in so many words, to take charge of one.
Lincoln wrote a reply the same day. It answered Seward’s points and made clear that if a policy had to be run, the president would run it. Historians believe Lincoln never sent the letter. He seems to have settled it with Seward in person, and the memo stayed private. He did not embarrass Seward in front of the cabinet, he did not demote him, and he kept handing him important work.
Within a few months, Seward wrote to his wife that “the President is the best of us.” He served as Secretary of State for the rest of Lincoln’s life and became one of his closest allies.
It did not work on everyone. Salmon Chase, another rival Lincoln brought into the cabinet as Treasury Secretary, got the same treatment and kept angling for the presidency anyway. He left the cabinet in 1864, and Lincoln then named him Chief Justice of the Supreme Court.
Lincoln made two moves. The first was private and early: a clear answer to the question of who leads, given in a way that let Seward keep his dignity. The second was public and lasting, because the whole country could see Seward doing important work with Lincoln’s trust behind him. Lincoln controlled the offer and whether he kept it, and Seward and Chase each decided what to do with it.
Go back to Marcus, and say he skips the two months of waiting. In his first week he asks Dana to get coffee away from the office and says something like this:
“You wanted this job, and you had every reason to. I’m not going to pretend that isn’t awkward. I need your help with the billing migration. You know why it’s built the way it is, and I don’t. I want you to own the plan and present it to the VP next month.”
That is the whole conversation, and it is short on purpose. Marcus should have it because Dana is owed it, since she built much of what he now runs. A conversation held only to neutralize a threat gets felt as exactly that.
Marcus starts by naming the thing. The disappointment is in the room whether or not anyone says it, and saying it out loud tells Dana that he noticed what this cost her.
The request is for help on a real problem, one where her history matters. A made-up project or a vague “I’d love your input sometime” gets read for exactly what it is.
He leaves out two things. He does not apologize for getting the job, and he does not offer some fuzzy co-leadership. Marcus is the director. Pretending otherwise only delays the fight about who decides, the same fight Lincoln had to settle with Seward. Being plain about it from the start, and kind about it, is cheaper.
The last part matters most. Dana presenting the plan to the VP puts her name in front of the people who decide careers, with Marcus’s backing.
Marcus controls whether he makes the offer, how he makes it, and whether he keeps his word. Dana’s answer is hers. Some people in her spot leave anyway, and the offer was still the right one to make.
The Greek word for the desire Dana is carrying is philotimia, love of honor, the wish to be recognized for doing something worthwhile. Aristotle noticed in Book IV of the Nicomachean Ethics that people use the word both ways. They praise someone as ambitious when the person wants the right honors for the right reasons, and blame someone as ambitious when the person wants too much or wants it from the wrong place. Dana’s desire for honor did not vanish when the promotion went to Marcus. Left with no outlet, it turns into keeping score. Real work with her name on it gives it somewhere honorable to go.
The honor has to be earned, though. Aristotle says the great-souled person takes little pleasure in honor from casual people or for trivial things, and someone who cares about honor will spot a consolation prize at once. The work has to matter to the business, and Dana has to be the right person for it.
The whole team knows Dana wanted the job, and they are watching what Marcus does about it, because it shows them what to expect from him. How a new leader treats the person who lost tells everyone else how they will be treated when they disagree with him or miss out on something they wanted.
A freeze-out teaches the team that losing to Marcus is permanent. Bringing Dana in and handing her real work teaches them that people can lose a round under this leader and still matter.
Michael Watkins makes the case in The First 90 Days that the impressions a new leader creates early are hard to undo later. The first conversation with the passed-over rival is one of the earliest and most watched impressions you will make. If you skip it, the team still forms an impression, just from your silence.
Have the conversation in week one. Do it before the silence hardens into a story. A month in, you are trying to reverse a conclusion the other person has already reached.
Name it in one sentence. “You wanted this job, and I know it.” You do not need a speech, and you should not apologize for getting the role.
Bring a real problem. Pick work where their history and relationships matter, the kind you would struggle to do without them. Then ask for their help with it.
Settle who decides, in private and without heat. If they push for the wheel the way Seward did, answer the way Lincoln did: privately, clearly, and without taking their work away.
Make their work visible. Put their name on the plan, have them present it to leadership, and give credit in public where the team can hear it.
Check which side of the table you are on. If you are the one who got passed over, your move is different, and I wrote about it in Sour Grapes Cost More Than the Grapes. If you are the leader handing the role to someone else, read Your Successor Can’t Lead While You’re Still in the Room.
The challenge for this week: if there is a person who wanted your job and you have not had the conversation, put thirty minutes on the calendar for coffee before Friday. Bring a real problem.
The person who wanted your job holds things you need: history, relationships, and the trust of people who have followed them for years. They will rarely fight you openly. They will pull back and wait, and waiting for their disappointment to cool reads to them as not caring.
Lincoln had every reason to keep Seward at a distance. He gave him the most visible job he had, settled the question of who leads in private when Seward pushed, and kept handing him important work. Within months Seward was calling him the best of them. Chase showed that the same treatment does not win everyone over. Make the offer anyway, because the person who lost is owed it. Have the conversation in your first week, bring a real problem, and give them something worth their name.
If you want to lead in a way that turns rivals into allies, MasteryLab is built for that.
Talk to them privately in your first week, before silence turns into a story. Say plainly that you know they wanted the role. Ask for their help on a real problem where their history matters, and give them work that leadership and the team can see, with their name on it. Be clear that you make the final calls, without apologizing for getting the job. Waiting for the disappointment to cool usually reads as indifference.
Keep it short and specific. Something like: “You wanted this job, and you had every reason to. I need your help with this project, because you know things about it that I don’t. I want you to own the plan and present it.” Name the disappointment, ask for real help, and follow through with visible work. Avoid apologizing for the promotion or offering a vague co-leadership arrangement you cannot honor.
Seward lost the 1860 Republican nomination to Lincoln and became his Secretary of State. In April 1861 he sent Lincoln a memo saying the administration had no policy and offering to take charge. Lincoln wrote a reply making clear that the president would run policy, but historians believe he never sent it and settled the matter with Seward in private. He kept Seward in his job and kept giving him important work. Within months Seward wrote to his wife that “the President is the best of us.”
philotimia is the Greek word for love of honor, the desire to be recognized for doing something worthwhile. Aristotle noted that people use the word both as praise and as blame, depending on whether someone wants the right honors from the right sources or wants too much. For a new leader, the point is practical: a person passed over for a promotion still wants honor, and that desire needs honorable work to go into.
]]>Then March comes and the number gets hit anyway. The rewrite ships with half its features behind a flag nobody plans to turn on. The incident count drops because three outages got filed as “degradations.” Somebody hit the number, and nobody can explain how in a sentence they would want their boss to read.
I have been in that planning meeting more times than I can count, on both sides of the slide. The pattern is reliable enough that I now treat it as a law of organizations. Most people do not cheat, and somebody usually does. If no one can hit the number honestly, someone will hit it anyway.
Put two goals side by side. Both are big, and both get applause at the all-hands. From the back of the room you cannot tell them apart.
The difference is whether someone close to the work can describe a route to the number, out loud, in a way they would be comfortable explaining to the person who set it. The route can be uncertain, but it has to exist. One goal has an honest path and the other has none, and they produce opposite teams.
Leaders reach for the second kind for understandable reasons. They feel behind, and a big number is the fastest thing to produce when you feel behind. It fits on a slide, and it sounds like a plan without requiring one. There is also research that seems to back it up. Edwin Locke and Gary Latham spent decades showing that hard, specific goals beat “do your best.” Plenty of leaders take that to mean harder is always better. The same research found the effect depends on people being committed to the goal and able to reach it. Leave those two conditions out and the finding means something different.
A goal with an honest path is hard and has a route someone can name. “If we finish the queue migration in January and take two engineers off feature work for the quarter, we can cut pager volume in half.” The team might miss. The migration might take until March. But the work goes into the route, and if they miss, they know which part of the route broke.
That is the first benefit: a miss carries information. The second is who stays. Strong engineers like a hard problem with a real chance of solving it. That is the job they signed up for, and a goal like this keeps them in the room.
The cost lands on the leader. Someone has to do the planning work, and someone has to write down what stops so the route can happen. The two engineers taking on pager volume are not shipping features this quarter, and the leader has to say that out loud to the people who wanted the features. Sometimes the honest number is smaller than the one that would have looked good on the slide.
John Doerr’s account of OKRs in Measure What Matters gets one detail right that most companies copy wrong. At Google, as he describes it, hitting about seventy percent of a stretch objective counts as success. That rule makes a miss safe to report. A team that knows seventy percent is a good outcome has no reason to fake the last thirty.
A goal with no honest path runs the other way, and the cleanest case is a bank.
Wells Fargo pushed its branch employees to sell about eight products to every household. The target came from the top and ran down to people working the counter for an hourly wage. In 2016 regulators fined the bank $185 million after finding that employees had opened accounts customers never asked for. The bank said it had fired about 5,300 employees over several years for it. Later reviews raised the count of possibly unauthorized accounts into the millions.
It would be comfortable to blame 5,300 bad people, and it would miss the number that sat above all of them. A study published in 2004 by Maurice Schweitzer, Lisa Ordóñez, and Bambi Douma found that people working toward a specific goal they had not reached were more likely to overstate their performance than people told to do their best, and the effect was strongest when they were close to the goal. Decent people, a target with no honest route, and a short distance left to go add up to people filling the gap with something other than the work.
Engineering teams have their own versions. A test coverage mandate produces tests that run the code and assert nothing, and a velocity target gets you story points that inflate a little every sprint until a three is an eight. Tell a team to halve incidents and you will get a long debate about what counts as an incident. The newest version is “half our code will be written by AI agents by June,” set by a leader who has not given the agents access to the repo. Engineers paste a snippet from a chat window into a commit and tag it AI-assisted. The dashboard goes up, and nothing about how the team works has changed.
Charles Goodhart noticed this in economic policy in the 1970s, and the anthropologist Marilyn Strathern later gave it the phrasing everyone quotes: when a measure becomes a target, it ceases to be a good measure. The number keeps moving after it has stopped telling you anything.
Every target is supposed to point at something good. Pager volume points at engineers who sleep through the night and customers who can use the product. Coverage points at code that does what it claims. Gaming the number cuts it loose from the thing it was pointing at, which gives a leader a second test: if the number moved and the thing it points at did not, the team achieved nothing.
Often, before anyone games anything, the strongest people leave. Some stay and argue with the number, which is the best thing that can happen to it. They can do the math faster than anyone, they have options, and they do not want to spend a year on a treadmill that cannot move. The people who stay have three choices. They can miss and take the blame, burn out trying, or close the gap some other way. A leader who sets a target with no route has told them, without saying it, that the honest way is not expected.
The cost to the leader is trust, and it is spent before the quarter even starts. The team learns that the leader’s numbers are not real. Next quarter’s number gets discounted on arrival, including the honest one.
Aristotle draws the line that separates these two goals, in Book III of the Nicomachean Ethics.
He separates wish from choice. Wish, boulesis, can be for anything, including things that are impossible. You can wish you would never die. You can wish your team would win the championship, which is not up to you. Choice, prohairesis, is narrower. Aristotle says nobody chooses the impossible, and anyone who claimed to would be thought foolish. Choice is only about things in our power, and it comes after deliberation, which is working out the means to an end you already have.
Put that next to the planning slide. A target with an honest path is something a team can deliberate about: which work first, who moves, what stops. A target with no honest path cannot be deliberated about, because there are no means to work out. The leader has written down a wish and handed it to the team as if it were a choice.
Long odds do not turn a goal into a wish. A goal can be chosen if every step of the route is something the team can do or can be given, even when the chance of finishing is low. It becomes a wish when some step depends on something nobody controls, or on headcount and time nobody is going to provide.
The team still has to choose something, because they are still accountable. So they deliberate about the only thing left in their power, which is how the number gets reported.
None of this lets the person who cheats off the hook. The employee who opened an account in a customer’s name did something wrong, and so does the engineer who writes a test that asserts nothing. They chose it. The leader who built the treadmill made a different choice, and it is the one that made the cheat likely. Both are real, and the leader’s is the one that repeats across the whole team.
The person handed the number has a duty before it comes to that, at any level. On the first day, say in writing that you do not see a route and what a route would need. At the end of the quarter, report the honest number. A new hire and a staff engineer owe the same two sentences, and so does the director who got the target from a VP.
Most of this is fixable before the slide goes up.
Ask for the route before you announce the number. Find the two people closest to the work and ask them how they would hit it. Listen for an answer they would be comfortable giving you in writing. If nobody can describe one, or the only route needs headcount, time, or cuts you are not willing to give, you are holding a wish. Shrink the number or give them what the route needs.
Say what stops. A real stretch goal comes with a list of things the team will not do this quarter. If the list is empty, the plan is to do everything at once, and you already know how that goes.
Ask how a decent person would cheat. Before you commit, ask the team: if someone good were desperate to hit this by the deadline, how would they game it? They will know in about thirty seconds. Then pair the target with the number that would catch the cheat. Track coverage next to bugs that escape to production, and incident count next to outages customers report.
Make an honest miss safe and a perfect score a question. Thank the team that reports seventy percent with a clear account of what broke. When a hard number comes in at exactly a hundred, ask how, with curiosity. Most of the time there is a good answer, and the times there is not are the ones you need to know about.
Put the agents on the work. If your goal is about AI, the honest path runs through access. Give the agents the repo, the tickets, and the test suite. Pick three workflows where the model does the work end to end, and measure the hours that come back. A team with agents in the repo will usually beat a “half the code by June” target without anyone tagging a commit. A team with chat windows and a mandate will hit it on the dashboard and nowhere else.
I wrote last month about celebrating decimals, the habit of applauding partial progress as if it were the finish. This is the same problem seen from the other end. A goal the team cannot believe produces numbers nobody can trust, and then the measuring stops meaning anything.
A big goal with a route pulls a team forward even when they miss, while one with no route pushes the best people out and teaches the rest to make the number some other way. Both look ambitious on the slide, and only one is a plan.
Aristotle’s line is the test. You can wish for anything, but you can only choose what is in your power, and you can only plan your way to something that has a route. A leader controls two things here, the design of the target and what happens when someone reports a miss, and those two are enough. Before the next number goes up, ask the people closest to the work how they would hit it. If they cannot tell you, someone will hit it anyway, and you will not like how.
Setting goals a team can believe is a leadership skill, and it can be practiced. That is what MasteryLab is for.
]]>This is the first law in a while where I agree with more than half of what Greene says. I agree with the part about strings and quality, and I reject the part about people. Greene wants you to owe nobody, and a person who pulls that off has closed most of the doors anyone could use to help them.
Law 40: “Despise the Free Lunch.”
Greene’s judgment runs like this. Anything offered for free is dangerous, because it usually carries either a trick or a hidden obligation. “What has worth is worth paying for.” If you pay your own way, you stay clear of gratitude, guilt, and deceit. It is often wise to pay the full price, because excellence has no shortcuts. And be lavish with your money and keep it moving, because generosity is a sign of power and a magnet for it.
He then sorts the people who handle money badly into types. The greedy fish wants money so badly it swallows the hook. The bargain demon spends hours hunting a discount and pays for it in time and shoddy goods. The sadist uses money to keep other people waiting and dependent. The indiscriminate giver hands out gifts to be loved by everyone and ends up valued by no one.
Keep your eye on the reason in the middle of the judgment. Greene puts gratitude on the list of things to escape, right next to guilt and deceit.
Most of the practical advice is right, and anyone who has bought software for a company has learned it the hard way.
The vendor’s free pilot is free for ninety days. On day ninety-one your team has built three workflows on it, nobody wants to rebuild them, and the renewal quote arrives. The free architecture assessment from a consulting firm ends with a recommendation that needs the consulting firm. The recruiter who takes your best engineer to dinner is paying for something. Free things at work usually have someone’s quarterly number attached, and Greene is right to tell you to find the number before you eat.
He is right about full price too. I have watched teams save money on the cheap contractor, the cheaper laptop, and the free tier, then pay for it three times over in rework. The clearest case right now is AI. A team runs its trial on the free tier of a model, gets a weaker model with tight limits, decides AI is overrated, and goes back to doing everything by hand. That is the bargain demon in its current form. Pay for the best model you can get and put it on the real work on day one.
Seneca, who wrote seven books on gifts and favors, agrees with more of this than you might expect. In On Benefits he says to be more careful choosing whom you owe a favor than whom you owe money. A loan gets paid off and ends. A favor stays with you, and you do not want to carry one for someone you do not respect.
So the strings are real, and full price is often the right price. The trouble starts with what Greene builds on those two facts.
Greene does not stop at “check the strings.” He tells you to despise the free lunch so that you stay clear of gratitude, and that reason is where the law goes wrong.
Start with the act itself, before any bill comes due. A person who refuses to acknowledge help is giving a false account of themselves, as if they had built everything they have alone. That account is wrong on a day when it costs nothing, and the costs below come on top of it.
The first cost is that you become hard to help. A person who treats every offer as a trick turns down the free help along with the free trap. I have seen a senior engineer wave off a junior’s offer to pair on a bug, because taking the help felt like owing. I have seen new managers decline coffee with a peer who had done the job for five years, because they did not want to be in anyone’s debt in their first month. The help they refused was worth more than anything a vendor was selling them. Suspicion cannot tell the vendor from the colleague, so it refuses both.
Second, the ledger takes over your friendships. If you pay for everything so that you never owe, every relationship becomes an account that has to be kept at zero. Aristotle saw where that ends. In Nicomachean Ethics VIII he says complaints and quarrels come up mostly in friendships of utility, the ones where each side tracks what they gave and what they got. Friends who care about each other for their own sake do not fight over the balance, because nobody is keeping one. Greene’s rule turns every friendship into the utility kind and then insists it stay square. The colleague who returns every favor within the week, so nothing is ever left open, has very clean accounts and very few people who would call them at midnight.
Third, look at what Greene does with generosity. “Be lavish with your money” sounds like the opposite of a ledger, until you read his reason: generosity is a magnet for power. A gift given for that reason is a purchase, and the person receiving it is the thing being bought. People can tell. Most of us have been taken to a dinner that turned out to have an ask at the end, and we remember how the dinner tasted once the ask arrived. Generosity aimed at power fills a room with people who came for the food.
Fourth, the rule changes the person who follows it. If you believe nothing is ever given freely, you stop giving anything freely yourself, because in your picture of the world a free thing is a hook. Your own favors start to carry strings, since that is what you believe favors are. Greene warns you about the person handing out lunches with hooks in them, and his advice is a reliable way to become that person.
The fifth cost is the oldest, and Aristotle drew it himself without meaning to. His great-souled man, the person of megalopsychia in Nicomachean Ethics IV.3, likes to do favors and is ashamed to receive them, because giving marks the superior and receiving marks the inferior. He remembers the favors he did and forgets the ones done for him. Aristotle presents this as part of the portrait of a great man, and readers have been uneasy about that passage for a very long time. Greene took the most uncomfortable trait in the portrait and made it a law. The pride that cannot stand to owe tends to travel with the pride that cannot stand to be corrected or taught, and together they cost most in exactly the places where you need the help.
Software is the clearest counterexample to the whole law. This site runs on Jekyll. The code I write every day runs on Linux, Ruby, Git, and libraries written by thousands of people I will never meet, who gave their work away. That is the largest free lunch in the history of the trade. Some of it has strings, like a license or a company that hopes you will move to its cloud, and you should read the license. Most of it was given because people who are good at something like to give it away, and the teams I have seen refuse it on principle spent years paying for worse software.
Aristotle has the word Greene needed. It is eleutheriotes, usually translated as liberality, the virtue of being free with money in the right way. In Nicomachean Ethics IV.1 the liberal person gives to the right people, in the right amounts, at the right time, and does it with pleasure, or at least without pain. On one side sits the prodigal, who gives too much and to the wrong people, and Greene’s indiscriminate giver sits close to that vice. On the other side sits the stingy person, who gives too little and takes too much from the wrong sources.
Two parts of Aristotle’s account correct Greene. First, liberality covers taking as well as giving. The liberal person takes from the right sources and refuses the wrong ones, which is everything sound in Law 40: check where the lunch came from. His rule is to know whose lunch it is, and that leaves plenty of lunches worth eating. Second, the liberal person gives because giving well is good in itself, and Aristotle says liberality depends on what a person has. Someone who gives a little out of a little can be more liberal than a rich man who gives a lot. Greene’s lavish spender, giving to look powerful, would not pass Aristotle’s test at all, because the gift is aimed back at the giver.
That gives a working line between a gift and a hook. A gift is given for your sake and does not depend on a later yes, while a hook is a sale whose price shows up later. The colleague who offers to pair on your bug wants the bug fixed and you better at fixing it, and nothing about the offer changes if you never return the favor. The vendor’s pilot changes completely the day you say no to the renewal. Ask what happens if you say no to whatever comes next, and you will usually know which one you are holding.
Seneca covers the receiving end, which Greene leaves out entirely. (Law 13 was about the asking end, and why reminding people of past favors backfires.) Early in On Benefits he spends time on the three Graces, the old image of a favor passed from hand to hand and coming back around, the cycle the Greeks called charis. His point is that receiving is half of the virtue. He says the person who accepts a favor gladly has already paid the first installment on it. A gift given in order to get a return, in his view, was an investment from the start. Seneca keeps Greene’s caution about whom you owe, and then spends most of the work teaching you how to owe well.
Then there is autarkeia, self-sufficiency, which both Aristotle and the Stoics prized and which Law 40 imitates. The difference is that self-sufficiency means you do not need the gift in order to be whole. That is exactly what lets you accept one without being bought. A person who has to refuse every gift to stay free is less self-sufficient than one who can take a gift, say thank you, and stay the same person.
Socrates is the case that settles it. He took no fees. Xenophon reports that he thought people who charged for their company made themselves slaves to whoever paid them. The best conversation in Athens was a free lunch, and it was free so that the man giving it would stay free.
Before the practices, here is the end they serve. You want a team where help and good work move to wherever they are needed, between people who trust each other enough to let them move, and you want to be one of the hands they pass through. That takes giving and receiving without either one costing you your judgment: knowing who you owe, choosing them well, and being glad to owe them.
Find the string before you eat. For any free thing from someone with a number to hit, write down what they want in one line. The vendor wants the renewal. The recruiter wants your engineer. Then decide whether you want the lunch at that price. For a free pilot, write the exit plan before you sign up, while leaving is still cheap. You are judging the terms you can see, and nobody can read a giver’s heart. Deciding on the visible terms is enough, even if you guess wrong now and then.
Pay for what your work runs on. The model, the editor, the books, the training. If your craft runs on it eight hours a day, pay full price for the best version. Running your team on the free tier of the thing they use most is the bargain demon’s mistake. None of this conflicts with taking gifts gladly, because you pay the people who are selling and accept from the people who are giving.
Say thank you by name, the same week. That is Seneca’s first installment. “Your review caught the race condition in the retry loop, and it would have paged someone at 2am. Thank you.” Say it specifically, and in public where it fits. The thank-you leaves the debt open and tells the person you know you owe them and are glad to.
Let people help you. The next time a peer or a junior offers to pair, review, or walk you through something, say yes. Notice the flinch, the small feeling that now you owe them. Owing a good colleague is a fine thing to carry.
Give with no ask, and check it. Before you give something at work, like an introduction, a careful review, or an hour on someone’s problem, ask whether you would still give it if they never found out it came from you. If the answer is no, you are buying something. Be honest with yourself about what.
None of this depends on rank. The intern who turns down a senior engineer’s offer to walk through the codebase, because taking it feels like weakness, is following Law 40. So is the CEO who will not let a board member make an introduction because she does not want to owe him. The manager who picks up every team lunch so everyone remembers who paid is following the other half of it. And the junior who says “I have nothing to give back, so I would rather not take it” has the arithmetic wrong. Gratitude never depended on the means to repay. You owe the thanks and the use of what you were given, and a junior can pay both. Receiving well and giving cleanly are the same duty at every level.
One more, from my own desk. When a colleague gives me an hour of their time on a code review, that hour is a gift, and I want it spent on the judgment calls. So before I ask, Claude goes over the pull request first and catches the typos, the missing test, and the obvious bug. The person reviewing gets a change that is already clean and spends their hour on the questions only they can answer. Start with the model, and use it to make the human gifts you receive count for more.
Aristotle gives the whole frame in Nicomachean Ethics IV. Chapter 1 is eleutheriotes, a virtue that governs taking as well as giving and is measured against what you have. Chapter 3 is the great-souled man, ashamed to receive, and it is worth reading as a warning as much as a model. Book VIII adds that the friendships full of complaints are the ones where both sides keep score.
Seneca’s On Benefits is the longest ancient work on gifts, and it is mostly about the person on the receiving end. Choose whom you owe with care, he says, and then owe them gladly. Gratitude in his account is part of what makes a person free, because the grateful person is not trying to get out from under anyone.
Xenophon’s Memorabilia gives the portrait of Socrates refusing fees. Read it next to Greene. Both men wanted to stay free of other people’s claims. Greene gets there by paying for everything, and Socrates got there by giving his best work away.
Lewis Hyde is not ancient, but The Gift belongs on this list. His argument is that a gift keeps its value only while it keeps moving. Hold onto it, or turn it into a sale, and something goes out of it. Anyone who has watched an open source project thrive because people passed their fixes back has seen that argument work.
When did you last accept help at work and not pay it back within the week? If you cannot remember, ask yourself why owing someone feels so uncomfortable.
Who do you owe right now? Are you glad to owe them, or are you counting the days until you can square it?
Look at the last three things you gave at work. Which of them had an ask waiting behind them?
What do you use every day that you are still running on the free tier, and what is that costing you in work you redo by hand?
And the one that decides: if everyone you work with forgave every favor you owe them tomorrow, would you feel lighter, or would you feel like you had lost something? If the answer is lighter, Law 40 has been running you for a while.
Greene is right that free things from people with a number to hit come with hooks, and right that good work is worth paying full price for. He is wrong that freedom means owing nobody. A person who owes nobody has turned down most of the help, teaching, and friendship that would have made them better, and has usually started giving with hooks of their own.
Aristotle’s liberal person takes from the right sources and gives to the right people, gladly, and Seneca’s grateful person owes well and stays free. Socrates went further and gave away the best thing he had so that nobody could own him. Check the strings, pay for your tools, and then let the people worth owing give you something.
Law 41 is next, and it is about stepping into a great man’s shoes.
Learning to take help as well as give it is part of the craft. That is what MasteryLab is for.
]]>Ask what she did in the hard one-on-ones, the ones where somebody was underperforming or about to quit, and you get a long pause. “I don’t know. She was just good at it.”
Now ask about the worst boss. The answer comes fast and in high resolution: the standup where he read their bug out loud to the whole team, the exact sentence he used, and who laughed. They can also tell you the rule they made that week, which they still follow: I will never correct anyone in front of other people.
The best boss left them a feeling, while the worst boss left them a scene they can replay and a rule they still live by. If you measure teaching by what the student can still say years later, the worst boss wins by a mile, and that should bother anyone who thinks leadership is learned by finding a great leader and copying them.
Most leadership development runs on one belief. Find good leaders, put people near them, and good leadership rubs off. Mentorship programs assume it. Leadership books are full of admirable CEOs for the same reason. The advice to young managers is nearly always some version of “find someone great and watch what they do.”
Part of this is true. People learn by imitating. Aristotle wrote in the Poetics that humans are the most imitative of creatures and that we learn our first lessons through imitation. The Greek word is mimesis, imitation, and it is how most of us learned to talk, walk, and run a meeting.
The problem is what imitation needs in order to work. It needs something you can see. And the best leaders, at their best, give you very little to see.
Think about what a skilled leader does in a hard moment. A senior engineer ships a migration that takes down billing for an hour. A good manager has the conversation the same afternoon. The engineer walks out understanding exactly what went wrong, what has to change, and that their job is safe if it does change. They leave corrected, and they still want to come back tomorrow.
From the outside, that looks like two people talking for twenty minutes. The craft is in the choices nobody saw: what she said first, what she left out, how long she let the silence run, the question she asked instead of the accusation she could have made, the decision to do it today instead of Friday. If it worked, it looked like nothing. A conversation that lands well looks like an ordinary conversation, and you cannot copy an ordinary conversation.
Michael Polanyi, a chemist who became a philosopher, put the general problem in one line: “we can know more than we can tell.” Skilled people run on knowledge they cannot fully put into words. Ask your best former boss how she handled those one-on-ones and she may not be able to tell you either, because a lot of the method lived below the level where she would think to describe it.
So the good boss’s lesson is real, and it is mostly invisible, which leaves imitation nothing to work with. You copy the parts you can see, the calm voice and the open door, and miss the decisions underneath that made the calm voice work.
The bad boss has the opposite problem: the damage you felt from them is easy to see. You remember the date of the standup where they humiliated you, and where you were sitting in the all-hands when the reorg they had kept to themselves finally landed. If they promised the promotion “next cycle” three cycles in a row, you probably still feel it every review season.
Psychologists have a well-studied version of this. A widely cited 2001 review by Roy Baumeister and colleagues, titled “Bad Is Stronger Than Good,” found that across a long list of areas, bad events have bigger and longer-lasting effects on people than good events of the same size. A bad boss’s worst afternoon gets stored in detail, while a good boss’s skill mostly shows up as problems that never happened, and nobody keeps a memory of a crisis that did not occur.
Side by side, the two teachers look like this:
| Your best boss | Your worst boss | |
|---|---|---|
| What you remember | A feeling: things were calm | A date, a room, and a sentence |
| What you can name | Very little of the method | Exactly what went wrong |
| What it teaches on its own | Nothing you can copy without asking | What to avoid, with nothing about what to do instead |
This is an old observation. Plutarch records that Cato the Elder used to say that wise men gain more from fools than fools gain from wise men, because the wise avoid the mistakes they watch fools make, while fools never copy what the wise get right. Confucius said something close in the Analects: walk with any two people and you can learn from both, copying the good points of one and correcting in yourself the bad points of the other.
Notice what both men are saying. The bad example teaches by showing you a mistake with its consequences attached, out in the open where you can study it. The good example only teaches if you already know how to see what is good about it, and most of us do not, especially early.
I see the same thing at my desk. When Claude writes a clean pull request against a spec I wrote, I learn very little. It worked, so I merge it and move on. When it writes a broken one, the diff shows me exactly which line of my spec was thin or which file I forgot to point it at. The failure has an address, so I know where to go, and a clean run never tells me which of my choices made it work.
That is why your worst boss taught you so much: the damage had a location. The pain pointed at one specific behavior, and you could see the cost of it in your own week. Pain teaches when you can say exactly what caused it, a point I made about which failures are worth anything. A bad boss is, among other things, a very precise source of that kind of pain.
If that were the whole story, a bad boss would be a free education and you could take the lessons and move on. The trouble is that a lesson learned only from pain tells you what to avoid and leaves out what to do. When people only know what to avoid, they build the exact opposite of the person who hurt them, which is often a different failure.
Go back to the new manager who will never correct anyone in public. That rule is fine. Watch what it turns into over a year. First they stop correcting people in meetings. Then they stop correcting people in Slack threads, because others can see those. Then the private one-on-one starts to feel risky too, because what if it lands the way that standup landed on them? A year later they are the manager who says “solid start” about work they know is broken. I wrote about what soft feedback costs the person receiving it. This is one of the most common roads into it: a person running away from one bad boss instead of walking toward a good practice.
The pattern repeats with the other flaws. The person whose boss hoarded information overcorrects by forwarding every update to everyone, and the team drowns. The person whose boss broke promises stops committing to any date at all, and nobody can plan around them. In each case the new behavior is the reverse of an old injury. Nobody chooses the injury, but the manager chooses the rule again every time they apply it, and after enough years it hardens into a management style. By a flinch I mean a rule you follow because of how something once felt, as opposed to a skill, which is a practice you can explain, test against the person in front of you, and change when it stops working.
The opposite of a flaw is usually another flaw, and the skill sits somewhere specific between the two. Aristotle built much of his ethics on this point in Book II of the Nicomachean Ethics: courage sits between cowardice and recklessness, and generosity between stinginess and waste. He was careful to add that the right point depends on the situation, the person, and the reason, so it is rarely the arithmetic midpoint. Splitting the difference between your old boss and their opposite gets you a lukewarm manager. The bad boss cannot show you where the right point is, because they were never there.
The old boss’s behavior was never yours to control, but the rule you built from it is yours, and so is the decision to keep it.
Cato and Confucius both name two teachers: avoid the fool’s mistake, and copy the wise person’s success. The bad example gives you the warning, and the good example has to give you the practice. So the worst boss did teach you more in one sense: more you can remember and name. On its own, it is the half of the lesson you cannot build a practice from. Managers who learned mainly by watching their own bosses tend to collect plenty of warnings and almost no practice, because the good boss’s practice was invisible to them. Fixing that puts a demand on two people.
If you had a great boss, go get the method. Do not settle for “she was just good.” Ask her what she was thinking before that conversation, what she decided not to say, and why she did it that day. Many skilled people have never been asked, and the question itself helps them put words to what they do. I made a related case for studying leaders who are still becoming instead of the polished ones, because the polish hides the method.
If you are the good boss now, your skill is invisible to the people under you, and they are learning more from your mistakes than your strengths. Explain the move after a hard moment. Tell the new lead what you were weighing before the reorg conversation. Leaders who make hard work look effortless hide the path from the people behind them, and in this case the path is the only part worth copying.
Write down what your worst boss did. Three things, with the date and the room if you can remember them. Next to each, write the rule you follow now because of it. Then ask about each rule: did I choose this, or did it happen to me? If the only reason you can give for a rule is a memory, you have found a flinch.
Find the middle for each one. Ask what you needed from that boss and did not get: respect while being corrected, information you could plan with, a promise you could count on. Build the practice around the need, and you can reach it even if you never saw a good boss do it. If your rule is “never correct in public,” the skill in the middle might be “correct in private, within a day, with the specific fix,” which still corrects. Write the skill version next to the flinch version and practice the skill.
Ask your best boss for the method. Email or call one leader you admired. Ask about one specific hard moment you watched them handle and what they were thinking. One honest answer is worth more than a shelf of leadership books.
Narrate your own good moves. After your next hard conversation, tell one person who reports to you what you were weighing and why you chose what you chose. Give them something visible to copy.
Check the rule against the person in front of you. Before you apply one of your old rules, ask whether this person, today, is in the situation your bad boss created, or whether you are reacting to a boss who is not in the room.
The challenge for this week: write down your three bad-boss rules tonight, and for each one write the skill that sits between the flaw and its opposite. Then ask one good former boss how they did the thing you never understood.
Your worst boss taught you clearly because the damage landed where you could see it, with a date and a sentence attached. Your best boss’s skill was built to look like nothing, so you walked away with a feeling instead of a method.
Take the bad lessons, and do not let them run your leadership by themselves. A manager made entirely out of reactions to one bad boss is still being run by that boss. Turn each flinch into a skill you chose and go get the method from the good leaders you had. Then make your own craft visible to the people learning from you now. They are already studying your mistakes. Give them your good moves to study too.
If you want to build leadership from chosen practice instead of old scars, MasteryLab is built for that.
Often you learn more clearly from a bad boss, because bad leadership leaves damage you can point to: a date, a meeting, a sentence that stung. Good leadership tends to make hard moments look ordinary, so the skill behind them is hard to see and harder to copy. The catch is that a bad boss only teaches you what to avoid. If you never study what good leaders were doing, you end up with a list of things you refuse to do and no idea what to do instead.
Research on memory and emotion has found that bad experiences tend to have stronger and longer-lasting effects than good experiences of the same size. A public humiliation or a broken promise gets stored with details: who was there, what was said, how it felt. A good boss’s skill mostly shows up as things that did not go wrong, and the absence of a problem leaves very little to remember.
Write down what your worst boss did and what you now do because of it. For each rule, ask whether it is a skill you chose or a flinch you never examined. A manager who was humiliated in public may stop correcting people at all, which is a new failure. Check each rule against the person in front of you today, and borrow positive practice from leaders you admire by asking them to explain what they were thinking during hard moments.
mimesis is the Greek word for imitation or representation. Aristotle argued in the Poetics that humans are the most imitative of creatures and that we learn our first lessons by imitating. For leaders, the word explains a problem: imitation needs something you can see. When a good leader’s skill is invisible, there is nothing to imitate, so people copy the visible surface or nothing at all.
]]>Nobody pushes back. The review spends an hour on the cache change. Someone writes a careful timeline that fits it, someone proposes a rollback guard, and the action items all point the same way. You leave pleased. You asked for disagreement twice and got none, which feels like evidence that you were right.
Two weeks later the site goes down again, the same way. This time the cause is easy to find, because an engineer on the on-call rotation had it on a dashboard the first time: the database connection pool was maxing out under a traffic spike. She saw it before the meeting. She did not mention it, and if you ask her why, she will say something honest and a little embarrassing. “You seemed pretty sure.”
You did everything the leadership books tell you to do, including inviting dissent out loud and asking what you were missing. The only thing you got wrong was the order, and the order decided the meeting.
This pattern shows up in most companies I have worked with, and it is strange once you look at it directly. The leaders who suffer from it are rarely tyrants. Tyrants get flattered and they know it. The leaders who suffer most are decent people who sincerely want to hear the contrary view, ask for it, and still never hear it. Then they mistake the silence for agreement and the agreement for proof.
There is a name for it in software circles. HiPPO stands for the highest paid person’s opinion, and the joke behind it is that data tends to lose when the HiPPO has already spoken. The joke points at the wrong problem, though. It suggests the boss overrules the data. In most rooms the data never gets a hearing in the first place, because the person holding it decided not to bring it.
So the question for this piece: why does a sincere invitation to disagree fail when the person issuing it has already said what they think?
In the early 1950s, Solomon Asch put people in a room with a group of strangers and asked a simple question about which of three lines matched the length of a fourth. The strangers were working for Asch, and on certain rounds they all gave the same wrong answer before the real participant spoke. About three quarters of participants went along with the obviously wrong answer at least once. Roughly a third of all the answers on those rounds matched the group instead of the lines.
None of those strangers had any power over the participant, and none of them had a say over a raise or the next project. Still, hearing the answer first was enough to bend people about a question they could check with their own eyes.
Asch found one more thing that matters more for leaders. When a single member of the group gave the correct answer before the participant spoke, conformity dropped sharply. One early voice that disagreed with the majority gave people permission to say what they saw.
Now add rank. Put the person who writes your review and decides your next role at the head of the table, and have them give their answer first. The pull Asch measured among strangers does not get weaker.
Rome built this problem into its procedure. When the Senate debated, the presiding magistrate asked senators for their opinions in order of rank, starting with the most senior. On December 5, 63 BC, the Senate met to decide what to do with the men arrested in Catiline’s conspiracy. Sallust, who wrote the history of the affair, notes that Decimus Junius Silanus was asked first because he was consul-elect. Silanus proposed death.
The senators who followed went along with him. Then Julius Caesar, a praetor-elect and much further down the order, stood and argued against execution. The room moved. Sallust records that Silanus, the man who had opened the debate with a death sentence, said after Caesar’s speech that he would support a different motion. Cato spoke later and swung the room back, but the lesson for leaders sits in the first hour: the opinion that held the room was the first senior one, and it held only until someone with enough standing was willing to pay the price of disagreeing out loud.
Most of your meetings do not have a Caesar in them, and they end after the first hour.
The Mishnah, the early collection of Jewish law, describes how the Sanhedrin handled capital cases. In ordinary cases the court could start with its most senior judges. In cases where a life was at stake, the court started “from the side,” with the junior judges. The later commentary gives the reason plainly: so that a junior judge would not hear the senior view and simply follow it.
That is a court that looked at the cost of being wrong, saw that the order of speaking changed the outcome, and rewrote the order for the cases that mattered most. The senior judges kept their seats and their authority, and they spoke later.
I have watched this in architecture reviews where the CTO says “I’d just use Postgres” in minute one, and the next fifty minutes become a discussion of how to make Postgres work. I have watched it in planning, where a founder names the top priority before anyone shows the usage data, and the data that arrives afterward gets presented as support.
I have also done it myself. When I ran Integrum, I was usually the first voice in planning. I had the most context and I liked to set the direction early so nobody wasted time. I took the quiet that followed as alignment. Looking back, some of it was alignment, and some of it was smart people deciding the argument was not worth having with the person who signed the checks.
The clearest place I see the effect now is at my own desk. If I tell Claude my theory of a bug before I give it the logs, it gets very good at building the case for my theory. If I give it the logs and ask what it sees, I get a different list, and sometimes the right answer is at the top of that list instead of the bottom. The model behaves like the most junior person in a room who has just heard the boss speak. The difference is that it shows me the effect in two minutes instead of two weeks, and I can rerun the meeting.
Put the evidence side by side and the mechanism is simple. Rank does not have to overrule anyone. When the senior person speaks first, their view becomes the default the rest of the room reacts to. Every person after that is now either agreeing with the boss or disagreeing with the boss, and disagreeing carries a cost the first speaker never paid.
That cost works on people in two ways. Some start doubting what they saw, the way many of Asch’s participants did once the group had answered. Others stay sure and decide the fight is not worth having. The evidence never reaches the table in either case, and the same fix works for both.
The second group weighs the cost quickly and mostly without noticing. Is my evidence strong enough to contradict the VP in front of the team, and is it worth it for a forty-minute outage? Most of the time, the answer feels like no. So the dashboard stays closed, and the leader, who did ask, hears nothing back and reads the nothing as confirmation.
doxa is the Greek word for opinion, from a verb that means to seem. It also meant reputation, how a person appears to others. The double meaning fits this problem well. The senior person’s first sentence is an opinion, and it arrives carrying the speaker’s reputation, so the room treats it as more settled than the evidence behind it. Plato set doxa against knowledge that can give an account of itself. An incident review is supposed to produce the second kind. When the boss speaks first, it tends to produce the first kind with a timeline attached.
The Athenians had a formal answer for who gets to speak. isegoria was the equal right of every citizen to address the assembly, and debate opened with a herald asking who wished to speak. I wrote about that right as a voice window before a decision. Here the lesson is narrower. Your team has that right on paper in every meeting you run. Speaking first takes it away from them in practice, even though nobody ever said they could not talk.
Three things follow from this, and they get worse in order.
First, you lose the information that would have saved the decision. The on-call engineer’s dashboard was the most useful thing in that incident review, and it never reached the table. That is a direct cost, paid two weeks later in a second outage.
Second, the agreement itself starts to look like evidence. A leader who hears no objections becomes more confident, and that confidence makes the next room even quieter. Over a year, you can end up surrounded by people who agree with you for reasons that have nothing to do with whether you are right. I have written about how organizations keep promoting the loudest voice in the room. Speaking first is the small daily version of the same habit, and decent leaders do it constantly.
Third, and most expensive: the leaders who believe they invite dissent are the ones most exposed. “Push back on me” is a sincere sentence, but coming from the boss, it is one more instruction from the boss, delivered after the boss has already told everyone what they think. Only a change in the order undoes that.
None of this settles anything for the person holding the evidence. If you are the engineer with the dashboard, the boss going first raises the price of speaking and leaves the duty where it was. An outage that will happen again is worth one uncomfortable sentence in front of the VP, and “you seemed pretty sure” is an explanation you will not want to give a second time. The order explains her silence without excusing it. This piece puts the weight on the leader because the leader sets that price for everyone in the room at once.
Two more lines are worth drawing so this does not get misread. When someone has a new idea, the idea needs a person willing to say it first and risk losing, and waiting for the room to bless it kills it. And some decisions are not discussions at all; when you are deciding, say so. This piece is about the meetings where you really are asking. In those meetings, your opinion is the one most likely to drown out the others, so it should go last.
You control the order of the room and the moment you speak, and you have no control over whether anyone in it is brave. So build an order that asks as little courage of them as possible.
Senior voices last. In any judgment meeting, meaning one where the group is supposed to work out what is true or what to do rather than hear updates or receive a decision, call on the most junior people first, then the people closest to the evidence, then the senior people, then yourself. Say out loud that this is the order and why. The Sanhedrin did not apologize for it.
Evidence before opinions. Have people read the data, the logs, or the proposal in silence at the start of the meeting, or write their view in a doc before anyone talks. Amazon’s habit of opening meetings by reading a six-page memo in silence works partly for this reason: the evidence is in every head before any voice is in the air. This matters more than the speaking order alone. Your silence is not neutral if the room already knows your view from last week’s Slack thread or a hallway conversation, and a view written down before the meeting cannot drift toward yours.
Write your call down and keep it in your pocket. Before the meeting, write what you think and why. Do not say it until everyone else has spoken. Then compare. If the room landed somewhere different from your note, you just learned something you would have suppressed by going first.
Ask the model blind. Give Claude or whatever you use the logs, the diff, or the data before you give it your theory. Ask what it sees. Then give it your hypothesis and ask what would prove it wrong. You get the junior-first meeting in miniature, and you see your own anchoring in two minutes.
Count the contradictions. For the next month, keep a tally of how many times someone in a meeting told you that you were wrong about something that mattered. If the number is zero, the explanation is almost never that you were right every time.
The challenge for this week: in every meeting where you want real input, write your view on paper before you walk in, say nothing until the most junior person has spoken, and compare your note with where the room landed. Count the differences on Friday.
The VP in the incident review was not arrogant and did not silence anyone. They spoke first, asked for disagreement, and got the polite quiet that rank produces when it goes first. The engineer with the dashboard took the easy side of a bad trade, a choice that made sense in the moment and was still wrong, and the company paid for it with a second outage.
Rank changes what it costs to say a true thing, and the first senior sentence sets that price for everyone else in the room. The rank stays with you either way, so stop spending it in the first minute. The incident review exists to find out what broke, and the order of speaking decides whether the room can find it. Let the evidence and the junior voices go first, and save your opinion for the end, when it can be tested against what they found.
If you want to practice leading rooms where the evidence gets heard before the rank does, MasteryLab is built for that.
HiPPO stands for the highest paid person’s opinion. The HiPPO effect is what happens when the most senior person in a meeting states a view early: the group starts treating that view as the default, and anyone who disagrees now has to argue against the boss. People weigh that cost, hold back contrary evidence, and the leader reads the resulting quiet as agreement. The senior person never has to overrule anyone. Speaking first does the work.
Leaders should speak last when they are genuinely asking for input, because their first sentence sets the default the rest of the room reacts to. Once the senior view is on the table, every later comment becomes either agreement or disagreement with the boss, and disagreement carries a social cost the first speaker never paid. Speaking last lets the evidence and the junior views arrive while they are still free to say. The leader loses nothing by waiting, because they still make the call.
Change the order before you change the speech. Have people read the evidence or write their view before anyone talks. Call on the most junior or most expert people first and the most senior people last. Write your own position down before the meeting and keep it to yourself until the room has spoken. Then count how many times in a month someone told you that you were wrong. If the count is zero, the room is repeating you back to yourself.
doxa is the ancient Greek word for opinion, from the verb dokein, to seem. It also meant reputation or glory, how a person appears to others. Plato set doxa against episteme, knowledge that can give an account of itself. In a meeting, the senior person’s first sentence is doxa in both senses at once: an opinion that carries the weight of the speaker’s reputation, which is why the room treats it as more settled than the evidence behind it.
]]>For every man, everywhere, the finest thing is to be rich, healthy, and honored by the Greeks, to reach old age, to give your parents a fine burial when they die, and to be finely buried by your own children.
That is a good life the way most people would describe it, and it could run on a retirement brochure today. Swap “honored by the Greeks” for “respected in your field” and you would not notice it was 2,400 years old.
Socrates accepts every item on it and asks what makes them fine. Hippias has nothing to say.
You can name everything on your good-life list and still not know what makes any of it good. I did that for years. When you cannot say why the first item is good, you have no way to tell when the fifth one stops being good, and the list keeps growing.
(Scholars still argue about whether Plato wrote the Hippias Major. The question in it works no matter who wrote it.)
Ask anyone what a good life is and you get nouns. Health, family, enough money, work that matters, respect, a decent house, a few good friends, time to enjoy them.
None of those are wrong. Hippias was not wrong about health either. The trouble is that a list answers “what” and says nothing about “why.” It also has no rule for what to do when two items fight. Health and money pull against each other in most careers, and the house and the time pull against each other in most marriages. A list cannot settle that, because it treats every item as equally good and never says what they are good for.
Socrates finds the first crack fast. He points out that Hippias’s list is fine for some people and shameful for others. Achilles was the son of a goddess. He could never bury his immortal mother, and Homer has him choose a short, famous life over a long one. By Hippias’s list, the greatest hero the Greeks had would count as a failure.
When a reason only works in some cases, it tells you about the person who likes it, and nothing about what makes the thing good. What Hippias had was a list of things successful people tend to have, and he was one of those people.
In my thirties I owned a company during a bad economy. My kids were in competitive soccer, which is not cheap. The business was barely breathing. We pared the family down to the basics. I bought a twenty-year-old Toyota 4Runner on purpose, and we kept the house simple.
There was no philosophy behind it. We had no money and very little choice. But those years were good in a way I did not understand until later.
Then the kids graduated and the economy turned, and I refilled. I bought multiple cars, did rounds of remodels, and took on more commitments. I told myself I had earned it, and in some ways I had.
Every one of those things was fine. A reliable car is good, and so is a kitchen that works. I could have defended each purchase in a sentence. What I could not do was say what made the first car good in a way that would tell me when the third car had stopped being good. I had a list of good things with no reason behind it, and nothing inside a list like that ever tells it to subtract. In my thirties the economy did the subtracting for me. Once it stopped, every new item passed, because good things are good.
It cost me something the lean years had given me. I had less attention and less time, and I lost the plain clarity of knowing what mattered because there was no room for anything else. I traded those away without noticing, one fine purchase at a time.
The Hippias Major has a test that would have caught this. All through the dialogue one rule holds: whatever makes a thing fine cannot be something that produces harm. When Socrates and Hippias try “the powerful” as an answer, it fails on that rule right away. Power to do bad things is not fine, and every later candidate has to survive the same test.
My growing list never faced that test. I asked whether each item was good, and the answer was always yes. I never asked whether the pile was doing harm. It was, slowly, in ways that did not show up on any one receipt.
After the list fails, the dialogue tries more candidates, and the order matters.
First comes the appropriate: what fits, what suits the occasion. It fails because what fits can make a thing look fine without making it fine. A man in good clothes looks better, and the clothes do not make him better.
Then the useful. It fails because a thing can be useful for bad ends, and those uses are not fine.
Then the beneficial, meaning whatever produces good. That one comes closest, and it fails in a way Socrates finds absurd. If the fine is the cause of the good, then the fine and the good are two different things, and neither man can accept that the fine is not good.
The last candidate is what pleases the eye and the ear. Of everything they try, it looks the most like an answer, and it fails too.
The Greek word in play is kalon, which means fine, beautiful, and noble in a single word. (In this piece I use “good” in the everyday sense for what the dialogue calls fine.) The dialogue keeps finding that the beautiful part of it is the easiest to fake. I think most people make that mistake once the list stops working. Their good life starts to mean the good life that looks good, like the photo from the trip or the kitchen in the after picture. A remodel can please the eye every time you walk in and still tell you nothing about whether your life is going well.
The dialogue never reaches a definition. Socrates closes with an old proverb: fine things are difficult. That ending frustrates people. I find it honest, because it admits the question takes work and that most people quit on it early.
Looking back, the economy made the choice for me in my thirties. A reason is what lets you make the same choice when nothing is forcing you, and there is no need to wait for a bad year to start.
In my late forties I came back to a pared-down life. Most of the credit goes to my wife. She is a hardcore minimalist, much more disciplined about it than I am, and she has been the anchor that keeps me there. Her example helps, and the reason still has to be mine, because the choice gets made in my own head when something catches my eye.
A shorter list would have grown back. What changed was that I started to have a reason I could use on something new before it came in the door. Mine came back to something I had felt in the lean years without saying it: there is beauty and simplicity in keeping less, and anything new has to earn its place against that. Something still shows up and I want it. I let most of it pass now, because I can ask what it would be good for and whether the answer is one I actually believe.
Run my reason against my own history and it sorts things fast. The twenty-year-old 4Runner I bought on purpose passes, and the third car fails, because it was one more thing to keep up and did nothing the first two could not.
Keeping less is not the account either. An empty garage can turn into one more thing to be proud of. In his later dialogues Plato puts the fine in a well-ordered soul, where good judgment decides what a person wants and things are good when that judgment uses them well. That is closer to what I mean by a reason than any count of what is in the garage.
The word for what Socrates kept asking people to produce is logos, which here means an account: a reason that holds across cases and tells you what to do with the next one. Hippias could list fine things all day and still could not give an account of them, so his list was useless the moment something new came along.
An account explains several items on your list and rules out something you have been tempted by. If your reason only explains the one item you wrote it for, you have a preference with a sentence attached.
| A list | An account | |
|---|---|---|
| What it answers | What is good | Why it is good |
| What it does with something new | Adds it if it looks good | Tests it against the reason |
| When it gets shorter | Only when something outside forces a cut | When the reason rules an item out |
| When two items conflict | Says nothing | Tells you which one serves the reason |
You see the same failure at work, and it costs teams real decisions. Most company values pages are a Hippias list: integrity, excellence, customer first, innovation. Nobody can make a hard call with them, because nobody wrote down what makes those things good or what to do when two of them collide. When customer first and excellence point in opposite directions on a Friday release, the values page is silent, and the loudest person in the room decides.
This takes an evening, and you will not finish it. Start anyway.
1. Write the list. Ten minutes. Use honest nouns. Put down what you chase, not what you think you should.
2. Finish one sentence for each item. “This is good because…” Write the real answer, even when it is embarrassing. “Because people will see it” is a real answer, and it tells you something.
3. Find the reason that repeats. Look across the sentences. Usually two or three reasons keep coming back. One of them is closer to an account than any item on the list. The reason worth keeping says what the items are for. If your answer to “what is it for?” is another item on the list, you have not left the list yet.
4. Run the harm test. For each item, ask when it would stop being good. A second car is good up to some point. Where is that point? If you cannot name it, all you know so far is that you like it.
5. Use it on the next thing before it arrives. The account earns its keep on decisions you have not made yet: the car, the job offer, the remodel you are pricing, the commitment someone asked you to take on. Run the reason against it out loud. If the thing passes, take it without guilt. If it fails, let it go and notice how little you miss it.
6. Expect to come up short. Socrates did not finish this and neither will you. A rough reason you use every week beats a polished list you cannot apply to anything.
I wrote about a cousin of this in It Was Never a Sacrifice, about the things you give up that were never the good in the first place. That piece covered what to drop, and this one asks why you keep anything at all. If the larger question of what a flourishing life is interests you, I took it on directly in eudaimonia versus happiness.
Hippias walked away from that conversation with his list intact. He was still rich and still sure of himself, and he did not have a single reason he could use.
Most of us are in the same spot. We know what is on the list and would defend every item, and we still cannot say what makes any of it belong there, which is why the list keeps growing and the life does not seem to get any better.
Only a reason can tell you when to stop adding to the list. Get one by tonight, even a rough one, and use it on the next thing that catches your eye.
If you want to work alongside people who want to know why the things they keep are good, MasteryLab is where that work happens.
Plato never gives a one-line answer, and the Hippias Major shows why. When Socrates asks what makes things fine, every list of good things fails, because the items are fine for some people and harmful for others. The working demand in the dialogue is that whatever makes a life good has to hold across cases and can never be something that produces harm. Plato’s later dialogues tie that reason to justice and wisdom in the soul, not to the possessions on the list.
The Greek word kalon means fine, beautiful, and noble at once. It covers a beautiful object, a well-made tool, and a brave act with the same word. Aristotle says virtuous action is done for the sake of the kalon. The Hippias Major is an extended attempt to say what the kalon is, and it ends without a definition.
The Hippias Major is a short dialogue in which Socrates asks Hippias, a famous paid teacher, what the fine is. Hippias offers a beautiful young woman, then gold, then a list of the good life: wealth, health, honor, old age, burying your parents well and being buried well by your children. Socrates tests more candidates, including the appropriate, the useful, the beneficial, and what pleases the eye and ear. All of them fail. Scholars still debate whether Plato wrote it.
Start with the list, because everyone has one. Then finish the sentence “this is good because” for every item and look for the reason that repeats. Test that reason by asking when each item would stop being good. A reason that explains several items and rules out something you have been tempted by is the start of an account, and you can use it on the next decision before it arrives.
]]>A month later she had stopped talking in design reviews. You had felt terrible the whole month, and none of it reached her.
Regret stays inside the person who made the mistake, and remorse goes looking for the person who got hurt. They start from the same sick feeling, but only remorse reaches the person you hurt. Most of us, me included, are very good at regret and have done remorse well a handful of times. We count the first as if it were the second.
Regret is busy, and that is why it is easy to confuse with moral work. It runs the replay at 2am and rehearses the speech you would give if you ever gave it.
Regret also talks. You tell a peer, “I was out of line in that review,” and the peer says you are being too hard on yourself, which helps a little. Now two people have heard about the harm, and neither of them was the one harmed.
Next comes the extra warmth. You praise her work in a channel she reads and add an emoji. You are paying a debt in a currency the other person never agreed to accept, and without telling them it is a payment.
Eventually you start avoiding her. You stop posting in her channel and let someone else run the next review she is in, and you wait for it to blow over. Sometimes it does, in the sense that nobody mentions it again.
Every one of those moves happens inside your head or in front of someone who was not hurt. From the inside they feel like suffering, and suffering feels like it should count for something. Whatever it counts for, she never sees any of it.
Regret does some good when it changes what you do next. The next engineer in that design review never gets talked over. She is still owed a sentence from you, because the change protects someone else and leaves her where you left her.
You can sort it out in about a minute.
1. Who is the sentence about? Listen to what you say when you think about it. “I can’t believe I did that.” “I’m better than that.” “What is wrong with me?” Every one of those has you as the subject. Remorse sounds different: “She lost the room because of me.” “He stayed up redoing the work I broke.” When the sentence in your head is about you, you are in regret.
2. Has the person heard anything? A specific sentence, from you, to them. Warmth does not count, and neither does a better tone in the next meeting or telling their manager you feel bad. If the person you hurt could not repeat back what you said to them about it, nothing has reached them.
3. Who is doing the work right now? This one catches people who think they apologized. If the person you hurt is reassuring you, telling you it is fine, telling you not to worry about it, the apology has turned back around. They took the hit, and now they are also managing your feelings about the hit.
Side by side, the difference looks like this:
| Regret | Remorse | |
|---|---|---|
| Who the sentence is about | You | The person you hurt |
| Where it happens | In your head, or with someone who was not hurt | With the person who was hurt |
| What they get from it | Nothing they can see | A plain admission and a fix |
| What it asks of you | Time and pain | A conversation you do not control |
Aristotle has a word for regret: metameleia (μεταμέλεια), literally a change in what you care about after the fact. In Book IX of the Nicomachean Ethics he uses it in a passage about bad people, and he says they “are full of regret.” They are at odds with themselves, he says, wanting one thing and doing another, and the regret is what that split feels like from the inside. In that passage heavy regret is a symptom of a divided person, and he does not offer it as evidence of a good heart.
Marcus Aurelius defines it more tightly. In Meditations 8.10 he calls regret (his word is metanoia) a kind of reproach of yourself for having neglected something useful. The reproach is aimed at yourself, and for Marcus that is fine, because it is a signal. Something useful got neglected, so go do the useful thing. The failure is stopping at the reproach, and most of us stop there.
The English word “remorse” comes from the Latin remordere, to bite again. It keeps biting, and it does not stop on its own. You stop it by doing something.
We stay in regret because it protects your picture of yourself. In the replay you are still a decent person who did one bad thing and feels awful about it, which is the kind of person who would never do it again. Repair puts that picture at risk. To repair, you have to walk up to her and say the plain sentence, “I called your idea junior in front of the team, and that was wrong,” and then stand there while she decides what she thinks of you. The replay is painful, but you control it, and the conversation is a risk you do not control. The replay hurts enough to feel like penance, and it never asks you to stand in front of anyone.
I wrote earlier this year about the difference between shame and self-pity. That piece covered the inside job, what you do with the gap between who you are and who you meant to be. You can do that job perfectly and still owe someone a sentence. It is also a different regret from the one I wrote about in asking fear the wrong question, which is regret about a life you have not tried yet. This one is about something you already did to someone.
Remorse can go wrong too, and it usually goes wrong by getting large.
A leader realizes she crushed someone in a meeting. She books a 1:1, and it runs twenty minutes. She explains her week, her stress, the pressure from above. She tears up, says she feels sick about it, and asks if they are okay. The engineer spends the meeting saying “it’s fine, really, don’t worry about it.” They walk out having comforted their boss about the thing their boss did to them.
The remorse was sincere, and it got so big that it turned back into regret with an audience. The person who was hurt ended up carrying it.
You see the same thing in a public apology timed for the apologizer’s relief, before the hurt person has decided how they feel, and in the apology that ends with “can we be good?” and wants the answer now.
Repair has a right size, and you can miss it in either direction. Too little leaves the person alone with the harm, and too much hands them your feelings to carry as well. A cutting remark in a review calls for one honest sentence and a corrected record. If your apology is bigger than what you did, ask whose need it is meeting.
1. Say what you did in one sentence. Name the act: “I called your idea junior in front of the team, and that was wrong.” Leave out “if,” “I’m sorry you felt,” and the paragraph about your week. If you cannot say it in one sentence, you are still deciding whether it was really your fault.
2. Go to them, inside a day. Go to the person before you mention it to their manager, a peer, or the team channel. The longer you wait, the more the silence says for you, and what it says is that it did not matter enough to mention.
3. Ask what it cost them, then listen. “What did that cost you?” is a hard question to ask. You may learn the damage was bigger than you thought. She might have stopped bringing ideas to reviews, or started looking at other teams. You need to know that, and you cannot learn it from the replay.
4. Fix it where it broke. Harm done in front of eight people gets corrected in front of the same eight: “In last week’s review I dismissed an approach too fast. I was wrong, and I want to hear the rest of it.” A private apology for a public hit leaves the room believing the hit.
5. Change the behavior where they can see it. The apology is a promise. The next three reviews are where she finds out whether you meant it.
6. Then stop. Do not ask for forgiveness on a schedule. Do not check in every week to see if you are okay now. Let her decide what the relationship is. Your job was the sentence and the fix, and what she does with them is up to her.
Sometimes you cannot reach the person. They left the company, they will not take the meeting, or they are gone for good. The duty moves instead of leaving with them, to what you can still fix: correct the record with the people who heard the wrong version, and treat the next person in that seat the way you should have treated them.
Repair gets cheaper with practice, and more expensive with every day of silence. A sentence you could have said the next morning is hard to say a month later, because now you also have to explain the month.
So practice on the small ones. Interrupted someone in standup? Tell them at the end of standup. Sent a sharp Slack message on a bad day? Reply in the same thread. Quick, plain repairs build the habit for the day you do something that actually hurts someone.
If you lead people, they are learning what happens after a mistake by watching you. A leader who never says “that was on me” out loud teaches a team to hide mistakes, and a team that hides mistakes finds out about them in production. Repairing in the open shows everyone that the sentence is survivable. It is one of the few leadership behaviors you cannot fake, because it only counts when the other person hears it.
The 2am replay does not have to go away. Treat it as a prompt instead of a punishment. When it starts, ask one question: who needs to hear from me tomorrow? Write the name down and go back to sleep. Then send the sentence in the morning.
Feeling bad about what you did tells you something real. It means you know the standard, and you know you missed it. That feeling has one job, which is to get you out of your chair and over to the person you hurt. If it stays in your head, it can run for years without helping the person you hurt, and you will mistake the hours you spent hurting for the work you never did.
The repair helps you too. In Plato’s Gorgias, Socrates argues that doing wrong harms the one who does it more than the one who suffers it, and that a wrongdoer should go looking for correction the way a sick man goes looking for a doctor. Staying in the replay keeps you stuck with what you did, and saying the sentence is what frees you, though your share of that only arrives by way of the person you hurt.
You probably already know who it is. There is someone who has been waiting a week, or a year, for one plain sentence from you. Write it today and send it, or say it out loud at their desk, and then let them decide what happens next.
If you want to work alongside people who say “that was on me” out loud and then fix it, MasteryLab is where that happens.
Regret is pain about what you did that stays with you: you replay it, you wish it undone, and the person you hurt never hears about it. Remorse is the same pain pointed at the person who got hurt, and it pushes you to tell them what you did and fix what you can. Regret centers your picture of yourself, and remorse centers the person you hurt.
metameleia is the ancient Greek word for regret. It comes from a verb meaning that what you care about has changed after the fact. Aristotle uses it in Nicomachean Ethics IX.4 to describe bad people, who he says are full of regret because they are at odds with themselves. In his account the regret is a symptom of that inner split, and it does nothing for anyone else.
Say what you did in one plain sentence, to the person, within a day. Ask what it cost them and listen without defending. Fix what you can in the same place the harm happened, then change the behavior where they can see it. Keep the apology shorter than the harm. If they end up comforting you, the apology has turned back toward you.
It is a sign that you care about something, usually your picture of yourself as a decent person. Caring about the person you hurt shows up as action toward them: a sentence they actually hear, a fix they can see. Self-blame that never leaves your head can run for months while the other person gets nothing from it.
]]>That order is backwards.
A faster way to learn something new is to decide who you are going to teach before you know anything about it. Name the student first. Then open the docs. Researchers call this the protégé effect. People who do this remember more of what they read, in better order, and they get that gain before they teach a single person. Teaching for real within the week is what makes it last.
This used to be a nice study tip. In a year when a specific skill can expire in months, it is closer to a survival skill, and the best student you could ask for is already open on your desk.
Look at how we build learning. Courses run intake first and the project at the end. Onboarding hands a new engineer a reading list and a calendar of shadowing before anyone asks them to explain anything. Certifications test you once you have studied. Every one of them assumes you fill up first and pour out later.
AI added a second argument on top of the first. If the model already knows the answer, why learn at all? Ask it, ship the result, move on. The other camp says the opposite: do not let the model do your thinking, learn it the hard way, earn it.
I am firmly in the first camp for most things. I ask the model for the regex, the flag, the date format, the config I will touch once a year. Learning those by heart would be a waste of a good afternoon.
Both camps miss the same thing. For the material you have to own, meaning anything where you will later judge someone else’s answer, what matters is the order you learn it in, and the default order is the slow one.
The protégé effect comes down to this: people learn more when they are learning for someone else. One of the cleanest tests of it came in 2014, from John Nestojko, Dung Bui, Nate Kornell, and Elizabeth Bjork. Two groups read the same passage. One group was told they would be tested on it. The other was told they would have to teach it to someone. Then both groups took the same test, and nobody in either group ever taught.
The people expecting to teach remembered more. They also remembered it organized. Their recall clustered around the main points instead of scattered details. The only thing that changed was what they believed was coming, and that belief changed how they read.
Watch yourself read something when nobody is coming. You highlight and nod along, recognizing each sentence as it goes by, and recognition feels like knowing. Then someone asks a simple question and you have nothing.
Reading for a student is a different activity. The whole time, part of your head is asking what you would say first and which example would make it click. The part you could not explain jumps out while the page is still in front of you and you can still fix it. You are building the explanation as you go, and building it is the learning.
A year earlier, Logan Fiorella and Richard Mayer found the second half. Students who prepared to teach did better on a test right away. Students who then actually taught, explaining the material to a camera, were the ones who still had it a week later. Expecting to teach gets you started, and teaching for real is what makes it stay.
Seneca put it in a letter to his friend Lucilius almost two thousand years ago: “Men learn while they teach.”
One detail matters for how you use this. In that 2014 study, the gain came from what people expected while their eyes were on the page. Telling yourself afterward that you now have to teach it cannot go back and change how you read, so name the student before you start.
I go fast with models. They write most of the code on my desk now, and that is the right call. What the model cannot do for me is know enough to tell when it is wrong. Someone has to say what done looks like, read the diff, and notice that the retry logic will hammer an API that rate-limits at ten requests a second. When my name is on the approval, that someone is me. “The model wrote it” will not explain a bug I signed off on, so I have to understand the thing.
The thing keeps changing. A new agent framework, a new way of wiring tools, a new API, a pattern that was best practice in the spring and dead by the fall. Being a beginner used to happen once or twice in a career, and now it happens every quarter.
The people who stay at the front edge have a faster way of being a beginner, and that counts for more than how much they already know. I wrote about learning velocity as the compound advantage a while ago. This is the method under it.
The Feynman technique, as people usually teach it, goes like this. Pick a topic. Explain it in plain words to someone who does not know it. Find the spots where you wave your hands. Go back to the source and fix them. Repeat until the explanation is simple.
It works. Its weak point is the student. Most people run it with an imaginary student, and imaginary students are polite. They never interrupt, and they never ask the obvious question you were hoping to skip.
In 2009, a Stanford group found that students worked harder to learn science when they had to teach it to a computer character than when they learned it for themselves. That study is where the name protégé effect comes from. It was software that could barely hold a conversation. The student on your desk today can.
Before I open the docs for something new, I tell Claude I am going to explain the topic to it in an hour. Then I read with that hour in mind. When I come back, I tell it to play a sharp new engineer who has never heard of the thing. Stop me every time I hand-wave. Ask the dumb question first. Push back if an example does not hold up. When I finish, list everything I got wrong or skipped.
My first explanation is always worse than I expected, and that is the reason to do it. Every gap it finds is a gap I would have carried into a design review or a pull request. Ten minutes with a student that talks back finds them while they are still cheap.
The model answers my questions all day. For the few things I have to own, I flip the roles and let it ask the questions.
1. Name the student and the date before you open anything. A person, the team, or the model. Write it down. “Thursday I walk the team through how the new deploy pipeline handles rollbacks.” The sentence changes how you read the first page.
2. Read for the explanation instead of the coverage. While you read, look for what you would say first, the one example that carries it, and the place your student will get confused. This is the same move as reading with a job in mind, pointed at learning instead of building.
3. Teach the model, then let it grade you. Run it as a beginner first and a skeptic second. The beginner exposes holes in how you explain it, and the skeptic exposes holes in what you understand. End every session by asking for the list of what you got wrong, then go back to the source for each one.
4. Then teach a human inside the week. The model gets you through day one. A ten-minute walkthrough at standup or a Slack post with a diagram is the half that still holds next week. Real people ask questions the model did not think of, and having said it out loud to a colleague makes it stick.
5. Leaders, assign the teach-back with the learning. “Go learn the new framework” usually gets you someone who skimmed the docs, and “Friday you walk us through the new framework” gets you someone who read them hunting for the explanation. The difference costs you one sentence. It is the teacher-side companion to leaving room for people to figure things out.
Run every question through a teach-back and you will never ship anything, so I sort with one question. Will I have to judge someone else’s answer to this later? If the answer is no, ask the model and move on. If the answer is yes, because I will review it, defend it in front of a team, lead people who build on it, or decide whether to bet on it, then I learn it, and I learn it by teaching it.
Over time the habit runs without the checklist. You start something new and you are already sketching the explanation in your head, already picturing who will be confused and where. At that point you are always reading as if somebody is coming, because somebody usually is.
We learn in the order we were taught to learn, and that order puts the most powerful part at the end, where most people never get to it. Moving the student to the front costs one sentence and ten minutes of hearing where you are wrong, and it changes how every page reads. The explanation is where you find out whether you know the thing or only recognize it.
The student can be your team, a new hire, or the model on your desk, which is the most patient student ever built and will point at every gap you skipped. Use it. The next thing you have to learn, name the student before you read the first page.
If you want a room full of builders who learn new things in public and teach each other as they go, MasteryLab is where that happens.
The protégé effect is the finding that people learn material better when they learn it in order to teach someone else. Stanford researchers named it in 2009 after students worked harder to learn science for a computer character they had to teach than they did to learn it for themselves. Related studies found that simply expecting to teach changes how people read: they remember more, and they remember it organized around the main points.
Yes. In a 2014 study, people told they would teach a passage recalled more of it, and recalled it better organized, than people told they would be tested, even though nobody actually taught. A 2013 study added the second half: expecting to teach helped right away, and actually teaching is what held up a week later. Name the student before you start, then follow through and teach within a few days.
Study with the expectation of explaining it. Then tell the model to play a sharp beginner who has never heard of the topic, stop you every time you hand-wave, and ask the obvious question first. When you finish, ask it to list what you got wrong or skipped. Go back to the source for those gaps. The model is a better student than an imaginary one because it talks back.
For most questions, let the model answer and move on. For anything you will have to review, defend, lead, or decide, learn it yourself, because you will be judging someone else’s answer to it, including the model’s. A simple sort: if you will need to tell a right answer from a wrong one later, teach it back before you rely on it.
]]>The series plan files this one as a law to oppose, even though half of it is good advice. Greene tells you to stay calm, and he is right about that. Then he tells you to make the other person lose their temper on purpose, and that half costs more than it wins. The cost lands on the person doing the stirring.
Law 39: “Stir Up Waters to Catch Fish.”
Greene’s judgment goes like this. Anger and emotion are strategically counterproductive, so you must always stay calm and objective. But if you can make your enemies angry while you stay calm yourself, you gain a decided advantage. Put them off balance. Find “the chink in their vanity” through which you can rattle them, and “you hold the strings.”
The title is an old Chinese military saying. Muddy the water and the fish cannot see the net.
His lead story is Napoleon and Talleyrand in January 1809. Napoleon came back from Spain convinced his foreign minister was plotting against him, and in front of his other ministers he raged at Talleyrand for half an hour, finishing with a line history kept: “You are shit in a silk stocking.” Talleyrand stood there and took it without a flicker. On his way out he remarked to the others that it was a pity so great a man had been so badly brought up. Greene treats that half hour as the start of Napoleon’s slide. The emperor showed the room he could be moved, and the man who stayed still came out ahead. I will come back to this story, because it can be read the other way.
Anger that someone else set off, or anger far out of proportion to its cause, makes people decide badly, and I have never met anyone who disputes that. The person who stays level in a heated room has an edge, and the edge is real. They hear more, and they still have their judgment after the other side has spent theirs.
Provocation works too. In board prep, someone asks the engineering lead a question built to make him defensive, and the defensive answer becomes the story of the meeting. In a pull request thread, one comment lands on a senior engineer’s pride and the next forty minutes go to who is right instead of what the code should do. I have done it myself, asked the needling question in a review because I knew it would get a rise, and for about an hour after I got the rise it felt like winning.
Aristotle wrote the manual for this, which is worth knowing before anyone calls Greene a cynic. In the Rhetoric he defines anger as a desire, mixed with pain, for revenge for a slight that was plain to see. Then he spends a chapter sorting slights into contempt, spite (getting in someone’s way for no gain to yourself), and insult. Anyone who has read that chapter can manufacture a slight, and Law 39 runs on exactly that: anger follows the slight, and the slight is something you can make.
So the calm half of the law is sound, and the provoking half does work. What Greene skips is what you get when it works.
First, what counts as the move. A hard question does not break this law. A code review that says “this will lose data” can make someone angry, and so can a true answer they did not want to hear. The test is what you were aiming at. Law 39 aims at the other person’s anger so that their judgment fails. An honest question aims at the truth, and any anger that follows is a side effect you would have avoided if you could.
Then the act itself, before any bill comes due. Working someone’s temper on purpose treats their judgment as a lever for you to pull. That is a wrong done to a person with a mind of their own, and it would stay wrong even if every trade below came out in your favor. The costs make it a bad deal on top of that.
The saying is about muddy water, so start there. The fisherman stirs the river so the fish cannot see him, and he is standing in the same river. Once it is brown, nobody can see the bottom, including him.
That is the first and biggest cost, and it is the one Greene never mentions. You provoke a colleague to win a meeting, and what you won is a decision made on a worse version of their judgment. The engineering lead in that board prep knew something about the migration plan. He was going to say it. Instead he spent his five minutes defending himself, the board heard a defensive man, and the thing he knew went unsaid. The person who asked the needling question walked out ahead and walked out less informed than they came in. If you run the company, you are the one who needed what he knew.
Aristotle saw this from the other side. In the first chapter of the Rhetoric, before he gets to the catalogue of slights, he says it is wrong to stir a judge to anger or envy or pity, because it is like bending a carpenter’s rule before you measure with it. You needed a straight edge, you bent it on purpose to win the argument, and now every measurement you take with it is off. When the colleague you provoked is the one whose judgment you rely on, you have bent your own ruler.
The second cost is to the part of them you provoked. Plato divides the soul into reason, appetite, and thumos, the spirited part, the part that gets angry. In the Republic he gives thumos a job. It is reason’s ally, the part that rises against injustice and then comes back when reason calls it, like a sheepdog to the shepherd. It is the part of a good engineer that bristles when the numbers in the deck are wrong. Law 39 finds that dog and sets it barking at nothing. Do it enough and the dog stops coming when called, and the person you trained that way is someone you work with.
The third cost is the room. Provoke one engineer in a design review and every engineer in the next review arrives braced. Some of the most useful things anyone says in a company are half-formed doubts: “I am not sure, but I think the retry logic is in the wrong place.” Sentences like that only get said out loud in a room where nobody is fishing. Once people know someone in the room pokes for weak spots, they bring finished positions and leave the doubts at their desks. I wrote about the fear version of this in Law 17. Law 39 gets there by a quieter road, and the silence it produces is the same.
The fourth cost is to your calm. Greene’s calm is a loaded weapon, which means it still has an enemy in it. You are as angry as you were, and you have handed the anger to the other person to carry while you keep the aim. I wrote in April about calling your own anger righteous, and this is the other half of that essay. A person who stays level only so the other side will blow up is still running on the fight, and the steadiness disappears the day there is nobody to beat.
The last cost is the list. “Find the chink in their vanity” is an instruction to keep an inventory of the people around you, sorted by where they bruise. Law 33 was about that inventory and what it does to the person keeping it. And the list does not hold. People learn. The colleague you rattled in March has a prepared answer by June, so the next provocation has to go further, and the one after that further again. Now read the Napoleon story the other way. A dressing-down in front of the other ministers was itself an attempt to rattle a man in public. Napoleon was the one stirring, Talleyrand refused to be stirred, and the room watched the man holding the stick lose control.
Under all five sits the cost Plato would put first. Law 39 puts your reason to work for the wish to win. The part of you that should decide what is worth doing spends its day finding where other people bruise, and a mind used that way for long enough gets very good at it and forgets its other job.
Hesiod and Aristotle each supply a word Greene needed.
Hesiod opens Works and Days by correcting himself. He had thought there was one kind of eris, strife. There are two. One is cruel and breeds war and quarrels, and no one loves it. The other is good for people. It wakes even the lazy man to work, because he sees his neighbor getting rich from plowing and planting and wants the same. The potter envies the better potter, Hesiod says, and the builder envies the builder.
That is the correction for Law 39 in a few lines. There is a way to stir the water, and it is aimed at the work: a better version put in front of the team, the pull request that does the migration in one pass, the benchmark posted where people can see it. The engineer who sees it goes back to their desk a little jealous and a lot more awake, and nobody’s judgment got bent on the way. Hesiod’s kind of stirring lands on the craft, which can take it, and leaves the person alone.
The second word is Aristotle’s. In Nicomachean Ethics IV.5 he names a virtue for anger, praotes, good temper. The good-tempered person is angry at the right things, with the right people, in the right way, at the right time, and for the right length of time. He is clear that the deficiency is a vice as well. A person who is never angry at the things that deserve it, who lets their friends be insulted and says nothing, is slavish, he says, and a fool. praotes means your anger answers to your judgment, and so does your calm.
Greene gets this part backwards. Calm earns the name virtue when it keeps your judgment working, and calm held out as bait has none of that in it. The same half hour of standing still can be praotes or a trap, and the difference is whether you were trying to see clearly or trying to make someone else stop seeing.
Then there is the other side of the river, when you are the one being stirred. The Stoics had apatheia, freedom from being carried off by a passion. It means you stop handing your judgment to whoever speaks loudest, and it does not mean you stop feeling. Epictetus puts it in one line in the Enchiridion. When someone provokes you, it is your own opinion that provoked you. The insult only lands if you agree that you were insulted. That cuts both ways. Being provoked does not excuse the answer you gave. What you said while rattled is still yours, and “he set me off” is an explanation you would never accept from someone on your own team.
Seneca has the demonstration. In On Anger he tells how Cato was pleading a case when a man named Lentulus, a known brawler, spat in his face. Cato wiped it off and said, “I will swear to anyone, Lentulus, that people are wrong when they say you have no mouth on you.” The joke is that “having a mouth” meant having nerve. Cato did not counterattack or save anything up for later, and the case went on. That is calm with no hook in it, and it is harder than Talleyrand’s, because Talleyrand had something to win.
Before the practices, the end they serve. You want a room where people’s judgment stays straight, including yours, because the decisions made in it are only as good as the measurements taken there. Every practice below protects the ruler.
Argue the claim, not the pride. Before you post the comment or ask the question, reread it and ask whether it would still land if the other person had no ego at all. If your point needs them rattled to win, it is not winning on the merits, and you will find that out later, in production.
Ask what you lose if they answer calmly. This is the test for when you catch yourself reaching for the needle. If a calm answer would hurt your case, the case is the problem. If a calm answer would not hurt it, you do not need the needle.
Cool the room on purpose. When someone else is heated, restate their point better than they did before you disagree with it. “You are saying the migration has to run in batches because the table is too busy to lock. Is that right?” It takes twenty seconds, it straightens the ruler, and half the time you find out they were right. Your part is the restatement and your own tone. Whether they cool off is theirs, and if they stay hot you have still done your job.
Stir the good strife. If the team needs waking up, do it with the work. Ship the better version or post the number, and let the potter be jealous of the potter. Nobody has to lose their temper for the bar to go up.
When you are the target, name it once. “That sounded aimed at getting a reaction, so here is my answer to the question underneath it.” Then answer it, plainly, and let the heat run out on its own. Naming the move one time takes the fun out of it. Naming it every time turns into a fight of its own.
This duty does not come with rank. The staff engineer who needles juniors in code review is running Law 39. So is the new hire who baits the tech lead in the retro, and the contractor who picks at a client’s pride on the kickoff call. The exec in board prep has more reach, and the move and the cost are the same at every level.
One more, from my own desk. When a thread has gone hot and I am about to reply, I paste the thread into Claude and ask it one question: what is the actual claim each person is making? The model does not take the bait, because it has no pride to hit, and it hands back the argument with the heat taken out. Most of the time the two sides are closer than the thread made them look, and I write a better reply. Start with the model there too, and use it to clear the water before you step in.
Hesiod’s two strifes are the oldest version of the argument, and still the cleanest. Strife is going to happen in any group that wants to be good at something. The only choice is which kind you feed. Feed the kind that makes the potter go back to his wheel, and starve the kind that makes him go after the other potter.
Plato gives the reason provocation is so effective and so costly. In Republic IV thumos sides with reason against appetite, and it is at its best when it is angry at something that is really wrong. A person whose thumos has been set off by a planted slight is using their best part on your behalf. You have borrowed their sense of justice to win your argument, and it takes a while to come back.
Aristotle gives both halves, and they fit together. The Rhetoric shows how anger is made, and he allows a speaker to show an audience that a real slight is a slight. What the first chapter forbids is stirring passion about things outside the case, which bends the judge like a ruler. Anger that fits the facts is allowed, and anger manufactured to stand in for the facts is exactly what Law 39 recommends. The Ethics gives the virtue, praotes, anger that answers to judgment, and says the person with no anger at all has a flaw of their own.
Seneca goes furthest. On Anger calls anger a brief madness and argues that no one should ever use it as a tool, in themselves or in anyone else. He had watched Caligula up close and would go on to advise Nero. He knew what it looked like when a powerful man kept a list of other people’s weak spots and pressed them for sport, and he did not think the man pressing was the winner.
Epictetus supplies the defense. You cannot stop anyone from stirring, and you can still refuse to let them muddy your judgment. His line in the Enchiridion is the one to keep for the next pull request comment written to get a rise: nobody provokes you without your own judgment going along.
Think about the last argument you won at work. Did you win it, or did the other person lose their temper?
If it was the second, go back and find out what they would have said if they had stayed calm. Ask them. That answer is the information you traded for the win, and you may not like how much it was worth.
Then look at the list you are carrying. Do you know which of your colleagues bristle at being called junior, which ones hate being interrupted, which ones get defensive about their estimates? If you know it because you have used it, Law 39 has been running on you for a while.
Four questions to run before the next heated meeting.
When I ask a hard question, am I trying to learn the answer or trying to see them flinch?
What was the last good idea I heard from someone who was angry at the time? If I cannot think of one, what am I doing to the rooms I am in?
Where does my team compete over the work, and where do they compete over each other? Which of those did I stir last?
And the one that decides: if everyone I work with stayed perfectly calm for a month, would I win more arguments, or fewer?
Greene is right that anger makes people decide badly, and right that the calm person has the edge. He is wrong about what to do with the edge. Using it to rattle someone else wastes the one thing calm is for, which is seeing clearly, and it wastes it for everyone in the river, including you.
Aristotle had the better use for calm. Keep your judgment straight, be angry when something deserves it and for exactly as long as it deserves it, and never bend someone else’s ruler to win a measurement. Hesiod had the better use for strife. Aim it at the work, where it makes the potter better, and never at the potter.
Muddy water hides the net from the fish for about a minute. Then the fisherman is standing in brown water with the rest of them, and the river he needed to read is the one he wrecked.
Law 40 is next, and it is about the free lunch.
Keeping your judgment straight in a hot room takes practice. That is what MasteryLab is for.
]]>You say, “This is a really solid start. I’d think a little more about the failure cases. Let’s keep going.” They leave the room feeling good about the work. You leave feeling generous, which is what soft feedback feels like from the giving end.
Three weeks later they present the finished design to a wider group, and someone who has never tried to be kind to them asks what happens when the provider times out. The room goes quiet. The engineer spends the next week rebuilding what they spent the previous three weeks building, and a long time after that wondering why you did not say anything.
You did say something. You said it softly enough that it did not land, and the softness was for you.
Soft feedback is any message where the verdict is present in your head and absent from the sentence. It feels like kindness in the room because the room is where you are. The other person pays for it later, in weeks of work on the wrong thing, in a public correction they could have had in private, and in the slow discovery that your praise cannot be trusted. That bill goes to them, and you never see it.
Most managers I know believe hard news needs a cushion. The instinct is fear of one uncomfortable minute, and the explanation about protecting people gets written afterward. Nobody wants to be the manager who crushes people, so the verdict gets tucked between two compliments, “this will not work” becomes “I have a few thoughts,” and the actual problem gets demoted to a suggestion at the bottom of an email.
The instinct is aimed at the wrong person’s pain. When you soften a message, the discomfort you remove is your own. Saying “this design will double-charge customers” is unpleasant for the person saying it and useful for the person hearing it. Softening trades their information for your comfort, and it does the trade in a way that looks like generosity from where you sit. There are two things the word kind can mean here. One is the pleasant feeling in the room. The other is wanting the other person’s good enough to spend your own comfort on it. Softening delivers the first and skips the second, because the second costs a minute.
Brené Brown puts it in four words in Dare to Lead: clear is kind, unclear is unkind. I would add the clause she leaves implied. Unclear is unkind to them and pleasant for you, which is why it survives every resolution to stop.
I have written before about how nice leaders ruin people over the span of a career. This is about the same habit measured in weeks, at the scale of a single review, where you can still catch it.
The first sign that your kindness did not land is that nothing changes. You said “think more about the failure cases,” and the next version has the same failure cases. You said “I’d tighten the scope,” and the scope came back the same size with a new coat of paint.
The tempting explanation is that they did not listen. Watch more closely and you will usually find they listened carefully. They heard “solid start” and believed it, because you meant them to. They heard “think a little more about” and filed it as a polish item, because that is what “a little more” means in English. They went and did what you said. What you said was not what you meant. An engineer who hears a vague review can ask whether it was a blocker, and the good ones learn to. That does not move the debt. You had the verdict, they did not, and asking is a workaround for a sentence you owed them.
This is where the cost compounds. Every day between your soft comment and the real correction is a day they spend building on a foundation you already knew was cracked, working hard and in good faith on the thing you told them was solid. When the correction finally arrives, they are handed a bill for a test that was never on the page, and the person who wrote the test is the one who told them they were passing.
Psychologists have a name for the gap. Thomas Gilovich, Kenneth Savitsky, and Victoria Medvec called it the illusion of transparency: people believe their internal states leak out far more than they do. You know how worried you are about the retry logic, so “think about the failure cases” feels heavy with that worry when it leaves your mouth. To the listener, it is seven neutral words.
The speaker walks out believing the warning was delivered, and the listener walks out believing the work was approved. Both are being honest, and one of them is about to lose three weeks.
A second gap opens later. Belief is what happens while the relationship is new, and discounting is what the team learns once the pattern is clear, and the information fails to arrive either way. People adjust what they hear for who is saying it. If you have a habit of softening, your team has already learned to discount you, and “a few thoughts” from you carries less weight than “a few thoughts” from someone who says what they mean. Your soft feedback lands even softer than you intended, because they have calibrated to you. Then comes the day you mean “this is fine,” and it sounds the same as every other time, and they cannot tell the difference. You have spent your ability to reassure them. The hours are the visible cost, and they can be made up. A team that has no way to know the truth about its own work, because the one person positioned to tell them speaks in one register no matter what they see, has lost something a refund of three weeks would not restore.
Stop asking whether the feedback was gentle enough, and start asking who the gentleness was for.
aletheia, the Greek word for truth, carries the whole argument. It means un-hiddenness, the state of something pulled out of concealment into the open. It is built from a- (not) and lethe (concealment, forgetting), the same root as the river the dead drank from to forget. Truth, in this picture, is something you do to information: you bring it out. Whatever you do not bring out stays in the fog, and the fog does not clear on its own. Someone eventually walks through it and finds the wall.
Soft feedback keeps the truth in the fog. You know the design double-charges customers, and the engineer does not. Between those two facts sits a decision you made, which was to keep the knowledge on your side of the table, where it costs you nothing and costs them everything.
Seneca opened his letters to Lucilius with a complaint about time. People guard their money and hand out their hours to anyone who asks, he wrote, because time does not look like anything. Nobody sends an invoice for it. When you soften a verdict, you are spending someone else’s hours, and there is no invoice for that either. Three weeks of an engineer’s work disappears, and it never shows up as a cost on your side of the ledger. It shows up as a hard month on theirs.
Plutarch, writing about how to tell a flatterer from a friend, put the damage in the same place. The harm the flatterer does is that the friend finds out too late to fix anything.
So the revelation is arithmetic. Softening saves you one uncomfortable minute. It costs them the days between that minute and the moment the truth arrives from somewhere less kind. Multiply that across every review and every draft you called a great start, and you get a team that is chronically late and cannot work out why, reporting to a manager everybody likes.
Once you see who pays, the fix is mechanical.
Lead with the verdict. The first sentence carries the conclusion: “This design will double-charge customers when the provider times out, and the retry logic has to move.” Then the reason, then what to do next. Compliments, if they are true, come after, where they cannot be mistaken for the message.
Run the surprise test. Before you close a review, ask yourself one question. If this person gets a hard no from someone else in three weeks, will they be surprised? If the answer is yes, you have not told them. Whatever you said did not count.
Name the size. “Small” and “big” are the two most useful words in feedback and the two most often left out. “This is a big one, it changes the middle of the design” gives them a budget. “This is small, ten minutes” gives them a different one. Leave the size out and they will guess small, because you sounded calm.
Write it down the same day. The illusion of transparency lives in spoken feedback. A written sentence with the verdict in it survives the walk back to the desk, and it is the sentence that will be repeated when you are not in the room.
Check what they heard. “Tell me what you’re going to change” costs thirty seconds and catches the gap while it is still free.
You control whether the verdict, the reason, and the size were in the sentence. You do not control how it feels to hear it, and how it feels was never the test of whether you did your job. A face that falls means the sentence arrived. The other way to fail the person is contempt, and a verdict delivered with scorn or a sigh teaches them about you instead of about the work. The target is the plain verdict with the reason attached, without a cushion and without an edge.
I learned the shape of this from working with agents all day. When Claude hands me a design and the retry logic is in the wrong layer, I do not type “solid start, think about the failure cases.” I type: the retry logic is in the wrong layer, move it to the payment client, here is why. The fix comes back in a minute. Vague feedback to a model wastes my own afternoon, so I stopped giving it within a week of starting. It took me far longer to notice I was still giving it to people, and that with people, the afternoon I was wasting belonged to someone else. The model made the cost visible by billing me directly, and people are polite enough to eat it.
The challenge for this week: for your next five reviews, write the verdict sentence before you walk in, say it first, and end with “tell me what you’re going to change.” Count how many times what they say back matches what you meant. Every miss you catch in that thirty seconds is work they would otherwise have done on the wrong thing.
The comfortable version of feedback is comfortable for one person in the room, and that person already knows the answer. The rest of the room is about to pay for the fog with their time.
A manager who says the hard sentence first, with the reason and the size attached, spends one bad minute and saves the other person weeks. That is the whole trade. It looks harsh from inside the minute and, from three weeks out, like the kindest thing anyone did for them all quarter.
If you want to work with people who say the verdict first and give you the reason, MasteryLab is built for that.
Vague feedback fails because the listener decodes it as approval. When a reviewer says “solid start, think a little more about the failure cases,” the speaker believes a warning was delivered and the listener believes the work was accepted with a polish note. The listener then spends days or weeks building on a foundation the reviewer already knew was wrong, and finds out later, usually from someone less kind and in a bigger room. The harm is measured in the listener’s time, plus the loss of trust when they realize the reviewer knew all along.
Soft feedback is comfortable for the person giving it and expensive for the person receiving it. Softening removes the speaker’s discomfort in the moment, which is why it feels like kindness from the speaker’s chair. The listener loses the information they needed, keeps working on the wrong thing, and pays for the softness later. Kindness in feedback means giving the other person the truth while it is still cheap for them to act on, even though the minute of saying it is unpleasant for you.
Lead with the verdict in one plain sentence, then the reason, then what to do next. Name the size of the problem so the person knows whether they are looking at ten minutes or a rebuild. Write the verdict down the same day, because spoken feedback is where the gap between what you meant and what they heard lives. Finish by asking them to tell you what they plan to change, which catches the misunderstanding while it is still free. Cruelty comes from contempt and tone, and a clear statement about the work with the reason attached is a form of respect.
aletheia is the ancient Greek word for truth, built from a- (not) and lethe (concealment or forgetting). It means un-hiddenness, the state of something being pulled out of concealment into the open. In feedback, aletheia is the act of bringing what you know about someone’s work out where they can use it. Soft feedback keeps that knowledge concealed on the speaker’s side of the table, where it costs the speaker nothing and costs the listener the time between the soft comment and the day the truth arrives from somewhere else.
]]>