In 1973, the United States Army adopted a habit that had nothing to do with weapons and everything to do with memory. After every significant training event, the people who had just lived through it would sit down...
In 1973, the United States Army adopted a habit that had nothing to do with weapons and everything to do with memory. After every significant training event, the people who had just lived through it would sit down together, while the details were still fresh, and answer a fixed set of questions about what had happened. Not a lecture from the commander. Not a hunt for the soldier who got it wrong. A structured conversation, with rules, whose only permitted purpose was learning.
The Army called it the After Action Review, and it has now run continuously for more than fifty years, which makes it one of the longest-lived deliberate learning rituals in any modern institution. The version of it this article works from comes through an unlikely messenger: Ronald J. Baker, a California accountant who spent his career advising professional firms, and who in 2008 published Mind Over Matter: Why Intellectual Capital is the Chief Source of Wealth (Wiley), a book arguing that the real wealth of any organization is what its people know, and that almost every organization lets that wealth evaporate because it never built a mechanism to capture it. Baker surveys storytelling, mentoring, communities of practice. Then, in Chapter 12, he lands on the AAR and stays there, because it is the one mechanism in his entire book that is concrete, repeatable, and teachable in an afternoon.
Baker wrote for accounting practices and ad agencies. He never once, in roughly four hundred pages, discusses a household. That gap is worth stating plainly at the start, because this article is going to walk his mechanism across it anyway, and honesty requires marking where his text ends and our translation begins. Here is the claim we will defend: of everything in the knowledge-capture literature, the After Action Review is the single mechanism that survives the trip from a firm into a family almost without modification, and a family that runs a twenty-minute AAR after its significant events, good and bad, converts lived experience into transmissible knowledge instead of losing it the way nearly every family does.
Readers of this blog have met the blame-free review before, as a sentence of advice inside pieces about family meetings and minutes. Those sentences were pointers. This piece is the machine they were pointing at.
Baker introduces the AAR by quoting Sullivan and Harper's Hope Is Not a Method, a book by soldiers about how the Army actually learns:
"An AAR takes place after every training event. Its purposes are simple: learning, improving, doing better the next time... By asking questions such as 'What did you think I wanted you to do?' (as opposed to questions such as 'How did you screw that up?'), one can get to the roots of both success and failure." (Ch. 12, pp. 203-204)
Stop on that contrast of questions, because the entire ethic of the tool is inside it. "How did you screw that up?" assumes the plan was clear and the person failed it. "What did you think I wanted you to do?" allows for the possibility that the instruction itself was the failure, that the person at the top of the hierarchy gave an order that landed differently than it left. An institution built on rank designed its learning ritual around the leader's willingness to ask whether the misunderstanding started with him. Any parent who has ever shouted instructions across a compound and been amazed at the result already knows why that question matters more than the other one.
The review itself, as Baker reports it from the Centre for Army Lessons Learned, is four questions asked in order:
"What was supposed to happen? What actually happened (the 'ground truth')? What were the positive and negative factors here? What have we learned and how can we do better next time?" (Ch. 12, p. 206)
Four questions. No forms, no facilitator certification, no budget. But each one does specific work, and the work is different from what an ordinary family postmortem does.
The first question, what was supposed to happen, is the quiet ambush. Families, like firms, routinely discover at this step that there was never a shared plan at all. Everyone walked into the wedding, the move, the school term, or the business deal carrying a private picture of what success looked like, and the pictures did not match. Half of what families experience as betrayal is really this: two unstated plans colliding. The question surfaces the collision without accusing anyone of causing it.
The second question, what actually happened, is where the Army's phrase "ground truth" earns its keep. Not what each person felt happened, not the version already hardening into family legend, but the sequence of events as the people in the room can jointly establish it. In a family this step has a discipline all its own: the ground truth is assembled from everyone's account, including the youngest person in the room, because the person with the least status often saw the most.
The third question, the positive and negative factors, forces balance. Notice that it demands the positives. An ordinary family inquest convenes only when something went wrong and looks only at the wrongness. The AAR examines what went right with the same seriousness, because a success you cannot explain is a success you cannot repeat.
The fourth question, what have we learned and how do we do better next time, is where the whole exercise has been heading, and the Army protects it with arithmetic, which is the next section.
Everything above collapses without one rule, and Baker states it as the heart of the tool:
"The AAR is not a tool to fix blame or point fingers; it is a learning tool, designed to increase the effectiveness of the firm. Its role is to be analytical, not critical." (Ch. 12, p. 207)
Analytical, not critical. The reason is not kindness. It is engineering. The AAR runs on accurate information, and the moment a review becomes a trial, accurate information stops flowing. People are not fools: if admitting "I misunderstood the plan" or "I sent the money to the wrong place" will be used against them, they will not admit it, and the review will proceed on fiction. Blame does not merely make the meeting unpleasant. It corrupts the data the meeting exists to gather.
The Army backs the rule with a second, structural one: "AARs should not be used for promotions, salary increases, or performance appraisals" (Ch. 12, p. 206). What is said in the review has no career consequences, by design, because the Army understood that candor and consequences cannot share a room.
Now walk this into a household, and here the translation is ours, not Baker's. A firm that botches this rule loses candor. A family that botches it loses something worse, because a family has no exits. In a firm, a person whose honesty was punished can eventually leave. In a family, the brother whose admission was thrown back at him at a graduation two years later does not resign. He stays, and he simply never tells the family the truth again. The people around your table will be around your table for decades, which means every review you run is also training the next one. Run one blaming review and you may have purchased twenty years of polite silence.
So the family version of the Army's two rules looks like this. First, the ground rule is said out loud, every time, before the first question: we are here to learn, not to blame. Saying it aloud feels ceremonial and slightly awkward, and that is the point; a rule everyone privately assumes is a rule nobody can invoke, while a rule spoken at the start can be pointed to the moment the meeting drifts toward fault. Second, the family equivalent of the no-appraisals rule: nothing said in the review may be used as ammunition in any later argument. What is admitted in the review stays in the review. The first time someone breaks that rule and is corrected for it, by the elder in the room if possible, the family will have taught itself more about trust than any lecture could.
One more reason the rule matters more at home than at work, and it is the quiet spine of this whole piece. A firm can, at its worst, resolve a problem by removing a person. A family cannot, and should never want to. When the people are permanent, blame has no function at all. There is no one to dismiss, so a verdict changes nothing. The only thing that can actually change in a family is the plan, the habit, the system. Learn, do not blame is not a softer option for households. It is the only option that does anything.
The Army prescribes not just the questions but the clock:
"The Army suggests that you divide your time in answering the AAR's questions into 25-25-50: That is, 25 percent reviewing what happened, 25 percent reviewing why it happened, and the remaining 50 percent on what to do about it and how can you learn from it to improve." (Ch. 12, p. 207)
Half the meeting, by rule, faces forward. This is the split families miss most reliably, and everyone who has sat through a family crisis meeting knows the failure mode: ninety percent of the time is spent re-narrating the event, each person polishing their own version, and the meeting ends when everyone is exhausted rather than when anything has been decided. The retelling feels like work. It is not work. It is the comfortable quarter of the exercise dressed up as the whole.
For a family, the arithmetic lands like this: in a twenty-minute review, five minutes on what happened, five on why, ten on what changes next. Someone watches the time. When the retelling runs long, the timekeeper does not argue with anyone's version; they simply move the meeting to the question the versions were postponing. The discipline sounds trivial. It is the difference between a family that processes events and a family that merely relives them.
And the output of the second half must be concrete: one or two changes, named, with a person attached. Next school term, the fees are confirmed two weeks before the deadline, and this daughter holds the reminder. Next harvest, the buyer is agreed before the crop is cut, not after. Next hospital stay, one person is the family's single point of contact with the doctors. A review that ends with "we must communicate better" has failed its fourth question. A review that ends with one sentence someone can act on has done everything the Army designed it to do.
Why does a household need a formal ritual for something as natural as talking about what happened? Baker's answer, built for firms, transfers to families with uncomfortable ease: because knowledge that is not deliberately captured does not linger. It disappears.
His framework rests on the philosopher Michael Polanyi's observation that "we know more than we can tell" (Ch. 9, p. 118). Most of what a person knows is tacit: it lives in their head, unwritten and mostly unspoken, and it moves to another person only through deliberate acts of telling, showing, or writing. Firms are catastrophically bad at this. Baker cites the consultant Betty Zucker's estimate that "as little as 20 percent of the knowledge in a company is actually ever used" (Ch. 12, p. 190), and then reaches for the most expensive example on record, a NASA manager reflecting on the Apollo program:
"If we want to go to the moon again, we'll be starting from scratch because all of that knowledge has disappeared. It would take at least as long and cost at least as much to go back." (Ch. 12, p. 188)
The richest agency of the richest country in history put people on the moon and then let the knowledge of how it was done dissolve, not through fire or sabotage but through the ordinary failure to capture what its people knew while they knew it. Baker reports that the AAR habit spread from the Army into companies such as Boeing, Toyota, and Microsoft precisely because it is the rare mechanism that harvests tacit knowledge on a schedule, while the event is fresh, instead of trusting memory to hold it.
Now consider a family's position, and from here to the end of this section the application is our translation, not Baker's. A firm at least has personnel files, procedure manuals, and overlapping staff who might reconstruct what was lost. A family archive is usually one or two heads. How a bride price negotiation was actually conducted, and why it succeeded. Why the land boundary was drawn where it was, and what was conceded to get it. How a debt inside the extended family was settled without breaking the relationship. Why this guardian was chosen for the children and not that one. In a family economy that runs on land, relationships, and mobile money rather than on documents, this is not sentimental heritage. It is operational knowledge, and when it is lost the next generation pays for it in real money and real disputes, relitigating boundaries and renegotiating settlements their parents had already paid to resolve.
The AAR attacks this loss at its cheapest point: the week after the event, when the knowledge is fresh, complete, and sitting in the heads of people already gathered in one place. Twenty minutes of structured questions, and what one person learned becomes what the family knows.
Two refinements from Baker's chapter complete the tool, and both matter more in a family than he could have known.
The first: do not reserve reviews for disasters. Baker tells the story of his colleague Paul O'Byrne, who ran AARs in a twelve-person accounting practice in London and found the habit only stuck once the firm required a success AAR alongside every failure AAR (Ch. 12, pp. 208-209). O'Byrne was a colleague at Baker's own institute, so treat the story as an illustration rather than independent proof, but the logic stands on its own. If reviews happen only after failures, then being invited to one is an accusation, and the ritual dies of dread. A family that reviews the wedding that went beautifully, the harvest that beat expectations, the business quarter that finally turned, teaches itself two things at once: exactly what produced the success, so it can be repeated on purpose, and that a review is a normal act of a learning family rather than a tribunal.
The second refinement: the review has a twin that runs before the event. Baker reports the Before Action Review, four questions drawn from Darling, Parry, and Moore in the Harvard Business Review: "What are our intended results and measures? What challenges can we anticipate? What have we or others learned from similar situations? What will make us successful this time?" (Ch. 12, p. 208). For a family, the third question is the payoff, because it is the moment the last AAR gets used. Before this year's harvest, before the next school term, before the shop expands: what did we learn the last time we did something like this? A family that asks that question is no longer merely storing its experience. It is compounding it.
One caution belongs here, and it is ours. Borrow the Army's habit, not its aura. Nothing in a family should aspire to military command, and the AAR imported with a commander's tone will fail exactly the way a blaming review fails. The instructive fact about the Army is historical and runs in the opposite direction: the most hierarchical institution in American life concluded, and has maintained since 1973, that when the goal is learning, the hunt for fault has to be switched off. If candor needed protecting there, it needs protecting at your table.
This month, run your family's first After Action Review.
Do not wait for a crisis; a good event works better for a first run. Pick one significant thing that has just happened or is about to conclude: a school term, a harvest, a family function, a business decision, a hospital stay that is now safely past. Open Family Council in LegacyPot and put a twenty-minute review on the agenda, then write the structure into the agenda item so the meeting cannot drift. At the top, the ground rule, to be read aloud before anything else: we are here to learn, not to blame. Then the four questions in order: what was supposed to happen, what actually happened, what were the good and bad factors, what will we do differently next time. Then the clock: five minutes, five minutes, ten minutes, with the second half protected for what changes next, and name a timekeeper who is not the eldest person in the room.
Close by recording the output in Family Council: one or two changes, each with a name attached, and the date of the next event they apply to. If the review surfaced something worth keeping longer than a season, the story of how the negotiation was actually done, the reason behind the choice, move it into the Wisdom Library so it outlives the meeting.
The Army has asked its four questions after every significant event for over fifty years, which is longer than most companies live and about two generations of a family. That is the standard worth borrowing: not a heroic one-time conversation, but a small ritual, run on schedule, that quietly refuses to let the family's experience die with the moment that produced it. Your family is already paying for its lessons. The review is how you keep what you paid for.