Latest / Soft Ratios Research / The prepared mind saw nothing (on scenario planning)

Transcript

The prepared mind saw nothing (on scenario planning) · Soft Ratios Research

• 0:09
  Figure: Soft Ratios Research.
• 0:12
  Ground: The voices are synthetic [pause] no act, just the thinking.
• 0:19
  Figure: Hoyd Breton, a designer at Soft Ratios Studio, teaching himself in the open [pause] which works better with a reply. [pause] If this is interesting, write a review with your thoughts.
• 0:31
  Ground: One subject at a time, twice a week [pause] Tuesday and Thursday. [pause] Today, scenario planning.
• 0:37
  Figure: Here's the story you've heard, even if you've never heard the name. October 1973. The Yom Kippur War, then the embargo. Oil goes from under three dollars a barrel to nearly twelve inside a few months — two dollars ninety to eleven sixty-five by that January. Every oil major gets caught flat. Except one. Shell. And the reason Shell wasn't caught, the story goes, is that a small group in London had already imagined it. They'd written the crisis down before it happened.
• 1:09
  Ground: The prepared mind.
• 1:11
  Figure: That's the phrase. While the competition scrambled, Shell moved — and across the seventies it climbs from something like the weakest of the big players toward the front rank. And the tool that supposedly did it has a name that sounds like nothing at all: scenario planning.
• 1:28
  Ground: I want to sit with the appeal of that, because it's strong. It's the fantasy of seeing it coming. Not luck — foresight. Somebody in a quiet room drew the map of a future nobody else would look at, and when the future arrived they already had the terrain memorized. [pause] And it flatters something in us. We'd all like to believe the disaster was legible in advance, if only someone had been brave enough to read it. It makes the world feel less like weather and more like a book with the last chapter already printed.
• 2:03
  Figure: Keep that feeling. Because almost none of it survives contact with the record. [pause] Start with the man who built the scenarios. Pierre Wack — an economist from Alsace who told people to pronounce it to rhyme with Jacques. He kept incense burning in his London office. He followed the mystic Gurdjieff, and he took long pilgrimages to a guru in India.
• 2:25
  Ground: That is not the consultant I was picturing.
• 2:28
  Figure: Not remotely. His partner, Ted Newland, was the opposite — a hard-eyed geopolitical planner with no sentiment in him. And by 1971 the two of them had noticed a set of forces the industry was ignoring. OPEC producers gaining nerve. Libya had just shown that a single country could throttle supply and get away with it. America had tipped into being a net importer. Demand climbing, spare capacity thinning.
• 2:55
  Ground: So they did see it.
• 2:56
  Figure: They saw the pressure building. In September 1972 Wack stood up and warned of a — his words — a major discontinuity. He drew the futures as a river breaking into a delta, branching. Then the war came, the embargo came, the price quadrupled. Which sounds like the prediction landing.
• 3:17
  Ground: But.
• 3:18
  Figure: But his timing was wrong. Wack expected the break around a 1975 OPEC meeting — roughly two years early. And here's the part the legend can't absorb. Shell's own scenario veterans have said, in print, that the prediction never happened the way we tell it. Angela Wilkinson spent a decade on that team, and with Roland Kupers, in 2013, she wrote it flat out: we have no solid examples of Shell's having anticipated future developments better than other companies — the mythology around anticipation of the 1970s oil crises notwithstanding. Kupers, elsewhere, calls it a persistent myth.
• 3:58
  Ground: So the founding story of the whole discipline is a good story about the danger of good stories. [pause] But then I don't understand what Shell actually got. If Wack didn't call the date, if the competition eventually paid the same price for the same oil — what exactly did the prepared mind prepare? Because we imagined a crisis two years early and got the timing wrong is not obviously an advantage.
• 4:23
  Figure: The honest answer's smaller and stranger than prediction. What Wack changed wasn't the future. It was the managers. Because his first scenarios failed — completely. He presented them, the executives nodded along, and then they went back to their desks and did nothing at all. Alternative projections washed right off them. [pause] And Wack realized the problem wasn't the quality of the forecast. It was that his stories never reached what he called the manager's microcosm — the private little world each executive uses to make sense of things. So he rebuilt them. Not to inform the managers, but to restructure how they saw. Second-generation decision scenarios, aimed straight at the microcosm. [pause] And when the shock came, Shell's people didn't have a better forecast in a drawer. They had already lived, in imagination, in a world where cheap oil was over. So they moved faster. They believed it sooner. The whole advantage was reaction time, not the crystal ball.
• 5:26
  Ground: That I understand in the body. It's rehearsal. It's what an actor does, or a pilot in a simulator — or the darker version, the way you lie awake and run a hard conversation before you ever have it. You're not predicting what the other person will say. You couldn't. But when the real thing comes, some part of you has been in the room before. Your hands already know where the door is. [pause] And the strange gift of that is speed. When you've rehearsed the bad news, you don't lose the first crucial hour to disbelief — and disbelief is the expensive part. Everyone else is standing in the wreckage asking how this could possibly be happening, and you're already three moves in, because for you it was never unthinkable. You thought it. [pause] That's what a scenario buys. Not the map. The nervous system that doesn't freeze.
• 6:23
  Figure: And there's a name for that machinery underneath, and it predates Shell's scenarios by decades. The idea that a mind carries little models it runs ahead of reality — that isn't oil-company jargon. It's a Scottish psychologist, Kenneth Craik, 1943, a book called The Nature of Explanation. He wrote it just before digital computers, so his models are mechanical. And his claim was this: if an organism carries a small-scale model of external reality inside its head, it can try out various alternatives, conclude which is best, and react in a much fuller, safer, and more competent manner.
• 7:03
  Ground: So the brain was already doing this. Wack didn't invent it.
• 7:07
  Figure: Wack industrialized it. He took the private simulation each of us runs badly and alone, and made it deliberate, external, shared. Reperceiving, he called it. The gentle art of reperceiving.
• 7:20
  Ground: Where does the word even come from — scenario? It sounds like screenwriting.
• 7:26
  Figure: It is screenwriting. The word is stolen from Hollywood — a scenario was a script outline. And the man who stole it was not gentle at all. Herman Kahn, a physicist at RAND in the early Cold War, used scenarios to think about thermonuclear war. Escalation ladders. Megadeaths. A Doomsday Machine. He wrote reports as if they were dispatches from people already living in the future.
• 7:53
  Ground: Thinking about the unthinkable.
• 7:55
  Figure: That's literally the title of his 1962 book. Two years after On Thermonuclear War. He was so willing to game out the end of the world that he became a partial model for Dr. Strangelove — Kubrick even named the film's BLAND Corporation as a jab at RAND. And his creed is the one thing that survived all the way down to Wack's incense: you can't predict the future, but you can prepare for it by imagining it vividly — especially the branches everyone refuses to look at.
• 8:27
  Ground: So the discipline is born twice. Once in a war room imagining annihilation, once in an oil company imagining a price. Both times the point is the same — get the unthinkable thought before it arrives.
• 8:43
  Figure: And RAND had a rival method growing right beside it that shows you the fork in the road. At RAND, Dalkey and Helmer built the Delphi method — published 1963, but first run in secret in 1951. Seven experts, polled anonymously, round after round, asked to reason as a Soviet planner and estimate how many atomic bombs it would take to cut U.S. munitions output by a set amount. First round, the guesses ran from about fifty to five thousand. By the fifth round they'd converged — a final median of two hundred seventy-six.
• 9:21
  Ground: That's the opposite instinct. Delphi refines a crowd's estimates — often toward one number, sometimes toward a clearer map of where they still disagree. A scenario opens one head up into four worlds.
• 9:36
  Figure: That's the fork in the road. One discipline exists to narrow — to pull a crowd toward a shared estimate, or at least a clear map of where they still disagree. The other exists to widen — to refuse a single answer for as long as possible. And you can feel the temptation in scenario work to secretly do Delphi. To collapse the four worlds back into the one you think is coming.
• 9:59
  Ground: Which would be cheating at your own game.
• 10:01
  Figure: The whole game. Now — the method itself is surprisingly plain. You start not with the world but with a real decision, a focal question. You list the driving forces behind it — social, technological, economic, environmental, political. You sort them on two axes: how important, how uncertain. The ones that are important but basically certain, you call predetermined — they're the same in every future. The ones important and genuinely uncertain become your engine. You take the two biggest, most independent uncertainties, cross them into a two-by-two, get four worlds, and write each as a real, internally consistent story with a name. Then you carry your decision back into each and ask how it fares.
• 10:47
  Ground: I notice you laid that out like a recipe, and I don't trust recipes for imagination.
• 10:54
  Figure: You shouldn't, and neither did Wack. He held that two-by-two in near contempt. His words: simply combining obvious uncertainties did not help much with decision making — it brought us only to a set of obvious, simplistic, and conflicting strategic solutions. He called the matrix no more than a staging post.
• 11:13
  Ground: A staging post to what?
• 11:15
  Figure: To the harder thing — finding the predetermined elements. The certainties hiding inside what everyone treats as uncertain. Anyone can cross two axes and generate four bland quadrants. The craft is noticing what has to be true in every single one of them. Libya had already proven a single producer could squeeze supply. That wasn't a scenario — that was a fact, sitting in plain view, that nobody had let into their microcosm.
• 11:43
  Ground: So the four stories are almost a device to smuggle the one uncomfortable certainty past your own defenses.
• 11:52
  Figure: Close. And if you want the machine for generating the worlds themselves, it predates all of them too — Fritz Zwicky, the astrophysicist, in the forties. Morphological analysis. You break a problem into its independent parameters, list every value each one can take, and lay them in a grid. His famous case was rocket propulsion — six parameters, five hundred seventy-six combinations, and the box surfaced engine types nobody had thought of, because they'd only ever imagined the handful already built.
• 12:26
  Ground: The grid imagines what you're too anchored to imagine. That's the thread through all of it, isn't it — Kahn, Zwicky, Wack. Every one of them is building a device to outwit the poverty of a single mind.
• 12:40
  Figure: And that's the deepest problem in the whole field, the one that never fully solves. Kees van der Heijden, thirty-five years at Shell, said scenarios are really the language of a strategic conversation — and a conversation needs difference in the room. Too much agreement and you get groupthink, one official future everyone salutes. Too much difference and you get fragmentation, no shared action at all. The whole art is holding those apart.
• 13:07
  Ground: Which is a lovely problem to have when you've got a room full of people. It's a brutal one when there's only one head. You can't argue with yourself and mean it. You already know how the argument ends. [pause] That's the loneliness of imagining alone — you keep hiring your own opinions and calling them a debate. The workshop version has a room full of people with genuinely different interests, who don't share a mental model. The solo version has one person nodding at their own reflection. And the fixes are almost embarrassingly deliberate — you have to formally play the pessimist, the regulator, the customer, in turn, in writing, or run a pre-mortem where you decree the thing already failed and you're writing the autopsy. [pause] It's 2031, this collapsed, tell the story of why. You're manufacturing the dissent a real room gives you for free.
• 13:58
  Figure: Let me give you the case that's hardest to wave away, because it isn't about oil at all. [pause] South Africa, 1991. The fragile window between Mandela's release and the first free elections, when everything could still fall apart. Adam Kahane — he'd been running Shell's global political scenario team — gets invited to facilitate an exercise at a conference center called Mont Fleur, outside Cape Town.
• 14:24
  Ground: Who's in the room?
• 14:25
  Figure: Twenty-two South Africans who had every reason to hate each other. ANC radicals, establishment business figures, academics, trade unionists — black and white, left and right. Three workshops, three days each. And they built four scenarios as a fable. Ostrich — a non-representative government that buries its head and refuses to see. Lame Duck — a weak, hobbled transition that satisfies nobody. Icarus — a popular government that flies too high, spends recklessly on a populist program with money it doesn't have, and crashes the economy. And Flight of the Flamingos — everyone rising slowly, together.
• 15:07
  Ground: Let me guess. Icarus is the one that stung.
• 15:10
  Figure: Icarus stung, and it stung the left specifically. Because it was the ANC's own economists being warned — by a story they'd helped build — that a spending spree in the name of justice could burn down the house it was trying to furnish. Two men in that room, Trevor Manuel and Tito Mboweni, went on to become the country's finance minister and its Reserve Bank governor. And years later Mboweni, running the Reserve Bank, stood in front of international bankers and said — quoting — we are not Icarus; there is no need to fear that we will fly too close to the sun.
• 15:46
  Ground: That gives me chills, honestly. A story invented in a room becomes the language a Reserve Bank governor uses to steer a nation off the cliff. That's the dream. That's scenario planning changing the world.
• 16:00
  Figure: Slow down — this is where I have to be the ungenerous one. That telling is mostly told by the people who ran the workshop. And there's a colder reading, where Mont Fleur didn't so much save South Africa as launder a decision already being made. The ANC's turn away from nationalization was also being shaped by Mandela and Mboweni sitting with Chinese and Vietnamese delegations who were busy privatizing. By 1996 the government adopts a strict fiscal-discipline program — GEAR — that a journalist called the Great U-Turn. And one scholarly account of the whole arc isn't titled the miracle of dialogue. It's called From Comrades to Capitalists.
• 16:44
  Ground: So the flamingos were also the sound of the left being talked out of redistribution.
• 16:50
  Figure: Maybe. I can't prove the scenarios caused the turn, and I won't pretend the workshop was the sole engine — the honest version is that it helped incubate a reorientation that had other, harder causes. But notice what just happened to the two of us. We got a vivid four-part fable, a chills-down-the-spine quote, and we both leaned toward believing the story did the work. That pull we just felt —
• 17:16
  Ground: — is the trap.
• 17:17
  Figure: It's the whole paradox of the field, and it has a name in the lab. Tversky and Kahneman, 1983. They give people a short description of a woman named Linda — a portrait built to fit the picture of an activist. Then they ask: which is more probable — that Linda is a bank teller, or that Linda is a bank teller and a feminist? Eighty-five percent of people pick the second.
• 17:43
  Ground: And the second can't be more likely.
• 17:45
  Figure: It's mathematically impossible for it to be more likely. Bank teller and feminist is a subset of bank teller. Every added detail can only lower the probability, never raise it. But the richer story — the one that fits the portrait — feels truer. That's the conjunction fallacy. And Schoemaker, who wrote the foundational account of scenarios as a debiasing tool, cites that exact paper as a hazard of the method.
• 18:13
  Ground: So a scenario can invite that very fallacy. The entire craft is to make it vivid, detailed, coherent, named — Icarus, Flamingos, the river delta. And every stroke that makes it feel real is a stroke that makes it feel more probable than it is. [pause] Which means it fails at both ends of itself. Too dull to reach your microcosm, and it changes nothing — that's Wack's first-generation flop. Vivid enough to change you, and it's vivid enough to fool you — to feel like a prediction, to crowd out the sober question of how likely it actually is. [pause] The very thing that makes it work is the thing you have to distrust while it's working. You have to be moved and suspicious at the same time. Believe the world enough to rehearse in it, disbelieve it enough not to bet the house. That's not a comfortable instrument. That's a knife you hold by the blade.
• 19:13
  Figure: And that discomfort is why the field has a rival that keeps it honest — Philip Tetlock. Two decades of research. His first finding: the average expert forecasting politics and economics was, barely exaggerating, little better than guessing. And the confident ones did worse. The hedgehogs — one big idea applied to everything — lost to the foxes, the self-doubting, many-model thinkers.
• 19:40
  Ground: So the confident hedgehogs — the ones committed to one big idea — were the ones to trust least.
• 19:47
  Figure: Then in Superforecasting he found something sharper. A minority of ordinary people, no clearances, no secrets, who reliably beat the professionals. In the Good Judgment Project, the superforecasters did about thirty percent better than CIA analysts working from classified material. The superforecasters looking three hundred days out beat regular forecasters looking a hundred days out.
• 20:10
  Ground: Now I'm confused which discipline to believe. Tetlock says quantify, put a number on it, keep score, be humble, update. The scenario tradition — at least the story-first, intuitive-logics strain — says don't force a single probability, tell several plausible stories, imagine the unthinkable. Those sound like enemies.
• 20:35
  Figure: They sound like enemies. They're a division of labor across time. Tetlock's method works on questions that resolve — will this happen in the next eighteen months, yes or no, and we'll know who was right. Scenarios work where there's no resolution and no base rate — the five-to-thirty-year questions where the world can turn unrecognizable and probability stops meaning anything. You use the forecaster's discipline for the near signposts, and the scenario's imagination for the far frame. One keeps score. The other keeps you from being blindsided by the thing that was never on the scoreboard.
• 21:11
  Ground: Does any of it actually work, though — or is that just the tidy way to make peace between them?
• 21:17
  Figure: Fair, and the honest answer is the evidence is thin. The field's own scholars describe it as suffering conceptual confusion, methodological chaos, and scarcity of evidence. The best clean experiment I know — Chermack and colleagues, two hundred fifty-two graduate students — did find scenario planning reduced framing bias and improved decisions more than ordinary planning. With a catch. Only when they ran the full method. A partial version didn't deliver it.
• 21:46
  Ground: So the debiasing is real, but only if you don't cut the corners everyone cuts.
• 21:52
  Figure: The least marketable finding imaginable. It works, at the edges, if you do the whole hard thing — and even then it changes how you decide, not what comes true. And yet look who's adopted it. The U.S. National Intelligence Council has published a Global Trends report to brief each incoming administration since 1997. By 2004 they're using the intuitive-logics methodology associated with Peter Schwartz's old Global Business Network to hand the White House four named fictional worlds. Davos World. Pax Americana. A New Caliphate. Cycle of Fear.
• 22:30
  Ground: The government of the most powerful country on earth is briefing its president with made-up stories that carry a warning label saying these are not predictions. That should be absurd. Why do it?
• 22:43
  Figure: For exactly the reason we've been circling. Not to know the future — to give a fractured room a shared vocabulary to argue about it. And the climate world did the same. Its scenario work moved through three IPCC-linked generations — the old emissions storylines the IPCC published itself, then the concentration pathways it called for and a modeling consortium coordinated, then the Shared Socioeconomic Pathways an international research community built, five narrative worlds, SSP1 through SSP5, each a different way the century's societies could go. Not one forecast. A spread of plausible worlds you test your policy against.
• 23:26
  Ground: So the strange part is that whole civilizations now run their planning on named fictions — and they've quietly agreed the fiction is the point. Nobody's pretending the story is true. They're using it as a place to stand and see.
• 23:42
  Figure: Which forces the real reframe, now that the prediction story is dead. It was never forecasting. It's closer to something else entirely.
• 23:51
  Ground: It's design. A scenario is a prototype — not of an object, of a world. It's a provisional story you build so that what breaks shows up before a real decision makes it costly. And there's never just one, because the point isn't to be right. It's to widen the room. [pause] There's even a diagram for it — the futures cone. A cone opening out from the present into rings: what's probable, what's merely plausible, what's barely possible, and off to one side, what's preferable. The whole practice is refusing to collapse that cone down to a single line and calling the line a forecast. [pause] And the people who make futures physical — Anthony Dunne and Fiona Raby, the speculative designers — they say the point of imagining a future isn't to predict it. It's to argue about it. To hold up an object from a world that doesn't exist so the people in this one can decide whether they even want it. That's the same move Wack made with a slide of a river delta. Make the unthinkable thinkable enough to have an opinion about.
• 25:05
  Figure: There's a phrase for what it leaves behind. Arie de Geus, one of the Shell planners, called scenarios memories of the future.
• 25:14
  Ground: I love that, and I don't fully believe it.
• 25:17
  Figure: Then hold both. A memory of a future you never lived is exactly the thing we've been circling all along — strong enough to speed you up when the real thing arrives, false enough to fool you the moment you mistake it for prophecy. Not a picture of what will be. A rehearsal you carry, knowing it's a rehearsal.
• 25:37
  Ground: The prepared mind, one more time — but honestly this time. Shell's people weren't prepared because they knew. They were prepared because they'd already grieved the cheap-oil world in imagination, so when it actually died they didn't lose the first hour to shock. That's so much smaller than prophecy. And it might be better than prophecy — because prophecy you can't practice, and this you can.
• 26:04
  Figure: And it needs no corporation, which is the part that stays with me. Kahn had RAND. Wack had Shell. But the machinery is the one Craik described in 1943 — a mind running small models of a world that isn't here yet. What it asks for is a real question, the nerve to imagine the branch you're avoiding, and the discipline to be moved by the story without being owned by it.
• 26:31
  Ground: Moved but not owned. That's the entire knife-edge.
• 26:35
  Figure: Which means the real work was never out there in the future at all. It was the quiet, slightly dangerous act of letting a world you don't want become real enough to feel — and then walking back into this one, changed, but not fooled.
• 26:50
  Ground: How do you let a possible future move you enough to prepare for it, without letting it convince you you've already seen the one that's coming?