BY MATTHEW HERBERT
I have a Jill Lepore problem.
Every time I read a new book by the Harvard historian, I can’t help thinking it’s her best. I just start effusing that she is America’s greatest historian. She writes with such ease and appeal, you can almost forget how erudite she is.
Joe Gould’s Teeth: Lepore’s investigation of whether the longest book ever written actually was written, by a mad man wandering the streets of New York?–her best.
The Mansion of Happiness: A History of Life and Death: her exploration of human flourishing as it has been defined on the American scene?–definitely her best.
And her magnum opus, These Truths: A History of the United States?–absolutely, unquestionably her best. You can’t be fully American if you haven’t read it.
Which leads me to her latest (2020) book, If Then: How the Simulmatics Corporation Invented the Future. It’s certainly her best.
In 1959, a company was born, Simulmatics. The name was a mashup of simulation and automatic. The guy who came up with it, Ed Greenfield, was hoping the neologism would take off, like Norbert Wiener’s cybernetics, but it didn’t.
Eventually, the company Simulmatics was no more successful than the name, but it is in the telling of Simulmatics’s ambition, its methods and projects, and the ways it failed, that reveal how this company–already lost to memory–anticipated almost every idea that drives social media, advertising, big data, and political campaigning today. Simulmatics truly did invent the future. And then it disappeared.
The basic idea was simple. “The company proposes to engage principally in estimating probable human behavior by the use of computer technology.”
And engage they did.
During the company’s 11-year life, its scientists–known as the “what-if men”–advised John F. Kennedy’s 1960 presidential campaign, nudging him to advocate for civil rights and lean into his Catholicism; they helped the U.S Army dream up the “strategic hamlets” program and other data-driven ideas for winning the Vietnam war; they helped the FBI predict race riots in the mid-1960s by combining data about networks of people with external political variables that affected their behavior; they got involved in the invention of the internet, at the Defense Department’s Advanced Research Projects Agency. Always trying to predict, systematically, what people would buy, what they would choose, or do.
All of it depended on one guiding idea, the algorithm. For the Kennedy campaign, Simulmatics analyzed voters into 480 categories (for example, white, middle-aged housewife in the midwest), coded their opinions on to punch-cards, and started asking their IBM computers what-if questions, like if Kennedy coddles white southern conservatives on civil rights, what would other categories of voters do? And that’s how they came up with civil rights as a campaign issue. It wasn’t because advocating for civil rights was the right thing for JFK to do; it was a vote getter.
One of Simulmatics’s scientists called this kind of modeling a “people machine.” It was a way of aggregating data about millions of small, individual choices to simulate the way whole groups of people–maybe even an entire country–would act.
The vision was grand, Lepore tells us:
The scientists of the Simulmatics Corporation acted on the proposition that if they could collect enough data about enough people and feed it into a machine, everything, one day, might be predictable, and everyone, every human mind, simulated, each act anticipated, automatically, and even driven and directed, by targeted messages as unerring as missiles.
Lepore is the rarest kind of historian. She writes from an unmistakable moral perspective (liberal humanism) but without rendering history as a black-and-white morality tale. Despite the atrocious personal behavior of Simulmatics’s what-if men (they drank like fish, cheated on their wives, neglected their children), and despite the monstrosity of what they wrought–a democracy-killing dystopia–Lepore retains an almost Vonnegut-esque gentleness about their foibles, which were, after all, products of history before they became forces of history.
Looking back, she observes, “[i]t would be easier, more comforting, less unsettling, if the scientists of Simulmatics were villains. But they weren’t. They were midcentury white liberals in an era when white liberals were not expected to understand people who weren’t white or liberal.”
Lepore tells the story of how these flawed men created a system that would become a monster, but she tells it with wit, wryness, and often with tragic beauty. The chapter of If Then entitled “The Things They Carried” is in itself a masterpiece of historical writing, not to be missed. It describes in captivating, exquisite detail how Simulmatics gathered so much data about people, analyzed it so thoroughly, and often accurately predicted their behavior in highly particular circumstances, but still failed to understand what was really happening in the United States. They missed the real world undisclosed by their data–the anguish over Vietnam, the draft, and civil rights: the whole direction that country was going.
Much of the “social science” behind Simulmatics was shoddy–or outright chicanery–and it 1971 the company filed for bankruptcy. “Simulmatics died,” Lepore reports. “The fantasy of predicting human behavior by way of machines did not. Instead, it took new forms, forms that depended on forgetting that Simulmatics had ever existed.”
She goes on:
Simulmatics failed, but not before its scientists built a very early version of the machine in which humanity would in the early twenty-first century find itself trapped, a machine that applies the science of psychological warfare to the affairs of ordinary life, a machine that manipulates opinion, exploits attention, commodifies information, divides voters, fractures communities, alienates individuals, and undermines democracy. . . . Long before the age of quarantine and social distancing, Simulmatics helped atomize the world.
So when the wundkerkids of the early 21st century went looking for the venture capital that would make them into zillionaires, they needed to portray everything they were doing as utterly new, and that they were leaving all of history behind. They needed to mystify. Otherwise they might have to admit that their disruptive “breakthroughs”–big data, social networks, targeted algorithms–were just better recycles of what had already been done. Moreover, they needed to believe, and probably did believe, that they had escaped the orbit of the old, left history behind for pure data.
“The only thing that matters is the future,” proclaimed Google’s Anthony Levandowski in 2018; “I don’t even know why we study history.” But in fact the innovations of the new tech titans embodied the same old flaws and biases that had pulled an earlier generation of technology prophets crashing down to earth. Facebook, Twitter and Cambridge Analytics are all better at what they do than Simulmatics, but what they do bears the same human stamp, revives the same mad ambition. They might have known that had they studied history. They might have learned that the abrogation of the past for faith in a transformative future “isn’t an original idea,” as Lepore concludes. “[I]t’s a creaky, bankrupt Cold War idea, an exhausted and discredited idea. The invention of the future has a history, decades old, dilapidated.”
2 thoughts on “Review of “If Then: How the Simulmatics Corporation Invented to Future” by Jill Lepore”
From that statement “The company proposes to engage principally in estimating probable human behavior by the use of computer technology.” I was thinking “Vonnegut” and sure enough you saw it too.
LikeLiked by 1 person
Yeah, every time I read something ‘serious’ about data science or machine learning, I go back and read Player Piano instead. Not Vonnegut’s masterpiece, but I think it’s a great first novel.