The history of the future and the demon who haunted it

October 26 , 2022

In this week’s issue of The Distance, we have a unique treat from writer Amanda Gefter, who specializes in discussing mind-bending scientific topics such as fundamental physics and cosmology. Today, she’ll be considering theories from Einstein and John Archibald Wheeler, to help us understand whether it’s possible to actually see into and predict the future.

Now, here’s Amanda:

I used to be terrible at making decisions. Sure, I could manage the little ones — picking food off a menu, choosing a new show to watch — but the big ones often left me paralyzed. Should I stay in Philadelphia or would I be happier in Boston? Should I keep dating the brooding artist or take my chances on someone new? Should I stick with the safety of my nine-to-five job or would I make it as a writer striking out on my own? The paralysis, I’m certain, stemmed from a feeling in my gut that there were right answers and wrong answers. That if I just had more information, I’d be able to work out what the future had in store, and then I’d know which was which.

I wanted, in other words, to be Laplace’s demon. Pierre-Simon Laplace, the French polymath, noted in 1814 that if someone knew the precise position and momentum of every atom in existence, they could use Newton’s laws of physics to predict the future of the universe and see how their entire life would unfold long before it ever happened. The atoms that make up the world around us — that make up us — would be destined to follow a traceable course, like a chain of falling dominos. “For such an intellect,” Laplace wrote, “nothing would be uncertain and the future just like the past would be present before its eyes.”

Laplace called his all-knowing creature a “vast intelligence” — it was only later that people started calling him a “demon.” You’d think they’d call him an oracle, or just “expert decision-maker,” but instead he was cast as something evil — out of fear, perhaps, that if the future really were contained in the present, then our choices, despite being the “right” ones, wouldn’t really matter. If Laplace were right, people feared, our choices would be as predetermined as their consequences. Still, knowing ahead of time that the brooding artist would cheat surely would have made my decision simpler.

However, Newtonian physics was superseded by Einstein’s in the early 20th century. Einstein’s “relativity” model did the demon one better. Not only could a vast intelligence know the future, Einstein said, but the future, in a very real sense, is already here.

Einstein’s key insight, back in 1905, was that if you ask whether two events happened at the same time, you’ll get different answers from different observers. It’s not that some are right and others wrong; it’s that the meaning of the word “now” changes depending on your state of motion. And since every observer’s “now” is as real as any other, every “now” — every possible moment in time — must coexist, all at once. “The distinction between past, present, and future,” Einstein said, “is only a stubbornly persistent illusion.” All that really exists is one eternal instant, and we’re stuck wandering around inside of it, believing, like chumps, that the future has yet to arrive. Physicists call it a “block universe.”

But even Einstein didn’t see the full picture. It was John Archibald Wheeler who realized that Einstein was wrong. The legendary American physicist, who had walked and talked with Einstein in Princeton, and who became famous for coining terms like “black hole” and “wormhole,” saw more clearly than anyone how our understanding of quantum mechanics took a sledgehammer to the concept of a block universe.

Quantum mechanics teaches us that the universe is not set in stone; a physical system exists as a multitude of possibilities, which only become actualities when we measure them. In the process of measurement, we ask nature what she is like, and nature hands us an answer. And the central lesson of quantum mechanics, Wheeler realized, was this: we can always ask more questions of nature than she has room in her pockets to carry all the necessary answers around.

We can ask, for instance, What is that electron’s position? And nature pulls out an answer. We can ask, What is that electron’s momentum? Again, nature pulls out an answer. But at any given moment, the electron doesn’t have a position and momentum — not at the same time. Which means that nature is just making up answers on the fly. And if that’s the case, then our measurements don’t reveal preexisting facts about the world. They create new facts.

“The act of measurement,” Wheeler wrote, “has an inescapable effect on the future.” The future isn’t just sitting out there, waiting for us to stumble toward it, because we participate in making it.

And that turns out to be a real problem for Laplace’s demon. After all, to predict the future, the demon needs to know the precise position and momentum of every particle, which, according to quantum mechanics, no particle actually has. That’s why the outcome of a quantum measurement — the unfolding of time from one “now” to the next — is profoundly unpredictable. Random. Participation is a double-edged sword: because our own choices of what to measure shape the future, we can never know ahead of time what the future will hold.

Strangely, it’s not only the future that we shape through our measurements — it’s the past, too. In 1978, Wheeler proposed an experiment called “the delayed choice,” in which we make a measurement on a particle of light that’s come to us from a distant star. Our decision to measure one feature of the particle, instead of another, can determine, Wheeler showed, which path the starlight traveled through the universe in order to reach us, despite the fact that it was traveling that path for billions of years before we made our measurement.

“In a delayed-choice experiment,” Wheeler wrote, “a choice made in the here-and-now has irretrievable consequences for what one has the right to say about what has already happened in the very earliest days of the universe, long before there was any life on Earth.” The lesson of the delayed choice is not that our measurements reach back into the past and change it. The lesson is that there is no past just sitting out there, fixed, independent of our actions.

To be fair, Einstein, par for the course, actually saw that twist coming. In a little-known paper titled “Knowledge of Past and Future in Quantum Mechanics,” published in 1931, he wrote, “The principles of quantum mechanics actually involve an uncertainty in the description of past events which is analogous to the uncertainty in the prediction of future events.” But it was Wheeler who appreciated the source of that uncertainty: it’s us.

Here is an empowering and terrifying fact: our decisions actually matter. “We used to think that the world exists out there, independent of us,” Wheeler wrote, “we the observer safely hidden behind a one-foot thick slab of plate glass, not getting involved, only observing. However, we’ve concluded that that isn’t the way the world works. We have to smash the glass, reach in.”

Einstein always hoped that the theory of quantum mechanics would turn out to be wrong. It didn’t. It turned out to be the most successful, accurate, well-tested theory in the entire history of science. Physicists even put Wheeler’s delayed choice to the test and confirmed it: measurements in the present can indeed shape the past. The French physicist Alain Aspect, who performed one of the delayed choice experiments, was awarded the Nobel Prize in physics earlier this month.

So, quantum mechanics has exorcized Laplace’s demon once and for all. And personally, thinking about it now, I’m glad. It’s true that I wanted to be like the demon, but I’ve since come to realize something: the demon couldn’t make any choices. He may have known how his future would play out, but he couldn’t do anything about it. His knowing made no difference. He could have no impact on the trajectory of his own life, or anyone else’s, because that trajectory was set in motion in the very first moment that the universe began. He was ineffectual. Lonely. Standing outside the world, looking in.

We, on the other hand, are on the inside. Finite, flawed, flesh and blood creatures, we are pieces of the universe, and everything we do requires us to grab hold of other pieces and shake them until something new falls out. A measurement outcome. A life. A past. A future. Uncertainty, unpredictability, and randomness — they’re the costs of being part of the world. The price we pay to have an impact.

So these days, when I’m faced with a decision, I no longer try to gather endless information and guess what the future holds, as if it holds anything. These days, I make my choices. Smash the glass. Reach in.

Amanda Gefter (@amandagefter on twitter) is a freelance writer based in Watertown, Massachusetts. She is the author of Trespassing on Einstein’s Lawn.

Questions We're Asking:

• If it were possible to predict the future, would you actually want to have that power? Why or why not?

• Now that you’ve read about Wheeler’s conclusions regarding the nature of time and causality, how do you think it’ll affect the ways you make decisions in the future?

Hearing from you is the best part of our week!
Write us, TheDistance@fundrise.com

And now, a brief musing that concludes with the author depicting himself as a robot centaur

Alex writes:

Almost exactly a year ago, we published an essay posing what turned out to be a shockingly pertinent question: “Are robots coming for your job?” As writer Ryan McCarthy revealed in that piece, there are very real possibilities and precedents for some workers to find their professional skillsets absorbed — and, yes, replaced — by new technological advancements. On the other hand, a McKinsey Institute report estimated that “very few occupations — less than 5 percent — consist of activities that can be fully automated.” Still, Ryan’s bottomline advice to readers was to pay attention to how the industries connected to their jobs were likely to become automated, and to actively learn new skills that are less likely to get robotified.

I recently came across a couple articles, both published earlier this year, that add a radical new dimension to this topic. Specifically, they’re concerned with the recent boom in platforms that harness A.I. to assist users in quickly and easily generating custom art. In their simplest forms, these platforms only require you to describe your subject matter in a few words or a phrase, select a style, and moments later an original image appears (with, admittedly, a wide-ranging degree of accuracy depending on platform and prompt).

The first article, a blog post titled “The AI Art Apocalypse,” considers the question of what these new technologies will mean specifically for people whose professions center around creating commissioned art or other creative visuals. Author Alexander Wales thinks thoroughly about which human artists might be displaced by new robot counterparts, and he presents economically minded questions that may not be obvious at first glance. “What I think is… likely,” Wales writes, “is that prompt engineering and image manipulation will become go-to skills, and the artists with those skills will displace the artists whose primary skill is in the manual craft.”

The second article was a September report by the New York Times: “An A.I.-Generated Picture Won an Art Prize. Artists Aren’t Happy.” The complication this article introduces is especially problematic, because when we imagine robots taking our jobs, most of us probably picture purely mechanical tasks, like driving a car or transcribing dictation. In these cases, whether work has been successfully accomplished is a purely binary question; did the taxi customer get from point A to point B? Yes? Problem solved. Art, however, is a matter of taste, refined aesthetics, and, we suppose, human expression. A.I. can produce custom, original imagery, that much is clearly true… but is that imagery actually any good? Does it move us, as people? Well, according to the New York Times, it might! That is, if the right person is there to help edit and curate the output into a compelling final product.

I understand if people feel anxious about robotics invading the world of art, especially when it’s clear robots can produce it more quickly and more cheaply. Questioning what we stand to lose is a very valid response. But it also reminds me of some of the data that shows that robotics are most efficient not as replacements for humans, but as enhancement tools. In the world of chess, studies have shown that the most powerful players are not human grandmasters, nor are they standalone A.I. opponents — instead, they are “centaurs,” moderate human players paired with chess software. Similarly, in the world of medicine, machines have shown better accuracy in the critical practice of diagnosis, making diagnosis software a powerful resource for docs who know how to wield it.

Like Wales’ quote above, the emergence of A.I. art may not occasion the obsolescence of human artists but rather the appearance of a new set of human skills. Namely, the ability to nurture and direct these robotic abilities, like a stablehand taking care of a fleet of robot horses, ready and able to do work better than we ever could alone. Who knows — maybe it’s only a matter of time before a robot horse replaces me, your trusty newsletter editor, writing and publishing an editorial like this one.

What might that look like? Here’s what one art robot drew up.

Does the idea of A.I. art worry you? Why or why not?
Let us know! Write to us at TheDistance@fundrise.com.

Keep the emails coming!

We love to hear from you: TheDistance@fundrise.com

See you in the future!

The Distance is supported by Fundrise, edited by Alex Slotnick and Helen Chandler, and produced by Katie Valavanis and Brett Wilson, with help from Motasem Halawani. Our design director is Erin Culliney.