What The Hell Is Going On With Dark Energy?
The latest blow-up over dark energy reveals the gap between theory and observation
Last week, the science media was abuzz with reports that dark energy might not exist. The astrophysics community has largely rejected the study behind the headlines, which argues the original evidence behind dark energy is flawed. This latest dark energy kerfuffle doesn’t necessarily reveal a problem with the physics, but it does reveal the widening gulf between what our theories say the universe should be and what we actually observe.
The universe needs dark energy, the enigmatic substance that accounts for nearly 70 percent of the total energy in the cosmos and causes the universe to expand at an accelerating rate, to explain one of the most surprising discoveries of the 20th century. Two separate 1998 studies of distant supernovas found that the universe was not just expanding but that the expansion was actually accelerating, with three of the physicists involved winning the 2011 Nobel Prize for their efforts.
But that discovery didn’t fit with our understanding of gravity, which should very gradually begin to pull the matter in the universe back together. There needed to be something previously unknown that was resisting gravity and speeding up cosmic acceleration, and that something is dark energy. It could be something hardwired into the fabric of the universe — what’s known as the cosmological constant — or it could be an as yet undetected force or substance. Whatever it is, dark energy defies easy explanation.
“I mean, dark energy is weird, okay?” UCLA astrophysicist Edward Wright told Vocativ. “Having this weirdness in the universe is disturbing. And that’s why a lot of people are always on the lookout for ways to avoid it.”
At first glance, the best way to avoid dark energy would be to refute the original evidence for cosmic acceleration. That’s precisely what a new paper in Scientific Reports co-authored by Oxford University physicist Subir Sarkar purports to do. In the paper, Sarkar and his colleagues analyzed an expanded dataset of supernovas and found what they call “marginal evidence” for cosmic acceleration. They didn’t disprove an accelerating universe, but they argued the apparent data for it didn’t meet the level of statistical certainty needed for a formal discovery. In their view, dark energy might just be a fluke of the data.
To Sarkar, that’s reason enough for astrophysicists to call the entire model into question, given how strange and poorly understood dark energy is.
“Until they are absolutely certain that that is the correct theory, then they should be more conservative and leave open the possibility that there are other models that can fit the data as well,” he said.
Talking to astrophysicists, you get the sense that they wouldn’t necessarily mind being rid of dark energy, assuming the data in this paper could show that.
“I will definitely confess that when I first saw the abstract for the paper, a part of me went, ‘Yes! We don’t have to worry about this problem anymore… maybe,’” University of Washington astrophysicist Chanda Prescod-Weinstein told Vocativ. “So I can understand the impulse even from the point of view of someone who’s worked on the cosmic acceleration problem.”
Astrophysicists like Prescod-Weinstein and Wright have criticized the paper’s statistical method, arguing it makes assumptions about the universe that aren’t supported by the observational data. Sarkar pushed back against that argument, telling Vocativ that his team’s methods actively avoided the assumptions that he said tilted prior studies in favor of an accelerating universe.
The disagreements here are significant, with Sarkar calling into question a model of accelerated expansion that Wright said has been settled science for nearly 15 years. But the issues involved are also technical and nonintuitive, beyond the easy understanding of those outside the physics community. For those of us on the outside, it’s probably best to say that future research will clarify which side is right here, though it’s worth noting in the meantime that the astrophysics community is overwhelmingly in favor of the accelerated expansion model.
Either way, this is just the latest round of an ongoing debate between theoretical physicists like Sarkar and astrophysicists like Wright as to how we should best go about understanding the universe’s most challenging and perplexing mysteries. Should we believe what our instruments tell us, no matter how preposterous something like dark energy might seem, or do we first need to understand a phenomenon’s theoretical basis before accepting it as part of our understanding of the universe?
“The concern that I’ve always had is that the cosmological constant or dark energy, which is an intrinsic part of the fundamental model, has absolutely no explanation in fundamental physics,” said Sarkar. “And I would rather not bias my analysis by already assuming that it is there in the data.”
Part of Sarkar’s objection to accelerated expansion is as much political as it is theoretical. In his view, awarding the Nobel prize moved dark energy into the realm of fundamental physics, which would mean it ought to be held to the same standard of proof as the subatomic particles in the Large Hadron Collider.
“I was hoping for a simple universe and I didn’t get it.” — UCLA astrophysicist Edward Wright
“Discoveries in astronomy don’t normally get the Nobel prize,” he said. “And if it is a real discovery, if the universe’s expansion is actually accelerating, then it certainly deserves a Nobel prize. All we are saying is the criteria for that should be a little higher. The stakes should be higher.”
For Wright, this distinction is immaterial: Dark energy is there in the data — not just in the supernovas but in the structure of the cosmic microwave background, the ancient afterglow of the Big Bang — much as he might have once hoped that weren’t the case.
“I was hoping for a simple universe and I didn’t get it,” said Wright. “I can give you good arguments for why dark energy has to be zero but it turns out to be wrong. So observational evidence trumps your theoretical prejudice.”
Yet how certain can astrophysicists be about what they observe? To understand that question, let’s go back to Sarkar’s paper and say, for the sake of argument, that it’s indeed correct in its analysis and that the evidence for dark energy is less certain than previously thought. Even then, the difference isn’t much, as the new paper argues what was once a 5-sigma level of certainty — a statistical term that translates to 99.99994 percent, the same certainty needed to announce the existence of the Higgs boson — is now down to 3-sigma. But that’s still 99.7 percent in favor of dark energy.
Wright told Vocativ that a drop to 3-sigma wouldn’t be a compelling reason to doubt cosmic acceleration, even if he believed the paper was accurate. But Sarkar contrasted that view with his background in particle physics, where even a result that reached as high as 7-sigma — that’s 99.999999987 percent certainty — turned out to be incorrect.
“Particle physicists have an advantage over astronomers of doing reproducible experiments,” said Sarkar. “You can do the same thing over and over again in the laboratory. And therefore we have access to a database of past experience that says that 3-sigma happens almost all the time.”
Astrophysicists like Prescod-Weinstein aren’t necessarily convinced by that argument, but she did say there is always a need for skepticism, even if this particular paper is flawed in its methods.
“I agree that we should continuously push against what we think our data is telling us and make sure that we’ve correctly understood it,” she said. “I think that the impulse is a good one. It’s a scientific impulse.”
She and Sarkar actually agree on one major philosophical point: Whatever the solution to the dark enery mystery turns out to be, it likely won’t fit with our preconceptions of what physics ought to be. There’s a longstanding idea that physics should be fundamentally simple, that there ought to be natural explanations for why things are the way they are.
For instance, as fiendishly complicated as the math in Albert Einstein’s general relativity might be, it’s still elegant in how comprehensively it explains the universe (at least before you get into quantum mechanics). Sarkar argued that cosmology’s mistake has been in continuing to refine Einstein’s original model, instead of starting from scratch as seemingly absurd results like dark energy started creeping in.
“We could have a more complicated universe than the simple one Einstein assumed in the 1930s,” said Sarkar. “And dark energy might just be an artifact of those simplifications.”
There’s historical precedent for such cosmological artifacts: epicycles, the complex orbits within orbits that astronomers before Copernicus devised to maintain the model that Earth was at the center of the universe.
Geocentrism best fit the universe as observed before the invention of the telescope, and Sarkar argued its undoing was not in its failure to explain the orbits of other planets. In fact, the initial models with the sun at the center of the universe were less accurate than the geocentric models. Instead, the problem was similar to the one now facing dark energy.
“It had no fundamental explanation for why the hell we should be at the center of the universe and all the planets and sun should be going around us,” said Sarkar. By contrast, the heliocentric model paved the way for Johannes Kepler to provide the first comprehensive theory for planetary orbits. The Greeks and their successors had allowed their need for a simple model that reflected the universe as they saw it to supersede the problems with the data.
“I don’t think any of us should be arrogant enough to think our sense of physics should dictate the nature of the universe,” he said. “It is not the case that simple things are necessarily correct. The Greek model was very simple.”
But that argument can be turned the other way, with Prescod-Weinstein arguing the reluctance to accept dark energy is more akin to the 19th century skepticism that light could travel without being surrounded by some medium. To explain certain properties of light, scientists of the time postulated the existence of a “luminiferous ether,” a since debunked theoretical medium through which light was thought to travel.
“What seems right and natural to us is contextual,” she said. “There was a time when the ether seemed more natural to a significant portion of the scientific community.” That belief that light simply had to travel through a medium persisted even after the 1887 Michelson-Morley experiment — which also won the Nobel prize in physics — provided compelling evidence against the ether’s existence.
“Ultimately, we had to change our view of what was natural,” said Prescod-Weinstein. “It’s a little bit funny for me to be making the case for that, because I think the cosmological constant is a very unnatural solution. But I try and have some humility about the difference between my gut feeling and knowing that the universe could just be what it is. The universe doesn’t care what my gut feeling about the situation is.”
It can be uncomfortable — both for physicists and for the public at large — to recognize that such a big and fundamental part of our understanding of the universe remains so completely unexplained. But in science, that’s not necessarily a bug. That’s arguably its defining feature.
“Most of what scientists do is not know things,” said Prescod-Weinstein. “Our focus is what we don’t know, not what we do know. I know that when we put things in textbooks, we make it sound like we just have everything down. But actually what drives us every day is we don’t understand a lot of things.”
Dark energy remains at the top of the list of things physicists don’t understand. And, in all probability, it’s not going anywhere anytime soon.
This story originally appeared on Vocativ on November 1st, 2016.