Beginning in the 1970s cosmologists started to uncover the structure of the universe writ large. They already knew that galaxies occasionally clump together into clusters, but over even larger distances, spanning 100 million light-years and more, they found superclusters. And in between those superclusters they saw something even more unexpected: vast regions devoid of galaxies, great and immense dark hollows. The first of these cosmic voids that cosmologists discovered was 65 million light-years across. No theoretical work had predicted their existence, and for years cosmologists thought they were creating patterns with their imaginations.
We now realize that the largest structures in the universe are not superclusters or any other agglomeration of matter. They are the voids: the negative spaces, the intergalactic Saharas.
I was introduced to voids by my mentor Ben Wandelt while I was in grad school. He was fascinated by them; they appeared too empty to be explained by standard cosmology. My graduate advisor, Paul Ricker, and I worked with Wandelt on this problem for a while. But like most cosmologists, even though I knew about the voids, I didn’t think much of them. I favored galaxies and clusters. Voids were nothing, after all.
You could stretch 25 Milky Ways side-by-side in the gap between us and the nearest galaxy.
But after I got my Ph.D. I moved to Paris to join Wandelt for a postdoc position. Voids began to grow on me. I remember vividly the reactions we would get when we presented our preliminary work. Curiosity and interest, for sure, but also skepticism—not just the healthy skepticism needed for fruitful scientific progress, but the acidic scorn used to put junior scientists and their wayward ideas in their proper place. One time a prominent cosmologist—a senior, tenured professor—walked up to me in the hallway after I gave a talk at a conference, said simply “This will never work,” and turned around and walked away.
I didn’t mind the criticism because I knew something they didn’t. Cosmic voids are cosmology at its purest. They are simple. The complications of star formation and black holes don’t impact them because they don’t have any stars or black holes. They are basically big fossils from the earliest days of the universe and their shapes encode the evolution of the universe as a whole. If you want to understand some of the biggest puzzles in physical science, such as dark energy, you don’t want to look where the matter is, but where it isn’t.
Our most distant spacecraft, Voyager 1, has been traveling for over four decades and is now about 14.5 billion miles away from home. It is well and truly outside the boundaries of the solar system, and the average density within that space is a mere 10-15 grams per cubic centimeter. If that seems empty, just wait.
As we zoom out, we meet scattered stars, some of them associating with each other in clumps, others drifting alone. We see faint sketches of the great spiral arms of our galaxy, twisting pinwheels of gas and blue-bright stars, our own solar system tucked into a single spur of a much vaster arch. Our entire Milky Way galaxy is roughly 100,000 light-years across—a great, complicated metropolis of a few hundred billion stars. It has an average density of roughly 10-22 grams per cubic centimeter.
Beyond the warm confines of our galaxy sit the empty depths of intergalactic space, home to the occasional dwarf galaxy and the random flows of far-flung gas streams. The nearest major galaxy to our own, Andromeda, is 2.5 million light-years away. You could stretch 25 Milky Ways side-by-side in the gap between us.
Some galaxies lead solitary lives, and others assemble into groups and clusters of increasing size. As we continue to zoom out, entire galaxies, each one home to hundreds of billions of stars, appear as tiny motes of light adrift in an unlit ocean. The universe begins to look like a vast web, a network of long strings and tremendous walls, and between them the voids. We have a name for the nearby assortment of large-scale structures: Laniakea, the Hawaiian word for “bountiful heaven.” A knotted, twisted tangle of vines made of galaxies stretches for over 100 million light-years. At these distances the average density of matter approaches 10-29 grams per cubic centimeter.
And in those cavernous gaps between the structures, we find: almost nothing. There the density plunges to perhaps 10–30 grams per cubic centimeter. That is, to be sure, not entirely empty. Voids do contain a few dim, dwarf galaxies. Molecules of cold hydrogen and helium float among them, some blown out of their home galaxies, the rest left over from the Big Bang.
The nearest void to our Milky Way galaxy is the appropriately named Local Void, bordering our section of the universe and continuing on for almost 200 million light-years. The aptly named Giant Void is eight times wider. Voids comprise a tiny fraction of the mass in the cosmos, yet completely dominate its volume.
A cosmos that generates large-scale structures must, by its very nature, also produce the voids. The culprit is gravity: simple, persistent gravity.
Once cosmologists realized this story of creation, they realized they must turn the tables.
Billions of years ago, when the entire volume of our observable universe, 90 billion-plus light-years across, was crammed into a volume smaller than an atom, the exotic high-temperature plasma that filled our cosmos was relatively uniform. There were no significant density contrasts, and really nothing significant at all. Cosmologists believe that at this time, our universe underwent a sudden, rapid expansion known as inflation. It enlarged microscopic quantum fluctuations into the seeds of all we see today.
From then, gravity worked in the shadows. Regions of ever-so-slightly higher density had a little more gravitational pull than their neighbors, which encouraged more material to flow into those dense pockets. With even more material, the gravitational pull grew stronger, building on itself in a feedback loop. Over hundreds of millions of years, it gave rise to stars, galaxies, clusters, and the majesty of the cosmic web.
But in a universe where the rich get richer, the poor must get poorer. As matter flowed into the dense regions that would light up with stars and activity, the cosmic voids, seeded in those very same humble epochs, grew to their present all-encompassing size.
Once cosmologists realized this story of creation and confirmed it with observations, they realized they must turn the tables. The large-scale structure was not a question, but an answer. By measuring it, they could solve some of their other puzzles. The shape of the cosmic web is sensitive to the amount of regular and dark matter, the abundance of neutrinos, the presence of dark energy, and so on. So cosmologists sampled hundreds of thousands, then millions, of galaxies throughout the universe to map the large-scale structure in an attempt to quantify the basic ingredients of the cosmos.
Those surveys focused on what astronomers knew best: the bright objects. The cosmic voids were forgotten.
Those galaxy surveys have revealed ever-more precise estimates of the amount of dark matter in the universe. They have also provided independent evidence for the existence of dark energy, the mysterious substance that is driving the accelerated expansion of the universe.
But it hasn’t been easy. Galaxies and clusters are messy, complicated structures. The primordial density fluctuations have been swamped by generations of stars, supernovae, stellar winds, cosmic rays, magnetic fields, and all the wonderful astrophysics that makes life so interesting.
Teasing pristine, primordial cosmological information out of something as complicated as a galaxy or a cluster is next to impossible. Take a simple cosmological test developed by Charles Alcock and Bohdan Paczyński in 1979. If you scattered perfect spheres throughout the universe, from our point of view the spheres would appear to be elongated along our line of sight toward them—a slight stretch induced by the expansion of the universe. If you could measure that stretch as a function of distance, you could reconstruct the expansion history of the universe with incredible precision.
Clusters of galaxies are, for the most part, relatively spherical. They formed long ago and have since settled into equilibrium, balancing the motions of their own gas and galaxies against their gravitational pull. But trying to apply the Alcock-Paczyński test leads to failure. The motions of individual galaxies are hundreds of times larger than the effect of cosmological expansion, completely swamping it and rendering clusters useless.
Cosmic voids are the largest and most ancient time capsules known to science.
But what of the voids? An individual void is anything but round; it occupies a random amount of volume in a wide variety of shapes. But we live in an isotropic universe: Our cosmos looks roughly the same, on average, no matter what direction we look. If we were to take a collection of voids, they should have random orientations, and if we were to find enough of them and stack them one on top of another, they should average out to form a sphere. If I give you enough scraps of randomly shaped paper and you stack them on top of each other, eventually you would build something round. Could the average void be good enough to teach us something about the universe?
This was what I worked on with Wandelt. It would consume three years of my research life. As more researchers and graduate students joined our team, I hit the road, living out of my backpack for weeks at a time, hopping from seminar to seminar and conference to conference, spreading what we were learning about cosmic voids and what they could teach us about the universe.
We learned that voids are amazingly simple. Unlike their cousins, the dense structures of the cosmos, they live relatively uncomplicated lives. They simply appear in the early universe and grow. When they do occasionally merge, it’s a brief affair: A wall of galaxies between two voids thins out, and two voids become one. Because of their simplicity, they remained relatively uncontaminated by astrophysical processes and the movements of individual galaxies. Within a couple years, we were able to realize the dream of Alcock and Paczyński, applying their test to real voids found in galaxy surveys and using the result to provide an independent measurement of the amount of dark matter and dark energy in the universe.
Because voids don’t change much through their lives, they retain a memory of the young universe. If you want to know what our universe was like billions of years ago, you can’t look into a galaxy or cluster—too much has changed. But a void? A void today is pretty much the same as a void billions of years ago. Voids are the largest and most ancient time capsules known to science.
When I say that voids are empty, I mean specifically that they are empty of matter. But that emptiness makes them full of something else: dark energy. Cosmologists aren’t sure what dark energy is; all we know is that it’s been causing the expansion of the universe to accelerate for the past 5 billion years. We believe that dark energy is somehow related to the vacuum of spacetime, a component of all the quantum fields that suffuse reality.
If you take a box and empty it of all the matter, you’re left with vacuum and the quantum vacuum fields that live in it. On a grander scale, the cosmic voids are brimming with quantum fields, including dark energy. You’ll never feel the effects of dark energy inside a solar system because the density of all the matter is more than enough to overwhelm the slight, subtle effects of dark energy. But out in the voids? It’s nothing but dark energy. Indeed, the cosmic voids are the places in our universe where the expansion is accelerating. It’s not happening in galaxies or clusters, but in the emptiness between them.
The lesson is clear: If you want to understand how dark energy works, you have to dive into the darkness.
Thanks to the dogged determination of my collaborators and other void-enthusiasts, voids have come out of the shadows. Every major upcoming galaxy survey now includes void science in their repertoire. Cosmologists are learning how to use voids to understand dark energy and dark matter, search for exotic fifth forces, dig into the conditions of the early cosmos, and more. Most recently, voids have provided the most powerful measurement of dark energy within the local universe, using simple analysis techniques developed by a handful of people, compared against the massive resources employed by giant collaborations to try to get the same information out of the bright structures.
No, we haven’t yet solved all the mysteries of the cosmos. But the answers to those questions are certainly waiting for us in the void.
Paul M. Sutter is a research professor in astrophysics at the Institute for Advanced Computational Science at Stony Brook University and a guest researcher at the Flatiron Institute in New York City. He is the author of Your Place in the Universe: Understanding our Big, Messy Existence.
Lead image: The dwarf galaxy KK 246 is a lonely island of stars in the Local Void, an otherwise empty region some 150 million light-years across. Credit: NASA.