This is where stability matters. The galaxies are pulled together by gravity, by the invisible strings (or deformations in space-time, if you're feeling relativistic) that insist on bringing objects closer together. Resisting that is the intrinsic velocities of the galaxies themselves. You can imagine a cluster of galaxies like a tremendous swarm of bees. Gravity pulls the swarm into the space of a sphere and really, really wants to make that sphere smaller, but if the bees have enough energy, they'll just keep buzzing around inside that sphere.
If gravity overwhelmed the kinetic energy, the cluster would have collapsed a long time ago. If the galaxies were too energetic, it would have exploded. Since the time needed to collapse or explode is short (again, just as a reminder, that's “short” compared to things like the age of the universe), it's a good guess that these forces are in balance.
This allows a person like Zwicky to make a relatively easy measurement (the average speed of the galaxies is given by their redshift) and convert that into a relatively hard measurement (the mass of the cluster). So he did.1
He got a number that was a bit too large, honestly. He knew how much mass was in all the stars surrounding the solar system, and he knew how much light they produced. So that gives a pretty simple formula: if you see this much light, then you can make a good guess about how much mass is generating that light.
But that number, based on counting all the hot, glowy stuff on the Coma Cluster, didn't agree with the number produced by the virial theorem. The virial theorem was based on simple kinematics; it was just motion that was affected and balanced by gravity, which was on pretty solid footing. But that calculation, based on physics as good as anybody's, resulted in a cluster mass five hundred times larger than you would have guessed by adding up all the light sources.
Uh, what? What could explain this discrepancy? Maybe stars in the Coma Cluster are weirdos and don't behave like any other star anywhere else. That doesn't seem to fit. Maybe the laws of physics are different way out there than over here. Kepler surely wouldn't appreciate that, but he's dead and can't complain, so we'll keep that in our back pocket. Perhaps the Coma isn't in equilibrium after all—maybe we caught it in a really awkward time in its life and the ground rules of the virial theorem don't apply. Maybe, but while possible, that's unlikely.
Maybe it's something else. Maybe there's material in the Coma Cluster that isn't all hot and glowy. There could be macroscopic or microscopic stuff that we simply aren't seeing. In the words of Fritz Zwicky himself, maybe the cluster contains vast amounts of dunkle Materie.
It's German, and I'll translate for you: dark matter.
It was about time for a shake-up anyway. While astounding and capable of bending even the biggest of brains, the observational results of Hubble coupled with the theoretical insights of Einstein and company at least made sense when taken together: we live in an unfathomably large universe that's getting fatter every single day. We may not like that answer, but at least it's an answer.
But Zwicky's dark matter wasn't an answer—it was a question. Other clusters were tagged and bagged by later astronomers, and they each showed this curious disagreement between different measurements of the mass. But while it was noted, nobody made a serious move to try to explain it. I can't blame anyone: if you were an astronomer interested in cosmology, there was all sorts of excitement about the big bang, the cosmic microwave background, and the creation of elements to pass your time. No need to worry about tiny niggling issues like giant discrepancies in cluster mass estimates, right?
And so Zwicky, and everybody else, moved on from the mystery of dunkle Materie to other, more pressing and interesting problems. A generation went by without much further thought on the subject. Then in the 1970s another astronomer, Vera Rubin, saw the same problem crop up in another situation, and what was considered a forgotten yeah-we-should-get-around-to-that-someday problem rose from the dead, setting the stage for one of the great enduring mysteries of the modern cosmological age.
The brief window of explaining and understanding the universe in a consistent, coherent story was quickly closing. Oh well, it was nice while it lasted.
Rubin was studying the motions of stars in other galaxies, which by any measure (and especially mine, since I'm writing the book) is an amazing feat of human curiosity. Just a century ago, we were struggling to identify the distances and motions of the stars right in our own galactic neighborhood, and by the late 1960s, Rubin was investigating, with great success, detailed interior movements in structures millions of light-years away.2
It's a pretty straightforward measurement once you get the hang of it. Pick a galaxy. Zoom in on various parts of it. Look at the spectrum. Identify some known elements. Measure the shifting of the spectral lines associated with those elements relative to what you know on the Earth. The same song and dance from a century before repeated over again, just at unimaginably large distances.
In the case of the entire galaxy, you can apply this technique to find that the whole structure is moving, and usually you'll find that it's in a direction away from us. That's what Hubble found. But if you focus your scope on individual bits of the galaxy, you'll pick up some extra motions on top of the general movement. If it's rotating (and, uh, they do) then one side of the galaxy will be slightly blueshifted—spinning toward us—while the opposite side will be a little bit redshifted—spinning away from us. You can repeat this exercise at different spots in the galaxy and build up what's called a rotation curve: the orbital speed at various distances in a galaxy away from its galactic center.
I won't hold back the plot twist: Rubin didn't find the speeds she was expecting. I know, shocking.
Here's the deal. When you just look at a spiral galaxy, and I really mean look, you'll notice that there's a big bulge of material (stars, gas, etc.) in the center, with a relatively thin and empty disk surrounding it. And our dear old friend Kepler can tell you what the orbital speeds of stars out in the disk ought to be. OK, fine, it's really Newton's universal gravity, but it was Kepler who first spotted the harmonious motions in the planets, and universal gravity being universal and all, the same method applies on these gargantuan scales. The speed of an orbiting object, whether a planet in the solar system or a star in a distant galaxy, depends on its distance from the central gravitating body and the mass of that body. A bigger sun = faster planetary speeds. Closer orbits also = faster planetary speeds.
Great, lovely. If we apply this to a galaxy, where most of the material is obviously concentrated in the center, then we ought to find that stars nearest the center orbit the fastest, with a general laziness setting in as we move farther out in the galactic disk.
But alas, no. Rubin saw something completely different: totally flat rotation curves. Stars out in the boondocks, at the very edge of civilization, were orbiting just as fast as their more centrally located cousins. The stars in galaxy are simply moving too fast. There isn't enough gravity to contain motions of that speed—the grand, beautiful spirals that we know and love should have flung themselves to pieces long ago, not stayed glued together for eons.
And this wasn't just the case in one galaxy, but across dozens. So, unlike with Zwicky's cluster, we can't appeal to a chance fluke of observations. Even if one of these galaxies were simply acting strangely, most of them are in equilibrium. Apparently, moving too fast for the gravity of your own galaxy is just the natural state of things.
And in the decades since Zwicky's initial proposal, we had much more confidently pinned down the relationship between stellar output and mass. Even accounting for the mass of nebulae, whether they glowed brightly or not, there wasn't enough stuff inside a galaxy. Once again, one measurement (counting all the hot and glowy bits) was disagreeing with another measurement (one based on motions). Zwicky saw it in a single cluster, and the world ignored it. Rubin saw in galaxies everywhere she looked, and the world started paying attention.
We're at a crossroads. Nature is not playing fair. Different measurements of the s
ame quantity—answering the simple question of the total mass of a galaxy or cluster—were revealing different answers. And not just by a little bit. At best, the amount of glowing matter in a galaxy or cluster, even accounting for all the wave bands from radio to gamma rays and all the possible sources from nebulae and stars to brown dwarfs and giant black holes, was about one-fifth the value required by other mass estimates.
Rubin's result was as simple and annoyingly counterintuitive as Zwicky's: either our laws of physics don't work at galactic scales, or there's a dim and/or invisible component to their recipe. What is nature trying to tell us? New physics or dark matter? With only the results of Vera and Fritz (now that's a sitcom) to work from, we can't tell the difference between the alternatives.
We've been in this situation before, with a set of observations contradicting what we expect. It's kind of how science advances, so at least this is familiar territory. And in the case of gravity, we've taken both paths in the past. For example, Mercury. Tiny little Mercury, the closest to the sun and the most swift-footed of the planets. It has the most elliptical orbit of all the planets (not counting the dwarfs like Pluto and Eris), and the point where Mercury comes closest to our sun lazily traces out a circle over the years. Most of this motion is due to the gentle but persistent gravitational tugs of the outer planets, but a small part of that motion couldn't be explained by the gravity of Newton.
Perhaps it was another planet in our solar system, an inner-inner world of fire, nicknamed Vulcan, orbiting close enough to the sun to remain hidden to the ancients, only making its presence known by gravitational flirtations with Mercury. But searches turned up empty. The answer here was new physics: one of the first clues Einstein had that he was on the right track was that general relativity could fully explain Mercury's orbital oddities.3 In other words, the portrait of gravity as painted by Newton breaks down close to the sun, and it takes a revolutionary new view of the cosmos to understand what's going on.
So maybe the quandary presented by Rubin and Zwicky could be explained by new physics. Maybe general relativity, just like Newtonian physics before it, can't cut the mustard past a certain scale. We love you, Albert, and your theory is a thing of beauty. But maybe it's just not good enough, pal.
But to be perfectly honest, the Vulcan approach has worked in the past. The orbit of Uranus was also behaving oddly, given the known denizens of the solar system and our Newtonian knowledge of gravity. Instead of modifying Newton to account for the observations, astronomers instead posited the existence of a new planet to explain the curious orbits in the outer solar system, and in due time the agent provocateur was found—the planet Neptune.4
In that case, it wasn't new physics but previously unknown matter that best explained observations.
But what to do with galaxies and clusters? Well, when nature fights dirty, fight back—with science.
In the decades since Rubin's reinvigoration of the dark matter debate, astronomers around the world have embarked on an all-out observational war, measuring, comparing, and testing galaxy after galaxy and cluster after cluster with as many methods as humanly (and, in the age of more advanced technology, robotically) possible.
I'm going to be polite here and warn you that I'm about to absolutely inundate you with more evidence for the existence of dark matter. It's not that I'm trying to beat this concept into your brain, but—no, wait, that's exactly what I'm trying to do. The reason is that the mystery of dark matter persists to the present day (at least, to the time of me writing this book). We have tons of rock-solid evidence that something funny is going on out there in the great expanse, but we're much more hazy on what's causing it.
Because of this, there's a lot of fear, uncertainty, and doubt among the general public (surely not you, but maybe someone you know) when it comes to understanding this facet of our universe. In the past few centuries, we've solved a lot of mysteries of the heavens above us, but continued observations have revealed deeper, perhaps more sinister machinations in the motions of celestial objects. One of them is the nature of the earliest moments of the big bang, when forces were so extreme and exotic that we have trouble even theoretically navigating them.
The other is here, in the old and cold universe of today. It's not some relic of the distant past, a problem that we can leave on the doorstep of future generations of scientists, safe in the knowledge that our overall picture is coherent. Dark matter is present today, in galaxies and clusters all around you. And when I get there, you'll see that it's probably inside you too.
We've already seen Zwicky's realization that dunkle Materie might be a thing inside clusters of galaxies based on the motions of galaxies whizzing around inside them. But threaded between those galaxies is an incredibly hot (up to a hundred million Kelvin—that's hotter than the core of the sun) but incredibly thin (about one thousand particles per cubic meter; compare that to the 1025 air molecules per cubic meter that you're breathing right now) plasma. It's so thin that the particles—protons and electrons—travel for about a light-year before interacting with each other, but when they do, they emit X-ray radiation in a process known as Bremsstrahlung. It's German for “braking radiation,” and it may or may not be one of my favorite words in physics.5
So the gas is hot and emits X-rays. The hotter it is, the more X-rays come out; ergo, we can measure the intensity of X-rays and estimate the gas temperature. And just as with the galaxies themselves, there's a connection between the temperature of the gas and the total mass of the cluster. If the cluster is in equilibrium, then the gas can't be too hot (or the cluster explodes) or too cold (or it collapses). Since that is obviously not happening, we can get a rough handle on cluster masses. Result: not all the cluster mass is visible.
Here's another one. Let's rewind all the way back to the first dozen minutes of the universe, when all the protons and neutrons were hitching up to form the universe's initial supply of hydrogen, helium, and lithium. The prediction of the abundance of those light elements, spawned from our understanding of nuclear physics, was (and is) a triumph of the power of the big bang model. Those same calculations place a hard upper limit on the total amount of baryonic (if you remember your jargon lessons, that's a particle like a proton or neutron) material available in the universe. When we compare that to all our cluster observations, we end up with a number that's about one-fifth too shy. So in principle, we're measuring the same thing, the mass of the universe, but at different times (in the first minutes versus the latest billion years). Those numbers ought to be the same, because where is all the mass going to go? But instead the clusters are much fatter than they should be, given the elemental building blocks available to them.
Check this one out. Remember that story of inflation and the early gravitational growth of structures? It's a good story and worth remembering. There's one big caveat that I deliberately neglected to mention, because I wanted to save it for this moment: in order for the galaxies to have the sizes they do, we need a form of dark matter. Specifically, a kind of dark matter that doesn't interact with light. The problem with pure baryons is that they get easily distracted. Gravity tries to pull them together, but the intense radiation pulls them apart, a process we saw play out with agonizing repetition until matter and radiation finally went their separate ways with the birth of the cosmic microwave background.
But by the time of recombination, a few hundred thousand years into the history of the universe, the seeds were already planted…by dark matter. Some form of invisible matter could freely ignore radiation, pooling itself together in those early years, creating the nests that baryons would eventually collect in to start building larger structures. In other words, even though we can't see it, the cosmic microwave background and the later structures to emerge depends on the presence of dark matter.
Given only baryonic processes operating in the early universe, there simply wasn't enough time to build galaxies, including the Milky Way. Including you. Yes, you. Without dark matter, high-density structures
(and I'm not calling you dense, per se, just noting your density relative to—never mind) couldn't have formed.
Not convinced yet? Let's take a look at gravitational lensing. Matter, whatever its form, will bend space-time, which will deflect that path of light. We've tested this like crazy. So if you're looking at, say, a gigantic galaxy cluster and wondering about its mass, you can use the bent light from background objects to figure it out. If you look at a distorted image through a regular glass lens, and you know what shape the undistorted image ought to have, you can use your Knowledge of Physics to figure out the properties of the lens doing the distortion.
We know what galaxies look like because we have, well, a lot of samples. So when light from a distant galaxy passes through a not-so-distant cluster, that image gets distorted, and we can compare it to what galaxies look like without a funhouse mirror and Sherlock out the properties of the cluster doing the lensing. You know, its mass.
So that's a completely totally different method for measuring the mass of big cosmological objects. And guess what answer it gives?
Yup: large objects in the universe are more massive than they appear at first glance.
I don't know about you, but I'm getting the feeling that nature is trying to tell us something.
The evidence has been piling up for decades now, and our options for explaining that evidence—what nature is practically shoving in our faces every time we go to ask the very simple question of how much a galaxy or cluster weighs—have severely narrowed in the time since Rubin's, and especially Zwicky's, initial results.
Can all these combined results—galaxy rotation curves, cluster masses, the cosmic microwave background, early-universe physics, the mere existence of galaxies, and gravitational lensing—be explained by new physics? Is Einstein not enough? It's always a possibility, but as the years go by, it becomes increasingly slim.
Your Place in the Universe Page 17