We have completed maintenance on DiscoverMagazine.com and action may be required on your account. Learn More

Crisis in the Cosmos

There's not one crisis but two: the universe seems to be younger than the stars in it, and a huge chunk of it is headed in the wrong.

By Sam Flamsteed
Mar 1, 1995 6:00 AMNov 12, 2019 6:06 AM


Sign up for our email newsletter for the latest science news

Fewer than 70 years have passed since astronomers discovered that the universe is expanding. It’s been only 30 since they first detected the cosmic microwave background--the feeble, universe-wide glow that is the main evidence for the Big Bang. Yet for most of that time it has seemed as if they were making steady progress. Like the universe itself, our understanding of it, having begun with a bang in the twentieth century, seemed to be expanding constantly, rapidly, and inexorably. One could almost imagine a time when we would have figured it all out.

These days that goal looks distant indeed; these days cosmology seems to be collapsing in on itself. In the past few years two new discoveries have come along, observations so disconcerting that they threaten to shake the field to its foundation. The first is that galaxies in a huge chunk of the universe, a region at least 1 billion light-years across that includes our own Milky Way, appear to be moving, all in the same direction, at about 435 miles per second, or 1.56 million miles per hour. Even if you’re a cosmologist and are used to such ridiculous distances and speeds, you have trouble explaining that observation. The only force that could set such a large fraction of the universe in motion is gravity--and there just isn’t enough mass around to generate that much gravity. Even when cosmologists resort to their favorite deus ex machina and say that 90 percent of the universe is made of some invisible dark matter--identity unknown but full of gravity--they can’t get that many galaxies to move that fast.

The second observation, made with the newly repaired Hubble Space Telescope, is even more startling. The Hubble has allowed astronomers to make the most credible measurement to date of the age of the universe--of how long it’s been since the Big Bang. They have found that the universe is somewhere between 8 and 12 billion years old. Yet there is almost no doubt that the oldest stars in the Milky Way, which live in the globular star clusters that orbit the galaxy’s central bulge, are at least 14 billion years old, and probably even more ancient than that. A universe younger than the stars it contains is, to say the least, a fundamental contradiction.

Too little mass, too little time--either problem alone would be disturbing. Taken together they raise the specter of a scientific revolution, a shift in the cosmological worldview in which some fundamental assumptions in cosmological theory--perhaps even the Big Bang itself--will have to give. It would be premature to panic, says David Weinberg, an astronomer at Ohio State University. But if these results are confirmed, we theorists will be in trouble. We really have no good ways of explaining these observations.

It would be one thing if the observers in question were inexperienced or untrustworthy. In fact, they’re quite the opposite. Take the age-of-the-universe measurement. The lead astronomer on the 14-member team that made the discovery is Wendy Freedman, of the Carnegie Observatories in Pasadena, California. Few observers in the business are more respected for care and competence than she is--and several of those are on her team.

Freedman and her colleagues didn’t measure the age of the universe directly; that would be impossible. What they did was to measure the distance from Earth to the galaxy M100 in the Virgo cluster of galaxies. By measuring that distance, they were hoping ultimately to deduce how much the universe has expanded so far, and therefore how long it has been expanding since the Big Bang.

As the universe expands, galaxies get farther apart. The farther apart they already are, the faster they separate, as new space pops up between them at a constant rate. That rate is the Hubble constant, named after Edwin Hubble, the man who discovered in 1929 that other galaxies are receding from our own Milky Way--their light waves were stretched out, he found, and thus shifted toward the red, long-wavelength end of the spectrum. The amount of a galaxy’s redshift tells astronomers how fast the galaxy is moving away from us. If they also know how far away it is, they can calculate the rate at which the universe is expanding between us and it: the expansion rate is just the galaxy’s velocity divided by its distance. From the expansion rate they can then determine how long ago it was that the universe was crunched into a single infinitesimal point, at the moment of the Big Bang.

Unfortunately it’s not as easy as it sounds. A galaxy’s motion is produced by more than just the expansion of space. Galaxies tend to come in clusters, and the members of a cluster tend to gravitate toward one another. Our close neighbor the Andromeda galaxy, for instance, is in the Milky Way’s thrall to such an extent that it is not expanding away from us at all: it’s coming toward us, and its light is blueshifted rather than redshifted. To measure the cosmic expansion rate, astronomers have to look at a galaxy far enough away for its expansion velocity to dwarf whatever local velocity it may have within its cluster. But then they encounter a second problem: the farther away a galaxy is, the harder it is to measure its distance precisely.

Freedman and her colleagues navigated a compromise path between these two problems. They used the redshift of the Coma cluster of galaxies, which is so far from us that its velocity is mostly cosmic expansion. But the Coma cluster’s distance is too great to measure directly. Instead Freedman’s team got the most precise estimate to date of the distance to Coma by using their measurement of the distance to M100 in Virgo as a yardstick. The distance to Virgo had never been accurately measured either- -until the Hubble Space Telescope made it possible for Freedman’s team to do it.

The technique they used was the same one that Edwin Hubble himself had used to prove that galaxies are star systems beyond the Milky Way. It involves a type of star called a Cepheid variable, one that varies in luminosity over a regular period. That period is a good indicator of the star’s intrinsic brightness--the longer the period, the brighter the star-- which makes a Cepheid a good distance indicator: just compare its intrinsic brightness to the much dimmer dot you see through your telescope and you know how far away it and its home galaxy are from Earth.

But picking out individual Cepheids in a galaxy as far away as Virgo is hard, if not impossible, with a ground-based telescope; Earth’s atmosphere blurs the light too much. That’s why the Hubble telescope was crucial to Freedman’s measurements. When NASA decided to go ahead with the Hubble telescope back in the 1970s, she says, finding Cepheids in the Virgo cluster was one of the things it was explicitly designed to do. The Challenger disaster in 1986 delayed the launch of the telescope, however, and when it finally reached orbit in 1990 it turned out to have a misshapen mirror. But Freedman and her colleagues were among the first observers to get access to the telescope after shuttle astronauts installed its corrective optics in December 1993.

We began looking at M100, our first target galaxy, as part of the testing the engineers had to do to make sure the repairs had been done properly, Freedman says, and right away we could tell that we were going to be able to find Cepheids. Since they had to catch Cepheids at several points on the cycle of waxing and waning to make sure they could nail down the pulsation periods precisely, they made 12 sets of observations over 60 days. By July, says Freedman, we had our first confirmed Cepheid. Before they were through analyzing data at the end of the summer, they’d looked at 40,000 stars and found 20 Cepheids they were sure of.

Those 20 were the markers that allowed Freedman and her collaborators to put the distance to M100 at 56 million light-years, give or take 6 million. To get from there to the Coma cluster, they assumed first that the distance to M100 was the average distance of a bunch of spiral galaxies in the Virgo cluster, and second that those galaxies had the same intrinsic brightness as a comparable bunch of spirals in Coma. The degree to which the light from Coma had been dimmed before reaching Earth told them the cluster was about 5.5 times farther away than Virgo--a little over 300 million light-years away.

Dividing the cluster’s recession speed by that distance gave them the Hubble constant, in kilometers per second of expansion speed per megaparsec (3.26 million light-years). Freedman and company claim the Hubble constant is 80, plus or minus 17. In other words, for each megaparsec farther out into the universe, the galaxies are receding 80 kilometers per second faster, or 50 miles per second. If space has always been expanding at that same rate, the researchers conclude, the universe is quite young, between 8 and 12 billion years old.

The biggest drawback to Freedman’s results is the relatively huge uncertainty--the plus or minus 17. Part of it comes directly from the uncertainty in the distance measurement to M100. But most comes from the researchers’ assumption that M100 represents the average distance to all the galaxies in Virgo--whereas in fact they don’t know whether it lies in the middle of the cluster, on the near edge, or on the far edge. Freedman and her fellow observers are quick to acknowledge that this single observation is not sufficient to nail down the Hubble constant to anyone’s satisfaction; they would prefer to have observations of many more galaxies, and they’re planning to get them. What’s really significant here, says Joel Primack, a theorist at the University of California at Santa Cruz, is that there are hundreds of galaxies at the distance of Virgo, and they’ve shown they can find Cepheids out that far. The more galaxies they’ve got, the less uncertain their numbers will be.

But in the meantime, anyone who is counting on the 17-point uncertainty to ward off a cosmological crisis had better think again. Freedman and her group are hardly the only ones trying to measure the Hubble constant; they’re not even the only ones looking for Cepheids in Virgo. Just a month before they published their results, in fact, Michael Pierce of Indiana University managed to spot a few Cepheids in another Virgo galaxy, NGC 4571, using a ground-based telescope equipped with adaptive optics--essentially a computer-controlled flexible mirror that compensates for atmospheric blurring. Pierce’s conclusion: the Hubble constant is 87. And while his measurement is considered less reliable than Freedman’s, several other astronomers have used entirely independent techniques to come up with their own answers, most of which point to a young universe. At Harvard, for example, Robert Kirshner and several collaborators are looking at clouds of hot gas generated in supernova explosions to try to measure intergalactic distances (they get a constant of 73). John Tonry of MIT is looking at the statistical variation in patches of light within bright galaxies to deduce the brightness of individual stars (he gets 80). Only Allan Sandage, a Carnegie colleague of Freedman’s, gets a significantly lower number. Sandage believes that some kinds of supernova can serve as standard candles of known intrinsic brightness, much as Cepheids do; he argues that the Hubble constant is around 50 and that the universe is a comfortable 15 to 20 billion years old.

You’d naturally think, says theorist William Press of the Harvard-Smithsonian Center for Astrophysics, that all these numbers and all these uncertainties mean we can’t really be definitive about the Hubble constant yet. But by using a statistical technique called Bayesian analysis, Press argues, you can at least decide where the most probable value lies. Since the technique works by selecting only those values that are not likely to have systematic errors, it is a more reliable synthesis of many different measurements than a simple average, which can easily be thrown off by a few errant values. I can say with high confidence, says Press, that there’s a 50-50 chance that the Hubble constant lies between 71 and 77, and a 95 percent chance it’s between 66 and 82. Even if it’s closer to 66, that makes the universe 10 to 14 billion years old.

The oldest stars, some cosmologists claim hopefully, might just be able to squeeze into a 14-billion-year-old universe--but the chances are slim. The stars are probably older than that. We are really happier with 17, says Pierre Demarque, a stellar evolution theorist at Yale. The cosmologists are constantly pressuring us to stretch this a little further, but believe me, we can’t. And our group consistently gets younger ages for the stars than most others do. The stars could easily be as old as 19 or 20 billion years, or even older.

Cosmologists acknowledge that their star-watching colleagues have a lot more data on stars and the nuclear reactions that power them, along with better theories to tie the data together, than they do. Knowing how long it should take for a star to burn hydrogen into helium, helium into lithium, and so on, researchers like Demarque can tell from the composition of stars how old they are likely to be. The idea that we really understand the ages of the oldest stars is a widely held belief in astrophysics, says Press, and I really don’t expect that it will change. With that easy escape not available to them, cosmologists are facing the unpleasant fact that some fundamental principle of the universe still eludes them.

The situation would be bad enough without the discovery that a billion light-years’ worth of matter is sliding sideways across the universe. This distressing news comes from two young observers who are every bit as respected as Freedman and her colleagues--though perhaps not quite as dignified at all times. Once a week, at least, Marc Postman answers the telephone in his office at the Space Telescope Science Institute, in Baltimore, and hears a strange robotic chirping emanating from the handset. Postman immediately grabs a key chain from his desk and holds it up to the phone. The key chain is a small box with a picture of a tree frog on one side. He squeezes it, sending an identical chirp back out to the National Optical Astronomy Observatories in Tucson, Arizona. Only now does Postman settle into conversation with the party on the other line, who, he knows, can be none other than his close friend and collaborator, Tod Lauer.

Back in 1989, Lauer and Postman embarked on the series of observations they called Project Warpfire. Their plan was to go out looking for cosmic convergence, which sounds like science fiction too but isn’t. The idea of cosmic convergence is implicit in Freedman’s strategy for measuring the age of the universe; it’s the idea that at some very large distance, the motions of galaxies due to their gravitational influence on one another would become completely negligible compared with the motion caused by the expansion of the universe. For decades astronomers had been confident that convergence would happen at distances of perhaps 100 million light-years. That assumption was destroyed in the mid-1980s with the discovery of the Great Attractor. A team of astronomers known as the Seven Samurai had found that all the galaxies in a volume of space 100 or 200 million light-years across were moving, not just within their own little clusters, but in concert toward a point in the sky between the constellations Hydra and Centaurus. The Samurai concluded that a huge concentration of mass--a supercluster of galaxies--was pulling them.

The Great Attractor was hard for any cosmological theory to explain; no theoretical model predicted mass concentrations that large. Given enough time to tinker with their equations, though, the theorists thought they could probably deal with the problem--as long as there was nothing bigger than the Great Attractor out there, no even larger troop of galaxies embarked on a forced march across the expanding universe. Lauer and Postman set out to test that proposition by seeing if they could find the point of cosmic convergence--the distance beyond which galaxies are essentially at rest with respect to cosmic expansion.

As a frame of reference they used something that is definitely fixed with respect to expansion: the cosmic microwave background. The CMB is the afterglow of the Big Bang itself. It is radiation that comes at us from every direction, from the edge of the visible universe, and it is essentially uniform in all directions. In one direction, though, it has a hot pole: its wavelength in that direction is ever so slightly shorter, bluer, hotter. Astronomers attribute that to the motion of Earth through space: our planet orbits the sun, which orbits the core of the Milky Way, which is being pulled by Andromeda and by Virgo and by the Great Attractor. The sum of all those motions means the Earth has to be moving with respect to the CMB and that a blueshift in the direction of motion is inevitable.

If you measured Earth’s motion against a collection of objects that were themselves at rest with respect to the CMB, though, you should see precisely the same degree of blueshifting, in the same part of the sky. That’s what Lauer and Postman set out to look for. We wanted to go very deep, says Lauer, so we needed objects that are visible at great distances and that are pretty uniform in brightness.

The objects they chose were the individual brightest galaxies in a large number of distant clusters. The brightest galaxy in any cluster, astronomers have found in studies of nearby clusters, always seems to have more or less the same intrinsic brightness. That meant Lauer and Postman could use the relative brightness of the brightest galaxies to get a rough idea of their distance; from each galaxy’s distance they could then calculate what its recession speed should be. (Recession speed increases with distance--that’s Hubble’s law.) Subtracting that from its overall redshift gave them the component of its motion that did not come from cosmic expansion. Finally, averaging together the leftover redshifts of all the galaxies would eliminate their purely local movements within clusters and leave only the overall movement of the entire collection relative to the CMB. Lauer and Postman fully expected that movement to be zero; they expected to find cosmic convergence. In that case, Earth would appear to be moving with respect to the framework of galaxies exactly as it is moving with respect to the CMB.

The Project Warpfire observations went on for a year and a half, with Lauer and Postman taking every precaution they could think of to keep themselves honest. To cut down on the chance of error caused by slight variations in brightness, they looked at 119 galaxy clusters arranged all over the sky out to a distance of around 600 million light-years. We had to be careful that the instruments were working exactly the same at telescopes in the Northern and Southern hemispheres, and from winter to summer, says Postman. We observed lots of galaxies more than once, from different locations, to check ourselves. By the fall of 1991, the data were in and analyzed--and they were incredible. What we found, says Lauer, was that the Earth is indeed moving with respect to these distant galaxies. But it’s moving in a different direction with respect to the microwave background--the two are off by about 75 degrees.

What could this mean? Earth can’t be streaking through space in two entirely different directions at once. The only plausible conclusion that Lauer and Postman could come up with was that some vast cosmic current, far larger even than the one caused by the hypothetical Great Attractor, is sweeping along the entire collection of hapless galaxies-- including, of course, the Milky Way and its neighbors--at a velocity of 435 miles per second with respect to the CMB. From Earth out to a distance of 600 million light-years on all sides, everything is being carried toward some distant point, somewhere beyond Orion.

An analogy may help you understand what Lauer and Postman are claiming. Imagine you’re sitting in a rowboat in some body of water. You’re rowing hard toward a friend you see sitting in another boat. Glancing off to one side, though, you notice that you’re also closing fast with some rocks on shore. How could you be approaching both your friend and the rocks at the same time? The answer is that you and your friend are in a river. As you (Earth) row toward your friend (Earth’s heading in the galaxy framework), you’re also being carried downstream. Your net motion is the sum of those two motions, and it’s across the river, toward the rocks (the hot pole of the CMB). Postman and Lauer are saying, essentially, that they’ve discovered a giant river in space. The hundred or so galaxies they’ve observed are like so many boats--each with its own local motion, yet all being carried downstream.

When you get a result like that, says Lauer, you have to doubt it. You try everything you can think of to make it go away. They tried for a year; they looked for every conceivable way they could be fooling themselves. They couldn’t find any. Neither could anyone else. We finally presented our observations at a meeting in Milan in 1992, says Postman. My wife says she’s never seen me so nervous--and Tod was the one giving the talk. As soon as he finished, 20 hands shot up. But the questions were really disappointing. People thought they could explain the whole thing away in 10 minutes, as though we hadn’t really anticipated the obvious problems.

Now, more than two years later, nobody is making that mistake. People have inspected the work minutely by now, says Sandra Faber, at Santa Cruz, one of the original Seven Samurai and, as it happens, Lauer’s thesis adviser. Either it’s some sort of statistical fluke that nobody could have anticipated, or it’s correct, which tells us something profound about the universe.

But what? It’s almost impossible to say at this point. Even before these distressing new observations came along, theorists faced a maddeningly long list of unanswered cosmological questions. What is the dark matter made of? How much of it is out there, and how is it distributed? How did the structure of the present-day universe evolve--the stars and galaxies and galaxy clusters and superclusters--from the smooth and uniform explosion of the Big Bang? Now they have two more questions: How can the universe be younger than its stars? And how can so much of it be charging off toward Orion? All these problems are intertwined. And a plausible solution to any one of them often has the unfortunate property of making a related problem worse.

Take dark matter, for instance. Astronomers know it’s out there because they can see the influence its gravity has on visible matter--on the rotation of spiral galaxies, say, or on the motion of galaxies within clusters. They think there’s at least 10 times more invisible stuff than stars, galaxies, and gas clouds. But precisely how much dark matter is out there? The answer may determine whether the universe is open or closed: whether it will expand forever or eventually collapse back in on itself.

Dark matter, after all, exerts a braking force on the expansion. When astronomers talk about dark matter, they refer to a number called Omega, which is the ratio of the amount of matter in the universe to the amount needed to keep the universe from expanding. A universe that will eventually collapse under the weight of its dark matter has an Omega greater than one (though few astronomers advocate this possibility). Likewise, an Omega of less than one refers to a lightweight universe that will expand forever.

But Omega affects the age of the universe as well as its fate. Say there’s a lot of dark matter and the expansion is slowing down significantly. That means the universe is even younger than it appears to be. The reasoning is simple: if the universe expanded quickly in the beginning, it took less time to get to its present shape. Freedman and her colleagues took this possibility into account. The universe, they say, is 8 billion years old if the Hubble constant is precisely 80 and there’s enough dark matter to slow the expansion to a crawl; if the universe has no more dark matter than we currently know about, it’s 12 billion years old.

With data showing the universe to be younger than the stars it contains, why not just assume that Omega is only a fraction of one and the universe is therefore older than it looks? Because then cosmologists might have to junk something else they’re fond of: the inflationary-universe theory, which says that the cosmos expanded exponentially fast for a fraction of a second immediately after the Big Bang. The theory came out of particle physics, but cosmologists love it because it explains some thorny problems of theirs, such as why the CMB is so smooth. And one of its most fundamental predictions is that Omega is extremely close to one.

So even though all the visible mass in the universe amounts to a paltry Omega of 0.01; even though the dark matter whose gravitational effects we’ve been able to observe only brings Omega up to 0.1 or 0.2; and even though no one has yet found clear evidence of the exotic particles or massive neutrinos that might provide enough dark matter to bring Omega closer to one--in spite of all that, there is a strong constituency for an Omega of precisely one. It’s not just the influence of the inflationary theory; to some researchers a universe balanced perfectly between expanding forever and collapsing under its own weight has a powerful aesthetic appeal. To others, though, Omega = one is a prejudice cosmologists are going to have to give up if they want a universe older than the stars in it. I really think that this is what the recent Hubble constant measurements are telling us, says Jeremiah Ostriker, chairman of Princeton’s astrophysics department. Omega is significantly less than one, and we live in an open universe after all.

If Omega is indeed much less than one, then it’s conceivable the age problem could go away. The error in Freedman’s measurement would have to be as large as possible and in the right direction to make the Hubble constant as low as possible--that is, M100 must happen to be well in the foreground of the Virgo cluster rather than in the middle or the background. And the oldest stars would have to be as young as they possibly could be, around 14 billion years old. Then maybe--just maybe--the universe would be old enough to contain them. But then it would run afoul of Lauer and Postman: the large-scale motion problem they’ve identified would become even worse.

A lot of dark matter, you see, may imply a universe that’s too young, but its gravity is handy for pulling the smooth soup of visible matter produced by the Big Bang into galaxies, clusters of galaxies, and clusters of clusters. And it’s positively essential for pulling matter into the huge gravitating structure needed to haul Lauer and Postman’s river of galaxies across the sky. That is not to imply that anyone knows how to do that even with an abundance of dark matter. A group of theorists, including Ostriker and Michael Strauss of the Institute for Advanced Study, recently used elaborate computer simulations to test the ability of various dark- matter theories to produce the galaxy streaming seen by Lauer and Postman. We basically concluded, says Strauss, that their result is incompatible with all the models at a 95 percent confidence level. In other words, there’s only a 5 percent chance that such motions could exist under any known theory. But the theories that do the worst are the ones with the least dark matter and the lowest Omega. Yet if you try to fix the Lauer- Postman problem by raising Omega, you aggravate the age crisis. It’s a cosmological catch-22.

Of course, it’s still not entirely out of the question that one or both crises will turn out to be figments of faulty observations. Freedman and her colleagues are already out rechecking their Hubble constant measurements. Lauer and Postman, collaborating with Strauss now, have recently embarked on an even deeper sky survey; they’re going to probe out to perhaps 800 million light-years this time and increase their sample from 119 galaxies to more than 600. If they still see bulk flows at that scale, says Weinberg of Ohio State, then I’m going to be pretty worried. It’ll be a dramatic step if we have to throw out our existing models. But it certainly could be true that our models are missing something very basic, some sort of physics we haven’t been thinking about.

Some of the new physics that has been suggested recently has the smell of desperation to it. And some of it isn’t new at all. For example, to escape the young-universe dilemma, some researchers are dusting off a notion that Einstein first floated more than 75 years ago--and later discarded. Realizing that his equations of general relativity implied an expanding or contracting universe, and lacking evidence for either, Einstein threw in a cosmological constant: an all-pervasive energy field, a kind of cosmic antigravity, to keep the universe at equilibrium. Then it was discovered that the universe really is expanding. Einstein called the cosmological constant the greatest blunder of my life, and for decades nobody argued with him.

More recently, though, and for completely different reasons, particle physicists have decided the constant wasn’t such a bad idea after all. And now some cosmologists think it could help explain away the age crisis. The energy of the constant would exert an outward pressure on the entire universe, making it expand at an accelerating rate. That would mean the expansion rate was once much slower than it is now--and that it’s been much longer since the Big Bang than you would find by calculating backward using the current expansion rate. Unfortunately, there is still no observational evidence that the cosmological constant is real; some evidence actually points against it. In any case, it wouldn’t solve the Lauer-Postman problem.

One unusual solution to that problem would be to conclude that the universe is lopsided. The conventional view is that after the Big Bang, space and time spread out more or less evenly. Uniformity is something we’ve always assumed, says Princeton theorist Bohdan Paczy´nski. But there is no reason, a priori, to believe it’s true. The evidence for a uniform Big Bang is the uniform CMB: its intensity is nearly the same in all directions--except for that subtle hot pole that loomed so large in Lauer and Postman’s calculations. They and everyone else have taken the hot pole as evidence of Earth’s motion with respect to the CMB. But, Paczy´nski asks, what if instead the Big Bang was really a lopsided explosion, hotter in one direction than the other? Then Earth wouldn’t be moving toward the hot pole at all. It would only be moving with respect to Lauer and Postman’s distant galaxies, which would themselves be more or less fixed with respect to the universe. The contradiction Lauer and Postman identified would vanish; their rushing river of galaxies would become a tranquil pond (albeit an expanding one).

Another possibility, other researchers suggest, is that the universe is lopsided in a different way. What if galaxies on one side of the sky were intrinsically dimmer than comparable galaxies on the other side? That too would nix the Lauer-Postman results. To estimate the distance to their galaxy clusters--and thus calculate how much of the clusters’ motion was due to cosmic expansion--they assumed that the brightest galaxy in each cluster was a standard candle of consistent brightness. If that assumption was wrong, their entire result is suspect. Harvard astronomers Press and Kirshner believe the assumption is wrong. They recently observed 13 supernovas scattered around the sky in distant galaxies and found them to be virtually at rest with respect to the microwave radiation. They suspect that galaxies vary systematically in brightness, that supernovas are the more reliable standard candles--and that Lauer and Postman’s river of galaxies is a misinterpretation of their own data.

If Press and Kirshner are right, that would eliminate one cosmological crisis. But in its place would come another. It’s certainly possible they’re right, says Postman. But that would be an equally dramatic result. You’d have to come up with an explanation for why one part of the sky is systematically different from another.

And no one, it seems clear, has good explanations right now, not for any of the problems that bedevil cosmology, and certainly not for all of them at once. The field is in a troubled state, a disconcerting or an exciting one, depending on your personality--a state in which even the most basic assumptions seem open to question. Maybe the microwave background has nothing to do with the Big Bang after all but is due to some entirely different phenomenon. Maybe redshifts aren’t really due to recessional velocity, and the universe isn’t expanding. Maybe we don’t really understand gravity, which would throw all cosmological theories into the trash--even the Big Bang itself. Few cosmologists think that’s at all likely. But no reasonable cosmologist would claim that the Big Bang is the ultimate theory. With enough contradictions and inconsistencies, it could eventually be overthrown in favor of a model--perhaps a more complicated and difficult one--that better fits the facts. We like to think of the universe as simple and comprehensible, says Joel Primack, but the universe is under no obligation to live up to our expectations.

1 free article left
Want More? Get unlimited access for as low as $1.99/month

Already a subscriber?

Register or Log In

1 free articleSubscribe
Discover Magazine Logo
Want more?

Keep reading for as low as $1.99!


Already a subscriber?

Register or Log In

More From Discover
Recommendations From Our Store
Shop Now
Stay Curious
Our List

Sign up for our weekly science updates.

To The Magazine

Save up to 40% off the cover price when you subscribe to Discover magazine.

Copyright © 2024 Kalmbach Media Co.