They say academic arguments are so vicious because the stakes are so low, but Andrei Linde and Paul Steinhardt are arguing about something huge, something truly cosmic: the fate of the universe.

Linde and Steinhardt are two of the premier cosmologists of America and, as far as we know, of this sector of the Milky Way galaxy. They study the origin, structure and destiny of the universe, all the classic 3-in-the-morning dorm-room questions that threaten to blow your mind. How big is the universe? Is it infinite or finite, and if it's finite, where's the edge? How did it begin? What will happen to it? Why does anything exist at all?

Linde, a Stanford professor, Russian-born, theatrical (he performs magic tricks and acrobatics in his spare time), warned two years ago that the universe might collapse in as little as 11 billion years. Recently he revised his prediction, saying that the end might not come for at least 24 billion years. "Cosmic doomsday delayed" reported one online science journal earlier this month, throwing a sop to Democrats desperate for good news.

Regardless of the timing, The End as Linde sees it would prove to be extremely fatal to any living thing. Linde says the entire cosmos will begin to contract, the galaxies rushing back toward one another, space itself heating up, our oceans boiling off, Earthlife frying, until everything slams together in a cataclysmic Big Crunch, all matter and energy, and indeed time and space itself, compressed into an infinitely dense black hole.

"All these galaxies will crash into each other, and the sky will be shining in blue light, ultraviolet, then X-ray; the temperature will be very large and everything will be just evaporated," Linde says.

He isn't panicking. He speaks with a serenity that comes from knowing that this bad stuff is a long ways away. He says, reassuringly, "Mankind will find many ways of killing itself much earlier."

But now comes Steinhardt, a Princeton professor, with a scenario that is just the slightest bit sunnier. He suggests that our universe may be just one of two universes, or "branes," that periodically collide. These two branes, Steinhardt hypothesizes, have been bouncing off one another for untold billions of years. The next time the invisible brane comes crashing back into our universe, we'll see the laws of physics start to change.

Physical "constants" will cease being constant. Gravity will strengthen (apples falling from trees will hit the ground faster). The universe will collapse, as in the Linde scenario -- "Unless we found a way of protecting ourselves, we would be evaporated in the crash between these branes," Steinhardt says -- but there's good news at his version of The End. There's a bounce! The whole enterprise begins expanding again, inflating, and becoming once again a universe that is sufficiently comfortable to permit people to wear shorts and T-shirts.

Here is what Steinhardt says of Linde: "He will tell you many things depending on which day you catch him."

Linde on Steinhardt's bouncing branes: "Oh my God. Oh my God." (Deep breath.) "It is a very interesting but a very wild speculation. . . . It's not science."

The two men don't exactly rage at one another, but in phone interviews each makes it clear he finds the other irritating and unpersuasive.

Every cosmologist would like to become known as the person who figured out the destiny of the universe, but the facts are sparse and ambiguous. Cosmologists tend to do a fair bit of arm waving. You get the feeling that a clever scientist could prove on paper that the universe is made of cheese.

At some level it's the ultimate esoteric topic. Our lives are lived in a certain slice of space-time, so narrow that it really doesn't matter in any practical sense whether the universe will expand or contract or contract-and-bounce. But we also wouldn't be human if we didn't try to decode the universe. There's a story there, written in starlight.

Cosmologists are rambunctiously creative, painting pictures that inspire their colleagues to engage in furious art criticism. There are art movements, and splinter groups, and mavericks. Someday the current fascination with string theory, for example (you know: the universe is made of very tiny strings that vibrate at different frequencies), might be viewed as something lovely but a bit beside the point, like Picasso's blue period.

"The observations are not that definitive, so we have to build stories behind our observations," says Alan Guth, a pioneering MIT cosmologist.

The race to figure out the fate of the universe has grown more intense in recent years as a result of a startling discovery. In 1998, the Hubble Space Telescope obtained images of distant supernovae (exploding stars) that revealed that the universe not only is expanding but is doing so at an accelerating rate. A few years later, new observations showed that, many billions of years ago, the expansion was actually decelerating. So the cosmos apparently downshifts and upshifts. Second gear awhile back, fifth gear today.

Something is causing the current acceleration, and this force, whatever it is, seems to account for about 70 percent of the total energy in the universe. Scientists call it "dark energy." That sounds better than "secret energy" or "invisible energy" or "we-got-no-clue energy."

Scientists had long thought that the universe might have just the right amount of stuff in it to allow gravity to slow the cosmic expansion to a nice, stable, steady cruising speed, such that the universe would always be around, always habitable. But what we see through telescopes doesn't support that happy scenario.

It looks for all the world as though dark energy will rip the universe apart. It might take a trillion years, but everything we see -- every star, planet, asteroid, comet, speck of dust and molecule of gas -- will decay and dissipate, and in the end there will be essentially nothing at all -- the universe ending with a whimper.

"The universe will go on expanding forever. And will get cold and dark and presumably ultimately lifeless," explains Guth, describing this scenario.

Scientists have to follow the data, but they'd also like to find a model for the future that's more interesting than an eternal wasteland. "I don't like it," Steinhardt says.

Perhaps, Steinhardt says, dark energy is not a constant force but something that can grow less dominant over time. If dark energy weakens, gravity could reassert itself as a shaper of the universe, and the whole business could contract again.

Unfortunately, none of this can be measured directly at the moment. There's no such thing as a dark energy detector. No one has figured out how to catch this particular kind of lightning in a bottle.

It could be that we're nowhere close to understanding what's happening in the universe. The place always turns out to be bigger and more complicated than we thought. In the 1920s we discovered that we live in but one of countless galaxies of stars, and that all these galaxies are moving away from one another. The universe is expanding. Hit rewind and we see that about 13.7 billion years ago, everything in the universe was compressed into an unimaginably hot, dense wad of energy. Something happened to transform that cosmic knot into a vast universe. It all went bang -- but why? How?

In the early 1980s, Guth, Linde, Steinhardt and others developed a provocative theory called "inflation," which suggests that the tiny embryonic universe went through an extremely brief (fraction of a second) inflationary epoch, increasing from something smaller than a pinhead to something cosmic in scale.

Steinhardt has more recently developed the hypothesis of colliding branes, an attempt to incorporate string theory into cosmology. Steinhardt's scenario has many moving parts and requires unseen dimensions of space. He admits it's still a maverick view, and perhaps it will always be hard to get converts to a new theory that is built on the assumption that we're going to collide with a hidden universe.

Steinhardt, however, has no monopoly on mind-bending theories. Linde has spent years talking about the possibility that a scientist might be able to create an entire universe in a laboratory.

You'd only need a speck of matter, Linde says. If you could get it to start inflating, it would eventually turn into a whole new gigantic galaxy-filled universe. For reasons that are pretty much lost in translation between scientist and journalist, the creation of this new universe somehow doesn't destroy the laboratory, or even break any beakers or flasks, but rather the new cosmos squirts off into unseen dimensions while continuing to look like a tiny little thing to the physicist.

Linde sees this as a potential means of surviving the demise of our own universe.

"Maybe we can create the universe and transfer our knowledge to the inhabitants of the universe that is created in our laboratory," Linde says.

Or maybe, he suggests, someone already did that. The creator of our cosmos might be a "physicist hacker" in some other universe, he says. Again, this is a theory that has yet to yield a lot of support.

The ultimate question is why the universe exists at all. Guth says the equations for inflation allow the universe to expand forever, with new universes constantly popping into existence, but the equations don't permit an eternal past. There has to be a beginning, back there somewhere. Where did that first pulse, that first little spark, come from?

Guth says he's worked out scenarios in which the laws of physics allow something to pop into existence from nothing. But he adds, "I was implicitly assuming that the laws of physics already existed, even when there was no space, no time, no matter. I think that's all I need to assume. Nevertheless I am clearly making a big assumption there. That does raise the question of what caused the laws of physics, where they came from."

What's his answer?

"I don't have the foggiest idea."

So let's quickly summarize where we are: We're not sure if the universe will keep expanding or start contracting. We don't know if The End, if there ever is such a thing, is many billions of years away or many trillions of years away. We don't know the nature of dark energy, this strange force that's shaping the universe. We don't know if constants are constant. We don't know why there are laws of physics. We don't know why there's something rather than nothing.

And thus we might guess that scientists will not soon put theologians out of business.

There's a story there, written in starlight: The deepest portrait of the universe ever recorded, by a Hubble camera. Premier cosmologists Paul Steinhardt, left, of Princeton and Andrei Linde of Stanford have dueling scenarios about the doomsday question.