Foremost among the Jason panel's concerns is the question of whether a fusion reaction that occurs inside a capsule smaller than a fingernail can provide an accurate indication of how a full-size nuclear weapon would detonate. "It's a mistake to assume that NIF experiments are going to be directly relevant to weapons testing," says panelist Richard Garwin, a semiretired IBM physicist who has served on the president's science advisory committee. "The temperatures in the NIF chamber are much lower than they are in actual nuclear weapons, and the amounts of material being tested are much smaller."

NIF physicists apply computer-generated formulas to their experimental data in order to account for such scale differences. But Garwin says that instead of relying on unproven formulas to indirectly assess the status of existing weapons stockpiles, the government would be better off simply replacing the fusion-dependent components of weapons on a regular basis.

Raymond Jeanloz, a physicist at the University of California, Berkeley, who plans to use the finished facility for experiments that replicate fusion conditions inside stars, agrees that the correlations between NIF experiments and real-world explosions aren't yet proven. "That's one of the areas people will want to investigate as the facility gets up and running," Jeanloz says. He points out, though, that change-of-scale calculations are nothing new for most physicists—earthquake motion is regularly simulated with large syrup tanks, for instance.

"When we first started trying to replicate the conditions inside stars on a 100-micrometer scale, people said it couldn't be done," scientific group leader Bruce Remington says. "A decade later, hundreds of laboratory astrophysics papers have been published in peer-reviewed journals, and about 50 percent of them address scaling issues."

The NIF's scaled calculations will be accurate, Remington contends, because staff physicists have an intimate knowledge of how the physical properties of matter can change depending on how much of it is present. For instance, small amounts of the ionized gases generated in fusion reactions produce proportionally more heat-creating friction than larger amounts, and weapons simulation formulas are tailored to account for that factor.

The Jason panel is also concerned about whether the NIF's lasers will be able to ignite fusion at all. Because of laser backscattering—incoming laser beams are distorted by the superheated clouds of plasma forming around them—light leaks out of the capsule that encloses the target molecules, lessening the intensity of the beams. NIF scientists claim, however, that the concern is overblown and that the lasers' focusing capabilities will minimize excess plasma formation.

Another potential issue is whether the facility's precision-ground focusing lenses will be able to stand up to the heat. In previous tests under conditions similar to those the NIF researchers hope to create, optical equipment has simply shattered. "There's a question as to whether the glass can withstand such a high degree of power," says Robert Civiak, a retired physicist who compiled an independent report on the NIF for the White House's Office of Management and Budget.

Although Remington acknowledges that Civiak's concern has some merit, he says the solution is for facility administrators to be selective about allowing researchers to operate the laser at full strength.

Novelist E.L. Doctorow has likened the process of discovery and new thought to driving at night, on an unfamiliar highway, with headlights that barely illuminate the road ahead. Because NIF scientists cannot foresee small or large problems that may arise as they move toward the goal of achieving fusion—or, for that matter, all the exciting science that could come from NIF experiments—the ultimate value of the project cannot be fully assessed in advance. But of course the government agencies bankrolling the NIF would prefer the entire highway to be brightly lit.

"We're going to go through some difficult learning exercises," Remington says. "The facility's hardware is working well, but we have to make all of the beams work together, shoot at the same time, and point at the same place. There'll be a lot of hand-wringing figuring that out."

Meanwhile, for other NIF employees, work continues as usual, though they remain cognizant that factors beyond their control—whether fiscal or physical—could cut short their years-long quest. "We have high confidence that we're building the right thing," NIF physicist John D. Lindl says. "But the bottom line is that NIF is still an experiment, and the only way to prove fusion can be done is to demonstrate it with a facility that works."