Radiance by Carter Scholz. I looked at this book several times before I picked it up from the library last week, trying to think of where I'd seen that name before. Eventually, it leaped out at me from Patrick Nielsen Hayden's weblog-- Carter Scholz was one of the authors in the excellent collection Starlight 1, and the similarly excellent Starlight 2, both of which you should buy at once via Patrick's web site.

I didn't remember Scholz's stories in any detail, but the subject matter of this one seemed promising, so I picked it up on the pre-vacation library run. It's a story about physicists doing weapons research at a thinly disguised Lawrence Livermore Lab (referred to only as "The Lab" throughout):

J Section. Research And Development In Advanced Nuclear Concepts. Concepts as in weapons. Advanced as in not working yet. Radiance's charter was to develop energy weapons of all types, but Highet's hope and pet was the Superbright: an orbiting battle station of hairthin rods webbed around a nuclear bomb. The bomb's ignition would charge the rods with energy, focused into beams that would flash out to strike down enemy missiles, all in the microsecond before the station consumed itself in nuclear fire.

So far the beams flashed out only in theory. The theory, originated by Null, seemed to Quine sound, but the more he studied his computer model, the less he understood why any of Null's tests had ever produced the ghost of a beam. Yet the farther tests fell behind expectations, the more strident became Highet's public claims. Warren Slater, in charge of testing, had resigned in protest. His letter of resignation was classified and squelched. Bernd Dietz was given interim charge of testing, and to Quine fell the task of finding in disappointing test data any optimism about the promised results.

Not the best vacation reading material, really, but I read about two thirds of it before we left, and only finished it on vacation.

The first section of the book is very good, covering a sort of mid-life crisis of conscience for Phillip Quine, who sort of blundered into the strange and tangled world of the Lab, and is beginning to have doubts about his career path. Not only because he questions the morality of weapons work, but because the crazed funding-based culture of the Lab, encouraged by its ambitious director, Leo Highet, is pushing their work from sound science, into shaky science, and eventually outright fraud. It's a good sketch of the culture of Big Science, particularly on the government side. I've never worked in Big Science per se, but even on the small scale on which my old group at NIST worked, budget pressures were fairly significant. On the billion-dollar scale where nuclear weapons work is conducted, it's easy to see how the need for funding could warp the whole scientific process, and Scholz's picture of it rings true.