Officials in charge of the space science programme at NASA, the US space agency, might be forgiven for feeling put-upon this budget season. Agency head Dan Goldin has for several years been a good soldier in the rhetorical war against government spending. While other science agencies saw their budgets rise, Goldin watched his remain flat, and professed pride that NASA could do more with less. And by some measures — number of missions launched, for example — the space scientists in his organization have succeeded.

Now comes their reward — a $184 million cut to space science proposed in the Senate, a $265 million cut in the House of Representatives. The two chambers must reconcile their different spending bills, but the best NASA can hope for is to escape with last year's budget.

Why such harsh treatment? Congressional appropriators certainly didn't go hunting for space scientists. There is no anti-NASA lobby on Capitol Hill. But when normally friendly appropriators found themselves without enough money to satisfy noisier constituencies, space science proved a convenient account to raid.

The agency's small, cheap science missions were the hardest hit, and this may be the most sobering message to emerge from this year's budget travail. Goldin's “better, cheaper, faster” philosophy — lots of small missions rather than a few big ones — is now firmly entrenched as NASA policy, a shift that Congress and the White House have strongly encouraged. Ironically, the strategy arose precisely because large, expensive projects were seen as more vulnerable politically. But once started, they are also hard to kill. A threat in the House to cancel the $900 million Space Infrared Telescope Facility lasted only a few days this summer before congressional champions saw to it that the money was restored. Projects costing only $150 million are easier to sacrifice because they have small budgets and employ small teams of scientists and engineers.

The “better, cheaper, faster” strategy is still an experiment. Its scientific and organizational merits have yet to be proven, and NASA managers have openly acknowledged that it adds risk to the already tricky business of space flight. But those controlling the space agency's budget need to let the experiment run its course. If small missions are always at risk of being scrapped or delayed, and if funding for new technologies that might reduce the cost of space missions is perpetually starved, Goldin's strategy will not have been fairly tested.