Recent efforts by US nuclear weapons laboratories to engage in deeper and broader collaboration with university researchers have encountered a number of practical obstacles, including the role of foreign faculty and staff in such partnerships (see Nature 391, 311; 1998). So far, however, they have not faced any serious resistance on university campuses to acceptance of the money.

Defenders of these partnerships have argued, with some success, that the research they will carry out will not support the development of nuclear weapons as such, but merely assist the weapons laboratories in their new mission of assuring the “safety and reliability” of existing weapons through the $4.5-billion-a-year science-based stockpile stewardship programme. The problem is that this programme is not needed to ensure either the safety or the reliability of the existing stockpile. That is already quite safe and reliable, and could be kept so indefinitely by means of a small remanufacturing programme (see Ray E. Kidder, Nature 386, 645; 1997).

The real aims of science-based stockpile stewardship appear to be twofold. The first, which cannot be publicly acknowledged, is to maintain the three nuclear weapons laboratories at their Cold War level, so that Washington does not have to make politically painful choices. The second — also seldom discussed, but at least acknowledged in Department of Energy documents — is the long-term maintenance of a US nuclear weapons design capability.

These aims overlap, but are not identical. The laboratories at Sandia and Los Alamos are by far the main source of federal largesse in the state of New Mexico, whose senior senator, Pete Domenici, happens to chair the Senate's Budget Committee. The Lawrence Livermore laboratory is seen in northern California as a vital technological resource. It became clear early in President Bill Clinton's second term that the laboratory would not close or change its mission, whatever advice the administration received to the contrary.

The need for a future weapons capability compounds these blunt political considerations. The Clinton administration has a policy of not developing new weapons at present, but the stockpile stewardship plan states clearly that the capability to do so must be maintained. Major facilities, such as the National Ignition Facility at Livermore, are intended to help realize that goal. So are the partnerships with the universities.

One of these, at the University of Utah, has kept critics at bay by arguing that it will model only issues pertinent to nuclear weapons safety. What, then, of the other partnerships that will model explosive shock, or turbulence — both problems more directly tied to bomb simulation? Scientists at all five university partnerships point out that they are doing unclassified work only in basic physics. But the objective of that work, from the funder's viewpoint, is to lay the groundwork for as complete a simulation of nuclear weapons as is possible in the absence of testing.

The university scientists also claim that, by supporting the stockpile stewardship programme, they are enabling the United States to comply with the Comprehensive Test Ban Treaty (CTBT), which most support. But the programme helps the CTBT only by injecting so much money into the weapons laboratories that their directors will desist from undermining the treaty in the Congress. And the argument that the programme reflects the will of the president and of the Congress, a political tapestry that patches together various special interests, is not a reasonable basis for a university to decide what it will and will not expect its staff and students to do.

Rather than telling the public the truth about the stockpile stewardship programme, however, the scientific community has put aside obvious questions and quietly accepted the money. In doing so, it has implicitly endorsed a military posture that remains heavily reliant on weapons of mass destruction. General Lee Butler, former chief of the Strategic Bomber Command, recently called for this posture to be changed. Scientists with a conscience should follow his example and suggest how nuclear weapons research can be curtailed rather than expanded.