Faced with continuing attacks and threats from radical environmental and animal-rights groups, scientists increasingly find themselves lumped in with an odd assortment of societal targets — from fast-food restaurants to mink ranchers to second-hand car dealers — branded as agents of global capitalism and eco-destruction. No amount of academy studies proclaiming the safety of genetically modified (GM) crops, or protocols to guarantee the humane treatment of lab animals, or public-relations campaigns explaining the benefits of biotechnology to a hungry world, are likely to persuade the vandals and arsonists otherwise. They have reduced the world's complexity to a few simple manifestos, and have traded debate for intimidation.

So scientists should focus instead on winning over the vastly larger number of people who may be sceptical or uneasy about GM foods or the need for animal research, but who still have open minds. And the best way to do that is to stick with two of science's traditional strengths — the open exchange of information and a faith in rationality.

Two stories in this week's News section illustrate how difficult that high-minded road can be. A local planning board has rejected a proposal by the University of Cambridge to build a new primate research facility for fear of protests from animal-rights groups (page 725). And scientific societies are pressing the US agriculture department to remove information from the Internet about scientists who use laboratory animals — information that had been posted in the interests of transparency and public accountability (page 723).

The first action is the more lamentable, in that it represents a citizens' council simply caving in to bullying tactics. The second action, proposed by the research community itself, is defensible in spirit, but sets a worrying precedent. No one can blame scientists for wanting to protect themselves. The tactics of groups such as the Animal Liberation Front (ALF) and the Earth Liberation Front (ELF) have moved well beyond spray-painting slogans and trampling crops, to arson. If that trend continues, it's only a question of time before someone gets hurt. So the request from the Federation of American Societies for Experimental Biology (FASEB) to remove detailed identifying information about animal research facilities from the web seems justified. And there should be a way to accomplish that while still satisfying the public's right to know. The US agriculture department, for example, posts information about GM crop trials on the web, as does the Department for Environment, Food and Rural Affairs in the United Kingdom. In neither case is the precise location of research plots given, presumably to make things difficult for would-be saboteurs.

Some 'secrecy', if it can be called that, therefore seems warranted. At the same time, scientists must scrupulously avoid giving even the slightest impression of taking their research underground. That would only further inflame the zealots, and make it harder to win over the uncommitted. Besides, it's not clear that arsonists rely that much on the kind of information posted on the agriculture department website. Groups such as the ELF or ALF don't seem especially diligent about doing careful research before they choose their targets. More likely, a short news story about a genetics lab being built at a nearby university, or the simple knowledge that a scientist uses animals — any animals — in their work, will suffice.

Take the case of Toby Bradshaw, the University of Washington plant researcher whose office was burned down last May by the ELF. The arsonists obviously misunderstood his work. Bradshaw has studied traditional hybridization of poplar trees, but has never done any genetic 'engineering'. And when the eco-vandals fire-bombed his office, they also destroyed 100 showy stickweed plants — a quarter of the remaining samples of one of the region's most endangered species. Which goes to show that if too much information is sometimes dangerous, too little can be even worse.