The problem with doing research on informers is that much of the information is kept secret

Stories about the controversial use of informers have hit the headlines recently, with the planting of police officers in environmental groups causing serious embarrassment when it was found they'd had relationships and fathered children with activists. Reports of MI5 repeatedly approaching people assumed to be Muslims at UK airports with inducements to inform on their friends and neighbours have also been disastrous for community relations.

The dislike ? hatred even ? people feel for telltales reflects the trauma that is felt when trust is broken, says Dr Steve Hewitt, senior lecturer in the department of American and Canadian studies at the University of Birmingham, and author of Snitch! A History of the Modern Intelligence Informer. When it's a matter of deeply felt belief, the trauma is magnified. "It involves betrayal, and it's a very powerful emotion, whereas technology used for surveillance is impersonal," he says.

The anger and alienation felt by Muslims in the UK and abroad since 9/11 as a result of individuals being bribed ? or threatened ? to inform on possible terrorist activity has been an unfortunate own-goal by the intelligence services, suggests Hewitt. Recruiting reliable informers is clearly harder when you have no "in" to a community, he points out. And the dearth of intelligence agents from the minority ethnic groups now of interest means the authorities have been trying to win people over without sufficient knowledge, insight or sophistication.

"The sign of it not being done very well at times is that people are going to the media about MI5 approaching them in airports," he says. "That's incredibly clumsy. If recruitment is done well, you know what the answer is going to be before you ask the question."

For authorities desperate to prevent future terrorist attacks, nothing can replace inside information from people involved in the groups they suspect of planning them.

There is a deeply unpleasant side to recruiting informants, if allegations by some of those who have come forward to tell of their experiences are to be believed. At the milder end of the spectrum, immigrants are bribed with offers of citizenship in return for co-operation. Somewhat nastier is the threat of deportation should the proposal to inform be turned down. At worst, those who decline to help the authorities, Hewitt explains in his book, have allegedly been turned over to the US, with their next stop being Guant?namo Bay, "where they faced sustained pressure to offer their assistance [and] MI5 made frequent trips to try to convince them to begin working as informers in exchange for their release".

There is some collaborative counter-terrorism being done now by agencies trying to reach out to communities where it's thought plots may arise, Hewitt explains, but the recruitment and use of informers undermines state efforts at building trust among those very same communities.

The problem with doing research in this field, he says, is that the world of handler and informer are secret, while operations are active and the records of what went on and the information received are kept under wraps for decades. In the UK, they would never be disclosed while an informer is alive. In the US, where informers are used as witnesses in criminal prosecutions, the hidden role of the snitch comes to the fore, though a new identity will be provided after the trial and the informer usually disappears from view.

Sometimes an intelligence operation goes badly wrong. The recent press coverage of UK police who infiltrated environmental groups, began relationships with female members, married and had children with them not only spectacularly blew the cover of a number of undercover officers but, says Hewitt, highlighted the complex ethical issues facing agencies wanting to use them.

"There was a discussion I heard of the Mark Kennedy case," he recalls, "and an activist made the point that if the police want to put a microphone in my house, they need the home secretary's authorisation, but if they want to put an officer in to marry me, even have children with me, it can be agreed by someone far less senior."

Another ethically disturbing problem authorities must face up to is that informing is not a neutral activity. "It's not just collecting information, but it can also involve shaping the actions of the group, and that's why it's so controversial," says Hewitt. "With terrorist informers ? a question that can be asked is: are they pushing the activity along, and would that have happened in that group if the informer hadn't been there?"

There have been reservations expressed about the state actively working with, and paying sometimes large sums of money to, people who by their very nature may follow a very unsavoury lifestyle. On the criminal side, some informers will be gangsters, even killers ? and may well continue in their activities throughout their time as informer.

There has always been a "huge issue" of squeamishness, says Hewitt, about who is recruited and what they are permitted to do with the knowledge of the state. But, he explains, "if you're going to have people acting as informants, they're not always going to be nice people".

Being outed as a snitch can be lethal, as IRA informers discovered. But no matter what the consequences for any individual of their decision to inform, or not to inform, the a state's decision to infiltrate communities can be counter-productive, giving rise to more anger and mistrust than existed before.

"In liberal democratic societies it's marginalised groups who disproportionately experience the use of informers and have less access to power to protest it ?," says Hewitt. "Governments should recognise that this is an intrusive form of surveillance and be more sensitive about how the issue is dealt with."