When I read the Intelligence and Security Committee's report on the July 7 terrorist attacks that was published this week, two points struck me.
The first is that the Security Service did not give due attention to key indicators which would now flash red lights. For example, that two of the would-be attackers had travelled to Pakistan and Afghanistan. The second is the Security Service's assertion that they did not actually have sufficient resources to devote attention to these individuals and had to prioritise other investigations that were considered more pressing.
The Committee said that for the Service to achieve coverage of all targets, it would need "to be a very different organisation, both in terms of its size and how it operates, which would have huge ramifications for our society and the way we live". This is undoubtedly true.
But recruiting or using informants is unlikely to produce a sufficiently rich picture or provide full coverage. When recruiting informants and making use of the information they provide, we must bear in mind three potential problems.
First, there is always a risk that human sources might be unreliable. Secondly, there is always a risk that they might have ulterior motives. Thirdly, it is unlikely that they will reflect all the complexities and diversities of the communities from which they come.
At worst, if known, people might resent the fact that the Security Service is recruiting informants specifically from their community. It could create a perception that they are all being stigmatised and spied on, and this could in turn create a culture of mistrust.
This is not to say that traditional intelligence methods have no place. But I would suggest that the aim should be to reach a situation in which people come of their own accord to the authorities – to the police, local councils, schools – with any concerns about those in their neighbourhoods. In this situation there would be less need for informants. Why is this point important? Changes in behaviour and in attitudes can be subtle and gradual, and communities and families are best placed to notice early on any behaviour that is out of the ordinary.
They can, for instance, help the authorities understand the significance of a recent visit to Pakistan. Information forthcoming in this way would fill intelligence gaps and provide context.
It would help with the early identification and intervention of individuals at risk. It would also help the security services better prioritise resources and investigations.
To get to this point, great attention will need to be paid over the coming years to integrate counter-terrorism work into an effective system of community policing. It also follows from all this that counter-terrorism programmes and tactics must be assessed against their likely effect on community relations.
Today's Independent contains allegations from young Muslim men that they have been harassed by the Security Service to become informants. I have no knowledge of the validity of these complaints but it is clear that promising initiatives can be easily wrecked in their implementation, resulting in the reverse effect from the one desired and intended – a reduction in trust and communication between communities and police.
In what is without question a delicate and difficult challenge for our security services, great skill will be need to be shown over the coming years to effectively integrate counter-terrorism work into successful community policing.
The author is the shadow Security minister and former chairman of the Joint Intelligence Committee