Why do they do it? What is it that turns young men, some with good life prospects, into suicide bombers?
Scott Atran, a US academic who has conducted scores of interviews with families, friends and neighbours of suicide bombers, points to one common factor: publicity. “The difference between terror and other forms of violence . . . is publicity,” he says.
Publicity helps to provoke governments into overreaction and turns terrorists into media stars, and heroes in their own milieux, he says. Mr Atran’s conversations with children in the poor neighbourhood of Mezuak in the Moroccan city of Tetuan, show they dream of becoming either Ronaldinho, the Brazilian footballer, or Osama bin Laden.
The Moroccan authorities believe at least 30 suicide bombers in Iraq have come from Mezuak, home to just 19,000 people, 11 of whom have been confirmed through DNA sent to the authorities by the US. Five of the seven 2004 Madrid train bombers who blew themselves up when cornered by Spanish police also came from Mezuak.
The Madrid gang and other suicide bombers had another attribute so obvious it is almost always overlooked: they belonged to groups. According to some academics, terrorist cells are best understood by thinking of “bands of brothers” or “groups of guys” and examining the interaction among them.
According to the synopsis of a report by Mr Atran and two colleagues delivered to staff in the White House National Security Council in March: “Small group dynamics – rather than personality, ideology, education or income – is the prime factor in deciding which few, among millions of potential jihadis will actually go on to commit violence.”
Mr Atran and Marc Sageman, a former Central Intelligence Agency officer, are building a database of terror groups based on court records and interviews. Their preliminary conclusions sometimes run counter to conventional wisdom. Terrorists, for example, are usually turned within groups and there is not much evidence of top-down recruitment or brainwashing of plotters.
About 70 per cent of terrorists enlist in groups through friendship and about 20 per cent through kinship. The preferred cell size is eight members and consists of friends made between the ages of 15 and 30.
Neither is social deprivation a factor. A 2004 survey by Mr Sageman showed more than 70 per cent of jihadis were from middle or upper class backgrounds. More than 40 per cent were, like the group allegedly behind last week’s attacks in the UK, in the professions: teachers, lawyers and doctors.
Often within groups, there is a leader or hands-on figure, often not the most devout, who converts radical words to actions. In the Madrid group, this appeared to have been Jamal Ahmidan, who married a Christian and was a notorious drug smuggler.
Once in the group, what leads them to resort to terrorism and suicide? Christopher Heffelfinger, a senior analyst at the West Point military academy, has identified four steps on the road: introduction to the group; immersion in extremist doctrine; an initial effort to effect peaceful change; and, lastly, the step from nonviolence to violence.
Mr Sageman has described three generations of jihadis: the first the foreign mujahideen in Afghanistan and the second a younger generation of educated youths, such as those that hatched the September 11 plot in Hamburg. The third wave consists mostly of semi-skilled or marginalised people, such as those behind the Madrid bombings or the London attacks of July 7 2005.
On this nomenclature, the “medical cell” allegedly behind the attacks in London and Glasgow look like a late example of the second wave. Early reports suggest one of the suspects was radicalised before he came to Britain in 2004. It is thought he might have met four of the other suspects while living in Cambridge. Some were already, apparently, hardliners: two of the suspects were brothers, and one their cousin.
A former friend of one of the members of the group in Cambridge described it as being “small and selective”, containing only like-minded people associated with Hizb-ut-Tahrir, the radical group seeking to establish a Muslim caliphate. Hizb-ut-Tahrir denies it supports terrorism.
That picture of isolation from society has been noticed in other groups. Mr Atran describes the Hamburg plotters as “a group of intense and intimate friends who created a parallel universe of jihad by isolating their behaviour from the surrounding society”.
The 7/7 bombers had a ringleader – Mohammad Sidique Khan, a youth worker – but the other three all seem to have become extremists before falling under his influence. Khan and Shehzad Tanweer are thought to have received training in bomb-making in Pakistan – but Pakistan had nothing to do with their radicalisation.
Three of the four came from the same district of south Leeds in West Yorkshire in the north of England. An investigation of the group by Shiv Malik, described in an article in Prospect magazine last month, suggests the youths suffered a double identity problem. They were alienated from society at large but – much less understood – estranged too from their parents’ generation, whose attitudes derived largely from the tribal customs of a district in Kashmir from which they hailed. This clash was epitomised in their different views on marriage: the parents insisting on arranged marriages and their sons believing they should be able to marry any good Muslim girl.
But becoming an Islamist is not the same as turning to terrorism. Apart from any other factor, one determinant is opportunity.
Mr Heffelfinger, who has carried out a study of six suspects arrested in May for allegedly plotting to attack the Fort Dix army base in New Jersey, says they appear to have chosen their target capriciously: because one of their number had delivered pizzas there and was familiar with the base. “It was opportunity over any other motivation,” he says.
All these studies suggest that resorting to terrorism has multiple causes, responding to specific personal and cultural dislocations experienced by young men for which a violent, global movement seems to offer an answer.
Unfortunately, they do not suggest a typology in which those at risk of turning to terrorism can be picked out. “There is no predictive model,” says one British counter-terrorism official.