When I talk about my work with ordinary people (i.e., those who don't eat, sleep and breathe terrorism), the question I get asked most often is "Should I be worried?" 

In the wake of the death of Osama bin Laden, and as my book on American jihadists nears release, I've been thinking a lot about how to put the threat of homegrown terrorism into the proper context. 

We've been hearing a low rumble of policy makers implying or coming right out and say that homegrown terrorism is the most serious threat facing the U.S. homeland today. I think it's difficult to make that kind of simple statement. The nature of the threat is different than the nature of the September 11 attack or the threat of nuclear terrorism. I've been literally sketching out some ideas about how to prioritize homegrown and lone-wolf terrorism, and the charts below are a first crude effort to illustrate the problem. 

Click to enlarge

The first slide, above, is a general statement of principle which boils down to this: Number of casualties X frequency of attack = priority. So one September 11 over 10 years with 3,000 casualties would take roughly the same priority as 10 bombings killing 300 people over the same period. It's a big more complicated than that, of course. The 9/11 attack was exponentially more disruptive than a series of smaller attacks, but on the other hand, a series of bombings on U.S. soil, each killing hundreds, would create a much greater sense of insecurity. If there was a scenario for terrorist attacks every couple of days that kill just a couple of people at a time, that would create similar issues. So we're only talking broadest principles here; you can't necessarily make this a hard numerical formula.

Click to enlarge

The second slide shows where, in general principle, to place different types of attacks on the threat matrix. The threat from American jihadists today falls mainly into two categories -- weak network ties and lone wolf or so-called "leaderless jihad" attacks. Before 9/11, major terrorist attacks in the U.S. often fell under "self-organized," such as the World Trade Center bombing and the thwarted Landmarks plot that aimed to bring down the Lincoln and Holland tunnels during rush hour. These self-organized cells took longer to act, but worked on plots of much greater ambition than we've seen since 9/11, so they're in the same general neighborhood of priority on the grid. 

Nuclear terrorism is literally off the chart here. Should such an attack occur, it's likely to be a unique, one-time event with causalties far in excess of anything we've seen before. It's hard to imagine a higher priority threat, but I'm keeping it out of this discussion which is trying to evaluate recurring issues.

Click to enlarge

The third slide shows how different incidents before and after 9/11 fall on the matrix. Here, I've replaced "frequency" with "length of planning cycle" to reflect the fact we're talking about historical events rather than projections. The length of planning cycle has obvious relevance to frequency but is not exactly the same thing, since a number of attacks might be planned simultaneouslty. But it's close enough to use as an approximation here. 

Although this list is not inclusive, I think it's useful to note the concentration of threats in the middle of the grid. Had the landmarks plot been executed, it would quite possibly have killed as many people as 9/11. If I had listed all the lone-wolf, homegrown incidents, there would be a corrsponding concentration in the lower right quadrant. However, a lot of the known homegrown cases are pretty nebulous wth many homegrowns being arrested before developing a credible plan for an attack. 

The Baltimore shooting and the Portland Christmas bombing plots are listed here as "self-organized" and "weak network" plots respectively because in Baltimore, Antonio Martinez was trying to recruit additional participants and in Portland, Mohamed Mohamud believed he was receiving assistance from Al Qaeda. In the latter case, we don't know if he would have acted without that belief, so it's not appropriate to categorize him as a lone wolf. 

Click to enlarge

Threat potential is one thing, but threat credibility is another. Slide four shows the same incidents listed on slide three, but with an additional category for failed attacks. You'll notice that cleans up the matrix significantly. Virtually all of the intercepted threats listed here were stopped by law enforcement or intelligence actions with the notable exception of the Times Square incident, which failed due to terrorist incompetence. 

You estimate the damage done by terrorists according to successful attacks, but you evaluate the threat based on potential damage -- for which I am using casualties as shorthand here. (Obviously there are other ways to do damage, but they go beyond the simple formulation I am making.) The thing to remember is that just becaue these and other threats were neutralized doesn't mean they don't count. If we hadn't intercepted them, most of them would likely have resulted in dead Americans. 

My quick read on this and drawing on the research from my book (Jihad Joe: Americans Who Go to War in the Name of Islam, coming out this week) is that we're basically doing the right things to prevent homegrown, lone-wolf and self-organized terrorists from carrying out successful attacks, but as the Times Square incident shows, we're not doing a perfect job. The next time we miss one, it might be someone who knows what he's doing. But what does all this perhaps long-winded thinking aloud say about our national security priorities?

1) Homegrown terrorists have failed largely because of good policing.

So "should I be worried?" The answer is that American jihadists have by and large been unsuccessful in carrying out terrorist attacks on U.S. soil since 9/11. That is not a reason to reduce policing -- it's because of good policing, primarily by the FBI, that the majority of these efforts have failed or been intercepted before they could even be attempted. Among the ones we missed -- Nidal Hassan and Faisal Shahzad -- we see one success and one failure due to incompetence. I say more about some of the implications of those two cases in the book. 

2) One element of good policing is preventing American jihadists from organizing

Since September 11, if you get six guys in a room talking about jihad, odds are one of them is an FBI informant. The FBI's success at winning cooperation from American Muslims as well as placing undercover agents next to key figures has significantly diminished the ability of radicalized Americans to self-organize into viable terrorist cells. The two most significant jihadist attacks on U.S. soil prior to September 11 were the World Trade Center bombing and the Landmarks plot. Both of these were primarily carried out by self-organized cells, with support from weak networks -- "weak" here meaning the attacks got expertise and partial funding from more organized terrorist networks like Gama'a Islamiyya, Egyptian Jihad or Al Qaeda, but they were not network-owned operations from cradle to grave. 

While allowing for individual brilliance and group incompetence, cells are more dangerous and capable than individuals as a general rule. The FBI's success in inhibiting the formation of cells has thus far been remarkable, and it's a key factor in neutralizing the homegrown threat.

3) Homegrown terrorists have not yet become a top tier threat.

Leaderless jihadists and lone wolves have two basic problems keeping them at a lower level. 

First, they're not very successful. Nidal Hassan's Fort Hood shooting spree was far and away the highest casualty attack by a leaderless terrorist. He killed 13 people. The total death toll from homegrown jihadists post-9/11 is just 17, and that is using a pretty generous definition of jihadist. 

Second, homegrown terrorists don't attempt to strike very often. There are two ways to get promoted on the threat matrix. You can kill more people, and you can attack more frequently. There are a number of reasons why radicalized Americans don't attack more regularly. I would suggest the most obvious reason is also the most likely -- they're lot more talk and little less action. The aforementioned good police work is part of the reason for this. American radicals are justifiably worried about infiltration, which inhibits their efforts to organize, and the lack of success stories for individual action denies them encouragement. 

This is a moving target, and it could certainly changed based on outside pressures such as world events or jihadist touchstones like the death of Osama bin Laden. So far, at least, there are very few jihad-inclined American willing to throw their lives away on small-scale attacks. 

As I wrote last week, the death of bin Laden will provide a real test for the leaderless jihad. It strikes me as unwise to make extravagant predictions about the future viability of the American jihadist movement in the immediate aftermath of that event (and especially before we've heard from Al Qaeda in the Arabian Peninsula). The picture so far is encouraging, but I think that within the next few weeks and months we're going to get a pretty clear picture of the limits of leaderless and lone wolf jihad. 

Finally, this will probably piss some people off, but I think it's important to keep some perpsective on how terrorism rates in a broader perspective. There are a whole lot of reasons why some of the additional entries on this chart are not necessarily comparable to terrorism. I'm not seeking to make an easy or direct equivalency. All I'm saying is that when we think about our national priorities, it might be useful to keep in mind that on a practical basis, there are a lot of other problems which lead to Americans dying, and maybe our funding and focus priorities could be usefully adjusted to reflect some of these other issues.

Click to enlarge