A new study from the RAND Corporation looks at the threat of homegrown terrorism and concludes that our so-called “lone wolves” look a lot more like “stray dogs”—and stray dogs with more bark than bite, at that:

The 82 cases [i.e., investigations culminating in prosecution for some form of support for jihadist terrorism] since 9/11 involved 32 plots. Few of these 32 got much beyond the discussion stage. Only 10 developed anything resembling an operational plan that identified a specific target, developed the means of attack, and offered a sequence of steps to carry out the planned action. Of these, six were Federal Bureau of Investigation (FBI) stings. Only two individuals actually attempted to build devices on their own. One was arrested while doing so, and the other’s device failed. The rest of the would-be terrorists only talked about bombs. In only two cases did jihadist terrorists actually succeed in killing someone, and both of these cases, which occurred in 2009, involved lone gunmen. [Emphasis added.]

That assessment dovetails with the portrait painted by an important package of feature stories in the latest issue of Mother Jones examining the FBI’s pursuit of the War on Terror, and in particular the way the Bureau has established a vast network of informants. In many recent high profile cases, the FBI or its “assets” appear to have gone beyond trying to detect terror plots to playing a substantial role in manufacturing them. In line with the findings of the the RAND report, Mother Jones’ survey of domestic War on Terror success stories shows that many of the highest-profile ones, including all but one of the supposed subway bomb plots “foiled” by the FBI, had first been orchestrated by FBI assets. While the targets of those operations were clearly boiling over with anger at the United States, it’s not clear how many of them would have translated their rage into violent action absent the government’s prodding.

The author of the Mother Jones article compares the contemporary hunt for “lone wolves” and domestic terror cells to another notorious FBI initiative: COINTELPRO, a series of covert projects, stretching over three decades during the Cold War, that targeted domestic “subversive” groups for infiltration. But the aggressive use of informants and infiltrators is not the only interesting parallel here. COINTELPRO projects like Operation Chaos targeted activist groups, especially anti-war groups, suspected of being controlled by foreign governments, consistently failing to turn up proof of foreign control. But Lyndon Johnson was convinced that the link had to be there—and the failure to uncover it only underscored how insidious and dangerous the adversary must be. Thus, over time, the bar for what counted as foreign “ties” was lowered, the program’s scope expanded to include civil rights and women’s liberation groups, and its methods grew more aggressive. Because the foreign communist control had to be there, failure to detect it was regarded as failure, period.

The attempt to detect “lone wolf” terrorists presents a similar—and perhaps in some ways a still more daunting—problem. The last decade has seen a drastic expansion of government surveillance powers in the name of “connecting the dots” to identify the affiliates of foreign terror groups. This has inevitably involved the monitoring of many innocent people, but that project is, in one way, finite: Investigators begin with a set of known starting points—people, phone numbers, Internet addresses, bank accounts, etc., that are believed to be tied to a group like Al Qaeda, and then begin tracing the links in search of unknown allies. As investigators learn more about the target group through capture, infiltration, or surveillance, they can begin to develop estimates of how many operatives might reasonably remain at large—and in particular, how many of those might have made it into the United States.

But a true lone wolf won’t be located that way since, by definition, he’ll lack the necessary links to known terror groups. And there’s no reliable way to know in advance how many solitary angry individuals might be plotting violence. The effort to preempt lone wolves, then, leads to much broader attempts to detect “suspicious behavior,” leading to the compilation of dossiers on many innocent people. But it’s hard to justify the hundreds of billions, if not trillions, of additional spending on a vast national security state since 9/11 without a few scalps. Pressure naturally mounts on informants to come up with the lone wolves the government just knows must be out there—even if some of them need a little push, a little help getting a plan together or a working bomb produced. Call it supply and demand: When billions are available to fight a terrorist enemy, you can be confident we’ll come up with enough scary terrorist enemies to keep the money flowing. Even if we have to make them.