The revelations of the extent of National Security Agency (NSA) surveillance on US telecommunications this week bring up a chicken and egg problem: Are Americans so safe from terrorist attacks because the surveillance program works, or are they sacrificing privacy so that the government can protect them from a statistically marginal and rare occurrence?

The NSA is very good at electronic surveillance, so good that smart terrorists avoid using phones and the internet to conduct their business. That’s one reason that Osama bin Laden stayed alive for so long and why his former number two and current Al Qaeda boss, Ayman al-Zawahiri, remains alive today.

The fact is, domestic terrorism is an exceedingly rare occurrence and miniscule cause of death in the US, notwithstanding the outsized response attacks draw from citizens and politicians. It’s very difficult to count alleged terrorist plots that may have been thwarted by the snooping effort, but nevertheless, terrorism has been rare in America throughout the sweep of US history.

President Barack Obama defended the program today. “My assessment and my team’s assessment was that they help us prevent terrorist attacks,” he told reporters. “And the modest encroachments on privacy that are involved in getting phone numbers or duration [of calls] without a name attached and not looking at content, that on net it was worth us doing.”

Obama also said “you can’t have 100 percent security and also then have 100 percent privacy and zero inconvenience” and “we’re going to have to make some choices as a society.” Both those points are so true as to almost be tautological--though it’s simply not possible to have 100 percent security under any circumstances.

America and its politicians have consistently chosen “more security” and “less privacy” in recent years, but the fear of terrorism, and what’s being given up in the service of that fear, seem all out of proportion to the actual threat.

In 2011 there were zero domestic US victims of “terrorism.” In 2012 there was a single incident of domestic terrorism, when a white supremacist shot and killed six people at a Sikh temple in Wisconsin. This year, there has been a single incident so far, when two Muslim brothers, apparently motivated by hatred of the US and inspired by Al Qaeda-type thinking, murdered 3 people and injured 260 others at the finish line of the Boston Marathon.

The government’s spying powers aren’t supposed to be turned on US citizens, so the program was irrelevant to the 2011 attack. President Barack Obama says that US residents are also exempt, so presumably the program wasn’t relevant to the marathon attack either.

As Micah Zenko wrote last year, a comparable number of Americans were killed by falling televisions as were killed by acts of terrorism in 2011. How bad has terrorism been in recent decades?

In the 1990s, there was an Al Qaeda linked attack on the World Trade Center in 1993 that killed six and the 1995 attack on the Alfred P. Murrah federal building in Oklahoma City, carried out by two militia movement sympathizers, that killed 168. There were a few other minor incidents in the 1990s that brought the decade’s toll of deaths from terrorism on US soil to about 180.

In the 2000s there are only really two terrorist incidents to speak of. The Al Qaeda attack on the World Trade Center and the Pentagon on Sept. 11, 2001 that killed 2,977 people; and Major Nidal Malik Hasan’s rampage at Fort Hood Texas in 2009, that claimed 13 lives, for a total of 2,990 for the decade.

Of course, this is horrific enough. And violence designed to terrorize in search of a political goal is, in fact, something different than crime, or accidents. But a sense of proportion is sorely needed in considering the danger. Stephen Walt wrote a brief post yesterday taking writer Andrew Sullivan to task for saying that the “consequences of its absence” could be “terrible,” which I think is spot on:

This claim depends on the belief that jihadism really does pose some sort of horrific threat to American society. This belief is unwarranted, however, provided that dedicated and suicidal jihadists never gain access to nuclear weapons. Conventional terrorism--even of the sort suffered on 9/11--is not a serious threat to the U.S. economy, the American way of life, or even the personal security of the overwhelming majority of Americans, because al Qaeda and its cousins are neither powerful nor skillful enough to do as much damage as they might like. And this would be the case even if the NSA weren’t secretly collecting a lot of data about domestic phone traffic. Indeed, as political scientist John Mueller and civil engineer Mark Stewart have shown, post-9/11 terrorist plots have been mostly lame and inept, and Americans are at far greater risk from car accidents, bathtub mishaps, and a host of other undramatic dangers than they are from “jihadi terrorism.”

Over 100,000 people died in drunk driving accidents in the US in the 2000s (the decade with the lowest number of drunk driving fatalities in US history; in the 1990s the toll was over 136,000). About 26,000 Americans die from unintentional falls a year, and about 33,000 from accidental poisoning.

Americans don’t live in fear of those dangers, nor should they. Excessive, exaggerated fear of terrorism is something that, perhaps, the US will get over as it considers the prices being paid.

Or as James Madison wrote to Thomas Jefferson in 1798:

The management of foreign relations appears to be the most susceptible of abuse, of all the trusts committed to a Government because they can be concealed or disclosed, or disclosed in such parts and at such times as will best suit particular views; and because the body of the people are less capable of judging and are more under the influence of prejudices, on that branch of their affairs, than of any other. Perhaps it is a universal truth that the loss of liberty at home is to be charged to provisions against danger, real or pretended, from abroad.