In President Trump’s speech to Congress Tuesday night, he claimed that, “[a]ccording to data provided by the Department of Justice, the vast majority of individuals convicted of terrorism and terrorism-related offenses since 9/11 came here from outside of our country.” The following day, the Justice Department elaborated to the Washington Post that it had obtained convictions “against over 500 defendants for terrorism or terrorism-related charges in federal courts” and a “review of that information revealed that a substantial majority of those convicted were born in foreign countries.”
The Justice Department did not provide the raw data to support its assertion, but if that data derives from the source it appears to come from – a Justice Department National Security Division (NSD) list of terrorism and terrorism-related convictions – the claim is highly misleading. That’s because the NSD list is designed in ways that seriously undercut its ability to shed light on the scope or source of the terrorist threat.
I recently obtained the NSD’s Chart of Public/Unsealed International Terrorism and Terrorism-Related Convictions from 9/11/01 to 12/31/15, in response to a Freedom of Information Act request I submitted in 2015. To my knowledge, this version of the chart has not been made public. The chart identifies 627 “terrorism and terrorism-related” convictions through the end of 2015. This list does not note the national origin or citizenship status of terrorism defendants, but it is the updated version of a list that the Justice Department provided in January 2015 to the Senate Subcommittee on Immigration and the National Interest. That subcommittee, then chaired by then-Senator Jeff Sessions, conducted its own “open-source research” to determine the national origin of those included on the list, and concluded in 2016 that “at least 380” people on the list were born abroad.
If the NSD data is now being used by the President and the Justice Department to claim that a “vast majority” or “substantial majority” of convicted terrorism defendants are foreign-born, there are several problems with this claim.
First, the NSD list only includes convictions resulting from “international terrorism investigations.” That includes investigations of terrorist acts planned or committed outside the United States and those within the United States “involving international terrorists and terrorist groups.” But the list, by its own terms, “does not include convictions related solely to domestic terrorism.” If you exclude all convictions for “domestic terrorism” at the outset, how can you draw any overall conclusions on the citizenship status or national origin of those convicted of terrorism?
The NSD list doesn’t say much regarding how it distinguishes between “international terrorism” and “domestic terrorism.” But the U.S. criminal code draws the distinction this way: international terrorism involves acts that “occur primarily outside the territorial jurisdiction of the United States, or transcend national boundaries in terms of the means by which they are accomplished, the persons they appear intended to intimidate or coerce, or the locale in which their perpetrators operate or seek asylum.” Domestic terrorism involves acts that “occur primarily within the territorial jurisdiction of the United States.”
If the NSD list excluded all convictions for acts occurring primarily within the United States, that would seem an especially unsuitable source for making claims about who presents the greatest threat to Americans. In fact, I think the distinction is probably blurrier in practice, and perhaps even more problematic. Human Rights Watch and Columbia Law School’s Human Rights Institute, in a 2014 report on human rights abuses in terrorism cases, criticized an earlier version of the NSD list for including individuals who “appear to have no international link.” It might be that, in practice, prosecutors classify terrorism cases as “international” whenever defendants appear to be motivated by Islamist causes, on the theory that even acts within the United States by Americans inspired by transnational Islamic extremism “transcend national boundaries.” One can imagine an opposite presumption applying to white supremacists or other far-right extremists, whose acts of violence might typically be excluded as “domestic terrorism” (even if there are foreign influences). One wonders whether racialized assumptions, rather than objective differences, drive some of these classification decisions.
To be clear, I don’t know how the National Security Division actually draws the line between domestic and international terrorism. The Justice Department did not respond to my request for an explanation of how it distinguishes between the categories, despite furnishing the NSD list in response to the same FOIA submission. At this point, I can only speculate whether the Justice Department doesn’t have any such explanation in writing, or whether it simply ignored that portion of my FOIA request. (It also ignored my request for any statistical information the department keeps on “domestic” terrorism convictions.)
Nora Ellingsen, who worked five years in the FBI as an analyst, has noted that the work of the National Security Division has an “overseas” tilt to it, and that “much of the violence committed by [white supremacist and sovereign citizen groups] is prosecuted under state authorities, and when federal, not necessarily as terrorism.” So perhaps the National Security Division simply doesn’t track “domestic” terrorists, because that’s not its focus. If that’s the case, however, it makes any reliance on NSD data to draw conclusions about the backgrounds of terror suspects highly dubious, since it skews the list towards defendants with international connections and (likely) those who are Muslim. (Of course, it also raises questions about whether our government disproportionately focuses on violence by Muslims. That’s a larger subject than I can address here.)
But there are additional reasons why the NSD list would be a flawed source for the claims the President and Justice Department are making, if they are indeed relying on this data – reasons that suggest the numerator is wrong, not just the denominator. For one thing, the NSD list includes a subsection of 119 individuals whose “convictions arose from the nationwide investigation conducted after September 11, 2001.” The introduction to the NSD chart explains that these individuals were included if they were “identified” during the course of that nationwide investigation and were “subsequently charged with a criminal offense” – but “regardless of whether investigators developed or identified evidence that they had any connection to international terrorism” (emphasis added).
That last clause bears repeating: more than 100 people appear on this list regardless of whether investigators had any evidence they were connected to international terrorism. Some context helps explain this startling admission: after September 11, 2001, the FBI engaged in a sprawling national investigation, arresting hundreds of South Asian and Middle Eastern men for immigration violations or garden-variety criminal offenses. This investigation largely focused on immigrants, as FBI agents pursued tips hastily called in by members of the public reporting “suspicious” foreigners in their midst. Describing the detention of more than 750 non-citizens from mostly Muslim-majority countries, the Justice Department’s own Inspector General scathingly faulted the department for its arbitrary labeling of many immigrants as terrorist suspects with no factual basis for suspicion. (In a recent law review article, I tell the poignant story of Javaid Iqbal, a Pakistani man arrested in the sweeps who later sued the U.S. government for discriminatory confinement and physical abuse).
For present purposes, the point is this: if the NSD list includes people convicted of minor crimes who were simply caught up in the post-9/11 targeting of immigrants, without any connection to terrorism, that same list cannot be used to conclude that most people guilty of terrorism have been immigrants. I don’t know if the President and Justice Department are including this group of men in their numbers, though then-Senator Sessions’ 2016 list certainly did (while notably omitting the Justice Department’s caveats as to reliability).
But there’s more. The NSD document tells us that, after the investigation into the 9/11 attacks, “additional defendants have been added to this chart only if, at the time of charging, they appeared to have a connection to international terrorism, even if they were not charged with a terrorism offense.” Leaving out the group I’ve just described – those who were identified in the post-9/11 investigation – that still leaves over 150 people on the list who were not charged with terrorism offenses. This is why the list refers to “terrorism-related” convictions in addition to “terrorism convictions”: many people on the NSD list were convicted of fraud, drug charges, making false statements, or other criminal offenses that are not explicitly terrorism crimes, because prosecutors apparently believed they had a connection to international terrorism at the time of charging. Now, some of them may well have had a real connection to terrorism, and prosecutors simply chose a non-terrorism charge for legitimate reasons (just as they brought down Al Capone for tax evasion). But in other cases, law enforcement officials may never have established a credible relationship to terrorism, whatever their original suspicions. Treating this category of offenders as convicted terrorists, without more support, is problematic.
I could go on. I could note, for instance, that numerous individuals on the NSD list were convicted of material support to terrorism or conspiracy, controversial charges that may be far removed from common perceptions of terrorist threats. I could note the many questions raised by the government’s aggressive use of sting operations to convince hesitant individuals to take steps in support of terrorism. One might note, for instance, the studies cited in the Human Rights Watch report concluding that nearly half of all terrorism convictions involved informants, and one-third of those cases involved an informant playing an “active role in the underlying plot.” The larger point is that terrorism conviction data do not necessarily tell us much about the distribution of the terrorist threat, because they also reflect the investigative and prosecutorial priorities and strategies of law enforcement. If federal agencies focus their efforts on any particular category of individuals – be it immigrants, Muslims, communities of color, or otherwise – then conviction data will partly reflect back those decisions.
Perhaps none of us is surprised, any longer, when the Trump administration distorts the facts to support its policies. But statements made by the President in a national address to Congress, and then justified by the Justice Department, deserve to be critiqued – all the more so when made in service of a xenophobic agenda.
Shirin Sinnar is an Associate Professor of Law at Stanford Law School. Her scholarship focuses on the role of institutions, including courts and executive branch agencies, in protecting individual rights and democratic values in the national security context. This essay was originally written for and posted on Just Security on March 6, 2017.