What the Data Really Show About Terrorists Who “Came Here,” Part I: Introduction and Overview
In late February, during his address to a joint session of Congress, President Trump claimed that “according to data provided by the Department of Justice, the vast majority of individuals convicted of terrorism and terrorism-related offense since 9/11 came here from outside of our country.”
A week later, in the revised executive order restricting entry from six countries, the administration declared that, “Since 2001, hundreds of persons born abroad have been convicted of terrorism-related crimes in the United States.”
The White House has offered no data from the Justice Department in support of either of these claims.
Both turn out to have serious factual problems.
A few weeks before the president’s speech, one of us wrote a piece in response to the first travel-ban executive order, a piece which analyzed FBI arrests of international terrorism subjects over the previous two years. The post identified a trend that undermined the security rationale behind the administration’s new immigration restrictions: Natural-born U.S. citizens accounted for 76 percent of international terrorism subjects arrested by the FBI.
As the original article acknowledged, that dataset was limited. When tracking arrests, the piece relied on the Justice Department’s own press releases over the previous two years. Cases that the Justice Department reasonably wouldn’t want to bring to the public’s attention—including cases that were sealed or operationally sensitive—weren’t included in the dataset or the analysis. But the limited data available seemed to cut against the idea that the criminal terrorist cases in the United States were predominantly a problem of foreign origin. And that made both Trump’s statement in his joint-session address—particularly insofar as it was sourced to Justice Department data—and the claims in the revised executive order itself a bit jarring.
Another thing happened around the same time: A new, more comprehensive dataset became available to the public. Shirin Sinnar at Stanford Law School received under a Freedom of Information Act request from 2015 the National Security Division’s list of public and unsealed international terrorism and terrorism-related convictions from Sept. 11, 2001, to Dec. 31, 2015. Writing at Just Security, she pointed out that this list had previously been provided in early 2015 to the Senate Subcommittee on Immigration and National Interest, chaired at the time by none other than then-Sen. Jeff Sessions. The subcommittee had conducted its own open-source research on the immigration status of each defendant and the results were published in February 2017 by the Center for Immigration Studies, a group that favors lower immigration numbers. Fox News and Stephen Miller ran with CIS’s headline: “Study Reveals 72 Terrorists Came from Countries Covered by Trump Vetting Order,” while The Washington Post fact-checked the study. The Post noted that some of the individuals on the list had entered the United States years before they conducted any crime and that many of those individuals were not bomb makers; they engaged in more innocuous activities, such as transferring money.
This new dataset provided an opportunity to evaluate both Trump’s statement in his joint-session address and the factual claims of the executive order itself. This series of posts detail the result of that inquiry.
Here’s the bottom line:
- The data Trump cited in his speech to the joint session of Congress simply don’t support his claims that a “vast majority” of individuals on the list came from outside the United States—unless, that is, you include individuals who were forcibly brought to the United States in order to be prosecuted and exclude all domestic terrorism cases.
- While the data do validate the executive order on its statement that hundreds of convicted individuals were born overseas, they actually don’t support the policy the executive order embodies.
- Of the hundreds of foreign-born individuals, the vast majority were born in countries not covered by the executive order.
- And of the relatively small number of individuals from covered countries—which total 43—the clear majority come from only two countries (Somalia and Yemen), while a vanishingly small percentage of that come from Iran, Sudan, Libya or Syria.
In this first post, we describe our approach to analyzing this data and give an overview of our factual findings. In later posts, we will analyze the data country by country to examine which countries are and are not exporting terrorists to the United States; consider what happens to the data if one includes domestic terrorism cases, as well as international terrorism cases; and offer concluding thoughts on what one can and cannot responsibly say about the relationship between terrorism and country of origin and also about whether “entry” into the United States is a meaningful feature of terrorist crimes in the first place.
This inquiry was labor-intensive because the Justice Department’s list does not include material about the country of origin of terrorist defendants—we suspect because the Justice Department may not actually keep or have such data, which generally are not relevant to the criminality of the suspects. Whether data exist within the Justice Department that support the president’s claim we do not purport to know. But if such data do exist, as we will show, they would have to be quite different from the list of cases NSD released in response to Sinnar’s FOIA request.
Here’s what we did: We went through the raw list of cases and, using our own open-source research, determined the country of origin for each individual defendant on the Justice Department’s list.
Our dataset comprises the National Security Division’s list of public or unsealed terrorism-related convictions from 2001 to 2015. There are 627 individuals on this list. We excluded from our analysis the convictions of 118 of these individuals, which “arose from the nationwide investigation conducted after September 11, 2001.”
It may seem odd to exclude everyone convicted in the 9/11 investigation from such an analysis; it is necessary to do so for a variety of reasons. Most importantly, the Justice Department is candid that it is unable to confirm whether these individuals had an actual nexus to terrorism: “Individuals whose convictions arose from this initial terrorism investigation were included on the chart at that time regardless of whether investigators developed or identified evidence that they had any connection to international terrorism.” This investigation was sprawling and lots of people got swept up in it and were prosecuted for matters having nothing to do with the underlying subject of probe. The result, as Sinnar explains, is that we can’t assume that a nexus to terrorism did exist for most of these convictions. Additional practical reasons advised against including these defendants in our analysis. Given the age of the convictions, fewer public records are available online, and determining country of origin for each defendant grows increasingly difficult with each passing year.
We used open-source material to attempt to determine the country of origin for the remaining 508 individuals. We were able to do so for 455 of these persons. We have excluded from our analysis the 53 persons for whom we were unable to determine places of birth. So unless otherwise stated, the analysis below considers a total set of the 455 individuals for whom we were able to identify nationality, who were convicted of terrorism-related offense between Sept. 11, 2001, and Dec. 31, 2015, and who were not charged in connection with matters arising out of the original 9/11 investigation.
It’s important to emphasize that the bounds of the dataset here artificially inflate the percentage of foreign-born subjects. In other words, the very dataset we’re consulting is the one most apt to support the administration’s position. As Sinnar pointed out, the list doesn’t include any individuals convicted of domestic terrorism offenses; the Justice Department only released the names of individuals who were convicted of international terrorism offenses; these individuals are dramatically more likely to be foreign-born than people convicted of domestic terrorism crimes. Consequently, Sinnar noted, we can draw only limited conclusions from the dataset: “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?”
We focus here on the country of origin for each terrorism defendant not because we think country of origin is an especially illuminating metric but because it is necessary to evaluate the veracity of Trump’s two claims: (1) that the “vast majority of individuals convicted of terrorism and terrorism-related offense since 9/11 came here from outside of our country,” and (2) that “[s]ince 2001, hundreds of persons born abroad have been convicted of terrorism-related crimes in the United States.” The immigration status of these persons is thus generally irrelevant to our analysis. We have, however, included notations on individuals' citizenship or immigration status that we discovered through our research for the purpose of showing that many foreign-born persons committed and were convicted of terrorism-related offenses only after spending years in the United States and, in some cases, becoming naturalized citizens. This finding cuts against the notion that foreign-born persons sought to enter the United States for the purpose of engaging in terrorism on U.S. soil.
What Do the Data Show?
Let’s start with the president’s statement that “according to data provided by the Department of Justice, the vast majority of individuals convicted of terrorism and terrorism-related offense since 9/11 came here from outside of our country.”
In some very technical sense, this is true. Of the 455 persons, 132 are U.S.-born and 323 were born abroad. So yes, since 2001, a substantial majority of terrorist defendants have been born abroad.
But there’s a big problem: 100 of the 323 persons born abroad were extradited, or brought, to the United States for prosecution. This is a group of persons that the U.S. quite literally imported for purposes of prosecution. Including such people in the count of foreign-born folks convicted of terrorism-related crimes in the U.S. is a little like considering Chinese-made products domestic products because we bought them and brought them here. Excluding that group leaves 220 foreign-born persons, which is not quite a majority at all, let alone an overwhelming one.*
(If you’re paying attention to the math here, you might have noticed that the numbers do not quite add up to the total universe of 455 persons. This is because for three individuals—Muhammed Abid Hussain, Yildirim Beyozit Tumer and Mohamed Suleiman Al-Nafi—we were able to determine country of origin but were unable to determine whether these persons had been arrested on U.S. soil or had been extradited from another state.)
In all of the extradition or capture cases, nothing in the defendants’ public court records suggests that the individual had ever tried to immigrate to the U.S.; and if they did, immigration policies at the time were sufficient to deny them entry. To be clear, we aren’t critiquing the use of capture operations or extraditions to bring subjects to face trial in the United States; on the contrary, we’re in favor of using both tools to combat the terrorism threat. But none of these subjects would have been affected by the travel ban. They only entered because we forced them to. So to include these groups as part of any justification for the ban is either disingenuous or sloppy.
Now let’s consider the claim in the executive order itself: that “Since 2001, hundreds of persons born abroad have been convicted of terrorism-related crimes in the United States.”
This claim is true as far as it goes. The problem with it is that only a small percentage of the hundreds of foreign individuals actually come from the countries the order covers. Of the 220 foreign-born persons arrested in the U.S. and convicted of terrorism-related offenses, only 43—or 19.6 percent—were born in the six selected countries.
To be clear, nearly 20 percent is a significant number, but as we’ll describe in our next post, it is actually a somewhat misleading number. The reason is that some of these six countries contribute only a tiny fraction even within that sample. Only two of the individuals were born in Iran, only two in Libya and only two in Sudan. Persons from each of these countries make up only 0.9 percent of the 220 foreign-born persons who were arrested in the United States. (There is one additional individual from Sudan for whom we were unable to determine whether his arrest took place inside or outside the United States.) And only three, or 1.4 percent, were born in Syria. In fact, as we will show, the bulk of that 20 percent involves persons from only Somalia and Yemen. Nineteen persons, or 8.6 percent, were born in Somalia and 15 persons, or 6.9 percent, were born in Yemen.
The data, in other words, may suggest that there’s a Yemen problem or a Somalia problem. But they certainly don’t suggest that there’s a problem with respect to entry from these six countries more broadly. Furthermore, we should emphasize again that this analysis incorporates only the 220 foreign-born defendants arrested in the United States. The representation of these six countries would be even smaller if we include the 132 U.S.-born persons in the total.
This is Part I of a three-part series. Parts II and III are available here and here.
*Editor's note: Several readers have asked us to explain our decision to keep the 100 extradited individuals in the denominator while removing them from the numerator. We made the deliberate decision to exclude them from the numerator because they didn’t “come here” to commit terrorism crimes, as President Trump claimed in his speech to Congress. Someone who was extradited to the U.S. for trial clearly did not "come here" in any sense that the sentence could reasonably be understood to convey. However, we kept the 100 extradited individuals in the denominator because, despite the fact that they were never granted entry to the country, they were convicted in the United States. Because we are assessing President Trump’s claim that “the vast majority of individuals convicted of terrorism and terrorism-related offense since 9/11 came here from outside of our country” it would be inaccurate to remove 100 “individuals convicted of terrorism” from the overall analysis.