From the news today:
A man charged with trying to kill a Danish cartoonist was arrested last year in an alleged plot to harm U.S. Secretary of State Hillary Clinton, officials said. [...] The suspect was one of four people arrested last summer in Nairobi in an alleged plot to harm Clinton during her tour of African countries, the newspaper Politken reported. The suspect was released from a Kenyan jail in September because of a lack of evidence and returned to Denmark, where he had been living, Sky News reported Sunday.
Just a few days ago, CNN reported:
That announcement led to questions about how many other former Guantanamo detainees may be planning to carry out terrorist attacks.Pentagon officials have not released updated statistics on recidivism, but the unclassified report from April says 74 individuals, or 14 percent of former detainees, have turned to or are suspected of having turned to terrorism activity since their release.
Of the more than 530 detainees released from the prison between 2002 and last spring, 27 were confirmed to have engaged in terrorist activities and 47 were suspected of participating in a terrorist act, according to Pentagon statistics cited in the spring report.
More at Wikisource.
These are actually lower than the general population, where about 65% of prisoners are expected to be rearrested within 3 years. The numbers seem lower in recent years, about 58%. More at Wikipedia.
Wait, did you just compare the rate of Guantanamo detainees engaging in terrorist activity after release to the rate of recidivism among released convicts!?
Did you forget that Guantanamo detainees were imprisoned there without a trial, whereas convicts have been been arrested, tried, and convicted of their crimes in open court?
This makes me uncomfortable, because your comparison implicitly assumes that the detainees are all guilty of terrorism.
I think my first example includes someone who was arrested but not tried or detained, so I couldn't possibly be making you uncomfortable :)
Aleks: Thanks for posting this; maybe the criminologists among our readership will follow up. There's a lot of confused thinking out there on recidivism among otherwise-knowledgeable social scientists (see here, for example), and maybe it will help to post these sorts of numbers (maybe in graphical form too).
I find it hard to compare the 14% of former Guantanamo detainees suspected of terrorism after release with the 65% of former prisoners re-arrested: not only is there a difference between being arrested and being suspected, but there surely must be some former Guantanamo detainees who have been arrested for other offenses than terrorism since their release.
Also, 65% seems to be the wrong number to be looking at, anyhow. You'd need to compare to premeditated violent crimes (which terrorism would partially be filed under). Rapists and murderers have a much lower recidivism rate according to that Wikipedia link: "Within 3 years, 2.5% of released rapists were arrested for another rape, and 1.2% of those who had served time for homicide were arrested for homicide. These are the lowest rates of re-arrest for the same category of crime."
Using the 65% comparison, it would look like the US is grabbing a lot of innocent people. Using the ~2% comparison, it looks like they're doing pretty well.
It depends more on the lifestyle choice of terrorism, I suppose. Burglary and assault appear to be habits. Is terrorism a habit?
The Guantanamo numbers are small, so it's hard to do statistics with them. Andrew's suggestion to look for related data is the Bayesian Way: try to find related data which is similar, so that we can smooth the model and understand the phenomenon better.
Many comments indicate unhappiness about how the phenomenons are different. A good model will take this into consideration. In the interest of the investigation of the phenomenon, we need more and better data. Data is always scarce and costly. A (legal) trial is just about getting more data, but we never have all of it.
At the same time, when you have someone get arrested twice for attempted assassination over the course of less than two years - we might not be using the data we already have very well.
Recidivism is the wrong frame for looking at the activities of former prisoners of the Guantanamo facility. The vast majority of them weren't terrorists to begin with. According the Pentagon's numbers, only 5% (27 of 530) have been confirmed to have engaged in "terrorism" after their imprisonment. I put terrorism in quotes because some of those 27 are Afghan nationals whose activities (attacking the uniformed troops of a foreign occupier) isn't considered terrorism under many definitions of the word.
Of the 41 prisoners who managed to get hearings before federal judges, 78% (32) have been ordered freed because there was simply no evidence that the government had any reason to hold them.
What constitutes "terrorist activity"? My understanding is that the DOD counts released detainees who try to sue the USA among them.
Look at Nelson Mandela. Spent 36 years in prison, but since his release, he's kept his nose clean. So, rehabilitation can work.
(OK, I stole it from Ricky Gervais)
I have one or two questions based upon the premise and objective or theoretical process used for this article: 1) statistically the numbers create a distortion of information. My question pertains to the statistical processes and control devices. Recidivism among the incarcerated compares thousands with a wide demographic range and diverse imprisonment formats (security, living arrangements, etc.). Guantanamo detainees (for the most part) were entirely housed under the same control methods and the number used for your statistic was less than 600. The greater the number used in any analysis the greater the accuracy of the assumptions, hypothesis, etc. What is the error percentage you accepted for this article? +/- x% Was it in keeping with formats used in government or accounting procedures? The question isn't a challenge of your information it is a challenge to your process. 2) I did not read your acknowledged bias or limitations. This would clarify how you used the information and to what end you sought to communicate understanding. Without it the data stands alone. That creates a problem. I would like to know what you used to define your criteria and what other data you used to validate your information: theoretically and with an understanding of +/- accuracy.
I did appreciate that you are interested in understanding the human side of this issue and I am interested in how we as a democracy demonstrate and model our commitment to the Constitution.