Schneier on Security
A blog covering security and security technology.
« Breaking Hard-Disk Encryption |
| I Seem to Be a Verb »
December 28, 2012
Becoming a Police Informant in Exchange for a Lighter Sentence
Snitching has become so commonplace that in the past five years at least 48,895 federal convicts -- one of every eight -- had their prison sentences reduced in exchange for helping government investigators, a USA TODAY examination of hundreds of thousands of court cases found. The deals can chop a decade or more off of their sentences.
How often informants pay to acquire information from brokers such as Watkins is impossible to know, in part because judges routinely seal court records that could identify them. It almost certainly represents an extreme result of a system that puts strong pressure on defendants to cooperate. Still, Watkins' case is at least the fourth such scheme to be uncovered in Atlanta alone over the past 20 years.
Those schemes are generally illegal because the people who buy information usually lie to federal agents about where they got it. They also show how staggeringly valuable good information has become -- prices ran into tens of thousands of dollars, or up to $250,000 in one case, court records show.
There are all sorts of complexities and unintended consequences in this system. This is just a small part of it:
The risks are obvious. If the government rewards paid-for information, wealthy defendants could potentially buy early freedom. Because such a system further muddies the question of how informants -- already widely viewed as untrustworthy -- know what they claim to know, "individual cases can be undermined and the system itself is compromised," U.S. Justice Department lawyers said in a 2010 court filing.
Plea bargaining is illegal in many countries precisely because of the perverse incentives it sets up. I talk about this more in Liars and Outliers.
Posted on December 28, 2012 at 6:37 AM
• 17 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
The other perverse incentive is that it encourages the government to imprison criminals for longer sentences, so that plea-bargaining is a sweeter deal; taken to its logical extreme, it is a form of state torture to extract information.
And of course there's also the benefits for conviction rate statistics. With harsher sentences, pleading guilty and generally kowtowing starts having positive returns in a game theoretic sense, even when you're innocent - and this should strike fear into citizens of a would-be free country.
Can't help but wonder if Watkins associates on the outside who gathered the data aren't on the wrong side of Georgia state licensing laws for private investigators and thereby ripe for prosecution.
Got to say it: this brings new meaning to the term "Prisoner's Dilemma"
To see how badly informants can screw up justice, look up the Enzo Tortora case. At the time, Italian law gave disproportionate benefits to informants and even allowed convictions with no other evidence than consistent indication from former organized crime associates.
False accusations and a commonly misheard name were enough to put a man behind bars (mistakes from several informants counting as "confirmations"), and as soon as cases went public you could count on all informants to accuse whoever happened to be the suspect, thus providing even more "evidence" and securing both a conviction and benefits for the accusers.
Eventually, benefits had to be cut and a time limit was put in place (once you start talking, you have six months to testify on everything you know). It's still heavily debated if these rules created more damage than they prevented, given how essential informants are to fight organized crime.
I see this as a strong positive:
It reduces the need for undercover agents and turns criminals into agents. This eventually further undermines the criminal networks setting each other against themselves.
The blowback is these networks (eg, prison and outside networks) become invariably more aggressive towards one another and more dangerous.
The job for undercover cops and technical surveillance becomes much more difficult.
But, the positive is that law enforcement has true penetration of the networks. Preferrably, they gain by this system not just informers, but agent criminals working for them.... effectively, moles.
More moles the better.
The larger the percentage of the network they have under their control, the more unstable that network is.
Really this is the model of foreign intelligence of any value -- moles and agents (who are basically moles)... and it is a very powerful model.
But it appears strongly they are not treating this as the same or similar thing, and that puts at risk any informer they have. This means that criminal networks have a priority to get insiders in law enforcement, just as foreign intel agencies need a "mole to catch a mole".
Law enforcement has a tendency to be extremely weak at this. Internal affairs departments and internal investigations tend to be frowned upon by law enforcement.
Law enforcement is relatively easy to get into compared to intelligence... and they very often have thankless jobs with very poor pay which routinely puts their lives at risk.
Where I live if a bunch of criminals are caught they kill one of their own in jail and then point to the dead guy in court saying it was all his doing. Everybody walks, no justice is served
I think the underlying cause lies outside of this conundrum, very keenly summed up with one unassuming phrase in the article - "wealthy defendants could potentially buy early freedom".
There's another problem with rewarding information which pops up in this article on mandatory sentencing. Talking about a woman who was sentenced to life without parole for having .5 kg of cocaine stored at her home:
Providing evidence to the prosecution is one of the few ways to avoid a mandatory sentence. Because the government formally credited the other defendants with “substantial assistance,” their sentences were all reduced to less than 15 years. Even though Mr. Dickey was the leader of the enterprise and had a much longer criminal record than Ms. George, he was freed five years ago.
Looking back on the case, Judge Vinson said such disparate treatment is unfortunately all too common. The judge [...] says he still regrets the sentence he had to impose on Ms. George because of a formula dictated by the amount of cocaine in the lockbox and her previous criminal record.
“She was not a major participant by any means, but the problem in these cases is that the people who can offer the most help to the government are the most culpable,” Judge Vinson said recently. “So they get reduced sentences while the small fry, the little workers who don’t have that information, get the mandatory sentences.
In other jurisdictions where plea-bargaining is illegal the purpose of the justice (at least declared) is not just to put anybody into prison for any crime (real or alleged by plea-bargaining) and close the case as soon as possible or let go free for cooperation with legal system (police-prosecution-courts), but to find out real circumstances of the crime commited, level of suspect involvement, properly qualify those actions/inactions and provide sentencing based on provisions of the Criminal Code for that particular commited crime. The purpose is truth, not case closing by legal system based on bargain with criminal. On the other (negative or positive based on your own views) side, there is no jury option for trial. Prosecution in those jurisdictions should prove in the court that suspect is guilty based on other evidence even when suspect pleaded guilty (it could be for money, under coercion/intimidation to him of family, etc.). Court when properly qualifying the crime really committed may consider cooperation of defendant with legal system as mitigative factor for sentencing pupose only, but not for changing crime qualification which is absolutely wrong.
Plenty of hackers received double their sentence for becoming informants. They wanted them to go match handles to names at defcon and that felt like personal snitching instead of their 'impersonal snitching' revealing tactics or logins so they refused and got double time.
Marc emery also blogged about an informant he met in prison who was blind yet got 20yrs for having his fingerprints on an empty bag of cocaine.
Then there's Albert Gonzales who was given a huge sentence for continuing his criminal empire under secret service watch. Even though he handed the SS almost two dozen hackers and fraud kingpins it didn't save him from life in jail.
I remember a wired or register article a year or so ago that claimed 1 in 3 hackers on u/g IRC or forums are paid informants. Giant snitchfest
An interesting model for such criminal and intelligence networks can be based on the leaderless cell model. These networks in the proper config cannot be penetrated since identity is not unknown. This reduces danger to all especially for mole and intel heavy threat models.
An interesting model for such criminal and intelligence networks can be based on the leaderless cell model. These networks in the proper config cannot be penetrated since identity is unknown. This reduces danger to all especially for mole and intel heavy threat models.
This reminded me of the failed argument used in one of McKinnon's many appeals against extradition, trying to object to the very concept of plea bargaining - and promptly getting slapped down by the UK's most senior judges, explaining to his lawyer that in fact, "it is accepted practice in this country for the parties to hold off-the-record discussions whereby the prosecutor will accept pleas of guilty to lesser charges (or on a lesser factual basis) in return for a defendant's timely guilty plea", and indeed under Goodyear, judges here have MORE latitude than their US counterparts in plea bargain negotiations, not less.
In a sense, the government itself is always 'buying' the information from somebody (the accused); there are a lot of interesting pitfalls there regarding warrants, due process (when they cross the line from getting information a criminal had already obtained illegally into causing information to be obtained) as well as the more practical matters about how trustworthy it might be.
An interesting model for such criminal and intelligence networks can be based on the leaderless cell model
Which is what Anonymous groups like Lulzsec tried, beit without much success. When the feds got hold of Sabu, he almost immediately turned informant and ratted out the others.
The problem with looking at plea-bargaining as an information market is that no one really has any incentive that the information be accurate. Plausible and potentially convincing to a jury, perhaps, but not necessarily accurate. And, as economists and engineers know, when you optimize for a particular measurable quantity, that's what you get. The quality that that measurable thing is ostensibly a proxy for, not so much.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.