On Friday, January 30, USA Today reported:
“For the first time Sunday, federal behavior-detection officers will team with local police to use a controversial technique on people heading to a major event, the TSA says. The officers usually work in airports. Behavior observation aims to find people in crowds acting unusually. A flagged person gets a casual interview from an officer who determines if he or she should be formally questioned or arrested.”
I am concerned with the use of “behavior-detection” by law enforcement, even by proxy (through the TSA, as is the case here). The standard for being detained by a police officer is that the officer is required to be able to demonstrate a reasonable, articulable suspicion that the person being detained has committed or is about to commit a crime. This type of detention–short of an actual arrest, but not actually free to go either–is known as a Terry Stop.
Currently, behavior-detection does not meet the standards of a Terry Stop because it is too vague and too unreliable, hence the two-pronged approach used at the Superbowl. If an officer does not have reasonable, articulable suspicion, he or she may not detain a person, but he or she may go up to a person and converse with them, just like any other citizen. So the behavior-detection specialist flags a person and the officer goes up and “casually interviews” them. If the results of the interview reveal enough information to escalate the situation to a Terry Stop or a full-blown arrest, then the officer will proceed accordingly.
All of this seems legal, but I still worry. Many people are not aware that, in lieu of a Terry Stop, they are not required to answer a police officer’s questions and are even allowed to end a conversation with a police officer and walk away. Such a person might feel compelled to answer the officer’s questions, inadvertently giving up their 4th and 5th Amendment rights and incriminating themselves or subjecting themselves to an otherwise-unlawful search.
Of course, a police officer has always been free to “casually interview” such a person, so are we really adding anything new here? I think so. Look at the list of “suspicious” characteristics that might cause me to get a “casual interview” from a police officer: sweating, avoiding eye contact, and talking evasively. “Sweating.” So if it’s a hot day and I’m sweating, will I get a “casual interview”? What if I just ran from the parking lot? What if I’m just a naturally sweaty person? “Avoiding eye contact?” Well, I know of at least one case where an officer stated that a person making direct eye contact with her was a sign of aggression, so looks like I’m damned if I do and damned if I don’t. If I make eye contact, I’m being aggressive. If I don’t make eye contact, I’m being suspicious. “Talking evasively?” I have zero interest in talking with a police officer, and I think it’s highly likely that any conversation that I have with a police officer will be interpreted by him as “talking evasively.” And how did the behavior-detection specialist identify that I was talking evasively anyway, when he or she was supposed to be just scanning the crowd? In order to determine that I was talking evasively, he or she would already have to be interviewing me, which rules out this criteria as a way of identifying suspicious people for a “casual interview.”
Given the vague standards and the high false-positive rate of today’s behavior-detection, the behavior-detection specialist can identify nearly anyone as suspicious, with little accountability or verifiability. If a cop says he saw me leaving a house through a broken window with a crowbar, those allegations can be verified. If a behavior-detection specialist says I was sweating and being evasive, that cannot be verified. I especially don’t like behavior-detection overlapping with law enforcement because, if behavior-detection becomes an accepted way of identifying suspicious people, it might later become accepted as reasonable, articulable suspicion, which could give officers carte blanche to Terry Stop anybody they like. Essentially, the whole concept of a Terry Stop would be threatened.
A polygraph test involves sensitive instruments being hooked up to your body to measure things like respiratory rate, respiratory depth, and galvanic skin response (sweatiness). Then a carefully-designed set of questions are asked, to determine truth. And that’s not admissible in court. So you’re going to tell me that a TSA agent can take a seven-day training and learn to tell guilty people from innocent ones just by scanning a crowd? I’m skeptical.

 
#1 by J Frank at December 30th, 2010
I find it interesting that people who are skeptical tend to leave out the OJT and advanced training involved in the BDO program.
#2 by Joshua Bardwell at December 30th, 2010
Frank,
I don’t think it makes a difference to me. My point stands:
Given the vague standards and the high false-positive rate of today’s behavior-detection, the behavior-detection specialist can identify nearly anyone as suspicious, with little accountability or verifiability.
My primary beef is with the lack of accountability or verifiability.
I’m also extremely skeptical of the ability of the average BDO program trainee to implement the technique in an effective manner. I don’t dispute that BDO is a skill that can be learned. Look at the Israeli airport security as an example of a BDO program that seems to work. I just doubt that BDO, as implemented by TSA and other government agencies in the US, is up to those standards. If you look at the level of training that is given to, for example, the TSA, it is pathetic, and the “miss rate” in tests of TSA agents’ ability to detect simulated threats bears this out. My fear is that BDO just amounts to an excuse to relax the usual evidentiary standards that protect citizens’ rights in interactions with law enforcement.