Sunday, November 28, 2010

Are Terrorist Threats Changing Constitutional Parameters of Police Behavior? Considering an Incident at the Minneapolis/St. Paul Airport

An opinion of a federal magistrate shows that narcotics detection dogs at airport screening must be deployed before a seizure of a passenger’s carry-on luggage has become unlawful under the Fourth Amendment. The facts of the case also show something about how TSA and airport police hand off cases to each other during screening procedures. Conducting a dog sniff during an initial search of the luggage at an inspection table would have been legal under the magistrate's analysis, but detaining the luggage and the passenger in a separate screening room without probable cause or reasonable suspicion meant that the subsequent sniff by the drug dog was fruit of the poisonous tree and had to be thrown out. Would the result have been the same had a bomb dog been used instead of a drug dog?

A Passenger with $90,000 in Cash. Johnny Chan was selected by an airline for heightened TSA screening on his boarding pass. Cosgrove was working at Security Checkpoint 5 at the Minneapolis/St. Paul International Airport on July 20, 2007. (The diagram from the airport's website shows the configuration of security checkpoints.) Chan’s belongings went through the X-ray machine and Chan walked through the metal detector. The X-ray operator told Cosgrove that there were three large opaque items in Chan’s bag. Chan was led to a screening table where he was frisked by Cosgrove in what the TSA calls a bulk item pat-down. Cosgrove searched Chan’s bag and discovered a black plastic garbage bag with three bundles of cash. Cosgrove asked Chan what the money was for and Chan said it was to buy copy equipment in Chicago. Cosgove called over his supervisor, Finch, because significant amounts of cash were considered unusual circumstances. So far, it appears there was nothing illegal about the procedures.

Finch told Cosgrove to return to his regular duties. Finch asked Chan what he was doing in Minneapolis. Chan said he had been looking at colleges with his daughter but according Finch was unable to name a single college they had visited. Another witness said that Chan had mentioned Concordia College and the court concluded that Chan’s answers had been adequate. Chan appeared nervous and was sweating during the questioning but the court apparently determined that this was not unusual under the circumstances.

Finch told Kohanek, the TSA supervisor on duty, about Chan, indicating he was not satisfied with Chan’s responses. Kohanek walked to Chan with two airport policemen. It was unclear who had called the airport police. Even thought the screening process was completed, Chan was escorted by the two officers into a private screening room, about 20 feet from the screening table. Chan was not told that he was free to leave. It was unclear if Chan’s boarding pass and ID had been returned to him. The magistrate found that the detention became illegal at this point. The purpose of the stop had been completed and the airport police “had neither probable cause nor even a reasonable articulable suspicion that the luggage contained contraband or evidence of a crime.” Chan had not been told he was free to leave and no reasonable person in his situation would have thought so. Arguably, however, Chan’s nervousness in the face of questioning and his inadequate answers, had the court accepted Finch’s version of events, could have justified further inquiry. Nervousness has been found by some courts to provide reasonable suspicion for additional detention during traffic stops and in other situations, allowing for a canine sniff.

Husby of the Airport Police Department spoke to Chan in the private security room until a K9 officer arrived with a dog. Chan was not asked if he would consent to having his luggage subjected to a dog sniff. Husby testified that it was normal procedure to have a narcotics dog smell large amounts of currency. The dog alerted to the cash and Chan was removed from the security checkpoint to the police operations center. (The dog’s alert was most likely to methyl benzoate, a substance in street cocaine. The canine evidence was not challenged by Chan and the reliability of the dog was not discussed in the opinion. Nor was it determined how long Chan had held the cash, which would be important in calculating whether the methyl benzoate got onto the cash while it was in Chan’s possession.) Chan was then told he was free to go but that his cash was being seized. By this time Chan had missed his flight to Chicago. Also about this time, Husby and other officers learned that Chan had an outstanding arrest warrant but the officers decided not to execute the warrant. The warrant was a felony arrest warrant for financial fraud.

The government contended that Chan voluntarily consented to the investigation into the facts surrounding the cash, but the court held that “any reasonable traveler in Mr. Chan’s circumstances would have perceived that he was not free to leave from the moment he and his luggage were seized by TSA Officer Cosgrove, until he was released after the APD refused to return his money to him after the dog sniff.” The magistrate concluded that the $90,000 seized from Chan was the fruit of an unlawful seizure of his luggage and an unlawful detention and the money had to be suppressed. Specifically, the luggage was unlawfully seized when the airport police took possession of it with neither probable cause nor reasonable suspicion. The magistrate stated:

"At the moment of seizure, the police knew only that Mr. Chan had $90,000 in U.S. Currency, that he made no effort to conceal, and that he was transporting to Chicago. Mr. Chan knew the amount of money and answered the officer's questions about what he intended to do with it. Until the dog alerted to the money, the police had found no drug related evidence on Chan's person, or in his luggage. While the police later learned that Mr. Chan had a criminal history, they did not know that until after the dog sniff…. Neither the TSA, nor the police officers were able to articulate a basis for believing the money carried by Mr. Chan was drug related. Their respective descriptions of why they continued to investigate the money rise no higher than the level of a 'hunch', which is insufficient to warrant an investigatory seizure of Mr. Chan's luggage."

The magistrate said that even assuming there was reasonable suspicion allowing continued detention, the length of the detention here exceeded Constitutional bounds of a permissible search and seizure. Had the police executed the outstanding warrant as soon as they knew of it, this might have provided a lawful basis to seize the cash incident to the arrest. U.S. v. $90,000 in United States Currency, 2009 WL 6327469 (D.Minn. 2009).

A Slightly Different Scenario. Let me change a few facts and discuss whether the result should be the same. Instead of financial fraud, suppose that the officials discover that the warrant was issued by a foreign country and may have involved providing financial support for terrorists. Also suppose that instead of a drug dog, the dog was a bomb dog and alerted to the cash. Some explosives contain highly volatile chemicals that could adhere to currency. (Admittedly, terrorists could more easily separate those handling explosives from those handling cash than is generally the case with cocaine on currency, where the currency is often received from cocaine users. I could get around such an objection by just stating that the dog alerted to a suitcase, rather than cash in it.)

It is obvious that the full body scans that are now so riling the air travel public would not be tolerated were their objective to find drugs. The threats of explosives on passengers or in luggage or cargo are so horrifying as to change the boundaries of police behavior, and I suspect, judicial perspectives. Chief Justice Warren, more than forty years ago in Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), put the “threat of physical harm” to the officer involved and potentially others in the balance for determining whether the officer’s actions in a stop were reasonable. If the federal district court had found that the passenger’s nervousness justified detention for the sniff, then the seizure might have been approved. Would the court have favored the version of events argued by the government if the situation had involved possible terrorism? This would not result in a change of the letter of the law, but rather in a court’s tendencies in dealing with the facts of a case. Yet change it would be.

At Berkeley I took a seminar from the great scholar of ancient law, Professor David Daube. In one class Daube asked us why so many of the great works of Roman law dated not from the height of the Roman Empire, but rather from well into its decline. Was there something about legal reasoning that allowed it to flourish as a civilization neared its end? I don’t remember that Professor Daube really answered his question. It was more to make us think but this was decades before terrorism became part of daily life and fear. Now it strikes me that as a society becomes more paranoid when the enemy gets closer, the interpretation of rights is going to change in a reflexive effort to deal with the threat, particularly if the threat is invisible. It may be cloaked as a return to an ancient austerity, from a time when a different enemy was at the gates. Although I cannot say exactly how it will happen, it is obvious to me that cases involving terrorism are becoming a mechanism for restricting freedoms we Americans hold so dear.

For security reasons, some cases are deflected from public exposure in various ways. A case like my hypothetical, for instance, might have been dealt with by putting a homing device in the cash to follow the passenger’s subsequent movements, rather than an arrest. There would have been no case. Unfortunately, we live in interesting times.

Monday, November 22, 2010

Navy Oversight of Canine Operations Needs To Improve: Hazings in Bahrain

The Log Cabin Republicans sued the United States and the Secretary of Defense, Robert Gates, alleging that the “don’t ask, don’t tell” policy violated the rights of its members under the First and Fifth Amendments. Federal Judge Virginia A. Phillips of the Central District of California held that 10 U.S.C. 654, stating Congressional policy concerning homosexuality in the armed forces, was unconstitutional.

One of the stories brought before the court was that of Joseph Rocha, who enlisted in the Navy in 2004 when he was 18. His family had a multi-generational history of military service and Rocha wanted to become an officer in the Marine Corps despite not having been admitted to the Naval Academy. He set his sights on entering the Officer Training School as an enlisted man. After basic training, Rocha was promoted to seaman apprentice and volunteered for a military mission to Bahrain. There he sought out the base’s canine handler position because he wanted to become an explosive-device handler. He tested for a kennel-support assignment, but his contact with other handlers was limited to a few on the night shift.

In time, Rocha passed oral and written examinations to qualify for an assignment in kennel support, where he cleaned, fed, medicated, and exercised explosives detection dogs. Meanwhile, Rocha also earned martial arts, combat, and swimming qualifications. Judge Phillips describes what happened next:

When the eighteen-year-old Rocha declined to participate in the unit's practice of visiting prostitutes, he was taunted, asked if he was a “faggot,” and told to prove his heterosexuality by consorting with prostitutes…. [The canine unit commander, Chief Petty Officer] Toussaint freely referred to him as “gay” to the others in the unit, and others in the unit referred to him in a similar fashion…. When Rocha refused to answer the questions from [the Petty Officer] and others in the unit about his sexuality, “it became a frenzy,” in his words, and his superiors in the canine unit would gather around him, simulate sexual positions, and ask if the U.S. Marine Corps soldiers performed various sexual acts on him…. [The commander] ordered all of the other men in the unit to beat Rocha on the latter's nineteenth birthday.

On one occasion that Rocha testified was especially dehumanizing, [the commander] brought a dozen dogs to the Department of Defense Dependent School for a bomb threat training exercise. For the “training exercise” he instructed Rocha to simulate performing oral sex on another enlisted man … while [the commander] called out commands about how Rocha should make the scenario appear more “queer.” …On another occasion, [the commander] had Rocha leashed like a dog, paraded around the grounds in front of other soldiers, tied to a chair, force-fed dog food, and left in a dog kennel covered with feces.

Rocha did not discuss his sexual orientation with his commanding officer, who was obviously not tolerant, but in any case he wanted to comply with the don’t ask, don’t tell policy. Eventually, Rocha returned to the United States and was assigned to the Lackland Air Force Base for Military Working Dog Training School. After completing this training, he returned to Bahrain and encountered the same atmosphere. The new commander was as prejudiced as the last, telling Rocha that he was everything the commander hated, “liberal, Catholic, and gay.” The new commander continued haunting and harassing Rocha and put an image of two men engaging in homosexual activity on Rocha’s computer. Despite the egregious mistreatment, Rocha at first declined to complain about a superior even when he was being investigated on another matter.

Rocha received numerous commendations and was awarded the Navy and Marine Corps Achievement Medal and in 2006 received the sole nomination from his Congressman for entrance into the U.S. Naval Academy. At that time, reflecting on the long-term commitment a Navy career would entail, Rocha informed the Navy of his sexual orientation. Several Officers tried to dissuade him from coming out, telling him he was being considered for various honors and leadership positions, including a battalion leadership. Rocha persisted and received an honorable discharge.

While it has not been my intention to enter into non-canine policy debates, I do think losing this man was a serious loss for the Navy. What concerns me here is that this case refers to practices that suggest that some military working dog commands overseas may not be run very well and that oversight may be deficient or lacking several steps up the chain of command. Why was there a kennel covered with feces? Why was a person dogs were expected to obey put into positions that the dogs might see as justifying a lack of obedience on their part? Why did a training exercise involve simulated sexual positions? What were the dogs being taught to do? Sniff terrorists at gay bars in Bahrain?

Maturity is crucial in working with dogs, particularly with explosives detection dogs expected to have a high level of reliability. Putting the command of a military dog unit in a dangerous location in the hands of individuals with the maturity of a bunch of drunken frat boys should be of grave concern to the military and taxpayers alike. The Navy has said the incident is under review. Rocha has written about his experience. Toussaint was forced to retire.

Log Cabin Republicans v. U.S., 716 F.Supp.2d 884 (C.D.Cal. 2010).

Wednesday, November 17, 2010

Using a Bite-and-Hold Trained Dog for Search and Rescue Leads to Lawsuit

The police receive a call that a boy is lost, drunk, and wearing very little on a cold night. An officer arrives with a dog trained in tracking and suspect apprehension. The dog apprehends with a bite and hold. The department has a tracking dog that is used in search and rescue, but it is not available because of a recent injury. Should the police use the bite-and-hold dog? The issue recently faced a canine officer in Maryland. I'll refrain from editorializing while the matter remains in active litigation.

Judge J. Harvie Wilkinson of the Fourth Circuit Court of Appeals tells the story better than I could:

Around 7:00 p.m. on the night of March 17, 2006, Oscar Melgar, then thirteen years old, learned about a birthday party from his friend, Brian Bentacur. Oscar received permission from his mother to go to the party with Brian, and Brian's mother agreed to take the boys. The party was in Gaithersburg, Maryland, approximately thirty minutes from Oscar's home in Rockville, Maryland.

Brian's mother picked up Oscar around 8:15 p.m., and the boys arrived at the party around 8:45 p.m. Approximately thirty minutes after the boys arrived, an older youth, who was about seventeen years old, started pouring mixed alcoholic beverages of rum and cola. The beverages were served in eight-ounce cups. Oscar knew he was being served alcohol, although he had never had a drink before. Within ten minutes of his first drink, Oscar consumed two additional eight-ounce rum and colas. He began to feel the effects of the alcohol in the middle of his third drink but still finished it.

Brian also was drinking, and one of the girls at the party complained that he was becoming “too touchy.” Oscar was told to take Brian outside to “walk it off.”

The outside temperature was in the upper thirties or lower forties, and even Oscar later admitted that it was cold. Oscar was wearing jeans and had a green warm-up jacket over his shirt. Brian wore jeans but only had on a long sleeved shirt.

The boys started walking, intending to circle the block, but they were not familiar with the neighborhood and got lost. As they walked, Oscar could feel the effects of the alcohol growing stronger, and Oscar testified that Brian also appeared to be growing increasingly intoxicated.

The two boys eventually sat down on a lawn. Oscar had a phone but wanted to sober up before calling Brian's mother to pick them up. After sitting on the lawn for a few minutes, Oscar saw a man walking a dog approaching. Oscar, who admits he was drunk and not thinking straight by this point, got up and walked away. He crossed the street, headed down a sidewalk, and eventually crossed a lawn toward a home, where he passed out under a holly bush. He stated that he went under the bush “so nobody could see me drunk.” His last recollection was “getting cold, and trying to get warm.”

Around 10:40 p.m., Jessica and Michael Sommerville were walking together when they discovered Brian Betancur lying on the ground. Ms. Sommerville testified that she saw another boy stumble away from Brian as they approached. Ms. Sommerville did not recognize Brian, and she began to question him. She testified that Brian did not appear to be aware of the cold weather, even though he was only wearing a tee shirt. In addition, his speech slurred at times during their conversation, although at other times it was clear. Because the situation appeared unusual, Mr. Sommerville called 911 and reported finding a boy lying on the ground.

Both paramedics and the Montgomery County Police Department responded to the call. Officer Todd Uvary, the first officer to respond, saw that Brian Betancur “was extremely intoxicated” and had vomited and urinated on himself. He was told by paramedics that Brian was suffering from hypothermia and a possible alcohol overdoes. Both Sommervilles told the police about the other boy they had seen, and Ms. Sommerville described his clothes. She also told the police that he stumbled away and sounded like he was vomiting, although Oscar later denied being sick. Officer Uvary communicated this information to the next two officers to arrive, Officers Holland and Camp. Concerned about the possibility that another intoxicated, lightly clothed boy could be outside, the three officers circled the neighborhood for approximately twenty minutes, using spotlights mounted on their cars to look in yards. At least one officer also got out of his car with a flashlight and looked in back yards.

As the officers were unsuccessful in the initial search, Officer Camp called a sergeant for advice and was told that it would be acceptable to bring a canine unit to the scene. Officer Camp called one of the appellants, John Greene, who was dispatched at 11:27 p.m. When Greene arrived, the other officers told him that Oscar was last seen approximately an hour earlier and that they had already made an unsuccessful search on foot and by car. Greene also was told that Brian had been taken by ambulance to the hospital for possible alcohol poisoning and hypothermia.

Greene decided to use his patrol dog, Carter, an animal trained to find individuals and to bite them when he came in contact, to make a quick search for the missing boy. While Greene normally would have used a bloodhound to perform such a search, his bloodhound was out of service with a knee injury. Because the police department's other bloodhound was off duty, Greene estimated it would have taken approximately another hour for it to respond. Additionally, because no scent item was available, a bloodhound search would have been difficult even if one had been available.

Greene also evaluated several other factors in reaching his decision. He considered that there was no definitive evidence that anyone was actually missing or endangered. However, if anyone was in fact missing, the cold weather, the presumed intoxication, and the elapsed time were all of serious concern. Greene also realized that no criminal suspect attempting to evade capture was involved, making it likely that he would see anyone before a dog got close enough to bite.

After deciding to use Carter to search, Greene put the dog on a fifteen foot lead. He did not muzzle Carter, and there is a factual dispute as to whether a muzzle would have significantly limited Carter's tracking ability.

Greene took Carter to the last point where Oscar was reported seen and commanded Carter to “track.” Carter cast about looking for a scent and then began to track down the sidewalk. Greene began to call out to Oscar, saying he was there to help and to take him home. Oscar states he never heard anyone, but the district court gave no credence to his assertions because it recognized Oscar was inebriated and not in a condition to hear. After tracking a short distance down the sidewalk, Carter turned sharply across a yard and went into a holly bush where Greene could no longer see him. Greene did not interpret the turn as indicating that anyone was close. The lead went slack, and only then did Greene realize that Carter had found Oscar, who was asleep. By the time Greene realized what was happening, the dog had already bitten Oscar's lower right leg.

Greene testified that he did not verbally call Carter off because Oscar was struggling, and Greene was concerned that if the dog released he might re-bite Oscar's face. Instead, Greene walked up the leash and physically removed the dog. While this was happening, the dog pulled Oscar some five or six feet…. Oscar suffered two lacerations on his lower right leg just above the ankle, one approximately 4.25 inches and the other 1.5 inches long.

The injuries to the boy’s leg resulted in a permanent disruption in sensation around the scars. He will have to wear a protective ankle brace for athletic activities indefinitely.

Oscar’s father, as Oscar’s guardian, sued Officer Greene and Montgomery County, both of which moved for summary judgment, which the district court denied. The defendants appealed the denial.

Judge Wilkinson first concluded that a seizure had occurred, noting that a seizure can take place “if police purposely detain a person but somehow use more force than intended.” Citing Supreme Court dicta in Brower v. County of Inyo, 489 U.S. 593, 109 S.Ct. 1378, 103 L.Ed.2d 628 (1989), the Fourth Circuit said that as long as the instrumentality (here, a dog) was intended, a seizure occurred even if the degree of the instrumentality’s effectiveness was unanticipated.

Officer Greene chose to use a dog conditioned to bite its target when found and thus cannot claim that there was no seizure when the dog he set in motion behaved exactly as it was trained to do.

The court noted that the handler believed the boy was not eluding pursuit and would likely be out in the open and visible before a dog could bite him. The Montgomery County Police Department guidelines allowed the use of patrol dogs to search for missing persons.

Nevertheless, the court concluded that there were factual questions that made summary judgment on the excessive force claim inappropriate. Why was the dog not muzzled? Why was a long (15 ft) leash used? Should the officer have recognized from the dog’s behavior that it was close to the boy?

The officer argued that the dog would not smell inside of a muzzle. He argued that the dog could not cast properly without a long lead.

Prior cases have referred to different lengths of leads. In Trammell v. Thomason, 559 F.Supp.2d 1281 (M.D.Fla. 2008), the officer used a six-foot lead to pursue a criminal, while in Peterson v. City of Federal Way, 2007 WL 2110336 (W.D.Wash. 2007), the handler used a 30-foot lead to search for a hit and run suspect with an outstanding arrest warrant.

Finally, there is a dispute over when a reasonable officer should have realized that Carter was close to locating Oscar. Officer Greene testified that he did not realize that Oscar was near when his dog reversed course and cut across the yard right before entering the holly bush where Oscar was lying. Plaintiff's expert, however, stated that the shift in course was significant and that a reasonable officer should have realized what was occurring.

The court felt that cases where a dog was released from a leash in pursuit of a criminal were not analogous in that letting a dog run free, knowing that it would bite its target when found, was a sufficiently different situation as to provide no useful guidance.

The International Association of Chiefs of Police (IACP) has a policy explicitly allowing the use of police canines to track missing persons so long as the dogs “remain on a leash of sufficient length to provide a reasonable measure of safety to the subject of the search without compromising the canine’s tracking abilities.” In a concepts and issues paper from 1992 (revised 2001), Law Enforcement Canines, the IACP adds that “caution must be taken when using law enforcement canines to track children or other civilians who are simply lost.”

The court held that Officer Greene was entitled to qualified immunity. State law claims survived, however. Judge Michael dissented on the determination that qualified immunity applied, noting the dangers of using a dog taught in the bite-and-hold method.

Melgar v. Greene, 593 F.3d 348 (4th Cir. 2010). On remand, the federal district court declined to accept jurisdiction on the state claims and the case has moved to the state court system.

Thursday, November 11, 2010

The Dogs of the Night Watch

In the urban centers of northern Europe in the Renaissance, a lowly and lonely profession was that of the night watchman. Those wandering the dark streets at night could often be presumed to be up to no good and night watchmen were usually armed against the dangers. It is therefore not surprising that night watchmen would acquire dogs to add the animals' senses to their own, as well as to combine the jaws of dogs with the weapons of their arsenels.

The title page of a book by Thomas Dekker printed in 1608, reproduced here, shows a night watchman of London carrying a lantern and a pike and followed by a dog. The dog, more alert than the master, has turned backwards towards a sound or a threat, its face showing almost human suspicion if not anger. No leash is depicted but the dog's collar appears unnaturally lifted towards the front. Perhaps the leash was left out because the artist wanted the teamwork of the pair, rather than a leather strip between them, to intertwine their fates. Or perhaps there never was a leash and I'm getting overly creative in my interpretation.

A woodcut from not too much later shows a night watchman of Antwerp subduing a criminal using just such a pike and a dog. The dog is biting the criminal in the leg while the watchman prepares to stab the man in the buttocks. Schoon and Haak, in their book, K9 Scent Identification, reproduce the woodcut and a greeting card from the night watchmen of Leuven in 1786, showing five men following or being followed by their dogs. These authors suggest that the companions of the night watchmen may have been the first police dogs and I have found no evidence to disprove this idea.

The dogs in these seventeenth century depictions are all medium to large, unlike the dogs leading blind men that I have described in prior blogs (October 9, September 14, July 19). Although the relative size of the dogs may have been adjusted for the artists’ purposes, it does seem that guard dogs were generally large, while guide dogs were small because of the poverty of their masters and the difficulty they would have had in feeding and caring for the animals.

Some of the dogs of the night watchmen may have been trained to attack. Ulysses Aldrovandus, in his natural history, De Quadropedibus, printed in 1621, writes that war dogs were trained as follows:

"The [war] dog ought to be trained up to fight from his earliest years. Accordingly some man or other is fitted out with a coat of thick skin, which the dog will not be able to bite through, as a sort of dummy: the dog is then spurred on against this man, upon which the man in the skin runs away and then allows himself to be caught and, falling down on the ground in front of the dog, to be bitten. The day following he ought to be pitted against another man protected in the same manner, and at the finish he can be trained to follow any person upon whose tracks he has been placed.... Blondus is even of the opinion that from time to time it is a good thing to go for this dog with drawn swords."

Aldrovandus reports that war dogs were more common in Spain than elsewhere in his time, and there are records of their murderous use by conquistadors in the New World. Stephanitz cites the 1473 edition of Mynsinger as stating that dogs were taught to "stand up to a man, who was clothed in a stout coat of skins lest the dog should bite him during his education."

Rembrandt’s Night Watch, displayed in the Rijkmuseum in Amsterdam, contains the figure of a dog towards the lower right of the painting. The dog was pointed out to me more than a decade ago by Milton Bellis when we stood beside the painting the night of a reception at the museum for the International Swaps and Derivatives Association. Milton, once a UPI correspondent in post-WWII Europe, knew some of the scholarship on Rembrandt's painting. The painting, earlier known as The Company of Frans Banning Cocq and Willem van Ruytenburch, may have nothing to do with night watchmen. The dog, the only animal in a painting of 34 people, barks at a drummer. It has been suggested by the art historian Georges Boka that the dog in the painting is a symbol of death. At a time when one could not turn on a light bulb in the night perhaps it was easier to imagine a dog barking at a soul searching for the boatman to cross the final river.

The tradition of night watchmen using dogs may have continued until the beginning of formal canine units in police forces. Policemen who used dogs generally bought their own initially, something mentioned in a number of nineteenth century and early twentieth century U.S. legal cases. J.J. Sullivan, official historian of the New York City Retired Transit Police Officers' Association, has found a notation in the records of the NYC police for 1853 that may indicate the police had in that year purchased "1 dog." It is not clear what the police department, almost 50 years before the first police canine units on record, would have wanted with a dog. Dogs were known to be used to guard warehouses and armories in the nineteenth century, and it is possible that some such use was being made of the dog.

Sources: Thomas Dekker, Bringing to Light the Most Notorious Villanies that are Now Practised in the Kingdome. Profitable for Gentlemen, Lawyers, Merchants, Cittizens, Farmers, Masters of Householdes, and All Sorts of Seruants to Mark, and Delightfull for All Men to Reade (limited edition (200) linoleum block print). For a reproduction of a woodcut showing a night watchman with a lantern, a staff, and a dog, see William Chappell, The Roxburghe Ballads, IV, 470 (Ballad Society, Hertford 1883); Adee Schoon and Ruud Haak (2002). K9 Suspect Discrimination: Training and Practicing Scent Identification Line-ups (Detselig Alberta, Canada). Mark Derr (1997 and 2004). Dog’s Best Friend: Annals of the Dog-Human Relationship (University of Chicago Press 2004), 153; Stephanitz, M.v. (1923) The German Shepherd Dog in Word and Picture, 336. Anton Kampfe, Jena, Germany, quoting Mynsinger, Heinrich (1473). Puoch von den valken, habichten, spreben, pfariden, und hunden. Stephanitz notes that Mynsinger was using older sources. Georges Boka (1994). Rembrandt's Nightwatch: The Mystery Revealed, 88 (Boulanger-LithoChrome Inc. Montreal).

Thanks to The American Kennel Club Library for providing me with the translation of portions of Aldrovandus contained in an Appendix XIV of Edward Ash, Dogs: Their History and Development (London 1927).

Tuesday, November 9, 2010

Canine Remote Explosive Scent Tracing, Long Used in Europe for Screening Air Cargo, Continues to Be Tested in U.S.

Federal law mandates searching checked luggage, which may be done by canine teams, certain equipment, and other means, but luggage is not the only thing that goes into the hold of a plane. Air cargo also goes there. As recent events have made clear, cargo may not be screened as effectively as luggage. Although Congressional and Administrative policy has been to increase the number of explosives detection canine teams at U.S. airports, one of the more sophisticated explosives detection approaches awaits implementation in the U.S. despite widespread use in Europe.

A landmine detection methodology first developed in South Africa in 1985 (see blog of May 29, 2009) has in the last decade been adapted to sample air cargo for explosives. This method, called remote explosive scent tracing (REST) or sometimes remote air sampling for canine olfaction (RASCO), involves using dogs to sniff air samples at stations that are located in sterile rooms, much like those used in advanced scent identification procedures. The samples are extracted from cargo containers, or from under the plastic wrapping that holds a large number of packages together for shipment, using a suction device so that the sample presented to the dog comes from a confined airspace. An alert by a dog has a high probability of identifying the presence of an explosive among the items within the confined airspace.

Dogs are not the only animals used in REST work. One organization involved in removing landmines in Tanzania and Mozambique has had considerable success with rats.

Remote explosive detection is particularly useful with cargo because of the fact that unbundling materials being shipped together is both expensive and time consuming, and in some cases, such as with pharmaceuticals, can damage or destroy a shipment. About 12% of air cargo in the U.S. travels on passenger planes, but about 16% of air cargo coming into the U.S. reportedly comes on passenger planes. Air cargo operators in the U.S., such as FedEx and UPS, tend to have dedicated cargo fleets, and some passenger airlines, such as Northwest and United, also have dedicated cargo fleets. Overseas, about 22% of air cargo travels on passenger planes. It thus appears that over 80% of air cargo in the U.S. is not subject to the screening requirements that apply to cargo placed in the holds of passenger planes, even assuming the latter requirements are consistently met.

Use of REST in Europe

In 2004, a commercial system began to be used by agencies of the governments of France, and the UK approved it for screening cargo. The Netherlands Ministry of Justice and Military Police approved it in 2006. (See written question of Jeanine Hennis-Plasschaert, a member to the Dutch House of Representatives, to the EC, February 28, 2008.) REST is also used in the Amsterdam Schiphol and Johannesburg airports. The picture (courtesy DiagNose) shows a dog checking stations in a REST system.

In 2007, the Government Accountability Office published a report on aviation security, Federal Efforts to Secure U.S.-bound Air Cargo Are in the Early Stages and Could be Strengthened. The GAO discussed remote scent tracing:

"[T]wo European countries [presumably England and France] are currently using canines in a different manner than TSA to inspect air cargo for explosives. Specifically, these countries are using the Remote Air Sampling for Canine Olfaction (RASCO) technique, which involves the use of highly trained dogs to sniff air samples collected from air cargo or trucks through a specially designed filter. The dogs sniff a series of air samples to determine whether or not there is a trace of explosives and indicate a positive detection by sitting beside the sample. According to foreign government officials representing two of the countries that use this technique, tests to determine the effectiveness of this practice have shown that RASCO has a very high rate of effectiveness in detecting traces of explosives in cargo. According to foreign government officials, this inspection method can be used on cargo that is difficult to inspect using other methods, due to size, density, or clutter, and does not require the breakdown of large cargo pallets. Further, officials stated that the dogs used in RASCO do not tire as easily as dogs involved in searching cargo warehouses, and can therefore be used for a longer period of time…. According to TSA officials, while the results of previous agency tests of RASCO raised questions about its effectiveness, they continue to work with their international counterparts to obtain information on the feasibility of using RASCO to inspect air cargo. TSA officials stated that the agency has not yet determined whether RASCO is sufficiently effective at finding explosive in quantities the could cause catastrophic damage to an aircraft and whether this technique will be approved for use in the United states."

In a 2010 report on air cargo security, the GAO referred to RASCO again, but made no mention of its possible implementation in the U.S. The GAO did state that RASCO might "produce different results from TSA's screening standards," but it is not clear if this was a reference to any research TSA had conducted.

The Scientific Working Group on Dog and Orthogonal detector Guidelines (SWGDOG), which suggests industry standards on the training and uses of police dogs, said in its 2007 Guideline SC7-Research and Technology that independent scientific evaluations of the capabilities and limitations of the REST systems are essential. A 2010 proposed revision includes the same urgent recommendation. In Norway, Rune Fjellanger, the preeminent researcher on the method, reported 95% detection reliability eight years ago, thus meeting the U.S. military's standard for explosives detection dogs.

How Effectively Is Air Cargo Screened Before Going on Passenger Planes?

The bombs hidden in printer cartridges found at airports in East Midlands and Dubai last week contained PETN (pentaerythritol tetranitrate). East Midlands Airport is the UK’s busiest pure cargo airport, handling 300,000 tons of flown cargo every year. It is the UK hub for DHL and UPS, and it was in a UPS package that the disguised explosive device was found. News reports stated that the bombs had been on passenger planes during their journeys.

The Implementing Recommendations of the 9/11 Commission Act of 2007 (PL 110-53, August 3, 2007) contained provisions (§ 1602) on the screening of cargo carried aboard passenger aircraft, which required that, by August 2010, “the Secretary of Homeland Security shall establish a system to screen 100 percent of cargo transported on passenger aircraft operated by an air carrier or foreign air carrier in air transportation or intrastate air transportation to ensure the security of all such passenger aircraft carrying cargo.” Screening may include various technologies including “explosives detection canine teams certified by the Transportation Security Administration.” The Act also encouraged the use of blast-resistant cargo containers (§ 1609).

In 2008, the GAO reported that there were 370 TSA-certified explosives detection canine teams cross-trained to work in multiple aviation environments, including air cargo. Teams specializing in air cargo were located primarily at the 20 airports that receive approximately 65% of the air cargo transported in the U.S. In June 2010, the Transportation Security Administration issued a statement, 100% Air Cargo Screening: Path Forward, admitting that “screening 100 percent of cargo inbound to the US will not be attainable by industry by the August [2010] deadline, however TSA is committed to ensuring industry obtain that level of security as rapidly as possible.”

A Report to Congress in 2007 noted that it was not only the amount of cargo that was shipped on passenger airlines that was a problem for screening everything, it was also the complexity of the system. About 50 air carriers transport air cargo on passenger airlines from nearly two million shippers every day. Approximately 80% of the shippers use freight forwarders who operate about 10,000 facilities in the U.S. The TSA has, therefore, relied on a strategy of establishing databases of known shippers and freight forwarders and comparing shipments to entities labeled as approved in databases. When a match is not found, a targeted inspection can be initiated. The report stated that TSA policy is to “screen, inspect, or otherwise ensure the security of all-cargo aircraft is to be established as soon as practicable, but sets no specific deadlines or time frame for compliance.” When air cargo arrives in the U.S. from abroad and is transferred to a domestic passenger flight, TSA's policy is that this cargo "must be made available to TSA canine teams when these teams are present in cargo facilities."

At present, when canine teams inspect cargo, the procedure primarily involves running a dog through a loading area or having it stand beside a conveyer belt where it sniffs the packages as they pass. One bomb dog handler advised me that sometimes air cargo facilities are concerned about a shipment wrapped in plastic but do not want the wrapping unbundled, so he slits the package on each side allowing the dog to sniff the air inside the wrapping. He acknowledged that a remote air scenting system would have been preferable.

In a 2009 report reviewing TSA’s progress in screening, the GAO said that there were only 37 canine teams dedicated to air cargo screening, but added that there would soon be 48 more, meaning that there are now somewhere around 100 dedicated air cargo sniffing dogs. The report said that TSA was attempting to identify peak cargo delivery times during which dogs would be most helpful for screening. This suggests that packages arriving at non-peak times may not be screened.

EU Requirements


The European Commission provides standards for aviation security in the EU. After April 2009, European Commission Regulation 272/2009 allows screening of passengers and employees at checkpoints to be performed by dogs, as well as cabin baggage and baggage carried at a checkpoint, hold luggage and other cargo and mail to be loaded into the hold. Vehicles and the aircraft themselves may also be searched by explosive detection dogs.

In Regulation 573/2010, the Commission elaborated considerably on explosive detection dog requirements. Dogs are to be single purpose dogs and are to be taught to give a passive response. The EDD and the handler must both be approved independently and in combination. Both are to receive initial and recurrent training. Recurrent training is to occur at intervals of no longer than six weeks. Recurrent training sessions are to be at least four hours long. Training records are to be kept on both the EDD and the handler. Finally, the Regulation provides that security screening may involve free running or remote explosive scent tracing. REST may also be used for screening cabin and carry-on baggage, vehicles, aircraft, in-flight supplies and airport supplies. Dogs used in REST work can have two handlers, presumably a recognition that dogs can do this work for longer periods than is the case in traditional sniffing assignments.

Where Does TSA Stand on Implementing REST?

A 2008 Report to Congress stated that TSA was working with the Department of Homeland Security on operational procedures to improve canine detection capabilities. The Report stated: "One technology being examined is Remote Air Sampling Canine Olfaction (RASCO) sensors, which can provide a concentrated sample from a container for a canine to inspect and has been used extensively in Europe. The DHS project plans to expand this concept to include chemical sensors carried on jackets worn by the canine that will be capable of transmitting data to remote monitoring stations."

Perhaps TSA's efforts to combine the dog's sniff with a chemical sniff explains the delay in implementing REST in the U.S.

The Technical Support Working Group, an interagency program for research and development into counter-terrorism measures, has posted a Combating Terrorism Technical Support Office 2009 Review. A section of the Review on "next-generation canines" indicates that TSWG is exploring "technologies that support and enhance canine detection." TSWG is under the Combating Terrorism Technical Support Office (CTTSO), which is under the Assistant Secretary of Defense for Special Operations/Low-Intensity Conflict & Independent Capabilities. Proposals for RASCO testing were invited by CTTSO in October 2008 (Broad Agency Announcement 09-Q-4500). The following describes the tests that were sought:

"The performing organization will be required to design, develop and operate a RASCO test-bed for air cargo screening. This test-bed shall be a model of activities of an air-cargo facility which has deployed the RASCO screening process. A warehouse where cargo can be assembled into air shipping configurations and then arranged for sampling will be required. The warehouse shall also have a separate area for canine interrogation of the samples. The test-bed shall have separate storage areas for used and unused commodities and shall have storage facilities for bulk amounts of explosives. The investigating entity shall be fully licensed to receive, store, handle and ship bulk explosives. (This can happen up to two months after award if necessary). The site shall have the capability to accept and ship large amounts of air cargo commodities in their shipping configuration or broken down as required. The contractor shall have knowledge of U.S. air cargo screening requirements. The contractor shall procure and train canines to detect trace amounts of explosives using a government furnished protocol. Canines must be maintained for long periods of time at a location remote from explosive storage. It is noted that the training procedures for RASCO may not be the same as for other types of canine detection. These training procedures will be provided by the government. The training aids [generally samples of explosives] used shall be fabricated by the performer using established procedures. Adequate facilities shall be available for the formulation of training aids. The procedures for fabrication will be supplied by the government. The option to use the test-bed as a permanent facility for training and testing is desirable. Authentic air cargo samples must be taken and used for training purposes. The awardee must be able to supply these samples to the test-bed. A location near an air cargo hub is desirable."

It would seem then that testing is going on, and may have been going on for several years.

Update Note:  In testimony before Congress in 2014, TSA officials stated:

"Recommendations of the 9/11 Commission Act of 2007 directed DHS to examine the use of third party explosive detection canine teams for air cargo screening. In 2011, TSA, in coordination with the DHS Science and Technology Directorate (S&T), conducted a Third Party Pilot Assessment to examine the use of these teams in the cargo environment. TSA and DHS S&T analyzed current industry detection canine capabilities through a pilot to determine the degree of modification to industry programs needed to adopt and implement TSA screening standards. The assessment revealed inconsistent results of industry programs due to unsatisfactory odor recognition and performance. However, TSA remains open to future proposals on third party canine use."

Conclusion

It is unlikely that the Transportation Security Administration would make a public statement regarding a specific commercial provider, but it should acknowledge the existence of the approach and place it in an advisory, if not a regulatory, context. REST has been tested by various countries and at various cargo facilities, both air and maritime (as well as for downloading air samples from trucks in some locations), and the risks associated with air and marine cargo argue for implementation as soon as is practicable.

Additional Sources: Fjellanger, R., Andersen, E.K., and McLean, I.G. (2002). A Training Program for Filter-Search Mine Detection Dogs. International Journal of Comparative Psychology, 15, 277-286. Fjellanger, R. (2002) REST—A Method for Detection of Explosive and Chemical Substances. Abstract in Proceedings, Vapor and Trace Detection of Explosives for Anti-Terrorism Purposes, Technological Centre, Russia. Uddqvist, A. and Roberthson, I. (2010). Improvement of Sampling System for Remote Explosive Scent Tracing. Bachelor’s Degree Project in Mechanical Engineering, University of Skovde, Sweden, Spring 2010. See also Fjellanger’s website; Meier, B. and Lipton, E. (2010). In Air Cargo Business, It’s Speed vs. Screening, Creating a Weak Link in Security. New York Times, November 2, A12 (citing the International Air Cargo Association that 16% of cargo carried into the U.S. comes on passenger planes); ICTS Europe Q1 Newsletter 2010 (canine trials at central search areas of East Midlands and other UK airports, indicating full implementation had not been approved); Elias, B., CRS Report for Congress: Air Cargo Security (updated July 30, 2007);TSA Air Cargo Programs Update (FY 2008 Q2) (showing that only 12% of U.S. air cargo goes by passenger aircraft. This was the TSA Air Cargo Programs Update describing the Certified Cargo Screening Program (CCSP)); Statement of Inspector General Kenneth M. Mead, DOT, Key Issues Concerning Implementation of the Aviation and Transportation Security Act, Report CC-2002-098 (February 5, 2002); Wickens, B. (2001). Remote Air Sampling for Canine Olfaction. IEEE 35th International Carnahan Conference on Security Technology, 100-102.

Addendum. I thank readers for sending me comments and additional leads regarding this subject since I first posted the blog two days ago. Perhaps the biggest advantage of an increased readership is that people are happy to point out errors and oversights. Even before I opened the New York Times this morning (November 11), I was directed to the story John Burns on A6, Yemen Bomb Could Have Gone Off at East Coast, stating that the package in Britain had been missed by sniffer dogs and explosives-detection equipment. I have been advised that there is no REST system at the East Midlands cargo facility, so if this is where dogs did not alert to the package, they must have been taken past it by a handler.

Second Addendum. Steve Lord, Director of Homeland Security and Justice Issues with the Government Accountability Office, testifying before the Subcommittee on Transportation Security of the House Committee on Homeland Security, on March 9, said that the Transportation Security Administration had, as of February 2011, 113 dedicated air cargo screening teams working at 20 airports, but would be adding seven additional teams. He also stated that a pilot program at two certified cargo screening facilities (CCSFs) was testing the feasibility of using private canine teams to inspect air cargo. Certain air cargo is not screened by dogs, however, as is made clear from the following portion of Lord's statement:

"TSA has not approved or qualified any equipment to screen cargo transported on unit-load device (ULD) pallets or containers—both of which are common means of transporting air cargo on wide-body passenger aircraft—both domestic and inbound aircraft. Cargo transported on wide-body passenger aircraft makes up 76 percent of domestic air cargo shipments transported on passenger aircraft. The maximum size cargo configuration that may be screened is a 48-by-48-by-65-inch skid—much smaller than the large pallets that are typically transported on wide-body passenger aircraft. Prior to May 1, 2010, canine screening was the only screening method, other than physical search, approved by TSA to screen such cargo configurations. However, effective May 1, 2010, the agency no longer allows canine teams to screen ULD pallets and containers given TSA concerns about the effectiveness of this screening method for those cargo configurations."

Lord states that TSA is continuing to test other screening technologies for ULDs. It appears that cargo may remain the most significant hole in passenger airline security.

On March 14, 2011, DiagNose announced that its trademarked system is now in use at the Amsterdam Schiphol Airport.

Third Addendum. Attachment 12-H to Regulation 573/2010 specifies size limits for consolidated consignments for explosives detection dog screening. The Attachment reads as follows:

“When screening cargo (including the cargo consignments of express parcels), the consignment shall be made available for screening in a non-containerised packing. The height of consolidated consignments shall not exceed 130 cm in height and 100 cm in depth [just over 4 ft. x 3 ft. 3 in.], unless a single piece of consignment is exceeding such dimensions. The nature of each consignment shall be taken into account before the handler shall direct the EDD to each item and evaluate the reaction of the EDD. The EDD shall be able to access and sample odour from all pieces of cargo.”

This means that if a consolidated consignment can be sniffed from both sides, it can be up to 200 cm wide (about 6½ ft.). For consignments higher than 130 cm, a ramp can be used for the dog to sniff the top of the consignment.

At least one company providing canine services for cargo screening has requested additional guidance from the European Commission. The EU responded that existing EU legislation on explosive detection dogs can only be reviewed after it has been in place for a reasonable period of time, but if other industry stakeholders believed the need to be more urgent, the Directorate General on Mobility and Transport will investigate "at an earlier stage." So far this has not happened.

Monday, November 1, 2010

Dispute Over $44 Million Contract Shows War Dogs Are Now Big Business

Contracts for providing the military with explosives, narcotics, and cadaver detection dogs, as well as patrol and sentry dogs, are becoming lucrative, with one contract reaching $44 million and another, just a month ago, $34 million, both going to the same contractor. With success comes competition, however, and with competition come disputes, with disputes come lawsuits, and lawsuits mean work for lawyers, unquestionably a good thing. (One must pay homage to one's guild from time to time.)

What the Army Expects from Dog Team Contractors

Contract working dogs (CWDs) are to be distinguished from military working dogs (MWDs) owned by the military. CWDs are provided by contractors as patrol dogs, narcotics detector dogs, and explosives detector dogs for work at Kandahar Airfield and throughout southern Afghanistan. There were at least 220 CWD dog teams in Afghanistan by July 2010, and the number is rapidly increasing. The picture shows a CWD team posted on an Army website.

To meet CWD contract requirements, dogs must be at least one year old and no more than seven years, and “similar in size to the German Sheppard,” though other breeds are permissible. Contractors are responsible for off-post kenneling, meaning that the U.S. Army Military Kennel Master does not provide kennel services. Each dog is to be deployed with a dog crate that can be used as a temporary kennel when necessary. Travel is paid for by the contractor, even on military planes. Naturally, such costs would be factored into contractors' bids.

Explosive detector dogs (EDDs) sent to Afghanistan must maintain a 95% proficiency rating in detecting a specified list of explosives:
  • Commercial dynamite (Gelatin and Ammonium Nitrate)
  • Military dynamite
  • Water gel (TOVEX)
  • TNT
  • Smokeless powder
  • C-4
  • Detonating cord
  • Potassium chlorate
  • Sodium chlorate

Canine Explosive Scent Kits (CESKs) are provided by the Army. If other explosives come into common usage by enemy forces in Afghanistan, contractors have at least 30 days to attain detection proficiency once the explosives are identified. Certification of dogs to the 95% proficiency level is made by an Army official (CJTF-101 MWD Program Manager). In a Federal Claims Court case described below, it was revealed that in 2008 the Army had only one person acting as a canine certification authority in Afghanistan.

Narcotics detector dogs are to maintain a 90% proficiency level (5% lower than EDDs), and must be able to detect concealed marijuana, hashish, heroin, and cocaine.

Patrol dogs must be able to stay at heel off-leash and attack only on the command, GET HIM. They must be trained to attack a person wearing an arm protector positioned at least 40 feet away, using a bite and hold method. Dogs can attack without command if a subdued person attacks the handler or the dog itself. On the command OUT, the dog must cease pursuit and return to the handler if commanded to HEEL. The description of services provided to potential contractors states that a patrol dog’s ability to continue an attack despite gunfire is critical. The dog must be able to detect the scent of a person hidden 50 meters upwind and follow the odor to the person’s location, and must also be able to detect and respond to a sound made by a person 100 feet downwind.

Four hours of proficiency training are to be provided by the contractor to each team each week. There is to be one handler per dog and the contractor must “replace lost capability within 15 days.” Contractors are to have contingent teams to replace a working team no longer deployed due to death, injury, vacation, decertification, or any other reason. Dog teams must be able to operate in “Hostile Fire or Imminent Danger Areas.” Canine personnel can be required to work “in excess of 10-hours each day, 6-7 days a week based on mission need,” including federal holidays.

The description of services specifies that CWDs need a diet significantly different from that of pet dogs and states that dogs should be fed “a food the same as or similar to, Hills Science Diet, NSN: 8710-01-415-6950.” That seems like a plug a marketing department should be able to do something with.

The contractor’s kennel master is an important part of any proposal submitted by a contract bidder. This person must have at least six years prior experience as a senior military or civilian working dog handler and two years experience as a kennel master. The kennel master participates in strategy sessions and must have “a valid and current security clearance SECRET, minimum, in order to attend specific Government meetings.” Handlers assigned to work with U.S. or Canadian Special Forces are also expected to have “SECRET security clearance for access to classified information from US, Britain, Canada, and Australia (4 EYES).” Handlers are to be certified and proficient in both long and short-barrel weapons. The contractor is required to supply uniforms that distinguish handlers as its employees. Performance Work Statement (June 5, 2008).

Contract Dispute Reaches Federal Court 

Disputes over a large contract began in October 2007 when the Army solicited bids for additional CWD services in Afghanistan (RFP # W91B4L-08-R-001). American K-9 was the incumbent contractor providing CWDs to the Army’s Special Forces in Afghanistan. EOD Technology, Inc. (EODT) also provided CWDs for use in Afghanistan, but not to the Special Forces. (According to a page on EODT’s website, the organization provides de-mining dogs for “local Afghan deminers.”) American K-9, EODT, and RONCO Consulting Corp. submitted bids. In December 2007, the Army awarded the contract to EODT. American K-9 and RONCO protested the award to the Government Accountability Office (Protests B-311008.1 and B-311008.2). RONCO has been mentioned in a prior blog (August 20, 2010).

Following the protests to the GAO, the Army terminated the contract with EODT and canceled the underlying solicitation, but then issued a new solicitation on February 10, 2008, this time specifying that in addition to the Army, the dogs would be supporting the NATO International Security Assistance Force Regional Command. The request sought two trained and certified patrol/narcotics detector dogs as well as explosives detector dogs (EDDs). The number of EDDs sought is redacted from the case where the dispute ended up, as are other facts perhaps deemed sensitive by the Army. Four offers were received to the February 2008 solicitation, The Army chose EODT (Contract # W91B4L-08-M-0232).

American K-9 again filed a protest with the GAO and again the Army stopped the contract, issuing a statement that it had become aware “of several problematic issues which threatened the viability of the solicitation.” Two days later, on March 28, 2008, the Army asked American K-9 to submit a proposal for a “sole-source, six-month bridge contract” to provide canine detection services. The Army entered a Justification and Approval for Other Than Full and Open Competition (FAR 6.302-2(a)(2) in the General Services Administration Federal Acquisition Regulations). The Army said that American K-9, as the incumbent contractor, was the only CWD supplier in a position to satisfy the immediate need “with no mobilization lead-time required.”

American K-9 was then awarded the contract on March 31, 2008, and EODT took its turning filing a protest with the GAO (GAO Protest B-311349.2). EODT’s protest triggered an automatic stay of performance on the contract, and on April 6, 2008, the Army issued a stop-work order to American K-9, but provided American K-9 with a three-month, sole-source “interim” contract. On April 8, the Army decided that the six-month, sole-source contract (W91B4L-08-M-0257) could proceed despite EODT’s GAO protest, saying that there were “urgent and compelling circumstances that significantly affect the interests of the United States and its Allies, and will not permit waiting for the GAO’s decision on the protest.”

EODT, seeing the GAO protest wasn’t working, filed suit in the Federal Claims Court.

The court began to sift through the evidence, including the history in the preceding paragraphs. It noted that after a certain redacted date, NATO would not be using CWDs in Afghanistan. Only the U.S. would be using these dogs. It was not explained why other forces would not be using CWDs, though the Federal Claims Court found that the need for CWD teams in Afghanistan was rising.

The legal issue boiled down to the Army’s ability to override a stay by the GAO, which required that the Army establish that the best interests of the United States outweighed Congressional policy. The Army argued that the CWD services could not lapse even for one day, noting that an EDD at one forward operating base had found a redacted but obviously high number of improvised explosive devices (IEDs) in the previous year. EODT, in protesting the award of the bridge contract to American K-9, noted that it also had CWD teams in Afghanistan and could provide the teams required by the Army Special Forces without any delay.

The Army, apparently forgetting its choice of EODT in December 2007, said that it only wanted to use one contractor and that bringing EODT or another contractor in while it was using American K-9 would “negatively impact mission effectiveness.” Further, it said that use of two or more contractors would “disrupt the integrity of the teams currently working together” and “elicit disruption and harmful competition.” Apparently, the handlers of the dogs, if not the dogs, would inevitably get in fights just as the corporations for which they worked were doing.

The Army also said that it had only one certification authority for dog teams in Afghanistan, a Sergeant First Class (name redacted, of course). The Army said that this soldier could only certify a redacted number of teams a week. It was not explained why this soldier could not certify teams from different providers as efficiently as teams from a single provider given that the certification standards are those of the Army, not of the providers.

The claims court decided not to override the bridge contract but said that it could not be expanded from its six-month bridge status to provide American K-9 with a longer-term contract without allowing other bidders to make offers. The Army would have to open up the bidding again. EOD Technology, Inc. v. U.S., 82 Fed.Cl. 12 (Ct.Fed.Cls. May 15, 2008)

Dispute Moves to GAO Turf 

On June 13, 2008, a new solicitation was issued (RFP # W91B4L-08-R-0025) to obtain CWD services throughout Afghanistan for one year, with four option years. The Army anticipated needing 50 teams initially (Responses to Contractor Questions, June 24, 2008). EODT was awarded the contract in this round and American K-9 protested to the GAO on August 15 (B-400464, supplemented September 11), attacking the Army’s evaluation of the past performance of the bidders, one of the factors to be considered in awarding the contract. The Army agreed to reevaluate past performance. After receiving letters from the bidders, the Army affirmed its award to EODT.

American K-9, not to be deterred, protested again, saying this time that EODT had not satisfied the experience qualifications required by the solicitation. Among other things, American K-9 argued that the kennel master/project manager that EODT planned to use did not satisfy Army standards. The Army concluded that there was a deficiency but gave EODT the opportunity to update its proposal. The Army also asked American K-9 to update its proposal. EODT provided an updated resume for its proposed kennel master. Meanwhile, the Army also questioned American K-9’s proposed kennel master, saying it had “reason to infer” that the kennel master referred to in American K-9’s proposal was not the one that was going to do the work.

Whether the Army was questioning the qualifications of the kennel masters on its own, or whether some corporate espionage was going on between the bidders, is not clear from the documents publicly available. In any case, American K-9 protested to the GAO that the Army was conducting “results oriented” discussions because it wanted to give the contract to EODT regardless of the merits. The GAO dismissed this protest as premature. EODT had given a price of $38,350,935 for the work; American K-9’s bid is redacted, though said to be “approximately equal” to EODT’s. The Army again preferred EODT.

American K-9 protested yet again, persuading the GAO that the Army’s discussions had not always been meaningful. The Army had not talked to American K-9 about some of the deficiencies it found in the company’s offer. The GAO sustained American K-9’s protest of an award to EODT on May 5, 2009 (Protest B-400464.6) and recommended that the Army request final proposal revisions and make a new source selection.

American K-9 Wins (At Least for Now)

A U.S. contracting website lists the final award given to American K-9 as being for $44,775,558.79 (Award # W91B4N-10-C-5001 to RFP # W91B4L-09-R-0025), though the site contains a pdf of a contract signed by Mark Mahler, President of American K-9 listing the total award amount as $25,830,292.74 (Award # W91B4N-10-C-5001, dated December 11, 2009). The $44 million seems to be the amount most commonly used in press reports. A separate Defense Department website refers to the same award (at least by number) and states its amount as $15 million. The discrepancies may in part reflect guaranteed and extension amounts. Two news releases on American K-9’s website do not list the amount of the award but state that it was the largest CWD contract ever issued.

American K-9’s ability to land government contracts for CWDs continued in 2010. In August, the company was awarded a contract to provide cadaver/human remains detection dog teams to Iraq. In September, it got a similar contract for Afghanistan. Also in September, a contract from the Navy was awarded to American K-9 at $18,426,926, which with contract extensions could reach $34,659,851.

The price for dogs may seem high, but working in danger areas is costly in many ways. EOD Technology had 34 deaths in U.S. government contract work between September 1, 2001, and June 30, 2010. EOD Technology was the subject of some negative attention in a Senate Armed Services Committee Report issued September 28, 2010. Whether this had any effect on EODT's efforts to get canine contracts is unclear.

Additional sources: Abby Brown, Discussions Were Improper When ‘Discussion’ Questions to One Offeror Seemed ‘Contrived’ (June 8, 2009); Kara M. Sacilotto, Is the Game Worth the Candle? The Fate of the CICA Override, The Procurement Lawyer, 45(1), 3 (Fall 2009) (discussing the Federal Claims Court decision and other cases). To see what roadside bomb detection actually looks like, see the excellent video on the website of DiagNose.

Additional Notes. The Associated Press reported on March 17, 2011, that the State Department ended a $274 million agreement with EOD Technology to provide protection for the U.S. Embassy in Kabul because the company was not going to be able to begin performing its functions by May 1. EOD had been selected to replace ArmorGroup North America, whose guards had been caught drinking excessively and engaging in lewd behavior at barracks a few miles from the embassy. ArmorGroup's contract is to be extended until September 2011. The Senate Armed Service Committee has been investigating the role and oversight of private security contracts in Afghanistan.

On December 14, 2010, the Defense Department announced that K2 Solutions, Inc., of Southern Pines, North Carolina, was being awarded a $24 million firm-fixed-price contract for IED detector dog procurement, training, sustainment and conditioning, and handler training. The contract is also to include kenneling and basic care of the dogs. The Marine Corp will designated "team integration training venues," perhaps at Twentynine Palms, near Indio, California. Team integration occurs prior to deployment.

On October 4, 2011, K2 Solutions won an award for providing the Marine Corps with approximately 359 improvised explosives device detector dogs, along with training of 647 Marine Corps personnel annually. The contract (M67854-11-C-3015) award dollar amount was listed as $34 million, but the contract includes options, which if exercised will bring its cumulative value to $91 million. Training will be performed in Southern Pines, North Carolina, Twentynine Palms, California, and overseas. If options are exercised, work will be completed March 2014.  U.S.-trained mine detection dogs are working in Jordan on the Syrian border (AUD-CG-13-42).