David Aylor On The Role of Data in Florida v Harris

A bombshell story broke in Propublica in early November when award-winning journalist Ryan Gabrielson wrote about his attempt to fact-check a random sampling of 83 majority decisions that the Supreme Court of the United States (SCOTUS) handed down between 2011 and 2015. Much to his shock and horror, Gabrielson discovered that the Justices cited bad data, incomplete information, and false assumptions as evidence in decisions on a number of pivotal cases, including the highly-criticized Shelby County v Holder case, which repealed a number of the provisions of the Voting Rights Act of 1965.

Gabrielson found a glaring error in a unanimous decision Justice Elena Kagan wrote for Florida v Harris, a case that revolved around illegal search and seizure as outlined in the fourth amendment. The case was argued in 2012 before the passing of Judge Scalia and decided in 2013.

Clayton Harris, a known and convicted drug manufacturer and dealer, was pulled over at a traffic stop in Florida. During the interaction between the officer and Mr. Harris, a police dog named Aldo alerted his handler that he had detected something on the driver’s side of the car. When the officers proceeded to search the car, they found no drugs, but an open container of alcohol. At his arraignment, Harris’ lawyer moved to suppress all the evidence found during the search that was initiated by the dog’s alert, saying that the dog’s training was insufficient for its call alone to qualify as probable cause. Long story short, the case got appealed again and again until it reached the Supreme Court.

The Court decided to weigh in only on one very narrow aspect of this case, specifically whether the rate of false positives in drug-sniffing dogs. In the realm of statistics, false positives, or Type 1 errors, occur when someone chooses to believe something incorrect – for comparison, a Type 2 error occurs when someone chooses to disbelieve something that turns out correct. Over the course of oral arguments, the Justices sought to determine the frequency with which police dogs falsely alerted handlers. Some studies had demonstrated that 4 out of 5 times a police dog alerts its handler, there is nothing to find – that is, the dogs committed a Type 1 error. Lawyers for the state of Florida defending the dogs’ judgement called on Aldo’s certifications, performance record, and past successes in passing early trainings with his handler.

In 2013, Justice Elena Kagan wrote the unanimous decision that a trained police dog’s notification did indeed qualify as probable cause for an officer to search the premises. Kagan called on the certification and rating system of the Police Dog Academy as evidence that the dogs had sufficient training to detect the presence of drugs.

Gabrielson’s investigation, though, uncovered a sore spot in Kagan’s decision. The certification that the police dogs receive has nothing to do with Type 1 errors, or the chances that the dogs alert their handlers to nothing. Rather, their certification tests for the chances that a dog commits a Type 2 error, of that the dog fails to alert its handler to the presence of drugs. Though there are various versions of assessments that dogs have to pass, a dog will be disqualified only if it commits a Type 2 error, and not a Type 1 error.

The legal implications of the court’s decision are widespread, as it establishes probable cause for a search where it appears there isn’t any cause whatsoever 80% of the time. As of yet, there is no mechanism to challenge a Supreme Court ruling except to bring up a similar case and hope that the details and new justice will rule a new way.

About the author: David Aylor is a Charleston-based lawyer with over ten years of experience in the field of law. His practice, Aylor Law, has been lauded for its exploits, and strives to bring an empathetic experience to its clients. David specializes in criminal defense, auto accidents, personal injury, and drug and alcohol offenses. Visit davidaylor.net for his blog on legal proceedings.