top of page
  • rdwall2221

Kansas v. Glover - PART I

How the Misuse of Probability Evidence Leads to Bad Decisions and Absurd Results


            A Douglas County Kansas Sheriff's Deputy stopped a vehicle driven by Charles Glover, Jr. after running the vehicles' plate and finding out that Glover, the registered owner, had a revoked license.  Apparently, this was not Glover's first time driving with a revoked license.  He was charged with being a habitual offender.  Glover filed a motion to have the evidence against him excluded from trial on the grounds that the stop violated the Fourth Amendment, which prohibits unreasonable searches and seizures.  The trial court granted the motion, and the Kansas Supreme Court affirmed. 


            The United States Supreme Court granted review and in an 8 - 1 decision "reversed" the Kansas Supreme Court.[1]  The majority held  that the stop was lawful under Terry v. Ohio, 392 U.S. 1 (1968), which requires that an officer have a reasonable suspicion of criminal activity in order to make even a brief investigatory seizure.  According to the majority,  "commonsense" supported an inference that the driver of any vehicle is likely to  be its registered owner.  Thus, the Deputy who stopped Glover reasonably suspected that the driver of the vehicle was Glover himself and that he was driving on a revoked license. 


            In dissent, Justice Sotomayor argued that the inference relied upon by the majority to uphold the stop was merely a generalized inference about how likely it is for persons with revoked licenses to drive a vehicle registered to them.   According to Sotomayor, that inference, even if well founded, did not satisfy the requirement  that the officer's suspicion of criminal activity be "individualized" to the person being stopped in order for the stop to be reasonable.   She took issue with the idea that where such a connection is lacking, an officer can substitute "commonsense" for particularized facts giving rise to a reasonable suspicion the person being stopped is committing a crime. 


            Sotomayor touched on, but did not fully identify the fundamental problem with Kansas v. Glover.  Under Terry v. Ohio, an officer must have a reasonable suspicion that the person being stopped is engaged in criminal activity.  An officer is also permitted to rely on "commonsense" when assessing whether the known facts are sufficient to justify such a suspicion.  But, what happens when "commonsense" is demonstrably wrong?  Can an investigatory stop be justified by "commonsense" when it is shown that a commonly held belief about the relationship between known and unknown fact conflicts with reality?  That is what happens when "commonsense" is based on misconceptions about the nature of probabilities their and their legitimate use  to describe and predict real world events. 


            One example of how the application of probabilities to real world events is often misunderstood is asking someone to predict the result of tossing a coin.  The probability that a single toss of a coin will result in either a heads-up or tails-up position is 50%, since there are only two possible outcomes.  Thus, if a coin is tossed repeatedly, the probabilities are that half of the time the coin will land heads-up and half of the time it will land tails-up.  But that probability does not predict what will happen on any particular toss of the coin. 


            If a coin is tossed a number of times and each time lands heads-up, most people using "commonsense" will believe the probability of the coin landing tails-up increases with each toss.  But, that is a misconception.  In fact, the probability that the coin will land either heads-up or tails-up on any particular toss is always the same - 50%. 


            To say that there is a 50% chance the coin will land heads-up or tails-up is true only when applied to a number of tosses.  To know whether the coin actually landed head-up or tails-up on any particular toss it is necessary to observe the coin after it comes to rest.  That is to say, we need to know something more than just the probability that it will land one way or the other.  This is true no matter what the probability is, even if it is close to 100%.  Even a high probability only predicts what will happen if the same event occurs a number of times.  It does not describe what has already happened.  Each toss of the coin is a separate event that results in either a heads or tails, but probabilities tell us nothing about the results of any one toss. 


            Applying these principles to the facts of Kansas v. Glover it is easy to see where the majority went astray.  Even accepting that there is some unknown probability any  vehicle on the road is being driven by its registered owner, that probability only predicts how often the driver will turn out to be the registered owner if multiple stops of different vehicles are made.  It does not predict whether the stop of any one vehicle will result in the driver being identified as the registered owner or someone else.  As to any particular stop, it is necessary to have some additional information to support a reasonable inference that the driver is actually the registered owner or even that there is a "fair probability" the driver is the registered owner.  Just as in the example of tossing a coin, the fact that a certain percentage of vehicles on the road are driven by their registered owner does not mean that the driver of any particular vehicle is by that same percentage likely to be the registered owner.  The driver is either the registered owner or is not the registered owner.  The "commonsense" notion that, because registered owners tend to drive their own cars, the driver of any particular car is likely to be its registered owner is simply wrong. 


            The misuse of probabilities as in Kansas v. Glover can lead to absurd results.  For example, if it could be shown statistically that in a particular neighborhood there is a "fair probability" any male between the ages of 15 and 25 if stopped and searched by the police would be found to be unlawfully in possession of contraband of some kind, the police could stop and search every male in that neighborhood who appeared to be of that age at any time without violating the Fourth Amendment.  That would amount to naked profiling, which has never been allowed by the Supreme Court, not even under Terry v. Ohio.  Yet, the justification for stopping all males of a certain age in a certain area is no different from the justification used in Kansas v. Glover to allow officers to stop any person driving a vehicle whose registered owner has a revoked license.  The person is being stopped not because they have done anything to arouse suspicion, but simply because they have been identified as belonging to a group or class of persons deemed likely to be engaged in criminal activity. 


            Even in the context of vehicle stops, Kansas v. Glover is likely to lead to clear violations of the Fourth Amendment.  Consider the following scenario.  An officer stops a vehicle because the registered owner has a revoked license.  It turns out, however, that the registered owner is not driving.  The driver is allowed to continue on because there is no evidence that any crime is being committed.  The vehicle travels a short distance and is stopped again for the same reason, but this time by a different officer.  Under Kansas v. Glover, both officers have a reasonable suspicion the driver is engaged in criminal activity even though law enforcement is now aware that there is no crime being committed.  The unfortunate driver is subject to being stopped repeatedly either by different officers or even by the same officer, since the facts justifying the stop never change: the registered owner of the vehicle has a revoked license and the vehicle is being driven by someone.  The repeated and potentially unending seizure of the vehicle and its driver by law enforcement simply because the registered owner of the vehicle has a revoked license is clearly unreasonable.  Nevertheless, it is entirely permissible under Kansas v. Glover.   



[1] It is a misnomer to say that a ruling by the United States Supreme Court contrary to the decision of a state court in the same case constitutes a "reversal" of that decision.  The Supreme Court is a federal court that exercises  judicial power conferred by the United States Constitution, whereas all state courts derive their judicial power from their respective state constitutions.  States are sovereign political entities.  Thus, state courts, unlike the lower federal courts (i.e., federal district courts and circuit courts of appeal) are not "inferior" to the Supreme Court.  The Supreme Court has no authority to alter a state court decision or otherwise directly order a state court to take any particular action in a case.  It can issue orders to the state itself requiring the state to do something or refrain from doing something, such as giving the accused a new trial or releasing the accused from custody, and those orders can be enforced by the executive action. But, the Supreme Court has no direct control over the conduct of any state court in the same way it can control the conduct of the lower federal courts by remanding a case with specific directions as to how the lower court is to proceed.   

1 view0 comments

Recent Posts

See All


How the Supreme Court's Recognition of a Religious Exemption to Antidiscrimination Laws Could be a Small Step Toward Establishing an American Theocracy Masterpiece Cakeshop, Ltd. v. Colorado Civil Rig


bottom of page