Because It’s Not Always Junk, Part 1
May 3, 2016 (Mimesis Law) — The supposed relative weakness of circumstantial evidence is a well-worn television trope. Often, during a forensic science driven drama like CSI, someone will denigrate circumstantial evidence, instead pushing the team for “real scientific evidence.”
While it’s hard to say, legal television shows may have slowly changed the expectations of jurors. Many may now expect that the law requires something higher than circumstantial evidence. So, in cases where circumstantial evidence is to play a key role, the lawyers usually educate the juror during jury selection about it, and the judge may give a preliminary instruction about it.
In Ohio, the preliminary instruction looks something like this:
The classic example of direct and circumstantial evidence is if a jury must decide whether a hypothetical boy named Johnny ate a piece of cherry pie. If a person walked into the kitchen and saw Johnny eating the pie, that would be direct evidence that he ate the pie. If a person walked in and saw Johnny with an empty pie plate in his hand, cherry pie around his mouth, and a smile on his face, that would be circumstantial evidence that he ate the pie.
This type of instruction serves a twofold purpose. First, it reminds jurors that you can make strong, certain conclusions about the past, based solely on circumstantial evidence. Second, it reminds jurors that we frequently rely on circumstantial evidence. If you’ve had kids, then you know frequently things get broken, siblings get punched, and messes get made far out of the sight of parents. So, parents frequently have to piece to together the truth of what happened using circumstantial evidence.
Moreover, the popularity of shows like CSI and informative murder porn have given rise to the legal conventional wisdom that jurors not only want, but also need, forensic evidence to return a guilty verdict. And if, as a lawyer, you’ve failed to receive a continuing legal education solicitation to teach you all about this so-called CSI effect, then you’re probably doing probate law.
The most recent, high profile murder case that was largely proved by circumstantial evidence was the Aaron Hernandez case. The facts were essentially as follows.
Hernandez texts some friends to drive to his place. They arrive and Hernandez is recorded by his own surveillance system holding a gun. At 1:12 AM all three men leave in a rented silver Nissan Altima. About an hour later they get gas, and Hernandez buys a pack of bubblegum. At 2:32 they arrive at the victim’s house, texting him, “We’re here.”
At 3:32, the car is seen on a security camera driving to a secluded area of an industrial park. Sometime within the next four minutes the victim is shot dead five times with a .45 caliber handgun. At 3:28, the car is seen leaving the industrial park. A few minutes later, Hernandez’s surveillance system records the three men arriving and the presence of the handgun. The surveillance system is turned off for six hours.
In addition to these facts of essentially a circumstantial case, two pieces of forensic evidence helped the prosecutor’s case. First, upon returning the rented Nissan Altima, an employee of the company found a piece of chewed bubblegum stuck to a .45 caliber shell casing inside the vehicle. As it turned out, the DNA recovered from the gum was a link to Hernandez. Yet more evidence that circumstantially demonstrated that Hernandez was there at the murder.
Second, the prosecutors were able to create a probable link to shoe prints recovered at the scene to shoes in Hernandez’s house that were also documented during the execution of the search warrant. In addition, these shoes, the Nike Air Jordan XI Retro Low, had only been on the market for three weeks. All of this evidence is probably what convinced his defense lawyers to acknowledge his presence during the murder, but point to another triggerman. The sheer number of what would otherwise be coincidences made it appear unlikely that Hernandez was not there.
In his book, Fooled by Randomness, Nicolas Taleb explains his disgust over watching a part of the OJ trial. The lawyer (probably Barry Scheck) argued that there were probably at least four people in Los Angeles that carried the same DNA profile as Simpson. According to Taleb, that this was a specious argument because it ignored the effect of joint and conditional probability. Joint probability is calculating the chances of rolling all sixes on 20 die all thrown together. The upshot is that the probability of joint events are lower than the individual events.
Unconditioned on anything else, it seems unbelievable that an NFL superstar would murder a regular guy in New England. Hernandez had fame, fortune, and a young family. We would assume that no rational person would flush that life down the toilet to kill a guy over what would seem to be a minor spat. But when you roughly consider the probability that someone else did it, then it’s hard to imagine anyone but he or his crew killed the victim.
You have the relatively small number of people who had the same shoes, you have the vanishingly small number of people with Hernandez’s DNA profile, the number of people that chew bubblegum, the probability of being murdered by someone known to the victim, and the number of probable people around the industrial park around the time of the killing. When you consider the other facts, such as Hernandez’s access to a handgun, the shell casing in the rental car, the disabling of the security system, the fiancé disposing of trash at an odd time and manner, and Hernandez’s penchant for being around murders, it is hard to imagine he was not involved that night.
In this sort of joint probability situation forensic evidence plays an important but not overly weighty role. For example, Nike identified three shoes that possibly matched the shoe print. So, at the very best, the shoes Hernandez owned only had a one in three chance of matching. And even if the DNA match was in the tens of millions, it at least was highly suggestive of Hernandez. Neither of those experts could testify that Hernandez was either there or killed the victim.
But all the evidence considered together makes it more probable that Hernandez was there. So, when the forensic evidence can add to the confidence in the defendant’s guilt, such as in a joint probability context, it can be quite helpful. On the other hand, it’s entirely conceivable that if the only evidence was the shoe print, Hernandez would not have been charged, never mind convicted.