Mucking up Causation with Identical Twins

Orlando Nembhard was charged with the February 12th shooting death of Sir Xavier Brooks, 19, outside a Phoenix nightclub called Leonardo’s Da Vincci Code. There were eyewitnesses claiming to have seen Orlando pull the gun and they had a description of the shooter which matched Orlando. Then the prosecution’s case went to hell when it was revealed that Orlando Nembhard has an identical twin brother, Brandon Nembhard, who was also at the nightclub the night of the shooting. Now with conflicting stories and suspicious behavior from Brandon, prosecutors are questioning whether they’ve charged the right man.

Barbara Jones, the victim’s grandmother, recently said in an interview, “They need to put them in a room and let them battle it out,” calling to light a pariah in evidence law — probabilistic evidence. Courts have never been comfortable with using statistics to convict defendants, permitting its use only in some employment law cases and tort cases. But with the stakes so high in a criminal murder case, the burden of proof of “beyond a reasonable doubt” cannot be quantified.

When studying the law of torts, students learn the case of Summers v. Tice: two hunters simultaneously shoot and injure a third hunter, though only one bullet causes the man’s injury. Because the guns are identical it is impossible to tell which gun fired the  round. Common law holds that the hunters will be held jointly and severally liable, each of them responsible for the entire amount of the man’s injuries. The idea is that the burden shifts to the negligent parties to prove that they were not the cause of the injury. The problem with the application of this case is 1) this is tort law (civil, not criminal) and 2) there was only one shooter. In the case in Phoenix, only one of the brothers allegedly shot the victim.

Judge Richard Posner has long been a proponent of the use of statistics in legal fact finding. In Posner’s Economic Analysis of Law (2007) he writes:

In the typical civil trial, there is no basis for supposing that Type I errors (false positives, such as convicting an innocent person, or in the civil context, erroneously finding the defendant liable) on average impose higher costs than Type II errors (false negatives, such as an erroneous acquittal or the denial of a meritorious claim). So it is enough in the usual civil case to justify a verdict for the plaintiff that the probability that his claim is meritorious exceeds, however slightly, the probability that it is not.

Why should a tie go to the defendant rather than to the plaintiff? The principle of diminishing marginal utility of income implies that the loss to the deserving plaintiff who loses is probably slightly smaller on average than the loss to the deserving defendant who loses…. In addition an erroneous judgment for the plaintiff imposes a cost that is avoided when the error goes the other way – the cost of collecting the judgment….

Type I errors in criminal cases involve additional cost because the cost of imprisonment is high, that costs is of course avoided when a guilty person is acquitted, though such an acquittal will reduce deterrence by reducing the probability of punishment for the crime. But the asymmetric effect of the cost of imprisonment on convictions and acquittals means that it probably takes several erroneous acquittals to impose a social cost equal to that of an erroneous conviction. This is one economic rationale for requiring proof beyond a reasonable doubt in a criminal as distinct from a civil case and another is the inherent advantage that the prosecution enjoys in a criminal case, compared to a private civil plaintiff…. In an inquisitorial system, where the search is conducted by a presumably disinterested judge, the need for a heavier burden of proof in a criminal than in a civil system is attenuated.

But an economic approach to causation has never been popular as few juries can sit comfortably with the idea that they have knowingly convicted an innocent person for the sake of easing the burden of proof. Because statistical probabilities will inevitably scoop up innocent people, their use in the criminal context is utterly impermissible. Perhaps it’s Posner’s economic theory of law that have kept him off the Supreme Court (or maybe it was that article characterized as Posner’s argument for selling children; why not verify that claim yourself here).

Via the NY Times (thanks Matt!).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s