CHAMPAIGN, Ill. - A disagreement among state courts on the subject of drunk-driving homicide can be resolved by requiring the prosecution to prove in these cases not that the driver's intoxication caused the fatal accident, but merely that it contributed to the causal mechanism behind the accident, says a forthcoming paper by a University of Illinois expert on criminal law.
Although philosophers and legal scholars have long recognized that conduct can contribute to a result without strictly causing it, this phenomenon of "causal overdetermination" has been almost entirely overlooked by criminal law scholars, says Eric A. Johnson, a professor of law at Illinois.
Even scholars in areas other than criminal law have failed to spot the connection between causal overdetermination and the kind of "wrongful aspect" causation at work in drunk-driving homicide cases, Johnson said.
"Nobody has put the two ideas together," he said. "With this paper, I want to make judges, prosecutors and defense lawyers notice the connection between the two ideas."
According to Johnson, in cases involving drunk-driving homicide, courts around the country are divided about equally on the question of whether the law requires a "causal nexus" between the defendant's intoxication - the wrongful aspect of the conduct - and the fatal accident.
"There's this division in the courts, and they don't even really appear to know that they're divided," he said.
It's a question that gets litigated frequently in state courts, each of which applies its own distinctive state criminal law on the subject, Johnson said.
"The courts have come to different conclusions about whether or not the government is required to prove that the intoxication actually caused the accident," he said. "So there's this stark division between the two positions, and I was interested in exploring the question, 'Who's right?' It seemed as though one side or the other had to be right."
When Johnson investigated this divide among the courts, though, he noticed that the courts on both sides seemed uncomfortable with the positions they had staked out.
"It seemed as though the courts that had decided not to require a causal nexus, well, they were nevertheless uncomfortable imposing liability in cases where the defendant's intoxication had nothing to do with the accident," he said. "On the other side of the spectrum, the courts that said a causal nexus is required, they didn't really seem to be demanding what the law appears to require, which is proof beyond a reasonable doubt that the accident would not have occurred but for intoxication."
Johnson's own intuitions about the problem also were divided. On one hand, the position that the intoxication's causal role shouldn't matter at all in these cases struck him as "intuitively wrong." But he also recognized that in many cases the government would find it next to impossible to prove that the accident would not have occurred "but for" the driver's intoxication.
"In some states, the prosecution needs to satisfy the 'but-for' standard - that is, it has to prove beyond a reasonable doubt that the accident would not have occurred 'but for' the driver's intoxication," he said. "Well, if somebody runs over a pedestrian who's walking on the side of a dark road with hairpin curves, it's very hard to say beyond a reasonable doubt, 'A sober driver would not have run over this person.' And in fact it seems as though the courts in practice aren't really willing to hold the government to that tough burden of proof. It seems as though the courts in those jurisdictions that require a causal nexus are hedging."
According to Johnson, the solution to this problem lies in the idea of "causal overdetermination." In cases of causal overdetermination, a person's actions contribute to the causal mechanism underlying the harm without necessarily playing a decisive role.
If causal overdetermination were at work in the drunk-driving homicide cases, it would explain why the courts seemed uncomfortable with the idea of requiring a "but for" causal connection, Johnson said.
"Causal overdetermination basically occurs when someone contributes to an already dangerous situation," he said. "Let's say that someone's gravely ill and you want to kill them. You steal their medication and a few hours later they die. In this case, we can't say for sure that the person wouldn't have died anyway. After all, they were gravely ill. But we can say that taking away their medication increased the likelihood that the underlying illness would kill them."
The theft of the medication in this case is an "overdetermining cause," Johnson says. So the court wouldn't require the government to prove that the victim would not have died "but for" the theft of the medication.
"The person might have died anyway," he said. "So the government can't prove that they would not have died 'but for' the defendant's actions. What the government can prove is that the defendant contributed incrementally to the causal mechanism behind their death. And that's enough in a case like this."
Drunk-driving homicide cases also involve causal overdetermination, Johnson says.
"In drunk-driving homicide cases the causal mechanism behind the accident usually is the interaction of a roadway hazard - a pedestrian on a blind curve, for example - with inherent limitations on the abilities of human drivers to perceive and react to hazards," he said. "That combination is what causes the accident. But the driver's intoxication contributes to the danger posed by that combination in exactly the same way that taking away a sick person's medication contributes to the danger posed by their underlying illness."
If drunk-driving homicide cases are causal overdetermination cases, then courts that require a "but for" causal connection between the intoxication and the accident are requiring too much. By the same token, though, courts that require no causal connection at all are requiring too little.
"The answer really lies in the middle," he said. "What's required is only that the driver's intoxication contribute to the mechanism behind the crash. So if the accident occurred under circumstances where the driver's intoxication made matters worse, where it's even possible that a sober driver would have been able to avoid the accident, then the crime's causation element is satisfied."
According to Johnson, this is the kind of issue that lawyers deal with all the time.
"It's probably one of the most frequently litigated questions in drunk-driving homicides," Johnson said.
Although it sounds purely academic on the surface, "it's easy to see how it could be reduced to a jury instruction in a criminal case," he said.
"You could do that very easily - you'd tell the jury that the only question is whether the accident occurred under circumstances where the driver's intoxication 'might' have made a difference. The government doesn't have to prove that his intoxication 'would' have made a difference."
As a former prosecutor who worked in the offices of the attorneys general of New York and of Alaska before becoming a criminal law scholar, Johnson says he would like to see the law go in that direction.
"The government shouldn't have to prove that the accident definitely wouldn't have occurred; it's enough that the driver was drunk, which increased the risk, and the accident occurred," he said.
The paper will appear in the Connecticut Law Review.