Notwithstanding the briefs of Georgia-Pacific and numerous amici, appellant Bostic, appellee Georgia-Pacific and (seemingly) most of the justices appeared by the end of oral argument on Monday to reach at least partial agreement on the big issue. Specifically, that somehow or another it ought not be the law when multiple defendants create conditions each independently capable of causing plaintiff's injury that none of them be held liable merely because it would be impossible for plaintiff to prove that any one of them was the "but for" cause of her injury. It wasn't exactly a Kumbaya moment but it was close. And the refrain was of Ford v. Boomer and of Merrell Dow v. Havner; and it made us very worried about the future of one of the most important defense victories ever - Borg-Warner v. Flores.

Until the last third of the proceedings a sensible guess as to which way things might go was hard to come by. Much of the briefing and a fair bit of the argument was hopelessly confused due to the tendency of the various parties to attribute decidedly different meanings to identical causal language. It's hardly the fault of the attorneys. Most causal distinctions made in legal opinions still consist, however solemnly invoked, of little more than "moonshine and vapor" (see "Proximate Cause in California" by William L. Prosser, 1950 - a really fun paper btw). Maybe the court in whatever opinion it authors will adopt a modern lexicon of causality with readily translatable and transportable ideas like "necessary cause" and "sufficient cause". We can only hope. But in any event Bostic's counsel cleverly suggested a compromise, the Davidson rule with a little Havner on top, and suddenly, alarmingly, everyone seemed in agreement.

The Davidson rule, like that of Ford v. Boomer, settles the question of where to draw the line for the outer limit of liability for asbestos-related disease by requiring that plaintiff's exposure from any potentially liable defendant's product (or premises) be in and of itself sufficient to have caused the disease. In Texas (see Havner) that would mean given the current state of uncertainty about the causal mechanisms underlying asbestos exposure and mesothelioma that plaintiff would have to show whatever exposure she attributes to a particular defendant doubled her risk of developing mesothelioma. Any defendant whose contribution could be shown to satisfy the risk doubling requirement would have to face a jury; and any defendant whose contribution did not double the risk could have its summary judgment and go home. Seems fair; what's not to like?

Plenty. Remember that causation is necessary but not sufficient for the determination of substantial factor causation. It is not enough for a legal causation showing to prove that grabbing a stumbling passenger's arm dislodged a plain brown package containing fireworks that fell to the ground, exploded and caused some distance away a scale to bonk Mrs. Palsgraf on the head.  The risk posed by the act of grabbing the stumbling passenger's arm must have been of such a degree as to generate a duty for the ordinarily prudent railroad employee to have done something different. Unless we missed it, not once during oral argument did anyone utter "legal cause" or "de minimis risk". Why? We suspect it's not because everyone forgot that the substantial factor test of legal causation requires causation plus a non-de minimis risk but rather because everyone assumes a 100% risk increase must be substantial - an all too common consequence of our cognitive blind spot for percentages.

Something that increases the risk of mesothelioma by 100% hardly sounds de minimis. But think about it this way, your odds, absent asbestos exposure, of developing mesothelioma are 1:1,000,000 (one in a million). Doubling the risk increases the odds to 1:500,000. Those are your odds of being struck by lightning this year according to National Weather Service estimates. And that risk is tiny compared to your risk of having say the wind drop a tree limb on your head.

So should a 1:500,000 risk be big enough for the imposition of a duty? That's for the courts to decide by way of public policy analysis but they ought to go into the exercise understanding what it would mean to say that someone who creates a 1:500,000 risk of death can be subject to liability and even punitive damages. Consider these examples of activities that increase the risk of death by 1:500,000: two days of snow skiing, going horseback riding 4 times in or eating one peanut butter and jelly sandwich per year (aflatoxin). If that doesn't make the point consider this: 1:500,000 odds are less than those of flipping a fair coin 18 times and having it come up heads every time.

So here's to hoping the court keeps in mind that whether or not a duty should be imposed in a given case is gauged by whether the defendant has created a substantial risk and not whether a risk has been altered substantially. Otherwise we'd wind up with the bizarre conclusion that doubling an infinitesimal risk creates a substantial one.

P.S. If you're interested in how the counterfactual or "but for" view of causation survives a multiple sufficient causes problem like that seen in Bostic see chapter 10 of Judea Pearl's book "Causality".