'Partridge' family reunion: California v. Allstate
I saw this opinion, California v. Allstate, a couple days ago but other things interfered with blogging about it until now. This is a case about coverage for environmental pollution under the "sudden and unexpected" language found in old liability policies, and that is an interesting thing in itself, but what chiefly interested me about this case was the extended discussion of the Partridge case, an influential case that, in days gone by, was in part responsible for the creation of the modern anti-concurrent cause language found in first party policies.
Partridge is not a first party case, of course, but its reasoning resonated beyond the third party liability context, as I discussed in this Appleman's Critical Issues article on anti-concurrent causation analysis from a couple years ago. (The Partridge discussion starts on page 61). Partridge was cited in a 1982 Ninth Circuit case, Safeco v. Guyton, that in essence overrode California's efficient proximate cause analysis in property insurance cases and endorsed a Partridge approach. (I know that to folks who may not be steeped in this particular aspect of coverage law, this may all sound like secret code at best or Klingon at worst, and maybe it is, but for good or ill insurance coverage has its own specialized language that is an initial barrier to entering the conversation or even following what it is being said. This is unfortunate in some respects, but it is a fact of life).
What a Partridge approach means is this: whenever two causes of liability exist, as long as one of them is covered, the liability will be covered, even if the covered cause is a subsidiary part of the causation chain and not the efficient or proximate cause. It just has to be a proximate cause. As the California Supreme Court later found in Garvey v. State Farm, however, the analysis of concurrent causes that might be appropriate for liability cases, where tort law sets the stage for analysis, is not appropriate for first party causation analysis, where the court said efficient proximate cause was the correct analysis. (California is one of just a few states where statutes have been interpreted as mandating efficient proximate cause and precluding the contractual use of anti-concurrent cause provisions).
But California v. Allstate is a liability case, and Partridge remains good law in California for those kinds of disputes. Now, it should be said at this point, that not every state has a case like Partridge, and courts in those states may not buy arguments in the liability context based on the reasoning of Partridge, as I have personally found out. A few words about what Partridge held: it involved an insured under a homeowners policy that contained, as most or all of them do, an exclusion for liability arising out of the use of a motor vehicle. The insured committed two negligent acts -- filing a gun trigger to a "hair trigger" -- so that it would go off with a slight touch, or even, as it turned out, without being pulled at all -- and driving erratically, which led to the gun firing and injuring a passenger. Partridge featured two independent causes, neither one of which could have caused the result by itself. You can formulate this circumstance in various ways, that is, you can use different words to describe what it is, but the usual way in this context is to say the causes were indivisible, incapable of being separated.
The Allstate court used the Partridge analysis to find that sudden and unexpected, or accidental, releases of pollution, which were covered, could not be divided under the facts presented from releases that were gradual and not sudden and which presumably would not be covered. The court picked this phrase from Partridge as the most apropos: "whenever an insured risk constitutes a proximate cause of an accident, even if an excluded risk is a concurrent proximate cause," liability coverage will exist. The result, apparently, is that all the pollution liability of the state is covered.
Before discussing this a little more in-depth, let me just add a curiosity here. I noted the analysis of the court on the "sudden and unexpected" language with great interest, because in Oregon, a case called Baxter & McCormick has largely read the "sudden" out of sudden and unexpected, and the analysis would have been quite different -- if this case occurred in Oregon there probably would not have been a discussion of covered vs. uncovered and concurrent tort causes at all.
Now, let's talk a bit about this idea of concurrent cause in the liability context. I have thought a great deal about this over the past couple years, and there is something intellectually unsatisfying in doing so. Trying to nail down the analysis in this area is kind of like sticking your arm in water -- the refraction makes it appear that your arm has become disjointed and separated from itself. Likewise, in this area, it can be hard to tell illusion from reality. One possible conclusion is that there is no reality at all, because the entire concept is artificial and a human construct. Many things are human constructs, of course, and that does not stop us from thinking of them as reality, or at least reflective of something that is or resembles objectively or empirically observable. But this area is somewhat different, it seems to me, and less prone to being pinned down, tested or verified.
Consider this, when considering Partridge: think about the motor vehicle exclusion in the homeowners policy in that case, or a "business pursuits" exclusion in a homeowners policy, or perhaps an earth movement exclusion in a contractors liability policy. In Partridge, the negligent act of altering the trigger mechanism was found to be a proximate cause. So was the negligent use of a motor vehicle. But no matter how you see this case, one thing has to be admitted -- the use of "negligence" as a cause is particularly prone to manipulation, and I don't say this with malice, I simply observe that this is so. Since any court case will involve the actions of human beings, no event that will be considered by a court is untouched by potential human negligence, and this can consume and overwhelm any other attempt at analysis. If you read the Appleman's article I linked to above, you can see this, and this was noted by the Fifth Circuit in some of the Katrina cases, such as In re Katrina Canal Breaches Litigation.
In other words, if a court will let you, you can always attack an exclusion as not really excluding liability because, although it may exclude liability "arising out of" business pursuits, or motor vehicles, or earth movement, it does not specifically exclude human negligence. So the question arises: is negligence on par with motor vehicle use as a cause? Are these like comparing apples and oranges, or even further apart, like comparing apples to monster truck tires? Again, I am not taking sides in this, I merely point out that they are not conceptually equivalent.
One final word, about the Allstate case. You have reflected by this point in this post that the facts of Allstate and Partridge are not the same. As the insurers pointed out in briefing, Partridge involved a single, one-time, injury, but Allstate involved various kinds of pollution releases, some sudden and some not sudden, over a period of a number of years. The court gives an indication it would have given credence to this argument, if the facts would have allowed the court to see the costs of remediation as divisible into costs associated with overflow of the containment pits versus costs associated with gradual seepage from the pits. That is a very hard distinction to make, if you are in the position to have to make it, and all the chemicals come from the same source and are the same chemicals. It is easier to do this, in environmental work, where, say, one polluter had gasoline at a site, and another had dry cleaning fluid. Test monitor wells and soil samples will allow a pretty decent estimate of how much of each kind of chemical makes up the pollution mix. So, not knowing more about this case, on the surface what the court asks to be done sounds incredibly difficult to do.
I do note one other difference with Partridge. The analysis of the Allstate court does not, at least it appears to me, go nearly as far as some courts do in separating negligence as a cause from actual events, or in other words what we often think of as "causes." I'm not sure if this different way of approaching the problem is because of the difference in the facts, or if the Allstate court has a somewhat different way of thinking about the issue than did the Partridge court.
UPDATE: Almost forgot, I saw this case analyzed in a piece by Robert M. Horkovich at the LexisNexis Insurance Law Center. Full disclosure: I'm on the advisory board of the ILC, so obviously I have a vested interest, I think it's a good product and I shamelessly plug it when I can.