The Judge and State Law: MTBE Multi-District Litigation

Imagine a product that, when used properly, is safe, valuable and environmentally sound, but when used improperly is dangerous to the environment. Many such products exist. Drano is safe and useful--except when stored in the pantry salt shaker. Firearms save many lives--but not when used by criminals. 1 Automobiles get us where we need to go--but in the hands of bank robbers and drunks they can be lethal weapons. And the gasoline from our automobiles can either get us rolling, burn our house down, or leak through rusty gas tanks into the ground.

In every one of these cases the product itself is not deemed defective or unreasonably dangerous. The product is considered fine if it is correctly manufactured and accompanied by adequate instructions for proper use. If harm occurs, we assign legal blame for the misuse of the product, for its negligent storage or for the underlying crime.

Keep this in mind as we discuss the Methyl Tertiary Butyl Ether (MTBE) multi-district litigation. For MTBE, when used properly, is inoffensive to the environment. Indeed, MTBE helps the environment according to the Environmental Protection Agency, since it is an oxygenate that allows gasoline to burn more efficiently. And oxygenates are required by law (as will be discussed below).

I. In re: Methyl Tertiary Butyl Ether ("MTBE") Prods. Liab. Litig . 2

A. The Plaintiffs' Claims: The Procedural Posture

Plaintiffs--(four Boards of Education or School Districts; one Church; one Company; fifteen Utility Companies or Water Districts; twenty-Six Towns, Cities, Municipalities or Fire Districts; three Home Owners Associations (HOAs) or Villages; and two individuals from fifteen different states)--sought relief in multi-district litigation assigned to the Southern District of New York, for defendants' contamination or potential contamination of groundwater by MTBE. The defendants--five petroleum companies that supply some form of petroleum and own and operate various gas stations--invoked rule 12(b)(6) of the Federal Rules of Civil Procedure to demur to all of the complaints, which had originally been filed in fifteen states but removed to federal court for reasons of diversity of citizenship. In essence, the defendants' demurrers point out that MTBE is not intrinsically harmful, and that in any case plaintiff s cannot identify the origin of any product which in fact harmed any one of them, as the chemical is fungible. If the plaintiff s cannot identify a product's manufacturer, they must rely on some theory of collective liability, which defendants claim that all fifteen states reject.

Other courts have consistently ruled that the Reformulated Gasoline (RFG) requirements of the Clean Air Act do not free petroleum companies from liability for damages caused by their oxygenate as a matter of law, for the Act only requires the defendants to use an oxygenate, and they chose to use MTBE.3 This is true. But the lack of immunity under the Clean Air Act does not mean, of course, that the defendants should not succeed on demurrer if there is no proof they produced a defective and unreasonably dangerous product that proximately caused harm to an identifiable individual.

B. The Science

In an effort to significantly reduce summertime smog pollution and year-round air toxic emissions, Congress required the use of RFG beginning in 1995 in the nation's most polluted cities. As part of these "clean gasoline" specifications, Congress required that every gallon of RFG contain cleaner burning fuel additives called oxygenates.

The two most commonly used oxygenates were Methyl Tertiary-Butyl Ether (MTBE) (used in approximately 85 percent of RFG), and ethanol (used in approximately 10915199244 15 percent of RFG). According to both the Congressional Research Service and the federal Department of Energy, ethanol is the more difficult and expensive way to meet the new RFG specifications than MTBE. This is true for two main reasons:

i. Pipeline Transportation: Ethanol's high affinity for water does not allow it to be blended with gasoline at the refinery, nor transported through the existing nation-wide gasoline pipeline infrastructure. Ethanol must be stored in segregated tanks, can only be transported by rail or truck to its final destination and must be blended into gasoline at the terminal or even the retail gas station. As a result, the cost of blending ethanol into gasoline is significantly higher than the cost of gasoline without ethanol.

ii. Blending Characteristics: RFG's clean fuel specifications call for limits on gasoline's ability to evaporate quickly in the summertime. Because ethanol blends evaporate more readily than MTBE blends, ethanol-using refiners are forced to spend additional resources and capital to produce a gasoline blend stock with ultra-low evaporative properties. This is a very expensive process that adds significantly to the cost of producing summertime gasoline ready for ethanol.

MTBE has none of these disadvantages. It is derived principally from natural gas, which is in abundant supply in the US. It can be safely and efficiently blended into gasoline at the refinery and efficiently shipped via the interstate gasoline pipeline system, already mixed with gasoline.

The differences in the efficiency and convenience between MTBE and ethanol are apparent, as witnessed by gasoline prices before more recent federal regulations mandated increased ethanol use. The primary ethanol/RFG market was the Chicago/Milwaukee area, where gas prices were over 40 cents per gallon higher than elsewhere; over 85 percent of the nation's gasoline providers used MTBE, their oxygenate of choice. In sum, defendants and many other companies chose to use MTBE for sound economic reasons. The cost efficiency of MTBE as opposed to ethanol is of course a good thing, unless one is an ethanol producer.

MTBE is highly soluble; once it enters a water source it is not readily biodegradable. When properly stored and used, MTBE never enters any water source. But beginning in late 1996, MTBE was discovered at low levels in groundwater sources in California, notably in Santa Monica and Lake Tahoe. Since then, MTBE has been detected at low concentrations in other parts of the country. Invariably, the presence of MTBE in groundwater was linked to underground storage tanks (USTs) that had been leaking gasoline for an extended period of time--several years, in many instances. These leaks are typically due to inadequate or non-existent UST inspection and/or maintenance practices. MTBE received an inordinate amount of attention from public officials because it is more water-soluble than other gasoline components and thus can be transported faster and farther in soil and water. As a result, MTBE is the canary in the proverbial coal mine: it is often discovered at the front edge of any gasoline plume traveling through the soil. An MTBE leak signifies that other, much more harmful, gasoline components, such as benzene, are in fact present as well. MTBE can persist for decades in water supplies and if foul-smelling a small quantity can make the water supply unfit for human consumption. Of course, this also applies to many other products, including non-oxygenated gasoline. To repeat, MTBE was never meant to get into water supplies in the first place and will not get into a water supply if stored in proper gasoline holding tanks.

As of March 2001, the California Department of Health Services reported that MTBE had been detected in 0.8 percent of all water sources sampled, with only 0.2 percent of those samples exceeding California's primary health standard for MTBE. In addition, a report by the engineering consulting firm, Exponent, Inc., concluded that, "Despite the negative publicity surrounding MTBE and potential aesthetic issues, MTBE in drinking water should not pose a significant public health hazard in California " MTBE has become a political scapegoat, one very attractive to the ethanol producing lobby, blamed for failure to enforce federal storage tank regulations. This has occurred despite the fact that it is far more cost-effective to ensure that UST systems are properly preventing leaks than it is to ban the use of MTBE. When UST systems work well, all leaks, not just MTBE leaks, are prevented. According to the EPA, compliance with the 1998 minimum UST installation/upgrade requirements and the 1993 UST leak detection requirements is a national priority. Data suggests that UST systems in compliance with applicable regulatory requirements are not experiencing problems with leaking gasoline or gasoline additives, including MTBE. The dramatic impact that UST upgrades have had on groundwater protection is evident in recent contamination data from the California Department of Health Services. This data indicates that as USTs are upgraded the concentration level and frequency of MTBE detections is leveling off and beginning to decline. However, as of 2000 more than 40 percent (304,000 USTs) of all USTs were still not in compliance with 1993 leak detection regulations. More than 15 percent (150,000 USTs) remain out-of-compliance with 1998 regulations for the upgrade of spill, overfill and corrosion protection requirements. By law, all non-compliant USTs must be closed; however, many remain operational. Clearly, those operating non-compliant USTs should fear the wrath of tort law.

Upon repeated oral exposure of very substantial amounts of MTBE, female rats demonstrated an increase in the combined tumor types, lymphoma and leukemia. Male rats developed an increase in testicular tumors. Results from each of the two long-term inhalation studies in laboratory rats and mice, respectively, showed an increased occurrence of kidney and testicular tumors in male rats and liver tumors in mice of both sexes. The relevance of these findings to humans, at concentrations of MTBE found in the environment, is questionable. Extremely high doses were administered to the animals, and it is not known how MTBE causes tumors in these animals. Nevertheless, several agencies have concluded that it is carcinogenic in animals. Plaintiff s claim that it may also be carcinogenic in humans, but no one has yet determined that.4 The EPA reviewed available health effects information on MTBE in its 1997 Drinking Water Advisory guidance and decided that there was insufficient information available to allow the EPA to establish quantitative estimates for health risks--and as such would not set health advisory limits. The drinking water advisory document indicates that there is little likelihood that MTBE in drinking water will cause "adverse health effects" at concentrations between 20 and 40 ppb or below. Those "adverse health effects" are essentially unpleasant odor and taste, but the vast majority of MTBE detections have been at non-sensory concentrations, under five parts per billion (ppb)--well below the EPA Consumer Advisory level.

The plaintiff s claim that the defendants were aware of dangerous contamination qualities of MTBE, and that the defendants are therefore each jointly responsible for contaminating their water, no matter whose MTBE actually caused the damage. Note that this reasoning could apply equally well to Drano, to gasoline itself, to fi rearms, and to automobiles. Ford and GM "know" that some of its cars will be misused by drunkards and criminals--but only the latter remain liable for the misuse of cars.

C. Collective Liability Theories

There are four recognized theories of collective liability: concert of action, alternative liability, enterprise liability, and market share liability. Each theory is often altered to fit a particular state's preferences, but they all have a basic definition from which the states begin their analysis.

CONCERT OF ACTION is described by the Restatement (Second) of Torts Section 876 (1979) as a vicarious liability, where one party is responsible for the acts of another if he (a) does a tortious act in concert with the other or pursuant to a common design with him, or (b) knows that the other's conduct constitutes a breach of duty and gives substantial assistance or encouragement to the other to so conduct himself, or (c) gives substantial assistance to the other in accomplishing a tortious result where his own conduct, separately considered, constitutes a breach of duty to the third person. For example, if you and I rob a bank together, I am liable for the entire amount of booty stolen, and injuries to the teller whom you struck.

ALTERNATIVE LIABILITY, which was first adopted in the controversial Summers v. Tice , occurs "where the conduct of two or more actors is tortious, and it is proved that harm has been caused to the plaintiff by only one of them, but there is uncertainty as to which one has caused it."5 In these cases, "the burden is upon each such actor to prove that he has not caused the harm."6 Both of us misbehaved, one of us harmed the plaintiff , and the other did not, but the plaintiff cannot make his case against either of us, so the court deliberately chooses to hold both of us liable. This derogation from the normal plaintiff 's burden of care is not upheld in all states, and where it is upheld it is done only in extreme cases (e.g., where misbehavior was virtually concerted).

Under ENTERPRISE LIABILITY plaintiffs must demonstrate defendants' joint awareness of the risks at issue and their joint capacity to reduce or affect those risks. As it were, each company in an industry gives cover to the others as they misbehave, much as a rampaging mob gives anonymity to each rampaging citizen.7 Enterprise liability is only applicable to industries composed of a small number of units, for "what would be fair and feasible with regard to an industry of five or ten producers might be manifestly unreasonable if applied to a decentralized industry composed of thousands of small producers," who could hardly concert.8

MARKET SHARE LIABILITY allows the plaintiffs to shift the burden of proof to the defendants when identification of the product manufacturer is problematic or impossible.9 "The plaintiff must join as defendant manufacturers representing a substantial share of the particular market and each defendant is liable for the proportion of the judgment represented by its share of the market unless it demonstrates that it could not have made the product that caused the plaintiff 's injury."10 Four states accepted market share liability for DES (the generic, synthetic female hormone, still used for many purposes but once incorrectly used to prevent miscarriage, causing ovarian cancer to female off spring). Most states squarely reject market share liability.

To these four established federal approaches to the issue, Judge Scheindlin added a fifth, dangerous legal innovation.

D. Judge's Scheindlin's Addition

Judge Scheindlin's COMMINGLED PRODUCT SHARE MARKET LIABILITY theory is an expanded version of market share liability.11 This theory was held by Judge Scheindlin to be applicable when "a plaintiff can prove that certain gaseous or liquid products (e.g., gasoline, liquid propane, alcohol) of many suppliers were present in a completely commingled or blended state at the time and place that the risk of harm occurred, [if ] the commingled product caused a single indivisible injury, [so that] each of the products should be deemed to have caused the harm."12 Under this theory damages should be apportioned by proof of the defendants' share of the market. Plaintiff s only need to identify those defendants they believe to have contributed to the 'commingled' product which caused their injury; they must conduct some form of an investigation so that they can make a good faith effort to identify the defendants whom they believe caused their injury.13

Again, note that unless the defendant manufacturers knew that their MTBE would be placed in a particular leaky tank, they are similarly situated to Ford, who "knows" statistically that some of its drivers will use their cars as a deadly weapon, but also that this is not the case for any particular driver. Unless such knowledge is proven, it is very hard to see how this new theory could plausibly apply without implying a revolution in products liability.

II. Federal Judges Predicting State Law

When a case is in federal court due to diversity and the substantive law of a forum state is uncertain or ambiguous, the federal court may certify the question to the highest state court.14 Even when certification is not available under state law, a federal court "not infrequently will stay its hand, remitting the parties to the state court to resolve the controlling state law on which the federal rule may turn."15 Of course, the federal court may also attempt to predict how the highest court of the forum state would rule.16 To make these predictions the federal judge must look to the state constitutions, statutes, judicial decisions, the Restatements, as well as law from other states.17

A. Judge Scheindlin on How to Predict State Law

While many federal judges take a cautious and conservative view when making predictions about the evolution of a state's laws, Judge Scheindlin states that a more liberal view is in order. For while "a court may not adopt innovative theories without the support of state law, or distort existing state law, when a case is removed to federal court, the plaintiff is entitled to the same treatment it would receive in state court."18 Judge Scheindlin's theory is that the fears about possibly distorting established law or wrongly speculating about trends in state law are only appropriate where the plaintiff s brought the action in federal court, for then their motive may be to "obtain a broader interpretation of state law."19 In this case the plaintiff s have not brought the case to the federal court, rather the plaintiff s objected to its removal there by the defendants.20 Judge Scheindlin argues that in this situation, a liberal construction of state law is in order, for it protects principles of dual sovereignty "by protecting a party who sought to obtain a resolution of state law claims from state courts."21 If a more restrictive view was adopted, this would lead to forum shopping--a main concern of the Erie Doctrine.22

Under this reasoning, if a corporate defendant properly removes (for diversity reasons) a state law case to federal court, for fear that as an out-of-state corporation it would be discriminated against (i.e., would not have state law correctly applied to it) by the elected judge and local jury, the federal court is now free to hold the corporation liable anyway. This is a significant new illustration of "darned if you do, darned if you don't."

Judge Scheindlin takes a very broad view of predicting state laws; this view is evident in her ruling.

B. An Opposing View of Predicting State Law

Many take the opposing view of federal prediction from Judge Scheindlin and implement a certification process or simply adhere to a conservative view of the state law and thus refrain from making innovative predictions when the state law is not easily discernable.

Certification occurs when a federal court, sitting for diversity purposes, is faced with an unclear question of state law, and asks the forum state's highest court for resolution.23 Proponents of certification argue that the process is beneficial and promotes comity and federalism and avoids prognostication by federal courts.24 Certification may "save time, energy, and resources and helps build a cooperative judicial federalism."25 The Supreme Court has held that for a "matter of state law [federal judges are] 915199255outsiders' lacking the common exposure to local law which comes from sitting in the jurisdiction."26 In Lehman Bros., the Court does not assert that certification is obligatory merely because state law is in doubt.27 The Court does say, however, that certification would seem "particularly appropriate" because of the novelty of the question posed and because the state law to be applied (in the instance, that of Florida) is in considerable distance from the federal court (New York).28

IV. "Erie Guesses"

A. Indiana

Among fifteen states, Judge Scheindlin predicts that Indiana will apply her commingled theory of market share liability.29 She cites two cases toward that end.

In City of Gary v. Smith & Wesson Corp. , the courts chose to reject market share liability theory, but stated that even if they chose to adopt it, it would not be applicable in their case, for guns are not fungible and there was a great remoteness problem between the defendants and the crimes that their products later caused.30 The "[market share] approach to allocation of liability has not been adopted in Indiana. . . Whatever the merits of 915199255market share' in other contexts, we do not believe it is properly applied in this situation involving such a wide mix of lawful and unlawful conditions as well as many potentially intervening acts by non-parties."31 Judge Scheindlin's theory is that this case is not determinative of what Indiana would do in the case of MTBE, for the two cases are too dissimilar, with the key being the fact that guns are not fungible, while MTBE is fungible.32 But of course MTBE only affects water supplies when it escapes through leaky tanks that are not the responsibility of MTBE producers--and leaky tanks are not interchangeable with non-leaky tanks, are they?

Judge Scheindlin relies heavily on E.Z. Gas, Inc. v. Hydrocarbon Transp., Inc. to reach her conclusion of commingled market share liability.33 In E.Z. Gas , a gas explosion caused injury to plaintiff when he lit the pilot of his gas heater, because of its lack of gaseous odor.34 Various gas manufacturers' non-odor-added products had been commingled and thus the plaintiff could not identify whose product caused his injury.35 Indiana allowed burden shifting because there was a product identification problem.36 Judge Scheindlin also relied on Indiana's Comparative Fault Act,37 which adopted Indiana's Products Liability Act,38 and requires that parties' level of liability must be apportioned. Joint responsibility is rejected in Indiana. But this is enterprise liability, and maybe even concert of action, not some version of market share.

Strangely, in a footnote, Judge Scheindlin reserves the use of market share liability, musing that Indiana may one day adopt ordinary market share liability based on the Restatement (Third) factors of market share liability (Section15).39 However, there is no evidence that Indiana has adopted this section of the Restatement (Third). While Indiana courts have used the Restatement (Third) in products liability cases, they have yet to specifically refer to Section15.40 In addition they continue to rely on portions of the Restatement (Second) of Torts.41 But mere musing is not sufficient to preserve the practice of dual federalism.

B. Kansas

The judge also claims that Kansas will leap to apply her theory of market share liability. MTBE Prods. Liab. Litig. 42 However, Kansas is silent on whether or when they would adopt any collective liability theory.43 Even more important is that Kansas' legislature has made it clear that policy-based alterations to well established tort law doctrines should be left to the legislature. Thus it would be inappropriate for federal courts to venture into realms of Kansas public policy.44

Despite this, the judge comes to her conclusion by analyzing McAlister v. Atlantic Richfield Co. , where a plaintiff sued various oil companies for polluting his fresh water well.45 She concludes from this case that collective liability is allowed where all defendants acted tortiously towards the plaintiff .46 In McAlister it is "not a prerequisite to recovery that it be shown that the [defendants] were the sole cause of the pollution."47 Rather, there simply must be enough facts from which the defendants could reasonably be inferred to be the source; once this is determined the issue may then go to the jury.48 Again, this is concert of action or enterprise liability--it has nothing to do with Judge Scheindlin's case.

In addition, Judge Scheindlin relied on the Restatement (Second) of Torts Section 433B(2) and Kansas' comparative fault statute in determining that Kansas will likely apportion the damages among the defendants.49 But apportionment of comparative fault still requires causation and individualized damages, facts the judge seemed not to address. McAlister appears to be saying that plaintiff s may name any one defendant as long as they can prove that the defendant is wrongfully responsible for their injuries; they do not have to list every possible defendant.50 Judge Scheindlin also relies on the "common law notion that oil companies should be liable for storing hazardous substances on their land and permitting those substances to damage the plaintiff 's property" to argue that the defendants have breached a duty to the plaintiff s.51 However, there is no proof whose oil damaged the plaintiff s' property, i.e., which tanks were leaky, and this fundamentally undermines the judge's analogy.

Respecting federalism would require a conclusion that Kansas would insist that plaintiff s first prove that each defendant violated a duty against them or that each defendant was reasonably responsible for their injuries before going forward with their case.52 The fact that the plaintiff s cannot prove this, and that the Kansas courts have not addressed the theories of collective liability would lead one to conclude that the Kansas plaintiff s' causes of action may not survive the defendants' 12(b)(6) motion.53

C. Vermont, Virginia and West Virginia

Vermont, Virginia and West Virginia have not had a chance to rule on the viability of collective liability in a products liability case, and Judge Scheindlin stated that she was not able to discover a "basis for inferring whether [those states] would accept or reject collective liability in MTBE cases."54 What this means, of course, is that potential plaintiff s (there were "DES daughters" in all 50 states) did not even dare sue in these states, knowing as they did that their suits would be dismissed on demurrer for failure to state a legally cognizable grounds for liability. However, despite this, Judge Scheindlin concluded that the states would adopt the theory of market share liability.55 Judge Scheindlin reached her conclusion by citing two cases where the states expanded on the common law "to meet the changing needs of their society" and allowed recovery where, in the past, tort victims were unable to recover.56 This is, however, in utter contention with our system of federalism; it is the states' right to expand upon the common law, not the federal court system. In brief, because these states' common law has not remained an un-moveable concrete block, Judge Scheindlin is authorized to treat it as moist clay moldable to her liking. D. New Jersey Judge Scheindlin predicts that New Jersey will adopt a market share liability theory, despite the fact that New Jersey has explicitly rejected market share, enterprise, alternative and concert of action theories in the context of products liability actions.57 Judge Scheindlin bases this prediction on the idea that Shakil v. Lenderle Labs left the door slightly ajar, for the use of a market share theory of liability in a different context.58 But Shakil made clear that its ruling was confined solely to the context of vaccines.59

Judge Scheindlin theorized that this MTBE case would nonetheless fall under the 915199255rule' of Shakil, which was never the rule and addressed only vaccines.60 She came to this conclusion by comparing Shakil and this MTBE case. First, the two cases have a mutual interest in public safety.61 Second, vaccines are required to save lives,62 and gasoline, while needed, is not required to protect lives.63 Placing liability on the defendants will not harm the public.64 Third, liability exposure would have hurt companies willing to make the vaccine--for only two existed--while with MTBE there are over fifty petroleum companies. This effort at legal reasoning turns on its head earlier New Jersey law. Now, the less likely it is that you caused individualized damage the more likely it is that you will be held liable.

There is solid reason to think that New Jersey will decline the wide-spread adoption of market share liability.65

E. Louisiana

Louisiana plaintiffs asserted seven causes of action: unreasonably dangerous design in violation of the Louisiana Products Liability Act (LPLA), inadequate warning in violation of the LPLA, negligence, public nuisance, private nuisance, trespass and civil conspiracy.66 Judge Scheindlin dismissed the Louisiana plaintiff s' five non-LPLA claims, for they were precluded by the statute.67 Her analysis was thus based on whether the two LPLA claims may survive on a theory of collective liability.

Judge Scheindlin claims that Louisiana has been silent on the adoption of collective liability. Again, what this reveals is that "market share" plaintiff s ("DES daughters") did not even dare sue in Louisiana. Indeed, the defendants claimed that other federal courts, when called on to apply Louisiana law, had consistently refused to recognize any theory of collective liability.68 However, according to Judge Scheindlin, such decisions are irrelevant. She concludes the Louisiana courts will adopt the market share theory of liability.69

Judge Scheindlin relies heavily on Gould v. Hous. Auth. of New Orleans to reach her conclusion.70 She argued that in this case the courts allowed the plaintiff s to move forward with their claim, despite the fact that they could not identify the exact manufacturers of the lead based paint, which caused the tenants' lead poisoning.71 Gould relied on Louisiana's statute, which allows for pleading in the alternative.72 The statute specifically states that "a petition may set forth two or more causes of action in the alternative, even though the legal or factual bases thereof may be inconsistent or mutually exclusive."73 Gould refrained from making any decisions on market share or collective liability, for the appeal was from the dismissal of the claims and based on Louisiana Civil Procedure; the issue of market share or collective liability did not need to be decided upon at that point in the proceedings.74

Judge Scheindlin is correct that the plaintiff s may move forward with their causes of action at this time due to Louisiana's alternative pleading rules. However, one questions why Judge Scheindlin would then take the extra step of invading the Pelican State's sovereignty to decide such a controversial issue as the adoption of collective liability: precisely what was reserved in Gould. What is more puzzling is that she stakes this claim on the idea that Louisiana has or would adopt the Restatement (Third) of Torts Section15.75 There is no evidence that the judge cites that Louisiana would adopt the Restatement (Third) or Torts. Rather, the Restatement (Third) of Torts Section 15 cmt. c specifically discusses how Louisiana has rejected market share liability. Section 15 cmt c, cites the Louisiana case Jefferson v. Lead Indus., which rejects market share liability.76

There are numerous Louisiana cases which reject the theory of market share liability.77 Judge Scheindlin, while correct in her ruling that the plaintiff s' cause of action may move forward at this time, had no need to make a prediction about Louisiana's adoption of the market share liability theory.

F. Connecticut

Connecticut has not considered the theories of collective liability--but again, Judge Scheindlin concludes that they will adopt her novel theory.78

She relies on two cases: Sharp v. Wyatt79 and Champagne v. Raybestos-Manhattan Inc .80 In Champagne -- Sharp relied on Champagne --th