Public nuisance is in the news again. Three important opioid cases have been recently decided. In November plaintiffs lost a bench trial in California state court, and eight days later, the Oklahoma Supreme Court reversed a $465 million trial verdict, holding that, as a matter of law, public nuisance does not extend to the manufacturing or marketing of prescription drugs. About a week later, a jury in a bellwether, the Ohio federal MDL, held that pharmacies caused a public nuisance by failing to respond to curb medically unnecessary prescriptions.
David Dana’s article offers a bold prescription to courts about how to approach public nuisance, including the opioid litigation. Dana’s argument should, in theory, make sense of November’s mixed bag of decisions. His argument operates at two levels, first about the relationship between public nuisance and democracy, and second about the specific wrongful conduct which the tort of public nuisance should address.
Dana begins his analysis at a familiar place: that the meaning of public nuisance is “contestable,” (P. 9). This is, perhaps, the only thing on which there is wide-spread agreement among scholars and courts. Not only has caselaw expanded to include conduct that earlier courts would have excluded, but there have also been episodes of genuine conflict over the best account of that caselaw, as when, in 1970, the American Law Institute refused to adopt Prosser’s more restrictive definition of public nuisance as arising from conduct that was criminal.
In light of this familiar challenge, Dana adopts a familiar strategy. He asks, “when have courts actually been willing to find a public nuisance and order relief?” (P. 9.) The three categories he identifies are (1) quasi-crime cases, (2) environmental cases and (3) product-based cases. (P. 9.) From this empirical observation, he draws the following conclusion: the three categories are united in that they “all do the work of what a well-functioning administrative state should do.” (P. 10, emphasis in original.)
Dana’s next analytic move is unexpected and clever. He observes that modern public nuisance – as defined as the universe of the three categories above – is “in tension” with the “ideal of the administrative state” (P. 11), because every public nuisance claim implies that the administrative state has failed. The function of public nuisance is to “make actionable unreasonable interferences” with “public rights”. (P. 11, paraphrasing Restatement (Second) of Torts § 821B.) This being the case, Dana acknowledges that the criticism of public nuisance as lacking democratic legitimacy—made, for example, by Donald Gifford—has some superficial appeal, and his project is to defend public nuisance against it. I say that this move is unexpected because by assuming that modern public nuisance properly understood appears to function like administrative law, Dana seems to be conceding Gifford et. al.’s chief objection. The move is clever because if he can pull off this defense, then the doctrinal scope of modern public nuisance law will be as broad as that of the modern administrative state.
Dana argues that when critics of modern public nuisance law claim that courts cannot hear claims concerning “public rights” because doing so oversteps the power of the democratically elected branches, they treat public and private rights differently without offering an argument other than stipulating that the former cannot be enforced by the courts. (P. 13.) Turning the tables on the critics, Dana argues that it would be odd—in fact, positively undemocratic—for courts to retreat from their traditional role of remedying all rights violations, including “public rights.” (P. 14.)
Dana’s defense is doubly clever, because it will appeal to torts scholars, like myself, who are committed to viewing torts as redress for rights violations. It is undeniable that public nuisance is defined in the Restatement Second? as a common law action in response to a violation of “a right common to the general public,” and, unlike Tom Merrill, I would not want to banish public nuisance from tort law. A right “common to the public” does not on its face identify? the state as the right-holder. The implication is that this “right” has the same basic private law character as other rights in tort: it is a private relational right running between persons. Unlike other tort rights, a right common to the public is “common” in that it runs from every member of society to every other member of society.
To the extent that public nuisance deals with relational rights held by persons (and not the state), it would seem to be consistent with tort law, which has also gradually imposed new obligations on various powerful members of society, especially businesses, without violating democratic principles. And from this perspective, the putative tension with administrative law that Dana places at the center of his defense can be resolved. To say that the law of negligence empowered MacPherson to demand that Buick carefully inspect wheels purchased from a supplier does not entail that New York tort law was “in tension” with the power of the New York legislature to require a seller of automobiles to carefully inspect wheels purchased from a supplier.
MacPherson had a right—personal to him—that Buick engage in specific conduct (carefully assembling a car) only under New York tort law. The fact that public law and tort law may address identical conduct and secure identical ends does not mean that they are in competition; it means that they are complementary (unless there is preemption, of course).
However, I have trouble with Dana’s further arguments that (1) for any harmful conduct which could be lawfully addressed by administrative law, there is a correlative public right to be free of that harmful conduct which could be addressed by public nuisance, and (2) whether that ‘could’ becomes ‘is’ depends on a balancing test that weighs rule of law values against social welfare. (P. 6.) The remainder of this Jot challenges the first half of this argument.
Dana unduly conflates the domain of administrative law with the domain of public nuisance. The former includes a wide range of ends which the state can secure under the police powers, including sanctioning conduct to promote policy goals. A decision by the Treasury Department not to classify cryptocurrency issuers as banks may be, in fact, a mistake, but that does not mean that the issuers have committed a wrong if they do things that would not have been permitted had the Treasury Department done its job. Public nuisance is narrower than administrative law “done right”: its wrongs must be grounded in an (unreasonable) interference with a “public” right.
Legal rights are correlative to legal duties. But not all legal duties generate legal rights. As Goldberg and Zipursky have explained, relational duties—based on directives about how to conduct oneself in relation to another—generate rights. Simple duties—based on directives about how one should act simpliciter—do not generate rights. The law contains simple duties, which are owed to the state, based on simple directives, and relational duties, which are typically owed to other persons but can also be owed to the state when the state’s right is based on its ownership of property, not its police powers.1
Public nuisance claims only embrace conduct that breaches a relational duty to private parties, not to the state. If the Park Service cannot drive one of its maintenance vehicles into a portion of a park because of a trespasser, it cannot bring a public nuisance suit to get an injunction to remove the trespasser (although it can, of course, sue in trespass or use public law). But the Park Service can (in theory) claim standing under public nuisance if the trespasser significantly interferes with the public’s ability to enter the park. Every public nuisance claim protects a public right—a right whose correlative is a special kind of relational duty between the defendant and members of the public.
Given that, by definition, a “public” right is a subset of the private rights extant in tort, what test can identify them? A test based on the interest invaded seems to be a non-starter. Even ancient examples of public nuisance, such as classic cases of “fouling a public waterway” can be characterized as conduct that interferes with either (i) an interest that could not ground a public nuisance action (the user of the waterway’s right to control what touches his property) and (ii) an interest that could ground a public nuisance action (the user of the waterway’s right to traverse the waterway). A better test, I think, is Merrill’s: a public right is grounded on the obligation not to cause a “public bad”. He defines a violation of a public right as “conduct that] produces undesirable effects that are nonexcludable and nonrivalrous.” (Is Public Nuisance a Tort, at 8.)
While I don’t agree with Merrill’s ultimate conclusion—that public nuisance is not a tort—I think he is absolutely right about what kind of rights violation grounds public nuisance. It is a private right—held by everyone in society—that the defendant conduct himself in such a way not to interfere with each rightholder’s capacity to x, where x is a “non-rivalrous” interest, such as access to the public roads or clean air. The correlative of the nonrivalrous nature of the right is that the remedy is “nonexcludable”: removing the obstruction in the road is relief to everyone, not just one person or the person who most immediately wants to use the road.
Once the precise character of a public right as a subject of a relational duty is established, it is easy to see how Dana’s conflation between administrative law and public nuisance gets things wrong. Administrative law’s domain is huge, and a moment’s reflection reveals the implausibility of saying that every interest protected by a legitimate exercise of administrative law becomes by definition a “public right”. The legitimate exercise of authority by New York to require employers to distribute an information sheet on sexual harassment, which is required by NYC Administrative Code § 8-107, sub. 29(e), protects the interests of employees. If the employer fails to comply with that regulation, and an employee suffers an injury cognizable in tort as a result, the breach of the relevant relational duty is not between the public and the employer but only between that employee and the employer (although the regulation can be used as evidence of liability).
This suggests that injuries that are only contingently connected to the violation of a public right, such as personal injuries, are distinguishable from injuries to the public. The right of the public in California that justified the injunction against gang members in People ex rel. Gallo v. Acuna2, contingently resulted in fewer personal injuries (and property damage) to be sure, but the court specified that the wrong redressed was the gang members’ unreasonable interference with the “right to public order”. Many important interests protected by tort law will not be directly protected by public nuisance. This understanding of public nuisance explains the skepticism of critics like Gifford towards using public nuisance to remedy the many evils done by powerful actors like the tobacco industry and the opioid industry.
Dana, like many who have defended recent public nuisance litigation in connection with tobacco, lead paint, and other commercial activities that caused widespread personal injuries, rejects the argument that because these cases involve defective products, they cannot also be brought under public nuisance. I agree with his general point; there is no reason to categorically deny a plaintiff a right to redress under public nuisance simply because the conduct at issue by the defendant is coextensive with conduct that also could be redressed under products liability law. But expansionists like Dana must still make the positive argument still must be sustained for holding that the conduct related to the manufacturing, and distribution, of a product violates a relational duty that is within the domain of public nuisance.
As Dana notes,
[P]roduct-based public nuisance claims differ from standard product liability claims, to the extent that … the producers of the harmful products were able to inflict harm on the public for profit because they misrepresented what they knew about the risks inherent in their products and thereby undermined the ability of the government – the legislature, agencies – to protect the public, as well as undermining the ability of members of the public to protect themselves. (P. 37.)
This is an interesting insight. Interference with the ability of the state to regulate products is conduct which simultaneously may violate (1) a simple duty to the state (e.g., criminal and civil commands of candor and disclosure); (2) a relational duty to the consumer in products liability (and perhaps fraud); and (3) a relational duty to each member of the public to provide truthful information in anticipation of public regulation of the product. The third duty is grounded on a right of each person to have an “unpolluted” information environment similar to the ancient right of each person to unfettered transit on the public roads and free and unpolluted waterways. Concerning this last duty, each person’s right is non-rivalrous, and the remedy for the invasion of the right would be nonexcludable.
But framed in this way, Dana’s public nuisance covers much less than, for example, has been claimed in the major opioid cases around the country. If the interest invaded is the right of each person in (for example) Oklahoma that both state and federal regulators receive information that is not false and misleading, then the plaintiffs’ case for liability and damages looks different than has been claimed in court.
For example, there may be a major difference between the claims against opioid manufacturers, like Purdue and Endo, and opioid distributors like McKesson, and opioid retailers, like Wal-Mart. The 2020 settlement with the Justice Department includes allegations that Purdue misrepresented to federal and state regulators the true purpose of its Abuse and Diversion Detection program, which was to push back against physicians’ concerns about addiction in their patients and thus encourage unnecessary prescriptions. (See Settlement Agreement, October 21, 2020, Addendum A, Section I.) In Alabama’s public nuisance suit, the claim against distributors is quite different: Since McKesson (a distributor) was “required by law to ensure that . . . opioids would not be diverted for illicit purposes,” Alabama claimed that it breached a public law duty “to report, investigate, and halt suspicious orders.” An Alabaman’s claim that McKesson breached its relational duty to take reasonable care to monitor orders of opioids faces significant causation problems, as well, as in the tobacco litigation, the defense of plaintiff fault. But even if these problems could be overcome, there is no colorable violation of public right. While the failure to assist Alabama in policing illegal prescriptions may have been a violation of public law, it is hard to see why it was an unreasonable interference with a right owed to every person in Alabama, unless we are to retreat to the position that every person in Alabama has a right that every person in Alabama fulfill their obligations to the state of Alabama.
If limited only to a manufacturer, framing the public nuisance in terms of a duty to provide uncorrupted information about opioids reveals the difficulty in proving liability. The trial judge in the California case focused on this hurdle when he found for the defendants. He stressed that the plaintiffs, four counties, ignored their burden at trial of proving causation. Even if it were true, as seems obvious, that the false and misleading marketing described above occurred, liability could only attach if the plaintiffs proved that the breach of the public’s right to uncorrupted information caused members of the public to suffer as a result of the absence of accurate information.
The critical problem for the plaintiffs was that they had the burden of proving that that “medically inappropriate prescriptions” were caused by the violation of the public right.3 The trial judge held that, unlike lead paint, which had no appropriate consumer use once its defect was known, opioids continued to have appropriate consumer use during the period the plaintiffs claimed the public right was violated. (The federal government and the State of California approved and encouraged the use of opioids.) Unlike tobacco, where, arguably, it might have been the case that but for the violation of the public right to information about cigarettes, tobacco use would have been dramatically limited, the claim about the causal relationship between the manufacturer’s violation of public right and liability is highly attenuated, and that proved quite problematic given that the plaintiffs offered virtually no evidence on causation.
Dana’s article is valuable at two levels. First, it illustrates why some have invested into public nuisance their hopes that tort law can effectively tackle seemingly intractable crises such as climate change and the creation of markets for harmful but popular products like guns and opioids. The hope is that the private law remedy of public nuisance can be as broad as the scope of administrative law. Although I have raised concerns that this strategy broadens public nuisance too greatly, Dana helps us see clearly what is at stake. Second, although not his primary purpose, Dana provides a roadmap for describing a modern public right that the seller of a product must provide “unpolluted” information to the public. For those of us who seek to use tort law to protect rights, this is something to take seriously.
- Sometimes the content of a simple duty and a relational duty are coextensive. “Don’t trespass” can be both a simple directive (criminal trespass) and a relational directive (the tort of trespass).
- 14 Cal. 4th 1090, 60 Cal. Rptr. 2d 277, 929 P.2d 596 (1997)
- See People of the State of California v. Purdue Pharma LP et al., case number 2014-00725287 (Nov. 1, 2021), at 15.