Photo of Eric Alexander

Those of us who took Con Law as first year law students may recall Marbury v. Madison as an early test of the Supreme Court’s place in our nascent republic.  Alliteration being a mnemonic device, some may recall that Madison was Secretary of State James Madison and the decision was written by Chief Justice John Marshall, both Founding Fathers and fairly famous fellows.  The other M was petitioner William Marbury, an erstwhile Justice of the Peace in recently created D.C.  Fewer will remember that the Supreme Court helped cement its role as an independent branch of government by denying jurisdiction through application of the political question doctrine in this seminal case.  Fewer still will actually remember the details of the case and decision.  We need not delve into those details for the purposes of this post, but we will pose one not-so-rhetorical question:  Isn’t just about everything at issue in just about every lawsuit a political question to some extent?

This reminds us of a scene from Fiddler on the Roof, a musical turned movie based on a collection of short stories by Shalom Aleichem that was set in a fictional town in what is now part of Ukraine (and is now the name of a real town founded to house refugees from the Russian invasion of Ukraine).  We will set aside discussion of the relevance of this tale in light of current events, but there is another connection to Marbury v. Madison:  the Ukrainian embassy in Washington, D.C., is located in a building once owed by William Marbury.  In any event, in the scene, an ardent socialist proposes to one of the titular character’s daughters:

Perchik:  There’s a question … A certain question I want to discuss with you.

Hodel: Yes?

Perchik:  It’s a political question.

Hodel:  What is it?

Perchik:  The question of … marriage.

Hodel:  Is that a … political question?

Perchik:  Well, yes. Yes, everything’s political. Like everything else, the relationship between a man and a woman has a socioeconomic base. Marriage must be founded on mutual beliefs. A common attitude and philosophy towards society…

Hodel:  And affection?

Perchik:  Well, yes, of course. That is also necessary. Such a relationship can have positive social values. When two people face the world with unity and solidarity …

Hodel:  And affection?

Perchik:  Yes, that is an important element! At any rate, I … I personally am in favor of such a socioeconomic relationship.

In a sense, a proposal of marriage is not a political question.  In another, it sure is, especially if you consider the number of state laws governing and referencing marriage.  Among the line of Supreme Court decisions on substantive due process that the overturning of Roe by Dobbs cast in doubt, multiple relate to state laws on marriage that undoubtedly reflected legislative answers to political questions.  Griswold related to Connecticut laws on the use of hormonal contraception by a married woman.  Loving related to Virginia’s laws on interracial marriage.  Bowers v. Hardwick related to Georgia’s disparate application of sodomy laws to married heterosexual and unmarried homosexual couples.  We could go on.

What about product liability claims?  Clearly, many political decisions are inherent in the laws that affect product liability claims.  For the sort of litigation involving medical products that we do, and this Blog discusses, decisions made by FDA are often at issue.  Similarly, decisions of other agencies feature prominently in litigation involving other kinds of products, such as decisions by EPA in connection with pesticides and herbicides.  The reality is that, shifting back to our kind of cases, one side—usually the plaintiff—is typically second-guessing some decisions made by FDA, whether product-specific or broader.  If those agency decisions involve answers to political questions, then should not the federal court apply Marbury v. Madison to refuse to allow a party to premise its relief on second-guessing the correctness of the agency’s answer to a political question, as doing so would deprive the court of subject matter jurisdiction?  If, as Perchik maintained, “everything is political,” then this should come up all the time.  But it does not.  Indeed, the decision in Caston v. F. Hoffman-La Roche, Inc., No. 23-cv-0092-TLT, 2024 WL 1548649 (N.D. Cal. Apr. 8, 2024), is the only decision we can recall that applied the doctrine to kick a product liability claim as to a prescription drug.  So why did the court in Caston decline jurisdiction over certain asserted product liability claims and how broadly applicable will this decision be?

As a preliminary matter, Caston was essentially the re-filing of a prior case called Nelson (which we discussed here).  The details of the claims asserted by the different plaintiffs and against the different defendants in Nelson and Caston would require some serious unpacking, which we will not do here.  Beyond the political question analysis, Caston is commendable for getting it right on personal jurisdiction and the preemption of warnings and misrepresentations claims against generic manufacturers.  On the former, there was no general jurisdiction over New Jersey entities because a related California entity’s contacts could not be imputed and there was no specific jurisdiction based on entering a distribution agreement with the Department of Defense that covered California bases along with all other U.S. military locations everywhere else.  Id. at *9-12.  On the latter, the plaintiffs who took generic drugs could not plead their way around the duty of sameness for labeling and the resultant impossibility conflict preemption.  Id. at *12.  For design defect claims, as to both the branded and generic drugs, the court did not analyze preemption.

Boiled down, those design claims were that mefloquine, whether branded or generic, allegedly had undue risks of a wide range of neurotoxic and psychiatric problems and that the use by plaintiffs in the 1990s or 2000s allegedly caused them to develop a wide range of neurotoxic and psychiatric problems.  (Statute of limitations would seem to be another huge issue for the plaintiffs because their alleged injuries were long before they initiated suit and the risks of this class of drugs was allegedly well-known for decades before they used the drugs.)  The commonality is that the mefloquine was prescribed to each plaintiff as an anti-malarial agent by military providers in connection with deployments to areas with high risks of malaria, pursuant to established military policy.

The defendants moved to dismiss on both the political question doctrine and the government contractor defense.  Looking at the motions to dismiss in Caston and Nelson, which was adopted by reference, the two arguments were paired together and the relief sought under the political question doctrine was the dismissal of the entire case for lack of subject matter jurisdiction.  The arguments were also focused on the issue of second-guessing the judgment of the U.S. military, with limited discussion of FDA decisions in connection with the government contractor defense only.  The Caston court, however, did not address the government contractor defense and focused its analysis of the political question doctrine on FDA more than on the U.S. military.  With assistance from our crack associate Avery Holloman, we checked to see if our instinct on the novelty of the conclusion reached in Caston was correct.  It seems to have been.  While the government contractor defense comes up relatively often in product liability and toxic tort cases, the few published decisions that analyzed the political question doctrine in product liability cases rejected it.  In In re Methyl Tertiary Butyl Ether (MBTE) Prods. Liab. Litig., 438 F. Supp. 2d 291, 300 (S.D.N.Y. 2006), the MDL court followed the general rule that “[e]ven when products are heavily regulated under federal law, tort suits involving those products may be brought absent a congressional injunction prohibiting such suits.” 

Defendants’ arguments boil down to the claim that holding manufacturers, refiners, and sellers responsible for MTBE contamination is highly controversial and thus should be left to the Congress, the EPA, and the President. But, the fact that the issues arise in a “politically charged context” does not convert this tort suit into a non-justiciable political question, given that there is no evidence that Congress has decided that it would resolve the issues. While regulation of the national fuel supply is surely not an issue for the judicial branch, these suits seek abatement and damages in addition to a ban on further contamination. Weighing the issues in a products liability claim is a quintessential judicial function.

Id. at 304.  In Lofgren v. Polaris Indus. Inc., 509 F. Supp. 3d 1009 (M.D. Tenn. 2020), which concerned product liability claims over an ATV accident in connection with Army training, the defendant raised the political question doctrine.  However, the record indicated that “the military was not controlling Defendant’s actions, but instead making mere suggestions and requests for certain design features. Defendant has repeatedly characterized the interactions with the military as a ‘collaboration,’ and the military as having ‘requests.’”  Id. at 1026.

Caston applied the same test from Baker v. Carr, 369 U.S. 186 (1962), as in Lofgren to reach a different conclusion.  It was also different than the same court had reached in Nelson.  As it explained, the difference was that the design defect claims in Caston “effectively challenge the safety and efficacy of active ingredients [and] implicate nonjusticiable political questions outside the scope of the Court’s Article III jurisdiction.”  2024 WL 1548649, *4 n.2.  Of course, FDA had made those decisions initially, even if military policy and individual prescribing decisions for each plaintiff agreed with them.

Pharmaceutical design defect claims challenging active ingredients necessarily implicate a nonjusticiable political question under the appropriate jurisdiction of the FDA because the FDA is a highly technical, scientific, and medical agency whose mission is to safeguard the public health by ensuring safety and efficacy of human drugs.

Id. at *8 (citing FDA mission statement).  The Caston court considered itself “unfit to review” FDA’s decisions on safety and efficacy because it “lacks the scientific and clinical expertise of the FDA.”  Id.  FDA’s decisions are “multifaceted,” considering things like public health and the “demand for a particular therapy at a given point in time,” rendering those decisions “heavily scientific, and indeed economic and political, but not judicial.”  Id. 

If adjudicated here, the Court would need to consult the broader scientific literature, and the content submitted to the FDA, and make its own determination of whether the studies submitted in the NDA were enough to warrant approval in 1989, a question designated for the FDA. The FDA has exclusive jurisdiction over the approval of New Drugs, including the designs of those drugs, pursuant to the FDCA.

Id.  “[A] contrary ruling would potentially call into question the FDA’s credibility. It would be inappropriate for the court to obfuscate a decision made more than thirty years ago by a federal government agency.”  Id. (citation omitted).

On first and perhaps second blush, the Caston court’s analysis of the nonjusticiability of design defect claims for an approved drug should apply broadly.  We have been saying for a long time that design claims for prescription drugs are almost always unsupportable and preempted, like here, here, and here, in part because of some of the same indisputable observations from Caston.  Yet, the political question doctrine has not been successfully applied to those claims before, at least in any published decision we could find.  The difference may rest on the timing.  The NDA held by one of the defendants was approved 1989 and the first ANDA was approved in 2002.  (It even appears that “U.S. Army Walter Reed” held the first NDA for the drug.  Dr. Reed earned his fame with yellow fever, but his namesake has apparently been working on malaria for quite some time.)  The plaintiffs’ use of the drug, which provides the relevant timeframe for analyzing design defect claims, was roughly 20-27 years ago.  Plaintiffs contended medical literature since the 1940s was relevant to the risks of the drug.  In short, proceeding with the case would involve quite a long look back in time.  Avoiding such a squinting look through the retrospectoscope is part of the reasoning for having statutes of limitations and repose, but the timing in Caston would clearly also make any second-guessing of FDA inherent in adjudicating plaintiffs’ design defect claims much more difficult and less reliable.  That may be why Caston was a good case to make new law.

The case is currently on appeal, so we expect to hear more on these issues, even if we do not see political question shots taken by other defendants with regulatory histories for their medical products going back decades and/or some military overlay in their fact pattern.  Litigation over those medical products certainly exists.  It will be interesting to see how this highest level of deference to governmental decision making–the court does not even have jurisdiction–plays out, especially at a time when the general issue of deference to governmental agencies is very much up in the air.