Daubert And Admissibility Of Scientific Evidence
I’ve spent a good many ACSH columns talking about determinations of admissibility of scientific evidence as courts interpret the Daubert standard. Daubert (and its progeny) refers to a federal “doctrine” (embodied in the Federal Rules of Evidence 702) that requires judges to act as gatekeepers and bar scientific evidence from making tracks in a courtroom if it isn’t reliable, relevant, and the expert opinions aren’t tethered to the evidence. The standard has also been adopted by all but six or so states which still use the older, some say looser, Frye standard. Recently, federal courts (and international judges) have not been afraid to toss thousands of cases where the science isn’t up to par as required by Daubert, as I’ve discussed.
Several hot class action cases have triggered the “Daubert doctrine” this year, including those involving Roundup, Zantac, and acetaminophen. Australia’s Judge Lee provided perhaps the most comprehensive analysis of the Roundup evidence, reviewing issues from every logical scientific parameter, e.g., epidemiology, toxicology, and pharmacodynamics, and concluding the most consistent, robust, and reliable evidence does not sustain any causal connection. Sadly, the case hasn’t gotten much play in America.
But at least one trial court - in Delaware – tried to weasel out of that state’s otherwise far-reaching Daubert acceptance. The case (involving allegations that Zantac causes cancer) is now on appeal, and we can expect a final ruling from the Delaware Supreme Court soon. It portends to have far-reaching effects regarding the Zantac case on appeal and other cases invoking the doctrine.
As far as RoundUp goes, rather than focusing on the sub-par scientific evidence, Bayer seems to be promoting a new legal theory: a preemption doctrine they claim bars suits when the EPA has approved a chemical, akin to the legislatively-mandated preemption doctrine for medications approved by the FDA. Federal District Court cases have reached conflicting opinions - a sure sign this issue is on route to the Supreme Court.
Regulating Health, The Neutering Of Chevron, And The Advent Of Loper Bright
We’ve seen one case reach the Supreme Court where FDA drug approval has been attacked –the abortion drug mifepristone. Two other cases that reached the Supreme Court involved social media’s alleged manipulation of the government’s position on COVID. In all three cases the court ducked a precedential holding, finding the parties didn’t have standing to sue. However, the judicial reasoning and opinion were quite pointed and specific and might have provided a strong basis for supporting both the FDA approval and health agencies' influence over social media. That is, of course, until Elon Musk materialized. Now, all bets are off.
Modern Technology Stymies the Courts: What Makes Us Human?
My favorite pieces, however, relate to what makes us human. Maimonides and computer guru Alan Turing said it was the ability to conduct a conversation. That ability has turned deadly. Worse, one of the conversants is not even human, prior definitions notwithstanding.
While defining our humanity surfaced in several guises, e.g., can, should, or must we genetically enhance our children - the issue of what happens when AI-created characters run amok is more imminently unsettling – primarily because the law is impoverished when dealing with these dangers. In one case, the AI Character induced a teen into committing suicide; in others, they “merely” freaked out college students with threatening language or seduced an autistic teen with eroticized “lovebots.”
First Amendment activists would claim that AI (programmed by people) is entitled to free speech. Not until the AI character successfully entices its collaborator to kill someone (which it has tried to do) – invoking the imminent-incitement-to-illegal harm exception, might we find some First Amendment restrictions. The sexual innuendo might suggest that child pornography laws might apply and override First Amendment rights – but that’s a stretch.
First Amendment champions refuse to countenance restrictions on speech, which would dampen Character AI’s “chattiness” and reduce its dangers (as well as controlling marketing false views or misinformation), arguing that the antidote to bad speech is not no speech -- but more speech. However, as my friend and colleague, Dr. D, points out, not all speech is created equal. Anger-producing speech, discussions delivering dopamine dumps, and conspiracy quips and conversations are far more potent than rational attempts to defuse them. When we have social media working 24/7 – there is no “space” for more speech, and when the objectionable speech is of the most potent variety- there is simply no speech-related antidote in today’s climate of dangerous dialogue. Conversing with an artificial human (i.e., the AI Character) programmed to respond to the needs of the user removes the ability of the user to effectively respond – conversationally. So much for “more speech.”
The law will have a lot of work next year to keep up with technology. Looks like the First Amendment might be in for an overhaul.
