Ever wonder why oil companies throw money at researching climate change? Or why creationism was rebranded "intelligent design," and packaged as a scientific theory? It's because the American legal system lets lawmakers fudge their science, so long as they can claim there's "scientific uncertainty."
In areas as diverse as criminal punishment, environmental protection, and health care, legislatures are confronted with areas of scientific uncertainty. And when those laws are challenged, courts have to determine how that uncertainty affects the law. And they've decided that the presence of "uncertainty," in the form of any kind of disagreement among the scientific community, means that courts will let legislatures justify laws, under whatever scientific interpretation they want.
Note that while this post is all about U.S. law, it is not legal advice. Consult your attorney before trying to claim "scientific uncertainty" in court.
RBR – Rational Basis for Review
Let's start at the beginning. When a law's constitutionality is challenged, there are a number of standards that a court can use to evaluate it. These standards range from the extremely deferential “rational basis for review” (RBR) to the much more skeptical “strict scrutiny” standard. When the court applies RBR, it's unlikely that the law will be overturned.
Since the stricter standards tend to be reserved for things like sex and race discrimination, the vast majority of cases involving disputes over scientific uncertainty fall under the rational basis for review standard. Rational basis for review just requires a law be “rationally related” to a “legitimate government interest.” And the government is not actually required to present its interest — the court hearing the case can hypothesize one for them.
And “rationally related” basically just means that it not be tangentially related. Congress can get it wrong. Or they can choose to address only part of the problem. Or, as Justice John Paul Stevens, quoting Thurgood Marshall, put it in 2008: “The Constitution does not prohibit legislatures from enacting stupid laws.”
How Scientific Uncertainty Fits Under RBR
In the 1974 case Marshall v. United States, the Court applied RBR to a challenge to the Narcotic Addict Rehabilitation Act of 1966. It determined that the law's exclusion from rehab for people with two or more convictions was rationally related to a goal of providing services to those more likely to actually benefit from them. In its RBR analysis, the Court stated:
“When Congress undertakes to act in areas fraught with medical and scientific uncertainties, legislative options must be especially broad.”
Basically, the Court not only applied a deferential standard, it took pains to point out that deference was particularly appropriate when laws tackle issues of science and medicine.
What Scientific Challenges to Laws Look Like
Challenges to laws based on science come in two flavors: where the law is using a scientific term incorrectly, and where the law is actually justified by science. Generally speaking, case law prefers to leave science in the hands of legislators, and not judges. As a result, both scenarios get the benefit of the doubt in court.
Kansas v. Hendricks (1997) typifies the cases where legislators redefine a term with a scientific meaning. Kansas had a statute allowing for involuntary commitment for "mental abnormality" rather than "mental illness." Prior case law supported "mental illness" as a reason for commitment, but not "mental abnormality." The Court rejected the argument that this invalidated the law, because the Court had "traditionally left to legislators the task of defining terms of a medical nature that have legal significance."
Scientific justifications for laws appeared as early as 1905, in Jacobson v. Massachusetts, which revolved around a challenge to Massachusetts' compulsory vaccination law. While also finding that the states' police power can be used to enforce the common welfare over individual freedom, the Court also held that states can choose among competing theories when crafting laws. The uncertainty raised by the plaintiff could not stop the legislature from acting.
Of course, this question comes up again and again in the context of some of the most contentious political issues, like abortion, global warming, and creationism.
Why the Court Defers
There are a few good reasons for deference to lawmakers when questions of science arise. Marshall pinned it to a danger in assuming that judges are wiser than legislators, and letting them rewrite laws. There’s also the problem of courts preventing government intervention in areas that need it, just because the science isn’t settled.
That latter issue came up in Jacobson v. Massachusetts in 1905, where the pressing public-health need allowed the state to choose between competing theories on the utility of vaccines. (Massachusetts was in favor of a smallpox vaccine, by the way.)
The theory behind this is that legislatures have the resources to gather more scientific information than a court. This is balanced by a fear that courts making determinations about science could fix certain scientific ideas into the law, even if they're later disproved.
How Do Courts Define "Scientific Uncertainty"?
Because this is the Supreme Court we're talking about, they've never actually defined "uncertainty." But when courts say “scientific uncertainty,” they can mean where the science is unresolved for whatever reason — but they usually mean where experts disagree on the science of a particular point. They are usually referring to "controversy." Here's a sampling of some language the Supreme Court has used:
- Lambert v. Yellowley (1926) upheld a Prohibition law restricting the ability of doctors to prescribe alcoholic medication because of “high medical authority being in conflict as to the medicinal value of spirituous and vinous liquors taken as a beverage.”
- Powell v. Texas (1968) determined that Texas's "public intoxication" law did not criminalize a disease where there was no “agreement among members of the medical profession about what it means to say that ‘alcoholism’ is a ‘disease.’”
- Stenberg v. Carhart (2000) found problems with Nebraska's partial birth abortion ban because the “division here involves highly qualified knowledgeable experts on both sides of the issue.”
- Gonzales v. Carhart (2007) partly used “documented medical disagreement” to uphold the Partial–Birth Abortion Ban Act of 2003, even though it was not “uncritical” of Congress’ findings.
Problems With the Legal Use of "Scientific Uncertainty"
However, the Court haven't actually created a framework for lower courts to determine either whether uncertainty exists, or what level of uncertainty rises to the level of requiring deference. As a result, there's no easy way to distinguish between different types of uncertainty.
For example, scientists often focus on what has been called “knowledge uncertainty” – that is, uncertainties in the research itself. Stephanie Tai runs through many different kinds of this kind of uncertainty in Uncertainty About Uncertainty: The Impact of Judicial Decisions on Assessing Scientific Uncertainty.
Moreover, scientists may still see a theory as "unproven" or "uncertain" — while still thinking that the bulk of research supports it and that it will eventually be proven. But that's still enough for a court to conclude that uncertainty exists, and give legislators a lot of discretion to choose what they want to believe.
But most legislators and judges aren't scientists. They aren't going to get those nuances. This means that they can labor under the misapprehension that a "theory" is "uncertain," even if it has wide acceptance. Or mistake an expert's unwillingness to say that something is a "fact," when it is generally believed to be true, for uncertainty. As above, "uncertainty" can mean this kind, but most judges and legislators are going to mistake it for controversy.
And then there’s the problem of controversy being produced by biased research. Not that scientists are manipulating research, but funding can create pressures which will steer research toward the interests of its sponsors. For example, analyses of biomedical research show that studies funded by drug manufacturers result... in more studies favoring drug manufacturers. Shocker. Research also shows that the appearance of disagreement is greater when policy advocates have incentives to fund research supportive of a particular result. Also a giant surprise, right? Another fun result is that a minority scientific opinion that supports one side of an issue will receive funding and attention out of proportion to its acceptance within the scientific community.
Much of the time, biases are seen as being offset by the collective nature of scientific research, through peer review and the "norm of skepticism." But a system that encourages special interests to invest in advocacy science slows acceptance and paradigm shift.
How It Stands Right Now
In the end, courts don’t often inquire too far into the reasons for “scientific uncertainty,” and just determine whether it exists and should lead to greater deference to lawmakers. Therefore, successfully challenging a law based on science would require showing that the science is "certain," and is the opposite of the legislature's conclusion.
Which, good luck proving that a scientific theory or conclusion is "certain." Chances are, it'll only be when a court wants to strike down a government action. For example, the two Carhart cases involved similar abortion bans, but in one case, the Court decided that uncertainty made the law a problem and, in the later one, it didn't even care that some of the findings Congress based it on were flat-out wrong.
A lot of the time, though, it's going to go back to Jacobson v. Massachusetts, where a government interest allows the legislature to choose between the theories presented.
Cited throughout: Stephanie Tai, Uncertainty About Uncertainty: The Impact of Judicial Decisions on Assessing Scientific Uncertainty. (2008).
Charles Weiss, Expressing Scientific Uncertainty (2002)