During last week’s oral arguments over the Biden administration’s vaccine mandate for private companies, Justice Sonia Sotomayor claimed that over 100,000 children are in “serious condition” with covid-19. Many, including The Post’s Glenn Kessler, pointed out that while omicron does seem to affect children more than other variants of the coronavirus, Sotomayor’s figure was wrong by a factor of about 20.

It is, of course, a big deal for a justice to make such an inaccurate claim during oral arguments (though some critics have shamefully revived bigoted nonsense from when she was nominated — that she’s dumb or was an “affirmative action pick”). I won’t pretend to know why Sotomayor made the claim. Perhaps she misspoke or misread some data. Or perhaps, as more partisan critics seem to think, she deliberately cited a fake statistic to defend the Biden administration’s vaccination policy.

But for all the furor over Sotomayor, her mistake came during oral arguments, not a written opinion. And that brings us to a bigger problem far more consequential but much less discussed: Supreme Court justices have a history of making factual errors in written opinions for which they have ample time to research and fact-check. Some of these errors have had sweeping consequences for constitutional rights. And the court has never bothered to correct them.

The court has repeatedly held, for example, that an alert from a drug-sniffing dog constitutes probable cause for a more extensive search. Over the years, the justices have been presented with statistics showing that drug dogs’ alerts are often no more accurate than a coin flip. The dogs’ sense of smell isn’t the problem. It’s that the animals have been trained to read their handlers’ body language, and they’re rewarded for confirming their handlers’ suspicions.

Despite multiple opportunities to address the issue, the justices have doubled down. In a 2013 unanimous opinion, Justice Elena Kagan wrote that fears about false positives are misguided because most dogs are certified by accrediting organizations. A ProPublica investigation found that certification groups don’t test drug dogs for false positives.

Because the court refuses to consider the possibility that it may have been wrong, drug dogs have become “probable cause on a leash,” making it easier for police to conduct roadside searches on the most rudimentary of hunches — exactly what the Fourth Amendment is supposed to prevent.

Justice Anthony M. Kennedy’s 2002 mistake about sex offenders and recidivism was even more catastrophic. Kennedy cited a pop psychology magazine’s claim that 80 percent of sex offenders go on to reoffend. The opinion has since been cited by lower courts more than 90 times to justify increasingly draconian punishments for offenders, including lifetime registries, residency restrictions that have left people living under bridges, and even indefinite detention after they’ve served their sentences.

In fact, there was little evidence for the 80 percent figure at the time. Research has since shown the rate to be somewhere between 8 and 35 percent. (And the very types of punishment Kennedy’s opinion has unleashed tend to make recidivism more likely, not less.) The court has since had multiple opportunities to correct Kennedy’s mistake. It hasn’t.

There are more examples:

  • The court continues to rely on a study from 1963 (!) about police officers killed during traffic stops. That study was plagued by sampling issues, and it co-mingled traffic stop data with other enforcement stops. But despite other data showing violence during traffic stops is very rare, the court has continued to cite this almost 60-year-old study, perpetuating the myth that traffic stops are inherently dangerous.
  • In a 2011 ruling, Justice Clarence Thomas claimed professional discipline by bar associations is sufficient to deter prosecutorial misconduct. The overwhelming evidence is that even brazen misconduct is rarely punished at all.
  • In a 2006 majority opinion, Justice Antonin Scalia argued that evidence obtained after an illegal no-knock raid should not be thrown out, in no small part because a “new professionalism” had taken root in policing that would deter officers from committing such violations in the future. One of the criminologists Scalia cited in his opinion wrote that the justice had badly mischaracterized his research. In the 16 years since, we’ve seen routine and systematic abuse of no-knock raids.
  • In a 2011 opinion, Justice Samuel A. Alito Jr. wrote that 88 percent of private-sector employers conduct background checks on potential hires. That was false. Alito had pulled it from an amicus brief filed by a consortium of people and companies who perform background checks.
  • In 2012, Kennedy wrote an opinion upholding a portion of Arizona law that requires police to verify the citizenship of anyone they detain. Kennedy cited a statement from an anti-immigration group that undocumented immigrants commit a disproportionate number of felonies. That claim wasn’t true at the time, and it has been thoroughly refuted since.

The irony here is that Sotomayor has often been a lonely but reliable voice for victims of police and prosecutorial abuse, the very people most often harmed by these misstatements of fact in written opinions. (Though she also joined in two of the opinions above.)

If Sotomayor’s claim about children and covid makes it into a written opinion, she’ll deserve all the scorn she’s getting. Until then, the charitable view is that she simply misspoke or was misinformed. That’s regrettable but also not uncommon in oral discourse. We should be far more concerned about the times justices have made provably false claims in written opinions — claims that have had profound consequences — and then refused to correct them.