Texas-Sized Problems
Mathematical edges to edgy laws
|
| Cropped from Lex Fridman podcast |
Scott Aaronson is one of the top researchers in complexity theory and especially in quantum computing. He is famous for more than his research, he is famous also for his blog Shtetl-Optimized. Its scope includes just about everything that touches on complexity theory, but includes many topics that span just about anything that is interesting these days.
For a whole week, he has had the topic: Exciting opportunities at Kabul University!
Scott is kidding about Kabul of course. Rather, his topic is the recent Texas law against abortion. Here are his initial comments:
Now, of course, Texas has effectively outlawed abortion—well, after the 6th week, which is before many women even realize they’re pregnant, and when the fetus is still the size of a grain of rice…
There are no exceptions for rape or incest, and—this is the “novel” part—there’s a bounty system, with $10,000 fines for anyone who helps in any way with an abortion, payable to anyone who snitches on them. Texas has openly defied Roe v. Wade and, for the first time in half a century, has gotten five Supreme Court justices (three appointed by Donald Trump) to go along with it. Roe v. Wade is de facto no longer the law of the United States.
We will not talk about abortion, but rather the novel aspect of the Texas law. This will not lead to quantum computing, not even along lines of a famous SMBC cartoon that Scott created with Zach Weinersmith, from which we give the second panel:
We promise it will lead to some math, however.
The Texas Law
What can we do to stop the Texas law? Many are worried that the fact that the US Supreme Court has refused to even temporarily block the Texas law is unheard of. Does this mean that the new Texas law somehow is valid? Does this mean that abortions will be made illegal? Not clear.
One part of the Texas law is clever—we can disagree with it and still note that the new law is clever. The law does not give the usual suspects to the local law enforcement, who are usually the ones responsible to enforce the law. Rather the law insists that anyone can raise the issue that X has violated the law against an abortion. Say S raises that X helped make an abortion happen. Then S gets $10,000 payment provided the case is upheld in court. Scott calls S the snitches.
As recognized by the Emergency Medical Network (Emnet), this step in moving the enforcement from the usual police to the general public is one of the reasons the Texas law is so hard to fight. The usual way to challenge a new law is to immediately file some counter suit stopping those charged with enforcement. This seems to be impossible with this trick. Who can be S?—by construction that could be just about anyone. There is no single body to force to defend the law in court. Clever.
The Texas Law—The Trick
We are experts in theory of computation. Can we use that ability to find a way to attack the new law? Can we see any weakness that is inherent in the law? Some trick? Or is the Texas idea somehow a breakthrough in ways to state a law against abortion?
Ken and I have an insight. A standard trick in theory is given a new idea can we generalize it. If we can use the new idea to solve other problems then we should shed some light on the new idea. If the idea is too powerful, then we essentially prove that it cannot work in general. If it does work elsewhere, then we can use it to solve previous unsolved problems.
The Texas Law for Speeding
Cars speeding is a serious problem just about everywhere. I live in a rural area and it is 35 mph here.
Yet when I am driving I am often passed by cars that cross a yellow line going 50 or faster. The issue is that the laws insist that some officer witness a violation. But since that rarely happens, the speeders almost always get no ticket.
Let’s use the same trick that is used in the new Texas law. If a car races by me then I will send a digital message to the police that says: “Today I was driving my car number 1236335356 at time 11:19.42am and witnessed a car with plate number ERRW5365 speeding at 55.11mph while passing me. I was at location …
This would be sent to the police and they would then have the task of checking it out. Provided it checks out, I would be awarded a bounty of $500, say. And the speeder would get a fine and …
How could I prove the speed? I wouldn’t have time to take a video in the moment. But Ken’s smartphone—and those of many of his colleagues—have an app that is solving problems of this type 24/7. His is a University at Buffalo product but there are many similar ones. Quoting the website makes everything quickly clear:
PocketCare S is a bluetooth low energy (BLE) solution which enables smartphones to send and receive anonymous beacon signals. It checks the distance between a smartphone and another beacon (or smartphone running PocketCare S) … [and] … records the duration a close encounter with another beacon. PocketCare S is designed to report social distance information without collecting or revealing any personally identifiable information about any specific individual.
The car would only need to have a detectable beacon, not the same app—and the info giving proof of speed over said duration might already be in the authorities’ data banks even before the complaint is filed.
Bipartite Epidemiology?
Of course, what we are trying to say is that this kind of lawsuit would become a new kind of virus. It might extend to many other transgressions besides speeding. What strikes us further is that this would assume a binary structure that we do not know to exist in medical pandemics.
The first point is that the people who voted for the Texas law believe they are not the ones who would be violating it. Call them Team A. The remaining people—Team 2—would want to get back at them. Maybe speeding is an equal vice for those two teams, so that wouldn’t work. We’ve seen suggestions like Team B suing for open-carry of firearms. Let us just suppose Team B finds something.
The second point is that a “snitch” lawsuit is like one person infecting another. But unlike our present pandemic, it’s one where people in A can only infect those in B, and vice versa. We have a bipartite graph 
One issue noted by Scott in a comment is whether the law allows someone to be sued multiple times. If not, then being sued once would be like acquiring immunity. The US recognizes principles of “no double jeopardy” but the Texas law appears to avoid them.
We’ve tried to look for related analysis of the popular college “Humans Versus Zombies” (HvZ) game. But in that game, the zombies only increase. Scientific American once published an analysis titled, “Modelling a werewolf epidemic” that also mentioned zombies. But it did not say the werewolves could only infect zombies and vice-versa.
Binary infection situations do of course exist when two groups of people suddenly mix, each having already acquired resistance to a disease that infects the other. There have been great tragedies when the mixing was a relatively small number of A colonizing into B. But we still do not know of models that would apply when the kinds of “infections” involve spontaneously arise en masse in the manner of a pandemic with high 
Open Problems
Is this type of snitching a viable adjunct to law enforcement for speeding? Could it save lives? What about the associated dimension of propagating personal information into massive repositories?
As I suggested on Scott’s blog, what we’ve got here is a fantastic new legal mechanism for making end-runs around the Constitution … just for one example …
Citizens have a constitutional right to keep and bear arms, or so we’re told.
But any State so inclined can now pass a law enabling private citizens to sue anyone who keeps and bears arms or aids and abets another person doing so.
Consider the posse-bilities …
The “bounty hunter” element of Texas’ new law concerns me. There is a federal law (HIPAA) that protects every American’s medical privacy. Would Texas’ anti-abortion law grant bounty hunters the authority to search any individual’s medical record?
2.
A lot of countries have speed cameras but in some countries, the speed cameras are required by law to be visible and warning signs posted. This leads to a funny situation where people slow down before the camera but speed up after the pass the camera zone.
The Netherlands seems to have solved this problem. They take the average speed approach:
Instead of using one camera, they use two separated by a fixed distance. Each time you pass a camera, it timestamps you. After you pass the second camera, it calculates your average speed and issue you a ticket if you go over the speed limit. For example, the cameras are 1,000m apart and the speed limit is 130km/h and the tolerance is 5km/h. Say you pass the first camera at time t = 0s and the second camera at time t = 10s, then the average speed is 100m/s = 360km/s…. In this case, you’d probably get your license revoked!
But if you speed in the beginning and slow down before the camera… you kinda defeat the purpose of speeding…
100km/s = 360km/h***
My guess is that the law will only get challenged in court once 1 or more people actually tries to collect the $10K, or somebody actually gets arrested for an illegal abortion. At that point you’ll have a clear, specific legal case for the courts to hear.
Double jeopardy would mean that EITHER only the “first snitch” wins all $10K OR “all snitches” for a specific defendant must split it, because the same defendant cannot be brought to trial more than once for the same crime (in this case an abortion) & payout of the $10K bounty is contingent upon a conviction for that crime. Odds are the courts will strike down the bounty if it isn’t very clearly and carefully explained exactly what happens in all these various situations, even if the larger law is allowed to stand. It might down to a legal test of “severability”- i.e. whether the judge(s) think the bounty part can be removed without massively altering the rest of the statute.
As for the law itself, if it does actually lead to a de facto or de jure overturn of Roe v. Wade, the immediate follow-up will be to argue that most abortions are legally equivalent to murder and should be therefore be outlawed. The legal burden would then shift from states trying to limit abortion to states explicitly permitting it, as states like California would almost certainly pass their own laws to explicitly legalize most abortions. I doubt the current SCOTUS could or would be willing to overturn Roe v. Wade and outlaw most abortions nationwide within the same ruling, but it isn’t impossible.