I.

Suppose that you, an ordinary person, open your door and start choking on yellow smoke. You think “there should be less pollution”, and call up your representative.

The technical expert hears “there should be less pollution” and immediately has dozens of questions. Do you just want to do the obvious things, like lower the threshold for hexamethyldecawhatever? Or do you want to ban tetraethylpenta-whatever, which is vital for the baby formula food chain and would cause millions of babies to die if you banned it?

The expert has a point. Any pollution legislation must be made of specific policies. In some sense, it’s impossible to be “for” or “against” the broad concept of “reducing pollution”. Everyone would be against a bill that devastated the baby formula supply chain for no benefit. And everyone would support some magical bill that managed to clean the skies without any extra hardship on industry. In between, there are just a million different tradeoffs; some are good, others bad. So (the technocrat concludes), it’s incoherent to support “reducing pollution”. You can only support (or oppose) particular plans.

On the other hand, ordinary people should be able to say “I want to stop choking on yellow smoke every time I go outside” without having to learn the difference between hexamethyldeca-whatever and tetraethylpenta-whatever.

I think you’re supposed to imagine the environmentalists’ experts and the industries’ experts meeting policy-makers and hammering out some compromise, then moving one direction or another along the Pareto frontier based on how loudly normal people protest pollution.

But if you’ve been demanding an end to pollution for years, and nothing has happened, then it might be time to hit the books, learn about hexamethyldecawhatever, and make sure that what you’re demanding is possible, coherent, and doesn’t have so many tradeoffs that experts inevitably recoil as soon as they have to think about the specifics.

II.

I’m not a pollution expert, but I’m a psychiatrist, and I’ve been involved in the involuntary commitment process. So when people say “we should do something about the mentally ill”, I naturally tend towards thinking this is meaningless unless you specify what you want to do - something most of these people never get to.

Let’s start with a summary of the current process for dealing with disruptive mentally ill homeless people:

1. A police officer sees a mentally ill homeless person and assesses him as disruptive. Technically the officer should assess whether the person is “a danger to themselves or others”, but in practice it’s all vibes. They bring this person to the ER of a hospital with a psychiatric ward.

2. In the ER, psychiatrists evaluate the person. If some number of doctors, psychiatrists, and others (it varies on a state-by-state basis, and most people defer to the first psychiatrist anyway) agree the person is a “danger to themselves or others”, they can involuntarily commit them. Psychiatrists know lots of tricks for getting the evaluation result they want. For example, wasn’t the person brought in by the cops? Aren’t cops infamous for shooting mentally ill people? Sounds like whatever they did to attract the cops’ attention put them at risk of getting shot, which makes them a “danger to themselves or others”. Again, in reality this is all vibes.

3. The patient gets committed to the hospital. The hospital makes an appointment with a judge to legally evaluate the commitment order. But realistically the appointment is 4-14 days out (depending on the state), and by then the patient may well be gone anyway, in which case the hearing can be cancelled. If it does go to trial, the judge will always defer to the psychiatrists, because they’re experts trying to do a tough and socially important job, and the defendant is represented by an overworked public defender who has devoted 0.01 minutes of thought to this case. This is part of why everyone feels comfortable making commitment decisions on vibes.

4. If the patient seems psychotic, the doctors start them on antipsychotic drugs. These take about 2-4 weeks to make people less psychotic. But one of their side effects is sedation, that side effect kicks in right away, and heavily-sedated people seem less psychotic. So realistically the person will stop seeming psychotic right away.

5. After a few days, the hospital declares victory and discharges the patient with a prescription for antipsychotics and an appointment with an outpatient psychiatrist who can continue their treatment.

6. The patient stops taking the antipsychotics almost immediately. Sometimes this is because they’re having side effects. Other times, it’s because some trivial hiccup comes up in getting the prescription refilled, or in getting to the doctor’s appointment - nobody is good at dealing with healthcare bureaucracy, but semi-psychotic homeless people are even less good at this than usual.

7. Repeat steps 1-6 forever.

This isn’t going to win any of the people involved Doctor Of The Year awards. I’m sympathetic to attempts to change the system. But it’s hard to find the right point of leverage.

If your plan is to change the case law around involuntary commitment - to expand the definition of “dangerous to themselves or others” - it probably won’t matter, because most of these decisions are based on vibes that only loosely connect to the written law. But also, even if you find a way to make doctors commit many more people, it still won’t matter, because those people will stay in the hospital for a few days, then come out with antipsychotics which they will immediately stop taking.

If your plan is to change the law around guardianship and get all of these people state-sponsored guardians, that could help around the edges. But guardians can’t directly physically confine people or force drugs down their throats, so this won’t get them “off the streets” or “medicated” without additional steps.

If your plan is to “lock them up long-term”, keep in mind that (for now) there are almost no institutions equipped to do this. Each state usually has one center with a 3-digit number of beds for the most recalcitrant patients. Getting into these is like getting into Harvard, only in reverse - you need a spectacular anti-resume proving that you’re among the worst of the worst in the country. If you want tens of thousands of people in institutions like these, then you’ll need some kind of vast nationwide building program (do you expect San Francisco to be good at this?)

But okay, suppose you build those institutions. How long are you keeping people there? Remember, someone’s going to come in, start taking antipsychotics, and (if the drugs work) appear significantly saner within 2-4 weeks. Best-case scenario, they’re completely sane. Now what? Do you keep a completely sane person locked in the mental institution forever? Or do you let them out, at which point they will inevitably stop taking the drugs and become psychotic again?

Okay, sounds like you need some kind of social services to keep patients taking their drugs. There are lots of these, all around the country. The problem is, if the person isn’t taking the drugs (and sometimes even if they are), they won’t attend appointments with social workers. And the social workers can’t show up at their door, because these patients are homeless and hard to track down.

Okay, sounds like you need to get them homes. But there’s not enough government-subsidized housing. And anyway, now we’re back to Housing First, the solution that all of these “We Should Do Something About The Mentally Ill” articles treat as their foil.

Okay, then can you threaten people into attending appointments and taking drugs? There are Involuntary Outpatient Commitment orders, which say basically “go to your psychiatrist and follow their recommendations, or else we’ll put you in jail”. But remember, a lot of the problems happen when these people fail, through bad luck and bad executive function, to get refills. Typical cases might be:

• They lose their prescription and don’t know how to get another one. Or their plan is to call their insurance, insurance whines “these drugs cost $500 and you lost your last prescription too, we’re not paying”, and they don’t know what to do.

• Another homeless person steals their pills thinking that they might be opioids, and they’re going to grind them up and snort them and have the worst day of their lives.

• The pharmacy/insurance demands the doctor make some trivial irrelevant change to the prescription before they’ll accept it. The patient doesn’t know how to call their doctor, or doesn’t own a phone, or they contact their doctor but the doctor is sick of pharmacies/insurances demanding trivial things. In any case, nothing gets done.

• The patient messed up their appointment with the welfare bureaucracy that was supposed to give them a free subway pass, so they didn’t get one, so now they can’t make it to their doctor’s appointment.

• The patient went to their appointment with the welfare bureaucracy that was supposed to give them a free subway pass, but the welfare bureaucracy was inexplicably running three hours late and closed before seeing them, and the bureaucrats didn’t apologize or make an alternative appointment, so the patient didn’t get a subway pass and couldn’t make the appointment.

• The patient went to their appointment with the welfare bureaucracy that was supposed to give them a free subway pass, but they saw a drug dealer who had a grudge against them in the waiting room, and they left because they thought they’d get beaten up if he saw them.

• The patient was in the hospital with sepsis during their psychiatrist appointment, and nobody told them how to get an alternative psychiatry appointment.

• The patient wrote their appointment time on a piece of paper, which they left in their tent, which got flooded in a rainstorm and all their stuff was washed away.

• The patient is confused and sedated, which is a common side effect of antipsychotic drugs, and maybe if they made it to the doctor then the doctor could recommend something that would prevent this (realistically only the top 5% of doctors catering to the homeless will go this far), but they can’t make it to the doctor because they’re confused and sedated.

(Am I just being gullible? Are these people just making fake excuses? I don’t think so. I used to work at a clinic where visits were free with insurance, but missed appointments incurred a $200 fine that insurance didn’t cover. My low-income patients would miss a visit, then freak out because they couldn’t afford the $200 fine. Then I would talk to people, get permission to waive the fine once, and tell them okay, we won’t charge them this time, but they had to show up the next time. Then they would miss their next appointment too, their bill would go to collections, and I would never see them again. I asked the clinic director why we had such a punitive policy, and she said that without the policy, patients would miss so many appointments that the clinic couldn’t make money or pay staff.)

There are an infinite number of ways that semi-psychotic homeless people can miss appointments. The half-life of these people’s contact with the medical system is a month or two. So they’ll miss their appointment and get off the drugs. The police aren’t going to start a nationwide manhunt for a psychotic homeless person who’s indistinguishable from all the other psychotic homeless people. So realistically what will happen is they’ll be back on the street, a year later they’ll get arrested for some other reason, the police will notice they violated the treatment order, and the judge will try to add an extra year to their sentence for the treatment order violation. Then if their lawyer is really good, he’ll spend his 0.01 minutes on the case arguing that his patient has one of the excuses above, which will always be true. Then the judge will either give them a year in prison or not, and everyone will feel kind of dirty and ashamed of themselves either way.

(Also, even in the best case where you successfully treat somebody, I’m afraid that “1995 - 2024: psychotic homeless person” doesn’t look good on a resume, so they probably won’t be getting high-powered jobs. Meanwhile, cheap apartments in SF are $1000/month. So the connection between “no longer psychotic” and “no longer homeless on the street” is tenuous unless you also have some plan to provide free housing.)

Okay, then can you just make it a crime to be mentally ill, and throw everyone in prison? According to NIMH, 22.7% of Americans have a mental illness, so that’s a lot of prisoners. “You know what I mean, psychotic homeless people in tents!” Okay, fine, can you make homelessness a crime? As of last month, yes you can! But before doing this, consider:

• In San Francisco, the average wait time for a homeless shelter bed is 826 days. So people mostly don’t have the option to go to a homeless shelter. If you criminalize unsheltered homelessness, you’re criminalizing homelessness full stop; if someone can’t afford an apartment or hotel, they go to jail.

• Most (?) homeless people are only homeless for a few weeks, and 80% of homeless people are homeless for less than a year. If someone was going to be homeless for a week, and instead you imprison them for a year, you’re not doing them or society any favors.

• How long should prison sentences for homelessness be? Theft is a year, so if homelessness is more than that, it becomes rational for people to steal in order to make rent. And realistically it will take police years to arrest all of the tens of thousands of homeless people, so if a sentence is less than a year, then most homeless people will be on the street (and not in prison) most of the time, and you won’t get much homelessness reduction.

• What’s your plan for when homeless people finish their prison sentence? Release them back onto the street, then immediately arrest them again (since there’s no way they can suddenly generate a house while in prison)? Connect them to social services in some magical way such that the social service will give them a house within 24 hours of them getting out of prison? If such magical social services exist, wouldn’t it be cheaper and more humane to invoke them before putting someone in prison?

I admit that if you’re willing to be arbitrarily cruel and draconian (life sentence for someone and their entire family the moment the bank forecloses on their home!) you can make this one “work”. But anything less than that and it becomes just another confusing bad option.

In practice, the government tries some combination of these things, each of which works a little. Sometimes they fiddle with the law around inpatient commitment around the edges. Sometimes they give people free houses. Sometimes they threaten them with Involuntary Outpatient Commitment orders. Sometimes they throw them in prison. Most of these things work a little. Some of them could work better with more funding.

Nobody thinks the current system is perfect. I respect people who want to change it. But you’ve got to propose a specific change! Don’t just write yet another article saying “the damn liberals are soft on the mentally ill”.

The damn liberals are soft because some of them are the people who have to develop an alternative plan, and they can’t think of a good one. If you’re going to write yet another article like this, and you want to change minds, you should skip the one hundred paragraphs about the damn liberals, and go straight to the part where you explain how you plan to do better.

If your plan is “be arbitrarily cruel and draconian”, then that will work, but please admit it. Don’t mumble something about “I just want these poor people to be able to get the treatment they deserve yet don’t know how to ask for” before going back to railing against the damn liberals.

If your plan is something else that will solve everything with no tradeoffs, then you owe it to your readers to explain what that is.