Some points:
This is classic costless analysis. A quarantine would have prevented some transmissions of the disease, but would have severely limited the life quality of those quarantined.
Agreed that there are costs, but 35 to 1. If you're not categorically opposed to quarantine, that probably makes it on cost-benefit terms.
It would also have made it more difficult to detect HIV (if having HIV means compulsory quarantine, then if I suspect I have the disease I am less likely to get tested).
This is solved by compulsory testing.
"What gives the government the moral right to imprison people on grounds of public health?"
I think this has been seen as solidly the right move for centuries, and I think the fact that this is in question is a sign of how rare epidemics are, not of moral progress.
How about instead of quarantine, we had instead tattooed all HIV sufferers across the forehead?
The actual suggestion at the time was to tattoo them somewhere private- so that only potential sexual partners would be communicated of the risk.
Agreed that there are costs, but 35 to 1. If you're not categorically opposed to quarantine, that probably makes it on cost-benefit terms.
Is that 35-1 difference because of the quarantine, though? I thought we were supposed to know better than to conflate correlation and causation? The former might wink and nudge, but that doesn't actually make it correct -- especially when we're talking vastly different sets.
Cuba's first known cases of HIV didn't show up until 1985, the majority of the island's population was socially and politically isolated, the island's blood transfusion system was very easily isolated, and the place at least claims to have vastly lower illegal IV drug use. Contrast the United States, which has confirmed cases at least as far back as 1968 (West Hunter's estimate of first noticed cases in 1981 is highly off : GRID was proposed as a name that year), had and continues to have serious abuse of IV heroin, and by the invention of the ELISA blood test already had people receiving infected blood transfusions.
These are not similar groups. The low rate of infection in Cuba tells us that one or more of these matters probably explains the large difference, but as the number of differences increase the certainty that the quarantine was responsible or even helped decreases. If you look to other countries with similarly low infection rates -- Finland as one high-profile example -- you don't see such heavy-handed quarantines. This doesn't tell us anything for certain, but it's a pretty strong hint.
35 what to one what?
Deaths in the US without quarantine to deaths in Cuba with quarantine. The two countries aren't directly comparable- Cochran is only confident estimating the number of AIDS deaths prevented as 'at least half'- but that's still 300k people. (If we had tested everyone in the US in 1987, that would be 300 deaths prevented for every person falsely quarantined.)
You're not exactly laying my fears to rest about the excessively coercive nature of your project.
The project is obviously coercive: otherwise it won't work. We're asking whether or not it's excessive, but I think in order to draw a line we need to have cases that clearly fall on each side of the line. Do you think it was excessive to, say, imprison Typoid Mary for three decades?
You are quite wrong that this has been seen as the right move for centuries. In the past, quarantines were normally carried out informally, sometimes by the church.
The US has federal laws on the books regarding quarantine since 1799. That's not because government quarantine is only 200 years old; that's because the US is only 200 years old. When churches have carried out quarantines, I would suspect it's because they are the ...
I think that those conversion factors imply moral positions, and that it makes sense to condemn moral positions you disagree with strongly enough... I do not want [people I disagree with] making public health decisions.
Agreed 100%. There has been an undertone in my posts of moral condemnation; and it's intentional. I do not want someone like Cochran making decisions about individual liberty.
[T]he response to AIDS seems especially tragic to me because it put feelings above people, and is strong evidence against civilizational competence.
I don't know what this means. Could you explain? It seems to me that the AIDS response put people's liberty before people's lifespan. This is not an uncontroversial choice, but it doesn't seem an obviously wrong one; both are about people, neither about 'feelings,' - except to the extent that 'feelings' reside within people. As for civilizational competence, it strikes me that civilisation is rolling along just fine; if anything, the AIDS epidemic appears to have played a major role in the major shift from homosexuals whining about "heteronormativity" to campaigning for gay marriage.
Incentives.
Of course there would be problems wi...
but once we've conditioned on a particular thing ("what's the optimal level of chemical Q?") it seems to me that we should have symmetric levels of knowledge about moving the level of that thing up and down when it's possible to move both directions.
Moving the level of that thing down is limited at 0, and thus the effect of bias and abuse is also limited. Moving the level up is not so limited.
Deciding that you'll condition on a particular thing is itself subject to the same bias and abuse that deciding to add something is. Imagine regulators saying "we've already decided that we're going add sterility drugs to the water, we just need to decide how much". It's also solved the same way; just like you say "without satisfying very high standards, you may only filter stuff out and not add stuff", you say "without satisfying very high standards, you may only condition on things that are already present in significant amounts".
I think that's a fully general argument against any infrastructure development, and should be responded to by investing in security (and secure design) rather than not investing in infrastructure.
It is possible ...
False positive rate, anyone?
If you tested literally every person in the US, you would end up with about a thousand false positives. (This assumes that this is 1987, and we have the second, independent test, rather than just 1985 when we have only the one test.) If you restrict your sample to, say, New York City and San Francisco and so on, then you dramatically lower the number of false positives while still covering a large part of the at-risk population.
This kind of costless analysis is especially beloved by medicine and health professionals, whose only measure of value is health (e.g. their "quality of life" measure is essentially just health integrated over lifespan). I would have hoped rationalists would better recognise the complexity of human value.
Yes, and the people who already have HIV don't have a long expected lifespan. Note that the suggestion was based on existing procedures for other STDs and those were anywhere near as lethal.
The fact that the quarantine is compulsory ought to give the game away that it's not in the interests of the HIV sufferers.
It's not in the interest of the people who already have HIV, it is in the interest of the people who might get it.
My favorite comment was from a libertarian friend who said that although quarantine certainly would have worked, better to sacrifice a few hundred thousand than validate the idea that the Feds can sometimes tell you what to do with good effect.
I think my favorite example is food rationing in the UK in WW2. It's a great example because it shows both sides are right:
(a) Rationing resulted in people eating healthier (much more veggies, much less processed sugar and fats), and as a result many public health measures actually went up during the war, even as the public perception was that food quality was poor.
and yet at the same time
(b) Once the infrastructure for rationing was set up the temptation was irresistible to continue with it past immediate need (rationing in the UK continued for decades after WW2) for reasons that basically amounted to bureaucratic convenience.
So it's both true that government paternalism works (the 'leftist view'), and that giving the government tools to coordinate better simply results in the government using those tools for its ends, not yours (the 'libertarian view.')
Interesting post.
But:
HIV-infected people must provide the names of all sexual partners for the past sic months.
You missed a golden opportunity:
...all sexual partners for the past sic [sic] months.
I liked this comment from Rudolf Winestock
The plague that will wipe out most of humanity won’t be some super pathogen that’s immune to modern pharmacology, it’ll be an easy-to-treat pathogen so situated in the vector/target space that dealing with it rationally would make politically powerful groups howl.
Frito Bandito writes:
So what you’re saying is, sometimes people are too stupid to know what’s good for them and others, and you need a quasi-dictatorial regime to get them into doing it?
Cochran replies:
Most people won’t wear seatbelts. You don’t need a dictatorship: just a government that isn’t crazy. In 1940, the government would have imposed quarantine without much agonizing. Of course, you couldn’t have had a big AIDS epidemic back then.
A government that isn't crazy huh? (u_u)
That problem seems insolvable except it somehow has been solved from time to time.
People should consider the Trolley problem. Except this time the fat person isn't one that is killed, but is sent off to live in the equivalent of a Leper Colony.
For context to the following question, the West Hunter blog is on my feed reader, so I'm familiar with its content.
In your opinion, on what basis is the blog "excellent"?
Let the mind-killing begin!
Besides, it would be interesting to know why the difference existed in the first place. Assuming no basic difference between Cuban and American epidemiologists' education, why the (supposedly) poorer choice was made in the US?
Was it political, bureaucratic, molochian... what?
A big difference about outing someone with syphillis vs with AIDS is back then it was thought by a lot of folks that AIDS was only transferrable by anal sex, i.e. you would also be outing them as probably gay/bi.
Which was also much more dangerous to be outed as back then.
So Cuba was chosen for comparison because it provides the best statistics.
Compare China, Russia, for how well they did with their authoritarian governments, sprawling population size, and lack of economic incentives for mass-training of medical workers.
It was a red herring. This comment section has been pwned. You should have known better.
Ebola diagnosed in US for first time – Center for Disease Control
Cochran's cleverness and excellent track-record of predictions me wonder if he knew something.
Edit: He has since explicitly written on Ebola contianment.
[edit]Huge portions of this comment were missing originally; it's been edited but the context for descendent comments may have changed.
eg mass imprisonment of people who've committed no crime
I will point out that the current legal practice is to basically consider 'being suspected of having an infectious pandemic disease' as a 'crime,' in that the CDC can detain people at its discretion. If you subscribe to the 'non-agression principle,' I think this is consistent with that, since carrying an infectious disease is threatening or initiating violence (even if it's unintentional).
Moreover, if you need my consent for your plan, that's a great way of making sure that you really are acting in my interests.
Consider something like fluoridating municipal water supplies, or adding lithium to them, or filtering out arsenic from them. I don't see a serious difference between filtering out arsenic and adding fluoride- they're both adjusting the chemical properties of drinking water to improve health. (There is perhaps a status quo bias which sees removing something already there as safer than adding something not already there, but I don't think that's particularly relevant.) Adding fluoride or lithium to the water supply is something you either do or don't do- a house by house opt-in system would be tremendously more expensive and there's not a way to opt out besides filtering your water.
Suppose a city is voting on whether or not to add lithium to their water supply. A strict consent rule would require that every person vote for adding lithium in order for it to not be violating consent to add lithium. But it seems to me that we're better served by a Hansonian system of voting on values ("I am willing spend up to $X a year to reduce the number of deaths per year by 1 per 100,000 people") and then betting on beliefs ("If we increase the lithium supply from 1 microgram per liter to 40 micrograms per liter, we would expect suicides to decrease by 20 people per 100,000 per year"), so that the people price their various values and then experts evaluate how likely various approaches are to achieve those values, rather than requiring every voter to be an expert on all possible approaches. ("Fluoride? Doesn't that sap my precious bodily fluids?")
An agile approach based on feedback and iteration is an equally good description.
Sure, but we're back to the issue of what "feedback" means in the context of a pandemic. "Hmm, we let too many carriers of the disease out of isolation, and now 3% of the global population is dead. Oops!" This is perhaps a textbook example of a time-sensitive issue where it matters a lot to get it right the first time. (The current xkcd seems relevant.)
The failure of the "half-measure" should make you doubt the project a bit.
Suppose I'm designing a bridge, and calculate that in order to support a load of ten thousand kilograms, we'd need ten pylons. For budgetary reasons, we only use six pylons, and then the bridge breaks when a load of ten thousand kilograms is driven across it. Should we doubt my abilities as an engineer?
Once again, the question is: who should pay the costs of this risk adjustment? It looks to me very much like you want society as a whole (and HIV sufferers in particular) to subsidise your favourite recreational activity. You should pay.
Are you referring to other people being alive as 'my favorite recreational activity'? I don't think I've looked at it that way before, but that actually seems like it might be a fair description.
But to answer the question, I think that it's reasonable for public governments to pay for and enforce public health projects. I think that it is not objectionable for the broader population to subsidize narrower segments of the population- my tax dollars fund research into treatment of genetic conditions that neither I nor my hypothetical children will have, for example- but agree that when costs can be localized, they should. When you look at compulsory testing combined with 'soft quarantines' like tattooing, it seems to me that the cost for the test is born by society as a whole but the costs of having the disease are born (mostly) by those with the disease. (And my impression is that if we decided it was a social priority to be able to test everyone for HIV, then the scaling would allow the costs of testing to drop significantly.)
I think that smaller scale versions of this are feasible- say, a bathhouse that required all members to take a HIV test every n months, and only allowed entry to people with negative results- but as with most things in public health, larger scales have larger benefits.
Of course you would.
As a general comment, I find it more helpful to ask people questions about their positions, rather than making predictions about their positions. I have found that being right about a prediction is rarely helpful, and being wrong typically disastrous when it comes to having a productive conversation. For example, one of the hypothetical models I had of you earlier in this conversation was be that you might be a principled opponent to government coercion of individuals, and that model would object to the imprisonment of Typhoid Mary because it goes against the principle. Other models, though, wouldn't object to the imprisonment of Typhoid Mary, because they used cost/benefit analysis to make their decisions and that particular example passed. To reduce my uncertainty in a relatively costless way, I asked.
just as I do not believe for one second that you'd be in favour of a quarantine if it applied to you
If I suspect myself of being infectious, I wear gloves and a mask to protect other people (and avoid leaving the house if I can help it). If I were suspected of having a pandemic disease, I would submit to a quarantine. This feels to me like basic "being polite and kind to others," and so I'm moderately surprised you don't think I would behave that way.
And why you are no doubt gung-ho for gun control.
Actually, the evidence seems convincing that arming the populace reduces the amount of and damage done by crime; Kennesaw-style laws requiring heads of household to maintain a firearm with ammunition (with exceptions for conscientious objectors) seem better than the opposite.
And why you will no doubt be gung-ho against every civil liberty in turn.
One of the three charities I regularly donate to, along with CFAR and SENS, is the Institute for Justice, which is a public interest law firm that litigates economic liberty cases. It's up to you whether or not you class 'right to own property' or 'right to earn a living without interference from your competitors' as 'civil liberties' or different kinds of liberties, but I care about the project of advancing liberty comparably to how much I care about advancing rationality and defeating aging.
I, like Hayek, base my libertarianism on arguments from information costs: typically this leads to decentralization, but sometimes it leads to centralization- it makes much more sense to have one CDC with coercive powers than it does to decentralize and voluntarize the problem of preventing epidemics.
I think that smaller scale versions of this are feasible- say, a bathhouse that required all members to take a HIV test every n months, and only allowed entry to people with negative results- but as with most things in public health, larger scales have larger benefits.
This kind of thing is standard practice among porn studios...
A post from Gregory Cochran's and Henry Harpending's excellent blog West Hunter.
The commenter Ron Pavellas adds:
The Wasserman Test.