Coronalinks 4/10: Second Derivative

The second derivative is the rate of growth of the rate of growth. Over the past few weeks, the second derivative of total coronavirus cases switched from positive (typical of exponential growth) to zero or negative (typical of linear or sublinear growth) in most European countries. Over the past few days, it switched from positive to zero/negative in the United States and the world as a whole. These are graphs of the rate of growth – notice how they go from shooting upward to being basically horizontal or downward-sloping (source).

This graph shows the numbers a little differently, (source), but you can see the same process going on in individual US cities:

It would be premature to say we’re now winning the war on coronavirus. But we’ve stopped actively losing ground. If we were going to win, our first sign would be something like this. Current containment strategies are working.

As before, feel free to treat this as an open thread for all coronavirus-related issues. Everything here is speculative and not intended as medical advice.

The Bat Flu

SSC reader Trevor Klee has a great article on why humans keep getting diseases from bats (eg Ebola, SARS, Marburg virus, Nipah virus, coronavirus). He explains that because bats expend so much energy flying, they run higher body temperatures than other mammals, which degrades their DNA. Their DNA is such a mess that the usual immune system strategy of targeting suspicious DNA doesn’t work, so they accept constant low-grade infection with a bunch of viruses as a cost of doing business. Sometimes those viruses cross to humans, and then we get another bat-borne disease.

Subreddit user nodding_and_smiling doesn’t quite buy it:

I don’t think deep-diving into the bat immune system, while certainly very interesting, is necessary to explain the number zoonotic diseases from bats. I think a more important point is there is just a crazy number of bats, and the post doesn’t seem to fully appreciate this.

There are over 1,250 bat species in existence. This is about one fifth of all mammal species. Just to get a sense of this, let me ask a modified version of the question in the title:

“Why do human beings keep getting viruses from cows, sheep, horses, pigs, deer, bears, dogs, seals, cats, foxes, weasels, chimpanzees, monkeys, hares, and rabbits?”

That list contains species from four major mammal clades: ungulates (257 species), carnivora (270), primates (~300), and lagomorphs (91). Adding all these together, we don’t even get to 3/4 of the total number of bat species…

Read the full comment (and the ensuing discussion) for more, including whether biodiversity vs raw numbers is the appropriate measure here.

Mail Suffrage

The Wisconsin Democratic primary (plus some unrelated elections) went ahead as usual this week, with people going out to voting booths instead of voting by mail. Democrats wanted to allow (mandate?) mail voting, but Republicans refused.

Presumably Republicans assumed mail voting would benefit Democrats? The last time a state instituted vote-by-mail, in New Jersey, it did seem to increase the Democratic share of the vote.

I’m surprised by this, because I would have expected mail voting, as opposed to booth voting, benefits people with good executive function who are familiar with doing things by mail – ie older, richer people, ie Republicans. It would appear that I am wrong.

What if the epidemic isn’t done by November? There will probably be a discussion of lifting the shutdown to have a normal election, vs. voting entirely by mail, vs. combination where people who want to vote by mail can but the polls are open for everyone else. I don’t know if the second option is in the Overton Window right now (or if it should be). The party lines here seem to be the same: Nancy Pelosi is already pushing for it, and conservatives are already denouncing it as a liberal plot.

I’m in favor, obviously, but also terrified that something goes wrong. In one scenario, failure to agree on vote-by-mail rules (or failure to implement them competently) delays the election, with no clear way to get it back on track. In another, the sudden panicked switch to a less-tested voting method goes wrong in unpredictable ways and creates ambiguity over election results. It could be Bush v. Gore x 1000.

The Neoliberal Project has an analysis of what we should do and how to make postal voting work. I just really hope it doesn’t come to this.

Charity Update

Last week I linked a list of potentially good coronavirus charities cobbled together by some random people on the EA forum. Now a more serious organization, 80,000 Hours, has posted their own list.

The top option is still the Johns Hopkins Center for Health Security, which researches and advocates for biosecurity policy. Last week someone in the comments doubted the quality of their work, pointing out that one of their flagship efforts is a ranking of how prepared different countries are for a global pandemic; their 2019 listing put the US at the top, which now feels like a cruel joke. But I’m not sure how much to hold it against them. Looking at their webpage, it mostly investigates whether a country has good plans addressing various issues of a crisis, and lots of resources that it can deploy if needed. As best I can tell, the US had great plans and didn’t follow any of them, and lots of resources which it totally failed to deploy effectively. Responsible think tanks are probably not allowed to add a -10000 points at the end of their analysis for “but its leaders are idiots”. This might still be a good time to reread Samzdat on hokey country rankings and no_bear_so_low on hokey country rankings.

Speaking of charity, you can read on Twitter about the trials and tribulations of people trying to donate face masks to hospitals, and here’s an article from three years ago about issuing pandemic bonds as a novel insurance-type way of funding global disease response. Pretty neat.

And you might think that a page called The COVID Challenge where you sign up to deliberately get infected with coronavirus is a bad idea, but it’s actually some volunteers trying to make a list of people who would be willing to get deliberately infected (if it came to that) in order to test vaccines, which they will hand over to vaccine-makers once they get to the testing stage. Rationalist John Beshir did something like this for a malaria vaccine last year and earned $3200 (plus the warm glow of having made a difference) by letting himself getting bitten by infected mosquitoes in an Oxford laboratory.

There Is No Coronavirus In Ba Sing Se

Turkmenistan is a strange country. You probably remember it for its wacky former dictator Turkmenbashi, who among other things renamed the month of March after his mother, and told citizens that anyone who read his book three times would enter Heaven. Or for its wacky current dictator Gurbanguly Berdymukhamedov, who NPR describes as a “dentist/rapper/strongman”. Or for its impressive accomplishment of beating out North Korea to be named the most repressive country on Earth by Reporters Without Borders.

Its coronavirus response will do nothing to improve its reputation: early reports claimed it had banned mentioning the word ‘coronavirus’ or acknowledging its existence in any way.

The Diplomat argues this is not quite true; some state media seems to be using the word. But they are definitely arresting people for talking about it outside official government organs, and they are definitely denying that there are any cases in the country. Since Turkmenistan is right next to Iran, which has had thousands of cases for months, this is pretty implausible.

The Diplomat also requests that people try not to focus on the country’s wacky dictators so much every time they talk about it, since that makes it hard to get people to take its suffering seriously. Sorry, Diplomat and Turkmen people 🙁

And SSC reader Castilho describes their home country of Brazil, which seems to be right up there with Turkmenistan:

We’re one of the few countries in the developing world that actually could handle the pandemic reasonably well (We have around 61.000 ventilators, or 1 ventilator per 3.300 people, which isn’t actually that bad and could be expanded for a decent epidemic response)…

However, our president has decided to go all-in on denying how serious the virus is. The Atlantic even called him “the new leader of the Coronavirus denial movement“. He’s accusing local politicians who have instituted lockdowns of plotting to destroy the country’s economy in order to use it against him later. His sons, who are local politicians in the wealthy parts of the country, have been saying this is all a plot by leftist politicians together with the People’s Republic of China to make him and Trump look bad. I wish I was kidding…

The worst part is that he’s led a nationwide movement telling people to leave their homes and go back to their normal lives. The government actually wanted to make “Brazil can’t stop” into a nationwide campaign, but when a significant part of the population didn’t appreciate it, they just deleted the social media posts and now they claim there never was such a campaign.

Read the full comment for more.

And last month I wondered about the surprisingly slow spread of cases in Iran. I can’t find anyone saying so outright, but it seems like the numbers are probably wrong. At least that’s what I gather from articles like this and Twitter accounts like this highlighting the scale of the crisis there, which seems at least as bad as anywhere in the world. I don’t know if they’re deliberately lying about case numbers (why start now, after the numbers were so bad a few weeks ago?) or if testing has just completely broken down there. See also this article on how their form of government has led to power struggles and a garbled response. I would say something mean about radical Islamic fundamentalism, except that the whole thing mirrors blow for blow what happened between Cuomo and de Blasio in New York.

And finally, here’s a great article on the mystery of Japan. Tl;dr: cultural traditions like mask-wearing and bowing helped it for a while, crowded trains aren’t as bad as you’d think because nobody’s talking, banning large gatherings very early was a really good move, their weak half-hearted version of test-and-trace worked for a while out of sheer luck, but now cases are finally starting to rise and there probably won’t be a mystery to explain for much longer.

Economic Unanimity

The IGM Economics Experts Panel surveys a view dozen top economists on the issues of the day. This month they’re focusing on coronavirus. Here are some sample results:

…they pretty unanimously support the lockdown, even when asked only to reflect on its economic impact.

Some socialists on social media are trying to spread a narrative where capitalists think the economy is more important than lives and want to lift the lockdown immediately, and it’s only socialists who are standing up for the importance of saving people. Top economists aren’t a perfect stand-in for capitalists, but it’s still pretty clear that they’re wrong.

Also, there are starting to be some econ papers trying to more rigorously analyze the pros and cons of lockdown. The Benefits and Costs of Flattening the Curve for COVID-19 says that “assuming that social distancing measures can substantially reduce contacts among individuals, we find net benefits of roughly $5 trillion in our benchmark scenario”.


Is there anything Americans can be proud of here?

@noahpinion reminds us of America’s long history of being late on the trigger but doing a great job once we get started (Churchill: “You can always count on Americans to do the right thing – after they’ve tried everything else.”). We were late entrants into both World Wars but had an outsized effect on both of them. In that spirit, although we were very slow to start testing, we’ve ramped up impressively fast – from almost none to 1/3 of South Korean levels per capita within a few weeks.

Also worth celebrating – during the Wuhan phase of the pandemic, China built an impromptu 1,000 patient hospital in ten days. US media reported this as unbelievable – a sign that a young and vigorous country could accomplish feats that a decadent America could never dream of. But last week in New York, the Army Corps of Engineers converted the Javits Convention Center into an impromptu 2,000 patient hospital in…about ten days.

I don’t know, maybe this was easier because they’re converting an existing structure instead of building a whole new one (though even the Chinese used prefab units). But it’s nice to know we still have it in us to do things quickly. There’s no civilizational decline. If the government ever legalized building things quickly again, we’d be mopping the floor with China within weeks.

Legal Immunity

There’s a Jewish legal principle called marit ayin, which means that it’s illegal to do something which is legal but looks illegal. For example, you can’t eat some kind of plant-based Impossible Bacon, because it would look like you were eating real bacon. Some authorities say it is sometimes permissible to eat the Impossible Bacon if you leave the box out in a prominent position so that it doesn’t look illegal; I’m not sure of the details.

The argument is that widespread flagrant unpunished violation of the law makes the law uncompelling and unenforceable, and this is true whether the violation is real or imagined. If you never see anyone eat bacon, you probably won’t eat it yourself; if everyone around you seems to be eating bacon all the time, it feels less taboo. Also, if you’re a police officer, it’s hard to identify the real bacon eaters if there are a bunch of people eating Impossible Bacon who get annoyed every time you question them.

I was thinking about this recently with the news that Germany is considering issuing immunity certificates for people who have gotten coronavirus, recovered, and are now safe to do normal activities. It’s a good idea, but suffers from the same problem as Impossible Bacon – if there are hundreds of people going outside maskless, eating at restaurants, and sunning themselves on the beach, it’s going to be hard for the rest of us to take lockdown seriously enough.

The equivalent of the rabbis’ put-the-box-out solution would be for governments to issue not just a certificate but some kind of unique article of clothing people could wear to mark their status. For example, they might give an unusually shaped red cap – if the beaches are full of people in red caps, that’s fine and doesn’t say anything about whether you personally should go sunbathe. And if the beachgoers see someone without a red cap, they can question them or keep their distance.

This would take a lot of centralized coordination, though. I’m not sure how you could send the same message without a government order explaining what the cap meant to everybody. Though (as per this Onion article) wearing a fake pangolin snout over your nose would send a strong signal.

A reader who has overcome the disease emailed me to ask whether there are any useful volunteer opportunities for people like him – anyone have any advice?

Short Links

Last week I expressed confusion about how to measure population density so that arbitrary choices of border don’t distort the results. Commenters delivered by finding me this article on population-weighted density, which solves my theoretical concerns but doesn’t really change any of the numbers much.

The Netherlands is another country which, like Sweden and Brazil, is volunteering to be the control group for the great experiment of whether national lockdowns work. Maybe someone should compare them to Belgium or somewhere like that in a few months and see how they did.

An aircraft carrier captain publicly complained that the Navy was failing to address an epidemic aboard his ship; the Navy fired him for whistleblowing. I’m having a hard time thinking of any perspective other than “the Navy is bad and should be torn down totally to the foundations, preferably using some sort of land-based weapon so they can’t fight back”, but here’s a different ex-captain trying his best to give a nuanced perspective.

Say what you will about the New York Times’ coverage lately, but their cover design remains second to none.

This Tumblr post has a discussion of how/whether a Clinton administration might have responded differently to the pandemic, but the part I like is the discussion of the phrase “follow the pandemic response playbook”. It turns out this is a literal document, called the Playbook For Early Response To High Consequence Emerging Infectious Disease Threats And Biological Incidents, and you can read it here.

Marginal Revolution: are hospitals really saving that many people?

UK clinical guideline body NICE now officially recommends against using NSAIDs for coronavirus. Still not completely proven, but I think they’re right to advise caution. While most experts themselves behaved appropriately, this is more egg on the face of the media, which until a few weeks ago was running stories telling people this was a myth and they should ignore it.

538 surveyed infectious disease experts around the US, asking them to predict the number of cases in X days’ time, with confidence intervals. The results are in, and the experts did worse than just continuing the exponential curve on the graph would have. EDIT: But see here.

If you’re following Robin Hanson’s variolation proposals, you can watch Hanson debate vs. Zvi Moskowitz and vs. Greg Cochran (and here’s Cowen on Hanson). Anyway, viral dose seems to have gone mainstream, though nobody seems to be doing anything about it yet.

The two different interpretations of “flatten the curve”. I think this explains why so much of the discussion around this phrase has been confusing.

Trump Asks Medical Supply Firm 3M To Stop Selling N95 Respirators To Canada, and also Key Medical Supplies Were Shipped From US Manufacturers To Foreign Buyers. I think we’re supposed to be outraged about both of those things simultaneously but I can’t manage it, maybe some of you will have better luck.

How much risk do young people really face from coronavirus? What are the risks of long-term complications? Sarah C investigates.

Last week, Elon Musk got widespread praise (including here) for donating a thousand ventilators he managed to procure through his Tesla supply chain. Now the picture has become more confusing. Reporters looking at a picture of his shipment noticed that the boxes pictured are for BiPAP machines – technically a kind of ventilator, but not the kind hospitals need to fight coronavirus. Was the whole thing a giant mistake or cynical PR stunt? But then some hospitals tweeted thanking Tesla specifically for delivering “Medtronic invasive ventilators”, which are the kind hospitals need to fight coronavirus. Some people are theorizing that maybe hospitals don’t want to offend Musk since he might have real ventilators later, other people that maybe Musk got both some useful and some non-useful ventilators in his shipment. I dunno. In any case, he’s still promising to make some at Tesla factories, though.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

829 Responses to Coronalinks 4/10: Second Derivative

  1. thisheavenlyconjugation says:

    “Top economists aren’t a perfect stand-in for capitalists, but it’s still pretty clear that they’re wrong.”

    It seems more reasonable to assume that when socialists refer to “capitalists”, they actually mean capitalists (possessors of capital) who have in many instances expressed (crudely self-interested) anti-lockdown opinions like claiming their companies are essential businesses when they obviously aren’t, or this one from Elon Musk:
    “My best guess, for what it is worth, based on the latest Center for Disease Control data, is that confirmed COVID-19 (this specific form of the common cold) cases will not exceed 0.1% of the US population”.

    • EchoChaos says:

      They have also in many instances gone against their self-interest to help others, e.g. Mario Salerno, who forgave rent for all his units.

      Broadbrushing any group is probably too strong in any case.

      • thisheavenlyconjugation says:

        Yes, but literally everyone constantly makes generalisations like that, it’s not newsworthy. Although, in this case I’ve not seen any actual examples of such generalisations (although surely there must be some) – if some socialists on social media are “building a narrative” by giving specific examples that hardly seems objectionable.

    • HeelBearCub says:

      I don’t think this really washes either.

      The appropriate response to the pandemic in the US was essentially initiated by possessors of capital. The NBA acted immediately and decisively when it became clear how bad it could get (via reporting from Italy) and how vulnerable people are (via Rudy Gilbert testing positive).

      LiveNation began cancelling concerts the next day.

      Different industries and companies have different cultures and incentives. You can’t depend on industry to respond correctly to a pandemic, but it isn’t capitalism or capitalists that are necessarily the problem either.

      • thisheavenlyconjugation says:

        I’m not saying that the socialists Scott is talking about are right, I’m saying that they’re making the true claim that some capitalists seem to be valuing profit over people’s lives and maybe incorrectly building a more general narrative from that (but I think whether the narrative actual is incorrect depends almost entirely on your prior beliefs about the evilness of capitalists); rather than making a false claim about what economists think.

        • HeelBearCub says:

          But isn’t Scott’s main point that there is evidence to say that their general claim about “capitalists”, writ large, is wrong? That’s why he says “Top economists aren’t a perfect stand-in for capitalists”.

          The individual cases of specific people downplaying the threat aren’t all that interesting. Rather, what is interesting, is correctly identifying what the causal links are to downplaying the risk.

          Socialists are identifying a causal link as “capitalism” and there isn’t good evidence that capitalism writ large is responsible. Sweden isn’t doing better on this, and they basically Democratic Socialists. I don’t know that we have any good examples of currently socialist countries and there response, but history tells us that they won’t deal any better than capitalists at responding to this kind of crisis.

          • truckdriver20 says:

            North Korea has zero confirmed coronavirus cases, what more evidence do you need

          • Radu Floricica says:


            For once, I believe them. They literally track every contact to a foreigner. And possibly second contacts as well.

          • Hoopdawg says:

            I don’t think the causal link is “capitalists writ large doing X”, as much as “your financial interests [cough, class, cough] give you an incentive to do X” (which, of course, can and hopefully will be mediated by other incentives). I don’t think this is a particularly controversial observation.

            I’m pretty sure you will find specific people on social media employing the actual “capitalists writ large are evil”, but, as you point out, they’re not particularly interesting. Social media gonna social media, invoking the worst arguments of the enemy side is one of the problems with the discourse on them. The solution is to not pay attention to social media and try to do better.

            All that said, there were specific people right here in this blog’s comments arguing against the lockdown measures, and hopefully the economists’ position can convince them.

          • HeelBearCub says:

            All that said, there were specific people right here in this blog’s comments arguing against the lockdown measures

            Still are.

        • Ttar says:

          What does it mean to “value profit over people’s lives”? Profit is the numerical representation of the voluntary fulfillment of human wants, and the point of life is arguably the fulfillment of human wants (even if a person’s wants are helping others and denying themselves pleasure). So are socialists accusing capitalists of valuing the point of life over life itself? I’m okay with being accused of this. The alternative points vaguely toward the Repugnant Conclusion.

          • Lambert says:

            Value the captalists’ profit over the workers’ lives.

          • Ttar says:

            But the capitalists’ profit represents the provision of value. “Their” profit is actually society’s accounting of the benefit received by others as a result of their action, much like their workers’ salaries represent the workers’ profit from the provision of value to customers (and workers’ salaries can only be paid if the business remains functioning and continues creating value). Keeping people working is valuing the workers’ profits just as much, and all the profit represents is accounting of transmission/creation of value. So I guess, isn’t protecting the capitalists’ (and thus also the workers’) profit the same as maintaining the provision of value? And aren’t the things we value the definition of the point of our lives? So shouldn’t “putting profits before lives” boil down to “putting our values ahead of our lives”? And isn’t that, basically, the definition of being human?

          • Lambert says:

            Assuming spherical cows, you are correct.

            This seems to be in the context of employees being required to come to their workplace.
            The signal to capital that the workers value not coming into the workplace would be employees changing jobs to ones that let them work from home more readily.
            In spherical cow world, this will happen and the market will clear instantly and without any friction.
            In the real world, it takes more than a month to change jobs and longer than that to get good at the new job. Also very few firms are hiring right now.
            For a worker, their options are to come into an unexpectedly infectious workplace or be unemployed till nobody knows when.

            Plus workers spread the disease as well as contracting it. This creates a negative externality, which wouldn’t be priced into a free market. (Even if I work from home, I don’t want everyone else in the neighbourhood to be getting infected at work)

          • Ttar says:

            Is there a course of action proposed by the “profits over lives” camp that doesn’t boil down to “lives over values”? I agree that workers’ choice is between coming to work and getting paid despite a small increase in risk, or staying home and not being paid with a small reduction in risk. Company management’s choices are between having people come to work and paying them, or sending their employees home without pay (even if this is called a furlough these days, it is roughly the same arrangement of things as if someone is fired; there’s just less paperwork involved). Company cash reserves generally aren’t high enough, except maybe at Apple or Google, to afford to pay most staff, without them performing their work, for more than a few days — and that payment would represent a wealth transfer from shareholders to workers; nothing of actual value is being produced in that arrangement. Most shareholders are pension funds/governments and retired seniors — robbing them during a pandemic seems far more immoral than asking workers to create something of value if they want to be paid…

          • A1987dM says:

            Whether it’s preferable for a given fraction of the population to die if that prevents a given amount of worsening in everybody else’s lives (more concretely, whether you’d rather have a 10% chance of losing your job or a 1% chance of losing your father) is not a tautological question, not everybody will have the same answer to it, and it’s not that implausible that the optimal policies under one set of answers are suboptimal under a different set.

          • Tatterdemalion says:

            Ttar, I don’t think your argument works.

            First of all, “profit” is not the same thing as “trade” or “capitalism”; profit is simply my incomings minus my outgoings.

            Yes, it’s true that the profit motive often compels people to engage in mutually beneficial trades with others, and can sometimes therefor be a force for good. But that doesn’t mean it’s an altruistic rather than a purely selfish motive.

            Desire for profit will lead you to make choices that benefit others when, and only when, those are also the choices that benefit you the most.

            “Valuing profit over people’s lives” expands out to “choosing to act selfishly, in ways that maximise your own personal wealth, rather than altruistically, by e.g. giving your employees terms more generous than you could get away with, or donating goods that your factory is well-placed to produce to the people who need them most at below-market prices”.

            What rich people are being accused of is valuing their own well-being more than those of others.

            I worry that you may be conflating “free trade is good, because it leads even people with selfish motivations to act in ways that benefit others” with “selfishness is a good thing in its own right”.

          • Guy in TN says:


            “Their” profit is actually society’s accounting of the benefit received by others as a result of their action

            No, it’s not. It’s an accounting of the economic value produced, not an accounting of aggregated utility gained. “Human’s values” has no relation to “economic value”, and it would be rather trivial to increase total economic value, while deceasing total human’s desires fulfilled.

          • Guy in TN says:


            Is there a course of action proposed by the “profits over lives” camp that doesn’t boil down to “lives over values”?

            To taboo the highly-unhelpful word “value”: the goal is to increase things human desire on aggregate, in contrast to maximizing the aggregated total of economic profits received.

          • Guy in TN says:


            So are socialists accusing capitalists of valuing the point of life over life itself?

            The belief that “maximizing economic value is the point of life” is almost exactly what the left is accusing capitalists of believing, yes. I couldn’t ask for a more succinct illustration.

          • Controls Freak says:

            “Human’s values” has no relation to “economic value”, and it would be rather trivial to increase total economic value, while deceasing total human’s desires fulfilled.

            Being able to show some instances of such a thing emphatically does not prove that there is “no relation”. In a similar way, one can’t leverage, “It is rather trivial to show how some government actions decrease total human’s desires fulfilled, therefore there is no relation between government action and human values.”

            To taboo the highly-unhelpful word “value”: the goal is to increase things human desire on aggregate, in contrast to maximizing the aggregated total of economic profits received.

            Can you think of a more basic human behavior than, “I desire X. In order to acquire X, I will provide a thing of value to someone who can provide me with X”?

          • The Nybbler says:

            Can you think of a more basic human behavior than, “I desire X. In order to acquire X, I will provide a thing of value to someone who can provide me with X”?

            “I desire X. In order to acquire X, I will take it from someone who has it.”

          • Controls Freak says:

            Good news! The Straw Libertarian that Guy likes to argue against totally wants to outlaw that behavior. Are there other basic behaviors in the set? …because that guy’s gonna argue that what we’re left with is those transactions along the lines of, “I desire X. In order to acquire X, I will provide a thing of value to someone who can provide me with X.”

          • To taboo the highly-unhelpful word “value”: the goal is to increase things human desire on aggregate, in contrast to maximizing the aggregated total of economic profits received.

            And increasing value, in the sense in which economists use the term, is increasing the degree to which people get the things that they desire.

            To define the latter, you need some way of doing interpersonal comparison, of balancing an increase in how much of what you desire you get against a decrease in how much of what I desire I get in order to decide if the net is positive or negative. I thought from things you had posted earlier that you realized that, and your only complaint was that you wanted to do interpersonal comparison using (unobservable) utility instead of (observable) willingness to pay.

            But you are now writing as though you simply don’t understand what economic value means.

          • Controls Freak says:

            Assuming spherical cows, you are correct.

            Totally agreed that harsh realities and frictions constrain our spherical cows and inhibit our can opener assumptions.

            This seems to be in the context of employees being required to come to their workplace.

            What I find interesting is that, sometimes, when we have genuine solutions to the frictions, society considers them unacceptable. An example here is Uber’s dynamic prices. There was a massive scandal about an active shooter area and Uber rides cost like $100 to get out of the area. It was super scandalous. “How can you possibly charge that much to save a person’s life?” Well, uh, one can simply look at the other side of the transaction – how much would it cost for you to decide to drive into an active shooter area and give people rides out of it?!

            Of course, the thought is that this is an area in which charity should come in. Well, frankly, nothing was ever stopping charities from organizing the ability to provide such rides (just regular frictions that make it not a spherical cow situation). So, there weren’t any charity rides. Some rides at $100 is better than no rides (the totally predictable outcome of a price cap is a shortage). Uber managed to temporarily overcome the frictions, improve the welfare of folks, and they were derided for it.

            I would also be more sympathetic to people who complained about workers not being compensated for the risk they were taking… if they didn’t tend to be the same people who complain about price gouging letting the price rise to the market-clearing price. I’m totally in favor of the price of goods rising, the wages of workers rising to compensate for their risk, and most importantly, additional money going to whoever can figure out how to get me some friggin’ yeast after passover.

          • Guy in TN says:


            I thought from things you had posted earlier that you realized that, and your only complaint was that you wanted to do interpersonal comparison using (unobservable) utility instead of (observable) willingness to pay.

            But you are now writing as though you simply don’t understand what economic value means.

            This feels like a motte and bailey. I have never heard of “economic value” being used as shorthand for utility before, and it is almost never used in such a manner in common language.

            I feel like if I said, “actually, a country that decreases it’s total GDP in order to provide welfare is increasing its economic value, I’m just using another different interpersonal utility comparison, one that doesn’t involve money” that…wouldn’t go over well. But that is what you are suggesting I should do here?

          • Guy in TN says:

            @Controls Freak

            Can you think of a more basic human behavior than, “I desire X. In order to acquire X, I will provide a thing of value to someone who can provide me with X”?

            I have no idea what you are talking about or asking.

            This isn’t a “spherical cow” problem, as Lambert suggested. He was being too generous. This is a “basic internal logic” problem, that’s missing the mechanism that explains why aggregated dollars should be used as a proxy for aggregate utility.

          • Controls Freak says:

            We build up to aggregate by analyzing components. I’m asking about components of human desires and how they’re fulfilled. Can you think of a more basic human behavior than, “I desire X. In order to acquire X, I will provide a thing of value to someone who can provide me with X”?

          • Guy in TN says:


            And increasing value, in the sense in which economists use the term, is increasing the degree to which people get the things that they desire.

            You understand why I shouldn’t use the term in this manner, right?

            Don’t you see how it would cause more confusion than clarity?

            Let’s apply it to the Coronavirus situation:
            A: We should be trying to maximize economic value, and therefore be very careful about how the shutdown effects the economy
            Me: Oh, don’t worry. The shutdown is actually increasing economic value.
            A: You mean like, in the long run, there would be better if we take a small hit now rather than a big hit later?
            Me: No. This has nothing to do with that trade-off
            A: Then I don’t understand. GDP is dropping. Unemployment is rising. Stocks are crashing. How is this “increasing economic value”
            Me: It’s simple! I’m using another mechanism for calculating economic value other than aggregated dollars.
            A: So your version of “economic value” has nothing to do with total profits, GDP, stock market value, unemployment, or anything like that?
            Me: Correct
            A: (…)

          • and most importantly, additional money going to whoever can figure out how to get me some friggin’ yeast after passover.

            You could start a batch of sourdough starter. It isn’t as if yeast is a rare and endangered order.

          • Guy in TN says:

            @Controls Freak
            I don’t know what you mean by “more basic behavior”, since I don’t have a conception of a “scale of basic-ness”.

            The components upon which human desires are fulfilled, like all things, have elements that are mutually agreed upon between parties, and elements that are not.

            The mutually agreed upon elements could increase or decrease total aggregated desire-fulfilling. You cannot determine whether the exchange increases aggregated desire-fulfilling by determining whether it was mutually agreed upon between the parties.

          • Controls Freak says:

            The mutually agreed upon elements could increase or decrease total aggregated desire-fulfilling. You cannot determine whether the exchange increases aggregated desire-fulfilling by determining that it was mutually agreed upon between the parties.

            Asserted without… how’d you put it… “basic internal logic”. We need to reason our way there, and we’ll need some building blocks.

            Can you think of any economic transactions that you engage in during your daily life that are not meant to satisfy your desires?

          • Guy in TN says:

            Can you think of any economic transactions that you engage in during your daily life that are not meant to satisfy your desires?

            This inevitable argument is why I kept using the word “aggregated” in an almost annoyingly-repetitive manner.

          • Controls Freak says:

            We’re building our way to aggregated. We have to build up basic internal logic.

            EDIT: You can’t complain about people not building up the basic logic slowly enough and then complain about someone walking through it slowly to build it.

          • Guy in TN says:

            Okay, so I do transaction that makes me and another person better off.

            Like, for example, I pay someone to let me rape a poor person. Person A and Person B both better off. Now what?

          • Controls Freak says:

            That’s a good example! It’s a transaction that involves what we call an extremely negative externality… one that is borne of massive coercion, to boot! (The Straw Libertarian you like arguing against wants to ban that coercion bit, so that’s right out for him.)

            We see transactions like that with such extremely high negative externalities. We ban them. We ban purchases of rape. Same with murder. We can pretty easily get rid of a lot of really bad ones. Other transactions have really positive externalities.

            So, now, can you think of transactions that you engage in on a regular basis (which aren’t banned, I assume) that have negative externalities that are more than the benefits derived by you and your trading partner? Can you think of a transaction someone else engages in on a regular basis (again, not banned) that affects you negatively more than they are benefiting? Do you have a measure for this?

            Can you think of a measure for externalities across a society? Can you estimate what percentage it is of the personal economic value? 5%? 10%? -10%? Can you measure the marginal externality as being approximate -100%? (If not, then perhaps there is “a relation” between satisfying desires and economic value. In fact, if it’s more negative than -100%, they’re still related, just negatively related.)

          • Guy in TN says:

            If not, then perhaps there is “a relation” between satisfying desires and economic value. In fact, if it’s more negative than -100%, they’re still related, just negatively related.

            Is this your only point? That I used the word “relation”, and you thought I meant “mathematical correlation”? I have no idea whether it is correlated, my argument was that it’s conceptually unrelated.

            “Economic value” and “human values” share the same root word but refer to very different concepts, and it’s a rhetorical sleight-of-hand to conflate them.

          • my argument was that it’s conceptually unrelated.

            Which, as I already pointed out, is false. They are conceptually closely related since they are sums over the same items, differing only in the weighting.

            You believe your weighting rule is superior, since it fits your moral philosophy. I believe mine is superior, since it approximates yours and, unlike yours, produces something we know how to maximize.

          • Guy in TN says:

            I believe mine is superior, since it approximates yours and, unlike yours, produces something we know how to maximize.

            We’re doing the same arguments, over and over. This is part where I give my spiel about how humans are actually very good at gauging interpersonal utility, and that we have evolved to have the very tools necessary for knowing how to maximize it, mothers and children, and all that. Most people would be morally unsatisfied with applying Coase Theorum to the rape example, for instance.

            This is also the part where I mention how you yourself once indicated that you would deliberately eschew maximizing aggregated economic value, by giving a loaf of bread to a starving penniless person over someone willing to pay, indicating that you must be aware of other tools in the toolbox. But haven’t we been there three or four times before, by now?

            Yes, it’s “easier” your way, to maximize something that fits nicely into mathematical equations. But whether that something you are maximizing is the thing that we ought to be maximizing (that word “approximation” doing a lot of the heavy lifting) is the whole question, isn’t it?

          • My point isn’t that you are wrong about what we should maximize. My point is that you are wrong to claim that economic value and human value had no relation to each other. Not only is that claim wrong, you knew that it was wrong when you made it, as you have repeatedly demonstrated by making the argument you have just made.

            Suppose we were talking about a presidential election. Would you say that the number of electoral votes a candidate got had no relation to the number of popular votes he got? That’s the same issue, since both involve summing the same items with different weightings.

          • Guy in TN says:

            @David Friedman

            Suppose we were talking about a presidential election. Would you say that the number of electoral votes a candidate got had no relation to the number of popular votes he got? That’s the same issue, since both involve summing the same items with different weightings.

            To counter Ttar’s conflation of the term “economic value” with the concept of human values, all I have to do is show that they are not synonymous. That’s it. Unfortunately, I used the squishier term “not related”, which has opened up all sorts of unnecessary lines of objection.

            All things are related from a high enough viewpoint, of course, since the concept of “relatedness” is just a category-boundary question.

          • Guy in TN says:

            @David Friedman

            They are conceptually closely related since they are sums over the same items, differing only in the weighting.

            If you conceptualize a monarchy as the summation of a people’s preferences, only with a strongly tilted weighting system (king at 1x, non-kings at 0x), then you could also say that a monarchy and a democracy are “closely related”. Since after all, they are both the summation of the same set of people, only with different weighting systems.

          • Controls Freak says:

            my argument was that it’s conceptually unrelated

            You’ve already admitted a conceptual relation – when two people engage in a transaction, they do so in order to fulfill their desires. This is a conceptual relationship. We have taken two concepts (economic value and desire-fulfillment) and linked them conceptually using a basic, logical story. The question is whether you can overcome the magnitude of that relation by constructing a measure of utility on externalities. You can plausibly overcome this magnitude by analyzing individual instances (thus, I asked about whether you have some individual examples of things that are not banned but which have sufficiently negative externalities) or by constructing a measure on the aggregate. You can’t just lie about the fact that we have established a conceptual relation to overcome.

          • anonymousskimmer says:

            Paid time off is fricking wonderful. I sincerely wonder how many people are getting it who didn’t have access to it before (improved unemployment), or were fearful of taking it before.

            How much of GDP increases go toward time off per year, and how much of the estimated GDP decrease from the closures does the improved unemployment balance out?

          • To counter Ttar’s conflation of the term “economic value” with the concept of human values, all I have to do is show that they are not synonymous. That’s it. Unfortunately, I used the squishier term “not related”, which has opened up all sorts of unnecessary lines of objection.

            Unfortunately, you said something that was false instead of something that was true, and someone objected.

          • Guy in TN says:

            All things are “related” from a high enough viewpoint, and all things are “unrelated” when zoomed in close enough. This isn’t a question that “true or false” applies to.

            “Economic value” and “human’s values” are conceptually unrelated enough that anyone who uses one as shorthand for the other should immediately be called out on it for the sleight-of-hand, and I will continue to do so in future debates.

          • Controls Freak says:

            All things are “related” from a high enough viewpoint, and all things are “unrelated” when zoomed in close enough.

            This isn’t true, and it’s telling that the directionality you’ve stated here is the opposite of that which you were trying to claim above.

          • HeelBearCub says:

            These things are in the Set of Things in the Universe. Done.

          • All things are “related” from a high enough viewpoint, and all things are “unrelated” when zoomed in close enough. This isn’t a question that “true or false” applies to.

            So when you wrote

            “Human’s values” has no relation to “economic value”

            you were making a statement that you knew was meaningless?

            Why can’t you simply say “I was wrong. I shouldn’t have said that. What I meant was that the two concepts were not equivalent to each other.”

            Which, of course, is true, if what you mean by “human values” is total utility or something similar.

          • Viliam says:

            Not sure whether this is the right moment to say the obvious, but there are a few things that make “human value = economic value” somewhat complicated:

            1) Some things are not exchanged for money. If I move my neighbor’s lawn for money, and my neighbor mows my lawn for money, we have both created value and increased GDP. If I move my own lawn, and my neighbor mowes their own lawn, the same value is created, but the impact on GDP is zero.

            2) How do you compare utility across different people? Philosophically, make your answer whatever you want (e.g. all people matter the same). Economically, if I have twice as much money as my neighbor, my “votes” on the market are twice as strong.

            3) Humans are irrational.

          • there are a few things that make “human value = economic value” somewhat complicated:

            Are you using “Economic value” to mean GDP?

            That isn’t what “value” means as used by economists. Alfred Marshall, who largely invented the concept we know as economic efficiency, pointed out that when a man marries his housekeeper national income goes down, since the exchange of services no longer involves a money payment. His point was precisely that value in the sense of interest to economists is not the same thing as national income or similar measures.

            The clear sense in which economic value differs from utility is your point 2, the point Guy has repeatedly made — that value in economics is willingness to pay, which means we do our interpersonal comparison as if a dollar represented the same amount of utility to everyone, which pretty clearly is not true. A further point is your final one. If individuals are not rational, willingness to pay may misrepresent utility even at the individual level — I may irrationally be willing to pay for something that is not actually of value to me in some psychological of philosophical sense.

            But the observation that lots of transactions don’t involve money is not a difference between economic value and human value, it’s a difference between both and GDP, national income, and similar measures. Your points 2 and 3 are correct, but your point 1 is a misunderstanding of what economists mean by value.

          • One further point expanding on the above. Everyone always puts the problem with interpersonal comparisons using willingness to pay in terms of rich vs poor, but that’s only one of the reasons that marginal utility of income differs among people.

            The implicit assumption is that everyone has the same utility function but different incomes. But it’s pretty clear that isn’t the case. Two people with the same income but different utility functions, an ascetic and a hedonist, may be willing to pay the same amount for a dish of ice cream although the hedonist gets much more utility from consuming it.

            Indeed, by making a different assumption, one can even reverse the usual argument. Assume everyone has the same ability to produce income and the difference is in their utility functions, specifically how much they value consumption. The person who has a high value for consumption works long hours, has a high income, and even with that high income a higher marginal utility for income than the ascetic with a lower income.

            In the real world, people differ both in their ability to acquire money and in their utility function for the things money buys. The common assumption that willingness to pay overweights the utility of the rich depends on assuming that the former is a more important difference than the latter, which may well be true but doesn’t have to be and probably depends on the particular society one is considering.

    • Incurian says:

      In Stubborn Attachments, Tyler Cowen makes the case that there’s nothing better for humanity in the long run than economic growth. While this may not be mainstream, it’s not quite fringe either, and I’m unsure why it isn’t brought up in this context more often.

      • Ttar says:

        I’ve been thinking about this a lot too. I’m not worried about 3.8% GDP loss in this year. I’m worried that if the response delays e.g. AI or genomic research by a couple of years it will represent a death toll, over time, in the hundreds of millions or more.

        • deciusbrutus says:

          What’s the expected delay of a 0.2% chance of death for each researcher?

          • HeelBearCub says:

            Not only this, but this kind of thinking leads to one of those repugnant conclusions.

        • anonymousskimmer says:

          Computerists can do most of their work remotely.

          For other people this is a time to catch up on paperwork, designs, plans, etc…, as long as it doesn’t last too long.

      • Edward Scizorhands says:

        Sometimes I think that is the lesson of

      • Hoopdawg says:

        I’ll go out on the limb and say that this is likely wrong. Or rather, only right insofar as economic growth facilitates technological progress. It’s safe to assume some positive feedback exists between the two, but the relation is not universal and for the most part they can and should be considered separately.

        To use an extreme example: war is bad for economy, but recent ones have been great for progress. Even in our current circumstances, no harm will come from productive forces being averted away from producing consumerist gadgets. Future of humanity depends on research labs. If they stop working, that’s bad, but this simply does not extend to the majority of economic activity. (And that’s not even getting into arguments how most of it is essentially useless either way.)

        • Juanita del Valle says:

          Economic growth is the effectively the manifestation of technological progress.

          • Not necessarily. It could also be capital accumulation, possibly human capital. It could even be climate change. I expect the end of the little ice age resulted in economic growth.

      • HomarusSimpson says:

        There is a clear correlation between GDP/ capita and life expectancy. Immediate caveat, causality may not flow from A>B here, but it doesn’t take much thought to say with a degree of certainty that it does somewhat.

        A bit of squinting at graphs on ourworldindata gives a general number of 3% of GDP/capita equates to one year of life expectancy worldwide, and 6% per year in the developed West.
        It is illuminating to express the tanking of the economy in those terms. Let’s for ease say that western economies shrink by 6% because of this, that’s one year of life expectancy, for everyone (on average).

        If say, 1% of people will die of corona, and lose an average of 5 years of life expectancy (figures plucked from my nether regions) that’s 0.05 population years, only a 20th of the loss from the economic shrinkage.

        • Juanita del Valle says:

          I think you might be confusing trends in long-run growth with business cycle variations.

  2. Lemon_Fantastic says:

    Re: Netherlands:
    >Sometimes this lockdown feels invisible. Cities may be quieter, but children still clamber on climbing frames and teenagers cycle side-by-side.

    Anecdata from Amsterdam: it’s been absurdly quiet and empty for the past 3 weeks. My experience does not reflect this article at all.

    The amount of people outside is less than 20% of what it normally would be. The amount of hospitalizations is steadily dropping, although the amount of reported cases is not [1]. This is probably due to increased testing capacity. NL seems to have been, and still be underreporting cases, and is finally catching up as reflected here [2]. Testing capacity has been massively increased since April 1st [3].

    Finally, according to the bureau of statistics, reported deaths are probably underestimated by a maximum factor of 2 [4]. This number should come down from 2 as testing capacity scales up further.


    • truckdriver20 says:

      Based on the article Scott linked, the Netherlands has bars and restaurants closed, schools closed, hairdressers and the like closed, but more “non-essential” businesses open compared to the US and no official stay-at-home order.

      This looks to me like what May/June in the US might look like as we gradually open up a little bit more while remaining cautious. If it works in the Netherlands then it’s a good sign that it could become a sustainable normal even without mass testing and tracing available yet.

      Do you know if the gym is open?

      • Aapje says:

        Gyms are closed.

      • crosswind says:

        There’s official stay-at-home “advice”, though.

        Everyone in the Netherlands should stay at home as much as possible. Work at home if possible. Go outside only when you need to: to buy groceries, to walk the dog, to get some fresh air, or when you need to run an errand or care for someone else. If you must leave your house, go alone as much as possible and stay 1.5 metres away from others. People who work in crucial sectors and critical processes can go to work if they cannot work at home.

        Moreover, congregating with more than 3 people in public places – and not keeping distance – can be fined by up to 400 euros.

        • SamChevre says:

          That’s very much the same as Massachusetts–not that much is actually closed, but people are following the “stay at home” recommendation.

    • Cliff says:

      I have been following closely since I know many people in Belgium, and at first Belgium seemed to be doing a bit better (especially in deaths) but now it is doing significantly worse overall (although the two have had quite similar outcomes so far). Belgium has done a bit more testing but its not a huge difference. If Netherlands does no worse that strongly suggests we can relax measures a bit, which seems right to me. There’s a lot you can do while maintaining a reasonable physical distance and wearing a mask.

    • uffe says:

      I live in Sweden and we’re following a similar strategy like you are. One big challenge is that deaths are counted with different quality in different countries. Sweden has one of the most rapid & comprehensive updates to deaths by Corona, and we also count people dying at home with a very generous definition.

      Our deaths per million, which are still below many Western European countries but well above our Nordic neighbours, would have to be compared once have the full data for the period and can compare with previous years’ median deaths.

      Another factor is immunity. We had a sampling study done in the Stockholm area which came to the conclusion around 2.5% of people were carrying the virus at the time(this was late March/Early April). But you also have to factor in that Sweden got its first case in late Feb and we’re now two weeks from that. We’re supposed to get a new statistical model after the weekend, but some early speculation is that we could already have 20-25% immunity in Stockholm (which has over half of all national cases).

      In addition to that, because testing has ramped up, the daily new cases isn’t as reliable anymore. Deaths are a lagging indicator. I look more towards ICU, where we have seen a very stable daily admittance rate for many weeks without much movement. In fact, we’re using less than 20% of our total ICU capacity and people are encouraged to come if they have any complications at all (free of charge).

      From where I stand, our strategy has worked fairly well thus far given that no vaccine is likely for at least a year if not longer and any 2nd or 3rd waves would now hit us less hard than many others due to much greater immunity.

  3. Radu Floricica says:

    Re the importance of initial viral load, and also related to using people who recovered as volunteers: what happens when you take either somebody who recovered, or somebody still fighting the disease, and put them without protection in an environment with high viral load? IANAD, but my guess is that immune response is always dose dependent, so adding more than it’s currently set up to deal with doesn’t help.

    And if that is true – wouldn’t this make a difference on how patients fare when treated in a hospital, in an environment where they’re likely breathing aerosols non stop?

    • Purplehermann says:

      If this were true wouldn’t vaccines and variolation fail miserably in general?

    • MilesM says:

      I am not a virologist, (I am actually a biologist, and worked with viral cultures quite a bit, a long time time ago – but hated it, changed fields and didn’t stay current, so…) but I’m pretty confident that even the heaviest environmental “viral load” you’re likely to experience is at least several orders of magnitude lower than what would have been present inside the body during acute infection.

      So if your immune system had been able to handle that, being exposed to some extra viral particles in the air should be a non-issue. (assuming this virus is something people can develop effective and lasting immunity to)

  4. Alsadius says:

    Regarding mail-in voting, remember that the ballots also need to be counted. If you can’t put several people in a room(at the very least, one counter and a couple scrutineers), you can’t count mail ballots, even if people cast them.

    As for the political angle on it, the usual complaints I see about it from the right side of the political spectrum all boil down to fraud. It’s easier to fake your identity/citizenship/felon status/etc. if you’re using ID sent by mail than if you have to show up in person.

    Here in Canada, we get voter cards sent by mail every election. These aren’t ID(they’re more to tell us about polling locations etc.), but they are sent out based on where Elections Canada thinks you live. And mine have been right about half the time, in practice. Part of that is that I’m younger(34) and have moved a fair bit, but more than once I’ve gotten cards for other people, or my cards have never shown up. So clearly, the system we have now is not sufficient to tell who lives where, despite Canada typically being not-too-bad at that kind of thing. You will need to add on the ability to say “Hey, I’m a new voter [or I moved, or whatever] – please let me vote from my new place”. And then prove your identity with a piece of paper, where they can’t easily check for fake ID.

    We also had a party leadership election a few years back, where one of the candidates had the support of the party apparatus, and a lot of volunteers for his leading opponent never got ballots in the mail, despite calling in to complain repeatedly. I don’t know if this was fraud or incompetence(it could really have been either with that group), but it really left a bad taste in the mouth of the supporters of the latter candidate when he lost in a close race.

    Obviously, the counter-argument here is that fraud is super-rare. The gold standard for fraud investigations(at least, so far as I’m aware) in the US in recent years was the 2004 Washington governor’s election, where the legal fighting resulted in a judge declaring 1,678 illegal votes having been cast, in a race with 2,810,053 total votes, or about 0.06%. I think they only checked about half the votes cast(since it was mostly Republicans making this argument, they only looked at counties the Democrat had won), so it might be up to 0.1%. And this was only the kinds of fraud that could be proven after the fact – mostly felons, with a few double voters and dead voters. So any fake-ID voting that was successful wouldn’t be in these numbers. Still, 0.1% is a pretty small amount of fraud as these things go, especially since a lot of them are honest errors (not systematic fraud efforts) that could largely cancel each other out. So perhaps they invest too much effort in worrying about it. But the number is clearly higher than the ~10 cases in the US per election cycle that actually result in criminal charges. What to do about it, I’ll leave as an exercise to the reader.

    • EchoChaos says:

      There are a couple of other concerns with mail-in voting that are harder to detect.

      The first is that it vitiates the secret ballot. A polling place legally must have poll watchers and ensure that you aren’t having someone go in with you unless you specifically need aid. Elsewhere it’s much easier to do “Honey, prove to me you voted for Trump”.

      The second is that combined with ballot harvesting (legal in some but not all jurisdictions with mail-in) it allows outsized and completely impossible to scrutinize turnout operations that specifically benefit Democrats. Even if you just harvest ballots and don’t do any pressuring, Democrats tend to live in tighter quarters, which means that ballot harvesting “Fill it out and I’ll mail it in” will benefit them per unit time.

      • Alsadius says:

        Fair. FWIW, there have been some cases of Republican ballot harvesting (North Carolina had a case last election, IIRC, despite it being illegal there), but it probably advantages Democrats overall in the US context.

        • EchoChaos says:

          Fair. FWIW, there have been some cases of Republican ballot harvesting (North Carolina had a case last election, IIRC, despite it being illegal there), but it probably advantages Democrats overall in the US context.

          To be clear, both sides do ballot harvesting (and probably both do illegal ballot harvesting), but Democrats get more bang for their buck out of it. Driving to each house in a suburban neighborhood takes more effort than canvassing an apartment building.

          And NC making it illegal is the only reason that voter fraud was detected. If ballot harvesting had been legal, that would’ve been completely undiscoverable.

      • DeWitt says:

        I don’t really know of a good way to completely assuage the secret ballot concern, but what keeps the US from mandating a federal ban on ballot harvesting to deal with the second?

        • EchoChaos says:

          but what keeps the US from mandating a federal ban on ballot harvesting to deal with the second?

          The Democratic Party.

          • DeWitt says:

            The Republicans just recently came out of a good few years in which the Democratic party has been toothless, so I’m going to want you to source such a claim.

          • EchoChaos says:


            Is joke. You laugh.

            Seriously, I’m being relatively flip, but at no point have the Republicans had a filibuster-proof majority at the Federal level in the past decade.

            Additionally, given that each state runs their own elections, there would be some strict scrutiny if the Federal government even had the right to mandate such a thing.

          • DeWitt says:

            Ha ha.

          • HeelBearCub says:

            Want to hear a joke?

            Echo Chaos.

            So hilarious.

          • DeWitt says:

            Aww, shucks. So funny. Wow.

          • EchoChaos says:


            Love you too. 🙂

          • Incurian says:


          • Simon_Jester says:


            The problem with laughing off @DeWitt ‘s question is…

            Did the Republicans even [i]try[/i] to pass legislation banning ballot harvesting? Do you have evidence that Democrats even threatened to filibuster such legislation?

            I can very easily picture Democrats agreeing to that, especially in the context of a broader compromise bill that makes mail order voting in general more feasible but also tightens up provisions to avoid actual (as opposed to imagined-as-excuse-to-avoid-letting-people-vote) fraud issues.

          • EchoChaos says:


            Did the Republicans even [i]try[/i] to pass legislation banning ballot harvesting?

            Yes, Republican states have banned it (e.g. Texas, Arizona, North Carolina) and Democrat states have legalized it (e.g. California).

            Arizona’s is still under court challenge from Democrats.

            There is no Federal law because there is no “Federal vote”. All votes in the USA are adjudicated and decided at the state level. It hasn’t been a Federal issue until very recently because it’s just a very recent thing.

            I can very easily picture Democrats agreeing to that, especially in the context of a broader compromise bill that makes mail order voting in general more feasible but also tightens up provisions to avoid actual (as opposed to imagined-as-excuse-to-avoid-letting-people-vote) fraud issues.

            I can’t. I’ll betcha a nickel that Democrats don’t offer a vote harvesting ban.

          • I was curious about exactly what ballot harvesting was, so googled for it and found a news story that quoted the relevant bit of California law. What struck me is that there is no requirement that the ballot be sealed before it is handed over to someone else to hand in. That makes it trivial to use ballot harvesting for fraud — just collect lots of ballots and only hand in the ones that vote the way you want. And doing that on a scale of hundreds of votes would only require one person acting where nobody else could observe the fraudulent part of what he did. Have a bunch of people collect a thousand ballots, hand them to one person to hand in, he goes over them in private and trashes the ones he doesn’t want to hand in.

            I would think that requiring the ballots to be sealed, ideally in some way that made unsealing and resealing hard, would be the obvious precaution for anyone who wanted to make voting easier without making fraud easier.

          • nkurz says:

            > requiring the ballots to be sealed

            While this is probably an improvement, I don’t know if it would help that much with fraud. Even if you can’t verify the ballots directly, probabilistically eliminating ballots based on demographics gets you most of the benefit. Simplistically, you could just collect ballots from a neighborhood that historically votes for your opposition, and throw them all away. It would seem much safer just to ban the practice altogether.

          • Simon_Jester says:


            Under present circumstances it would be fairly easy to create a federal law regarding ballot harvesting.

            “Due to the coronavirus outbreak we are giving each state X dollars per registered voter in the state in 2018, for the purpose of setting up infrastructure for universal mail-in voting. This money will only be distributed under the following conditions:”

            1) [Antidiscrimination boilerplate ensuring that a state cannot, for instance, ‘forget’ to pass out mail order ballots to specific groups of voters]

            2) …
            3) …

            N) No person shall mail a ballot on behalf of another person, unless they sign two statements representing that they will faithfully mail in that ballot. One will be left with the voter, and one enclosed with the ballot.

            Then you hold the election. If the Republican Party is worried about people harvesting ballots and conveniently leaving out the Republican ballots, they send someone around with a clipboard a few weeks later (or take advantage of their existing mailing/email/phone/etc lists):

            “Hey, Joe Redhat, did you mail in a ballot?” “Yep.” “Did you send it in yourself?” “No, I gave it to Bob Sleazydonkey.” “Did he give you a receipt?” “Yep.”

            [Check to see if a receipt saying “Bob Donkey is mailing this ballot” came in with Joe’s ballot. If not, or if Joe’s ballot never arrived at all, Bob faces criminal charges]

            This exact system may not be airtight, but you get the idea. Given that we still have roughly six months to prepare, and that the federal government has plenty of tools to force states to adopt standards on something like this, it wouldn’t be hard to set up.


            If the Democrats don’t offer a vote harvesting ban along these lines, I predict it’s because they never get a chance because the Republicans refuse to even consider mass mail-ins of absentee ballots as an anti-coronavirus measure.

      • nupi says:

        Secret ballots by mail is trivial. You put the certificate that you are allowed to vote in the envelope (needed t9 guard against double votes, etc) and then the actual vote nextyto it inside of a second envelope that gets only opened upon counting.

        This has literally been done for two plus decades in Switzerland every 3 to 4 months.

        Of course, if you cannot manage to issue certificates properly it is harder but that has nothing to do with secrecy of the ballots

        • EchoChaos says:

          Secret ballots by mail is trivial. You put the certificate that you are allowed to vote in the envelope (needed t9 guard against double votes, etc) and then the actual vote nextyto it inside of a second envelope that gets only opened upon counting.

          That is also the US system in states with mail in ballots. But how do you verify that a wife didn’t fill it out with her husband watching?

          In a polling place, she goes in alone and poll workers prevent her husband going with her. When they live together, how can you do that?

          • nupi says:

            With a sufficiently malevolent/violent husband (in all other cases it would seem up to the wife), you cannot. But you also cannot completely prevent malfeasance of that sort in the case of in person ballots (‘Hey, where did my car keys go?’ at least prevents her from voting)…

          • EchoChaos says:


            This assumes that the husband knows the wife is not going to vote the way he wants. She could tell him that she’s voting Kang and really vote Kodos in the booth.

            With a mail in ballot, this is not possible.

          • The Nybbler says:

            I think EchoChaos is using that example because he thinks it will resonate with Democrats. But I suspect the actual kind of pressure that Republicans are afraid will result is the other way — not husbands forcing wives to vote Trump, but wives/girlfriends/fiancees pressuring husbands/boyfriends/fiances, acting as the enforcers of social mores, not to vote for that awful orange man. Non-secret ballots could break the “shy Trumper” effect.

          • albatross11 says:

            Well, if you propose a new voting system with better security, but it allows voters to prove how they voted to others, nobody is adopting that system. There are clever cryptographic schemes to let you get verifiable polling-place elections without leaving voters with a way to do that, but mail-in ballots don’t[1].

            Now, this is a security/usability tradeoff and it’s pretty easy to explain to voters and voting officials, and if the voters want it, I think it’s reasonable. But it’s accepting that we lose resistance to coercion and vote buying, in exchange for more convenience and higher turnout.

            [1] Some examples: Scantegrity, Pret-a-voter, Helios, Starvote. If you’re interested in this, go find Ron Rivest’s ThreeBallot paper and read it–nobody would actually use ThreeBallot directly, but it gets most of the desired properties of cryptographic voting schemes without making you untangle mix-nets or zero-knowledge proofs or homomorphic encryption schemes.

          • Edward Scizorhands says:

            “Secret ballot, unless someone wants it to not be secret” is not secret ballot.

            I don’t like voting outside of a booth, but sometimes we need to do it anyway. Some people are out of town. Some people cannot get to the polls. And sometimes there is a virus going around.

          • viVI_IViv says:

            Forget husbands and wives. What does prevent people from selling their mail-in ballots?

            And yes, this sort of thing happens in various countries.

            Polling stations don’t guarantee 100% secrecy either: there are people who take pictures of their marked ballots, which is illegal in various countries and some US states, but with mail vote this would go up to eleven.

          • ksvanhorn says:

            Need I point out the bigotry on display here that assumes it’s a malevolent husband trying to control his wife’s vote, and not a malevolent wife trying to control her husband’s vote?

          • EchoChaos says:


            No, because @The Nybbler already did.

          • Simon_Jester says:

            The bigotry comes from a very natural place: namely, that for most of history men had the authority to tell women what to do and make them miserable if they disobeyed, without legal consequence.

            I mean, it was even deployed as a probably-not-facetious argument against women’s suffrage: “Letting women vote is redundant because they’ll just vote how their husbands tell them to vote.”

            I can’t comment as to what the actual prevalence rate is of coercive women versus coercive men who would try to pressure a spouse into voting against their preferences. I suspect it’s a problem that largely cancels itself out, and also one that would exist anyway because many spouses aren’t going to realistically be able to conceal their vote choices.

          • John Schilling says:

            The bigotry comes from a very natural place: namely, that for most of history men had the authority to tell women what to do and make them miserable if they disobeyed, without legal consequence.

            Based on my experience reading popular literature going back at least a thousand years or so, I’m pretty sure women have always had the ability to tell men what to do and make them miserable if they disobeyed, without legal consequence. And will at least three centuries into the future. That both genders have this ability, has resulted in no small amount of misery – some of it dramatically or comedically entertaining when viewed from a safe distance.

            There are asymmetries in the method, of course. But mostly in ways that relatively empower women on the domestic front, and we’re proposing a change that makes voting into a domestic activity.

          • John made the point I was about to make.

            There’s a great anecdote by al-Tanukhi, a ninth century Qadi, which implicitly assumes that, in a legal system where men could divorce their wives but wives couldn’t divorce their husbands, a wife can force her husband to divorce her by making being married to her sufficiently unpleasant.

          • anonymousskimmer says:

            because many spouses aren’t going to realistically be able to conceal their vote choices.

            In California certain political parties require you to be registered within them in order to vote their ballot.

            If you are registered with a political party in California, you’ll receive a voter’s guide for the primary ballot of that particular party (at least in my county).

          • Simon_Jester says:

            The wife CAN coerce her husband in societies with what we would call ‘traditional’ marriage laws.

            However, she usually has rather fewer social and legal protection from adverse consequences if she decides to escalate.

            It’s like a nation that has no nuclear weapons, deciding to fight a war with a nation that has them. Under some circumstances that can be a survivable and winnable experience, but it there is a significant power imbalance at work there, and one side has a lot more options for ending the conflict permanently than the other does.

          • John Schilling says:

            In every society whose stories I have read, it is a cliche that the woman has the right to make the man’s life a domestic hell if he does not arrange domestic affairs to her satisfaction, and that the man has to either suffer this or conform. The man who beats his wife over this is a pathetic loser, the man who kills his wife over this even more so, and the man who is killed by his wife over this still more so. The man’s legitimate authority to beat or kill his wife, where it exists, is for e.g. sexual or economic affairs, and if he is seen as exceeding that authority then people will likely either imprison him for murder or tut-tut with mild regret about how he fell backwards onto that kitchen knife so many times.

            I don’t know that this is a universal human norm, but I’m pretty sure it’s far more widespread than the bit about how until our enlightened era it was all domineering evil men imposing absolute submission on their women in all aspects of life.

            Also, we aren’t talking about archaic traditional societies, but about a possible modification of our own enlightened one. If you move voting into the domestic realm, then you’re moving it into the realm where women are allowed to henpeck and men are enjoined from violent retaliation, now more than ever.

        • Alsadius says:

          I understand the principle, and it’s not a bad system. But it’s at least a bit easier to cheat than in-person (since you can Photoshop an image of a driver’s license a lot more easily than you can make a fake driver’s license), and for people who worry about fraud, that’s going to be concerning.

          • nupi says:

            I agree the fraud part is potentially an issue (more so than secrecy of ballots in my view – people living under circumstances where that would be an issue have bigger problems to solve). Less so if you have a good idea where your voters live which I understand is less of a given in the US…

      • Corey says:

        Husband-wife is a much smaller concern than employer-employee, or employee-union if unions were still a thing.

        Or vote buying for that matter. Basically any way one can prove one voted a certain way can lead to coercion (I know ancaps don’t consider vote buying coercion).

        • albatross11 says:

          Right. I can just pay you cash for your unmarked mail-in ballot, and you never learn who I am or how your vote was changed.

        • Simon_Jester says:

          While this is possible in theory, it is also entirely feasible to make it super illegal and harshly punish anyone who tries.

          As forms of election fraud go, I’m less worried about it than I am about the old standby of “use your buddies’ news agency to downplay the severity of an epidemic, then chuckle as your supporters turn out to vote in larger numbers while your opponents stay home to avoid getting sick from the epidemic.” Which seems if anything a MORE unethical, and importantly a less neutral, form of election rigging.

      • thedufer says:

        Anyone who is anti-mail-in voting needs to explain how switching from partial mail-in voting to all mail-in voting would change anything. Surely anyone who would commit fraud (or break ballot secrecy, or whatever the argument of the day is) is already doing that? Mail-in absentee voting is legal in every state. Doing so requires no excuse in most cases, and merely requires that you explain why you’ll be outside of your county on election day in the rest. Anyone who would do nefarious things via mail-in voting already can. At this point it’s mostly about making mail-in voting more accessible so that everyone does it, instead of only those who need to plus those who are committing fraud. Sure, maybe there are a few fraudsters at the margin who want to commit fraud, but only if getting a mail-in ballot is really, really easy, but that can’t possibly be a large group.

        • The Nybbler says:

          Anyone who is anti-mail-in voting needs to explain how switching from partial mail-in voting to all mail-in voting would change anything.

          And they say slippery slopes aren’t a thing.

          • Simon_Jester says:

            I dunno, to me this argument comes across as “if there’s a slippery slope here, we’ve already fallen down it anyway, because it’s trivially easy to get a mail-in ballot.”

            If there was someone out there willing to just buy votes for $100 each, they could already go to a low-income neighborhood and finding people willing to sell their votes by ordering a mail-order ballot and handing it to the buyer. That’s already a thing that can happen.

            So we have to ask ourselves, why doesn’t that appear to be happening on any large scale in real life, and why would it suddenly start happening if we issued more mail-in ballots?

          • John Schilling says:

            So we have to ask ourselves, why doesn’t that appear to be happening on any large scale in real life

            Because the Russians weren’t sure what they could get away with in 2016, and the Chinese weren’t sure they could get away with anything at all along those lines. Now that everyone who has been paying attention knows that the US response is 98% yelling at the domestic outgroup for being in collusion with foreign “election-meddlers”, and 2% yelling at the foreigners in question, and 0% doing anything that actually punishes the foreigners in question, their tactics will probably change accordingly.

            Security against human adversaries cannot be maintained by assuming a static threat model where that what was not harmful in the past will remain harmless in the future. That way leads to saying that, since nobody has ever tried to rob Fort Knox, we can dispense with the guards and alarms and vaults if they prove inconvenient.

          • Edward Scizorhands says:

            John points out the fundamental issue when someone wants to enhance election security: the other side sees it as an invalidation of their victories — or at the vest least, a validation of the wretched outgroup — and insists that it was never a problem and never will be one.

            You can imagine tying the two things together into a common threat — say, what if the Russians print up a bunch of absentee ballots just to fuck with our election process? I suspect the result would be both sides declaring it impossible and you’re evil just for asking.

          • Simon_Jester says:

            I don’t think the Russians, or any foreign agents, will create legwork-intensive schemes for meddling in US elections.

            The nice thing about slipping the candidates money or free oppo research, or creating Facebook bots to praise them, is that there’s less legwork. You don’t actually have to create huge networks of foreign agents who all keep their activities secret while in the country whose elections you’re manipulating.

            Plans that involve things like fraudulent ballot harvesting, fraudulently impersonating another voter, and so on, take legwork. You need people on the ground who are familiar with the local system, who live in the target country. Large numbers of them, who know they are committing criminal activity on your behalf.

            The Russians and Chinese may try to game the next election, but I suspect they’d be doing it by manipulating someone else who’s already here.

          • John Schilling says:

            I don’t think the Russians, or any foreign agents, will create legwork-intensive schemes for meddling in US elections.

            It takes very little legwork for the Russians to data-mine for susceptible targets and call or email them with the offer “Send us a signed, valid, but otherwise blank absentee ballot and we’ll send you $100”, or some variation on that theme. They may be able to leverage the legwork of American petty criminals to that end, but no Russian need ever set foot on American soil.

            Or maybe a few Russians to drop off the now-completed ballots in American mailboxes for them to be suitably postmarked, but not many and with very little exposure.

          • Edward Scizorhands says:

            They don’t even need to succeed at altering election results with a bunch of random fake ballots for that to have the desired behavior of sowing chaos and doubt about our election results.

            You want to be able to show, after the fact, “this is what the enemy did and this is how our system thwarted them.” You need to have this ready ahead of time without knowing specifically what the enemy is going to do. “I only planned for the enemy to do things that align with my political priors” is non-responsive.

            HBC did a pretty good of changing my mind about many parts of the North Carolina absentee ballot process using facts. A hostile actor would need to take steps that are detectable before the election by requesting a ballot get sent to a given person at a given address. I’d probably want more effort put into attempting to detect fraud, but I suspect a lot of other battleground states have worse security that deserve attention.

    • Dan says:

      “Fraud” is the excuse Republicans use but it’s not what they’re really concerned about.

      The problem is that essentially all retirees vote Republican and essentially all struggling poor people who are holding down three part time jobs and can’t afford a babysitter would vote Democrat, if they voted, which they don’t because ain’t nobody got time for that.

      A system where you have to take time out of your day and go wait in line for a unknown length of time to vote (ie, the current system) therefore benefits Republicans because more Republican voters can/will go out to vote because the retirees have nothing else to be doing anyway. If you allow voting by mail then you will boost Democratic Party turnout by allowing people to vote who otherwise couldn’t have found the time to do it.

      • DeWitt says:

        I volunteered for a voting booth, once, and it was literally right inside of a retirement home. It oughtn’t be a mystery to anyone that old people vote a lot when they get genuinely and literally wheeled into the booths.

      • EchoChaos says:

        essentially all struggling poor people who are holding down three part time jobs and can’t afford a babysitter would vote Democrat, if they voted, which they don’t because ain’t nobody got time for that.

        Incidentally, that’s a modestly Republican demographic.

        Democrats own the truly poor (<30k), but the "multiple jobs and struggling" folks are usually making more than that.

        You are probably mentally adding "minority", which would indeed make it a Democrat demographic, but there are actually more poor whites in the USA than any other poor minority. And poor whites are very Republican.

      • Alsadius says:

        I talk to a lot of Republicans. And yes, they are genuinely worried about fraud. “We don’t just have to beat the Democrats, we need to beat them by the margin of fraud too” is a fairly common sentiment. You can say it’s overblown, or that they’re believing conspiracy theories about the scale of it. You could even say that it’s a real issue in principle, but a smaller issue than ballot access, and therefore we shouldn’t fret too much in practice.

        But their beliefs are honest, just like yours and mine.

        • DeWitt says:

          People are very good at genuinely believing they are right about wanting to come out on top. I’m sure their beliefs are genuine, but that doesn’t necessarily make them less misguided for it.

          • Alsadius says:

            Of course. “Their claims accurately reflect their conscious beliefs” is an extremely low bar to clear, for any large group of people. TBH, for a large group to consistently say things that do *not* reflect their beliefs is borderline conspiracy theory territory.

            This is why I get so annoyed by how frequently the accusation is made that $LARGE_GROUP is lying about $ISSUE, and that their real concern is $MALFEASANCE. It’s a fairly crazy accusation, but I see it in like 80% of political debates even so. Because people lack understanding, empathy, and/or common sense.

          • DeWitt says:

            Yeah, no, I absolutely agree. Even so, their being genuine in no way reflects well on what they are saying being valid despite it.

          • smilerz says:

            “People are very good at genuinely believing they are right about wanting to come out on top. I’m sure their beliefs are genuine, but that doesn’t necessarily make them less misguided for it.”

            That penchant for motivated thinking works both ways – not just the side that disagrees with you.

          • DeWitt says:

            Yes, it does. ‘My beliefs are genuine’ is as bad an argument as ‘their beliefs are genuine’.

          • smilerz says:

            @DeWitt – I assume you meant “their beliefs are not genuine”

          • DeWitt says:

            No, I meant what I said. Whether or not someone really believes, honest in what nonsense they’re talking about has little bearing on whether you should also believe it.

        • Dan says:

          I should have clarified: Republican voters are worried about fraud, because they have been told repeatedly that it’s a major problem by Republican politicians and Fox News. But it’s not actually a problem, and the Republican leadership (other than Trump) knows this perfectly well; this is why concern for voting fraud is so perfectly aligned with partisanship. If there were actually problems with voting fraud, you would expect some Democrats to care, since you’d expect some of the fraud to benefit Republicans. But there is no actual fraud going on, other than the “fraud” talking point itself

          • Alsadius says:

            Democrats care quite a lot about Republican fraud. Just look up a lot of the fights in North Carolina in recent years.

            I think there’s a healthy dose of confirmation bias here. Our Team is good, noble, and would never stoop to fraudulent actions (other than maybe to counter Their Team’s well-known love of cheating). But Their Team is a pack of scoundrels who’d sell their mothers for a nickel.

            Also, the people who can’t legally vote – non-citizen immigrants, under-18 children, and convicted felons – are all in groups that lean far to the left. If it’s pure partisanship at play, both parties would behave like they are behaving. Republicans don’t have much opportunity to cheat on voter ID laws, because frankly most people who’d need to cheat in order to vote wouldn’t support them. (There’s a whole host of other ways they can cheat, obviously, but not this particular one very much).

            Finally, be careful saying “They’re all blinded by their media sources of choice”. It’s a fully general counterargument against any position anyone has ever taken, regardless of details. Clearly some fraud exists – note that Washington data I cited above at the beginning of this thread. (0.06% may sound low, but it’s about four orders of magnitude higher than the estimates I see from random left-wingers of the amount of fraud in US elections, because they use data like criminal charges to say it’s a single-digit number of cases per election)

          • HeelBearCub says:

            I’m concerned about individual cases of perpetration of fraud. I’m not worried that these are particularly wide spread.

            Again, the ballot fraud in NC isn’t the kind of ballot fraud Republicans have said they have been concerned about generally. It is the kind of ballot fraud that is actually a risk. But only when vote totals and margins are low.

            And I’ve said before, the particular case in NC was perpetrated by someone who was doing it for the Dems when NC was basically a one party Dem state, and started doing it for the Republicans when party control changed. It has a lot more to do with the specifics of Bladen and Robeson counties than anything else.

          • Gobbobobble says:

            But it’s not actually a problem

            Not actually a problem currently. The United States didn’t always have secret ballots: “In the U.S., voting by secret ballot was universal by 1892 but criminal prohibitions against paying people to vote were only instituted in 1925.[19]”

            But, no, go ahead and rip down the fence. The cruel irony will be when the capitalist class intimidates/bribes/etc the wage slaves into voting their way… again.

          • Mary says:

            But it’s not actually a problem,

            And how do you know this?

            Furthermore, it is not enough for it to not be a problem. You need to be able to show it. Justice must not only be done, it must be seen to be done.

          • Edward Scizorhands says:

            I think there is very little education done about how the people who run our elections secure them. We need regular news stories that describe this, in more detail than “we take care of it, honest.”

          • Mary says:

            It’s not education. It’s inspection. Furthermore, that means the relationship is the opposite.

            They do not educate us about what they do to make the election secure.

            We inspect whether they are making the election secure.

        • DaveK says:

          I would say this is a bit more complicated then the dichotomy that is being discussed here.

          It’s true that people aren’t generally completely lying about their conscious beliefs. That does seem to be a thing people often assume about the other side.

          For example, with the Brett Kavanaugh case, I saw a symmetry between the sides where not only did they believe his guilt/innocence was obvious, but also believed it was obvious to the other side, and they simply didn’t care.

          That would be a case of conspiracy theory type thinking.

          But there is something more subtle when people make the claim “x group claim to say they care about this, when they really care about this.”

          I think there a lot of cases, and this is probably true a significant amount of the time, where people give a “most favorable” explanation of their motivations, when there motivations are actually complex.

          Take something like the abortion debate. Pro-choicers will often argue that pro-life people “don’t really believe fetuses are being with a right to life, there actual motivation is to control women.”

          I think what’s often true in that case is that pro-lifers actually do believe fetuses should have a right, but for many whose views are motivated by religious beliefs, also feel that the pro-choice promotes promiscuity which they view as immoral. And if you read/listen to conversations, this becomes clear. But understanding the politics, won’t bring up that aspect of the motivation, or maybe even will be somewhat in conscious denial of that motivation while recognizing it on a sub-conscious level. I thin for rationalists, the “other minds fallacy” plays big here, especially in regards to people not being aware of their own mental processes.

          So it’s not a case of absolute true/false statements.

          I used the abortion debate as an example, but I think this applies to lots of these sort of discussions on both sides. People aren’t usually completel lying about their motivations, but presenting them in a way which reduces their complexity.

          An interesting aspect of this is that often what a tribe accuses the other side of secretly believing says a lot about how the accusing tribe views the situation.

          In anarchist circles, both right-leaning and left-leaning anarchists accuse the other side of secretly being authoritarian. Since both groups describe themselves as being motivated by anti-authoritarian beliefs, this accusation should in spherical cow world represent a common ground for discussion.

          But what you find is that both sides have a very different notion of what “authoritarian” means.

          Right leaning anarchists tend to view authoritarianism as the power of the collective over the individual, whereas left leaning anarchists tend to be more worried about the authority of an individual over another individual. One could make the argument that both ultimately amount to the same thing, but both sides clearly are much more concerned with one then the other, which leads both sides to accuse each other of being motivated by authoritarianism.

          This is complicated in terms of its status as “true/false” because the relative truth or falsehood of that claim depends on slightly different definitions of a similar word that are based on a priori sentiment, not objective measurable fact.

          And of course, tribes can very easily see the duality in the other tribe’s thinking, but miss it in their own.

          Take the issue of “freedom of speech.”

          It is very common for right wing people to be concerned about freedom of speech, but be accussed of only caring about it in a way that benefits their interests.

          What I have observed is that their not “lying” about caring about- but they tend to mentally view it in such a way that it supports their own partisan positions. This isn’t lying about being motivated by free speech concerns, it’s a “politics is the mind killer” thing where those concerns should apply on a first principle basis to a concern for the free speech of their opponents, but in practice doesn’t, because they view it in such a way that the definition slightly changes to benefit their views on any given instance of a related debate.

          Left-wingers tend to so the same thing on the same issue- while the use of the term “free speech” is not currently in vogue amongst left wingers, they do complain about the idea of “censorship” when it applies to their views.

          So trying to sort this all out as a rationalist-(bottom line), it’s rare that people are simply “lying” about their motivations- but it’s also true that their motivations tend to be more complex then they openly describe, or possibly are even consciously aware of. People that accuse them of “lying” are usually wrong that they are motivated by what they claim to be motivated by, but are correct if you define “lying” as not giving a full and honest account based on self-analysis of their total motivations and beliefs regarding a subject.

          • Alsadius says:

            This is a good point, but I’ll also add that most people’s views form a coherent whole (though not always a consistent one). Typically, there’s feedback that tries to get rid of the really obvious inconsistencies. So it could just as easily be that people who worry about collective dominance become right-anarchists, and people who worry about individual dominance become left-anarchists. Or that right-anarchists hang out with other right-anarchists, read a lot of Reason and Murray Rothbard and such, and thus hear a lot of horror stories about collective dominance and begin to push their politics in a direction to counter that threat.

        • deciusbrutus says:

          I say that they’re lying, a strategic lie that causes people to put in more effort so that they end up believing that they will win by a perfectly rational estimate of winning chances plus their own bias towards winning.

          Because it’s much more motivational to tell people ‘our enemy is cheating’ than to tell them about planning fallacy, typical mind bias, and all of the other reasons why they think they are doing better than they are.

        • Simon_Jester says:

          Most Republicans have spent decades listening to ‘trusted’ right-wing news sources that make a point of telling them every election cycle that the Democrats are engaging in massive fraud. Said right-wing news sources do not report that all investigations turn up no evidence of these huge fraud schemes.

          Suppose Fox News started holding daily programming on how Bigfoot was scary and real. Suppose that ten years later the Republican Party started passing the Anti-Bigfoot Act to protect Americans from Bigfoot. You could easily go around polling Republicans and find that they genuinely believed that Bigfoot was real and a threat.

          But it would not prove “the Republican Party as a collective entity is being sincere here.” It would prove “the Republican Party contains or is allied to a mechanism for manufacturing tactically useful beliefs among its own supporters,” which we already knew.

          • EchoChaos says:

            Same thing exists in the opposite direction with “white supremacists”.

          • Simon_Jester says:


            I question the implication that white supremacists are as rare as Bigfoot and largely fictitious as a threat.

            Now, if the left wing media were claiming something like “thousands of people are murdered every year by white supremacists and it gets covered up,” THAT would be parallel to, say, “Democrats commit voter fraud to the tune of several hundred thousand votes a year, ro even millions, and it gets covered up.”

            But I cannot recall seeing such claims.

          • anonymousskimmer says:

            White supremacists and election fraud both had their heydays in the mid 1800s to early 1900s in the US.

            Open organizations of white supremacy still exist, though in not nearly as great numbers, and do not engage in racial voter suppression to any extent that I am aware of. Open organizations of election ‘fraud’ (e.g. Tammany Hall) no longer exist – the closest thing are union or PAC endorsements, which are nothing more than endorsements.

      • aristides says:

        So I’m a Republican who has had a lot of closed door conversations with other Republicans about fraud. These range from average uneducated republicans, to law students that went to work for think tanks and corporations. I am on the side that thinks we should make voting easier, and even when they are arguing to convince me, a Republican, they do not argue that voting restrictions will make it easier for Republicans to win. There argument is very deontological, voting fraud is bad, and we need to stop it no matter the cost. It does not matter if the restrictions are inconvenient for some people and lead to 10 citizens that should have voted not voting for every one fraudulent vote that is prevented. The consequences do not matter to them. What matters is preventing fraudulent voting.

        As another example, my father has voted straight Republican for decades, and has never voted for a constitutional amendment in Florida, except one. The only constitutional amendment he voted for was to let felons vote in elections after there time was served. Florida elections are famously close, so he knew full well doing so, might make Florida turn blue. But he didn’t care, it was the right deontological thing to do.

        Disclaimer, I am only talking about average voters. Party establishment politicians may very well be as cynical as you are suggesting. I wouldn’t be the least bit surprised.

        • HeelBearCub says:

          Of course they aren’t going to make the cynical argument. There are so many reasons not to make the cynical argument out loud.

          But if you scratch the surface you can hear what the actual fear is. Someone is putting the idea that high levels of voting mean Republican losses into Trump’s ear.

          You can then say that there is a sincere belief that what that really means are high levels of fraud, but when it is expressed the way it is, it’s hard to credit as the core, underlying belief.

          • Alsadius says:

            The funny thing about listening to dog whistles is that you’re really guessing about what it is that the dog hears. For that quote, I’d wager >90% of Republicans would hear “They had levels of voting, that if you ever agreed to it you’d never have a Republican elected in this country again”, and immediately append “Because the Democrats would cheat so much” to the end of it.

            Perhaps Trump actually did mean it like you say – I don’t know the man’s mind. But you’re claiming that a large part of the country is all thinking that they need to rig elections, and yet there’s absolutely nobody saying it out loud (even behind closed doors, with allies and friends), other than one vague comment from a man famous for being incoherent.

            It strains credulity to say that when Trump talks about cleaning forest floors to prevent forest fires it’s incomprehensible gibberish to much of the left, but that as soon as he talks about voting in a tangential comment you know exactly what he means, how he means it, and that it’s proof of the secret plan that you’ve suspected all along. Again, these are fully general arguments. They could be made about literally anything. For example, does this tweet mean that the Sanders campaign secretly wanted to enact a communist dictatorship?

          • Mary says:

            Except that you, right here, make the cynical argument, thus undercutting your claim that people wouldn’t.

          • Alsadius says:

            @Mary: Which cynical argument, sorry?

            Edit: You’re saying that HeelBearCub is arguing a cynical political argument, and thus not credible as an anti-cynicism advocate?

          • Theodore Ehrenborg says:

            I think that voting by mail will raise turnout, but I’m not sure whether either party will benefit.

        • DaveK says:

          I posted another comment, but I think the thing is- in this case, people are actually motivated by the argument they say they are motivated by- but those arguments tend to be things that will also benefit their political tribe.

          The process is something like, influential republicans see that a certain position is bad for their pragmatic interests, make an objection based on those interests that also coincides with some genuine principle, and that argument gets “passed down” from leaders or influential voices to the “ordinary” republicans, who then genuinely believe in said argument.

          I think though that any member of any political tribe finds it very easy to believe in the specific principles their tribe is arguing for.

          I have speculated that had Trump locked down the country back in February, the somewhat partisan split regarding republicans being more concerned about lockdowns/the economy and dems being more concerned about the virus would have been reversed. The arguments wouldn’t have been exactly the same, but i could see dems arguing that TRump was a mad dictator, and republicans arguing the lockdowns were necessary for national security, and dems didnt care about the lives of old people.

    • HeelBearCub says:

      As for the political angle on it, the usual complaints I see about it from the right side of the political spectrum all boil down to fraud. It’s easier to fake your identity/citizenship/felon status/etc. if you’re using ID sent by mail than if you have to show up in person.

      Conservatives in the US have historically generally been in favor of by-mail absentee ballots.

      For one very simple reason, their use has generally favored conservatives. The older and more well to do you are, the more likely you are to apply for and complete an absentee ballot.

      They’ve generally been against in-person early voting.

      The arguments about fraud in this context ring quite hollow.

      • Alsadius says:

        I can’t speak to the history of it per se. But in the time I’ve been paying attention to this issue(maybe a decade?), I don’t recall Republican views sounding particularly pro-mail-ballots or anti-early-voting.

        Someone who makes those arguments cannot consistently complain about fraud on mail-in ballots, I agree.

        • HeelBearCub says:

          Republicans are anti vote-by-mail. They are pro absentee ballot by mail. You need to make that distinction if you want to understand what I am saying.

          If you really need examples of all of the places that Republicans have reduced access to early in-person voting, I can dig them up. In NC, they reduced in-person voting days, locations and even reduced the times to be during typical work hours. They als cut off a specific Sunday access to in-person voting.

          You know what demographic specifically organized around voting on that Sunday? Black churches.

          • albatross11 says:

            HBC: I think everyone is in favor of allowing absentee ballots for people with a good reason, and that’s distinct from wanting universal mail-in ballots.

            OTOH, I’m 100% sure that both parties are calculating their advantages when considering any change to how elections are run. I think there are sensible arguments for and against universal mail-in ballots. I think it’s very hard to make similarly sensible arguments against, say, keeping the polls open later or having early voting available or stretching the election out over 2-3 days or making election day a holiday.

          • Scott Alexander says:

            What is the difference between absentee ballot by mail and vote by mail?

          • HeelBearCub says:

            @Scott Alexander:
            Distinguish between these four types of common non-election day voting:
            1) Requested absentee, with excuse (i.e. you are required to give one of several defined valid reasons why you can’t vote on election day), ballot by mail
            2) Requested Absentee, no excuse, ballot by mail
            3) Requested Absentee, no excuse, in person ballot
            4) Universal (without specific request) ballot by mail

            Those are, roughly speaking, in order from favored by Republicans to not-favored by Republicans, with the break line most commonly being at 2, or between 2 and 3.

          • HeelBearCub says:


            HBC: I think everyone is in favor of allowing absentee ballots for people with a good reason, and that’s distinct from wanting universal mail-in ballots.

            Please do note the number of comments in this thread of people saying they are firmly, sincerely, against requested absentee ballots by mail due to fraud.

            The reason I am making the point that traditionally Republicans have been in favor of absentee ballots is to point out that these claims are novel.

          • EchoChaos says:


            Note that the order 1-4 also selects strongly as you move up the list by people who can plan ahead.

            E.g. if I already know what I’m doing November 3rd well enough to request an absentee ballot ahead of time, I am the sort of person who is put together and high time preference.

            On the other end, permanent mail in ballot just requires you to remember you have it at least once in the time period between receiving your ballot and Election Day. You can even be reminded and have someone else take all the effort necessary short of actually filling it out in states with legal harvesting.

            The reason that Republicans want a higher percentage of voters with high time preference seems obvious.

          • HeelBearCub says:

            @Echo Chaos:
            If that’s your argument, I think you just conceded mine.

            It is also not quite true as stated, but I don’t need to articulate how it is untrue to point out that you just gave away the game.

          • EchoChaos says:


            If that’s your argument, I think you just conceded mine.

            What is your argument? Mine is that voting is a weighty matter and if those who can’t take it seriously enough to actually plan ahead in order to do it lose that ability, it doesn’t bother me at all.

            Note that I have no moral attachment to universal suffrage. Which system produces the best results is most important to me.

          • Lambert says:

            The thing about universal suffrage is that it’s a good chestertonian fence.
            Keeps out everyone on both sides who would make it harder for their opponents to vote.

          • EchoChaos says:


            Nit: universal suffrage doesn’t exist and never has.

            Universal adult suffrage is relatively young. Less than 50 years in some very well governed countries (e.g. Switzerland), less than 100 in others (France), almost exactly a hundred in the USA.

            For most of those years the USA had everyone voting on election day unless you had a specific excuse and we were very well governed.

            There is a real cost to voting, and I understand that. I also like that, because the weightier it is, the more people will value it. Someone who values the vote less than an hour of their time is probably not someone whose vote is a net positive.

          • zzzzort says:

            Ah yes, poll taxes, what could go wrong.

          • thedufer says:

            > Someone who values the vote less than an hour of their time is probably not someone whose vote is a net positive.

            I’m not sure why you think this is true. An hour is much cheaper for someone who is retired, wealthy, etc. For someone who doesn’t eat this Friday if they don’t work that hour, it’s pretty damn expensive. I wonder which side the latter votes for…

            This is why universal suffrage is so useful – without a clear bright line, anyone can come up with a rule and a corresponding story about why that rule is valuable. Oops, that rule disenfranchises their opponent’s voters? Just a coincidence, they’ll assure you.

          • Rick Jones says:

            Alexandra Petri explains the answer to Scott’s question below. “An absentee ballot is sent in the mail, whereas a mail-in vote is mailed in. Sending ballots in the mail is famously secure, but mailing in votes is ripe for all kinds of fraud and confusion.“

          • An hour is much cheaper for someone who is retired, wealthy, etc.

            Cheaper still for someone on welfare.

            Which group do you think is more likely to vote for which party — people on welfare or the working poor?

          • anonymousskimmer says:

            What do you mean by “welfare”, David? And does this concept even exist as a unitary thing anymore in the United States?

            Unrelated to my above questions: If SNAP/TANF/whatever only allows you to buy fresh groceries, then more of your time is being spent preparing meals versus those who can afford pre-prepared meals. Is this the case? Demographically I don’t know.

          • Both people on welfare and people in low-paid jobs are more likely to vote Democrat than Republican; the former by more than the latter.

            Thanks. I’m not surprised.

            The point being discussed implies that the switch to mail-in voting will reduce the fraction of voters who are on welfare, increase the fraction who are in low-paid jobs, making the effect on the D/R ratio indefinite.

          • albatross11 says:


            I think it’s entirely possible that:

            a. Republicans widely believe (or the party hierarchy believes) that X will hurt Republicans and benefit Democrats.

            b. X will actually have no effect on the balance of power, or will even benefit Republicans and hurt Democrats.

            As an example of this, if I recall correctly a later analysis of the 2000 election in Florida showed that if the Bush campaign’s original proposed counting rules had been followed, Gore would have won. As another example, basically all political insiders in the Republican party were convinced that a Trump campaign was a complete joke and would lead to a disastrous defeat of the party if he somehow got the nomination.

            I suspect this is relatively common–unless they’ve actually run the experiment and seen who benefits, the political consultants are probably wrong almost as often as they’re right. (The exception being when they can lean on some very simple heuristic like “keeping more blacks away from the polls helps Republicans” or “getting poorer people to vote helps Democrats.” )

          • Tatterdemalion says:

            The point being discussed implies that the switch to mail-in voting will reduce the fraction of voters who are on welfare, increase the fraction who are in low-paid jobs, making the effect on the D/R ratio indefinite.

            I am extremely sceptical of this. I think that the easier/harder you make voting, the higher/lower the proportions of both unemployed and low-waged voters you will see, and the more you will benefit Democrats/Republicans.

          • albatross11 says:


            This year, it will also increase the number of old and sick people who are willing to come to the polls. It’s not at all clear to me which side this benefits, but a hell of a lot of people are going to say “do I want to stand in line with 100 other people for half an hour to vote today, knowing that if I get this virus, I’m liable to die of it?” And then they’re going to slip their shoes back off, sit back in their easy chair, and go back to reading a book or watching TV.

      • craftman says:

        For one very simple reason

        Dude, it’s 2020, the “one very simple reason” is that I don’t want to go stand in line with a bunch of schmucks to poke a piece of paper when I could just do it from the comfort of my own home. I’ve done mail in ballots in Colorado since at least 2008 (I remember doing my first vote, in 2004, in person).

        I am 1,000% on the side of mail-in voting and it is purely due to convenience. I don’t care whether Team RED or Team BLUE benefits more. Why are we lining up at a school?

        Maybe I’m alone in this feeling. Certainly being a small-l libertarian who mostly votes for the propositions and ballot initiatives helps…

        • HeelBearCub says:

          Again, this was a response to the claim that conservatives have been generally against mail-in absentee ballots due to concern about fraud. That claim is not true.

          I’m glad you are in favor of mail-in ballots. I wasn’t arguing against them. I believe mail-in ballots can be made to work, as born out by the wide range of political entities successfully using them.

      • Mary says:


        Why on earth would different people at a different time supporting something impose a duty on anyone to support it?

        • HeelBearCub says:

          Well, Trump did support absentee mail-in ballots a whole two day ago. You are quite right that he may have already changed his position.

          Then again, Trump is the political quantum particle, so it’s impossible to know his current position and what his next position is likely to be, so I won’t take any guesses.

          Seriously though, it’s not like this isn’t a well known, consistent fact. The GOP spends consistent dollars on encouraging absentee vote by mail.

          • Corey says:

            Soldiers are a big constituency for vote-by-mail, and lean conservative, is the conventional wisdom.

          • HeelBearCub says:

            More impactful is the elderly.

            Armed Forces members aren’t as reliable (to vote) or as numerous, although I believe those that do vote are believed to favor conservatives in their voting patterns. IIRC, there should be vote totals in presidential years that would reflect this, as certain overseas ballots are counted separately (as the overseas voter has a state, but no local address). But I can’t remember the particulars.

          • Alsadius says:

            IIRC, the military vote and the elderly vote are about equally conservative. Maybe 70% or so, in both cases. (There’s no demographic where the Republicans win 95% like the Democrats do with blacks, so far as I’m aware)

    • eremetic says:

      You seem to be under the impression that America requires ID to vote. Actually, proposing this is considered extremely right-wing.

      • HeelBearCub says:

        Every voter registered in America has to provide some form of ID at the time of registration.

        • EchoChaos says:

          But not to actually vote. @eremetic is correct that requiring the ID to vote is indeed a very right-wing proposal.

          • HeelBearCub says:

            Can you vote without providing some form of ID is false. You can’t vote unless registered, you must provide some form of ID to register.

            The proposal under contention is requiring an ID each time you vote. In addition, usually not just some form of ID, but a picture ID provided by the government.

            It’s important to distinguish between these positions.

          • Alsadius says:

            FWIW, in Canada, having government-issued ID to vote is expected. There is an alternate process where you sign an affidavit that you are legally entitled to vote, and if you do that you can vote with no ID, but it’s incredibly rare. (I did it once, for a municipal election while I was in university, but I don’t know of anyone else who has).

            We also do not have any formal registration process the way the US does. TBH, I don’t know how they maintain the voter list, but they do keep it at least somewhat up to date. Maybe they dump tax/census records into their database?

          • Eltargrim says:

            @Alsadius: Census doesn’t happen often enough. It’s a combination of information from tax returns, drivers licensing/ID, Canada Post, and the provinces’ electoral information.

          • sharper13 says:


            You’ve been generally very reasonable in this voting by mail discussion, but this is inaccurate:

            Can you vote without providing some form of ID is false. You can’t vote unless registered, you must provide some form of ID to register.

            You’re assuming that the person voting is the same as the person who registered to vote under that name, which is the thing actually in question. Assuming that your opponent’s argument is false doesn’t prove it to be false.

            In reality, investigative reporters have demonstrated that it’s fairly easy to walk into a polling place without ID and end up with a ballot designated for someone else who previously registered to vote.

            In terms of vote by mail, the information a voter must currently provide to validate their ballot is the type of information also available to a partisan who wants to vote for them. There have been cases of people successfully registering non-existent people online to vote by mail, of hundreds of absentee ballots requested for addresses where no one lives, etc… but those are penny-ante compared to organized legal in some places Party efforts for ballot-harvesting, which provide the opportunity for mass vote swings.

            Where everyone voted by mail, nobody can prove that a ballot-harvesting operation collected properly signed and sealed ballots and never looked at them, but it’s also pretty tough to prove that they didn’t discard the ones who didn’t vote the way they liked, or didn’t substitute a different ballot, or didn’t just collect ballots from mail boxes and fill everything out themselves.

            Our current system has a lot of room for improvement, but vote-by-mail isn’t an improvement, it’s a regression, in terms of security. Only thing worse might be voting online (not in the theoretical sense, but in the sense of how real-world State governments would implement it).

          • Corey says:

            In reality, investigative reporters have demonstrated that it’s fairly easy to walk into a polling place without ID and end up with a ballot designated for someone else who previously registered to vote.

            Sure! And serious efforts have failed to find the traces this would leave behind if successful. (see upthread: people on the list of who voted, who did not actually vote)

            ETA: Correct that vote-by-mail is less secure. Attempts to secure it more (requiring witnesses and, God help me, notarization) push down the response rate while not adding a lot of security.

          • Steven J says:

            “You can’t vote unless registered, you must provide some form of ID to register.”

            Not true in all states.
            E.g., Minnesota has same-day registration, and there are several ways to prove residence that do not require an ID, and are easier to fake.


          • HeelBearCub says:

            @Steven J:
            The one thing there seems to be inconsistent with what I’ve said elsewhere about how one proves identity is the “registered voter in same precinct/elder care worker can vouchsafe”.

            That’s a new one on me.

            However, I believe it’s likely those same day registration ballots are typically provisional ballots. If so, the registration process still needs to be completed in order for the cast ballot to counted, which would allow party election monitors to challenge a pattern of fraudulent vouchsafing.

          • HeelBearCub says:

            That’s technically correct (best kind).

            Similarly, someone could obtain a fraudulent ID document, so we still wouldn’t be able to say that person voting had a valid ID. If you say this presents to large a hurdle, remember we are positing a mass, coordinated, fraud that swings an election. If you can organize thousands upon thousands of people to show up at the polls impersonating voters on the rolls without that plot being revealed, you can probably make some fake IDs.

            The question which I’ve been primarily engaged with is whether in-person voter fraud represents a risk to election integrity, not whether it’s technically possible.

          • Steven J says:

            “However, I believe it’s likely those same day registration ballots are typically provisional ballots.”

            Incorrect. There is a bill to make same-day registrations into provisional ballots (with the predictable partisan split), but as of now, same-day registrations are treated the same as any other registrations, with no additional scrutiny after being cast.


          • HeelBearCub says:

            Interesting. As I said vouchsafing isn’t something I’m familiar with. Something of an elegant weapon from a more civilized age. I wonder if there are any statistics on how many registrations are vouchsafed.

            I’ll note that the article you are pointing at doesn’t necessarily support the idea that the split here is purely partisan, although I’m inclined to believe it is generally true.

            Julie Hanson, who oversees elections in Scott County and is a Minnesota Association of County Officers director, said a provisional ballot system would require extra personnel to process the provisional ballots within a tight deadline. She said that could be particularly hard on smaller counties that don’t have a big full-time election staff.

            “This is going to be a big deal for us. Our taxpayers are going to bear the burden of what this is going to cost,” Hanson told the committee. “Just the cost of envelopes and ballots and all of these kinds of things alone. Elections are very expensive.”

            This position, that we should resist spending on the machinery of elections, is frequently a conservative position, as we would tend to expect. Scott County seems to be fairly solidly Republican, although they do seem to have some DFL (Democrat-Farm-Labor Party reps in the state house. Obviously that doesn’t mean Hanson is a conservative, but it’s worth noting the conservative argument.

            Also worth noting that, generally speaking, Minnesota is a little bit of an odd bird in US politics, as the fact that primary left of center party is a unique-to-Minnestota coalition party.

          • Alsadius says:

            I can’t speak for other conservatives, but I tend to respond to complaints about the cost of elections with eye-rolling. Democracy is worth it. An election might possibly cost 0.1% of the nation’s annual budget, and they don’t happen annually. Good government is worth orders of magnitude more than that.

          • Edward Scizorhands says:

            whether in-person voter fraud represents a risk to election integrity

            I vaguely mentioned it elsewhere, but I’ll be more explicit here: one advantage of doing most of an election on one day is that potential fraudsters don’t have time to figure out how much fraud they need to win.

            Say that with exit polling on election day you might figure out that you needed to fraud up around 40 votes to flip some key position. If you figure that out while there is still time for showing up in person, you can probably fraud those up.

            But the more votes you have to fraud up, the bigger the chances of your scheme being discovered, so you would rather not take the risk if it doesn’t matter.

            There is a multiplier effect where tiny costs on voters, like showing an ID, translate to big costs on fraudsters, who would need to prepare a bunch of fake IDs ahead of time, which creates more risks of their scheme being discovered, especially since they don’t really know how many they would need. (I support provisional ballots for people who can’t get or forgot their ID, provided we do some verification and allow challenges for allegations of fraud prior to counting; and, after the election when we have more time, audit a statistically significant number of them.)

          • Controls Freak says:

            But the far more common form of ID is providing the last four of your social and having it matched electronically with your name.

            I don’t think this is what anyone is talking about when they talk about Voter ID, which is why I think it’s sort of disingenuous to say that, “Every voter registered in America has to provide some form of ID at the time of registration.”

            FWIW, registering to vote is by far the easiest official document to procure, such that when I changed state residencies, I looked through the list and planned out how I was going to switch over my documentation and realized, “Wow. Even if I don’t need the voter registration for a long time, I pretty much have to start with that one. They require nothing for it, and I can use it for all of my other documents.”

            If the bar to any sort of online identity fraud was, “Can correlate a name with last four of an SSN,” and this could be done in any sphere connected to something we care about (where it, ya know, cost someone money or something), it would be rejected in literal nanoseconds.

        • J Mann says:

          If that’s the case, then what’s the problem with showing ID at the time of voting?

          Is it that you can register with forms of ID (bills sent to your house, etc.) that aren’t accepted at the polls, or is there a material group of people that is capable of producing an ID to register to vote but not to vote?

          • DeWitt says:

            Is it that you can register with forms of ID (bills sent to your house, etc.) that aren’t accepted at the polls

            As best I can tell, yes.

            The other part of the issue is that the people for whom this is true most often vote Democrat, which is why this has become a partisan matter.

          • Corey says:

            Honestly it depends on how the requirement is done.

            Does the address have to match? If so, how many people update their driver’s licenses when they move to different apartments in the same town? (Nobody even updates their voter registration, almost everything we do on election day that’s not a perfect scenario voter involves unreported moves)

            Does it have to be unexpired? What if you don’t drive anymore?

            For photo IDs, do you trust poll-workers to be empowered to turn you away if they don’t think your picture looks like you? (Consider how literacy tests used to be applied). In NC the one election I ran when the ID requirement was live, all 3 judges (who always include one D and one R) had to agree it didn’t look like you, then you could still vote a provisional ballot even if rejected. But do other places do that?

            Also college students. (I know lots of people don’t think college students should vote in local elections, but I don’t think that’s coherent unless you go full landowners-only, which at least is an ethos)

          • HeelBearCub says:

            One form of ID is providing utility bills, etc.

            But the far more common form of ID is providing the last four of your social and having it matched electronically with your name. In fact, that system is also what provides evidence of citizenship via your Driver’s License Number as well. The DMV is collecting those last four digits and verifying them, IIRC.

            Altogether, the objection is that these create impediments to voting while also providing no substantial reduction in actual voter fraud. The people who can’t imagine that showing a driver’s license at the polls could possible be a burden are basically just committing a kind of typical-mind fallacy. There are tons of people who don’t have DLs, they just aren’t you.

          • albatross11 says:

            I think provisional ballots are required everywhere.

          • Edward Scizorhands says:

            When Jill Stein raised $2 million from suckers to “audit the election,” NPR had stories about how nearly every election board already audits their own election. It’s part of the job. Auditing a statistically significant sample of absentee/provisional ballots should[1] be part of this audit, too.

            [1] as in “ought”

          • Corey says:

            @albatross11: That’s right. Everyone gets the option of a provisional ballot, no matter what, even if they tell you upfront they’re not registered. The folks downtown sort it out later (in front of party reps, candidates’ lawyers and such). (Disclaimer: I only know for sure about NC)

          • Alsadius says:

            If you look at the details of ID proposals, they’re usually a lot better than just “show a license”. For example, I looked into North Carolina’s proposal the other day. And for reference, this one has been quashed by a judge who felt it too onerous – I’m quoting from that judge’s decision here.

            Ten different forms of ID are authorized:
            1. North Carolina driver’s licenses;
            2. Other nontemporary IDs issued by the Division of Motor Vehicles (“DMV”);
            3. United States passports;
            4. North Carolina voter photo ID cards;
            5. Tribal enrollment cards issued by state- or federally recognized tribes;
            6. Certain student IDs issued by post-secondary institutions;
            7. Certain employee IDs issued by a state or local government entity;
            8. Out-of-state driver’s licenses and nonoperator IDs (if voter is newly registered);
            9. Military IDs; and
            10. Veterans IDs.

            #4 in particular is worthy of some analysis, since that’s the one designed to deal with these issues.

            S.B. 824 further provides for the issuance of free “voter photo identification cards” upon request. Voters can obtain these IDs in two ways. First, voters can visit their county boards of elections and receive IDs “without charge.” 2018 N.C. Sess. Laws 144 § 1.1.(a). To obtain an ID from a county board, a voter must visit in person and provide her name, date of birth, and the last four digits of her social security number; no additional documentation is required.

            Second, voters over the age of seventeen are eligible to receive a free nonoperator ID card from the DMV. Id. § 1.3.(a). Although this method does require certain underlying documentation to prove identity, such as a birth certificate, the state must supply the necessary documents free of charge if the voter does not have copies. Id. § 3.2.(b). Relatedly, if a voter’s “driver’s license, permit, or endorsement” has been “seized or surrendered due to cancellation, disqualification, suspension, or revocation under applicable State law,” the DMV must automatically mail that voter a special replacement identification card which can be used for voting. Id. § 1.3.(a).

            So basically, there’s an office in every country that can give them a free ID that enables voting, and in addition to that every DMV office is also able to give them one for free (and in many cases, will do so proactively). And they have a lot of DMV offices, at first glance.

            This sure looks like a good-faith effort to give every legal voter a chance to vote while still implementing an ID requirement. Maybe they can do better somehow, but I can’t think of any way that it could be improved offhand.

          • HeelBearCub says:

            When I’ve gone to the DMV in NC (in populated areas) it’s a minimum one hour wait. Frequently longer.

            And in order to stay compliant with the voter registration laws, you have to do it every time your address changes.

            This is not cost free, even if they charge no dollars.

          • Evan Þ says:

            @HeelBearCub, where and when were you going? When I was going to the NC DMV back twelve years ago when I still lived there, I never had to wait anywhere near an hour. I was in the Triangle area, which I think counts as fairly populous.

            Sure, it’s an additional burden. But as “things to do when moving” go, it’s nowhere near the most onerous.

          • Alsadius says:

            @HeelBearCub: What’s your proposed alternative? If you need to re-register at the new address, the waiting time is the same. If you don’t, where does the security come in?

          • HeelBearCub says:

            @ Evan Þ:
            The DMV’s funding for offices isn’t keeping pace with population growth. IIRC, Republicans cut the DMV office budget when the came into power in 2011.

          • HeelBearCub says:

            Changing your voter registration is as simple as filling in a form and mailing it in. Requiring a picture ID at the polls procured with a trip to the DMV is not as friction free.

        • Matt says:

          In 2001, my college roommate (German Citizen) presented his German Passport to get a Georgia driver’s license, and was immediately registered to vote in Georgia because he didn’t check the ‘opt-out’ box stating that he was NOT a US citizen. Not ‘you must prove you’re a citizen’. It was ‘you have used ID that proves you’re not a citizen to get your driver’s license, and you must take an extra step for the secretary of state to act on that information and deny you the ability to vote’.

          I have no idea whether he actually voted in 2002 or 2004, but I know he was registered to do so, and he told me he intended to do so. What would stop him?

        • SamChevre says:

          I don’t think this is true–I certainly don’t recall providing any ID when I registered after moving.

          I was at the RMV, but the registration form was separate from the license forms.

          • HeelBearCub says:

            A) Moving within a state? Or moving to a different state? Different rules apply to a new registration and a changed registration.
            B) What information did you provide to the MV otherwise? Was the registration form printed for you, pre-filled by the MV department, and you simply signed it? All of that information is (likely) transmitted electronically and a scan of the form is also transmitted. If they had last four of your social already, the result of that check would have been passed on automatically. That’s a part of the “Motor Votor” legislation that came with HAVA post-2000, along with the requirement to be able to register at the DMV.

          • SamChevre says:

            I’ll take that back. I just don’t remember clearly enough; I think the registration form was blank, but don’t remember for certain how it was connected to the other forms.

            I was moving to a different state.

            I do know (because the voter role is by address) that the former owners of my house stayed on the voter roles for 5 years after they moved out of state and I bought it.

          • HeelBearCub says:

            Yes, voter role cleanup after someone moves out of state, or in state but doesn’t register in the new location, is not nearly as simple a process as one would like. Perhaps ironically because the voter roles belong to the states independently. It’s also important to remember that voting is a constitutional right, and the government can’t deny it to you arbitrarily, so various things you might think would be reasonable to do probably aren’t, actually.

            When you register in a new state, removal from the voter roles in the previous state is entirely dependent on you providing accurate information to the new state about your registration in the new one. The new state then informs the old state that the voter has indicated they were registered in the old state. Absent a mandated federal voter register, there isn’t any way to guarantee that voters aren’t registered in two states at once. Resistance to a federal voter registry would, IMO, very likely cut across party lines

            In NC, IIRC, failing to vote in two federal elections in a row will cause a non-forwardable letter to be sent to the voter’s address. The USPS will return that piece of mail unless the voter is still at the address, and a process is followed to remove the voter from the roles. I can’t remember precisely the steps in that process. I think you remain in the role for one more election, but in an inactive state, and then are removed if you don’t vote.

            I want to say that HAVA did something to standardize that process, but I can’t remember the specifics. I don’t think it actually materially changed the process in NC, but I could be wrong.

          • Alsadius says:

            A friend once looked into getting himself removed from the voter registry in the state where he went to university (and that was >20 years ago). It was such a headache that he didn’t bother. That is not a good sign about the cleanliness of the voter rolls.

          • HeelBearCub says:

            It depends a great deal on the particular state. States that have robust, consistent, process based voter roll maintenance will be better.

            As to the difficulty of removing someone whom you claim to be you from the voting rolls, that sword has at least two edges. Again, nationwide, automatic voting roles would solve the bulk of this, but you it’s not particularly feasible to enact in the US at this time.

    • Corey says:

      Election nerd / polling place supervisor here with everyone’s periodic reminder that impersonation CAN be detected after the fact.

      If someone impersonates you and votes, either you show up on the List of Who Voted even though you did not, or you try to vote and get told you already voted, or you voted before the attempted impersonator and they get thwarted because you already voted.

      All of these things can be counted, by anyone (including highly motivated parties like losing candidates or “buses of illegal aliens” truthers) – go download the List of Who Voted, ask people on it if they actually voted, follow up on inconsistencies. (Yes that’s a big and error-prone job, and everything will probably be swamped by Jr/Sr-at-the-same-address mixups).

      • Edward Scizorhands says:

        We can do all that. Do we?

        • EchoChaos says:

          I do, but I am weird.

        • Corey says:

          OP’s linked article about the Washington gubernatorial election sort of implies they did – the GOP raised the alarm about odd things in the List of Who Voted. The local news found 8 dead people on it IIRC, though one died after election day and one was a husband/wife mixup. It doesn’t say explicitly that they tried to find people who didn’t actually vote, as opposed to e.g. felons on the list.

          • DaveK says:

            8 seems like it would be far more likely due to error then fraud.

          • Alsadius says:

            8 was the number in a newspaper report, and it turned out most of them were erroneous (or in one case, legal – the guy voted before election day, and died after voting but before the election). I think there were a few more in the official complaint, because they checked into more counties than the newspaper did, but you’d expect the results to be the same.

            The bulk of them were felons, FWIW. Non-citizens were the next largest group, I believe, and there were minuscule numbers of dead people and double voters. So about what you’d expect of non-systemic whoopsies. No really obvious fraud was seen in the data, just normal human errors.

            It was still enough to swing the election, and in principle if it could be stopped without imposing costs elsewhere we should do it. But it provides no evidence for the “millions of illegals are voting to try to keep the GOP out of office” arguments that Trump likes to make. Presumably the systemic frauds are better at covering their tracks if they exist, but it’s also quite possible that they don’t exist.

      • HeelBearCub says:

        Note that the issue of people trying to vote and being told they already voted is the primary reason we can rule out significant numbers of voter impersonations. Sure, it would be better if everyone looked at their voter history, but that applies the most when turnout is the lowest.

        If 1% or even 0.5% of voters showed up at the polls and were turned away due to supposedly having previously voted it would be a gargantuan, obvious, blatant signal.

        • Edward Scizorhands says:

          If people are taking names from the voter rolls random to impersonate them, we would notice quickly.

          But people trying to win elections have advanced data analytics that tell them the odds of someone voting. They would select the people that are already unlikely to vote.

          • HeelBearCub says:


            But note that the people unlikely to vote are the ones who will be most targeted by get-out-the-vote campaigns.

            Thus, the closer the election (where fraud is most likely to make an impact) and the higher the impact of the election, the higher the chance that you encounter a great deal of these double votes.

          • Corey says:

            That’s still detectable but it takes heavy effort. On the other hand if you’re a losing candidate, you already have those same advanced analytics, so you can identify a nice subset of doors to go knock on to ask if they actually voted…

          • viVI_IViv says:

            But people trying to win elections have advanced data analytics that tell them the odds of someone voting.

            The “advanced data analytics” could boil down to just downloading the last 3 or so Lists of Who Voted and cross-check it with any list of residents (e.g. phone books). Somebody who didn’t vote in the last 3 elections is unlikely to vote this time.

          • Simon_Jester says:

            Unlikely to vote in the upcoming election is not the same as won’t vote.

            Suppose I was a malevolent Democrat who wanted to hand the state of Swingstatia to Joe Biden in the 2020 election. I go down the voter rolls and identify ten thousand people who have not voted in the last three elections.

            Swingstatia resident Joe Redhat is the Republican mirror image of all those ‘purity test’ Democrats you get who tend to scream at each other’s candidates for insufficient leftism.

            He didn’t vote in 2008 because he thought John McCain was a namby-pamby liberal, and didn’t vote in 2012 because he thought Mitt Romney was a Mormon and Mormons aren’t real Christians and it just wasn’t worth bothering. He didn’t vote for Donald Trump in 2016 because all the news outlets said Clinton was going to win and he couldn’t bear the prospect of going out to vote for Trump and having That Woman win the election anyway.

            But in this election cycle, he thinks the Republican presidential contender is both worthy and has a fighting chance. And while a bunch of liberal news idiots are rambling about coronavirus, Joe knows it’s all overblown. He’s gonna go vote.

            If I am a Democrat trying to fraudulently vote in Joe Redhat’s name, I am in trouble.

            Any large scale effort to vote in the name of other people is doomed to failure. MOST of the people in whose name you vote will not vote, and you’ll get away with it. But if you’re faking votes on a large enough scale to be reasonably sure it’s even going to make a difference, you’ll get caught.

            One of the hundreds or thousands of people you have going around claiming to be someone else will spill the beans to avoid going to jail for election fraud. Or they’ll claim to be someone the poll workers personally recognize and get caught on the spot. Or one of the people you’re impersonating- or rather, dozens of them- will decide to vote after all.

            A strategy for successfully stealing one person’s vote does not necessarily scale up into a strategy for committing effective election fraud.

          • If I am a Democrat trying to fraudulently vote in Joe Redhat’s name, I am in trouble.

            Are you? If you show up at the poll as Joe Redhat and the poll officials note that he has already voted, what happens? Do you get arrested? Held until they can contact Joe and make sure he really did vote, that it wasn’t a mistake by someone crossing out the wrong name on the list?

            I don’t know the answer, but my guess is that none of that happens. Assume I am right. If you want to commit the particular version of fraud being discussed, you and your friends vote early, making it unlikely that the person you are pretending to be has voted first even if he ends up voting. One time out of fifty you lose your gamble, he decided to vote this time and came even earlier, the poll officials don’t let you vote, and you go home.

            Anyone know if my assumption is mistaken? Any cases of someone being arrested just for showing up as a name already crossed out?

          • Simon_Jester says:

            Here’s how I would make it work:

            Joe Redhat II fills out a provisional ballot. Some days after the bulk of votes have been counted, the election officials notice they have two votes claiming to be Joe Redhat. Joe Redhat (the actual Joe Redhat, at a known address) is questioned on whether the provisional ballot reflects is valid his or not.

            If you tried to impersonate Joe Redhat, you are now an anonymous wanted criminal. You may well get away with the attempted crime, but you have failed to block Joe Redhat’s vote. If you impersonated Joe Redhat at several different polling places, you are now an anonymous wanted criminal in several places, with more witnesses. If this happens on a large enough scale to be even vaguely relevant to the election, you have the basis for a state election commission to freeze the overall state vote count and hold an investigation.

            The people doing this MAY get away with it personally, but it is far from certain that they will do so. Certainly the incident will gain national attention and probably backfire horribly on the fraudster.

          • You may well get away with the attempted crime, but you have failed to block Joe Redhat’s vote.

            That assumes that Joe votes. I thought we were considering the case where you were pretty sure, but not certain, that he wasn’t going to vote, since he hadn’t voted in the last two or three elections.

            So you impersonate ten people. One of them shows up to vote after you do. After suitable investigation it is determined that he is the real Joe Redhat. He gets to cast his vote, but there is no way to identify and eliminate the vote you case for him.

            If you are sufficiently unlucky, he shows up first. This time your vote becomes provisional and is eliminated. So the result is that a tiny fraction of the bogus votes are eliminated, the fact of bogus voting, or errors, is established, and nothing happens to the people who cast the bogus vote. They are almost certainly imported from out of town, making it vanishingly unlikely that anyone will see one, remember that he was the one who cast what turned out to be a bogus vote, and call the cops.

          • anonymousskimmer says:


            but there is no way to identify and eliminate the vote you case for him.

            I believe the method to eliminate this is to identify all ballots in some manner, ensure that all registered voters have at most one ballot to their names, and only then count all single-ballot-per-voter ballots, while investigating the multiple-ballot-per-voter cases in order to determine which ballots get counted and which don’t.

        • Alsadius says:

          Impersonation of real people, yes. False registrations are not obvious in that fashion. (And that’s true whether they’re registering a fake human, or whether they’re a real person who isn’t legal to vote.)

          But you’re right. Some genres of vote fraud can be ruled out fairly well. (They may exist in ones and twos, but there’s no systemic amount of them)

          • Corey says:

            Registrations are public record, so registrations of fake people and/or the ineligible can be rooted out in a similar manner. And of course if they vote they’re on the list of who voted.
            Though that’s a hard problem, e.g. my county used to have a Donald Duck registered (a real person, of similar age to the cartoon character; he’s died since I started working elections), but if members of the public were combing through the lists they’d probably file that as “obvious fake”.
            Election boards don’t have perfect insight into who is who and who is eligible, but I don’t know of any institution in a better situation (e.g. see how much gibberish is probably in your credit report).

          • Mary says:

            Can be and a 99 cents will get you a cup of coffee.

      • Election nerd / polling place supervisor here with everyone’s periodic reminder that impersonation CAN be detected after the fact.

        The fact of impersonation can be detected after the fact, although it sounds as though it usually won’t be. But who did the impersonation cannot be detected, at least by the methods discussed here, so cannot be punished.

        Which starts me thinking about ways of changing that. Face recognition is now pretty good, judged at least by Facebook’s ability to recognize pictures that I am in. So we could have a system where everyone’s face was on record from when he registered, a photo of the person claiming to be him was taken and compared in real time, and if they didn’t fit further checking would happen. Alternatively, if we develop databases with everyone’s face in them, which unfortunately may happen, photograph everyone before he votes and, if it turns out it was a bogus vote, seek out the person who did it.

        One could do the same thing with fingerprints, if there were an adequate database and recognition software.

        For a simpler approach … what happens if someone shows up to vote and the list of names shows that he has already voted? He could be either an impersonator, someone impersonated, or neither, due to some error, perhaps someone crossing out the wrong name. Is there currently a procedure by which, if he is an impersonator, he ends up arrested?

    • MisterA says:

      Conveniently, Donald Trump has a hard time remembering when he’s supposed to lie and give the fake reasons Republicans are for or against things, and when he’s supposed to just say the real reason. So he went on Fox & Friends last week and said, “The things they had in there were crazy. They had levels of voting, that if you ever agreed to it you’d never have a Republican elected in this country again.”

      Which is always the real reason Republicans want to make it harder to vote. Same as voter ID laws, opposition to making election day a holiday, and so on down the list. Every now and then a Republican politician will forget themselves and admit that it’s purely the fact that they know that if more people vote they are more likely to lose, so they want to make it harder to vote, end of reasoning, but usually they make up some stuff about fraud because it sounds better.

      • Alsadius says:

        I’ve already discussed this:

        I’ve been in the trenches in politics a fair bit over the years, and know more than a few stories about outright-illegal maneuvers. Even the people who tell me openly about their vandalism and sabotage never say “We need tough voter ID laws so that the lefties can’t vote”. More often, they’ll use fraud to justify why they’re doing their illegal actions, “to level the playing field” or whatever.

        When people confess to crimes but don’t confess to trying to use ID laws tactically, it makes me think that people mean what they say.

        • MisterA says:

          It’s not like it’s just that Trump quote, that’s just the most immediate one. There’s of course the famous bit where PA’s House Republican leader, Mike Turzai, rattling off a list of conservative accomplishments he could take credit for at a Republican State Committee meeting, got recorded saying this –

          And of course, there was that whole bit recently where the White House tried to change the census questions about citizenship – a move which on its face would have been totally legal and within their power, if not for the fact that one of the Republican strategists behind it died, his daughter found his files where he helped map out said strategy, made them all public, and so it was possible to provide documentary proof to the Supreme Court that all the administration’s stated reasons for the move were bald-faced lies and the sole actual reasoning was that suppressing the minority vote helps Republicans win elections.

          • Alsadius says:

            That seems exactly the same. “They will cheat unless we implement real security”.

            Regarding the census, the citizenship question seemed perfectly fair. TBH, I would have thought it was required since 1870 under the 14th amendment. The constitution clearly specifies how districts are allocated between states, but not within states.

            Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed. But when the right to vote at any election for the choice of electors for President and Vice-President of the United States, Representatives in Congress, the Executive and Judicial officers of a State, or the members of the Legislature thereof, is denied to any of the male inhabitants of such State, being twenty-one years of age, and citizens of the United States, or in any way abridged, except for participation in rebellion, or other crime, the basis of representation therein shall be reduced in the proportion which the number of such male citizens shall bear to the whole number of male citizens twenty-one years of age in such State.

            So given that, you can draw districts within states fairly freely, other than by Supreme Court rulings, and arguably some federal election laws (under the “Republican Form of Government” guarantee). There’s a long-standing requirement that districts be drawn with equal populations, from one of those sources(I don’t know details). However, I would think it a plausible argument that districts can be drawn to have equal numbers of voters, instead of equal numbers of adults or residents. In order to do that, though, citizenship needs to be measured in the census. I’d probably prefer that system, on principle, because the goal is to make all votes equal. Not all adults, because people who can’t vote should not have equal democratic power to people who can vote – that’s why they’re denied a vote in the first place.

            So yeah, the citizenship question was fine by me. It had obvious partisan implications, but it was IMO within the realm where disagreement is reasonable, not malicious.

            There is a line here, to be clear. It’s been a while since I looked into that strategist’s work, but I recall thinking some parts of it were over the line. But if your goal is to make sure that legal voters can vote, illegal voters cannot, and all votes are counted equally, I’ll let you wrangle over whether to hold your advance poll on Saturday or Sunday, even if I know it’ll have an effect on Jewish turnout.

          • HeelBearCub says:

            However, I would think it a plausible argument that districts can be drawn to have equal numbers of voters, instead of equal numbers of adults or residents.

            I believe SCOTUS disagreed with this, although I think it’s possible they decided on more narrow grounds. IANAL.

            They basically signaled that the citizenship question might have been fine. But the ruling against the question was dependent on the evidence of the intent to use it in this manner coming to light.

          • Alsadius says:

            Yeah, it may have failed. But my view is that everyone has the right to make plausible legal arguments on questions that have not been explicitly decided before. This feels very plausible to me, even if it ultimately failed.

          • MisterA says:

            That’s sort of my point, though – it’s not that there aren’t plausible arguments, it’s that the arguments keep being shown to be fig leaves deployed for plausible deniability to cover the actual motivations.

            Roberts basically put in his decision that this would have been legal, if someone hadn’t turned up ‘Evil Republican Plan to Screw Minorities.docx’ and submitted it into evidence, but since they did, everyone had to stop pretending the plausible fiction was true and look at the actual reason.

          • Controls Freak says:

            However, I would think it a plausible argument that districts can be drawn to have equal numbers of voters, instead of equal numbers of adults or residents.

            I believe SCOTUS disagreed with this, although I think it’s possible they decided on more narrow grounds. IANAL.

            It was more narrow. They viewed the petitioner’s request as trying to Constitutionally require states to use voter-eligible population instead of total population. SCOTUS said that they weren’t Constitutionally-required to use voter-eligible population; that is, they were allowed to use total population. They did not hold anything on whether states were required to use total population or were prohibited from using voter-eligible population.

            As it stands, “one man, one vote” is still an utter conceptual mess from the standpoint of Constitutional jurisprudence. Maybe we’ll come out of it with a modern day Three-Fifths Compromise.

        • Corey says:

          I gave my home State of NC the benefit of the doubt with the racial gerrymander a while back. I assumed they were just shooting for disenfranchising Democrats rather than blacks per se.
          Then the court case came out, where it turned out they were researching racial not partisan turnout, and generally targeting blacks as blacks.
          An example where this actually disadvantaged conservatives: The religious-black vote was crucial in passing NC’s anti-gay-marriage constitutional amendment. Nonetheless one of the changes was to eliminate Sunday-morning early voting, when black churches would organize “souls to the polls” drives.

  5. RC-cola-and-a-moon-pie says:

    Does anyone know if this radical “distancing” regime that is implemented to address the coronavirus is having the incidental effect of drastically reducing the rates of flu and other communicable diseases during the time it has been in force? It would be nice if at least a silver lining was a large reduction in non-coronavirus death and suffering from these sorts of contagions.

    • HeelBearCub says:

      It’s going to decrease pretty much most communicable disease.

      I’m sure there are some specific diseases that are exacerbated, due to their specific nature, but I would have to think that’s the small minority.

    • The Nybbler says:

      Yes, according to networked digital thermometer company Kinsa. Their map shows that even in Queens, NY (the county worst hit by COVID-19), total influenza-like illness (I presume really just fever, as that’s all they can meaure) is below normal.

      • RC-cola-and-a-moon-pie says:

        Thank you, I appreciate it. Very interesting.

      • Jake says:

        How much of that can be explained by a larger percentage of healthy people taking their temperatures just to see though? I’d be interested in seeing that as a total number of people with a high temperature, instead of a predicted percentage.

      • Douglas Knight says:

        It’s below average, but is it abnormal?
        There’s a lot of variation from year to year. If the flu came and went early this year, then it would be low compared to other years at this time, but that doesn’t mean much.

        The CDC Flu Dashboard has data from the last 20 years. Different years are on different schedules. This season peaked in week 4, before social distancing. That’s early, but it’s not extraordinary; the 2017-2018 season looks similar. (The upper right corner shows visits for influenza-like-illness (ILI), which includes covid, while the lower right corner shows absolute numbers of positive flu tests which doesn’t. They diverge in the past few weeks.)

    • nupi says:

      Apparently it also reduces ER admissions because of accident related trauma which by extension must do something on accident related deaths, too. Sorry, hearsay so far but I am sure we will get data on that.

    • Aminoacid says:

      The April update of the New England Journal of Medicine starts by discussing how the flu season has been skipped this year, since people are not bumping into each other

    • DaveK says:

      The policies in theory should also lower many causes of fatality, like traffic accidents.

      I read something about how reduced pollution probably will save a lot of lives.

      I would guess the tradeoff there might be some people are forced to more sedentary, and in the long run mental health effects, particularly for those who live alone without partners or families.

      • HeelBearCub says:

        Traffic accidents are quite likely lower. I think fatalities will be confounded in many cases by the increased speed of the accidents which do occur. I’ve seen anecdotal evidence that this has occurred in MN, although I didn’t check it myself.

  6. bean says:

    An aircraft carrier captain publicly complained that the Navy was failing to address an epidemic aboard his ship; the Navy fired him for whistleblowing. I’m having a hard time thinking of any perspective other than “the Navy is bad and should be torn down totally to the foundations, preferably using some sort of land-based weapon so they can’t fight back”, but here’s a different ex-captain trying his best to give a nuanced perspective.

    First, using a land-based weapon won’t really help these days. They have Tomahawk (series coming next month) and aircraft carriers.

    I think Modly had no choice but to relieve Crozier for the reasons McGrath lays out, although it could have been handled better. Writing something like that in a way that’s likely to lead to disclosure to the media (we don’t pick idiots to command our carriers) is not something they can let him get away with, unless they want to allow the Navy to be dictated to by every O-6 who can make himself look sympathetic to the media. (Not Crozier, but there are some really weird Captains in the Navy.) Suspending Crozier pending investigation would probably have been better from a PR point of view, but I wouldn’t have second-guessed Modly before his speech a couple days ago. (Note that McGrath’s thread was pre-speech.)

    The big problem for Modly was the speech he delivered to the Roosevelt’s sailors, which was the most spectacular example of political suicide I’ve ever seen. His direct attacks on Crozier in front of a very sympathetic audience were the sort of thing that make me very confident in his complete lack of judgement. There are lots of things he could have done to put out or at least dampen the fire, and he instead doused himself in gasoline and jumped in.

    The Navy is definitely not in a good place right now, and the SecNav position in particular is increasingly looking like a singularity of stupidity. But the proper response is to find a better SecNav, not to destroy the entire Navy. (Yes, I am available.) If that’s not an option, I suspect that a cabbage would be the proper choice to the string of people who failed to be better than that. (Mabus, Spencer and now Modly.)

    • tgb says:

      I’m sympathetic with Modly having to relieve Crozier for reasons that don’t his innocence into account and I’m also annoyed seeing the misrepresentations of his speech in media (he did not, in fact call Crozier stupid, and the fact that everyone seems to think he did makes me sad because it means you cannot use even slightly nuanced sentence structures without being misread. Is an “If A then B” statement too complicated for everyone? That’s not the same as saying that B is true, particularly when you’re next statement is that A is not true. Modus ponens versus Modus tollens, etc.).

      But the part where he blames China belies Modly’s bad intentions and I haven’t seen it called out. The whole “China lied about how bad this was so we can’t be blamed for failing to respond” argument is political hogwash. It was abundantly clear that the threat was real and that a serious response was needed due to the numbers we did get from China, if they lied it was no excuse to not respond. That’s like blaming the weatherman for you not taking an umbrella since they only said there’s an 80% chance of rain.

      • HeelBearCub says:

        The argument about Modly’s statement about Crozier applies just as well to Modly himself.

        But then the conclusion would be that Modly is an idiot, unless you somehow how believe his intention was for the audio of this speech to leak. So, I don’t think “he was just making a nuanced argument” really flies.

        • tgb says:

          Modly was stupid to make that speech but the part being quoted was just a “nuanced” (if it even gets that far) argument that everyone else (intentionally?) misconstrued. It’s the second half of the statement that frustrates me more because I expect Trump-appointees to torpedo their careers in an idiotic explosion but I want to hold the rest of us to a higher standard than that. And it’s not a high standard, this is like 8th grade reading comprehension level.

          • HeelBearCub says:

            It doesn’t matter that the part of the speech quoted was nuanced. It was nuanced in precisely the way the statement itself states is deliberate or idiotic. It is self defenestrating.

            “For thee but not for me” is always a bad look.

          • tgb says:

            Well, he resigned afterwards, so you can hardly call him hypocritical of removing Crozier.

        • DaveK says:

          yes, it was an extremely ironic case.

          Roughly paraphrasing- “If Captain Crozier was naive or stupid enough to believe his letter wouldn’t get leaked to the media in the information age we live in..” As Moldy clarified, he didn’t think Crozier was stupid, meaning he believed it was Crozier’s intention that the letter be leaked, which Moldly was trying to cite as a reason his dismissal was neccesary beyond just “he violated the chain of command.”

          The irony of course being this means Moldly himself is too naive or stupid to be the Navy Secretary as it didn’t occur to him that the audio of his own speech would get leaked.

          I think the whole “the information put the crew in danger because China realized the ship had limited capacity” doesn’t pass the smell test. It stretches beyond belief to think that China would invade Guam or something just because the carrier dedicated to defending it had limited operational capacity.

          You could make an argument that what he actually meant was too specific to be clarified. That right now China is taking advantage of the relative weakness of the US. For example, some weeks ago the Navy tried to send some ships through an area of the Ocean China had claimed and built artifical islands to protect (in order to establish that China’s claim was invalid and that the US would not respect it), and China used sound and electronic weapons to repel the Navy and they had to turn around, meaning the mission failed. So the actual reason could be something like “letting China know this ship and other might be having issues that would effect their operational capacity might embolden them to assert their power more then they are already doing”, which might not be something you would want to make publicly explicit. The “too naive or stupid” argument still applies to Moldly unless his intention was for the audio to get leaked. Even then, failing to realize it would be misrepresented still leads to the “too naive or stupid” conclusion.

          The other “actual” motivation for the sacking that I could think of has to do with precedent.

          If Crozier demanded help for his crew, other commanders of military forces might be emboldened to do the same, which WOULD have a pretty big impact and potentially put troops in danger (image troops in places like Iraq or Afghanistan where Islamic militants might be emboldened to strike.) Again, this rationale isn’t something you could state explictly, as it would be a self fulfilling prophecy.

          To be clear, I’m not arguing in favor of the sacking or of Moldly’s speech, rather thinking about what the Navy and Moldly might have been thinking.

      • bean says:

        One of the things someone at Modly’s level should know is the potential for bad soundbites, particularly when delivering a speech to a hostile audience on a matter that you’re under substantial public fire on. He chose to say what he said anyway. For that matter, “Either Crozier was an idiot for thinking it wouldn’t leak, or he committed a serious UCMJ violation” is really only marginally better than “Crozier was an idiot”.

        The rest of the speech was just confused, like it was half written by a good speechwriter and half either off the cuff or by someone trying to destroy Modly. Things like “Imagine if every other CO also believed that the media was also the proper channel to hear grievances with their chain of command under difficult circumstances. We would no longer have a Navy. And not longer after that, we’d no longer have a country.” pretty much needed to be said. But that should be early on, and coupled to a “More in sorrow than in anger” tone about Crozier, not bringing up stuff about how (for instance) he dropped the Governor of Guam in it. Likewise, “it’s good that you love him, but you’re not required to love your leadership” is incoherent and doesn’t actually make any point Modly needed to make. The thesis of the speech basically should have been “The proper order of priorities for a military force is mission, men, self. Crozier lost sight of that, and got 1 and 2 reversed.” But written by a better speechwriter than me.

        • tgb says:

          I agree that Modly was dumb to make the speech and should have known that it would end badly. However, I don’t think that excuses people/media from not having the very basic skill of understanding such a simple argument (or of intentionally taking it out of context to deprive others of that). I really want to live in a world where politicians can make statements with conditionals in them without being misunderstood.

          • bean says:

            I can sort of see where you’re coming from, but I don’t think most of the blame lies with the media. There are ways to make a nuanced argument that people will understand as nuanced and there are ways to make it so that they ignore the nuance. Guess which one he did? Specifically, if you say something as inflammatory as the “stupid or naive” bit, people are going to ignore most of what’s around it because they’re too busy watching the fire. This is just kind of how most people work, and good practice if you’re making a nuanced point is to make sure you don’t set half your sentence on fire.

          • @tjb:

            My reaction as well, as I think I said in an earlier comment. Any writer who, having read the speech, claimed that Modly said Croizier was naive or stupid was lying.

    • J Mann says:

      Does the Navy have competent Communications/PR department? It seems like this is the kind of situation that they should be handling.

      • bean says:

        The Navy’s never been great at PR, but in this case I think it’s less that and more Modly deciding to take matters into his own hands. The speech has some really good parts and some really bad parts, and the bad parts look like something he came up with on his own. There’s not much a PR department can do if the boss decides to say something stupid.

        Edit: That said, Kenneth Braithwaite, whose name has been submitted to the Senate for the SecNav job, spent most of his naval career in public affairs. If he’s willing to stand up for the Navy (he’s close to Mark Esper, the SecDef, so I’m a bit skeptical on this) he could be just what the Navy needs to turn its image around.

    • eremetic says:

      I think Crozier was probably deliberately committing career suicide in order to change the situation, no?

      • bean says:

        I think that’s exactly what he was doing, and I salute him for his moral courage in doing so. But I also don’t really want to let him off the hook, because it normalizes going to the media when you have problems in your chain of command. That is a very bad thing, and making it less taboo means that we’ll have people doing so when they don’t have equally good reasons for it.

        • Eric Rall says:

          Relieving him of command strikes me as a pretty well-calibrated punishment. It’s severe enough to send a clear “seriously, guys, don’t do stuff like this” signal, but he isn’t getting court-martialed, demoted, or even (as far as I know) formally reprimanded, so the door should be open for the Navy leadership to rehabilitate him a year or two down the road.

          • Matt M says:

            Being relieved of command is a big deal. It means his career is effectively over. He will be shuffled to some boring desk job, there is zero chance he’ll ever be promoted again, and he will be strongly pressured to just go ahead and retire.

            Now yeah, that’s hardly the worst thing that can happen. They aren’t throwing the book at him in any sense. But they are basically saying “You need to go away” on the strongest terms they possibly can without having to launch some sort of formal disciplinary procedure where the lawyers all get involved.

        • Simon_Jester says:

          You don’t want military officers going to the media with problems in the chain of command when the chain of command is competent.

          But sometimes the problem with the chain of command is “the people appointed at the top are nincompoops.”

          Suppose your military is in the process of being disabled by massive coronavirus outbreaks aboard its most powerful warships. Suppose the president isn’t doing anything about it because he’s afraid his balls will fall off if he admits there’s a problem. Suppose the SecNav he’s appointed is supporting this approach because he’s a sycophant chosen largely on the basis of how good he is at flattering the president.

          At some point, this becomes the kind of problem the public SHOULD be aware of, because “the politicians we elected and the commanders they’ve appointed to run the military are making decisions so profoundly bad that it’s compromising national security” is a pretty important piece of information the public should be aware of when the next election cycle comes around.

          • Clutzy says:

            Just FYI, most grunt level people I know have said we have an Admiral problem for a decade+ The rot at the top of the armed services is widespread and well known.

          • A much larger example of the same issue was the exposure of mass surveillance by Snowden.

      • DaveK says:

        Reading the wording of the letter, it seems clear he was inspired by his ship’s namesake, Teddy Roosevelt, who wrote the “Round Robin” letter under comparable circumstances.

        When Crozier was waying his options, that bit of history likely influenced him. As it turned out for Roosevelt, the letter was not career suicide, but different cirucmstances.

        Another note- the military reactions I have seen have been split, although often along partisan lines and it’s difficult to tell because of the issue of representative sampling. For what it’s worth, most of the publications aimed towards military members struck a sympathetic tone, although that may have been due to PR calculations as opposed to actual sentiment.

    • sharper13 says:

      Crozier knew what he was doing. He copied 20-30 extra people on his letter. He was also infected, so was likely going to be relieved soon no matter what.

      The fact that Crozier had been in frequent contact with Modly’s chief of staff, even on the day he sent the letter, and had been given Modly’s personal cell phone number to call if he wanted to raise additional concerns argues that the Navy wasn’t simply ignoring him, which would’ve been the only possible justification in my mind for going public.

      In reality, it appears that all he accomplished was to make any potential enemies/attackers aware his carrier group was crippled, while making himself and the Navy look bad in general. Hardly a recommendation for continued command.

      • DaveK says:

        “in reality, it appears that all he accomplished was to make any potential enemies/attackers aware his carrier group was crippled, ”

        That rationale strikes me as implausiable. As I said in another comment, there very well may have been other more logical motivations for his firing which wouldn’t have worked to air publicly, but the idea that China was going to decide to attack his ship specifically just because of this news stretches credibility.

        • LesHapablap says:

          I don’t think it stretches plausibility to say that if China has evidence that US forces are weakened they might get more aggressive in the south china sea. Obviously if every US ship that ends up with coronavirus is taken out of action then soon the whole navy will be out of action temporarily.

          • Simon_Jester says:

            Given that the Chinese themselves, even if they’ve got coronavirus cases down to a minimum right now, realistically have to be worried about a resurgence…

            This would be an extremely bad time for them to start indulging in acts of war. Nobody wants to have to fight a war and deal with a pandemic at the same time.

            A falling meteor shower sinks all boats.

        • sharper13 says:

          There’s a reason unit readiness information is considered a secret. Surely you can appreciate why that would be in general….

          … and thus to apply that general order to this specific case is pretty straightforward.

          Sure, do we know of a specific threat from a specific enemy against his specific carrier group (not just his ship)? No.

          But then, we wouldn’t, would we? We also wouldn’t know about plans enemies might have for say, taking over more of the spratley’s, etc… which would be encouraged by knowing the U.S. doesn’t actually have a carrier group which is in a position to respond or help our allies, etc…

          It doesn’t have to be about an attack on China directly on his specific carrier.

      • John Schilling says:

        Guys, the Chinese have satellites. And spies and SIGINT and all the rest. They already know which US carriers are tied up at the dock, and which ones are offloading their crews into quarantine facilities ashore. If you want to say that Crozier deliberately or negligently mishandled classified information, then that’s technically true in the way that the whole “But her Emails!” thing is technically true. And in both cases it should probably result in that person being disqualified from command as a matter of principle. So, mission accomplished.

        But if the claim is that Crozier actually endangered national security or emboldened our enemies, then no, not that.

        • LesHapablap says:

          You clearly know a lot more about this than me, but didn’t Crozier publicly pressure the Navy into changing their COVID strategy to one that takes ships off the line? Successfully?

  7. Yovel says:

    Re: Iran:
    The opposition organization Mojahedin Khalgh have estimated 23,100 dead in Iran [1], about six times the official number. I’m not sure they are absolutely reliable, but they were the source for some important leakages, including exposing the Iranian nuclear weapon plan, so they have at least some credibility. Plus they have boots on the ground, which nobody else does.
    The WHO estimated the Iranian dead as 5 times the reported number [2], and it has been leaked that the Israeli inteligence estimates the ratio as 4 [3].
    So while the exact number is not clear, it is clearly extremely underestimated.


  8. Purplehermann says:

    On variolation and mainstream talking about dose:

    The main issue with this method is the PR essentially.
    Most people hear “we can infect people and have lower death rates” and they react as if it was “let’s kill people, then we’ll have less dead. It’s just numbers, and this way we get better numbers”.

    So the mainstream putting the importance of dose, and how high doses are killing healthcare personnel into the water supply will make people a lot less freaked out about variolation (coronation?) for doctors and nurses, and after a little while other people as well

    This also means Robin Hanson should design his experiment to look as safe and benign as possible while still having some useful info

    • zby says:

      There is an important improvement/alternative to the variolation proposal – the virust must be mutating all the time and some of the strains must be mild – let’s find one that is mild and still give immunity and then vaccinate people with it: (via HackerNews:

      • Purplehermann says:

        I don’t think this is super important honestly.

        I’m guessing young healthy people have something like a 0.01% death rate in the wild, and hopefully the lower doses in safer infection sites would take us the rest of the way to ~0%.
        Either way if the first trial is robe right no one will die.

        Right now time is more important than getting a slightly milder strain.
        If it can be found fast enough to be incorporated in the first trial, great.
        If not then maybe later on.

        [Edit: it should definitely be checked too, but one test shouldn’t wait on the other]

        • fallenscien says:

          This is hugely important. The point of the milder strain isn’t as much about the lower chance of harming the person we’re putting it in. It’s about transmission from that person to other people.

          We do not have any reasonable way to guarantee that someone has completely cleared a virus. There is always the possibility – and it’s a pretty strong possibility – that they will start shedding the virus again later.

          If the virus they’re shedding is an attenuated strain, great! It probably won’t infect anyone because of low copy number, but it it does, they’ll fight it off no problem.

          If the virus they’re shedding is full strength, and will kill any elderly or immunocompromised people who catch it from them, we’ve just made them a potential spark in a powderkeg.

          • Purplehermann says:

            It is pretty unimportant to trials for effect/correlation of/to dose on severity.

            If variolation ends up being used because it is useful in getting more minor cases, it will be used on either those most at risk or enough healthy people to get a chunk of herd immunity.

            When days are important you don’t wait for testing one mechanism, just in case someone finds a different piece, or in case it turns out stopping the virus spreading is impossible. You start doing what you can, if someone finds a milder strain that confers immunity as well, great. It turns out people are infectious at random times years later, variolate everybody if variolation works. If not, great.

      • tgb says:

        I saw this and isn’t it saying “Vaccines are too slow to develop, so let’s do this other thing instead” but the reason vaccines are slow to develop is testing them (depending on type, some are already being tested) so the slowness is inherent to anything that needs to be tested. And the new proposal is at least as risky as vaccination and needs to be tested at least as much. So the article totally failed to justify it’s position, as far as I could tell.

        • Purplehermann says:

          IAMNAD, everything in this comment is my impression.

          Vaccines are not the virus (They are based on it but important changes have been made). This means that
          1. They don’t always provide immunity
          2. They sometimes make things worse than the virus itself.
          3. The vaccine needs to be produced

          With the original virus we don’t really have the first two issues, and the third is less relevant.

          If the dosing and placement are done correctly then the upper bounds of damage done is just COVID-19 infection.

          The lower bound is immunity that getting the virus gives. (If the dose is too small to provoke an immune reaction, that’s just more testing)

          If the group selected is young and healthy this really isn’t an issue.

          The death rates for 20-39 year olds Sarah C gives are 0.1%-0.2%. I think 0.1% is about right, but let’s take 0.2%.

          In NY(C?) it looks like only a twentieth of the young dead had no underlying health conditions.
          That gives us a 0.01% death rate.
          Keeping in mind that high doses seem to be particularly lethal and the virus starting off in the lungs in some cases, it looks like young healthy people are very unlikely to die from the variolation tests.

          Tl;dr this isn’t actually very risky, and is a pretty sure bet that it works to some degree, and probably will need much less testing.
          It’s not as good as a working vaccine, but trying it is probably a good idea

          • fallenscien says:

            Vaccines can just be the virus. That’s what an attenuated vaccine is – a mild version of a virus, whether produced intentionally or found in the wild. See the Sabin Oral Vaccine for an example.

            Using live virus – especially when it’s a novel virus that’s poorly-characterized – can go extremely wrong very easily.

            There are a huge number of problems with this proposal. But I should be making that argument at the people who are trying to do it.

            I’m just wondering where all the real epidemiologists are at.

          • Purplehermann says:

            @fallenscien an important difference between manufactured attenuated strains and low doses is ADE. Novel issues may also result, Erich is also an important difference. Attenuated live viruses are not just the virus.

            I feel that you gave a basic explanation for a concept that, while related, wasn’t actually that pertinent (or accurate) and was also generally understood.

            Then you asserted that things could go very badly wrong very easily, while dismissing this discussion as an important place to make your argument.

            This was probably just a low effort post, not dunning-kruger/jerkitude/trolling, but it’s coming off a bit like it is to me.

            Could you explain what exactly could go horribly wrong when using the normal virus?

            Screwing up the dosing is the main thing I’d be worried about, followed by infecting a vital area (like the lungs).

          • albatross11 says:

            The main thing that’s likely to go wrong is that you end up making your test subjects very ill and then you have a new concentration of infection to deal with. You might even screw up and give them a worse version of the infection than they’d have had normally, though I think that’s pretty unlikely if you’re careful and don’t do something like try to infect them by having them take some deep breaths of very small aerosolized droplets full of virus or something.

          • Purplehermann says:

            500 people being infected isn’t a huge deal if you know exactly who they are.

            You realize most of the group would probably be asymptomatic right?
            That’s even if super low doses aren’t better than low doses

          • fallenscien says:


            Yeah, it was a low-effort post.

            I’m frickin’ tired. I’m spending a lot of extra effort trying to keep people close to me from panicking, and I’m doing my best to be the primary interface between my elderly relatives and anything they need from the world, in order to minimize their risk.

            I was, and kind of still am, an infectious disease researcher. Not virology, but adjacent to it. Literally. In the wake of the 2009 swine flu pandemic, I worked a couple doors down from researchers who were working on the influenza virus, and we’d talk about the possibility of another pandemic, the challenges of vaccine design, and the possibility of creating a universal influenza vaccine (which was one of the things their lab was trying to fill in the basic research needed to produce).

            I (mostly) left academia because the pay and hours are terrible, funding has been continually shrinking and that trend showed no signs of reversing, and I have some issues with fundamental ways that incentives are structured which cause our most talented researchers to spend all their time applying for grants instead of doing research. But that’s an entirely different conversation.

            In my experience, if you argue with a rationalist and hit them with a slow trickle of information against a centrally-held belief, the prior that they’re going to update is the one that says they should listen to you and seriously consider your arguments. Coming back over and over with minor quibbles to try to undermine a single belief is what biased sources of information do.

            Instead, if you want them to reconsider a firmly-held belief, you need to hit them all at once with a figurative sledgehammer of data, meticulously constructed and rigorously cited.

            In this case, that’s a whole lot of work. Because getting a person to the right place in this specific argument requires what is more-or-less a short course in viruses and the human immune system, it’s more on par with writing a chapter in a textbook than a single research paper.

            I suspect you’re not going to be sympathetic to this, but the comments section of a blog post is absolutely the wrong place to post a book chapter worth of material.

            The thing is, I’m a low-level expert in this subject as far as credentials go. I’ve never cared about those. I’m exhausted and don’t understand how it could possibly end up being my responsibility to derail the activity that Hanson and you seem to be somehow seriously trying to move ahead on.

            I am not assuming I would succeed. I’m just saying that it would be pointless to half-ass an attempt. I don’t even know if anyone would listen to me if I tried.

            At the end of the day, I’ve got a voice in my head that sounds like Dante from Clerks: “I wasn’t even supposed to BE here today.”

            I don’t /want/ to have this argument, but for some reason, it seems like no one better-qualified in epidemiology is around. So if I have to try, I’d want to actually try as completely as I can, and aim as directly and efficiently as I can at the people doing it.

            Are you a direct path to the source of this effort?

          • Purplehermann says:

            I get where you’re coming from now.
            (For the record, low effort posts just telling people they’re wrong don’t seem helpful.)

            I am not a direct path.

            If you do decide to try convincing Robert Hanson I’d like to see it.

            Imo you could have done a much better low effort post…

            A hypothetical way things could go wrong that is outside what’s already taken into account, a link to some information on the risks of experimental vaccines vs variolation explaining relevant issues with the method beyond what is already being discussed…

            This isn’t a centrally held belief in any way.

            This is a “well it looks right, and most of the people protesting it are pretty much saying vaccines are better in every way, and ignoring most of the claims made, maybe a few million lives can be saved and there is very little risk. Any objections (besides: it’s stupid, for reasons)?”

            The assumption here is (should be) that the ones with central beliefs here are those saying we shouldn’t infect people, and the experts are already doing everything that makes sense.

            So while actual info should be looked into and appreciated, low effort “you’re wrong/evil/stupid/arrogant claims would be ignored, and the people making them paid less attention to on this subject.

            [Added: as far your suspicions, that was just obnoxious. An actual text book chapter is not expected by almost literally anyone.
            Also, I really am not sympathetic to a claim of “it’s so much work, and a very simplified reason it wouldn’t work is impossible, so I resorted to a fly-by ‘you’re wrong’ and explaining a concept already discussed in an unhelpful way, because this isn’t my problem.”]

      • truckdriver20 says:

        This is just fast-tracking a vaccine. If they do end up doing a “herd immunity” approach (which should be last resort) they should be variolating people as Hanson describes while also essentially enrolling them in massive vaccine trials. My understanding is that the reason a vaccine will take so long is because they typically need to prove that it’s safe enough to not accidentally cause terrible effects in 1/10000 people. But if we’re resigned to the fact that everyone is going to end up getting the virus without it, surely a much lower threshold for safety is acceptable (for healthy young volunteers giving informed consent, of course).

        • albatross11 says:

          Right. If you want to intentionally expose people to the virus in a world where we already have some experimental vaccines ready to test out, then the right way to do it is to give a bunch of young, healthy people the vaccine, wait 3 weeks, check their antibody levels, and if all looks right, challenge them with the virus and see if they {show any symptoms, have detectable virus levels on any day} over the next 14 days or so.

          Then, you not only get a pool of immune people (for however long immunity lasts–something else we don’t really know), you also now know whether your experimental vaccines provide protection against getting infected. That probably massively speeds up getting a vaccine out into use. At least the mRNA vaccine is apparently pretty easy to make in large quantities, too.

          • Lambert says:

            And you can get a bunch of different candidate vaccines in the hope that at least one is sucessful.

        • tgb says:

          But in what way is it fast-tracked? I don’t see any reason to expect it to be faster than vaccine development (given that some are already in the testing phase).

          • albatross11 says:

            You get information about whether the vaccine prevents infection much more quickly when you intentionally expose your test subjects to the infection. Note that this includes exposing the control group, who probably got a saline injection, and so who will probably all get sick. Without that, you don’t find out for sure whether the vaccine does better than just being young and healthy enough to get into the vaccine trial.

            The alternative is to wait a year or so and see what fraction of the treatment and control groups got symptomatic infections.

    • John Schilling says:

      The main issue with this method [variolation] is the PR essentially.

      The main issue with variolation is the stupidity. Variolation is literally the stupidest way to try and develop a vaccine. Variolation is “let’s grab what we think is mildly-infected material from a natural source whose true potency is unknown, and see if there’s a dose that produces immunity without serious infection”. Every other technique for developing vaccines involves at least trying to apply cleverness to modify the virus for a higher immune-response-to-actual-infection ratio, culturing or manufacturing the result so that you know what you are getting, and then seeing if there’s a dose that produces immunity without serious infection.

      Every other technique for producing vaccines is better than variolation. And we’ve got close to a hundred vaccine candidates coming out of those better techniques, four of them already in human testing.

      The great advantage of variolation is that, being stupid, it can be applied by stupid people. Or, more precisely, smart but ignorant people who don’t know how to do any better, like eighteenth-century doctors. Or maybe mid-twenty-first century doctors after the apocalypse puts us all back in Mad Max land and we’ve still got plagues that need dealing with. But if you’ve got actual professional virologists and epidemiologists and proper research facilities, you don’t need and don’t want to play with variolation.

      And if you play with variolation anyhow, because you imagine yourself to be the great biohacker who is going to vanquish the plague where the professionals have failed but you don’t have the knowledge or resources to do anything better, so what? You aren’t going to be allowed to roll it out for general use without going through the same testing as the professionals, and the professionals now have a head start on that. If those testing requirements are relaxed for anyone, they’ll be relaxed for the professionals first and they’ll still beat you to the finish.

      • nkurz says:

        @John Schilling:

        I think you are being too harsh. Or perhaps, your attack on classic variolation might be completely justified, but classic variolation is not exactly what Tillett is proposing.

        > The great advantage of variolation is that, being stupid, it can be applied by stupid people.

        Tillett might be wrong, but he’s not stupid. He also has a new article up describing what his approach might look like in practice:

        As I see it, the advantage of his approach is that by starting with a wild but mild strain, you have the advantage of being able to do observational outcome testing that might be impossible (or at least unethical) to obtain for a novel vaccine: Large X people were infected with this strain, small x people died. If the numbers are good compared to the base case of natural infection, helping this strain to spread might be a societal win.

        And unlike in the past, with modern sequencing it’s possible to distinguish strains beyond just looking at outcomes. Rather than just simply for strains associated with mild symptoms, Tillett also puts a lot of emphasis on checking the sequences for genetic deletions that 1) are likely to cause a change in lethality, 2) are unlikely to affect immunity, and 3) are unlikely to mutate back to a lethal form.

        > and see if there’s a dose that produces immunity without serious infection

        I haven’t seen Tillett place any emphasis on dosage. To the contrary, he deemphasizes it on the theory that what matters most is not the effect on the person that you intentionally inoculate, but on the people that are infected downstream from that person. Since you have no (or very little) control over that downstream dosage, you don’t want to be dependent on dosage. It’s probably worth re-emphasizing that Tillett’s approach is trying to make use of the natural spread of the mild strain, with the goal of outcompeting more dangerous strains through person-to-person transmission.

        > Every other technique for producing vaccines is better than variolation.

        I think this is a stronger argument. The counter-argument is that if a mild natural strain exists, and if a large number of test subjects have already been exposed to it with positive results, why wouldn’t we make use of it? We shouldn’t stop working on modern vaccines, but if it’s relatively low effort, we probably should (in parallel, using non-competitive resources) search to see whether such a strain already exists. Unless there’s a solid reason to believe that no such strain can possibly exist?

        > because you imagine yourself to be the great biohacker who is going to vanquish the plague where the professionals have failed

        I’d rather evaluate the ideas on their own merit, but Tillett looks like a fairly well-credentialed professional in the field: You might be right that his suggestions lead nowhere, but I think it’s worth looking closely at the specifics of what he suggests, rather than quickly rejecting it based on historically similar precedents.

        • Purplehermann says:

          Hanson is talking about dose specifically.

          I see little reason not to try both.

          People with the virus are hopefully being isolated, you should be able to constrain their movements. It’s only people who aren’t confirmed to have it who are an issue

      • Purplehermann says:

        1. When you say “you know what you are getting”, what do you mean? Supposedly the virus has been pretty stable, with variolation you should know what you are getting, more than a manufactured or altered vaccine.

        2. Assuming the dose can be measured very accurately and the doses given are lower than normal wild infections, what do think the risk is?

        Assume healthy young people who probably would have gotten it anyway.
        Please be specific, with numbers if possible.

        3. Do you think variolation is as bad as or worse than just letting natural herd immunity occur while isolating high risk groups? (There seem to be plenty of advocates for this, there’s a chance that this will be policy). Why?

        4. Do you think variolation is less likely to give immunity than vaccines in the next year? Do you think vaccines are a sure bet i the near future?

        5. Stupidity of solutions is indeed an advantage. They are less likely to go wrong and more limely to work, even if they often give subpar results 😉
        Here subpar results may be enough considering how relatively little the virus does to young healthy people

        • John Schilling says:

          Stupidity of solutions is indeed an advantage. They are less likely to go wrong and more likely to work, even if they often give subpar results

          I’m pretty sure this is false. The stupidity of variolation is that you are more likely to accidentally give people full-fledged COVID-19, of the sort that kills people. Smallpox variolation killed 1-2% of the people it was tried on, far more than proper vaccination. And it won’t be just the people who volunteer for your experiment, but also the health care workers who wind up treating the ones who wind up in an ICU, and the family members of those health care workers, etc. And those aren’t all going to be healthy twenty-somethings. But I’d love to see you defend your claim that injecting people with a natural strain of COVID-19 is less likely to go lethally wrong then injecting them with a deliberately-attenuated virus or a dead virus or a protein stripped from the virus.

          As for the rest, no, I’m not claiming that it is better than trying for herd immunity. Using the stupidest possible vaccine is somewhat less stupid than doing nothing at all. Whee.

          But, with a hundred or so smarter vaccines in the works, I think the odds are very good that one of them will work. And if none of them work, it will almost certainly be because there’s something about SARS CoV-2 that makes it highly resistant to vaccination in general, in which case variolation isn’t going to work either.

          • Purplehermann says:

            I’ll try to defend it alright.

            Let’s assume the variolation is no better than a regular, wild low dose. It’s main benefit is being a constant low dose that can be administered at a determined time in a body area of choice.

            Young healthy people barely ever die from this. Most of them are actually asymptomatic. Their systems usually handle it pretty damn well.
            There are some plausible mechanisms that cause the few deaths, which can easily be avoided in a lab.
            1. Unusually high doses.
            2. The infection starting off in the lungs (or other vital area with a bunch of ACE2 receptors I guess).
            3. Worn down health before infection (lack of sleep, being overworked for example)

            With 500 people the chances of death are really low. No one should die low.

            I also think bringing the nurses who would be infected into this odd. About 2-4% young adults need ICU, and this includes unhealthy people. Unhealthy people were dying at ~20X the rate as healthy for this age bracket, if that holds for ICU admissions then we should expect 0 or 1 people in need of an ICU.
            There is no reason a single person should infect the nurses either (or at least raise the chances of an infection all that much).
            Thinking a single ICU patient (who recovers) will make a serious difference when a pandemic is going on is silly imo.

            The comparative dangers of vaccines: Immune enhancement.
            Which is why they need more testing.

            As to your last point, if natural infections give immunity but vaccinations don’t, variolation could very plausibly end up in the natural infection camp – it is a natural infection, just controlled.

            That was all assuming variolation doesn’t give any benefits beyond the givens (that you decide on health and age, make sure they are rested when it comes, able to isolate, decide where it goes and the dose).

            If it makes the disease much milder too, that would be wonderful, and it seems likely.

            Finally, we should really be trying most anything that isn’t very risky if it lessons the chances of a natural herd immunity policy imo, and I think you should agree.
            Countries are very uncomfortable waiting half a year to start their economies up again.

          • John Schilling says:

            OK, first, you seem to be assuming reliable control over dosage. The only way you can get this is by isolating the virus in the lab before you dose someone with it. If you’re doing that why aren’t you (as Lambert and albatross11 note) using one of the reasonably quick and reliable techniques for inactivating the virus at the same time?

            Second, yes, if you test it on 500 healthy young people in controlled conditions, it’s highly unlikely that you’ll kill anyone. But what good is the result? All you can establish with any confidence is that variolation has a fatality rate no greater than 0.2% in the lab. If you then use the technique on the general public, you may wind up killing a hundred thousand people. Or more, because even if you only give it to healthy young members of the public for some sort of augmented herd immunity, you won’t have nearly as much control and will be e.g. variolating people who only superficially appeared healthy under a much less thorough examination than your initial test subjects.

            And when some of your people wind up in an ICU, they’ll be treated by doctors and nurses who aren’t all carefully-selected young men in perfect health. Some of them will be older men and women, some of them will be parents who go home to their families with whatever you just gave them, and pretty much all of them will be overworked and not getting enough sleep.

            Nobody is going to allow you to do any of that, especially not when you deliberately skipped the “here we inactivate the virus” step. The only way variolation can be used outside the laboratory, is if it goes through the same testing that e.g. inactivated whole-virus vaccines go through. Which we are already doing. Since inactivated viruses are rather definitively safer than live viruses, your proposed technique is more dangerous than what we are already doing. And, since we’re already doing it but aren’t doing variolation, it’s going to be ready faster. What benefit does variolation offer to offset these disadvantages?

            Again, the alternatives aren’t “variolation or nothing!” or even “variolation now or more sophisticated techniques later”. It’s variolation against all of the less stupid ways to make a vaccine and make it sooner.

          • The Nybbler says:

            It’s variolation against all of the less stupid ways to make a vaccine and make it sooner.

            Nobody with the ability to use any of those less-stupid ways has shown any particular interest in making things sooner, at least not since some early press releases.

          • Lambert says:

            Who are you expecting to be doing the variolating? People who are not able to kill all the viruses in a sample?

            At least with smallpox you have the low-tech solution of snorting a crushed up scab. How do you expect to reliably give people a small dose of SARS-COV 2 without aseptic technique?

          • John Schilling says:

            “Showing a particular interest in making things move sooner”, doesn’t make things move sooner. For that, you have to convince the FDA to waive at least some of the rules against making things move sooner. I expect the reason most of the people engaged in vaccine development aren’t showing any interest in that, is the same reason I don’t generally show my interest in e.g. not paying taxes.

            If you are interested in making things happen sooner, and you think there’s any significant chance of that happening, then going to the FDA with any vaccine development plan other than variolation and asking “under these emergency circumstances, can we make this happen sooner”, is more likely to succeed than going to the FDA and saying “I want to try this old technique that was abandoned more than a hundred years ago due to the safety issues and I want you to waive some of your safety rules in the name of speed”.

            Ability to convince regulators to step aside so you can move fast and maybe break things, is specifically one of the ways variolation is stupider than every other way of developing a vaccine. Take whatever lobbying-fu you think was going to let you convince the FDA to let you do quick & dirty variolation, go to any of the other people doing less-stupid vaccine development programs, and offer to lobby the FDA on their behalf.

          • The Nybbler says:

            Variolation can be accomplished by people who don’t go to the FDA; that is, who do it entirely illegally. That is its sole advantage.

            Going to the FDA and asking them to speed things up is a waste of effort; if it were possible, the actual vaccine makers would have done it already.

          • John Schilling says:

            Variolation can be accomplished by people who don’t go to the FDA; that is, who do it entirely illegally.

            To what purpose? If you’re just looking for a way to protect yourself from COVID-19, masks hand sanitizer and social distancing are safer and more effective than any homebrew variolation. If you’re looking to meaningfully contribute to ending this plague, or to create a community in which you don’t have to socially distance yourself from everyone, or just to do Purplehermann’s 500-person trial so you can know that variolation works, then you are correct, you don’t have to go to the FDA for that. They’ll be coming to you.

            Anything that looks like you giving people the medical advice to get themselves infected with COVID-19, organizing a collective effort to that end and actually going out and doing it on any but the most trivially insignificant and irrelevant scale, is going to get you thrown in jail. If it’s just spitballing on the internet and hoping someone else actually organizes the effort, sure, you can do that under the aegis of the 1st amendment, but nobody else will step up to organize the effort because then they’ll be the ones going to jail. So what are you trying to accomplish here, and how do you expect it to happen?

          • The Nybbler says:

            I won’t be doing anything, since I’m terrible at organizing things. But there are people who are good at organizing who might be able to make a go at variolation without getting thrown in jail right away. Hell, I’m not even deliberately infecting oneself is illegal, except in the way that everything’s illegal nowadays — specify the act and a prosecutor can find a law.

          • Purplehermann says:

            @John Shilling

            First, if the dose can’t be controlled then the whole “low dose” idea doesn’t work, so I’m assuming it. Inactivated vaccines seem to have a higher chance of immune enhancement than an actual infection.

            Second, no one needs to die or go into ICU to check if the cases seem to be much lighter than usual. Cases aren’t binary – “dead 0 alive 1”.
            If it shows promise (not just “nobody died”…) in the first trial, a trial with a broader pool would be acceptable.

            I’m not sure why you are a priori assuming the spread from know infected people can’t be contained, the opposite seems intuitive to me.
            (You can variolate by household, make “Hero Hotels” like Hanson has proposed, etc..)

            As for doctors and nurses working with coronavirus patients, they are already working with coronavirus patients. They could benefit from variolation policies in particular – they’re already at risk.

            Please explain to me why you think immune enhacement risks are not a serious disadvantage to vaccines compared to natural infection, which are a big reason for longer testing schedules.
            Doing anything to change the way the virus works means a chance of something going wrong in a way wild infections alone wouldn’t.

            Right now “nothing” is exactly what is being done for doctors and nurses dealing with infected patients. (Obviously they are taking precautions, but no immunity)

            “Nothing” has been and is being discussed as a legitimate herd immunity strategey in multiple countries

          • albatross11 says:

            I think the best-known instance of immune enhancement is from dengue, and there, it happens as a result of the disease. You get one strain, and then I think you’re protected from future infections with *that* strain, but future infections with one of the other strains usually makes you sicker. I think the way this works is that the virus becomes a lot more serious if it can infect macrophages (cells in your adaptive immune system). Macrophages ingest things that have antibodies stuck to them. If the antibodies have stuck to the virus in a way that prevents it from effectively infecting a cell, then you get immunity, but the antibodies to the wrong strain don’t prevent the virus infecting cells, they just attract vulnerable macrophages so the virus can infect them.

            Is there some reason to think that enhanced infections are more likely as a result of vaccination than infection? I’ve never heard of a reason, but I’m certainly no expert.

          • Purplehermann says:

            Live attenuated vaccines rely on creating a different strain, and hoping it confers immunity to the original apparently.

            Dosing is using the same strain.

            Beyond that, if enhanced infections were as likely from prior wild infections, why would human testing be such a big deal?

            People are getting exposed multiple times all around the world in high numbers, if the experimental vaccine risk was no larger than the virus risk to an individual, why the strict testing – if it’s no worse than getting infected?

        • Edward Scizorhands says:

          Variolation has the advantage that you can try it on day 1. We are well past day 1.

          • John Schilling says:

            This also. It’s not much of an advantage if you’re going to do the full year-plus of testing that is the current standard, but if you’re in a tearing hurry and you are starting from scratch, variolation would be a bit faster.

          • Lambert says:

            You can try whole-virus vaccines on day 1+ however long it takes to boil some SARS-COV 2 (like 5 minutes if your kettle is slow?)

          • albatross11 says:


            Inactivating virus isn’t all *that* hard–people do it with heat and chemicals all the time. Heat-inactivated COVID-19 + adjuvants injected into your bicep probably has a pretty good chance of getting an immune response, and essentially zero chance of giving you the disease. OTOH, producing enough virus to vaccinate everyone may be a big engineering problem.

          • matkoniecz says:

            Are you sure that it is *that* easy and 12-18 months minimum for vaccine is just wasting time?

            I would expect that properly inactivating (proper immune reaction, low infection rate), selecting proper adjuvants, low rate of side effects and all other things are actually not so easy.

            And that mass scale production is not a sole issue.

          • Clutzy says:

            12-18 months is clearly an untenable timeline. Its up to the pro vaccine folks to figure out other plans that are not insane.

          • Garrett says:

            > OTOH, producing enough virus to vaccinate everyone may be a big engineering problem.

            I’m not sure that matters in the short-term. The biggest immediate hurdle from the inside right now is the lack of PPE for critical healthcare workers, trying to keep their whole staff from being out sick at the same time or being spreaders to other patients. If we’re able to get a good number of relevant healthcare workers immunized such that they no longer require the extensive PPE and can then fall back to standard precautions/BSI (of which I think we have adequate supplies) we’re then able to care for more people in the hospitals more efficiently and can focus more on the next hurdle.

            Combined with a minimally-risky test which can determine immunity (eg. serum antibody levels) would allow for significant improvements in our ability to manage this, even if at the hundreds-or-thousands of doses a day rate.

          • Edward Scizorhands says:

            12-18 months is the time to watch for side-effects. Is it any different for variolation or vaccine (or using a weakened strain, like elsewhere on this page)?

            The time for variolation was January or February. Our medical system was not under strain and we could have watched the initial volunteers extremely closely.

          • The Nybbler says:

            12-18 months is clearly an untenable timeline. Its up to the pro vaccine folks to figure out other plans that are not insane.

            Not as long as they have power. It it apparent to me that the public health establishment thinks that lockdowns until a vaccine is developed (and without cutting any corners, cowboys) are perfectly acceptable. And for now the politicians are going along, because ruling arbitrarily and capriciously by decree is the highlight of their careers.

          • albatross11 says:

            Getting a proper vaccine takes that long because you need to verify that it doesn’t make people sick and that it gives good protection. You can speed that up a lot with a challenge study (intentionally give the subjects the virus–basically what you’re trying to do with variolation, but you do it after you’ve tried the vaccine on them and given them time to get an immune response).

          • albatross11 says:


            I think elected officials are in over their heads, trying to figure out what to do in an untenable situation. The lockdowns are being advised by a bunch of respected experts and the governors and mayors are basically all guys with a law degree who don’t remember what a mean is, so they’re not reading epidemiology papers on pandemic modeling and trying to reason about them.

            The universe has served us all up a nice, steaming shit sandwich, and the fact that some of the stink has landed on you isn’t actually because of the malevolence of the authorities. To some extent, it’s due to the incompetence of the authorities and the brokenness of our institutions, but most Western countries’ institutions and authorities aren’t doing so well right now, either–we seem to be about the middle of the pack overall. Would this have gone better if all the powerful people were 20% smarter and 20% better-informed about science? Probably, but everything else would go better in that world, too. We elect a bunch of third-rate used car salesmen and reality TV stars to positions of power here, so the smart people can go into science or technology or finance.

  9. tgb says:

    Can anyone explain the discrepancy between FiveThirtyEight’s covering of how well the experts did and the tweet Scott links to? 538 makes it look like the expert’s did great, well within the consensus region, and if anything were underconfident. Yet my reading of their initial chart also seems to suggest that all but three of the 18 experts had it outside their confidence bands, with most being completely mistaken.

    Also, how do you create a consensus confidence interval from expert opinions? My first thought is using the same methods for meta analysis of papers like Fisher’s method, but it’s clear that doing that here would have generated a ridiculously narrow band and in general, surveying 1000 experts shouldn’t make you much more certain of your consensus estimate than surveying 50. Meta-analysis methods would really be estimating the average expert opinion, and the confidence interval around that is not what we should have for the confidence interval of actual value – but I don’t know what we should use instead.

  10. meh says:

    i am going to assume that having a segment of the population mark their clothing to be distinguishable is pretty taboo in germany

    • noyann says:

      My first thought, too.

      • Purplehermann says:

        Mine was maybe a yellow star would work

        • Leafhopper says:

          “Okay, everyone, get your stars out again — this time, they confer privileged status.”

          • Edward Scizorhands says:

            When the Star-Belly children went out to play ball,
            Could a Plain- Belly get in the game…? Not at all.
            You only could play if your bellies had stars
            And the Plain-Belly children had none upon thars.

        • bzium says:

          The biohazard symbol is kind of star-shaped.

    • DaveK says:

      Also easy to fake.

    • soreff says:

      Would badges work as a visible marker of immunity?
      It is kind-of special clothing, but has a long history as marking a special status, e.g. police.
      (How widespread is using badges? USA/UK – but how much broader?)

  11. TheTurtleMoves says:

    On the election: the two most likely scenarios in my mind are…
    1) The President calls for a delay, which only Republican governors abide by, yielding Bush v. Gore X 1000
    2) Hardly anyone votes because vote-by-mail is barely implemented anywhere, yielding Bush v. Gore X 1000

    • meh says:

      how will the recent supreme court decision effect any calls for a delay?

      • smilerz says:

        That depends on how the delay came about.

        The SCOTUS decision was really narrow and amounted to “judges can’t change the rules, only legislatures can”

      • Clutzy says:

        It simply means there is no US Constitutional provision or statute that guarantees a voter the right to a delay of an election. The date of the election is set by Congress already, so that can only be changed with a statute that passes both houses and gets signed in time. Governors have no power there.

        This is going to have to be a state-by-state discussion as to what provisions they want to have to make voting less infectious.

        • anonymousskimmer says:

          It simply means there is no US Constitutional provision or statute that guarantees a voter the right to a delay of an election.

          I believe it’s broader. It explicitly doesn’t allow for the votes (except at in-person polling sites) of people who followed the law as written and asked for absentee ballots before the cut-off. It can be presumed that some of these late, but in-time, absentee requesters actually cannot vote in person.

          So this ruling states that the incompetence (for whatever reason) of the government to provide ballots means that portions of the people can be disenfranchised of their votes. How this passes the “republican form of government” clause, the 14th amendment’s “equal protection” clause, or in the near future the 14th’s section 2, I do not know.

          • Clutzy says:

            Its not incompetence, its a policy choice. There is no constitutional right to an absentee vote (although some courts will undoubtedly conjure it at some time). The state provides that privileged at their discretion, and if a flurry of people driven mad overwhelms the system that is not incompetence, its just a systemic weakness, no different than the TP shortage.

          • HeelBearCub says:

            Once the state sets the rules for an election, they aren’t supposed to be able to arbitrarily change them. Granting only some people who followed the rules, and not all, an absentee ballot is the part that is disenfranchisement.

          • Clutzy says:

            Only if those people legitimately had no other avenue for exercising the franchise. You could probably bring an equal protection claim if you were a bubble boy or something like that, but in that case you would have requested an absentee ballot far earlier.

            The constitution does not mandate impossibilities.

          • anonymousskimmer says:


            There’s no Constitutional right to an in-person vote, either. I believe a strict reading of the 14th amendment section 2 (“when the right to vote” [for] “the Executive and Judicial officers of a State” [is] “in any way abridged”) should reduce representation of Wisconsin by the fraction of voters who lawfully requested an absentee ballot in time yet were mailed that ballot too late to send it back to be counted.

            We have a spreading virus. How can you be certain a person didn’t become sick, or have a household member become sick, the last day to request absentee ballots, and immediately request an absentee ballot in order to vote?

            Likewise, how can you be sure that the person doesn’t work for a company that called them in to work during election day as essential personnel? Given the line lengths presumptive on election day (with the closing of so many polling stations), the three hours mandated time off to vote may be insufficient to vote.

            No, the constitution does not mandate impossibilities. Which is a reason why the Supreme Courts of Wisconsin and the US were wrong to insist upon post-by date. And why Wisconsin should have it representation in Congress reduced proportionately.

          • Clutzy says:

            Again, no. The reduction clause wasn’t even used when states had poll taxes. To invoke them you need intentional and discriminatory infringement.

            C-19 is no different than gang violence or seasonal flu or traffic in its reduction in your ability to go to the polls. Traffic fatalities are probably the most apt. The chance of dying in an accident going to the polls is probably within an order of magnitude of the chances of catching and dying from C19 in a single poll visit. People are pretending that going to the polls creates nearly 100% chance of catching C19. It isn’t close to that. If you were particularly concerned you can wear gloves, trashbags, goggles, and a mask and reduce your chance of catching it to something like .1% on a trip. If you do that, your chance of dying on the drive is much higher.

          • anonymousskimmer says:


            The huge decrease in the number of polling stations is not equivalent. It is action taken by the government which immediately abridged the ability to vote. The government needed to staff all polling locations, or create an alternate method of voting. It decided not to, thus it decided to abridge the ability to vote.

            I am fully aware that the text of the law doesn’t matter, what matters is how judges rule on how people should interpret the text of the law. This is unfortunate.

          • matkoniecz says:


            If you were particularly concerned you can wear gloves, trashbags, goggles, and a mask and reduce your chance of catching it to something like .1% on a trip. If you do that, your chance of dying on the drive is much higher.

            0.1% chance of dying on one commute translates to 30% chance of dying over 340 commutes.


            And 97% chance of dying over 3400 commutes.


            If 0.1% is a real chance then it is vastly higher than chance of dying in a traffic accident.

            (even after adjusting that not all infected from COVID suffer/die it is likely still higher than a traffic accident risk)

          • Clutzy says:

            Catching, not dying. You have to multiply by another .004 (at worst).

    • Some Troll's Serious Discussion Alt says:

      I doubt Trump would call for a delay. Hes got a big advantage in voter enthusiasm, pulling record turnouts in primaries that are basically just there to confirm him as the incumbent while Democrats are down from 2008 despite the wild battle their primaries were.

      It being a bit dangerous and scary to go vote favors him.

    • Edward Scizorhands says:

      Every state has absentee ballots. They need to change the rules so that we have absentee-voting-on-demand, and then scale up the number of ballots they print.

      (A reminder I hate absentee-ballot-on-demand. But there are lots of things I normally hate that we absolutely need to be doing right now, and this is one of them.)

      • zzzzort says:

        Really we only need to change it in a few states. Most states already have no-excuse absentee ballots, and most states don’t have competitive national elections because the US electoral system is dumb. So really it’s just Pennsylvania and Virginia and maaaybe Texas and South Carolina.

    • Tarpitz says:

      Leaving aside the question of what will happen, from my perspective as a Brit who has little enthusiasm for Trump or Biden but would marginally rather see Biden in the White House, what should happen if all coordination problems could magically be solved is a delay. We didn’t have a general election between 1935 and 1945. Neither Chamberlain nor Churchill (that time) won one. It didn’t do any lasting harm. No-one should be trying to change governments in the middle of a genuine externally imposed national emergency.

      • EchoChaos says:

        Americans would be furious. Both the USA and the CSA held elections on schedule during the Civil War.

        Our national character is such that we do this because we are proud of our heritage.

        • Matt M says:

          We also allowed bars to remain open and baseball games to be played and weddings and funerals to happen during the Civil War, too.

          We’ve already crossed lines we’ve never crossed before. What’s one more?

          • John Schilling says:

            Cancelling baseball games doesn’t cause half the United States Army to believe that its sworn duty is to enforce martial law like President Trump ordered while the other half believes that its sworn duty is to put down the Trumpist mutineers like President Pelosi ordered, or vice versa.

            Seriously, everyone who thinks the elections should be postponed: If the United States of America does not hold a presidential election in 2020, A: what is the name of the person who you expect will be the President of the United States on 21 January 2021 and B: what fraction of the people who disagree with you do you imagine will put down their guns just because people like you explain to them that they are wrong?

          • Tatterdemalion says:

            Seriously, everyone who thinks the elections should be postponed: If the United States of America does not hold a presidential election in 2020, A: what is the name of the person who you expect will be the President of the United States on 21 January 2021 and B: what fraction of the people who disagree with you do you imagine will put down their guns just because people like you explain to them that they are wrong?

            I don’t think it will or should happen, I think that the odds of it happening without bipartisan consent are virtually nil, and I think that if one side tried to postpone it against the wishes of the other the results would be just as catastrophic as you imply.

            But if the outbreak has gotten worse rather than better in November, it doesn’t strike me as beyond the bounds of possibility – vanishingly unlikely, sure, but not impossible – that the senior Democrats and the senior Republicans would agree to postpone it.

            And if that happens, whatever deal they strike will include an answer to that question, probably along the line of checks and balances that made it impossible for anything to happen without bipartisan consent, and I’d expect most people on both sides to accept it.

            What happens if Democrats and Republicans come up with a mutually acceptable deal to postpone the election, but the SCOTUS – not unreasonably – says that actually you can’t just ignore the constitution and lock in two-party rule, even if the alternative is an election where voting in person is dangerous and irresponsible, I don’t know.

          • HeelBearCub says:

            If we somehow don’t hold a general election in November, which I would put at a very small percentage chance, just over zero, I believe the constitution and the law would make Patrick Leahy President. There would be 67 senate seats whose term had not expired, making Leahy the ranking majority member. There wouldn’t be any House members, and the US would be fairly severely hamstrung in terms of making very much of anything non-executive happen, legally speaking, owing to no sitting House.

            I don’t think it’s germane, but it’s somewhat interesting.

          • John Schilling says:

            it doesn’t strike me as beyond the bounds of possibility – vanishingly unlikely, sure, but not impossible – that the senior Democrats and the senior Republicans would agree to postpone it.

            The problem is that “the senior Democrats and the senior Republicans” don’t have the legal authority to postpone it. It would take at least one constitutional amendment to do that, with three-quarters of the state legislatures signing on. Or, I suppose, a constitutional convention, but if we’re not having elections because it’s a bad idea to have lots of people coming together for a common purpose…

            I believe HBC is correct that a pedantically literal interpretation of the laws and constitution we currently have, would make Patrick Leahy president at noon on 20 Jan 21 if there is no election this year. But, good luck getting a broad coalition behind that. Good luck getting Donald Trump to agree to that.

            And if you do, remember that the lesson of both Donald Trump and Bernie Sanders is that many and possibly most Americans now believe that “the senior Democrats and the senior Republicans” no longer represent the will of the American people and should not rule over the American people. If “the senior Democrats and the senior Republicans” come to a backroom deal that makes a senior Democrat the new President in a way that looks not at all like a legal presidential election and requires a full-bore Voxplainer to walk people through why it might be legal anyway, then it’s the law that is going to break.

          • Doctor Mist says:

            John Schilling, can you elaborate? What I see in the Constitution as currently amended suggests a few workarounds.

            First, per Article 2 Section 1 part 4: “The Congress may determine the Time of chusing the Electors, and the Day on which they shall give their Votes; which Day shall be the same throughout the United States.” So I don’t see why Congress could not change the date.

            Now Inauguration Day is specified per Amendment XX: “The terms of the President and Vice President shall end at noon on the 20th day of January, … and the terms of their successors shall then begin.” So changing Election Day gives you only two and a half months of grace.

            But also in Amendment XX, “Congress may by law provide for the case wherein neither a President elect nor a Vice President elect shall have qualified, declaring who shall then act as President, or the manner in which one who is to act shall be selected, and such person shall act accordingly until a President or Vice President shall have qualified.” It doesn’t say anything about why neither one qualified. So it seems like legislation could, say, postpone the election until fall of 2021, when we might plausibly have a vaccine, and could even state who would be the President Pro Tem until then. (I can’t see any prospect of such agreement, since I assume “by law” means the House and Senate would have to pass a Bill and the President would have to sign it. But, you know, if we’re coming up on November and Covid is back worse than ever, and Biden or Trump dies from it, well, who knows? If both sides thought they might have a better chance a year later, you might see bipartisan agreement even neglecting the potential for saved lives.)

            I’m a little unsettled to conclude from this that for ninety years there has been a path for the Congress and President to appoint a permanent dictator any time they wanted to. What am I misunderstanding?

          • anonymousskimmer says:

            @Doctor Mist

            IMO only, and plenty of times I don’t understand how the government can get away with what I see as unconstitutional things.

            I think you’re missing that “shall have qualified” is past tense, not future tense. And that the Constitution itself is the sole arbiter of said “qualifications”. Congressional statute is not. So Congress cannot move the voting date past the 20th of January. It doesn’t have the power to eliminate this qualification in this manner.

            And two: the Congress which can choose a President is not the outgoing Congress, but the incoming Congress (which also wouldn’t exist if the election was moved).

            And three: Congress can’t do diddly about its own election (Article 1 Section 4 notwithstanding). So at most the longest they could appoint a dictator is not quite 2 years (unless sufficient numbers of the electorate were happy with the situation), or 4 years in the extreme event that they eliminate Congressional elections entirely by pushing them far into the future (and thus eliminate Congress). However given that the states are Constitutionally guaranteed Congressional representation, I don’t believe Congress can eliminate Congressional elections by pushing them into the future, and even if they could the states could just special elect or appoint new Congress people. So here’s your limit – 17 days shy of 2 years (Jan 20th to Jan 3rd) or the will of the people.

            And 4th: Presidential term limits.
            Aside: If you’ve got a popular president termed out, it may be worth running them again. Their VP pick would automatically become President.

          • Doctor Mist says:


            Well, the business of the dictator was a throwaway coda to my message. I’m really more interested in whether there is really anything that would prohibit a principled and bipartisan delay of the election in the current emergency.

            But since you are so kind as to engage with my absurd hypothetical:

            the Congress which can choose a President is not the outgoing Congress, but the incoming Congress

            “Shall have qualified” is definitely the future perfect tense, not the past tense. The language of the XXth is what enables the current legislation that specifies the order of succession, which is definitely an earlier Congress specifying who gets to be President at some unspecified point in the future. I don’t know anything that would prevent analogous legislation, except that instead of specifying the algorithm it names names.

            So at most the longest they could appoint a dictator is not quite 2 years

            There’s no automatic sunsetting of Congressional acts, so the appointment of a dictator wouldn’t just automatically expire when you got a new Congress. You’re right that a new Congress could end the whole thing by passing (with a veto-proof majority) legislation that actually names a new Election Day.

            The XXth does say explicitly, as noted, that a Presidential term ends on Jan 20, so I guess the hypothetical dictator would be bound by that.

            Term limits affect whether a person who served previously can be elected, which probably technically doesn’t affect whether someone could be appointed by Congress. Ooh, so bring Obama back, or Bush. That would be trippy.

            Another thing I ignored, both for my real question and for my absurd hypothetical, is that the election that matters is that of the Electoral College, on the first Monday after the second Wednesday in December. The Constitution doesn’t say anything about how Electors are chosen by the states; theoretically a state that is (desperately) worried about Covid could choose them in some way that does not require a state election, though probably some (or most?) states specify the process in their own constitutions.

          • Doctor Mist says:

            To be honest, if we* conclude that Congress can change the date of the election, it might well make more sense to move it up to late July or early August rather than to delay it until next year. There seems a fair chance that the danger from Covid will be much smaller in midsummer than it will be in early November.

            * The august body of SSC, who can surely be entrusted with that decision.

          • anonymousskimmer says:

            You’re correct about the tense. I meant that the indicated event took place in the future past, not in the future future. Therefore it’s possible Congress is prohibited from negating that future past event. This would be an unusual reading of the prohibition of ex post facto laws, but given the additional Constitutional weight against eliminating the election of a President, it seems a plausible interpretation.

            Moving the election up would cause the writer’s of the 20th amendment to roll in their graves, given the primary purpose of the amendment was to decrease the lame duck period. 🙂

          • Garrett says:

            Could individual States change their laws locally to “chuse electors” in a different way for this election? As one hypothetical example, they could randomly select Electors from a complete Statewide list of registered voters. Then you’d only have at-most a few hundred people who need to gather for the process.

          • Edward Scizorhands says:

            In theory, the Constitution says the States can do that.

            As soon as someone does that, though, you’ll get a lawsuit saying people are being disenfranchised, and you only need one Federal judge to agree to stop it up.

          • Doctor Mist says:

            Then you’d only have at-most a few hundred people who need to gather for the process.

            I think Edward Scizorhands has the right substantive answer, but it’s interesting that the Electoral College doesn’t actually meet as a body. According to

            The electors meet in their respective States, where they cast their votes for President and Vice President on separate ballots. Your State’s electors’ votes are recorded on a Certificate of Vote, which is prepared at the meeting by the electors. Your State’s Certificate of Vote is sent to Congress, where the votes are counted

            Sort of obvious when you think about it. The Electoral College is not a deliberative body — they aren’t going to debate the question, they’re just going to vote. It’s almost more of a surprise that even the Electors for a given state assemble. I guess it means there doesn’t need to be another entity that vouches for the results of the vote, and saves the Congress from having to validate 538 individual ballots.

            Anyway, this means the largest assemblage would be California’s with 55 electors.

          • Garrett says:

            > they aren’t going to debate the question, they’re just going to vote

            I wasn’t arguing that they would debate the issue, so much as that they would be required to gather. Hell, they could all be given level-A hazmat suits.

            I took the Constitutional phrase “The Electors shall meet in their respective States” as a requirement that all the Electors end up in the same place at the same time to sign/seal/deliver their ballots.

          • Doctor Mist says:


            Yeah, I think that’s right. One might imagine some interpretation of “gather” for modern/coronavirus times to include videoconferencing, as I gather even the Supreme Court is using to hear cases. But my main observation was that you won’t see an assemblage of 538 Electors even in normal times. Not that that matters at all for what we are discussing.

      • Edward Scizorhands says:

        Our next Presidential term starts on January 20th. It’s written into the Constitution.

        We have regular elections on a schedule, not when there’s a vote of No Confidence.

      • Lambert says:

        I, for one, would love to see what forcing the republicans and democrats into a Government of National Disunity would look like.

      • Loriot says:

        As an American, the idea of canceling the election is simply unthinkable. No matter what happens, half the country would be (metaphorically) up in arms.

    • DaveK says:

      Worth noting that even before the pandemic, people opposed to Trump expressed concern with a scenario where the vote was close (which it probably will be), there are irregularities, (which there will be) and Trump and a significant contingent of his supporter refuse to accept the results, alleging a democrat/deep state conspiracy.

      Given that I’ve heard republicans already discussing fears about a conspiracy to steal the election from TRump and how this must be resisted or it’s the end of democracy, and how a number of people on the left have thought about this as a real possibility, I don’t think such a scenario is so remotely unlikely as to not be worth considering.

      • EchoChaos says:

        Trump and a significant contingent of his supporter refuse to accept the results, alleging a democrat/deep state conspiracy.

        Well, we’ve already seen what happens when the Democrats and a significant contingent of their supporters refuse to accept the election, alleging a Russian/Republican conspiracy.

        Why do you think the opposite way would have any effects any more substantial?

        • John Schilling says:

          Aside from the four more years of political polarization, and the differences between Republicans and Democrats, and the differences between vote-by-mail and in-person voting, there is the difference between the Russians of 2016 and the Russians of 2020. The Russians have had four years to study what does and does not work in terms of reducing the perceived legitimacy of America’s elected leaders, and are almost certainly better at it than they were four years ago. The Chinese, North Koreans, et al, have probably been paying close attention as well.

          Unless the United States takes steps to increase the perceived security of the electoral process, it is reasonable to expect that there will be more vocal and active dissent this time around – regardless of who wins. If instead we basically admit that we are going to increase our use of less-secure electoral processes because coronavirus, that’s just going to make the legitimacy problem worse.

          • Garrett says:

            > reducing the perceived legitimacy of America’s elected leaders

            From a political science perspective, all legitimacy is perceived. This is why the Divine Right of Kings needed as much support from the Church as possible. And if you couldn’t get it, you founded your own Church which taught the message you wanted taught.

            In more enlightened times we attempt to legitimize the structure on rational grounds. Sure, we don’t have a command from God, but at least we have a system that people can look at, shrug, and go “I can’t think of a way to do better across all possible metrics”. Also, we have the school system teach about how totally awesome our government structure is. Add a bit of history to add gravitas and away you go.

        • eric23 says:

          Why do you think the opposite way would have any effects any more substantial?

          Because Democrats (and Republicans not named Donald Trump) have historically showed a lot more concern for the law and the national good than Donald Trump has?

          • EchoChaos says:

            @Scott: Accidentally reported. Nothing wrong with the comment.


            I am going to suggest that partisan effects may be blinding you a smidgen here.

            From my perspective the Democrats spending 3+ years on a clearly spurious investigation was worse than anything Trump has done in terms of lack of concern for either the law or the national good.

          • eric23 says:

            a clearly spurious investigation

            I’m going to suggest that partisan effects are blinding you, not me. I find it hard to describe an investigation that convicted Trump’s personal lawyer, campaign chairman, and several close advisors, and did not clear Trump of accusations but rather concluded that it lacked the authority to charge Trump, as “clearly spurious”.

          • Doctor Mist says:

            Why blame me, Doc? You’re the one with all the dirty pictures.

          • Garrett says:

            > historically showed a lot more concern for the law and the national good than Donald Trump has

            Counterpoint: one of the things that makes Donald Trump particularly interesting is that he seems to be a master at (mostly) following the letter of the law while spitting in the face of the spirit of the law.

            As for concern for the national good, I find it hard to develop a model which separates someone who doesn’t care for the national good but claims to care, someone who cares but only secondarily to other goods, and someone who cares for the national good but simply has the wrong policies to go about it. That is, separating malice from stupidity is very challenging, especially with someone like Trump.

  12. TheTurtleMoves says:

    Volunteer opportunities: here in NYC a lot of volunteers are just running errands for people, such as getting groceries for people at high risk of infection and doing no contact deliveries. Seems like that would be necessary everywhere.
    If there’s no one organizing that wherever he or she is it seems like word gets around pretty fast with as little as a Facebook post.

  13. noyann says:

    What ex-infected, immune helpers can do

    – Run errands for people that are high risk (at home, in nursing homes) or in hospitals with no visitors. Some communities and churches may already have organized something to join.
    – Shops may want someone for delivery services.
    – Entertain the kids for a few hours/day in large families where both parents work from home.
    – Volunteer in hospitals or nursing homes take low qualification tasks off the staff.
    – Keep company or entertain patients/seniors who don’t get visitors.

    • SolipsisticUtilitarian says:

      Working in nursing homes would be extremely beneficial, as nursing homes are both full of at risk people and a perfect breeding ground for a fast spread of the virus. Therefore reducing the likelihood of a sick nurse introducing the virus should be a key concern; I can imagine a team of stand-by volunteers with a quick training in the basics of nursing temporarily replacing a team of nurses when there is a suspected case among the nurses.

      • Don P. says:

        We should note that an immune person can still transmit the virus by touch, so they’re not a panacea.

        • HeelBearCub says:

          No one is, or at least should be, suggesting that they not engage in proper hygiene.

  14. Gilbatron says:

    People who are immune should consider donating blood/plasma. A transfusion of antibodies can help.

    Many universities and large clinics in europe are looking for these donations right now. Pretty sure it’s the same in the US.

    Edit: this is not about regular donations. It’s about donating to a place that makes works on that specific treatment.

  15. The Nybbler says:

    Case numbers which include the US are still not meaningful. New York and New Jersey at least are not ramping up testing nearly fast enough to produce numbers which aren’t mostly artifact, and these states have the largest number of cases in the US. Daily positivity in NJ has been around 50%, NY was 50% but has dropped to around 40%; they’re still only testing people with severe symptoms.

    There’s no civilizational decline. If the government ever legalized building things quickly again, we’d be mopping the floor with China within weeks.

    Big laconic “if” here. That’s never going to happen; the government regulations are the mechanism of the civilizational decline. It’s good that we haven’t forgotten how to build things, but that red tape isn’t going away.

    From the risk to young people article:

    Young people exposed to COVID-19 are as likely as old people to become infected and contagious.

    The data I posted in an earlier thread from the Diamond Princess suggests otherwise, and nothing in the article supports this statement that I can see.

    • DaveK says:

      “Big laconic “if” here. That’s never going to happen; the government regulations are the mechanism of the civilizational decline. It’s good that we haven’t forgotten how to build things, but that red tape isn’t going away.”

      I do think the pandemic is likely to create political pressure for significant changes.

      I’ve already heard politicians and people from both sides of the partisan divide pointing out that it might be time to rethink the cost/benefit analysis of having all our supply chains be globalized and often under the control of nations which may be hostile.

      Of course these kind of discussions could go either way. There have been a lot of people arguing that China should be a model for their effective response, and the US should be organized more along centralized lines. It’s somewhat ironic, as these people are often the ones to be most critical of the failures of the federal government, but seem to be arguing a solution is to give the federal government more power and more responsibility.

      • The Nybbler says:

        I do think the pandemic is likely to create political pressure for significant changes.

        Sure, like you say, changes towards the Chinese authoritarian model. More power for public health authorities, stronger price-gouging and rationing laws, maybe government stockpiling of masks. If Trump wins in November, probably some made-in-America requirements. Lasting travel restrictions and intrusive anal physical probes at the airport (used mostly to catch drug mules of course). But we’re not going to see a loosening of FDA regulations, let alone any sort of of relaxation of the various restrictions which make building hard — the National Environmental Policy Act, the Americans with Disabilities Act, various gold-plated building codes, union requirements, etc. Those are outside the Overton window and getting further outside.

      • Christophe Biocca says:

        I’ve already heard politicians and people from both sides of the partisan divide pointing out that it might be time to rethink the cost/benefit analysis of having all our supply chains be globalized and often under the control of nations which may be hostile.

        Mercantilist (exports good, imports bad) positions have been held by the majority of Americans since before independence, though. Of course people who already want to decrease imports will claim that coronavirus demonstrates the wisdom of that position. Everyone is interpreting this thing more-or-less as a confirmation of their priors.

  16. smilerz says:

    Setting up mail-in voting nationwide will not be a trivial task.

    There is essentially no chance that it will be done well, and will open itself up to scores of lawsuits and court battles and like Scott, my biggest concern will be Bush v Gore times 1000.

    • albatross11 says:

      Changing the rules or operation of an election on short notice is always going to be painful. You’re dealing with local and state laws, but also with what equipment and resources local election officials have, and a lot of local election officials have very little money, not much staff, and a pool of mostly old retired volunteers who are probably not showing up for the COVID-19 election when this virus is so dangerous for old and sickly people.

    • Edward Scizorhands says:

      archive link to get past paywall

  17. Gilbatron says:

    the entire german immunity certificate thing has been blown completely out of proportion and badly translated.

    a reporter at a press conference asked about that and the answer was that such certificates were possible.

    that’s not the same as “germany to introduce immunity certificates”

  18. RC-cola-and-a-moon-pie says:

    I hate to post two separate questions under the same post, but I can’t resist posing one more. Has anyone seen any further analysis of the zinc question? My takeaway from last time I saw this issue discussed here was (1) there were decent theoretical (or analogical from other related conditions) reasons to think that early consumption of the right kind of zinc lazenges may reduce the severity or duration of the illness caused by this virus, but (2) there was no direct evidence of an effect on this particular virus and in any event any effects were unlikely to be large, and (3) there were some worrying risks, including even frightening neurological-related effects that might come with ingesting these things. My tentative bottom line was that if I contracted covid-consistent symptoms that I would suck a lazenge or two given the seriousness of this virus, but not without some trepidation. Have there been any sort of developments in this area? Is my takeaway at least reasonable?

    • keaswaran says:

      My takeaway of the discussion was that I should take a zinc lozenge if I think there was a possibility I was infected in the past couple hours, but that if I wait until symptoms develop, then it’s too late for the zinc to do much.

      However, the discussion of the importance of viral load is making me think there might be something relevant. If we think of the body as a city and the cells as people in the city, then the idea is that the height of the peak of the epidemic is determined by how many infected people there were at the point that protective measures started being put into place (which appears to be what we are seeing with the differences between European countries and between American states).

      The zinc would then function as something that prevents spread to the cells that it directly coats. In the early phase of the infection, this might help the body’s response overwhelm the outbreak, but even in somewhat later phases it might do something. But the analogy suggests that it would still only be relevant in the ramp-up, when the number of infected cells is increasing quickly. Which still suggests that once you’ve developed symptoms might be too late, if a lot of people don’t notice symptoms until a few days into the infection.

      • albatross11 says:

        I’m pretty sure a lot of hospitals are currently trying zinc with their COVID-19 patients, but it’s possible it either doesn’t help much or doesn’t help much at that stage in the infection.

      • Rick Jones says:

        My understanding is that Zn interferes with rna transcriptase but this only takes place inside the cell so having lots of Zn around In the plasma doesn’t help unless there is a way to transport it into the cell. One report suggested that might be how plaquinil works, if it does. And there’s also the possibility that the is a Zn transport system we don’t know of.
        But if it’s this simple then why doesn’t it work with other retrovirus like HIV?

        • anonymousskimmer says:

          If HIV has already integrated into the host genome it doesn’t need functional reverse transcriptase anymore.

          So if your understanding is true, its plausible that sufficient intracellular zinc would prevent HIV from incorporating at all. But why would anyone take this amount of zinc in the forms allowing it to be present intracellularly before already being infected with HIV?

          Coronaviruses aren’t retroviruses, so the mode of action of zinc would have to be something else.

          Edit to add:
          Zn2+ Inhibits Coronavirus and Arterivirus RNA Polymerase Activity In Vitro and Zinc Ionophores Block the Replication of These Viruses in Cell Culture

          • Rick Jones says:

            Good points. And thanks for correcting my misconception of sars2 as a retrovirus.
            But you still need a transport mechanism for zn to enter the cell which I guess the pyrithione effects?

    • Lambert says:

      Speaking of things that might stop it in the early stages of infection, has anyone studied the effect of trypsin on coronavirus?

      Coldzyme AB make a trypsin-based throat spray that’s been found to be effective at stopping colds, if you use it as soon as symptoms are beginning. Trypsin is a protease that damages the viral proteins so they can’t infect cells.
      But the studies tend to be about the effect on things like rhinovirus and influenza.

      • fallenscien says:

        Seems unlikely to help.

        SARS-CoV-2 is a lipid-enveloped virus studded with glycoproteins. The proteins are so covered in glycans that there’s hardly any protein surface that’s accessible – that’s one of the reasons it’s so good at evading the immune system.

    • DaveK says:

      I think mega dosing vitamins and mineral is a dubious proposition. While nutrition science is generally pretty weak, it is clear that balance is key.

      I would think the best option is to split the difference. Take zinc supplements, but not lozenges.

  19. dark orchid says:

    It would be premature to say we’re now winning the war on coronavirus. But we’ve stopped actively losing ground. If we were going to win, our first sign would be something like this.

    Sounds like a pretty good parapharase of: “This is not the end. This is not even the beginning of the end. But it is, perhaps, the end of the beginning.”

  20. Jaskologist says:

    Varad Mehta has a pretty balanced piece on what went down with the Wisconsin election.

    tldr; Democrats had scheduled the elections in order to give themselves and edge in the court elections. Gov Evers refused to delay the election because he doesn’t have the legal authority. He finally asked the Republican legislature to move the election (possibly after polling indicated the Democratic judge was going to lose), and they gave him the finger, so he declared it moved without them.

    Courts ultimately agreed with Evers’ earlier statement that he doesn’t have the authority to move elections unilaterally.

    The details of mail-in ballots were probably of secondary concern.

    • fallenscien says:

      It bothers me that Mehta states, “Evers’s dithering and fecklessness is the primary cause of the imbroglio” when Evers never had the legal authority to delay the election.

      Delaying the election would have required Evers and the Legislature to both agree to a new plan. Evers flopped all over, but the Legislature stuck to their guns in making the election move forward as normal.

      Evers was assuredly dithering and feckless. But how is he the primary cause of an outcome he had no ability to change?

      I’d think differently if the Legislature had passed a bill and he’d vetoed it. But that’s not what happened.

  21. sclmlw says:

    Question: Which of the two interpretations of “flattening the curve” are we actually achieving? If it’s one where the same number of people get infected but we’re just prolonging the crisis (the one from the now-famous infographic) isn’t there a potential downside to flattening the curve?

    If we prolong a pandemic of a slowly-mutating virus, aren’t we giving it time to mutate and thereby ensure we experience subsequent waves? What if the next wave is worse than the first? Will people accept ongoing waves of social distancing, or would there be rioting in the streets? What if prolonging the crisis is actually causing additional crises down the road? It seems likely that a virus capable of spreading as fast as COVID-19 would flame out quickly and disappear if it were allowed to spread unimpeded – as painful as that might be. Such that the difference between the null response and flattening the curve is a difference between a bad pandemic of 2020 and a newly-emergent endemic disease. If our default response to this kind of pandemic is to slow down the spread, does that ensure newly-emerging pandemics will all become endemic?

    If, however, flattening the curve reduces the overall number of people who get infected – by controlling spread to the point of eliminating the virus – we would expect a lower overall number of cases and therefore fewer opportunities for the virus to mutate. If this happens, we should expect to see less mutation in populations that flatten the curve. (cf. Brazil, Turkmenistan, etc.)

    If flattening the curve means exactly what the infographics people usually refer to are implying – namely that the area under the two curves is essentially the same – it’s potentially a bad thing, creating new crises down the road. If it means fewer overall cases, it’s probably preventing new crises down the road.

    Disclaimer – I’m not suggesting we change course. Even if you think this is the wrong response we should make sure it’s as effective a response as possible.

    • HeelBearCub says:

      The answer is that we don’t know which version, because it depends on future actions.

    • albatross11 says:

      I can think of a few reasons it makes sense to delay infections even if we end up with the same number overall:

      a. We can avoid having hospitals get overwhelmed with COVID19 pneumonia patients all at once, which is important because it means that the patients are more likely to survive. Your chances of surviving go way down when you’re on a gurney in the hallway with ten other very sick people, and one beginning-to-show-sympoms nurse is trying to keep track of you all while hoping the Tylenol she just took gets her fever down a bit.

      b. We buy time to learn more about how to treat the virus. Wait long enough, and there may be a protocol that everyone adopts that doubles the survival rate of people who end up in the hospital with the virus. Lots of people are trying lots of different things to treat the infection at different times, and hopefully some of them will work out.

      Both of those will pay off even if we end up with the same number of total infections, as the later infections will be a lot less likely to end up dead or permanently disabled.

      We also buy time to work out how to adjust our world to slow down the spread of COVID19 with measures short of total lockdown. The virus isn’t all *that* great at spreading–estimates are that each new infection causes about 2.3 new infections on average. Get that number down below 1, and the total number of infections should keep dropping.

      Places with lockdowns have probably got R way down, Unfortunately, they’re not sustainable long-term. But some less restrictive changes probably can keep R low enough that we can keep on top of it with testing and selective quarantines and maybe occasional localized lockdowns. My hope is that we can get the number of active infections down enough to get this mostly under control. But that seems to require, at a minimum, lots of continued measures to slow transmission, along with really extensive rapid testing (tests that give you an answer in a couple hours instead of a week).

      • glorkvorn says:

        What sort of “less restrictive measures” are you imagining, that would keep the spread slow but be tolerable for the long haul? I can’t imagine that we can keep the current situation up for more than a couple more months at best. I’m worried that this massive effort to flatten the curve is all in vain, because we’re going to end up just letting it spread anyway.

        • Edward Scizorhands says:

          Just spitballing, but something like “End most government-enforced lockdowns except 1) still ban events over 1000 people, 2) keep schools closed, 3) keep a hyperactive eye on nursing homes and other concentrations of the most vulnerable.”

          That would stop a lot of super-spreader events. And even with no government orders in place, lots of people and businesses are still vary of the virus, so would take a bunch of actions on their own.

          I don’t know the models. The modelers probably don’t know the models, and not because they are stupid, but because it’s an incredibly hard problem. They probably won’t know for sure until they get a chance to compare the data from different countries and regions that did various mitigation efforts, and then further watched in more controlled fashions as regions and countries try unlocking things and observe what we can get away with.

        • albatross11 says:

          It depends on how the virus spreads most of the time–something that’s not entirely clear. What I’ve read so far suggests that:

          a. There are known cases where people have spread the virus before developing symptoms. The claim I read said that when there’s asymptomatic spread, it’s typically 2-3 days before symptoms start. My guess is, there’s almost always a period when you could transmit the disease but don’t yet have symptoms, but you’re probably not all *that* contagious.

          b. There are many cases where people never got all that sick–they felt like they had a pretty bad cold, but didn’t want to miss work/school/their cousin’s wedding/their grandma’s funeral.

          c. There are cases where we know airborne transmission has taken place in a community setting–that choir meeting where everyone did social distancing and avoided close contact and used hand sanitizer, and a bunch of people got sick anyway is an example. At a guess, singing out like you do in a choir is probably producing a lot of droplets that have a lot of energy coming out of your throat and travel far. (Coughing and yelling probably work about the same way.)

          d. We know there is airborne transmission in hospitals, and also that hospitals have become huge sources of additional infections in a lot of places. I would bet a lot of money that right now in most of the US, nearly all new infections are coming from hospitals to staff/EMTs/other patients, and then from them back out into the community.

          e. The cases people have been able to track down don’t seem to show a *lot* of airborne transmission, and the R_0 of COVID-19 doesn’t look like it probably would if every sick person was leaving a cloud of infection behind them for a couple hours.

          The reason I care about airborne transmission is that it’s the hardest thing to stop. For most workplaces, I think it’s relatively easy to keep people 2m apart, make them wear masks (once we get the mask situation untangled), get them to wash or sanitize their hands many times a day, and disinfect all the high-touch surfaces. At that point, you’ve eliminated most opportunities for close contact and contaminated surfaces (“fomites”) like doorknobs and elevator buttons. But if everyone who breathes, eats, coughs, yells, or sings is putting very small virus-laden droplets into the air that can travel across the room and infect someone 10m away, that looks very hard to prevent.

          The reason I care about pre-symptomatic spread is because it’s pretty easy to tell people to go home if they come into work visibly sick. A fever check with a forehead thermometer at the start of the day and sending anyone who’s coughing/sneezing home catches most of the symptomatic infected people. Until we have cheap rapid virus tests, that’s probably the best we can do, but it doesn’t do anything to prevent spread by people who don’t have any symptoms.

          If airborne spread is either very rare from asymptomatic people, or blocked by making everyone wear a mask, then I think it’s not so hard to reopen most workplaces. Everyone gets a fever check in the morning, anyone showing symptoms has to go home or maybe go into a separate building/room to finish out the day, everyone stays 2 meters apart, everyone wears a mask, surfaces get disinfected regularly, we space out lunches and such to avoid close contact, etc. Similarly, you can imagine reopening restaurants for in-house dining, just spacing the customers pretty far apart and sanitizing surfaces and making the servers and kitchen staff wear masks.

          If airborne spread from asymptomatic people is relatively common, then that stuff doesn’t work so well. In that case, it’s hard to keep stores open and allow restaurants to open back up their dining rooms without having rapid spread. It’s also hard to go back to work in an office or warehouse or factory, because just spacing people out isn’t enough–you need to redesign the ventilation system or make everyone wear respirators or something.

          • Edward Scizorhands says:

            Based on the choir event, and the fact that Japan kept the spread low by people wearing masks on subways, maybe airborne transmission is more important than shared-surface transmission.

            I vaguely recall that the Biogen cluster or the early Korean church super-spreader was considered to be by common serving utensils at a buffet. If that’s true, it would mean my model is wrong. But was it really the common utensils at fault, or was that just an idea while we were more concerned about hand-washing and shared surfaces?

      • sclmlw says:

        I’m not disputing the potential benefits of reducing the spread; I’m familiar with all the arguments you outlined. I think we’ve discussed those upsides a lot recently, and you’re probably right that they should remain part of the discussion as we talk about pros/cons.

        Meanwhile, I’m asking whether there’s a con we haven’t considered yet: what if the end result of flattening the curve is everything you predicted, plus there’s an extra wave? Presumably we’ll be more prepared for the second wave, but it might be worse to go through this 2+ times than to go through a particularly horrible single event.

        • HeelBearCub says:

          It entirely depends on what the results of an uninterrupted first wave would have been. We have a sense of what that would be based on Wuhan, Lombardy and NYC. The UK started down this path and seem to be paying a heavy price. We may be about to get very familiar if Sweden is on the course they are reported to be on.

          So, assuming that an uninterrupted course really does result in 2 million deaths in the US … well, you won’t get an uninterrupted course. Lots of people, corporations, etc. will be be making the choice to voluntarily attempt to isolate.

          And that could end up in the worst outcome, because you get much of the economic damage without the the benefits of a coordinated response.

          • albatross11 says:

            Right, that’s the point of the graph I linked to elsewhere in the thread–restaurant reservations were falling off like a rock everywhere as news of the virus spread, not just in places with lockdowns.

      • nkurz says:

        > I can think of a few reasons it makes sense to delay infections even if we end up with the same number overall

        Another possible reason is that by waiting, you might be infected with a less harmful strain of the virus. Approximately, the argument for this is:

        1) There appear to be multiple strains of the virus.
        2) Severity of symptoms (and mortality) depends on strain.
        3) Strains with lots of asymptotic carriers are more likely to spread.
        4) Over time, the milder strains may outcompete the severe strains.

        I don’t know how likely this scenario is, but given expected human behavior I think it’s at least plausible. Consider two strains, one of which causes 20% severe bedridden cases and 80% asymptomatic; and the other 80% severe and only 20% asymptomatic. While there is probably an early window in both strains where the unknowing asymptomatic carrier is out spreading the virus, in the severe strain the carrier falls ill and is bedridden with limited outside contact. The asymptomatic carriers of the mild strain continue their normal daily routine, infecting others for a longer period of time.

        If we assume that a large amount of spreading is done by a small number of super-spreaders, this effect magnifies. An even minimally aware person isn’t going to drag themselves out of bed to attend their weekly choir practice, but a feverless individual who things they are healthy individual might choose to attend. In the right circumstances, this seems like it might lead to more people being infected with the mild strain, until herd immunity is reached and the severe strain becomes much rarer. There are lots of assumptions necessary for this to happen, but it seems worth modeling.

        • sclmlw says:

          4) Over time, the milder strains may outcompete the severe strains.

          This is not a timescale we’d want to rely on. Mutations aren’t directed, they’re stochastic. So the mechanism that selects for milder strains over time is the aggressive nature of the more severe strains limiting their spread and/or persistence within the population. In other words, it could easily get worse before it gets better.

          • albatross11 says:

            One thing that could cause milder strains to outcompete severe ones is if everyone who shows any symptoms is quarantined and doesn’t give the virus to anyone new, whereas asymptomatic people can still pass the virus around.

          • sclmlw says:

            Sure, but that could easily be an extended incubation period, or increased viral shedding in asymptomatic people; there are lots of potential mutations that could increase the spread but not impact lethality.

    • DarkTigger says:

      Those two interpretations don’t need to be mutual exclusiv. We can hope that we are able to flatten the curve to a point were R<1 which means the number of new infections might fall so far, that we can return to no lockdown and contact tracing all new cases.
      We can still prepare for the case, were we do not have that much luck, and the area under both curves are the same. It still gives us more time to find a vaccine or a cure, while our public health sector is still in a shape to administer it.

    • anonymousskimmer says:

      If we prolong a pandemic of a slowly-mutating virus, aren’t we giving it time to mutate and thereby ensure we experience subsequent waves?

      Not my expertise, but a virus doesn’t need “time” to mutate, it needs generations to mutate. If it infects the same number of people, and reproduces inside those people the same number of generations, it will have statistically the same number of mutation events.

      So theoretically the choice is between rapid exposure to a variety of coronaviruses (each of which can make you sick, unless you’re lucky in antibodies), or waves of exposure to a variety of coronaviruses (each of which can make you sick, though less frequently).

      If you’ve got a variety of strains circulating at the same time you have the opportunity for recombination between strains (in individuals infected with both), which may actually accelerate population diversity (e.g. flu H1N1 and H3N2 recombining into H1N2 and H3N1).

      • Rick Jones says:

        Where does anybody get the idea that the virus has mutated? According to Vince Rucinello at twiv, the virus has changed 15 bases since it was first sequenced some months ago.15 of 30,000. That’s 26 bases changed in a YEAR. And most, if not all, of those changes are silent. If you are talking out you rear about strains, please cite some actual references showing sequencing. The virologists I listen to claim that one of the characteristics of SARS2 is it’s stability. Which shows how little selective pressure it is under right now.

        • anonymousskimmer says:

          Darn it Rick. See the post I am responding to.

          That’s 26 bases changed in a YEAR.

          And also, read my post. Viruses do not have a mutation rate per YEAR, they have a mutation rate per GENERATION. While viruses aren’t my expertise, I know enough about polymerases to know about mutation rates.

          And yes, coronaviruses are one of the kinds of viruses that can swap domains between strains.

          SARS-CoV-2 is under as much selective pressure now from human immune systems as it ever will be (short of drug therapy, which add new molecules to select against, or which theoretically could increase the mutation rate of the virus by increasing the error rate of the RNA polymerase that replicates the virus).
          Don’t mistake mutation rates with selective fitness. Every new antibody generated by a human immune system is another attacker selecting against the virus. But this doesn’t change the fundamental mutation rate of the virus, it merely means that any mutant virus which survives the antibody binding and immune system attack will be selected for.

          • Rick Jones says:

            I’m learning my virology on the fly so be patient. I’m basing the 26 base per year change on listening to The week in virology podcast by Vincent Rancaniello and others. It was Vince who used the 26 number but I take your point that it’s a bogus unit and talking in terms of replications is better. So if we’ve had 15 base changes since the first sequencing, what should be the correct denominator? Megareplications? But there seems to be a pretty strong narrative that the virus is pretty stable.
            Also, if sars2 is an unsegmented virus, doesn’t that make changes less likely to either happen, or if they do, to be less virulent?
            I have to admit that I find what I read about adaptability of virus in vivo to be confusing. Richard Epstein was ridiculed for suggesting US deaths would be less than 5k, which he based on the idea that the virus would slowly become less deadly as it spread through humans. The disastrous New Yorker interview he gave even included real time refutation by a scientist. But I’ve also read that’s something like this process does occur. Probably not in the short time we are talking about here tho.
            Finally, I have a hard time tracking comments in these long comment strings. Any tips would be appreciated.

          • anonymousskimmer says:

            Now you’re asking questions I can’t answer. 🙂

            Thanks for the link. I’m learning things as well.

          • albatross11 says:

            One datapoint here: This preprint claims a different strain circulating in Singapore with a chunk of the viral genome deleted. This makes an argument that different strains may be arising, and it’s even possible that the less virulent strain was the one that hit Singapore first, and that accounts for some of their success in avoiding many deaths. But that’s very speculative.

            As I understand it, the argument that we’re probably not seeing very different strains arising is that coronaviruses tend to be pretty stable, and that COVID-19 seems not to have changed much over six months of circulation. That suggests that the virus is unlikely to split out into radically different strains that have very different properties.

          • albatross11 says:

            I think the idea here is that if most of the possible mutations for COVID-19 make it less fit, then those mutated strains will tend to die off, while the strains that stayed about the same will continue successfully spreading.

          • Douglas Knight says:

            No, this virus is special. It has a very low mutation rate, but most other RNA viruses have very high rates. I don’t know about coronaviruses in general, but SARS-1 had typical RNA speed.

          • nkurz says:

            > the argument … that COVID-19 seems not to have changed much over six months of circulation

            I haven’t read the underlying paper ( but this quote from a recent press release ( seems to say the opposite: “There are too many rapid mutations to neatly trace a COVID-19 family tree.” “The viral network we have detailed is a snapshot of the early stages of an epidemic, before the evolutionary paths of COVID-19 become obscured by vast numbers of mutations.”

            Is this just different standards for “too many/vast numbers” versus “not changed much”, or are these two in contradiction?

          • matkoniecz says:

            Maybe COVID-19 mutates too fast to trace family tree, vastly slower than other similar viruses and mutations are not affecting important parts so “not changed much” is still true?

        • fallenscien says:

          @Rick: Check nextstrain’s tracker here:

          Remember that this is an underestimate of the genetic diversity of SARS-CoV-2 in the wild, because most countries are massively undertesting and not sequencing samples from people who are infected but asymptomatic (likely giving us a narrower slice than random sampling). We’re in the millions for total cases, and in the thousands for sequences.

          We’re also always running behind. Incubation time is a delay, as is time to sample and sequence. All our results represent the state of things as they were a couple weeks ago. Cases are still growing exponentially.

          The virus is pretty stable, but the number of infections is huge. There’s a big pool out there.

          Viruses are very information-dense, with massive selective pressure to have highly-efficient genomes. Very small changes can cause major ripple effects – a single nucleotide change can cause a conformational shift in a protein which dramatically alters its effectiveness at binding host receptors, or cause changes in glycation so the virus looks totally different to the immune system. 15 base pair changes out of 30,000 is a lot.

          • Rick Jones says:

            I think the nextstrain site was where Dr. Rancaniello got the idea of the 15 base change. Also they say that sars2 has a proof reading enzyme to correct for errors. But the last episode of the podcast they mention the Singapore data with a 34 base deletion.
            I’ll write them an email asking about this. Let’s see if they answer on air which they do from time to time. I’ve got a lot to learn for sure. Also thanks for all the helpful and respectful comments. A refreshing change from other sites.

      • Purplehermann says:

        If the immunity doesn’t last long time could be an important factor

        • anonymousskimmer says:

          True. I don’t see how that changes things much, though. At least in the short to medium term.

          • Purplehermann says:

            It could change the calculus for how fast we would want to get herd immunity naturally.

            If going slow can’t do it then flattening the curve in the “everyone will get it sense” besides being harder would also allow for more virus generations

          • anonymousskimmer says:

            Or you get people catching multiple strains at once.

            About the only effective way to utilize this would be to synthesize the virus in a high-fidelity manner (eliminating most chances for mutation), and infect everyone with this unmutated strain at once. Though it’s possibly also effective to have designated super-spreaders spread it to everyone simultaneously (thus minimizing the potential for mutations to build upon prior mutations).

            We may as well “flatten the curve” and wait for a poly-antigen vaccine.

      • sclmlw says:

        Yeah, but there’s a difference between serial and parallel approaches here.

        To take an extreme example, think of a virus’ mutation if one person spread it to a billion people at a time. It has a few billion chances to mutate, but only about seven rounds to accumulate mutations. Now compare a virus that is passed from one person to the next billions of times. There is redundancy built into the genetic code, such that most mutations will be benign. Even mutations that change an amino acid usually just change from one hydrophobic aa to another similar one, not impacting overall function. Certainly the antibodies of the person who passed the virus to you will still be able to recognize the slightly mutated version.

        However, if you continue to accumulate mutations over time – by slowing down the reproduction rate to >1, but less than what it would be without intervention – you’re ensuring the virus goes through more serial rounds of mutation events. The resulting virus, like the classic game of ‘telephone’, will carry dramatically more mutations than one with a much shorter path through the population. Such that by the time it makes it back around to you it has mutated beyond the point where your immune system is capable of recognizing it anymore.

        • anonymousskimmer says:

          I recognize this. This is why I mentioned “super-spreaders” in an earlier comment.

          You are ignoring the fact that for one person to be that contagious they have to have many replication (thus potential mutation) events in their own person. One person cannot spread the virus to a billion people.

          Ultimately I still believe it’s better to minimize the number of infections until a vaccine or successful treatment is developed.

          • sclmlw says:

            Right, like I said the billion-person super spreader is an extreme example used to illustrate the concept of how serial versus parallel infections impact mutation rate.

            The comment of yours I was responding to stated, in part,

            a virus doesn’t need “time” to mutate, it needs generations to mutate. If it infects the same number of people, and reproduces inside those people the same number of generations, it will have statistically the same number of mutation events.

            As I read it, that comment is claiming that it’s a simple matter of adding up the total number of infected people and asking whether it’s the same in one scenario as in another. I think that’s a justified interpretation of the comment, since this was in response to my initial query about whether slowing down the rate of infection (bending the curve) could be causing additional waves of virus if the reproduction rate remains above 1.0. Regardless of super spreaders, if you lower the reproduction rate, but not below 1, such that the virus remains in circulation longer you’re increasing the probability of creating additional waves. This is because the path matters in accumulating mutations, which your earlier comment assumed was unimportant.

            (Obviously, if our efforts successfully reduce the replication rate get it below 1, then fewer people overall will be infected and the calculation gets a bit more complicated.)

            Whether the current approach to limiting the spread of disease is the best solution is speculative. If we go through another round of SARS-CoV-2, where a similar number of people die to the first wave, and our current vaccine is no longer any good, it’s possible that more people die than if we acted differently and experienced only one wave. Of course, you can never know whether the counterfactual would have prevented the second wave, so in part the question of which approach would be best will likely remain unknown.

            My intention is just to introduce this idea into the conversation as it isn’t something I see generally discussed. Each potential solution has both positive and negative consequences. That doesn’t invalidate the current course of actions, but it would be irrational of us to dismiss a concern because we don’t like the way it complicates matters. It’s possible to both admit that the current course is likely increasing the probability of a second wave and support the current global approach as the best strategy available given the current state of the evidence.

    • DaveK says:

      You failed to mention one of the most common justifications for “flattening the curve” and the one that seems to have the most validity, in that I haven’t really seen any counter-argument against it.

      If the virus spread unchecked, hospitals will be overwhelmed, and thus many more people would die not from the virus itself, but from the lack of medical care for everything else.

  22. HeelBearCub says:

    Starting today, LA County is going to randomly test for antibodies to Covid-19 in order to estimate what the exposure and spread has been in the general population.

  23. Paul Zrimsek says:

    Sorry, Diplomat and Turkmen people 🙁

    And SSC reader Castilho describes their home country of Brazil

    While I’m in favor of plural they, this is one of those places where it can be a stumbling block, at least in its current state of novelty. I’m probably not the only reader who was momentarily astonished to learn that Brazil is the home country of the Turkmen people.

    • Le Maistre Chat says:

      The other Turkic-speakers were just as astonished when a horde of male Brazilians canoed across the Pacific, landed in Siberia, and asked to become Turk men.

  24. truckdriver20 says:

    What’s going on in South Africa? Cases were rising exponentially and the situation looked scary. On March 27th they went into lockdown. Immediately after, the number of new cases dropped about a third and has been flat ever since. It’s been linear like this for over a week.

    South Africa seems to have the highest testing capacity in Africa at about ~1000 tests per 1M residents, compared to the US’s ~6000 and Korea’s ~9000. So this shouldn’t be an artifact of insufficient data, especially because their overall number of cases is low.

    Any bets onto what this is? Statistical fluke? Corrupt officials cooking the data? Climate and/or demographics making social distancing work especially well? Any predictions for where their curve goes next?

    • matkoniecz says:

      They run out of tests and now are testing at reduced rate? Number of daily performed test would help in interpreting data.

      They switched to testing medical personnel? They switched to testing random samples of a population?

      Other testing change?

    • DaveK says:

      Random Guess- Because of Apartheid, South Africa has a weird social setup where there are already barriers in place that reflect pre-apartheid social divisions.

      If the result of lockdown was strengthening these barriers, testing may only be getting a done on an unrepresentative group of people who are less likely to get the virus.

      Secondly, these people already had strong barrier type systems in place to safeguard their relative wealth from the poor people who hate them and have extremely high levels of criminal activity from various levels of organization.

      Related to that, there are places in South Africa where paramilitary private militias, of the sort favored by anarcho-capitalists, are an actual thing, who may be able to create relative “deadzones” between rich and poor areas which would more effectively prevent the spread from the latter to the former

      The poor areas are much more dense, and we’re talking here about extremely bad third world slum conditions versus relatively wealthy “western” conditions that already have active war level protections against such slums.

      The situation is a bit more complex as there are middle class and upper middle class neighborhoods that have declined in terms of economic status and relative safety, but have over the years had their own militias to force certian people out. The old racial divisions and rivalries are still very much a thing.

      If this model is true, it’s basically that the initial data was based on a more robust sampling of the country, and the more recent data which came after the lockdowns is based on effectively reinstuting apartheid like meausures and using data from the wealthy areas that are isolated from the rest of the country by miitary means.

      Again, no actual evidence for this, just a speculation/guess informed by knowing a little about RSA post-Apartheid (in the sense of by no means being an expert or even a lay expert, just having a little more famililarity then the average american.)

    • keith2000 says:

      I’m South African. Yes, the apparent low numbers of confirmed cases, deaths, and growth rates is puzzling.

      Probably data-snooping / wishful thinking, but there have been some studies indicating that the BCG (Bacillus Calmette-Guérin) vaccine may help in some manner. BCG is a vaccination is given at birth to all South Africans to protect against tuberculosis.

      See for example

      There have been several other similar studies over the last few weeks.

      It seems to have the unintended benefit of protecting against some viral infections.
      Studies indicating BCG’s positive effects on some viral infections (not related to tuberculosis) have been around before covid-19 e.g.

      Again, probably just wishful thinking, but interesting to consider

      Private militas are not really a big thing here. I have not encountered any in my 40 years in SA.

  25. One thing i worry about with low dosage variolation proposals is that the the intrisic immune system will wipe up the infection without the body needing to developed a learned immune response. I was infected with Lyme disease once but I never developed any antibodies to it because I went on antibiotics the same day I developed the bullseye rash.

    The large majority of the pathogens we’re exposed to are stopped by our skin and other barriers. The large majority of the pathogens that make it past that are stopped by intrinsic immune responses before they get a hold. It’s only in rare cases that you get sick enough that you need to learn to make antibodies to resist a disease.

    I would be somewhat less skeptical about route of infection making a difference. In smallpox it seemed that infection via the skin was less serious than infection via the the lungs, say.

    • albatross11 says:

      That’s true for other diseases, too. Pneumonic plague is even worse than bubonic plague, and inhalation anthrax is way, way nastier than regular on-the-skin anthrax.

  26. James Miller says:

    Drive-by voting proposal: Write down your name and address and tape it to the inside of a car window along with your ID. While your window stays closed, an election worker will verify that you are eligible to vote. Then you move your car to a machine that gives you a ballot through your now opened window. You fill out the ballot in your car and then drive to the place where you put your ballot in a box. Zero human contact. Alternatively, have in-person voting take place over a month to minimize crowding thus making it safer than going to a grocery store.

    • noyann says:

      Would limit to 1 person/car to ascertain that it’s a secret vote.

    • Edward Scizorhands says:

      This proposal isn’t necessarily wrong, but a problem with spreading out voting over multiple days is that it gives people who want to fraud themselves a win indications of how hard they need to work to steal it.

      Also, Milwaukee has tried drive-up voting, and there were problems since they were the first, but it seemed to work and if people are willing to learn from them it’s probably worth a shot.

      • albatross11 says:

        A practical issue that applies most years but maybe not this year: a lot of times, polling places are put in schools. That’s workable for one-day elections, but maybe not for 2-week elections.

    • Lambert says:

      Keeping polling stations open for longer is a good idea.
      And maybe double the number of them too.
      There’s a load of empty buildings and furloughed workers around.

      • Edward Scizorhands says:

        Wisconsin ran out of poll-workers, because they didn’t want to show up. Given that poll-workers tend to be very old, I don’t blame them.

        We need to actively recruit young people to be poll-workers for November, and we need to start doing it right now. Throw money at them if you need to. Money machine go brrrr.

        • albatross11 says:

          Yes, most poll workers are retired people. Not a group that is going to want to come out and get exposed to a few hundred voters just now, and maybe half a dozen of them are coughing the whole time they’re in the polling place. I’m somewhat high risk, and I wouldn’t do that for the couple hundred bucks I think they normally get paid!

          I have no idea how to do this administratively, but getting healthy young people who don’t have anyone high-risk in their household to help out with the election would be a huge win for the upcoming election.

    • Corey says:

      There’s already curbside voting for the disabled / elderly / etc, but it would be hard to scale that up – it involves a poll worker making multiple trips to your car while being you in the polling place (stating name and address, getting a ballot, going through the address-change process if necessary, etc.)

      In NC where I work, you have to sign an affidavit that says you are unable to enter the polling place due to age or physical disability, but nobody is going to challenge you about it.

    • semioldguy says:

      What would the plan be for someone who doesn’t own a vehicle and/or doesn’t have a driver’s license?

  27. eqdw says:

    I’m surprised by this, because I would have expected mail voting, as opposed to booth voting, benefits people with good executive function who are familiar with doing things by mail – ie older, richer people, ie Republicans. It would appear that I am wrong.

    I am not here to argue, merely to inform. I don’t represent the following as true, I’m only relaying that some people think it’s true

    I have spoken to like four Republican state senators, mostly in MN and TX, and they all seem to be strongly convinced that there are ongoing Democratic party operations to file fraudulent votes on a mass scale via mail-in ballots. They seem to believe the primary mechanism of action is to use the voter registration of elderly people who have since died, and file votes by mail for those individuals.

    Is this true? I don’t know. Probably it is at the margin, just because the US is massive and so every one-in-a-million thing you can think of happens _somewhere_. But I can say that these currently elected politicians are utterly convinced that it is true. Their opposition to vote by mail is not out of stupidity, and it’s not out of some scheme to disenfranchise voters. It’s out of a sincerely-held concern for the validity of their elections.

    • DeWitt says:

      No, really, just to reiterate my earlier question: why should anyone care whether or not people hold these concerns genuinely? Can’t we check whether or not they’re valid concerns, or at least the extent to which they are true, all without checking if our opponents are lizardmen or not?

      • mitv150 says:

        We should care because it determines how we shape our own arguments. Of course, it does not / should not lend any additional weight to the actual veracity of the argument.

        Situation 1: Professed concern of Party A is genuinely held, Party B believes it is genuinely held. Party B can reasonably proceed by checking whether professed concern is valid and marshalling appropriate evidence.

        Situation 2: Professed Concern of Party A is a front for True Concern. If Party B addresses only Professed Concern, no progress will be made because proving that concern invalid will never shift Party A’s position. In this situation, it is in Party B’s interest to spend more time addressing True Concern.

        Situation 3: Professed Concern of Party A is genuine, Party B believes it is a front for True Concern. Party B, in this situation, is unlikely to behave as in Situation 1, because, if they believe the situation is more akin to situation 2, they will tailor their behavior appropriately. Thus, Party B will spend their time and energy addressing what they think is True Concern, and no progress will be made.

        In the present context, the left* spends more time denouncing the voting concerns of those on the right as being about disenfranchisement, rather than focusing on proving them false. Typically, the left* quickly states that voter fraud concerns are clearly overblown and then spends a bunch of time explaining why its all just a cover.

        If the right’s voting concerns are indeed based on attempted disenfranchisement, this may be the correct approach.

        If, however, the right is genuinely concerned about voter fraud, the correct approach is to provide evidence that voter fraud is not an issue and to work constructively on ideas that might satisfy the right without causing disenfranchisement.

        *the “left” here refers to the general theoretical monolithical left, and not to any person specifically in this discusssion.

        • DaveK says:

          That doesn’t take into account what I think is the most likely scenario. People tend to develop True Concerns about things that have actual multivariant concerns. If the actual, initial motivation is disenfranchisment motivated by pragmatic power concerns, but the expressed motivation that gets spread and becomes a True Concern is resistance to a plot by the other party, that’s a different scenario.

          It’s not that stating the unsaid concerns will help. It’s that addressing the True Concern that is expressed won’t help either because there are additional motivations that motivate said True Concern subconsciously or for complex reasons, and people will still hold onto those motivations, possibly subconsciously, and conclude the opposition arguments to adress the True Concern are insincere.

          This also underestimates the degree to which both sides distrust the other side. It is at the point where both sides true beliefs are that the other side is inherently evil, manipualtive, and untrustworthy, thus anything they say must be untrue and at best a partially true statement that conceals the fullness of their actual evil agenda.

      • Corey says:

        It can be difficult to believe people hold the concerns genuinely, because they don’t stand up to even the most cursory examination. Or even just asking “How?”

        In this case, besides the ways it would be detected after the fact, how would anyone keep an operation of the necessary scale secret? Lots of people of all political persuasions are watching. The candidate and his wife are not going to be able to forge tens of thousands of ballots on their own. If even one person on your team blows the whistle, everyone goes to jail.

        (Yes there are some elections you could swing with a small number of votes. My unpopular opinion, as with Bush v. Gore, is that those are ties, morally speaking.)

        • DaveK says:

          You are greatly underestimating the extent to which people are “mindkilled” by politics. Both sides at this point believe in entirely separately realities, neither of which correspond especially closely to the territory. One can certainly argue that one sides map is closer to the territory then the others, but it varies on different issues.

          I think you’re absolutely wrong about the ability of other people to hold genuine beliefs that seem inherently unbelievable to you.

          In another post, I used the Kavanaugh case as an example, but a common thing I used as an example was that both sides were not merely convinced of his guilt/innocence, they believed his guilt/innocence was so glaringly obvious that the people who expressed an opposite view didn’t actually hold that view, they just didn’t care.

          I saw many Republicans/Democrats express how that case was a turning point for them, where it became clear that the other side knew that Kavanaugh was guilty/innocent, but were willing to lie about what they knew to be true to advance their agenda.

          It was so extreme that even suggesting to either side that perhaps the other side actually believed what they claimed was seen as evidence as being on the other side because no reasonable person could possibly come to a different conclusion then the obvious one supported by the facts.

          • albatross11 says:

            Just as an aside (we don’t want to hijack a COVID thread with CW stuff), I noticed this phenomenon, too–I remained skeptical of the claims about Kavenaugh, while most of my coworkers seemed to be 100% convinced that every accusation was true. It was clearly hard for them to understand how anyone could remain skeptical of his guilt in light of the overwhelming certainty they had he was guilty.

    • HeelBearCub says:

      Democratic party operations to file fraudulent votes on a mass scale via mail-in ballots

      Probably it is at the margin

      These statements are in disagreement.

      • Doctor Mist says:

        Well, not precisely.

        The fraud we know about is of a scale that is highly unlikely to affect a national election. To scale that up enough to change a national election (i.e. tip it from 49/51 to 51/49, i.e. affect it at the margin) would be something we would probably all agree is “on a mass scale” — at least two orders of magnitude more than what is happening now, and probably more than that.

        The question — and I think there are good arguments in both directions — is whether universal vote-by-mail, implemented largely from scratch in a matter of a few months, would enable that level of scaling up.

        I don’t see that we have much choice but to try, but I worry.

        • HeelBearCub says:

          I completely agree that implementing vote-by-mail presents challenges, and the less time you have to do it, the more likely there will be issues. There are, of course, states who do it successfully which proves it is possible.

          But you’d really like to see a good faith attempt to start preparing now, and even see states running primaries this way (as a way to work out kinks before the general). Some states will do that, but I have a feeling we won’t see many mixed-control or Republican led states do it. And that could lead to quite bad results if they are forced into it by events at the last hour.

    • Corey says:

      Upthread I talk about how this sort of thing would be easy to detect after the fact. In this case it would be trivial – look at the public-record List of Who Voted, find dead people, lawsuit ensues. (In the gubernatorial election linked in that subthread, local media found 8 such people, though one died after election day and one was a husband/wife mixup)

      ETA: At least in NC, absentee ballots (whether early voting or by-mail) are trackable, in a way election-day ballots are not. So if someone were to do this, and it got discovered after the fact, those ballots could be found and un-counted.

      • anonymousskimmer says:

        Unlikely to be the case in those other six instances, but still possible: What it’s like to be declared dead by the government. On the other hand, given the number of wrongly declared deaths per year, perhaps those other 6 cases are living persons.

        Rivers’ plight as a falsely-categorized deceased person is not singular: it is estimated that every year, some 12,200 very much alive U.S. citizens are declared dead by the Social Security Administration due to “keystroke errors.”

    • Etoile says:

      This sort of thing would be of greatest concern in swing states/districts, where a few dozen votes can swing it one way or the other — and where the difference is too fine to prove anything. This sort of thing wouldn’t flip Texas in the presidential election (yet) but it would flip a local or House election, push a barely-holding-on opposition incumbent out, etc.

      • Simon_Jester says:

        Yes, but there already exist plenty of methods by which such elections could be flipped.

        If you can flip an election by casting a few dozen fraudulent ballots, and you somehow know this in advance (which you wouldn’t, of course)… There exist viable ways to set up election fraud on that scale.

        The reason almost nobody bothers is that even though quite a few local elections turn out to be decided on a 50.1 to 49.1 split with a margin of less than a hundred votes or whatever… You can’t predict well in advance which elections will be that close. And since election fraud is a crime that can get you sent to jail even if (arguably especially if) you lose… It’s sheer insanity to commit such a petty fraud given that he odds are very high that it won’t decide the election one way or the other.

        Nobody wants to go to prison in exchange for a 4% chance of tipping the election for county dogcatcher in favor of their party’s candidate.

  28. Edward Scizorhands says:

    The Tumblr post says that Clinton wouldn’t have done better “because up until approximately the third week of March, no one in power was taking this seriously.” But even in Trump’s White House, there were people warning about it.

    Azar tried to talk to Trump about it on January 8th, instead of vaping. On February 5th he wanted a $2 billion order of PPE.

    Navarro warned on January 29th that a serious pandemic hit would cost us $3.8 Trillion and if there were just a 1% chance of it happening it was time to spend billions now. (A once-a-century event already has a 1% chance of happening in a normal year, and a year where there is a novel virus moving through other countries pushes that up even more.) And on February 23rd he said “This is NOT a time for penny-pinching or horse trading on the Hill.” (Emphasis in original.)

    I would be pretty confident in George W Bush, too:

    • tvt35cwm says:

      This is right. Navarro tried and he was ignored. The buck stops on the president’s desk.

      All those key bureaucratic posts left unfilled for years, all the rest of it. Down firmly to one person’s whim.

      The USA needs a career (non-elected) civil service and to remove the appointed positions. And limitations on presidential powers.

      • Doctor Mist says:

        Navarro is the boy who cried wolf:

        In context, however, much of the blame shifts to Navarro himself, and to his fellow economic nationalists in the White House. Their constant reliance on “national security” and “national emergency” hysterics caused others in the Trump administration (and the country more broadly) to tune them out. As one White House adviser put it, “The January travel memo struck me as an alarmist attempt to bring attention to Peter’s anti-China agenda while presenting an artificially limited range of policy options.” In short, when the real China emergency arrived, nobody paid attention to the guy who spent the last three years running around the West Wing yelling that everything China touched was an emergency.

        Given Covid plus China’s obfuscatory, obstructionist, and threatening stances afterwards, I’m a little tempted to think Navarro was right about all of it. But I can’t find it in me to blame people for writing him off.

        (Note that The Dispatch, the source of the article cited, is very far from Trumpist — as near as I can tell they split off from National Review because the latter was too tepid about its Never-Trumpism.)

    • mtl1882 says:

      There’s no way to know how others would have responded, and whether that change would have made an actual difference. It’s certainly possible that Clinton would have done better, but my intuition is that she would not done anything meaningfully different. (I’m sure she would have acted and certainly spoken differently than Trump in some ways, of course!)

      There is a big difference between being aware of a problem and choosing to take major action to address it. A president gets notified by advisors as to a long list of concerns, and the advisors disagree among themselves, and they eventually decide what to prioritize. (Someone can express concern Al Qaeda is planning a major attack; that doesn’t mean people believe them or that other things don’t seem more pressing). Navarro brought it up, but it is probably true that most people in power were unimpressed and pushed back (I see reports that many thought he was fixated on China). I could see a very similar situation happening under Clinton, with one or a few advisors expressing great alarm, but most of the leadership class pushing back. I don’t know what would have happened. The 2016 Clinton campaign leadership was criticized for operating somewhat in a bubble and not correcting course–not sure how valid that criticism is, but that’s a very common dynamic.

      There is always a major push not to do anything that might disrupt the economy, or at least to believe that anything that threatens the status quo is overblown. I can see the argument that Trump is more sensitive to this than Clinton, but certainly Clinton wouldn’t have been insensitive to it. There was clearly strong resistance to possible overreaction for other reasons, including an aversion to seeming xenophobic or draconian in taking quarantine measures. Being warned just doesn’t equate to taking it seriously, but I have no reason to think this is a Trump-specific thing. And while there were definitely people taking seriously like Azar, it was by no means the majority view in political circles. Trump has to listen to a bunch of conflicting viewpoints and then make his choice in prioritizing. These are ultimately political decisions, and always will be. The same types of mistakes happened in other countries and in various states and agencies. Some were better than others, but it wasn’t clear cut. Most Trump-specific mistakes getting attention are “stupid remarks” that people find inappropriate but were unlikely to have made a meaningful difference in the course of the pandemic. Clinton probably would not have made those remarks. The partisan framing of the matter helped no one (cable news talking about hoaxes to take out Trump, etc.), but it’s hard to say what the culture war would look like if Clinton had won, and those dynamics were in place long before Trump made these calls.

      Every president had a responsibility to take this seriously—it seems that Bush is the only recent president who really focused on it, because he had read a book about the 1918 pandemic. All one needed was knowledge about that event, let alone SARs and MERs. Other pressures and priorities won out–the funding and political will wasn’t there across multiple administrations. As everyone is aware, Bush didn’t spend his presidency actually tackling pandemics or other domestic natural disasters, even though I believe his concern was very real. The people around him saw the Middle East as a higher priority. A lot of people seriously seemed to think pandemics couldn’t happen anymore or were extremely low risk compared to other things. I believe Bush and Obama actually had a pretty good idea of the risks because they have more realistic views of what is within human control than either Trump or Clinton, or their own advisors, but that doesn’t necessarily make them good at governing in a world where people push them in a different direction. Also, while I believe preparation and some other choices could have made a meaningful difference, I still think we’d have an uncontrolled pandemic on our hands no matter who was president. This is a very difficult situation.

      • Edward Scizorhands says:

        I don’t think any President would have stopped the virus from gaining a foothold in the US. That would require strict border inspections and even Trump’s were too little, too late.

        But when a trillion-dollar-tragedy gets its odds bumped up to a few percent chance of happening, spending and attention activates. So there would have been more attention paid to PPE production and to test generation early on. They’d never be ahead of it [1], but they would have been much less far behind. I can buy that the test screw-up was entirely random chance, not something that necessarily could have been prevented by a White House without a crystal ball. But the screw-up would have been noticed sooner and other resources brought to bear to attempt to make up for lost time.

        [1] Every President would have, pretty much by definition, underestimated the odds of an American outbreak for a long time, since we know with the benefit of hindsight that the odds were going to turn out to be 100%. I think the latest point where a competent administration could have doubted this was February 21st, when we had the confirmed outbreak in Italy. At that point, it was definitely going to hit Europe and definitely the US.

        • HeelBearCub says:

          It’s worth pointing out that Trump’s restriction of Chinese travelers was so little and late that a) it cam after all the airlines stopped flying from China anyway, and b) thousands of people came in anyway, because the came through a different country first.

    • BBA says:

      Clinton wouldn’t have done better for a number of reasons, first and foremost being that she isn’t nearly as competent as her online fan club thinks she is.

      In particular, she’d be looking at the aftermath of a Republican wave election in 2018, a Congress even more implacably hostile towards her than the current one is towards Trump, and a bunch of state governments that would not be inclined towards cooperating with the Feds. And it’s an election year and the economy is on the verge of a recession. Those are long odds for anyone.

      • anonymousskimmer says:

        A part of the reason for the Democratic wave election (which only netted the house) in 2018 was the previous control of both chambers and the white house by the Republicans (I distinctly recall reading of people voting Republican for Congress because they wanted to balance against the expected Clinton win). Had Clinton come in with Republican controlled chambers there likely wouldn’t have been a Republican wave in 2018. And its possible the Democrats would have regained control of the house (though with a smaller margin), had the Republicans been especially hostile towards her.

        With Clinton we may have had a smaller recession already, and be in a post-recession recovery.

      • Simon_Jester says:

        “Clinton wouldn’t have been able to do a better job of managing this epidemic because the Republican Party collectively would have stopped her” may be true.

        But it has interesting implications for which set of politicians the voters of the United States should continue to allow in power if they want ot enjoy luxury goods such as ‘not dying of horrible diseases.’

        Another relevant point is that President Clinton would probably not have deliberately fired or failed to appoint replacements for federal officials who have responsibility for epidemic-handling, or on coordinating on China with epidemic-related problems. Even if Clinton herself had absolutely no more managerial ability than Trump (and Trump has done shit like “sign Defense Production Act authorization and then forget to actually order anyone to produce anything,” as far as I know)… There would be a much more intact and functional system in place to deal with the virus.

        Clinton also probably wouldn’t be telling her voters things that make the CDC people responsible for using fact-based rational methods to fight the virus facepalm and tell people “no that isn’t true” so often. Because Clinton doesn’t have a habit of routinely making shit up and telling it to the public to see if she can make it stick.

    • DaveK says:

      To be fair (and this isn’t an argument defending Trump, just a general point)- the POTUS and others are probably presented with a lot of scenarios that is they happened would be disastrous.

      The reasoning “this has a very small chance of happening, but since the consequences would be enormous, we should spend resources on it” that is often used by LW folks doesn’t account for limited resources and political will that have to be divided amongst many different unlikely concerns.

      Now I don’t think this argument applies to coronavirus specifically, because the odds were significantly higher.

      It’s more a thing that people in general don’t factor.

      I remember there was a lot of criticism of Bush and the administration in regards to their unpreparedness of 9/11 in regards to the warning signs. To be clear, I am not suggesting they don’t have blame. Rather, the way it is often presented is by critics is to assemble all the various information/warnings they did have and make it seem as it was incredibly obvious. That doesn’t account for all the other information/warnings people get that don’t pan out.

      I think it definitely suggested a better threat evaluation system needed to be in place, where individual pieces of information from different sources could be correlated and risk assesments be made. And there were some reforms made along those lines, although of course improving such also threatens individual liberties.

      I think most people here should understand what I’m getting at. Learning that there was information that pointed to bad outcomes in hindsight must be balanced against the amount of information, often conflicting, about potential threats and the understanding that sufficient resources can’t be spent on all of them.

      It’s especially problematic because dividing up resources based on likelihood and risk may not even make sense, as one could argue a better strategy would be to very prepared for some scenarios/threats and be unprepared for others, rather then have a small amount of preparation for many which wouldn’t be sufficient to prevent any individual scenario.

      • mtl1882 says:

        Perfectly said. As you said, coronavirus was a clearer case than many others, but a president can’t possibly just act on every warning—a huge part of the job is trying to sort out conflicting warnings or warnings paired with dismissals of warnings, even among advisors. I tried to make a similar argument above, but didn’t think to make this point, which is very important:

        It’s especially problematic because dividing up resources based on likelihood and risk may not even make sense, as one could argue a better strategy would be to very prepared for some scenarios/threats and be unprepared for others, rather then have a small amount of preparation for many which wouldn’t be sufficient to prevent any individual scenario.

        And if you try to address everything, it quickly becomes way too hard to coordinate the efforts relative to each other. Better to have a coherent focus and an understanding of trade-offs–if you know you’re prioritizing, it’s easier to shift priorities if needed.

        I recently read a book that had a really interesting breakdown of pre-9/11 intelligence issues and the dysfunctional way we evaluate and address threats:

        • albatross11 says:

          The thing to remember is that the president basically can’t do much about a crisis like this when it’s happening. His job is in the years before the crisis arises, when he should be ensuring that the machinery for dealing with this kind of crisis is in place, and that some competent person is in charge of that machinery.

          And of course, the problem there is that crises are by definition rare. A president, governor, congressman, etc., just needs to get through his term in office without catching the blame for a crisis, whereas many measures to prepare for a crisis are going to be expensive, fodder for attacks from the other side, annoying and tedious to deal with, etc. And indeed, it’s probably pretty hard for the president or governor to know for sure whether his health apparatus is well-prepared for a pandemic without actually having one, or whether his army is well-prepared for a war without having one, or whether his state disaster relief apparatus is well-prepared for an earthquake without having one, or whatever.

          • Edward Scizorhands says:

            One big reason to always react seriously to a new disease (at the government level — you don’t need to put in locks at the populace level right away) is that you find out where you have gaps.

            There were after-action reports for the Ebola outbreak where the Federal government looked back at its own response and said “here are the gaps that would have really hurt us had this been much worse.”

        • Simon_Jester says:

          Interestingly, a lot of the things the US could have done early to prepare for this crisis are things that would NOT be overreacting and would remain useful even if it all blew over.

          For example, mobilizing US industry to mass-produce PPEs and test kits and respirators in January would potentially save lives and win the US a lot of goodwill overseas in countries affected by the virus, even if it never got to the United States directly, because we could then be the “Arsenal of Health” or whatever.

          Reaching out to local and state governments and working with them to update their antivirus plans would help ensure that in any future pandemic, local communities have good plans they can easily pull off the shelf and implement to protect themselves.

          Working to spread accurate information about the virus and why it is dangerous, as opposed to bloviating and lying about it and acting as if it literally couldn’t reach the United States, would help to ensure the public remained aware, and potentially reduce the risk of it coming to us in the first place by making sure people coming into the country are taking quarantines seriously.

          Sure, President Trump (or Clinton) would have almost certainly been out of their minds to declare a full national shutdown in January in an attempt to fight the virus at that time, effective though such a strategy might have been. But there is so very very much that could have been done, and wasn’t done, that no matter WHO failed to do it, that person really ought to be condemned for their inaction.

          • mtl1882 says:

            There are definitely things that could have been done better; I just don’t think it would make a big difference in where we are right now, which is what many people are trying to argue, or that the president is the best person to be doing them. A president can get people to focus and take things seriously, but if he or she is not willing to say things are very serious, just reaching out to local communities and issuing guidelines just-in-case often does very little, because no one pays attention. We had plenty of plans we could have used. That’s a job for other agencies that shouldn’t need much involvement from the president unless there’s a need to mobilize and raise alarm. The agencies themselves, which had the most responsibility for those things, were mostly unconcerned, and that is where the ball was mainly dropped, and all the pressures were against using the presidential pulpit to alarm people. I don’t think Trump would have been “out of his mind” to raise the issue of some sort of shut down at the time, but most people would have *thought* he was, so it wouldn’t have gone very far.

            I do think ramping up PPE production would have been a really good thing that would have made a difference, but apparently we don’t have like any machines in the U.S. that make the required material for N95 masks, if I understand this correctly. Trump could have, and I would even say should have, ordered emergency construction of such a machine and put up government funding to cover the costs, and put industry to work on it. But that would be the equivalent to declaring this a major crisis that would disrupt everything, and the whole point was no one was ready to to acknowledge this scenario. It would have prompted extreme backlash, and could not be done as a quiet back-up plan (we still seem to be having trouble getting some of the basic practical supply manufacturing stuff going, because it disrupts business as usual). It required extreme action, and also revealing how little equipment we had.

          • Edward Scizorhands says:

            The defense of Trump I got a few weeks ago was that, by not addressing the virus or advising people to do anything, he was preventing a panic.

            Now I hear that no one would have listened to him and that’s he powerless and impotent.

          • mtl1882 says:

            The defense of Trump I got a few weeks ago was that, by not addressing the virus or advising people to do anything, he was preventing a panic.

            Now I hear that no one would have listened to him and that’s he powerless and impotent.

            There’s a difference between thinking Trump did a great job and thinking he couldn’t have made a big difference. I do think Trump was trying to avoid panic, under the hope that this was addressable, and with almost everyone around him thinking the same way. I don’t think this was the right choice, and I think there was some denial involved (rather than understanding the seriousness and deciding to prioritize avoiding panic), but I think attempts to get people to take it seriously would likely have failed. That doesn’t necessarily mean he should not have done it anyway, and at least tried. I’m not saying he should have incited panic, but tried to get people to take it seriously and prepare them for actions that might happen–this would definitely have resulted in some panic, but well worth it, IMO. He had the power to break the group think, and if he understood the risk, he probably should have done it.

            But Trump’s influence is limited and easily presented as sinister or goofy. Freaking out about the virus would have been easy to portray as “crazy Trump” or xenophobic/authoritarian Trump, both by the opposition that is always waiting and by his economy/stability-concerned advisors. It would probably have been ineffective, or even backfired, for him to have personally raised an alarm if the rest of the “establishment” or leadership/media class was not on board. They would have worked overtime to dismiss him and the threat, and then it would have been even harder to reverse course and coordinate once they were convinced it *was* serious. It’s already been an awkward reversal. (Someone also pointed out that at this time, Democratic primaries were going on. It would have been so easy for even people not particularly focused on Trump to see this as an attempt to distract people from the message and turning out.) Some people would have listened, but a lot of people would not have, and the most influential people would probably have been in the latter group. There were major screw ups, but they happened largely in concert, due to long-entrenched norms in the leadership class that make me doubt the president could have made a difference. We weren’t in a good position to respond to this.

          • Clutzy says:

            Indeed, the biggest problem with the anti-Trump theory of C19 is that if Trump had been more reactive than the media generally, he would have been attacked and “discredited”.

            Trump could have done better, but every instance that he could have done better would have been him acting opposite the institutional consensus.

          • EchoChaos says:


            And note that the places where Trump was ahead of elite consensus (e.g. cutting of travel to China in January) were places that the Democrats DID criticize him.

            His reassurances were mostly repeating things that people like Fauci were saying at the time, which was “we have this under control”.

            I understand why they were saying that. We had experience stopping SARS from entering the US in a big way, so it was plausible that we could also stop COVID-19.

          • anonymousskimmer says:

            Who gives a crap if the media craps on a President? Whoever set up the antagonistic (and protagonistic wrt Fox) relationship between them, it’s the current situation, and everyone knows this.

            What matters is how the public views him. Specifically the part of the public that isn’t already strongly polarized.

          • Edward Scizorhands says:

            Who gives a crap if the media craps on a President?

            Trump does.

            If you had asked me during the 2016 election “Which candidate would do what he thinks is the best course of action despite the media” it would have been “Trump, of course.”

            But apparently the prospect of doing the right thing and made the libtard media out to be complete fools was just too much winning and he had to pass it up.

          • mtl1882 says:

            Who gives a crap if the media craps on a President? Whoever set up the antagonistic (and protagonistic wrt Fox) relationship between them, it’s the current situation, and everyone knows this.

            What matters is how the public views him. Specifically the part of the public that isn’t already strongly polarized.

            It matters because they have a huge impact on how the issue is viewed, and what gets momentum. Without any media support, such a major effort would likely have failed. Trump certainly is willing to go straight to the people sometimes and ignore the media, with some success, but this is usually more on narrative stuff, not things that require actual effort and coordination with others. And in this case, he would have had way more people after him than usual, because business/economy-focused republicans would have come after him hard. The public is unlikely to believe things are that dire based on Trump’s belief alone, even those receptive to him. They want to believe nothing is wrong, no major changes are required. If they had been concerned about this beforehand, if it had been a hot button issue where some disruption was desired in some quarters, like tariffs, that would be one thing. That had a history of support, and other experts would have supported it, even if they were in the minority. But almost no one was ready to talk about the pandemic. I’m not saying the problem was Trump being afraid of the mean media—that’s nothing new for him.

  29. J says:

    I’m predicting a surge when we get to 10-20% of population out enjoying themselves because they’ve already had it, and everyone else is sick of being at home.

    Also my brother is probably just getting over it, but nobody will test him, so he doesn’t even know if it’s safe to go out at all, much less volunteer for hazard duty. We really need antibody tests!

    • HeelBearCub says:

      Note that LA County is now rolling this out in a randomized way,

    • keaswaran says:

      My understanding is that most antibody tests have something like a 2% false positive rate. Given that the base rate for “has been infected” appears to be substantially below 10% even in the worst-hit areas, this means that antibody tests don’t give you very clear actionable information, unless you had a disease with the right symptom profile at the right time, to add some more information beyond the base rate.

      • The Nybbler says:

        Given that the base rate for “has been infected” appears to be substantially below 10% even in the worst-hit areas

        This is not a given.

  30. orthonormal says:

    Former Wisconsinite. Re: voting by mail in Wisconsin, the Republicans’ calculus there was that many poll workers in cities would stay home, and indeed Milwaukee went from 180 polling locations to just 5. For the entire, very Democratic, city. People stood in very long lines, braving the rain as well as COVID-19, to vote in Milwaukee and Green Bay. Probably not nearly as many people as would have voted otherwise.

    Voting was not as impacted in redder, less urban areas.

    Moreover, it’s not the presidential primary they cared about but the state Supreme Court election (yes, they elect Supreme Court judges to ten-year terms there)- the same state Supreme Court that issued the ruling letting the election go forward, which was upheld by the U.S. Supreme Court.

    • Polycarp says:

      The Wisconsin Supreme Court did issue an order that let the election go forward (Wisconsin Legislature v. Evers). But that ruling was not (as far as I know) upheld by the U.S. Supreme Court. The U.S. Supreme Court order was issued in a different case (Republican Nat. Com. vs. Democratic Nat. Com.), which had to do with the date by which absentee ballots would have to be postmarked. See the short blurb in SCOTUSblog.

      • Jacobethan says:


        The US Supreme Court has no jurisdiction over cases arising under Wisconsin law. The Wisconsin Supreme Court is the final arbiter of what the Wisconsin constitution defines the governor’s emergency powers to be. The two cases were utterly separate and involved legal issues of wholly different substance and scope.

        • HeelBearCub says:

          Well, it’s not quite as cut and dry as that. SCOTUS can over rule state supreme courts about their own election laws, and who has what powers, precisely when they feel like it.

          Usually in a fairly principled way. But not always.

          • Jacobethan says:

            Only if a federal question is somehow implicated. Obviously state statutes are challenged on their constitutionality all the time, and there can be questions that arise out of conflict between state and federal law. But if one thinks a state court has misapplied state law, the recourse is through the state appellate system.

            There are instances where a federal court can hear purely state-law claims, diversity jurisdiction (i.e., where the parties are in different states) being the most obvious. But those exceptions don’t apply to the case at hand. SCOTUS does not and cannot act as a roving arbiter of state constitutional law.

          • HeelBearCub says:

            There were Federal implications in the Wisconsin election in much the same way there were Federal implications in Bush V. Gore. There were primary contests for federal offices, which count just the same, if I understand United States v. Classic (1941) .

          • Jacobethan says:

            Bush v. Gore was a 14th-Amendment Equal Protection case. It wasn’t Bush claiming the Florida Supreme Court had misconstrued Florida law.

            The question addressed in the Wisconsin case was whether the governor’s emergency powers under the state constitution extend so far as to allow him to override/suspend statutes, or only to do so with regard to “administrative rules.” This is a really fundamental separation-of-powers issue in terms of the basic organization of state government, with no obvious federal jurisdictional hook.

            It’s certainly true that US law constrains the conduct of elections in many ways. Most relevant here, perhaps, is that the Constitution provides that the manner of voting in both congressional and presidential elections (since interpreted to include primaries, as you rightly point out) is to be determined by “the legislature” of each state. There is some wiggle room as to what exact entity might count as “the legislature” (e.g., an independent commission), but it’s pretty clear that a governor acting in direct contravention of the actual legislature’s wishes is definitely not it. So if the ruling in RNC v. DNC had gone the opposite way, that could conceivably have formed the basis of a GOP appeal to SCOTUS. I don’t know, though, if the issue was raised in the arguments before the Wisconsin Supreme Court; in any case, the court in its ruling chose to focus only on the state-law question.

          • HeelBearCub says:

            And if a liberal SCOTUS had granted cert on whether a pandemic constituted a threat to rights under the 14th amendment and decided that the Wisconsin governor was correct to halt the election due to this, well, who would be able to stop them?

            Like I said, usually principled. Not always.

          • HeelBearCub says:

            … and speaking of not necessarily principled.

            Republicans, meanwhile, successfully asked the Court to impose a new limit on voters — that absentee ballots had to have been postmarked by April 7 or else they would not be counted.

            As it turns out, the post office does not place any postmark at all on some mail, such as ballots that are sent via metered mail. In other cases, ballots had postmarks which merely said that the ballot was mailed at some point in April of 2020, without giving a specific date.

            One of the striking things about the Supreme Court’s order in Republican is that the majority opinion makes no mention of the standard that normally applies to stays pending appeal, nor does it acknowledge that lower court judges typically should be treated with a degree of deference in emergency stay proceedings.

            So the 5 conservatives who control SCOTUS fubarred something they probably can’t or at least won’t undo because they didn’t follow the normal principles involved.

          • Jacobethan says:

            To be clear: I don’t think the quality of legal reasoning in either the per curiam opinion or the dissent in RNC v. DNC is terribly high. I think both sides were working under extraordinary tiime pressures and produced somewhat shoddy work that looks a bit like a first draft.

            To say that “the majority opinion makes no mention of the standard that normally applies to stays pending appeal,” though, is facetious. The opinion is, for better or worse, thoroughly premised on the Court’s precedents regarding the specific sequence of events that brought about this appeal, namely eleventh-hour changes to election procedure ordered by lower federal courts. The whole dispute between the majority and the dissenters turns on the majority’s desire to follow the “normal standard” that’s been established around the distinctive considerations involved in last-minute changes to electoral procedure.

            Ginsburg’s dissent, by the way, makes a compelling case that this is not the time to be curtsying to “normal standards.” But she also manages to insert an obviously wrong account of the precedent on which the majority relies, which vitiates not the overall moral force of her position, but much of the specific legal reasoning she depends upon.

            As I said, not a case that does much credit to anybody.

  31. AnteriorMotive says:

    Isn’t there a simpler-still explanation for the preponderance of bat-linked diseases? Bats can live in huge colonies of over a million members. This is the perfect environment for the spread of disease: plenty of potential hosts for the virus to jump to and from. A new virus infecting a herd of antelope would infect all ten members, and then go extinct once it had no new hosts to jump to, now that the herd has acquired herd immunity.

    The only other mammal with social groups that large, (who we can expect to be similarly disease-ridden), is humans.

    • Yep. There’s a great Matt Ridley article in the WSJ (free version here) which looks at this:

      There are good reasons why bats spread so many viruses. Bats are long-lived mammals, like us, and live in large crowds, like us—ideal for spreading respiratory infections in particular. One bat roost in Texas houses 20 million bats at certain times of year, a concentration of mammals paralleled only by people in cities. There are lots of different species—one-quarter of all mammal species are bats—so they have lots of different viruses. And they fly, carrying diseases long distances, allowing viruses to indulge in “host-shifting” between bat species. This especially suits viruses that can “recombine” with related strains, like coronaviruses.

      • Le Maistre Chat says:

        Bats are long-lived mammals, like us, and live in large crowds, like us—ideal for spreading respiratory infections in particular. One bat roost in Texas houses 20 million bats at certain times of year, a concentration of mammals paralleled only by people in cities. There are lots of different species—one-quarter of all mammal species are bats—so they have lots of different viruses. And they fly, carrying diseases long distances, allowing viruses to indulge in “host-shifting” between bat species.

        Wow, Today I Learned.

        • keaswaran says:

          I’m not sure if this is one of the ones in the caves of hill country, or if it is the urban bat colony that lives under the Congress Ave bridge in Austin. If you’re ever able to travel to Austin again, then some time between March and October, you should definitely go downtown just before sunset to watch the bats come out. It’s an amazing sight, and a great social event too.

  32. orthonormal says:

    Re: the second derivative, any popular distillation of that happy news needs to lead with the fact that until* the daily cases decline significantly in your area (not just plateau), the risk of getting infected if you slack on your social distancing is higher than ever.

    Now isn’t just premature to declare victory to the general public, it’s the exact worst time.

    * Of course, number of confirmed cases lacks true number of cases because people take a while to get symptomatic enough to get tested, and we’re too dumb/too low on tests to do contact tracing. But still.

    • DaveK says:

      Maybe. But that has to be balanced by other psychological factors.

      People will be more likely to take better precautions if they see the situation as temporary. If the length of the time is indeterminate, many people will do a different type of risk assesment. Also, with the bombardment of bad news, people need some positive news, as anxiety levels and spirals also impact physical health.

  33. Matt M says:

    Re: The poll of economists.

    I think the question is phrased in a sneaky manner that’s designed to produce the result we see. (As is often the case with these things.) Allow me to repeat it:

    Question B: Abandoning severe lockdowns at a time when the likelihood of a resurgence in infections remains high will lead to greater total economic damage then sustaining the lockdowns to eliminate the resurgence risk.

    The italics are my own.

    I would submit that the “lockdown skeptics” (a group in which I include myself) dispute whether the italicized portion necessarily describes where we are today.

    That said, if a genie came down and told me, for certain, that abandoning the lockdowns today would lead to a resurgence of the disease and an environment of which all of the former dire predictions became true, and would require either starting the lockdowns over again from scratch, or having the entire world face NY-like waves of disease and death, then yes – I’d favor keeping the lockdowns going.

    But many of us are uncertain that describes the reality we are currently in.

    For a sports analogy, it’s like asking American football coaches something like “If your opponent was going to blitz the quarterback, should you call a screen pass (for non-American football fans, a screen pass is a play known to be very effective against a blitzing defense, but risky otherwise)?” noticing that most coaches say “Yes!” and summarizing the results of “Coaches agree that more screen passes should be called!” No – they just agree that in a specific hypothetical circumstance wherein a screen pass would be expected to be the best call, it is what should be called.

    • ltowel says:

      I sort of agree with you and sort of don’t – I think we’re so fucked that these lockdowns aren’t saving any lives, and just stretching the inevitable deaths that will happen out – hence why I think that they should end.

      “Why call for a screen on third and long when you’re gonna get strip sacked anyways”

      • DaveK says:

        “I sort of agree with you and sort of don’t – I think we’re so fucked that these lockdowns aren’t saving any lives, and just stretching the inevitable deaths that will happen out ”

        I really have hard time even guessing what your reasoning is here, as well as the reasoning the previous poster is using to guess the ending the lockdowns within a week wouldn’t make a substantial difference.

        • ltowel says:

          That’s fair – I’ve definitely been drinking too much during this (including during this post), in a way that will certainly make my reasoning incomprehensible if not worthless.

          I basically think that those who are harmed by these approaches are consistently ignored, and I’m bothered that nobody even seems to be thinking about or considering the ‘end’ of this approach when I’m an alcoholic, my grandfather has died from a heart attack because he can’t golf, the person around the corner is killed by their abusive partner, etc, etc. I feel incredibly ignored, and don’t trust the records of the people proposing eternal lockdown as a solution.

          I might be wrong, but the only people publicly espousing my fears in the ‘mainstream’ are fucking idiots – it’d be nice to see them expressed and respected.

      • Matt M says:

        I don’t disagree with anything you said.

        I think there are two plausible reasons to oppose lockdowns… one is basically “they weren’t necessary in the first place. Two is basically “they won’t do any good because we’re eventually going to need to go ahead and get herd immunity anyway.”

        If either of those scenarios are true, the lockdowns are doing more harm than good already, right now, and should be ended.

        • anonymousskimmer says:

          We never needed herd immunity for SARS-1 or MERS. The principle reason we need herd immunity for other common diseases is that people and animals never bothered to nip them in the bud (as we did smallpox and are doing to polio).

          Herd immunity is far safer when vaccine derived.

          • Matt M says:

            Herd immunity is far safer when vaccine derived.

            Yeah, obviously.

            But the question becomes, if a vaccine is 1+ years away (as almost everyone believes it is), is it more harmful to keep the entire world shutdown for a year, or to just let the disease go and have ~0.1% of the population die or whatever would happen?

            Anyone who favors “keep everything locked down until vaccine” should be required to specify what they believe the time cutoff would be before it stopped being worth it? 3 months? 1 year? 2 years? 100 years?

          • anonymousskimmer says:

            False dichotomy.

            At this point I might be more in favor of literal sanitation police than continued lockdown (at least past the first hump). As well as mandatory remote work for those jobs that can function that way, and shift work for those jobs that can function in shifts (with sanitation between shifts).

            Targeted lockdowns still seem like a good idea to me going into the future.

          • keaswaran says:

            “is it more harmful to keep the entire world shutdown for a year, or to just let the disease go and have ~0.1% of the population die or whatever would happen?”

            Let’s assume that 0.25% of the world population would die if the virus were just left unchallenged. (That would represent a 50% infection rate and 0.5% fatality rate.) Let’s assume that the average victim has 10 years left to live, and that people that get the virus and don’t die have no loss of quality of life. Let’s assume that the lockdown removes 10% of the average value of life for the time it is in place. Then a lockdown of half a year would be justified if it could prevent most of these deaths.

            If we don’t have to sustain the full lockdown until vaccine, but only until infection rates are low enough for testing and contact tracing to continue the protective work, then it looks like we’re fine.

            If you disagree with some of the numbers I used, then we may end up with a somewhat longer or shorter time horizon. (I’m most skeptical about the claim that 49.75% can get the virus and have no loss of life value – from my friend who had it, it sounds like the three weeks that she had it were far worse than three weeks of lockdown.)

    • DaveK says:

      I think your argument cut both ways.

      If we are at a point where the epidemic hasn’t even peaked, it’s techincally true that relaxing restrictions now won’t lead to a resurgence.

      But if the surveyed economists are answering that question literally, a major factor in their reasoning may be that relaxing restrictions prematurely would ultimately lead to more time spent under lockdown.

      As a very libertarian leaning person, I do hate the current situation. However I acknowledge it may be the lesser evil, as pandemic mitigation is probably what minarchists would say falls under the few limited legitimate functions of the state.

      I would much prefer to “get it over with”. In other words, I would prefer lockdowns that lst sequentially longer rather then periodic lifts that ultimately mean more time spent under lockdown.

      I understand there may be an economic/sustainability argument that the nation has to do the latter. I think the truth is however no one can be sure of the ultimate consequences of lockdown, as it’s essentially an unprecedented experiment.

      I do question the value of the surveyed economists guesses if the length of the lockdown as a factor doesn’t yield significantly different estimates.

      • albatross11 says:

        I think we have to work out how to relax parts of the lockdown sooner or later, because we can’t function for a year with like half the population unable to go to work.

    • Purplehermann says:

      I’m not sure all the economists were measuring the question from an economics point of view either

  34. ermsta says:

    I would be keenly interested in any help finding raw all-cause mortality data for the purpose of estimating excess deaths during the pandemic — even if that help is just having better google fu than I do. Most of the data I have found is aggregated by year, which is too coarse (and updated too rarely) to be useful. I have a preference for US data but am also interested in non-US data. Ideally data would be broken down by sub-national level and by cause of death but this is not necessary.

    For the US, I have found monthly data (which is still a bit too coarse), but the most recent available is 2018. The best data I have found is weekly data for the UK but it is aggregated by date of death registration and not date of death, which for various reasons causes systematic bias. However I did find something interesting in it: in late March, approximately 1.5 times as many death certificates refer to covid-19 each day as there are official covid-19 deaths.

    • zzzzort says:

      Not the most obvious source, but this page has downloadable total mortality on a weekly basis.

      But note there is a lag, so the most recent few weeks aren’t trustworthy. The ny times had a good interactive, but just for the new york area

      • ermsta says:

        Thanks for the link. Unfortunately that is specifically pneumonia and influenza, and I am looking for all-cause mortality. I might still find something to do with it if I can find nothing else.

        • Douglas Knight says:

          It has three columns: influenza, pneumonia, and total deaths.

          Also consider Euromomo, which is up-to-date (with provisional data for the last, relevant, 4 weeks) all-cause mortality, but which is heavily transformed and only available as graphics. Where is it getting its data? Probably the raw data is available from the individual countries, whereas the above CDC link might be the only window into the CDC data.

          • ermsta says:

            Ahh, thank you! (and zzzzort) I must have glanced over the “total” column thinking it was just the sum of the influenza and pneumonia columns. I hope I can make use of this US data (although I would still appreciate any other sources people find).

            The euromomo data is a good find but tantalizing how they clearly have the data but aren’t releasing it. Maybe if I ask nicely they’ll just let me have it? Hm.

            The quite high variation in deaths due to the flu from one winter to another is going to make it rather harder to create a suitable model. The strong correlation within a year but low correlation across years drastically lowers the “effective” amount of data, and will make the error bars huge.

  35. Bugmaster says:

    Regarding the economy, my prediction is that the pandemic will greatly accelerate our pre-coronavirus economic trends. I expect to see the USA well on its way on becoming a wholly owned subsidiary of China in the next six years. We’ll probably join the Belt and Road initiative in some form, extend special trade partner status to China, clamp down hard on anti-CCP sentiment in the media and social networks, etc. We might even officially declare Taiwan a Chinese province, although grand gestures like these might take longer. In any case, expect China to greatly accelerate the pace at which it buys up key pieces of American real estate and business market share.

    • matkoniecz says:

      I expect to see the USA well on its way on becoming a wholly owned subsidiary of China in the next six years.

      Are you ironic here? Because if that is serious then it seems a ridiculous predicion.

      • Bugmaster says:

        How so ? Italy had already joined the Belt and Road initiative, so it’s not an unprecedented event by any means. It looks like France will allow Huawei to build their network infrastructure. China already wields a significant amount of control over the US media, so they can swing public opinion if they play their cards right. I agree that “wholly owned subsidiary” is perhaps hyperbolic — meaning that yes, it will happen eventually, but not in six years. That’s why I added “well on its way” in front of it.

        • HeelBearCub says:

          Does “wholly owned subsidiary” ever cease to be hyperbolic? Or do you mean this literally?

          • Bugmaster says:

            Ok, granted, it’s always a little hyperbolic 🙂 But what I mean is something like Hong Kong, where the country is technically autonomous, but even more technically they have to run every major decision past the CCP.

          • matkoniecz says:

            Congo Free State and Leopold II of Belgium?

            BTW, Congo Free State is a real “gem”.


            The baskets of severed hands, set down at the feet of the European post commanders, became the symbol of the Congo Free State. … The collection of hands became an end in itself. Force Publique soldiers brought them to the stations in place of rubber; they even went out to harvest them instead of rubber … They became a sort of currency. They came to be used to make up for shortfalls in rubber quotas, to replace … the people who were demanded for the forced labour gangs; and the Force Publique soldiers were paid their bonuses on the basis of how many hands they collected.

          • HeelBearCub says:

            What set of steps do you imagine happening that lead to the US being formally under the sovereignty of China? That’s the status of Hong Kong.

            That’s why Hong Kong isn’t truly independent, because their independence is only contingent on Chinese treaty obligations to the UK, who formerly took control of Hong Kong via treaty with China.

            Or do you really just mean “highly influenced by China with some treat pay obligations”, similar to the relationship the US enjoys currently with many countries?

          • matkoniecz says:


            So you expect USA to lose independence in 6 years? Are you serious?

          • Bugmaster says:

            No, that’s why I said “well on its way”, not “completely there”. Specifically, here’s some of what I expect to see (in no particular order):

            * Media/social network censorship of any negative coverage of China (already happening to some extent)
            * The US joins Belt and Road officially, or
            * The US does not join Belt and Road officially, but begins leasing major ports and resource operations to China anyway
            * China owns controlling stock in at least 2 of the top 5 US corporations
            * US government routinely and officially praises China for their (allegedly) effective government policies
            * US government either officially recognizes Taiwan as a province of China (as I said, above, somewhat less likely), or
            * US government clamps down on any mention of Taiwan whatsoever
            * 10 or more US Senators receive the majority of their campaign contributions from China
            * Tariffs on China are revoked, favorable trade policies are enacted instead
            * Consulting with official CCP representatives becomes a routine SOP in most major corporations, especially media/entertainment (already happening to some extent)
            * College curricula adjusted to include classes on US/China relations, Chinese history, etc., presenting generally favorable coverage of the CCP
            * Some version of Chinese shoots up into the top 3 list of America’s most commonly learned foreign languages (for non-native Chinese speakers, that is)
            * Huawei or some other Chinese company provides the majority of the US telecom infrastructure (or, at least, is contracted to do so — infrastructure takes time to build)

            I could go on, but you get the gist, I think.

          • HeelBearCub says:

            You think all of that and more is going to happen in the next 6 years?

          • matkoniecz says:

            “well on its way”, not “completely there”

            OK, that still seems really unlikely but has some chance of happening.

          • Lambert says:

            > US government clamps down on any mention of Taiwan whatsoever

            Sounds like a good way to get the unwanted attention of the ‘2nd protects the 1st’ crowd.

    • DaveK says:

      Why? The general political sentiment seems to be the opposite- that this has proven we should be less dependent on China.

      I understand practically this has probably increased China’s relative economic power, but it would seem that would likely accelerate the political will to fight back.

      China is still dependent on the USD having value. A US economic collapse would also hurt China badly.

      • Bugmaster says:

        I agree that we should be less dependent on China, but I don’t see it happening. The US is headed into a major depression the likes of which have not been seen since the 30s; meanwhile, China is poised for major economic recovery and expansion — due in part to their general disregard for human lie and a penchant for slave labor.

        • eric23 says:

          Nonsense. The US economy will recover quickly once coronavirus is gone. Nearly all the people unemployed now will go right back into employment once social distancing is ended.

          China, on the other hand, has benefited massively from international trade in the last 30 years, which effectively caused wealth to diffuse from the West to poorer countries. But now that China is middle income, it can no longer grow much by this method. Also China has about 15 years left in the demographic optimum (many people of working age and few old or young) before it has a massive number of old people and fewer workers to support them. So they will be facing massive problems in not too long.

          • The Nybbler says:

            Nearly all the people unemployed now will go right back into employment once social distancing is ended.

            Doesn’t the bailout bill make unemployment pay better than employment, at least for four months?

          • albatross11 says:


            Maybe, but that’s sure not obvious to me. A whole lot of small consumer-facing businesses are going to have gone two or three months with little or no income. Many will be in arrears on rent, loan payments, taxes, health insurance premiums, etc. Many of their employees will also now be in arrears on payments, will have run up debt to buy groceries. etc. Others will have moved on to other jobs and may or may not come back.

            After the lockdowns, there will still be people reluctant to be exposed to those businesses. Eating out, going to a bar, getting your hair/nails done, getting your morning coffee, going to a pool hall or bowling alley or theater or concert–all those are things that the most susceptible people, and those who care for highly susceptible people, will be reluctant to do. Demand will fall off quite a bit.

            End the lockdowns, and those businesses are still in trouble. They’re behind on all kinds of bills, their employees are also in financial trouble, and they have fewer customers and less business than they did before the shutdown.

            Maybe that will all just work itself out seamlessly, but I sure don’t see why that’s the way to bet.

  36. atticade says:

    That last graph is really hard to read. Seems like a lot of those colors are really close to one another. Anyone have a better (high contrast, or labelled) version?

  37. belvarine says:

    Some socialists on social media are trying to spread a narrative where capitalists think the economy is more important than lives and want to lift the lockdown immediately, and it’s only socialists who are standing up for the importance of saving people. Top economists aren’t a perfect stand-in for capitalists, but it’s still pretty clear that they’re wrong.

    Indeed these socialists are mistaken. However, the socialists trying to spread a narrative where capitalists think the economy is more important than lives and want to lift the lockdown immediately appear to be correct.

    • Putting the question as “is the economy more important than lives” is either fuzzy thinking or demagoguery. If you could save one life at the cost of the economy collapsing, should you? Surely not — if only because indirect effects of the economy collapsing would include the loss of many lives. If you could save a million lives at the cost of the GNP being .01% lower, surely yes.

      • DaveK says:

        Indeed. The narrative seems to be that such people are only concerned with their person stock portfolios.

        Concerns about the economy are genuine, and and economic consequences will hurt poor and middle class people more then rich people (unless you measure in trivial ways)

        • Hoopdawg says:

          economic consequences will hurt poor and middle class people more then rich people

          This is obviously true. Also, obviously unnecessary, we live in the wealthiest societies in history which could make sure everyone survives (economically, the sick will of course still die or see their health deteriorate) the pandemic without significant harm. If swaths of people do get harmed, it’s by choice. A choice unrelated to (necessary) counter-pandemic measures, but related to who is currently in power.

          Now that I think of it, this is probably a much clearer case of “only socialists (…) standing up for the importance of saving people” than a factual discussion about the economic effects of lockdowns could ever be.

          • Clutzy says:

            This is not so obviously true because the unemployment effects will be extremely hard to separate afterwards. Some poor and middle class people will be able to regain employment, i expect many in the middle class will be able to obtain labor at a less prestigious and less well paying job quickly IF they want to. Indeed, a laid off attorney or accountant would have no trouble getting a job in amazon procurement right now. But that is lower compensation than unemployment benefits. As we saw with the 2008 extension of benefits, this clearly happened. But the long unemployment creates a loss of human capital that caused long term effects where those middle class people never got back into the job market.

            In addition some poor will likely never get back to work if we keep supplementing them, that is why we have amended the rules of welfare over and over.

  38. glorkvorn says:

    Here’s my guess for what happens next:

    Rest of April: infection/death rates decline almost everywhere. Everyone begins patting themselves on the back.
    May: Gradual end to social distancing and shut downs. It varies by city, and by individual. Some people try to warn that the virus is still there and it will come back, but there’s no way to enforce it on an entire country when people aren’t scared anymore.
    June-July: Most places open and back to business as usual, infection/death rates still low. Everyone forgets this thing ever happened.
    August-September: It grows exponentially again, and we’re right back to where we are now. Thousands of deaths per day, not nearly enough testing or ICU beds, no vaccine in sight.

    So what happens then? Do we do another 2-3 month shutdown? Do we try and shut down for a year or more, until a vaccine is developed? Do we just accept the mass deaths of a “herd immunity” strategy? Or what? I want to be optimistic because so far it has grown much less than the initial warnings, but I’m worried that we’re just delaying the inevitable. Social distancing has worked better than expected, but how long can we really keep it up?

    • truckdriver20 says:

      We’ll need to slowly open up to a state of “mostly-normal” until we get the vaccine. This could look like: masks, bans on large gatherings, voluntary social distancing urged, crowd control, mass testing and tracing concentrated in high-density areas, white collar workers continuing to work from home, bars and restaurants opening up at half capacity…. schools could even remain closed for another year if necessary and it wouldn’t blow up society, the kids would love it. Is there a reason to think these mitigation tactics won’t work or won’t be applied? Right now I think cautious optimism is very much warranted.

      • glorkvorn says:

        I feel like if we don’t enforce those measures, not enough people will actually follow them. Everybody will want to go to church, have a wedding, have a party, etc all at once. Maybe 50% of people will still avoid big gatherings, but the other 50% will still spread it.

        • truckdriver20 says:

          Yeah I think bans on gatherings of over x people should probably be enforced

          The churches kind of worry me, these were the sites of the main contagion events in Korea. I think large centralized religious organizations eg the Catholic church will be socially responsible and have people space themselves out on the pews, drink communion wine out of disposable cups, etc. but at least a few of these independent Baptist churches and megachurches and whatnot are probably going to act like retards and it will be politically unpalatable to enforce actions against them. Don’t think this is insurmountable but idk how I would deal with this if I was a public health authority, it’s tricky.

          • albatross11 says:

            I think the choir practice case demonstrates that we need to keep churches, concerts, and sporting events shut down for awhile longer–singing, shouting, yelling, cheering, etc., probably launches tiny virus-laden droplets a great distance, where they happily float around until some previously-uninfected person inhales them.

          • truckdriver20 says:

            Idk if I’m entirely convinced that the choir practice story is true but yeah if it is then it becomes an even bigger issue. Good luck getting hardcore religious people to see the logic though

          • DaveK says:

            While you hear about a lot of individual pastors and churches wanting to defy orders, those people are getting a disproporionate amount od media attention. The vast majority of churches and religious leaders support social distancing and are not holding in person services.

          • keaswaran says:

            I live in a suburban town in Texas, and the four churches within two blocks of my house have all been shut down since mid-March, with no signs of any protest. One of them is still doing their Thursday food kitchen for the homeless, but with lots of people enforcing some sort of social distancing (I’m not entirely sure how effective their methods are, but they really at least seem to be earnest and partially effective, and it suggests that they aren’t champing at the bit to get the congregation back in).

        • Edward Scizorhands says:

          There’s likely a bell curve, where 10% of people are going to do extreme social distancing even if not ordered; 10% are going to be dangerous despite orders; then a spread in-between.

          So maybe that dumbest 10% is getting infected now, despite our best efforts. (Note: being infected now doesn’t mean you are in that 10%, just that you are their victim.) So if we start relaxing rules, they would be the first to crowd the bars and beaches, but they already have gotten some herd immunity to slow the spread.

          That’s a long string of “likely”s and “maybe”s, though.

          • albatross11 says:

            This graph shows a massive fall-off in demand for tables at restaurants–note that this includes places that didn’t have lockdown orders. People everywhere suddenly thought “Geez, I don’t want to catch the plague just to go on a date with my wife tonight–guess I’ll order out or stay home instead!”

            There were surely people on the right tail of the “go do stuff” distribution who still wanted to sit in a crowded bar or restaurant, or go hang out on a crowded beach, or whatever. But not all that many.

          • Clutzy says:

            Yeah, we went to dinner the Friday before my birthday and it was dead. Probably a top 3 restaurant experience I’ve had.

          • albatross11 says:

            The last time my wife and I went out to dinner, just before the real panic started, the restaurant was separating tables by more than 6 feet and there were very few customers. It was already starting then. Most people don’t want to catch something that’s like a really bad case of the flu but then it sometimes goes into pneumonia and kills you.

    • Clutzy says:

      This is my expected result as well, and it is the worst case.

    • Le Maistre Chat says:

      We’re going to get into a very bad situation in the lead-up to Christmas. Elected leaders ordering their people not to go to church and avoid all large gatherings for Christmas? That’s asking for open rebellion.
      I know the Catholic Church and a lot of others have chosen to shut to the public past Easter, but the medical experts were advising that like only one month before Easter. Christmas more than 9 months into the lockdown and a much bigger deal to everyone but the devout core.

      • HeelBearCub says:

        If we make it to Christmas still under lockdown, I think it will (almost but not quite certainly) be because of society wide consensus. Churches will have had a lot of time to formulate how to serve the needs of their faith and community.

  39. advaitv says:

    Similar to what you’ve mentioned “Flatten the curve” can have 2 meanings : Bulldoze the curve or let it simmer

  40. hnau says:

    Warning- possibly uninformed and unhelpful take:

    In the last few weeks the media, internet, public and elite opinion, etc. have crossed over from “insufficient attention to coronavirus” to “excessive attention to coronavirus”, due to hype and herd mentality effects. At the margin, spare attention / time / money would be better devoted to other causes.

    Anyone care to change my mind on this?

    • Edward Scizorhands says:

      Like watching the plane crash into the WTC over and over again, we keep watching the news long past the point where it’s healthy for us.

      A prepper blog shared here emphasized that tuning out of the news was going to be very important to one’s mental health. They said that a few months ago, before everything became all-corona-all-the-time.

      • Corey says:

        On one hand, reading Twitter daily and threads like this helped me get my panic-buying in before the rush, and masks made before it was cool.
        On the other hand, like with politics, the day-to-day minutia is of no consequence whatsoever, and will just crank up anxiety.

    • eric23 says:

      I think the amount of media attention is in proportion to the effects on people’s lives (no leaving the house except for a few specified purposes!)

  41. fion says:

    Wait, you don’t have postal voting in the US? In the UK anybody who wants can sign up for a postal vote as long as they register for it far enough in advance.

    • hnau says:

      Voting in the US is managed by the individual states. Some states have postal voting as you describe but I’m not sure how widespread it is.

    • Corey says:

      Everyplace in the US has a procedure for “absentee” ballots, so people can vote if out of the area on election day, or homebound etc. (Houston has a procedure involving proxies but not the mail, for ISS astronauts to use).
      Time was, in most places, you had to have a reason (affirm you would be out of town, etc.) to request one.
      Nowadays, many places have “no-excuse” absentee voting, so anyone can request a mail-in ballot.
      No-excuse absentee voting is what enables early voting – at least in NC early voting is legally an absentee ballot that you happened to get, fill out and drop off in person at an early-voting site. (They’re also labeled with an ID that tracks back to the voter, unlike Election Day ballots. This enables same-day registration and voting for early voting here; if the registration doesn’t go through, the ballot can be un-counted).
      Oregon (and possibly other places) have went entirely vote-by-mail.

  42. Rana Dexsin says:

    I’m surprised by this, because I would have expected mail voting, as opposed to booth voting, benefits people with good executive function who are familiar with doing things by mail – ie older, richer people, ie Republicans. It would appear that I am wrong.

    (1) I have mainly biased experience here, nothing well-analyzed or calibrated, but that’s the opposite of my intuitive priors. I would have expected more social-order-oriented people (like Republicans) to prefer things like “getting there on time” and “showing your face” (and “showing your ID”…), whereas leftists (often Democrats) who have or want to support people who have the aforementioned poor executive function, or possibly financial tenuousness, etc. would want people to be able to dither about with the mail-in ballot at home, have the opportunity to see it lying there on their desk for several days in a row as a reminder, and not have to figure out: where the polling location is; how to get there; possibly how to pay for getting there; how to not have the entire time it’s open blanketed by work shifts at the service job someone can’t afford to lose; and so on.

    (2) Alternatively: rich people with good executive function who care about voting will manage to do it regardless of the hoops they have to jump through. Poor people with bad executive function who don’t have much energy to care about voting might only manage it if the vote comes to them rather than the other way around, so if the latter group is more often Democrats than the former, then I would expect representation of Democrats to increase when voting by mail is available.

    (3) Alternatively: is your expectation partly based on the idea that it takes more executive function to handle voting by mail than voting in person? Because I’ve usually had a pretty poor executive function myself, and that difference runs the other way around for similar activities for me. If I had to go somewhere in person to pay my electric bill every time, I’d wind up with a lot more late fees.

  43. SolipsisticUtilitarian says:

    The article on Japan’s new rise in confirmed cases does not mention the explanation that is the most obvious to me: Japan tried to suppress confirmed numbers to not endanger the Olympics. But when that failed, Japan basically said: “No point in keeping up the facade now”, and started to report more accurate numbers.

    • eric23