That's funny in a way because it's literally Pinker's first argument against: "Cognitive biases inflate the subjective odds of disaster: If a scenario is imaginable, it seems probable."
"Cognitive biases inflate the subjective odds of disaster: If a scenario is imaginable, it seems probable."
Let's take a minute to remember that our brains are shaped, by millions of years of evolution, to model existential threats and act accordingly. I would say that what we deem "imaginable" is a pretty good proxy to what is "probable", especially in the long run.
There's also a dishonest sleight of hand here - when determining a course of action, it's not the probability alone you look at, but the probability multiplied by the expected harm. The probability of getting in a car accident is low, but the expected harm is high enough that the hassle of wearing a seatbelt each time you go for a drive is still worth it. If he were being consistent, Pinker would say: "why are you putting on a seatbelt? Just because you can imagine a car accident, doesn't mean it will happen today".
The purpose of the 'long bet' exercise isn't really about the money - it's about our ability to reason about the future. So talking about these things as abstract probabilities as if they are the outcomes of sporting events, not the lives of a million people, is dishonest bordering on morally repugnant.
> Let's take a minute to remember that our brains are shaped, by millions of years of evolution, to model existential threats and act accordingly. I would say that what we deem "imaginable" is a pretty good proxy to what is "probable", especially in the long run.
It is not a good proxy at all. The environment in which we lived for millions of years does not correspond well to our modern civilization. Being killed by a member of another tribe was a likely event for most of human and animal history. Today, being killed by a toaster is many orders of magnitude more likely than being killed by a terrorist. Yet many people fear terrorists, while no one fears toasters.
>Today, being killed by a toaster is many orders of magnitude more likely than being killed by a terrorist.
I don't think this is true. Googling it, it seems to originate from the claim "you should be more scared of toasters than sharks" based on the number of toaster deaths (source of this data is unclear - Fermi estimate?)
The only data I could find cites 21 deaths in the US in 2000 from "countertop cooking." By comparison, there were 211 mass shooting deaths in the US in 2019 per Northeastern University.
https://www.cpsc.gov/s3fs-public/hazard_housewares.pdf
The number of people killed in a year by toasters never deviates more than a couple of standard deviations from the mean. The number of people killed by terrorists in history could be exceeded in a single event. Likewise, at the beginning of the pandemic nobody had been killed by COVID-19 - were you making your toaster argument then too?
> The environment in which we lived for millions of years does not correspond well to our modern civilization.
Who would have thought that a brain evolved by taking in and balancing things based on interacting with a couple hundred other members of their species and maybe a 100 miles in any given direction wouldn't be very good when taking in information from billions of other people and millions of miles of territory.
It's not bullshit, and it really does apply to anything. Observe people who had only theoretical awareness of some issue (e.g. kidnappings, bad treatment of animals) who shrug off statistics, but if they read a detailed story about a particular example, they're suddenly super upset about it.
Specifically in the context of this bet, it's semi-bullshit, because we have no number of how inflated the estimate of danger is.
You can't really say "Rees must be having a cognitive bias, and his estimate of the bioterror/bioerror has to be grossly overstated". Rees is not a population. the Law of large doesn't apply. He might have been too optimistic. Even if he is overly pessimistic, maybe the inflation of risk isn't that significant. Or it is significant, but the risk is so great, that even after subtracting the inflation, it is still dangerous.
The point is, applying a statistical heuristic without supplying the actual parameters is meaningless. The heuristic itself is not bullshit. The application in this case certainly is.
Taking the bet in isolation, it would only make economic sense to make any bet at even odds if the probability of the event occurring is > 50%. If Martin Rees was honestly asserting that the probability of a "bioterror / bioerror" event in that 4-year window was >50%, I think he was clearly wrong, and falling victim to the "availability bias" that Pinker describes.
Now, this may be a situation where the sides think they're talking about different things: Pinker may be arguing that the probability is <50%, while Rees may be arguing that the probability, while low, is non-negligible. In which case, in my opinion, they'd both be right: But Rees' bet (again in isolation as a bet) isn't rational (since the expected payoff is negative), while Pinker's is.
I find both their arguments completely unpersuasive because neither provides an argument for or against the "bioerror" happening by this specific date?
Why or why not by the end of 2020? Why not by the end of 2021 or 2025 or 2050?
That seems the whole premise of the site: "Long Bets". You can't just say "some day" as given a long enough time horizon many improbable things will occur.
I may be misunderstanding, but I believe that the proposal was made much longer ago than 2017 and its original duration was closer to 20 years than 4. It became limited to 2017-2020 because the proposal was only accepted (by Pinker) in 2017, having lain unclaimed until then.
Hence Pinker's remark about 15 years of hindsight, and the fact that some of the comments below the line were posted 19 years ago.
With this in mind, I'd suggest that 2020 was picked as an arbitrary "sci-fi future" date rather than for an arbitrary term of 4 years. Still arbitrary yes, but more understandable in my view.
Let me try a different approach to explaining what I mean.
Let's say you're trying to argue why we'll run out of helium in, say, 10 years.
You could say: we have 10X quantity of helium, we use 1X per year, therefore we'll run out in 10 years.
Or you could say: we use a lot of helium. We're bound to run out of helium sometime, therefore we'll run out in 10 years.
The arguments presented in this bet are of the second sort. They do not say anything about why what they're betting on should happen in 10 years or in any other timeframe.
Sure, the bet is for the end of 2020, but in order to be persuasive you actually have to make an argument for why what you predict does or does not happen in that timeframe.
No such arguments were even attempted here. That's what's so frustrating.
No such argument is possible at the time they made the bet. If such an argument is possible, there wouldn't be a bet, because the other side would have been persuaded (or very foolish).
And whether an argument is persuasive depends on what the argument is and on the people evaluating it.
Arguments with using hard data and actual reasons for picking a certain time frame are not automatically persuasive.
Take a look at the Peak Oil arguments that were made. They had pretty hard numbers and reasons for their predictions, but plenty of people were not persuaded.
Another example is Ray Kurzweil's predictions about the singularity, which also had their reasons, statistics, etc, but plenty of people disputed them and made counterarguments.
But least you could tell why they believed their predictions would happen within that timeframe.
No such evaluation is possible with the "bioerror" predictions, however, since they didn't make arguments for why it should happen by the end of 2020, but instead just expressed their "gut feeling", as you point out.
A gut feeling is absolutely an argument. It may not be a well-reasoned, evidence-supported, or persuasive argument but it absolutely 100% is an argument.
You can think of it in terms of rates. For example, if you think some event has a constant chance of happening of 5%, you should bet for durations >13 years (for p>0.5). If your counter-party thinks the rate is instead 3%, then the duration is >22 years (for p>0.5). This means that two people with these two beliefs have a zone of possible agreement for making a bet in the time period of 13 to 22 years, where exactly it lands depends on the negotiation.
Neither is that compelling; otherwise there wouldn't be a bet.
Rees basically said, I bet this is going to happen before 2020 ends, because my gut said so.
Pinker multiplied a bunch of statistical principals without a number. I really don't understand how to reach a conclusion based on his line of reasoning.
At least the gut feeling of an expert is somewhat a reasonable point.
Often intelligent academic people can justify almost anything using their toolset of theories from different fields when they intuitively feel something is right. That intuitive feel is what really matters most anyway
This line of reasoning is seemingly missing the idea that we were prepared and informed about the epidemic before relevant groups were disbanded and it spread so widely. Also, gross incompetence with handling public policy is almost indistinguishable from malice. ie: The origin of the virus might not matter if the bioerror is considered to be the incompetent handling of the event. Given how many people are dying everyday in the US, I think we have definitely reached Rees’s bioerror definition.
That's just mangling the definition of bioerror until one side loses or wins. At that point the bet is about what the judge thinks instead of what the bet is actually about. Since the bet is now up to the judge the smartest decision would be to decide against the ambiguous argument to encourage future betters to avoid making more ambiguous bets.
> By "bioerror", I mean something which has the same effect as a terror attack, but rises from inadvertance rather than evil intent.
I don’t think it’s a stretch. In an optimistic view, the current administration could be said to have inadvertently allowed things to get so bad. In a more pessimistic view, there is malice involved.
Agreed. I think the burden of proof would be on the person alleging an "inadvertent" mishandling of COVID when there's plenty of direct and circumstantial evidence to the contrary: blatant denialism, ridicule of basic preventive measures, assertions that the virus would "disappear," withholding of crucial information early in the outbreak, the list goes on. All of these intentional steps were taken by high US government officials, including the president.
There is no evidence it is from lab, while it likely originated from bats, pangolins have very similar virus extant in the wild. The burden of proof is heavily on showing that it was released from the lab, Pinker wins by default.
The US state department has said that they have intelligence that tells them that it originated in a lab, but do not believe the release was deliberate. Now, I take that with a grain of salt, state agencies are known to sometimes say what they need to to further their agenda. But it is worth noting.
I've seen a few interviews and articles by evolutionary microbiologists who have examined some of the genetic and behavioral traits and found several that indicate that it evolved under selective pressures of being in a controlled environment like a lab. I don't have links to them, but I'm sure they can be found with an internet search, once you wade through the mountains of bullshit that you're sure to pull up when you do. Two behavioral traits I noted were the significantly lower number of cases that are thought to have been spread outdoors and the propensity of the virus to move across different tissues, a trait that is deliberately selected for in lab testing of viruses because it enables ease of testing in different scenarios.
So there is evidence is came from a lab, but so far no proof that the public is privy too.
Your [0] link mainly refutes one specific theory, of Li-Meng Yan, about intentional military engineering. It does not refute the wider range of possibilities where a lab escape was part of the path from nature to human circulation.
(When that article points out things like "Scientists have yet to find the direct parent of SARS-CoV-2 in feral beasts", or that with regard to some specific Chinese-military-catalogued viruses Yan fingered, "bat coronaviruses were circulating in wild bats and could have been discovered by anyone", it's saying things very consistent with the chance that this could be a wholly natural virus that was carried to Wuhan & then escaped into that urban environment via lab collection then escape. There were a lot of researchers carrying novel natural strains from caves back to labs!)
Your links [1] & [2] are interesting but seem to include no opinion or reasoning for or against a lab escape or lab-enhancement, either way.
I can see the articles, but personally can’t help but be a little suspicious.
The coronavirus pandemic, which started from a coronavirus in bats swapping species, started in the same city that happened to be actively researching the ability of coronaviruses from bats to swap species.
This doesn’t mean it was “non-natural genetically”, but I don’t see how we can eliminate it happening in the lab as a possibility, particularly considering if it was true there would be a large incentive to perform a cover up.
I mean personally I believe it starting in the lab is the more obvious answer applying Occam’s Razor.
I agree with you, but I don't see Occam's Razor applying here. There really is no more likely point of origin with just a cursory look at the situation. There's a bio testing lab at the point of origin, but there are also numerous examples of viruses jumping species.
Looking at the behavior of the virus and the arguments of some evolutionary geneticists (and I'm not well versed in genetics so I'm trusting their credentials here) changes things a bit of course.
My logic for Occam's Razor is that if you see someone dead on the floor it's most likely they died of a heart attack. If you see someone dead on the floor and they are next to an open bottle of poison, they most likely died of a heart attack.
Now maybe statistically it's more likely that someone just happened to have a heart attack next to an open bottle of poison, but Occam's Razor isn't to select the most likely answer, it's to select the simplest answer. The simpler answer is that he ingested poison, rather than he had a heart attack near a bottle of open poison.
If a pandemic involving a bat-virus swapping species starts in a city then it was probably just a virus jumping species. If that city happens to be doing active research on how much bat-viruses can swap species, then I think Occam's Razor can apply.
Obviously there is no objective truth to Occam's Razor, and what I think is the simplest explanation may differ from what you think is the simplest explanation.
I understand where you’re coming from. However couldn’t the even simpler explanation be that viruses frequently travel from bats to other species, this lab is in this city because there are a lot of bats with viruses there, and the thing that they are there to study ended up happening?
I agree that’s a high possibility totally, I just don’t see how we can exclude the other possibility as confidently as everyone else seems to.
Also, I think doing active research on something that can jump species and is highly contagious in a city would massively increase your risk of it jumping species in that city. That risk is increased if the lab is known for poor controls and was flagged as a risk, as the one in Wuhan was in the us cables.
Additionally if a researcher did get infected, there is a high chance that they would be asymptomatic and wouldn’t even know it was them.
The only “proof” that it wasn’t from the lab is the dna analysis, but that only proves that the research wasn’t altering the genome / that there was bio-engineering, there is still loads of research you can do without playing with the genome.
I disagree. It’s easy to prove natural origin through a phylogenic tree, and that information is impossible to protect from the public. So now tell me what publicly-available evidence would guarantee to exist to prove lab origin.
Well we know it had some natural origin, nobody is claiming the virus is synthetic. The question is whether it was inadvertently spread to humans through the actions of scientists, rather than pure serendipity.
We also know that the Wuhan lab was studying samples of novel coronavirii found in wild bats.
There is plenty of suggestive evidence the pandemic arose from a lab escape. Novel natural bat SARS-like coronaviruses have been found in caves hundred of miles from Wuhan... but are brought to Wuhan for study.
> we’ve made great progress with bat SARS-related CoVs, ID’ing >50 novel strains, sequencing spike protein genes, ID’ing ones that bind to human cells, using recombinant viruses/humanized mice to see SARS-like signs, and showing some don’t respond to MAbs, vaccines...
Lab accidents leading to investigator infection & escape are distressingly common. So why should "wet market vendor" or "wildlife trader" or "farmer" be a more presumptive index-case than "researcher modifying bat coronaviruses searching for those that are most dangerous in humanized animal models"?
Sister reply only seems to address the Yan report & malicious/military engineering - nothing about more generic lab escape.
In particular, your NatGeo link's description of constant natural discoveries, and continuing searches for natural origins, is totally consistent with the most trivial & non-malicious lab escape scenario:
(1) lab interested in this topic collects hundreds of novel natural strains that wouldn't normally cross from bats to human;
(2) lab accident infects researcher;
(3) researcher infects community.
Isn't Peter Daszak, author of the tweet I quoted, a domain expert?
We now know, from the pandemic, that a bat-related SARS-CoV2 virus that's highly transmissible in humans
actually exists, somehow.
There are human-staffed error-prone labs that specifically search for those kinds of viruses in bulk – for example, sending people deep into rarely-visited bat caves, collecting strains, then replicating (& remixing!) those strains to test them against human (and 'humanized') tissue for pathogenecity & treatment-resistance.
Given those plain facts, what makes you think some other bat-human (pangolin-human?, something-else-human?) contact should be presumed a more likely route into humans than something involving a lab researcher? Until there's more evidence, why aren't they all live hypotheses?
While I agree it's unlikely it came from the lab, as the dreadful, disgusting wet markets have been known to be a risk for those pandemics for a long time and are more likely source, you can't just dismiss it out of hand. We certainly can't trust the communist dictatorship to clear up the issue with their aversion for openness and fondness for censorship.
What if Covid was just released through bat research? Given the proximity of two bat research facilities to Wuhan, accidental release of a wild virus into human population seems much more likely than a gain of function scenario.
That'd also count as a 'bioerror'. But, what makes you think those Wuhan labs weren't doing gain-of-function research?
Or even: inadvertent gain-of-function selection, via other research methods? See the proud tweet about testing ">50 novel" "bat SARS-related CoVs" against "human cells", and using "recombinant viruses/humanized mice to see SARS-like signs, and showing some don’t respond to MAbs, vaccines" that's linked in my other comment: https://news.ycombinator.com/item?id=25370869 )
Given that, we have to think about prior probabilities: what is more likely, and requires fewer assumptions? i.e. that the virus emerged via the bountiful natural mechanisms that evolution has bestowed upon the world (and that have wrought pestilence upon humans many times), or via those channels plus a mistake in a lab in Wuhan?
It seems like a key part of the theories around the Wuhan Institute of Virology is that the virus emerged in Wuhan; in fact, we do not know where the virus emerged in humans. There's evidence that the virus was in Italy in December 2019: https://wwwnc.cdc.gov/eid/article/27/2/20-4632_article Obviously it would be a bit of a leap to link that to the Lazzaro Spallanzani National Institute for Infectious Diseases, a BSL4 facility.
It is an entirely possible theory, but it cannot and will not become the consensus view of scientists without some real evidence. The consensus view among virologists is that the virus originated in bats, where it either spread to humans directly or through an intermediate host.
None of your links identifies a "consensus on the true origin" you claimed. Instead you highlight multiple recent open questions that indicate there is no consensus.
The claim it escaped from a lab is not a contradiction with a bat-based origin; in fact, emphasizing all the research evidence tying SARS-CoV2 to bat strains already sequenced years ago in China – like RaTG13 – makes the outbreak's connection to prior research more plausible.
The November 2020 Proceedings of the National Academy of Science article, by Stanford & Veterans Administration Doctor/Professor David A Relman, in the comment to which you are replying (https://www.pnas.org/content/117/47/29246), lays out the mainstream understanding very well, and it leaves oven 3 main possibilities:
> There are several potential origin scenarios. First, SARS-CoV-2 may have evolved in bats, which are known reservoirs of immense coronavirus diversity (2), and then spread directly, or indirectly via an intermediate host, to humans through natural mechanisms. The degree of anticipated but undiscovered natural diversity clearly lends support to this scenario, as well as support to other scenarios. Second, SARS-CoV-2 or a recent ancestor virus may have been collected by humans from a bat or other animal and then brought to a laboratory where it was stored knowingly or unknowingly, propagated and perhaps manipulated genetically to understand its biological properties, and then released accidentally. …
> The third scenario, seemingly much less likely, involves laboratory manipulation or release, with the clear intention of causing harm. …
> Even though strong opinions abound, none of these scenarios can be confidently ruled in or ruled out with currently available facts.
He's one of those YouTube sleeper channels. On the surface a DIY/science tinkerer, then you pop on a livestream and he's hacking DNA on a website and creating mutant organisms.
If I'm reading correctly, the bet covered events in a period of 4 years. Given this, I think Pinker had the better side of the wager. Of course, with hindsight he was wrong, but with hindsight I'm a lottery winner.
With any starting year between 2000 and 2016, Pinker would have won. I'd guess that's the case for the next decade, too.
Rees's argument specifically mentions that it must be an error relating to biotechnology (possibly caused by "wierdos [sic] with the mindset of the people who now design computer viruses"). There is no proof that the coronavirus was created or unleashed intentionally or accidentally.
I feel like it's a wash. Sure we have a million causalities but is it really a bioerror? The problem is that the definition of bioerror is not in the bet itself. I personally think this bet should either be invalid, someone should flip a coin or the one who submitted the bet should pay because he can always word the bet in a way that favors himself. If bioerror is kept ambiguous then he would win through redefining the terms of the bet by a later date.
By that definition, Tuberculosis which kills 1.5 million every year. Doesn't quite satisfy the bet by itself but a few lopsided months could easily be 1 million in a 6 month period.
Regardless, I don't think the lack of disease prevention was the intent of bioerror. The article says that bioerror is supposed to have the same outcomes as bioterror but I believe the implication is that it is also caused by a single human or organization.
> Also you need citations for you claim that it would have been possible to stop the spread.
There's plenty of countries who stopped the spread early on. With a concerted, worldwide, coherent effort, we could have done much better and, likely, have stopped the spread months ago.
Unfortunately, this all breaks down in the face of rogue states like the US.
There are only some islands who are in the position to effectively restrict travel, and totalitarian states whose numbers can't be trusted.
Infection rates vary by countries, but a lot of it is driven by hotspot outbreaks . And some who seemed to be doing good at first are suddenly getting their hotspot outbreaks at later points.
Those were different incidents, iirc he proposed shutting down travel when news about Covid came from China.
The other "closing of the borders" is about immigration laws. I'm not from the US, but I never understood the anti-Trump position there. Surely, if you have immigration laws, you should try to uphold them. You can criticize the laws and try to change them, but criticizing attempts to uphold the laws makes no sense.
It's surprisingly difficult to find data for previous years. I think we need a bigger picture, including the data for 2021 (when it has passed).
I found one CDC number of 12% excess deaths in the US this year. It's not obvious if that is an extreme event or not. I know for some other countries there were also big events in the past decade, but before 2017. The nature of averages is that some years will below them and some years will above them.
Here is one criticism but I don't know if it has merit, I only found it by googling for data of older years and I don't have the time to dig into it right now.
It is easy enough, and a 12% increase in anything YoY is significant enough to wrry about. That number is, my I remind you, with all the measures and lockdowns put in place.
And picking some random link of a random google search confirming your opinion without having read it s propably the best example of confirmation bias I have seen in quite a while.
I specifically added that I haven't validated the numbers, how is that "confirmation bias". I googled merely for the numbers of deaths in previous years. Do you know the numbers, and are the numbers in the article wrong? Or are you using your "confirmation bias" to assume they have to be wrong?
Using their numbers, I am not sure how CDC arrives at their estimate of 300K excess deaths, either. The average 2017-2019 appears to be 2833507. With three weeks to go in 2020, according to the CDC there are 2703232 deaths so far in 2020. So using the average of 60k deaths of the previous weeks, we should be arriving at around 2900K deaths, maybe 2950k. That's 120k more than the average, not 300k excess deaths. Jumps of up to 90K more deaths in the yearly account also seem to be not uncommon (among other things because of rising population size).
Maybe there is a delay with death certificates registering, but then the precise numbers don't make sense to begin with, and we'll really have to wait a couple of months for final results.
12% big enough to worry about - worry maybe, but will it register as a significant event with hindsight (like 10 years down the road)? And the argument of it being "with lockdowns in place" is not really convincing, as other countries with fewer lockdowns have not necessarily fared worse (afaik Sweden does not have more excess deaths than in some especially flu heavy previous years, for example). Also, the claim was that we will wonder about our stupidity of not having more draconian lockdowns.
Sweden showed much more excess deaths than neighbourig countries. Sweden is also implementing stricter measures, having basically admitted that their light handed approach didn't work too well.
And looking at these graphs (https://ourworldindata.org/excess-mortality-covid) gives you a pretty clear picture. You see clear mortality spikes and, in the case of the US, an overall above average mortality rate. Just why anybody is trying to ignore that is just beyond me by now. And again, all this with measures in place.
The news about a change in "the Swedish strategy" has been greatly exaggerated in foreign media. The rules are basically the same as in April. Only main difference is 8 people instead of 50 in public gatherings. I can still arrange a private party in my own house with hundreds of participants if I wanted to.
Some of the recommendations have changed, but not by a lot. And not all recommendations have become stricter. For example: in April we were advised not to travel further than 2h from our homes, and the recommendations for 70+ were harsher than the rest of the population. These recommendations are no longer in effect.
Shopping centres are still open with unlimited capacity. Masks neither mandated or recommended. I was in a hospital a month or so ago, and saw one or two masks in total, and none among the nurses or doctors. Last time I was in a pharmacy nobody wore a mask.
It's not that there are no excess deaths in the US, the question is nevertheless if the scale is really so outrageous. And as I said, you have to look at the next year as well. It is not simply "Covid is fiction" vs "Covid will kill us all", it is the question of "how dramatic is it really". The narrative that every death in the world can be prevented is simply nonsense.
The change of policies in Sweden seems to be entirely political, not driven by the health authorities who stand by their decisions (I have a German article about it, but perhaps you can find something in English). The curve of deaths the second wave is much smaller than the first wave, nothing that really warrants such a change of course. https://www.worldometers.info/coronavirus/country/sweden/
As for excess deaths, you really have to look at it over the years, not just at single years. Comparisons with "neighboring" or "similar" countries don't automatically make sense, given that deaths are often driven by local hotspot events (like infections spreading in nursing homes).
If Reese doesn't get to collect $400 then this bet is an academic wankathon.
Ignoring clear global threats, gutting the programs that exist to respond to those threats, and piping all comm through private algos that (inadvertently) degrade basic practices of self-preservation is functionally equivalent to whatever movie plot these two thought they were betting on coming true.
Going forward-- dealing seriously with all three of these problems must decrease the risk of whatever movie plot these two thought they were betting on coming true.
Reese's definition of bioerror appears to be confined to accidental leak of a lab/home lab grown pathogen. I remain unconvinced of the allegations that SARS-CoV2 leaked from a lab.
The first sentence of Reese's position is "Biotechnology is plainly advancing rapidly, and by 2020 there will be thousands-even millions-of people with the capability to cause a catastrophic biological disaster." He also specifically states that a bioerror "rises from inadvertance", a.k.a. negligence.
The most likely explanation is that one of the many widespread contacts between wild animals and humans in the area, likely at a wet market where wild animals were being sold as food, resulted in the virus jumping to humans.
SARS-CoV2 makes a poor bioweapon. You'd want something with a higher fatality (or severe morbidity) rate, both for effectiveness and also to limit it's geographic impact to the target country.
If it's not an engineered virus, the lab isn't a smoking gun. We have identified a very similar virus where bats are the reservoir species, and have identified a couple of very plausible routes by which it jumped to an intermediate wild animal that was transported to Wuhan for human consumption. If it's a wild virus, a lab escape seems less likely than the bushmeat hypothesis.
Both HIV and genital herpes are believed to have jumped to humans due to humans/homonids eating bushmeat. It's known to be a very dangerous thing to do, and it's known to be somewhat widespread in Wuhan.
Another possibility is that it is a virus engineered for research instead of direct use as a bioweapon, and it escaped. It's close enough to the wild bat virus that it's not obvious any engineering has been done.
In any case, we have a very simple and likely explanation that matches the official story.
It's comforting to believe we're in control and that simply being more stringent about lab safety will prevent this in the future. It's also politically convenient to blame foreigners (a favorite tactic of politicians everywhere for every occasion). Those two factors are why the lab escape rumours are so persistent.
The truth is that (1) we keep having dangerous contact with wild animals and farm animals and we will not have control until we cut down on these practices and (2) U.S. officials are largely to blame for the sad state of things in the U.S. Yes, the Chinese authorities should have been more open from the beginning and botched the opportunity to snuff it out early, but there are plenty of countries doing much better than the U.S.
Chinese authorities have tried very hard to cover up the origin of the outbreak. They have refused to publish the travel history or close contacts of the 55-year old first known patient. Furthermore, they have issued a gag order to all doctors about discussing the early stages of the outbreak https://english.kyodonews.net/news/2020/11/9833532bb925-chin...
Given the above, and how I have little faith the WHO will independently investigate the origins of the virus, it's very likely there will never be a concrete answer to how SARS-CoV-2 first spread to humans.
Interesting information. This is much better than the supporting information I had seen earlier (debunked story about HIV genes being spliced in, etc., etc.)
From what I've seen, the Chinese government has a very authoritarian mindset, and assume any hint of pushback is a sign of agitators. They tend to obstruct all investigation, even when they're not trying to hide anything specific. They try and get anything even mildly embarrassing out of the news as quickly as possible by stonewalling. I was in China 2 months before the melamine milk scandal broke, and even after the story broke, they tried to suppress information, even though I'm not aware that there were ever any government connections to the scandal. They just hated the embarrassment.
How do you think a level 4 biosecurity lab ends up in a place like Wuhan? That lab is there because Wuhan happens to be a hotspot for dangerous novel tropical diseases. Regardless of the direction of causality for COVID, the original reason the lab was put there was to look for things like COVID. Your implied argument is bad. Link bombing with some random powerpoint may support direction of causality between the lab and COVID, but it does nothing to support your insuation of the lab being there in the first place.
> If Reese doesn't get to collect $400 then this bet is an academic wankathon.
You know, there is a conflict of interest. It was Reese that created the academic wankathon in the first place.
The bet is a sure win if he had predicted 1 million casualties through a virus. However that's not what he did. He added extra restrictions that decrease the probability of his bet turning out to be true. The challenger is obviously taking those restrictions into account and thereby estimates that it is easier for the challenger to win. If you now go and change the terms so that the restrictions are gone then if the predictor wins he has merely won through deception. A shallow victory that undermines longbets as a platform.
All money except the $50 filing fee goes to the winner's charity of choice. But I disagree, complacency or incompetence over a long period of time is not the same as a deliberate attack or a single error.
If the coronavirus did originate in a lab, then Reese wins.
This makes me wonder. The ability to explore the origins of COVID can easily outlast its impact. If someone finds evidence of a bioterror attack then his bet might turn out to be true well after 2020 but since there was no date given for this special case one would assume that it should be widely known by the end of 2020.
> Duration 4 years (02017-02020) ... a single event within a six month period starting no later than Dec 31 02020
Am I the only one who finds the terms above terribly ill-defined?
Let's suppose an event happens on Dec 31 2020, and a million casualties accumulate over the first few months of 2021. Does that count? It counts because it's within a six month period starting Dec 31 2020. It doesn't count because the duration of the bet was 2017 to 2020, and therefore wouldn't include 2021. Wouldn't have been less confusing to state the terms as "within a six month period ending no later than Dec 31 02020" (if that's what was intended)?
What if the event happened on Jan 15 2021 followed by a million casualties. Does that count? Jan 15 2021 is within "a six month period starting no later than Dec 31 02020". If that's the case, the duration of the bet is 2017 to Jun 30 2021.
If the analysis in Firstenberg's "The Invisible Rainbow" book is correct, then I would suggest that it has already happened and almost nobody is aware. And affecting more than just humans.
Perhaps Rees should read the book if he wishes to make good on the bet.
Looks like we may have a surprise winner, though I doubt we'll know for sure, for a while. We've certainly cleared the 1M deaths marker.
There is a decent chance that COVID was an accidental release from the Chinese research lab, just as SARS was released accidentally from a Chinese research lab in Beijing - on two separate occasions. These releases were confirmed by the PRC. [1]
To be clear there's not specific evidence or an admission in the case of COVID, just saying there's clearly established, documented precedent.
("By ‘bioerror’, I mean something which has the same effect as a terror attack, but rises from inadvertance rather than evil intent." -- an accidental release would of course fit this bill).
The SARS release was AFTER it'd spread worldwide and I'm guessing was from a lab studying the already spreading disease. I'm sure plenty of labs over the last 6 months have released COVID as well but it doesn't much matter since it's already widespread.
With that said, the Wuhan lab had both BSL-2 and BSL-3 units studying bat coronaviruses before the pandemic [1]. After the COVID pandemic hit the PRC mandated all COVID research be carried out in BSL-3+ conditions; the research they're doing at the moment on COVID happens to be done in their BSL-4 facility but only because of lack of sufficient resources in their BSL-3 facility; that's likely temporary [2].
I'm not sure why a secondary leak after a pandemic already began would be just fine though.
> With that said, the Wuhan lab had both BSL-2 and BSL-3 units studying bat coronaviruses before the pandemic [1].
That's interesting indeed. Thanks.
> I mean, none of them should leak.
In theory, there's no difference between theory and practice. In practice there is.
BSL-3 shouldn't leak, yet we don't put super scary viruses in there, because we know the probability it could happen isn't that low.
> I'm not sure why a secondary leak after a pandemic already began would be just fine though.
Fine isn't the word. It's still an industrial accident, but not necessarily a catastrophe. Do we even have an example of such leak that would have triggered an epidemic in the vicinity of the lab?
That's not easy to say. "Leaking" can be as simple as a researcher catching COVID outside the lab and then spreading it to a coworker who then spreads it outside the lab.
I nominate CDC's bioerror: allowing a test reagent to become contaminated and then distributing that contaminated reagent as part of its first batch COVID-19 test kits[1].
My recollection is that this delayed widespread availability of COVID-19 testing in the US for something like a month during the crucial early stages of the pandemic; that may well cause a million excess casualties from the pandemic -- especially if you count economic casualties (lost jobs). The FDA's regulartory approach (being slow to license tests and quick to crack down on unauthorized testing) contributed to this bioerror.
That's a good point. The arguments based around complacency can't count because each countries' response is different. There is no global preparation against the virus so all the additional deaths that can be attributed to local lockdowns not being sufficient cannot be attributed to a global cause. Each of them would be attributed to the individual countries' politicians.
10000 politicians each causing 100 deaths is not the same as an incompetent or malicious researcher leaking the virus on his own.
There is essentially zero evidence that SARS-CoV2 was engineered in a lab. In any case, the evidence that it's natural born doesn't have to be believed by everyone, only by Rees and Pinker. Then again, 2020 still has 21 days left to fuck us one more time.
"released from a lab" and "engineered in a lab" are two different things. Leaking a sample you took from some cave full of bats is one but not the other.
The standard for the bet here is "bioerror" not "bioengineered". I think both of these cases would be covered under "rises from inadvertance rather than evil intent".
Taking "engineered in a lab" to mean modern genetic technology, there are possibilities beyond that and all-natural. Remember that we modified wild polio to create a vaccine many decades ago, long before modern genetic technology. Passing bat viruses through human cell lines would not leave any kind of mark to indicate intentional modification.
True, but for the purposes of the bet, you'd have to effectively _prove_ that it escaped from a lab, which seems virtually impossible. In any case, the linked articles talk about that possibility too.
Then again one could make the argument that incompetence in the world's collective response to the pandemic is "bioerror" although again, the debate and terminology only really have to be resolved by the two bettors.
The incontrovertible facts are that deliberate and/or grossly negligent actions caused COVID to spread as far as it has. That’s a step beyond “a mere lack of preparedness”, so it’s not clear what the outcome should be.
There is a huge outdoors lab where we run a gigantic experiment "what happens when humans encroach on natural habitats of mammals on a never seen before scale and humans also are highly mobile, able to spread novel diseases across the globe before they even get discovered or their severity assessed?".
I'd qualify that statement. There is zero hard forensic evidence that it existed in a lab somewhere before the uncontrolled outbreak. If such evidence did exist, the PRC would never let it see the light of day anyway. There is, however, plenty of circumstantial evidence.
Remember that guy who said we should treat the Hunter Biden story as if it was a foreign disinfo op, even if it probably isn't? Yeah, that was crazy, but I actually see similar logic here. If the virus was engineered, that's really really bad for the world. It's like Iran suddenly making a nuke under everyone's nose... except if Iran had the largest population and largest economy in the world. It's so insanely important that even a tiny chance that it is true--like 0.1%--is enough to warrant strong diplomatic pressure on China to be totally transparent about the early days of the virus.
As someone else pointed out, it doesn't have to be engineered--China has already shown that it has a cache of previously unrevealed viral samples, so it could also just be a natural virus found on one of Dr. Shi's hunting expeditions that got loose from the lab.
What is sufficient to judge a policy error? Would an ineffective but rationally justified policy count? Given limited information, there can be any number of reasonable paths to choose from. Should it only count as an "error" if it's incorrectness was (or should have been) recognized at the time?
There is no global policy. Each local policy is only responsible for a fraction of the deaths. The originator of the virus can be a single person and thereby satisfy the criteria.
It's impossible to prove the negative, and if it's the positive the people responsible both have all the evidence in their hands and plenty of reason to destroy it.
Yes, if it happened then the hard forensic evidence is gone forever. The only remaining evidence would be witness testimony, and if anyone even remotely credible ever says a word they will be destroyed. So... yeah. The world will never get a solid answer on this one, sadly.
It has been more than a year since SARS-CoV-2 began spreading in China and the WHO is still too busy cowering to state that it originated in China. That's the first pitch of the first inning and they can't even do that. It's absolutely a safe bet the zero-credibility authorities will never provide an answer as to the origination of the virus. Their entire obvious goal is to never - not under any circumstances - blame China. They still dodge and evade that at all costs. They've collectively humiliated themselves and should be mocked frequently for it, for a long time. Every other sane person already knows where it came from: Wuhan, and likely inadvertantly escaped from their biolab after having been sampled from bats. It's the only credible, rational explanation given what is known (including about their sampling program and horrific lab security); every other explanation requires farcical, laughable stretches to go far out of the way to avoid blaming China (it was India; it was imported via frozen food; it was the US army; etc).
The job of the WHO is not to hold member states accountable during a pandemic, it's to provide support, services and intergovernmental collaboration. To share learnings. To provide their expertise to those involved. It's to provide health care to the poor and to share understanding.
They're not "cowering" to a country that provides them a teensy tiny fraction of their operation budget. Not to mention, they're not the "Health Police," they're doctors and epidemiologists.
The WHO has a ton of credibility. They've only been mocked by people spreading misinformation guided by their personal political beliefs. They certainly aren't the ones humiliating themselves.
If you want someone to hold China accountable for... honestly, there's no clear evidence the state of China did literally anything right now... take it up with the UN Security Council. You don't get mad at your doctor for the flu existing, and you don't get mad at the WHO for doing their jobs.
It might, or it might not. Nobody's saying the WIV created this virus de novo, they studied it in the lab. That meant they of course found it out in the wild, and collected samples. There existing cases in the wild doesn't preclude the involvement of the WIV. Although, again, there's no evidence the WIV is involved.
It seems almost as though Pinker was confused about what he was betting on. Most of his arguments are against deliberate bioterror attacks, not accidents.
I think it's generally not worth paying attention to renowned experts opining about stuff outside their field.
If an epidemiologist and a biochemist had different views on this question, and posted their reasoning, that would be interesting. What does Jennifer Doudna think of the probability of catastrophic bioerror? (I hope that's in an interview somewhere.) Or perhaps there's a retired national security expert who could weigh in on which groups are most likely to attempt that sort of bioterrorism.
But these two are only in a position to make uninteresting and unconvincing arguments on behalf of their respective positions.
> I think it's generally not worth paying attention to renowned experts opining about stuff outside their field.
I disagree. It is actually sometimes quite illuminating listening to an expert using heuristics outside of his field of expertise. As a mental excercise.
Ya, and to add to that, it's often difficult to define what exactly the 'field' is. Is the field biology? Is it statistics? Risk? The anthropology of terrorism? Answering a complex question like "will there be a bioterror event that kills 1 million people" doesn't really fit neatly into any of these categories, because it combines information from a number of fields to synthesize the final answer.
This is not a bet on biochemistry or epidemiology solely, but rather a bet on the likelihood of a tail event happening on a complex intertwined system of health, humans, society, mobility and many other issues.
I don't think this kind of bet/prediction really gives a big advantage to experts. And that includes any long prediction in any field, not just biochemistry or epidemiology.
I would put more faith in the predictions of a statistician or actuary (possibly with a expert in consult).
I'm curious about that perspective. Isn't it really hard for statisticians to give probabilities to things that have never happened before? Especially in the presence of important non stationarity ("the technology is proliferating")? Like, one can construct a lot of models which give non-zero probability, but the available data gives you very little ability to distinguish between them.
He might be, but this comment provides no argument as to whether he is, nor does it say anything about the subject matter.
Not to mention that even a "jackass and a jerk" can be absolutely correct, since those accusations are mostly about behavioral and/or moral failings, not intellectual.
He provided defence for Jeffrey Epstein at his trial for being a pedophile rapist pimp, and then remained friends with him after he was convicted of that.
There are many more points than that that could be made, but it does not seem necessary.
If they are similar this point, yes, they're not necessary, as they are totally irrelevant to the longbet in discussion, or to Pinker's work in general.
If Einstein was a serial killer on the side, relativity would still be a great theory...
There are millions of people who can provide at least as much insight as Pinker can ever manage. Let's listen to them instead of the creep who hangs out with rapist pimps.
>There are millions of people who can provide at least as much insight as Pinker can ever manage
That takes for granted what it should instead prove. Pinker, regardless of who he fraternizes with or what moral causes he supports or not, is an acknowledged scientist.
>Let's listen to them instead of the creep who hangs out with rapist pimps.
Who people hang with is irrelevant. We don't and shouldn't change who we listen to on technical/scientific/etc matters based on their personal morality (even less so based on someone's idea about that). We change who we listen to on those matters based on the quality of their arguments and domain knowledge.
In any case, we should not bring moralizing and posturing into discussions and into who we listen to, except if the morals in question are related to the subject matter.
E.g. I would be suspect to economic arguments from an economist who has been found to take money from states to push some agenda. But I wouldn't be suspect to economic arguments from someones who enjoys killing kittens. I just wouldn't take environmental / pet advice from that person...
I think he's a bit of a prick too.. but seriously, this isn't how HN works. If you're not a linguist or in stats then the critique is at best third hand, and basically common knowledge sour grapes. (All of the ego scientist stuff makes me mad, but I try to keep away from that)
Guilt by association is really toxic. Please don't open this door. Epstein had a little black book with thousands of names and you're basically starting "the crucible" because you want to win an argument.
I don't like Pinker as a public intellectual and not because of the company he kept, because of how sloppy he is with numbers, and an extremely argumentative egotistical posture. Some of his ideas are interesting.
If you're going to start bone pointing to evil, do it somewhere else.
If my understanding is correct, unless you are a journalist, you're basically saying "this comment is not worth paying attention to", which feels paradoxical to me.
I generally think Pinker had the better argument, but oof this turned out to be highly country-specific:
----
>>> Multiply further by the chance that the Black Hat will defeat the White Hats, namely the world's medical and epidemiological networks, who would combat an outbreak with containment, vaccines, antibiotics, etc.
The product is not 0, but I don't think it's high.
I try to understand "intellectual curiosity" here but I just find it very off-putting to bet on whether millions will die, no matter how sound the arguments of both sides might be.
The winner of the bet does not profit personally, so it's not really a bet. They use the term "bet" because it's easy for people to understand. And it makes it more interesting than alternatives like "prediction & rebuttal".
Long Bets and the Long Now Foundation try to get people thinking about the future so we can make it better.
Maybe the quick and wide rollout of the covid vaccine is the bioerror - say the mrna translation products causing some kind of amyloid-buildup. One year after vaccination people will be showing signs of dementia. Since the elderly get vaccinated first, it isn't noticed that fast.
For precedent of "new drugs have tragic effects you only notice in a ~year" see thalidomide.
To me it seems more like we developed, and approved, more vaccines in a shorter time period than ever before. So your argument would be much stronger if you provided sources and more details on which vaccines have been blocked by the FDA.
2040 is a better bet and it shouldn't be low. Enough time to mature biotech for novel biologics outside of nature and the technological means(method + access) to pose risk. Ugh pinker is such a pseud in the ivory tower.
To would be downvoters, list contributions of pinker that's novel and long lasting? Until then I'll continue to view him as another overrated commercial intellectual.
The US frequently has it's own lab incidents including even with anthrax. China must be having lab incidents as well. There isn't a smoking gun to indicate either that this is a lab outbreak or natural. However, in the absence of evidence we need to ask ourselves which is more likely:
1) A disease traveled hundreds of miles to a large city without leaving a trace of evidence
2) A lab had yet another outbreak
Almost all of the outbreak attention has been focused on the Wuhan lab that received international grants [1]. However, that lab is miles away from the initial outbreak and there's another lab that's a block away from the Wuhan seafood market that also does bat research (but not gain of function). This lab is also adjacent to the hospital that dealt with initial outbreak. This was pointed out by Chinese researchers in a preprint that was quickly taken down [2].
>>>1) A disease traveled hundreds of miles to a large city without leaving a trace of evidence
This doesn't strike me as intrinsically unlikely. Bats can apparently fly ~50 miles for food, a person on a bus can easily travel ~400 miles in a day, etc. I don't disagree with the premise that a lab outbreak is possible, but so is "guy got infected by bats or something, didn't realize, hopped on a bus to visit family/find work".
And that bus happened to be traveling to a city which hosts a top-tier biolab with the world's foremost experts on bat-borne coronaviruses. Not a smoking gun, of course, but anyways I think your logical fallacy here is the "reasonableness" heuristic, in which you base the likelihood of some incredibly unlikely event on whether you can "imagine" some scenario where it happened. The fact that you can come up with a story to explain... well, I don't even know exactly what you're explaining, because it's really just a fragment of a chain of causation which has been mostly left unsaid. But anyways, I'm willing to wager that you know very little about disease transmission patterns so your intuition here is worse than ignorance.
The OP specifically said "in the absence of evidence we need to ask ourselves which is more likely..." - and suggested (strongly) that a lab outbreak is much more likely than the alternatives. I was trying to point out that a) once the virus is in humans it's easily spread and b) though the lab's research may have led to the pandemic, it doesn't necessarily follow that there was a "lab outbreak".
Here's my understanding on what (probably?) happened:
1) Bat(s) carrying the novel coronavirus were taken from Yunnan or Zhejiang province.
2) They were brought to the lab in Wuhan (WHCDC) (~500+ miles away).
3) They were experimented upon, dissected for tissue samples, etc.
4) Some/all of these samples were disposed of.
And some time later, the Huanan Seafood Market (near this lab) was likely an origin point of the epidemic.
Now, the actual disease release could happen at any of the 4 steps above. Someone who captured the bats could have been infected at that time. Someone could have been exposed during transport of the bats. Someone could have been infected during experimentation. The materials could have been improperly disposed of, tainting food/water/clothes/whatever. Would we consider all of those "lab outbreaks"?
Also, I want to point out - "a city which hosts a top-tier biolab with the world's foremost experts on bat-borne coronaviruses" - we may also expect such a biolab to be located in an area where bats, and bat-borne viruses, are more likely to be, no? Where else would be a more logical place to put it? :)
But is it more likely for the virus to have undergone sufficient natural mutation from its closest known relative* (RaTG13) in order for it to have become both capable of and very effective at infecting humans (ACE2 receptors) and for it to have then been so carelessly disposed of from the BSL-4 Wuhan biolab that someone came into contact with it and became infected, or is it more likely for it to have been the product of the Gain-of-Function research on bat coronaviruses in that lab which the NIH began funding in 2014 due to the US-wide ban on Gain-of-Function research that lasted until 2017? [0]
Consider the fact that lab safety protocols are designed to contain pathogens, which creates very poor conditions for viruses to naturally evolve and transfer zoonotically, whereas Gain-of-Function research has the express aim of creating ideal conditions for evolution which is both directed and can proceed many orders of magnitude faster.
*Also, there's plenty of doubt as to whether RaTG13[1] actually exists; we just have the RNA sequence and we're currently just taking Beijing's word for that.
> 1) A disease traveled hundreds of miles to a large city without leaving a trace of evidence
It was certainly circulating in, say, California, for a couple months before it was actually detected [0]. During flu season, seems to me a new respiratory disease can probably get pretty far before anyone notices it's not the flu. It probably crossed the Pacific before it was detected and reported in China, what's a few hundred miles?
All you have to go on are symptoms, and mild/moderate covid doesn't look like anything special, it's not Ebola.
Neither of the people claiming that seemed to me be prevaricators/hoaxers/people-with-an-axe-to-grind. And, that sort of discovery would be a plausible outcome of a review of all autopsies back to December that California ordered in April – https://talkingpointsmemo.com/news/newsom-orders-all-ca-coun... – on which I've seen little followup reporting. But, I've seen no other confirmation of these UCSF rumors in official statements.
So: still wondering what hard evidence, if any dates COVID-in-the-US to December 2019 or earlier. (There's a survey of December blood donors which found 2% immune reactivity - but the imprecision of that test, especially given known cross-reactivities with other 'common cold' coronaviruses has caused some experts to discount it as definitive proof that COVID was circulating then.)
If you push the first death to Feb 6, then they were probably infected mid-January, and without any known links to China, it seems much more likely than not that there were a cases that percolated around California in December-January before it caused that death:
> “What it means is we had coronavirus circulating in the community much earlier than we had documented and much earlier than we had thought,” said Dr. Sara Cody, Santa Clara County’s public health officer. “Those deaths probably represent many, many more infections. And so there had to be chains of transmission that go back much earlier.”
We're agreed on what's probable, but I'd love to find reliably-reported evidentiary confirmation, especially about magnitude of pre-February-2020 or even 2019 US circulation!
> Coincidences can and do occur, even wildly unlikely ones.
You could gather some data and do some math on "Given that an outbreak occurred in humans, what is the expected range of the nearest bio research lab?" If the Wuhan lab(s) is a lot closer to the initial outbreak than you'd expect, then that alone merits a closer look.
I have no idea about the distribution of bio research labs, but seems like it's probably relatively easy to come up with that number.
Sadly, it looks like that the Chinese won't let that happen until it's impossible to figure out, either because they have something to hide or because it's just their usual modus operandi.
Rees's argument was short. It was one plausible scenario.
I often find the latter more compelling than the former.