There's a paper from 2014 that tried to estimate the annual chance of a pandemic from a lab leak. They estimated it at 2%.
I assumed they overestimated a bit for effect and put it at around 1%.
Pandemics have historically happened somewhere around ever 100 years. What's that annual probability? 1%.
So if you knew NOTHING else, from a bayesian standpoint, if you have to differentiate a once in a 100 year spillover or a lab leak, you would put it at 50/50.
Now add in the fact that the outbreak occurred right next to a BSL lab. Now add in the fact that this lab specialized in the exact type of virus that caused the pandemic. Now add in the fact that lab workers at that lab got hospitalized with respiratory symtpoms in November. Now add in the fact that China responded with tons of secrecy, pulled down their genomic database of known viruses in their Wuhan lab, Xi issued a proclamation in February that they were revamping safety at BSL labs to prevent leaks, and that the virus in question is remarkably similar to one they found in 2013 thousands of miles away, and add the fact that there was a proposal to modify cornaviruses to have a furan cleavage site to perform gain of function research and that the COVID virus has just such a furan cleavage site, and this virus emerged remarkably well adapted to humans very very quickly. Now how about the fact that China pushed the wet market theory even after they'd figured out that probably wasn't the case? Now add in the fact that China let SARS escape from the lab TWICE in the previous decade.
How does that affect your truth value? All the facts above push the probalistic truth value toward a lab leak. There are a few facts that push it back a little the other way, but there aren't very many that I've found.
> Now add in the fact that the outbreak occurred right next to a BSL lab.
The outbreak was likely to happen in a large population center near where the bats were. The actual probability that it went north and happened in Wuhan was probably 1-in-12 or so. And the first time there was a coronavirus spillover and pandemic, it happened in Guangzhou. So we rolled 1d12 once and didn't get a 1 and then rolled it again and did. Not that mind-blowingly improbable.
Also, not surprising that the lab was in a major city somewhat central and closer to the bats than e.g. Beijing. Because that is what it was set up to study.
> Now add in the fact that lab workers at that lab got hospitalized with respiratory symtpoms in November.
This has been asserted by a story in the NYT, but never proven and denied by WIV. There's literally no evidence of this.
> and that the virus in question is remarkably similar to one they found in 2013 thousands of miles away,
Still roughly a thousand base pairs and a few decades of evolution away from SARS-CoV-2. You can't get from RaTG-13 to SARS-CoV-2 in a lab, and there's no evidence they ever had live virus RaTG-13 in the lab.
> and that the COVID virus has just such a furan cleavage site
It had a novel PRRAR furin cleavage site which had never been seen before. Not one that humans would have ever guessed. That is actually strong evidence AGAINST it being lab-made.
> Also, not surprising that the lab was in a major city somewhat central and closer to the bats than e.g. Beijing. Because that is what it was set up to study.
As we've discussed, the greatest abundance of related viruses occurred around Yunnan and Southeast Asia. Dr. Shi herself didn't expect spillover in Wuhan:
> We have done bat virus surveillance in Hubei Province for many years, but have not found that bats in Wuhan or even the wider Hubei Province carry any coronaviruses that are closely related to SARS-CoV-2. I don't think the spillover from bats to humans occurred in Wuhan or in Hubei Province.
Wuhan was "closer to the bats" only in the sense that New York City is "closer to the alligators" than Boston. There's little reason to choose that phrasing except to deliberately mislead.
I've warned repeatedly that the failure of competent scientists to engage with the real possibility that their research caused this pandemic will result in a blunt and damaging backlash. We're watching that damage now in real time.
> As we've discussed, the greatest abundance of related viruses occurred around Yunnan and Southeast Asia. Dr. Shi herself didn't expect spillover in Wuhan:
And SARS-CoV-1 occurred in Guangzhou. The closest known relative virus (WIV16) is 96% homologous to SARS-CoV-1 and was found in Yunnan as well, which is over 1,000 km away from Guangzhou. Either the range of the bats carrying these coronaviruses is much larger than anyone in the world (including Dr Shi) knows about, or else the "blast radius" of the animal trade in China is considerably larger than anyone knows about.
I think the usual theory for SARS-1 is spillover from bats to other non-human animals outside Guangzhou. The virus was then brought to Guangzhou by wildlife traffickers, like in the live infected civet cats found in markets there. A similar conduit is possible for SARS-CoV-2, but we still haven't found that proximal host.
I don't think spillover of SARS-1 from bats in Guangzhou is commonly proposed. If you've seen it (and didn't just include that option for completeness), then I'd appreciate the reference.
I agree that unexpected things sometimes happen. Nobody expected spillover in Wuhan pre-pandemic though, and the WIV absolutely wasn't situated based on any such expectation.
You're not clarify anything I said or telling me anything I don't already know.
The SARS-CoV-1 virus moved over 1000km from Yunnan to Guangzhou.
The SARS-CoV-2 virus moved over 1000km from Yunnan to Wuhan.
This is the same fucking problem. One way or another we know it has a solution that doesn't involve WIV due to the SARS-CoV-1 spillover.
If we can explain SARS-CoV-1 without WIV then we can explain SARS-CoV-2 without WIV.
> I don't think spillover of SARS-1 from bats in Guangzhou is commonly proposed.
I never suggested that was definitely what happened, and I kind of doubt it, I think the wildlife trade is more likely. At the same time, I wouldn't be surprised if the range of the viruses in bats is larger than we know right now.
> Nobody expected spillover in Wuhan pre-pandemic though
Which doesn't mean it didn't happen.
> and the WIV absolutely wasn't situated based on any such expectation.
The central location made Yunnan a lot more accessible than if WIV was in Beijing, and puts it around about the same distance from Yunnan as Guangzhou is.
I'm aware that you already know everything I've written here. I agree that spillover from bats in Wuhan is not impossible (nature is big and mysterious), but your implication that proximity to such bats affected Dr. Shi's choice of working location just isn't correct. She can be wrong about a lot of things, but she can't be wrong about her own intentions.
I guess we're just endlessly arguing the same uncertain technicalities now. I miss the days when actual new information was becoming available, and appreciated the chance to discuss with someone informed with opposing views. It would be nice to confidently learn the truth someday. Perhaps the new administration will release something, but I think it's much more likely they'll just poison the topic politically even more.
To me it barely moves the needle, specifically because as you have said we have found other remarkably similar viruses in the past.
We have known about coronaviruses for almost 100 years, and we have been studying them due to their dangers for atleast 60 years, and it has likely existed since before humans could be called human. And it has been shown to be a highly virulent numerous times in many different forms. No matter how low the chances of winning the lottery, when literally billions of people are playing it daily, it is only of matter of when, not if, it turns into something more dangerous.
Now none of that comes anywhere near proving anything, but the fact that we have had multiple coronavirus infections in the past, many of which have come dangerously close to pandemic level infections, makes a natural occurrence seem the most likely source.
On top of all that, even if it did come from a lab, why does that matter? A handful of extra lottery tickets were sold and someone won from that pool. This isn't some bioweapon modified virus purposefully bred from a farm of human subjects, you would never be able to get away with such a program in a large publicly known virology lab and we aren't knowledgeable enough in viral genetics to make something like that without testing it on farms of people. There are no markers indicating engineering it in any way that we are actually capable of. And worst case it is something a mere handful of unguided generations away from a sample that was pulled from the public already and the lab got "lucky" with a random mutation on a petri dish they were studying.
You're not saying it, but you're obviously starting with a very different bayseian prior than my 50/50.
As someone that makes forecasts for a living, I'd like to see what you're assumptions are about the base rates there.
As for how the facts produce modifications to the base rate, personally I think that a virus identified in a cave over a thousand miles away popping up in a very urban area right next to a BSL facility that specializes in researching that type of virus moves the needle toward it being more likely a lab escape. You seem to feel otherwise and I don't really agree. In particular, if you take the converse: suppose the virus wasn't known at all to man, you'd probably argue it also pushes you toward the conclusion that it was a natural spillover. (In which case, I'd be in agreement.) So I don't think that argument is really a logical one.
Your objection in the last paragraph, describing this as a "bioweapon modified virus" is really a classic strawman, and since it isn't the argument I was making I see no reason to indulge it. It is indeed a relatively ridiculous notion.
Also, just for precision, my own assessment of the truth value there is about 70% in favor of a lab-leak.
Its origin is at a huge meat market full of both hunted and farmed produce and animals both live and dead from all over the country and attended by thousands of people daily. I find that far more convincing than the fact that there was a virology center in an urban area, there are virology centers in tons of large cities and all of them hold samples of many corona viruses because they are incredibly common. The fact that the one that potentially escaped just happens to be incredibly dangerous would seem like an astronomical coincidence if it wasn't released on purpose, and there are many problems with that idea. But the fact that a market full of both live and fresh slaughtered animal products ends up being the origination point of a dangerous virus does not seem coincidental at all, just a mere matter of time.
A Bayesian prior of 50/50 seems high to me. It assumes that 50% of new disease variants come from lab leaks.
In the last few decades there have been 1-2 confirmed lab leaks per year. And they're often thing like "we found a vial of smallpox we didn't know we had" not new diseases.
Nature very capably produced colds, flus, a bunch of nasty diarrhoeal diseases, the many and varied sexually transmitted diseases, the hemorrhagic fevers, and so on. For "some new disease variant that I don't know anything about", my prior would be more like 1/99 lab leak to natural origin.
Not that it will fit a western centric ideology but there is zero mystery with people going into hospitals in Nov. It would be surprising if it weren’t so.
It’s flu season and Chinese don’t go zoom their Dr, they go and check into the hospital.
In other countries, it would be considered sociopathy to go to work with a flu, but we’re all Real Americans so anything different means… conspiracy.
I'm actually aware of the cultural differences there.
Going to the hospital is not usually the same thing as being "hospitalized" though, and it would be relevant for us to make that distinction to determine how much that tidbit pushes us one way or the other. I had originally read "hospitalized" some time ago, but i just checked the intelligence briefing and it definitely does not indicate actual hospitalization.
So that intel doesn't push us one way or the other very much. Although perhaps the absence of known lab-worker hospitalizations is an argument against the lab-leak though.
If one were either the director or a senior leader of a more-or-less covert biolab doing research that is definitely supposed not to be discovered, would you have done your job if you had not established a procedure for medical treatment of sick or infected employees using local and probably covert resources? - And likely including local isolation of infected or potentially infected people? Whether or not people from this supposed type of biolab turns up at public hospitals does not seem to indicate much.
I assumed they overestimated a bit for effect and put it at around 1%.
Pandemics have historically happened somewhere around ever 100 years. What's that annual probability? 1%.
So if you knew NOTHING else, from a bayesian standpoint, if you have to differentiate a once in a 100 year spillover or a lab leak, you would put it at 50/50.
Now add in the fact that the outbreak occurred right next to a BSL lab. Now add in the fact that this lab specialized in the exact type of virus that caused the pandemic. Now add in the fact that lab workers at that lab got hospitalized with respiratory symtpoms in November. Now add in the fact that China responded with tons of secrecy, pulled down their genomic database of known viruses in their Wuhan lab, Xi issued a proclamation in February that they were revamping safety at BSL labs to prevent leaks, and that the virus in question is remarkably similar to one they found in 2013 thousands of miles away, and add the fact that there was a proposal to modify cornaviruses to have a furan cleavage site to perform gain of function research and that the COVID virus has just such a furan cleavage site, and this virus emerged remarkably well adapted to humans very very quickly. Now how about the fact that China pushed the wet market theory even after they'd figured out that probably wasn't the case? Now add in the fact that China let SARS escape from the lab TWICE in the previous decade.
How does that affect your truth value? All the facts above push the probalistic truth value toward a lab leak. There are a few facts that push it back a little the other way, but there aren't very many that I've found.