[shadowgovt]

The point of Occam's Razor is you can't prove the things the Razor leans you to. Not in a way sufficient to remove the need to invoke the Razor. But you can say that one explanation is simpler than another (such as a pandemic virus being more closely patterned to every other pandemic in human history than to a novel mechanism that has never become a pandemic before).

I see a bright glow on the eastern horizon about 7AM and it's probably the sun coming up. It could be the first strike in a world-ending nuclear exchange. I can't prove it isn't.

... but it's probably not.

[MockObject]

I wasn't asking for a proof of the phenomenon, but a proof that the razor points to that phenomenon.

[shadowgovt]

That's going to come down to an individual observer's priors on probabilities of "pandemic virus being more closely patterned to every other pandemic in human history" vs. "pandemic introduced via a novel mechanism that has never become a pandemic before."

[MockObject]

Wasn't the lab working with coronaviruses? So maybe some of it escaped. I really don't see how that's an unnecessary multiplication of entities. Is the objection simply that, pandemics have emerged from markets, but not labs? But we know that an escape of a coronavirus could lead to a pandemic.

I see nothing needlessly complex here, and certainly not extraordinary.

[divided]

I don’t believe anyone is saying it’s needlessly complex, but Occam’s Razor isn’t “the second simplest explanation is usually the correct one.”

On one hand you have the market which housed animals that throughout its history some have had and spread different coronaviruses naturally. Every prior pandemic has been believed to have been from animal to human transmission, it’s seemingly the simplest explanation.

The lab leak theory of course assumes a lab accident. Biolab accidents are fairly rare, some years have no recorded accidents, others a handful. They’ve all been successfully contained with no more than a couple fatalities each.

It’s not that it’s needlessly complex. It’s that you have a market with animals who have coronaviruses and you have a lab that studies coronaviruses. Both are reasonable explanations, but a pandemic similar to all other prior pandemics and lots of humans being exposed to lots of potential virus carriers with no protection _is_ a simpler explanation.

Having said that, I’ve remained open minded and willing to consider both reasonable and unproven. Some pandemic will be the first from a lab leak, this may have been it. We will probably never definitively know.

[MockObject]

> It’s not that it’s needlessly complex. It’s that you have a market with animals who have coronaviruses and you have a lab that studies coronaviruses. Both are reasonable explanations, but a pandemic similar to all other prior pandemics and lots of humans being exposed to lots of potential virus carriers with no protection _is_ a simpler explanation.

The likelihood a lab accident is independent of "all other prior pandemics", because it is not adjusted by whether there have been two natural pandemics, or two million of them. Rather, it's adjusted by the safety practices at the lab.

It's still not obvious that the wet market explanation is simpler than the lab leak. I don't see Occam's Razor making a selection here.

[divided]

When you hear hooves, think horses not zebras.

All throughout history viruses have spread from close animal contact with humans. Including—and I cannot stress this enough—a novel coronavirus circa 2002 that spread from bat to intermediary host to humans in a wet market in the very same country.

In comparison, for the lab leak you’re assuming a lab accident (rare), that it wasn’t noticed and quarantined immediately (even rarer), and that the wet market still randomly became the epicenter of early cases despite at that point the transmission being human to human (rare, there’d be no reason for it to be any more significant than any other place people gather).

So when we’ve never seen a significant outbreak from a lab accident (zebras) and throughout all of human history we’ve seen viruses spread to humans from close contact with animals (horses)… yes when there is ambiguity to what the cause is, defaulting to what we’ve always seen before—including literally just 20 years ago—is simpler.

[xyzzy123]

Entities should not be multiplied without necessity.

Put another way, the simplest explanation that fits the available evidence is most likely to be correct.

At this point it seems to me that a lab origin that superspreads at a market looks rather simpler than a market origin with double zoonosis and no trace of an intermediate host.

[divided]

To be fair, there’s no trace of it in the lab either /s

Double zoonosis? The virus jumped to us and then to dogs, cats, tigers, hippos, pigs, etc. 29 known species! An unfathomable triple zoonosis that happened many, many times (thousands?) over the years. I’d say double zoonosis isn’t too uncommon.

The initial SARS presumably passed from bat to intermediate host to humans in a wet market. This exact type of event happened in China just 20 years ago.

And there isn’t “no trace” of an intermediate host. There are suspected intermediate hosts but none confirmed. Similarly SARS had a couple potential intermediate hosts, and MERS had a potential intermediate host but unanswered questions around that, too. Scientists had more confidence in those theories, but still uncertainties.

Again, I think both theories are plausible. But being a repeat to what we’ve seen before _is_ simpler. When you hear hooves, think horses not zebras.