[Zuider]

OP's question appears to invert the burden of proof in assuming that a consensus, even of experts, ought to be believed unless there are compelling reasons not to accept it.

Where is the evidence that a consensus is a good metric for the truth of a matter? Has there been any work done to establish such a correspondence?

Asimov seems to be arguing that there are degrees of wrongness, and one can become less wrong by a process of rational error elimination and criticism. How does the mere fact of a consensus have any bearing on this process?

[silverspoonin]

> OP's question appears to invert the burden of proof in assuming that a consensus, even of experts, ought to be believed unless there are compelling reasons not to accept it.

Again, I pose the question, what is a reasonable alternative?

>Asimov seems to be arguing that there are degrees of wrongness, and one can become less wrong by a process of rational error elimination and criticism. How does the mere fact of a consensus have any bearing on this process?

The scientific process IS a process of rational error elimination and criticism. Scientific consensus is the result of the majority of experts in a field using the scientific process to agree on the "least wrong" information possible given the data available at that time. If you claim not to see the bearing on this process, you're being dishonestly obtuse.

> Where is the evidence that a consensus is a good metric for the truth of a matter? Has there been any work done to establish such a correspondence?

I think you have the causal chain backwards; the evidence of a good metric for the truth of a matter is what builds consensus. If the evidence is incompatible with consensus, and it holds up to scrutiny & reproduction, then the consensus changes to match. Either way, the end result is that consensus is an improving approximation to "the truth". If you have a better way of obtaining "the truth", I would be very glad to learn about it.

[Zuider]

>Again, I pose the question, what is a reasonable alternative?

Again, this inverts the burden of proof. There is no obligation to provide any alternative, and no obligation to defer to a consensus in the absence of one.

>Scientific consensus is the result of the majority of experts in a field using the scientific process to agree on the "least wrong" information possible given the data available at that time. If you claim not to see the bearing on this process, you're being dishonestly obtuse.

I see no evidence that this narrative is generally true, and it is a very weak argument for taking consensus as a metric of truthfulness.

>If you have a better way of obtaining "the truth", I would be very glad to learn about it.

One does not require a better way of obtaining the truth in order to avoid committing oneself to accepting unjustified claims. Again, such a requirement inverts the burden of proof.

[silverspoonin]

> There is no obligation to provide any alternative, and no obligation to defer to a consensus in the absence of one. ... One does not require a better way of obtaining the truth in order to avoid committing oneself to accepting unjustified claims.

In order to act on a claim (inaction is also an action; "avoid committing oneself to accepting" is just ornate wording for choosing inaction) one must evaluate the "probability of truth" of said claim. To my knowledge, scientific consensus is the most reliable source to inform that evaluation. Since you fail to provide a more effective alternative, then it seems that scientific consensus is at least non-inferior to any other method either of us is currently aware of. Thus it is the rational choice on which to base action; basing them on a less reliable source gives a higher probability of being "wrong", and thus would be irrational.

> I see no evidence that this narrative is generally true.

Honest answer: Then I invite you to attend a consensus conference, or enough of them for the p-value on your evaluation of how they run to reach whatever level is acceptable to you. Tongue-in-cheek answer: You also have no evidence you aren't a simulated agent in an advanced species' super-computer (Descartes's "brain in a jar"). At some point, you have to choose some axioms and build your world-view from there.

[Zuider]

>In order to act on a claim (inaction is also an action; "avoid committing oneself to accepting" is just ornate wording for choosing inaction) one must evaluate the "probability of truth" of said claim.

There is no requirement to evaluate the truth value of an ill defined, nonsensical or self-contradictory claim. In fact, a claim of that nature may as well be regarded as being meaningless. On those grounds, it makes no sense even to speak of probability in regard to such a claim.

>Honest answer: Then I invite you to attend a consensus conference

A 'consensus conference' may do one of two things:

1) It may dedicate itself to the evaluation of rational argument and evidence, in which case it is the argument and evidence that matters, and the consensus is irrelevant.

2) Some other means may be used to arrive at consensus, in which case it may be irrational, or at the very best, unscientific.

Note that the position that OP advanced, and which you appear to be defending is that the bare fact of the existence of a consensus should be treated as primary evidence on its own. At best, a consensus is secondary evidence, a pointer to the real crux of discussion, and treating it as primary evidence is a category error which is only compounded by the air of authority implied by a consensus which adds psychological coercion into the mix.

[silverspoonin]

>There is no requirement to evaluate the truth value of an ill defined, nonsensical or self-contradictory claim.

Calling a claim "ill defined, nonsensical or self-contradictory" is already an evaluation of it's probable truth value.

> Note that the position that OP advanced, and which you appear to be defending is that the bare fact of the existence of a consensus should be treated as primary evidence on its own.

Actually, the person you responded to clearly stated: "how should I evaluate truth in a subject matter that I myself don't have a doctorate in". For someone well-versed in the area, or at least with enough free time to become sufficiently acquainted with it, going to directly to primary evidence is the optimal approach. However, in the situation described by OP, were they are limited in the amount of time and cognitive capacity they can invest, then basing decisions on such "Secondary evidence" as expert consensus is the rational approach.

[atleta]

Nope. It's not that it's ought to be believed. The thing, in general, is that you always make a choice with your beliefs and decisions. When you make a decision (vaccinate or not to vaccinate, wear a mask, don't wear a mask) then you choose between two options and in the end who you believe.

A lot of people like to formulate this for themselves as being open and not accepting some kind of "dogma". Which sounds good, you stay open, what do you have to lose? (BTW, accepting the scientific consensus doesn't mean that you can't change your mind when new information comes in, but that's a tangent.)

But the thing is that when you have to make a decision then you have to accept some claims as being true (for at least the duration of making the decision). And that's when you have to ask yourself (as a layperson): who is more likely to be right? Which model is going to be more close to the reality? Science, the majority of scientists or some random dude? (Including maybe those few seemingly legit scientists who "speak up" in social media stating that all other scientists have gone completely wrong, including virologists and epidemiologists. But that one bloke is right.) And can that solo dude be right in theory? Yes, of course. Is he likely to be right? Well, looking back at the history of science it's rare, it doesn't last too long (until others realize and change their opinion) and it's getting more rare (as the scientific method evolves).

So yeah, no burden of proof on anyone, just a meaningful strategy for those outside of the field (any given field). Those doing research should always prove they are right. But, to get back to the original topic, that's not done in social media. That's done in scientific publications through peer reviewed papers. And, of course, the scientific consensus itself is supported by those so talking about shifting the burden of proof makes no sense at all. The proof is there for the consensus. That's the point.