Thursday, February 03, 2022

Skeptical About Skeptics

A lot of energy is being spent on conservative sites, even more now than previously, when it was already considerable, in being skeptical of experts, especially those with any authority. It's a fine Western and especially American tradition, and we should all be ready to ask ourselves questions about what we are hearing, and sometimes - to ask those questions aloud. Notice, however, that I framed this as being around questions. These must precede the skepticism.  Skepticism is not in and of itself a good thing.  It is a tool, very useful at times.  These days it seems to be prized for its own sake.  I remember thinking years ago the more leftist bumper-sticker Question Authority, immediately wondering what the intended replacement would be.  Because of course, it can only be a different authority, however disguised.

It's one of the reasons I so enjoy Razib Khan, who is a targeted rather than general skeptic. I listened today to his interview with Dr. William Gunn, a microbiologist who worked for Mendeley, then Elsevier, and is now Director of Communications  at Quora. (The website says for paid subscribers only, but it popped up for free on the podcast app on the device I carry with me when I walk.) In reply to Razib's observation that it seemed that Indians were now the primary experts replying in many fields on Quora, Gunn noted that people were leaving Google and Microsoft to come work on the AI at Quora because it is so fascinating. I did not know this, not an inkling.

Gunn is thus clearly tuned in to the change over the last twenty years in how science information gets out. He gave very clear information about how peer-review was an excellent system in the 50s and 60s but is now overwhelmed.  A million papers a year require peer review, and in some specialties, only 10-20 are qualified to review a new paper. He described his amusement at the first time one of his own papers was given to him to review - an accident that is not common but not at all unheard of. He knows the complaints from very close up at his previous jobs and observes how even those whose primary output is academic science publication do not entirely understand the system and make invalid complaints. He reviews some of the new media possibilities of how dissemination of science is to be done instead (including Quora), noting the immediate difficulties that many of the favorite panaceas present. Those of you familiar with academic publishing will likely appreciate the entire discussion. 

The very last part of the interview will be of more general interest. Razib and Gunn discuss all the problems with covid information coming out. Gunn talked about the difficulties of knowing what to emphasise and how to frame things.  Even when there is no particular political angle, only one person in a thousad will hear what you say directly in the moment.  The rest will get their information second, third, fourth hand, and even if they go back to the original source, they will already have been influenced by the intermediary sources. The original sources try to anticipate the many ways that the information will be misheard and misunderstood, trying to head off potential very wrong criticisms, and highlighting what they believe to be the most important points. As that is something I find myself doing when I write - perhaps not very successfully - I understood the problem immediately. There are ways that people bend information other than partisan politics - and nearly all of them are wrong. I had thought it might be 90% of the skeptics that just didn't have a good understanding, with a hefty portion of those being flat-out loons, but Gunn thinks it's 99% of skeptics have no business weighing in on a subject. He says this while acknowledging that the 1% often have insights that are very helpful.

Our discussion can go in a few directions from here. Is Gunn's impression colored too much by his personal experience of being on the receiving end of skeptics and trying to solve that? I thought I was pretty skeptical of the skeptics by estimating 90% inaccuracy and have admonished folks here that we aren't being skeptically enough, moving to all manner of unstable watercraft if we believe the main vessel has a leak or steering problem. 99% would be an amazing number...and yet he is much closer to this than I am.  This is how he makes his daily bread. 

I thought next of the subjects that I weigh in on, here and elsewhere. If the true number is 90%, then I am reasonably qualified to weigh in on a few things. Yet even on those, I am always aware of aspects where I only know "much more than average," not an amount that I should be seriously questioning people who have clear expertise. Mental health treatment and mental health law would be good examples of that. In general, I can comment, even taking on experts, because I have known and worked with experts who were incompetent, or so tied to one belief or another that they could not see the rest of the questions clearly. Yet I am not a prescriber, and so comment only in general on medications and promising new treatments.  I quote people who seem to me to make sense and have done their homework. That seems within my grasp. But if the real number is 99%, then maybe I'm one of those just-misses who also just doesn't get it and should shut up. Or on another topic, I read pastors and Christian writers who get things just wrong, sometimes badly, and will comment on interpretation.  Mostly, I follow CS Lewis, who I regard as more trustworthy than most others writing. If the percentage of reasonable skeptics is 10%, I qualify, especially as I will defer to others in places where my knowledge is only approximate, as in translation. But if the number is only 1%, I may be less helpful than I thought.

And strength is usually in discerning who is showing motivated reasoning, who is fighting fair, who is listening to and processing pushback information versus seizing on small points and treating them as definitive, or worse, disregarding pushback altogether and just bulling forward, reciting their claim more loudly and angrily. Having knowledge in an area helps alert me to people doing that, as I can know they are likely flat wrong right out of the gate.  Yet even in situations where I start out neutral because my knowledge is slight - epidemiology would be a good example - I can usually tell who is making the better case. All of this bears further thought.

On the other, other hand, I worked with doctors all my career, who had this voice they would adopt, clearly based on the attitude they had of "I'm a doctor." This was used not only in medicine outside their specialty - some justification there - but outside of medicine altogether. It roughly translates as "I'm one of the smart ones who has to know about a broad range of things." It ain't so.  (There is a pattern of the demographics of who are the worst offenders in my own generation - males from anywhere, the older the worse - compared to the worst offenders among the younger doctors - American females. It's an amazing switch.) Pastors also have that tone, based on that attitude of "but it's my job and my training to understand and apply Biblical principals to society at large and the Real Lives that people live." And thus they speak as if they are at least semi-experts on sociology, economics, criminology, psychology, history, and everything else where they had some carefully curated academic instruction. Again, it ain't so. A few lecturers and professors I have known do the same, even outside their fields.  They are usually smarter than the average bear. The often have read widely and their knowledge does extend beyond their specialty.  But they overreach repeatedly, as I can plainly detect when they come into my territory. If I am harsh about them, how much more should I step back and ask "Do I know what I am talking about here?"  William Gunn used the phrase epistemic humility.

Where Razib went was related to my thoughts about the validity of my own skeptical challenges, but I think he went one better. He asked if there was some difference when we had things like covid, where political considerations were pretty obviously coloring the views of the experts. Gunn had already dropped a few hints that suggested he is liberal, and his defense of experts did start out sounding like special pleading of the "they only made such a confusing mess of things because they were trying to steer around all the unreasonable skeptics" variety. Yet that did not last long, and he agreed pretty thoroughly with the idea that this had happened. Sidebar: Gunn mentioned the masking switch (I would add depressing the curve), which is a common complaint, but Razib hit a different moment.  His recollection, following both conservative twitter and science twitter at the time, is that the health professionals endorsing the George Floyd protests was a tipping point. They aren't the CDC, so it's not entirely fair, but he thought people were willing to overlook the mistakes about masks and even the wrong predictions, because it was all uncharted.  But for people to go out chanting in protest and getting applauded for it was a stark declaration of "We don't really mean what we are saying when push comes to shove." I had not thought of it that way, as people being jerks is a given in my book, and thus usually irrelevant. Yet I trust Razib's memory and Gunn agreed that this was very big. Huh. This is how history is remade in memory. Or my memory, anyway. Also, if health professionals were doing this, perhaps the CDC had the responsibility to call them out?

Ironically, outdoor transmission, even with chanting, turned out to be minimal, which is perhaps why the hypocrisy has receded in my memory.  But we did not know that then, and were still encouraging people in all sorts of caution.

Gunn did not think this was a one-off, and thinks it will happen on political issues again, because all the things that bend experts remain stable: where their funding comes from, where professional status comes from, professional ideologies that they have built careers on, or defending those they trained under and their school of thought. String theory eating up funding or the professional danger of destruction for attacking Chomsky's Universal Grammar is real in all fields. Politics is merely one of those ever-recurring problems.  However, Gunn still thinks that attending to the experts is a better bet than attending to the skeptics.  Because the skeptics are mostly loons.  He believes that the 1% of skeptics are usually the only ones who can bring down the Conventional Wisdom. The rest of us are just chirping. He thinks we should save our fire for targets we can hit.  He may be trying to reduce the irritation of people who have to put up with nonsense. But he's got a solid point here, and he has seen the crap flow into question boxes or onto bulletin boards for quite some time.

I see his point and think I will shade in that direction, but I don't think it is going to turn me around. I might be more cautious.  But I can still know an ad hominem argument when I see one, detect at least a possible conflict-of-interest in a person answering, or sniff out views that are more fashionable - or the delicious fashion of being anti-fashionable - than reasonable.


james said...

I wasn't trying to measure skepticism of the CDC et al, but the Floyd protest endorsements did cause quite a stir, and I think Razib and Gunn are right that there was a lot of trust irrevocably lost then.

There's a bit of variation in types of skepticism.

Flat disbelief is one thing, suspecting that there's more to the story is another.

If the experts take the time to show their work, does that make a difference? (Given how easy it is to lie with statistics, maybe that shouldn't make a difference for people without some background in statistics or experiment design.)

Can you accept a claim provisionally, and be able to discard it easily if it's disproved? (First report out of the gate tends to stick in the mind better--mine too.)

Grim said...

I would tend to say that 99% might even be an underestimate, depending on what you define as 'skepticism.' In large political matters, almost everyone has an opinion; almost none of them are experts in the field.

Yet there is an alternative to accepting a skeptical authority if you are inclined to reject the authoritative authority. This is to hold the question open, and to take reasonable steps to hedge bets. I think this is often the wisest position in cases in which experts disagree, given that science is always open to revision and in fact is always revising.

In the case of Climate Change, for example, there is a large scientific consensus about it; but we who are not experts notice there is also a massive amount of power and money behind that consensus, such that careers are easily made and funding easily acquired in one direction and not the other.

(That by itself is reason for skepticism, even if there were zero experts raising skeptical considerations -- and one that requires no special expertise.)

There are also a few skeptics, far fewer among the experts in the field, raising what appear to be reasonably serious questions.

As a voter and a citizen, or as a father and a homeowner, you could choose to cast your lot with one group or the other. Of course, being inexpert yourself, the choice is just as blind whether you go the one way or the other way. It's a gamble.

Alternatively you could do what we usually do with risks, which is try to hedge them. Perhaps you might decide to buy that beach vacation home, even though it could be underwater in... 5 years? 50 years? But you could also buy a flooding insurance policy. Or perhaps you could choose to adopt 'green' lifestyle choices where they make sense and are economical, rather than bending your entire life in the one direction or totally refusing and buying one of those 'rolling coal' trucks to trumpet your disdain for the consideration.

(Do you know about rolling coal? It's a highly dubious practice even from a mechanical perspective as it can damage your engine, but it's popular in some areas -- I don't know if yours is one of them. I never see real country boys do it, but suburban cowboys seem to think it's cool.)

Since no one is an expert on every question, we all have to do this kind of calculation all the time in various areas. And if we are to remain a free and self-governing people, it is necessary to entrust decision-making on even very important questions to non-experts -- because almost all of us are not experts on the particular question, but our equality lies practically in that our votes are equal in weight. If we change that, and make 'some more equal than others' (even only on some questions) we lose that equality and with it the capacity to govern ourselves as a people rather than accepting being governed by others 'for our own good.'

(What constitutes a "good" to be pursued is a question in which philosophers are especially expert, so you might think I was arguing against interests here; like Plato, I could set myself and my class atop the heap by trying to reserve questions to experts, and then pointing out that we happen to be the experts on the very highest and most central question of all. You have heard many of my arguments against taking that position, however.)

So: do question authority; don't necessarily accept anyone else's authority in lieu of it, either. Hedge your bets as seems prudent and hold questions open, perhaps even forever.

Christopher B said...

You do seem to have a much higher tolerance for people in authority acting like jerks. The risk of false positives is admittedly high but when the actions of the people giving orders are not in line with the orders being given most people's BS detectors go to the peg, and past it when lame excuses like "I was holding my breath" are offered.

Assistant Village Idiot said...

Mere wrongness, especially if it was easily avoidable, bothers me more than hypocrisy. Wrongness in a hard choice, not so much. I have known lots of competent pricks. Being wrong for personality reasons does bother me a lot.

Zachriel said...

Assistant Village Idiot: Skeptical About Skeptics

It's important to distinguish between healthy skepticism and unsound denialism.

"In science, 'fact' can only mean 'confirmed to such a degree that it would be perverse to withhold provisional assent.'" — Stephen J. Gould

Zachriel said...

Grim: I think this is often the wisest position in cases in which experts disagree, given that science is always open to revision and in fact is always revising.

While experts may disagree, especially at the margins, they often agree on other fundamentals. In other words, just because there may be expert disagreement about the exact shape of the Earth, experts may nonetheless agree that the Earth more resembles a sphere than a flat surface. Not all views are equivalent just because there is scientific debate.

Grim: In the case of Climate Change, for example, there is a large scientific consensus about it; but we who are not experts notice there is also a massive amount of power and money behind that consensus, such that careers are easily made and funding easily acquired in one direction and not the other.

Ad hominem is not always a fallacy when addressing an appeal to authority. However, we would point out that the consensus on climate includes scientists from many different countries, under various political systems, and in dissimilar cultures. More important, the consensus spans entirely different disciplines using independent methodologies.

How do we know something is a valid scientific authority? One of the most important measures is that a valid field of study overlaps with other fields. So evolutionary science overlaps with zoology and ecology and geology and molecular biology and so on. And each of these fields also overlaps other fields. This overlap gives a strong foundation to the entire scientific edifice. There will inevitably be tension between and within the different fields, and this tension can lead to new scientific insights or even new fields of study.

More generally, an appeal to authority is a type of inductive argument {eta: based on the experience that experts are more likely to be correct than non-experts in a field, though not infallibly so} and is evaluated as follows:

• The cited authority has sufficient expertise.
• The authority is making a statement within their area of expertise.
• The area of expertise is a valid field of study.
• There is adequate agreement among authorities in the field, and the authority is expressing this agreement.
• There is no evidence of undue bias}.

The proper argument against a valid appeal to authority is to the evidence.

stevo said...

One of the weak links that first made me skeptical was the PCR tests. The evidence seemed very convincing that these tests were not intended to diagnose infection, and could easily produce false positives, especially if the number of cycles used was higher. As far as I can tell these concerns were never addressed and the same test continues to be the "gold standard"

Assistant Village Idiot said...

I will let others weigh in on their value for individual diagnosis. But what I have most often heard this objection used for is to suggest that our overall count of covid death is way too high, as the test does not fully differentiate among covid, colds, and flu, so therefore it might not be covid. But we haven't had a million people die of colds, not even 0.1% of that, and if these were really flu (Mike Kennedy over at Chicago Boyz used to claim that when I still read his comments), then we have had two flu seasons that are 20x average back-to-back, with no explanation. I don't need an expert to tell me that's just not true. I can do big picture stuff, if nothing else.

Small chinks in the narrative seldom bother me, unless it can be shown that they actually point to bigger problems, rather than just show "Well, there could be problems, right? Right?"

Tom Bridgeland said...

We certainly could be having a few cold/flu/etc cases slipping through diagnosed as Covid. But at least in my experience in the hospital, not many. A few months back we had several cases of RSV in older adults. Normally a little kid virus, it was quite harsh in these adults. But, while superficially the symptoms were very similar, we pretty quickly figured out that something else was going on and did a 'respiratory panel' which nailed down the diagnosis.

The lab values are different. Covid is pretty unique and not hard to pin down even without a positive test. At home, I doubt could differentiate a mild covid case from any of several dozen possible diseases. I had a mild 'cold' last week and after a few days used an at-home covid test. Positive. So I am isolating until Thursday when I should go back to work.

Assistant Village Idiot said...

Thank you again.