NꙮW WITH MꙮRE MULTIꙮCULAR ꙮ

Learning To Love Scientific Consensus

[Related to: Contrarians, Crackpots, and Consensus, How Common Are Science Failures?. Epistemic status is “subtle and likely to be misinterpreted”.]

I.

There’s a list of scientific mavericks who were ridiculed by hidebound reactionaries but later vindicated that’s been going viral. I examined the first ten mavericks on the list to see if its claims held up. Overall I wasn’t too impressed. Let me go over them in more detail.

SVANTE ARRHENIUS:

His idea that electrolytes are full of charged atoms was considered crazy. The atomic theory was new at the time, and everyone “knew” that atoms were indivisible (and hence they could not lose or gain any electric charge.) Because of his heretical idea, he only received his university degree by a very narrow margin.

Sure, the professors who were judging his PhD thesis weren’t too convinced. So Arrhenius sent his proposal to the world’s top chemists at the time, and they were super-interested and started fighting among themselves to work with Arrhenius on it. Top chemist Wilhelm Ostwald received the paper the same day his daughter was born, and suggested that the paper was the more exciting of the two events. He journeyed to Arrhenius’ hometown of Uppsala, Sweden to try to convince Arrhenius to work with him; Arrhenius refused for personal reasons but later got a scholarship and worked with the top physicists in Europe. Arrhenius became a professor in a prestigious university about ten years after presenting his “ridiculed” paper, and won the Nobel Prize ten years after that.

HANS ALFVEN:

Astronomers thought that gravity alone is important in solar systems, in galaxies, etc. Alfven’s idea that plasma physics is of equal or greater importance to gravity was derided for decades.

This isn’t a great description of Alfven’s conflict with the establishment, but the list seems basically right insofar as Alfven’s ideas were ignored for thirty years before being proven mostly correct. I will give them this one.

JOHN BAIRD:

When the first television system was demonstrated to the Royal Society (British scientists,) they scoffed and ridiculed, calling Baird a swindler.

I can’t find any reference to this in various Baird articles and biographies. The closest I can come is this article by someone who was there at the demonstration, who said “They didn’t believe it…the pictures were a bit of a blur but it was amazing, they were all absolutely flabbergasted by it.” It looks like he is using “they didn’t believe it” in the colloquial way of “they thought it was amazing”. A TIME magazine article from the time described the same scientists as “deeply impressed”, though the wording is kind of unclear and they might have been referring to a different demonstration a year later.

In any case, it seems very clear that within a year everyone agreed he was legitimate and overcame their initial shock.

ROBERT BAKKER:

Everyone knows that dinosaurs are like Gila monsters or big tortoises: large, slow, and intolerant of the cold. And they’re all colored olive drab too! 🙂

Bakker did help produce the paradigm shift in paleontology from cold-blooded dinosaurs to warm-blooded dinosaurs. But he was not a lone maverick being ridiculed by everyone else. He learned that dinosaurs were warm-blooded from his professor at Yale, who was also part of the minority-but-totally-existing faction that believed dinosaurs were warm-blooded. He himself got a PhD at Harvard from professors who were apparently sympathetic to the same theory. And within seven years of his first paper being published, Scientific American was calling his ideas “the dinosaur renaissance”, which doesn’t leave a lot of time for him to be ridiculed and ignored in.

BARDEEN & BRATTAIN:

Not ridiculed, but their boss W. Shockley nixed their idea for a non-FET “crystal triode” device. When they started investigating it, he made them stop. They were supposed to be working on FETs instead.

ARG, I GOT THIS WRONG, THIS PART BELOW IS A BELL LABS STORY REGARDING ZONE REFINING OF SILICON, NOT THE BJT TRANSISTOR PROJECT: So, they assembled their ZONE REFINING experiment on a wheeled cart and continued. Whenever the boss was scheduled to check up on them, they could shove it into an adjacent unused lab.

Okay, it looks like the guy compiling the list admits he was wrong on this one. Moving on…

BRETZ:

Endured decades of scorn as the laughingstock of the geology world. His crime was to insist that enormous amounts of evidence showed that, in Eastern Washington state, the “scabland” desert landscape had endured an ancient catastrophy: a flood of staggering proportions. This was outright heresy, since the geology community of the time had dogmatic belief in a “uniformitarian” position, where all changes must take place slowly and incrementally over vast time scales. Bretz’ ideas were entirely vindicated by the 1950s. Quote: “All my enemies are dead, so I have no one to gloat over.”

This one is basically right and I’ll give it to them.

CHANDRASEKHAR:

Chandra originated Black Hole theory and published several papers. He was attacked viciously by his close colleague Sir Arthur Eddington, and his theory was discredited in the eyes of the research community. They were wrong, and Eddington apparently took such strong action based on an incorrect pet theory of his own. In the end Chandra could not even pursue a career in England, and he moved his research to the U. of Chicago in 1937, laboring in relative obscurity for decades.

Sort of true, but he was hardly shunned by the scientific community. He made his discoveries about black holes in the early 1930s, was well-received by many people, and won a Bronze Medal in some physics competition. In 1935, Eddington attacked his theory, possibly because Eddington was racist and didn’t like Indian people. But many other scientists, including Niels Bohr and Wolfgang Pauli, continued to support him (quietly, so as not to offend Eddington, which will be a recurring theme in these kinds of situations). Chandrasekhar was made a Fellow of the Royal Society in 1944, won the Royal Astronomical Society Gold Medal in 1953, and generally led a long and prestigious life. His theories were resurrected once people had better evidence that black holes existed. I’ll give this one half a point.

CHLADNI:

The scientific community regarded Meteorites in the same way that modern scientists regard UFO abductions and psychic phenomenon: quaint superstitions only believed by peasant folk. All the eyewitness reports were disbelieved. At one point the ridicule became so intense that many museums with meteorites in their geology collections decided to trash those valuable samples. (Sometimes hostile skepticism controls reality, and the strongest evidence is edited to conform to concensus disbeliefs.) Finally in the early 1800’s Ernst Chladni actually sat down and inspected the evidence professionally, and found that claimed meteorites were entirely unlike known earth rocks. His study changed some minds. At the same time some large meteor falls were witnessed by scientists, and the majority who insisted that only ignorant peasants ever saw such things were shamed into silence.

As the quote points out, this is a kind of weird one as meteorite work was ridiculed for a long time, but Chladni was taken seriously and helped change minds. Looking at Wikipedia, a lucky meteorite fall two years after Chladni first published his theory helped turn the tide in his favor, and by ten years after publication Chladni’s meteorite theories were pretty well-regarded. Even when people disagreed with him about meteorites, Chladni remained widely respected for some of his other work in acoustics.

There is a story here, but it’s probably not right to center it around Chladni, and his work was only scorned for a few years before everyone agreed it was true. I’ll give this another half a point.

CRICK & WATSON

Not ridiculed. But they were instructed to drop their research. They continued it as “bootleg” research.

The list admits they were “not ridiculed”. They were told to stop their research because there was all sorts of academic politics around who was going to be the first to discover DNA, and the guy in charge of their university was rooting for another team.

DOPPLER

Proposed a theory of the optical Doppler Effect in 1842, but was bitterly opposed for two decades because it did not fit with the accepted physics of the time (it contradicted the Luminiferous Aether theory.) Doppler was finally proven right in 1868 when W. Huggins observed red shifts and blue shifts in stellar spectra. Unfortunately this was fifteen years after Doppler had died.

I haven’t been able to find anything about this in various short online biographies of Doppler (1, 2). Doppler tested the effect himself by having someone play a trumpet on a train (really), someone else successfully tested it in 1845, and it was independently rediscovered in 1848. Doppler himself was made the head of the Institute For Experimental Physics in Vienna and died about as prestigious and beloved as a physicist can get.

So my impression is that only a third of these people really fit the pattern. Most of them were doubted for very short periods, continued to be respected in their fields for their other accomplishments even during those periods, or were part of medium-sized movements rather than being lone geniuses. After a few years – maybe an average of ten, very rarely as long as thirty – their contributions were recognized and they assumed their rightful place in the pantheon. Science isn’t perfect. But it is darned good.

[EDIT: Bill Beatty, author of the original list, responds here. My response to the response here.]

II.

I bring this up in the context of my last post on progress in the rationalist movement. There used to be a stereotype that rationalists were too quick to challenge scientific consensus. I think that was exaggerated, but based on a core of truth. Given that we’re interested in the ways that bias can prevent people from accepting truth, it’s unsurprising that we would focus on cases like these.

But I personally have changed my thinking on this a lot. Not in any way that I can explain explicitly – I’ve always thought something like:

Scientific consensus is the best tool we have for seeking truth. It’s not perfect, and it’s frequently overturned by later scientists, but this is usually – albeit not literally always – the work of well-credentialed insiders, operating pretty quickly after the evidence that should overturn it becomes available. Any individual should be very doubtful of their ability to beat it, while not being so doubtful that nobody ever improves it and science can never progress.

– and I still think that. But I’ve shifted from being the sort of person who shares viral lists of maligned geniuses, to the sort of person who debunks those lists. I’ve started emphasizing the “best tool we have” part of the sentence, and whispering the “isn’t perfect” part, rather than vice versa.

I’ve changed my mind on this because of personal experience. Rather than trying to describe it, it might be more helpful to give the most salient examples.

1. The Replication Crisis: I previously thought the scientific consensus was flawed because it failed to take the replication crisis seriously enough. I later learned that everyone else took the repliaction crisis exactly as seriously as I did. A poll in Nature shows that 90% of scientists believe reproducibility issues constitute a “crisis”, compared to only 3% (!) who don’t. For every person complaining about “methodological terrorists”, there are a dozen who are very concerned and trying to change the way they practice research.

This is especially impressive because as far as I can tell the whole shift happened in about ten years. I would date the beginning of the crisis from Ioannidis’ original 2005 paper, although it was only aimed at medicine. It got into high gear in psychology sometime around 2011 with Simonsohn’s False Positive Psychology. A Google Trends analysis suggests people only started searching the relevant keywords around 2013.

I started thinking about this sort of thing in 2009 after reading this LW post. At the time I thought this was some sort of exciting failure of modern science that I alone had figured out. But this was well after sharp people like Ioannidis were talking about it, and only a few years before everyone was talking about it. Framing this as “I was right and scientific consensus was wrong” seems grandiose. Better might be “I started betting on a winning horse about a quarter of the way between the beginning of the race and when its victory became blatantly obvious to everyone”.

2. Nutrition: The Bad Old Paradigm of nutrition says that obese people just have poor impulse control, that weight is a simple matter of calories in vs. calories out, and that all calories are equally good except fat, which for some inexplicable reason is the Devil. Anybody who’s read a few good books about nutrition science knows that the Bad Old Paradigm is woefully inadequate. I read a few of those books and became convinced that I was right and scientific consensus was wrong.

Unfortunately, this whole issue exploded when Gary Taubes published Good Calories, Bad Calories, which as best I can tell combined the first publicly available good critique of the Bad Old Paradigm with a flawed and basically false attempt at a new paradigm. There were lots of confused attacks against Taubes’ bad information which did collateral damage to his good information, and lots of confused defenses of his good information which inadvertently shielded his bad information from criticism. I previously focused on defend the good parts, but recently shifted more towards criticizing the bad parts.

After reading some more good books here (one of which I hope to review soon), my impression is that most nutrition scientists don’t believe in the Bad Old Paradigm and haven’t for a while. At the very least, most of them seem to believe in the lipostat and think it’s important, which is my proxy for “basically has their heart in the right place”. Insofar as the Bad Old Paradigm continues to be popular wisdom, it’s because of the diet industry, the government, social inertia, and nobody really having a good new paradigm to replace it with. I’m gradually seeing popular wisdom shift, and nutrition scientists themselves seem to be helping this process rather than hurting it.

Maybe somebody in this area has discovered the new paradigm and is a maverick being persecuted by hidebound reactionaries. But it isn’t Gary Taubes. And it certainly isn’t me.

3. Social-Justice-Related Issues: Another narrative I used to believe was that a lot of sketchy ideas were being flattered because they spoke to left-leading academics’ biases in favor of social justice. Implicit association tests, stereotype threat, the idea of zero meaningful psychological differences between men and women, et cetera.

When I started worrying about implicit association tests, I thought I was defying some kind of broad scientific consensus. But the meta-analyses showing the Implicit Association Test didn’t do what people thought had been around since 2009 and have only gotten more numerous since then, with broad media coverage. Problems with stereotype threat research are getting mainstream coverage and even airtime on NPR.

The problem here is that there was no equivalent of the Nature poll on the replication crisis, so I didn’t realize any of this was happening until just recently. For example, in 2016 this Voxsplainer made it sound like there was a monolithic consensus in favor of Implicit Association Tests that no sane person had ever disagreed with, even though by that point there were already several big meta-analyses finding they weren’t practically useful. The correct conclusion isn’t that this is really what scientific consensus thinks. The correct conclusion is that Vox shouldn’t be trusted about any science more complicated than the wedge vs. inclined plane. Once I realized that there was all this intelligent analysis going on that I’d never heard about, my claim to be boldly defying the scientific consensus evaporated.

Yes, Cordelia Fine is still around and is still writing books arguing against gender differences. But she’s starting to sound really defensive, basically the literary equivalent of “I know I’m going to be downvoted to hell for this, but…”. Meanwhile, other scientists are doing a good job pointing out the flaws in her books and conducting studies like this biggest-ever look at male vs. female brain differences, this magisterial look at personality differences, et cetera – not to mention great and widely-accepted work on how intersex people take on more characteristics of their hormonal than their social gender (honestly, we should probably thank transgender people for making this field socially acceptable again). People talk a lot about how Larry Summers was fired from Harvard for talking about male vs. female differences, but Steven Pinker did a whole debate on this and remains a Harvard professor.

Even things about genetic psychological differences between population groups are less bold and maverick-y than their proponents like to think. The relevant surveys I know trying to elicit scientific consensus (1, 2, 3) all find that, when asked anonymously, most scientists think these differences explain about 25% – 50% of variance.

I hate to bring that up, because it’ll probably start a flame war in the comments, but I think it’s important as a sign of exactly how hard it is to politicize science. Global warming skeptics talk about how maybe the scientific consensus on global warming is false because climatologists face political pressure to bias their results in favor of the theory. But scientists studying these areas face much more political pressure, and as long as you give the surveys anonymously they’re happy to express horrendously taboo opinions. This is about the strongest evidence in favor of the consensus on global warming – and scientific consensus in general – that I could imagine.

4. Nuture Assumption and Blank Slatism: The prologue of the first edition of The Nurture Assumption is Judith Rich Harris telling her “maverick genius kept down by hidebound reactionaries” story. But the prologue of the second edition is her being much more hopeful:

To some extent at least, times have changed…there is now more acceptance of the idea that behavior is influenced by genes and that individual differences in behavior are due in part to differnces in genes. People are more willing to admit that children can inherit behavioral quirks and personality characteristics…was it this cultural shift that led to greater acceptance of my theory? Or was it the fact that new findings, consistent with the theory, kept turning up? Over time, the early, angry response to The Nurture Assumption has softened noticeably, both within and outside of academia. Today, the book is widely cited in textbooks and journal articles. It’s assigned and discussed in courses in many colleges and universities; it shows up in exams…in his foreward to the first ediction of The Nurture Assumption, Steven Pinker made a rash prediction about the book: “I predict it will come to be seen as a turning point in the history of psychology”. Perhaps it is too soon to judge whether psychology has rounded a bend; perhaps it will take the perspective of twenty or thirty years. Even at this point, though, there are signs of a slight shift in direction. Within developmental psychology, I’ve noticed that descriptions of procedures and results are beginning to sound a bit defensive. Greater progress has been made in other areas of psychology. And the email I receive from students gives me high hopes for the younger generation coming up.

There were ten years between the first and second editions of The Nurture Assumption. In the almost ten years since the publication of the second edition, my impression is that its ideas have become even more widely-accepted. This month’s edition of the American Journal of Psychiatry, onbe of the top journals in the field, has a great study showing that child abuse does not cause cognitive disability, in contrast to several previous studies in the area. It cites Deary, Plomin, and Ioannidis, hits all of the talking points about genetic confounding of developmental outcomes, and receives glowing endorsement in the journal’s editorial section, which says that “if our causal explanations are wrong, we may be wasting our effort or even doing damage”. Every single psychiatrist in the country is getting exposed to this way of thinking.

And this has real results. I got to present a summary of behavioral genetics to a meeting of psychiatrists, including a lot of psychoanalysts, and I was shocked that most of them were at least a little receptive. I think they misunderstood it. I think they carefully raised caveats in exactly the right places to ensure they didn’t have to change anything they were doing. But the overall response was “Oh, yeah, we’ve heard stuff like that, it seems plausible, good thing that for various hard-to-explain reasons none of it applies to us.” This is what the first stage of progress looks like.

5. Intelligence Explosion And AI Risk: This was another place where I and many of my friends thought we were right and the consensus was wrong. It was another place where a lot of self-appointed defenders of the consensus told us we were crackpots and needed to listen to what real scientists thought. And again, when I looked into it, there was no consensus against the idea and lots of prominent researchers were in favor. Going to the Asilomar Conference and seeing a bunch of people from MIT and Harvard talk about how concerned they were really opened my eyes on this. Google now has an AI Ethics Board, Berkeley, Oxford, and MIT have foundations working on it, and people like Elon Musk and Bill Gates are involved. Bostrom’s survey of AI researchers and some more recent and rigorous not-yet-published surveys I’ve heard about confirm the impression. Nobody would ever say there’s a scientific consensus in favor of Bostrom’s theories. But at this point I think it’s also indefensible to say there’s a consensus against.

Bostrom first started writing about these sorts of things extensively in the early 2000s, so there was really only a ten-year gap between entering the intellectual environment and it becoming a (mostly) accepted part of the established field. Those ten years felt pretty long while we were in them, but the ability of a field to accept an on-the-face-of-it completely-insane-sounding theory within ten years seems to me a very strong argument against the hidebound-reactionaries theory and a very strong argument for considering scientific consenses to be unreasonably effective.

6. IQ: Another case where I worried about apparent failure of scientific consensus due to politically bias. I certainly encountered a lot of falsehoods around this when I was younger. My high school psychology textbook included a section claiming that all IQ tests were biased towards rich white people because they were based entirely on questions like “how many shots below par is a bogey?” Then it presented an “alternate IQ test” which “proved” that poor minorities had higher IQs than rich whites by asking some other questions with the opposite bias (I think they were about slang for drugs – certainly an interesting way to fight stereotypes). This kind of thing naturally made me assume that nobody had any idea what was actually in IQ tests and scientists were idiots.

But more recently I’ve been reading actual surveys, which find that about 97% of expert psychologists and 85% of applied psychologists agree that IQ tests measure cognitive ability “reasonably well”. And 77% of expert psychologists and 63% of applied psychologists agree IQ tests are culture-fair (with slightly different numbers depending on how you ask the question, but always about 50% of both groups).

This seems like less of a problem with expert consensus, and more of a problem of nobody else (including textbook writers!) listening to experts who are continually trying to beat reality into people’s heads. But I have a vague memory of having recently seen a survey (which I can’t find) that even experts in softer fields like sociology are generally in favor of IQ and admit that it has its uses. And even some left/liberal sources like Vox and Freddie deBoer are aware of the consensus and willing to respect it.

At the same time, I’ve encountered some people like Borsboom and Nostalgebraist who have relatively sophisticated (and limited) critiques of IQ, and who have allowed me to round off other people’s less-well-framed critiques to something more like what they are saying and less like the stupid things my high school textbook said.

So it seems to me that generally experts agree with reasonable statements about IQ, and where they seem to disagree they may hold reasonable disagreements rather than unreasonable ones. Again, where this fails is not in the experts but in the ability of people who don’t listen to the experts to get disproportionate social power and hide the existence of the expert consensus.

III.

Last week I wrote about universally-known criticisms of economists, like “they’re silly for assuming everyone behaves perfectly rationally”:

My impression is that economists not only know about these criticisms, but invented them. During the last few paradigm shifts in economics, the new guard levied these complaints against the old guard, mostly won, and their arguments percolated down into the culture as The Correct Arguments To Use Against Economics. Now the new guard is doing their own thing – behavioral economics, experimental economics, economics of effective government intervention. The new paradigm probably has a lot of problems too, but it’s a pretty good bet that random people you stop on the street aren’t going to know about them.

The same pattern explains a lot of my concerns above. I knew some criticisms of a scientific paradigm. They seemed right. I concluded that scientists weren’t very smart and maybe I was smarter. I should have concluded that some cutting-edge scientists were making good criticisms of an old paradigm. I can still flatter myself by saying that it’s no small achievement to recognize a new paradigm early and bet on the winning horse. But the pattern I was seeing was part of the process of science, not a condemnation of it.

Most people understand this intuitively about past paradigm shifts. When a creationist says that we can’t trust science because it used to believe in phlogiston and now it believes in combustion, we correctly respond that this is exactly why we can trust science. But this lesson doesn’t always generalize when you’re in the middle of a paradigm shift right now and having trouble seeing the other side.

I realize I’m (ironically) risking making my narrative of scientific success unfalsifiable. Suppose someone wants to argue that scientific consensus is wrong. If they point to something it used to be wrong about, I can respond “Yes, but it self-corrected and it’s correct now, so that’s fine.” If they point to something where cutting-edge scientists say it’s wrong but nobody else agrees, I can respond “Yes, this is what the beginning of a paradigm shift looks like, so that’s fine”. And if they point to something where nobody in the field thinks it’s wrong, I can say “You’re a crackpot for going against all reputable scientists; the problem is with you.” And if later they turn out to be right, and everyone acknowledges it, I can say “Yes, but it self-corrected and it’s correct now, so that’s fine.”

(and I’m making it even easier for myself in that I say “scientific consensus for” when I probably mean “no scientific consensus against”. I don’t claim that 90%+ of scientists always believe true things, only that there are very few cases where 90%+ of scientists believe things which smarter people know to be false.)

Against this I can only offer a personal narrative: the only light I have by which to judge scientific consensus is my own Inside View assessment of what seems correct. Again and again I have tried to defy scientific consensus. And every time, I either find that I am wrong, find that I am a few years ahead of a trend that most scientists eventually agree with, or find that what I thought was “scientific consensus” was actually a fiction peddled by biased industry or media sources slandering a scientific community which actually had a much more sophisticated picture. My history of trying to fight scientific consensus has been a Man Who Was Thursday-esque series of embarassments as I find again and again that my supposed enemy agrees with me and is even better at what I am trying to do than I am.

Scientific consensus hasn’t just been accurate, it’s been unreasonably accurate. Humans are fallible beings. They are not known for their ability the change their mind, to willingly accept new information, or to put truth-seeking above political squabbles. And our modern society is not exactly known for being an apolitical philosopher-kingdom with strong truth-seeking institutions completely immune from partisan pressure. I feel a deep temptation to sympathize with global warming denialists who worry that the climatological consensus is biased politicized crap, because that is exactly the sort of thing which I would expect to come out of our biased politicized crappy society. Yet again and again I have seen examples of scientific fields that have maintained strong commitments to the truth in the face of pressure that would shatter any lesser institution. I’ve seen fields where people believe incredibly-bizarre sounding things that will get them mocked at cocktail parties just because those things seem to be backed by the majority of the evidence. I’ve even seen people change their minds, in spite of all the incentives to the contrary. I can’t explain this. The idea that scientific consensus is almost always an accurate reflection of the best knowledge we have at the time seems even more flabbergasting than any particular idea that scientists might or might not believe. But it seems to be true.

(note that I’m talking about “scientific consensus” to mean a very high-level pattern, consisting of hundreds of scientists over the space of decades evaluating a broad body of work. Any individual study is still probably total garbage.)

Given how weird all of this is, I realize there’s another possible bias here that should be taken very seriously – which is that I’m wrong about one or both sides of this. Which is more likely: that Science always agrees with Truth? Or that one guy’s perception of Science always agrees with that same guy’s perception of Truth? The latter gives me two degrees of freedom: I can either cherry-pick experts who agree with me and declare them to be Consensus, or I can conform my opinions to consensus so slavishly that I end up discovering only that Consensus agrees with itself. I don’t feel like I’m making this kind of mistake. But then again, nobody ever feels like they’re being biased.

But if I’m making this mistake, I think it’s at least a better mistake than the one where people dream up stories about being mavericks persecuted by hidebound reactionaries. This mistake at least sets the terms of debate as “let’s try to ascertain what the scientific community thinks” and forbids me from believing completely crackpottish things. And it encourages trust in one of our more trustworthy public institutions, always a prosocial sort of thing to do. I would rather have a world of people debating who agrees with scientific consensus or not, than a world of people debating whether scientific consensus is even valuable.

There are two caveats to the above. First, I think it’s dangerous to promote a norm of agreeing with scientific consensus, insofar as that helps encourage exactly the mistakes about the nature of consensus that I discussed above. When poorly-informed diet industry gurus support the Bad Old Paradigm, their rallying cry is usually “You’re a stupid crackpot, bow to the scientific consensus which agrees with me”. I gave three examples above of cases where I would have gotten the scientific consensus 100% wrong if I didn’t have access to a formal survey of scientific experts. In a world where these surveys had never been done – or some existing field without these surveys – or some field where these surveys have been done inaccurately or in a biased manner – people will often believe the consensus to be the opposite of what it really is. In those cases, demands that people respect consensus can be used to shut down people who are actually right – the field-wide equivalent of calling true facts you don’t like debunked and well-refuted. I see this happening all the time and I worry that waxing too poetically about the unreasonable effectiveness of scientific consensus will only serve to empower these people. Goodhart’s Law says that a measure which becomes a target ceases to be a useful measure, so we should be reluctant to target scientific consensus too strongly.

And second, I think that even when the Outside View tells you that the consensus is correct, you should continue pursuing your Inside View hunch that it isn’t. This avoids awkward situations like every individual scientist doubting the consensus, but suppressing their doubts because the “scientific consensus” has to be right.

So maybe the things I’m saying about scientific consensus aren’t very actionable. But respecting scientific consensus in a non-actionable way is a lot less exhausting than believing yourself to be against it, and talking about how you’re against it, and taking flak for being against it. And in the same way it’s helpful to believe that God is good, even if He never really gets around to doing much about it, so it’s reassuring to be able to have faith in our institutions every so often.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

615 Responses to Learning To Love Scientific Consensus

  1. dk says:

    RE: Judith Rich Harris

    The following is an excerpt from a 2009 Time Magazine interview with her. Harris deliberately argued an extreme position to provoke a response from developmental psychologists and other academics.

    Link

    Q: How strongly do you believe The Nurture Assumption’s assertions hold up a decade on?

    A: They’ve held up quite well. I took an extreme position: that parents have no important long-term effects on their children’s personalities. By doing this, I was making myself an easy target, inviting developmental psychologists in the academic world to shoot me down. But their attacks have been surprisingly ineffectual. One traditional developmental [psychologist]even admitted, not long ago, that they still can’t prove that parents have any long-term effects on children. She continues to hope, however, that someday they will find the proof they are looking for — proof that can stand up to the scrutiny of skeptics like me. (See the best and worst moms of all time.)

    • enkiv2 says:

      I like the idea that one can “troll for truth”, but it seems like a dangerous technique, and one that requires you play the heel.

      In order for shouting an exaggeratedly extreme position to be a useful mechanism for producing nuance, you need to be *very* confident that the consensus is universally against you, and *very* confident that it shouldn’t be. Also, you need to not much care about your reputation (and in some cases, your physical safety). Good to have in the toolbox, but if you’re pulling it out frequently it’s a bad sign.

      • notsobad_ says:

        you need to be *very* confident that the consensus is universally against you, and *very* confident that it shouldn’t be

        I’m not sure I understand this. As Randy mentioned, she said she took an extreme position, though not necessarily one that was exaggerated or one she didn’t believe in. Even if this wasn’t the case, taking such a position would force experts to look at the evidence Harris (or someone else taking such a position) has presented and thus evaluate the current consensus and see how it holds up to the proposed theory. I’m not trying to say you’re wrong, I just don’t see why this is a “dangerous technique”. If the theory is bogus people will probably quickly identify it as so and move on.

        Unless you just meant it’s dangerous for the person’s reputation, in which case I agree and was just confused by your message.

    • Randy M says:

      Well, she says she argued an extreme position, but not one that she didn’t actually believe.

    • alwhite says:

      This whole discussion seems like a bait and switch and really sketchy.

      From the quote posted above:
      I took an extreme position: that parents have no important long-term effects on their children’s personalities.

      From a section Scott quoted above:
      To some extent at least, times have changed…there is now more acceptance of the idea that behavior is influenced by genes and that individual differences in behavior are due in part to differences in genes.

      We go from “parents have no effect” to “behavior is influenced” and “differences are due in part”; and this is grounds for “I was right all along!” Sketchy.

      • Desertopa says:

        The apparent conflict isn’t necessarily an actual one. The claim Harris presents is that parents have no long-term effect on the personalities of their children, but not that only genetics influence people’s long term behavior. The evidence distinctly does not support that position, about half of the variation in people’s personalities seems to be non-genetic, it just seems that parenting doesn’t contribute noticeably to that half.

        • wysinwygymmv says:

          Let’s remember in this context that we’re perfectly willing to throw out good scientific evidence as long as it points to a conclusion that looks intuitively unappealing.

          Children learn a lot while they’re quite young. They also spend almost all their time with their parents. They also tend to learn through observation and imitation. Is it intuitively plausible that these experiences would have virtually no impact on people’s personalities? Is it more intuitively plausible than the suggestion that PSY research is measuring something real?

          • Deiseach says:

            Is it intuitively plausible that these experiences would have virtually no impact on people’s personalities?

            Again, this is “do as I say, not as I do”. People who have door-slamming arguments who nonetheless teach their kids “You should always try to be calm and reasonable” are not alone passing on the genes for door-slamming arguments, they’re teaching by example: Dad says don’t shout and stamp your foot but he does exactly that.

            These positions can be summed up in folk sayings: “What’s bred in the bone will come out in the blood” – which is the “Genetics are everything” side – and “Little pitchers have big ears” – which is the “Blank slate/nurture” side.

            It’s a bit of both – people with grasshopper genes will find it very hard to settle down and grind away like ants, to use a metaphorical example, but also what children see and experience the adults and peers around them doing as against saying is what they take away as ‘you’re supposed to say this stuff but you don’t have to believe it or do it’.

          • INH5 says:

            There’s also the fact that the same logic used to argue for a lack of parenting effects would, if consistently applied, lead one to accept a number of patently absurd conclusions. After all, “shared environment” is, in theory, just any environmental factors that are shared between members of the same households, so if you want to dismiss it as unimportant, you have to dismiss a lot more than just parenting.

            For example, internet access. As late as 2014, 1 in 5 American households did not have regular internet access. So if you want to stick to the assertion that “shared environment has no significant effects on personality or life outcomes” you would have to conclude that the internet has no significant effects on personality or life outcomes. This in spite of all of the social changes in the past few decades that sure looked like they were caused by the internet.

            More generally, there’s the issue of reconciling this idea with changes over time. Over the past few decades, we’ve seen very significant changes in thinks like obesity rates, crime rates, and IQ (see the Flynn Effect; and it’s irrelevant whether it’s real or not, because the Flynn Effect is seen in IQ tests that are commonly used in twin studies and the like). These changes happened far too fast to be due to genetics, so they must be due to widespread changes in environmental factors. But how could these environmental factors, whatever they are, not vary between households and thus show up as a shared environment effect?

            I’ve brought this up before here and on the subreddit, and the usual response has been to insist that these changes must be due to some factor that varies across time but not across space. But I don’t see how that’s possible. Everything that varies across time varies across space, even if you restrict the range to, for example, the continental United States. The idea that whatever changes have increased IQ scores by 30 points of the past 100 years happened exactly the same way in both Manhattan and Lewisburg, West Virginia is beyond preposterous.

          • objectofclass says:

            @INH5,

            Re: Effects of unequal internet access —

            A 2013 experiment found no effect of having a computer in the home on any educational outcome for California middle-schoolers.

            “Even today, roughly one out of every four children in the United States does not have a computer with Internet access at home (NTIA 2011). While this gap in access to home computers seems troubling, there is no theoretical or empirical consensus on whether the home computer is a valuable input in the educational production function and whether these disparities limit academic achievement. Prior studies show both large positive and negative impacts. We provide direct evidence on this question by performing an experiment in which 1,123 schoolchildren grades 6-10 across 15 different schools and 5 school districts in California were randomly given computers to use at home. By only allowing children without computers to participate, placing no restrictions on what they could do with the computers, and obtaining administrative data with virtually no attrition and measurement error, the experiment was designed to improve the likelihood of detecting effects, either positive or negative. Although the experiment substantially increased computer ownership and usage without causing substitution away from use at school or other locations outside the home, we find no evidence that home computers had an effect (either positive or negative) on any educational outcome, including grades, standardized test scores, or a host of other outcomes. Our estimates are precise enough to rule out even modestly-sized positive or negative impacts.”

            (emphasis mine).

            This particular null effect of computer/internet access might sound patently absurd, but there it is.

          • Desertopa says:

            Personally, I find it very intuitively implausible, to the point that I still assign non-negligible probability to the prospect that parenting will turn out to have significant long term impacts in ways we haven’t yet assessed.

            But, I have to acknowledge that based on my intuitive model according to which it seems like parenting should have a significant effect, the data has been very consistently surprising. And I’ve had to modify my expectations very substantially in order to get better at predicting the results of these sorts of studies as they come out.

          • Tracy W says:

            They also spend almost all their time with their parents.

            That’s as young children. But if a child survives to puberty they are reasonably likely to outlive their parents. For their adult life, they normally will be dealing mostly with people who aren’t their parents.

            Plus parents have a strong genetic incentive to love their children regardless.

            So Harris’s argument is that children preferentially learn from their peers, not their parents. She cites cases of the toddler children of immigrants bringing home their peers’ accents from play groups, or insisting on only speaking English. Apparently one German lecturer’s daughter would speak English to him in America but German when they were visiting his family in Germany. She was treating her father like she saw her peer group treating their fathers.

        • sconn says:

          I just wonder if scientists (or perhaps I ought to say, the impression we have of the opinion of scientists) are quite right in equating “shared environment” with family and “non-shared environment” with school, peers, etc. I mean, within a single family, you can have a lot of variation in how kids are raised (hence differences based on birth order, family size, etc.). And on the other hand, within a single community, the families might not differ that much. Most middle-class white families raise their kids more or less the same way. Sure, there are small differences, like whether the kids are spanked once a month or never, or whether their parents actively help with homework or just double-check it every night, but there are still basic similaries like the values kids are raised with, the current expectation that parents help their kids in school, cultural beliefs about how much attention children should be paid and how much freedom they should have. And of course most adopted children are adopted into well-off, small families, which is a big commonality right there — it’s like not we randomly assign twins into massively different types of families.

          Of course I imagine scientists know all this. I recently read a book called Beyond Human Nature which was all about how nurture affects kids, but the first chapter admitted that basically all scientists understand that both nature and nurture have a role. What appears to be a schism in the field between “naturists” and “nurturists” is actually a very slight shift in emphasis because EVERYONE knows it’s both.

    • Deiseach says:

      People are more willing to admit that children can inherit behavioral quirks and personality characteristics

      I think this has to be read more as “[Professionals in the field of X Y Z] are more willing to admit that children can inherit behavioral quirks and personality characteristics”.. Ordinary people have always said “She has her grandfather’s temper” or “He reminds me so much of his Aunt Mary” but Educated Specialists liked to pooh-pooh that as “old wives’ tales” and “stuff only peasants believe”.

      I think often “lone maverick challenges hidebound reactionaries” is rediscovering common sense and what ordinary people have always thought 🙂

      • Conrad Honcho says:

        This was the same impression I got from the story about the meteorites. The peasants always thought meteorites were otherworldly. Chladni didn’t invent a new idea that the hidebound reactionaries squelched, he just proved what all the peasants already knew.

        Similarly on social justice issues. Anyone who’s ever had a child (or a family for that matter) could tell you there’s differences between boys and girls. It takes 10,000 pages of gender studies textbooks to “prove” there aren’t.

  2. Douglas Knight says:

    If you think those examples are bad, you should check out his inclusion of Thomas Gold (twice!) for ideas that are still considered crackpot. (I think Gold’s deep hot biosphere does not depend on abiotic oil and it is a pity he did not try to disentangle the ideas.) Freeman Dyson wrote a forward to Gold’s book in which he describes three previous times Gold put forward radical proposals. Once he was ignored and vindicated decades later (psychoacoustics), once he was mocked and quickly vindicated (pulsars) and the third time he was maybe embraced and wrong (steady state universe).

    Anyhow, Dyson is open-minded and smart, so compiling his endorsements is probably the best way of generating a list of promising crackpots.

    • Machina ex Deus says:

      But pulsars are evidence against the steady-state universe! So one of his big ideas killed another.

      (Not an astrophysicist, but: because there are a whole lot more very-red-shifted pulsars than somewhat-red-shifted pulsars, we know they were more frequent billions of years ago, which is hard to square with a steady state.)

      • Douglas Knight says:

        Are pulsars old? I wonder if you are confusing pulsars with quasars. As a historical matter, it was quasars that killed off the steady state theory; and this was years before pulsars were detected. Gold was trying to explain pulsars after they were detected (though others had predicted them), but before enough had been found to judge age.

        • The Nybbler says:

          I thought it was the cosmic microwave background that killed off steady state?

          • Douglas Knight says:

            Yes, CMB was more convincing than quasars. I’m not sure, but I think quasars were pretty convincing already. CMB was also before pulsars. Anyhow, my point was that Gold did not have a conflict of interest because (1) the theory was thoroughly dead; and (2) even if pulsars were evidence against it, they were already observed, and he was just trying to explain them.

      • seladore says:

        Astrophysicist here — Douglas Knight is correct, you’re confusing pulsars (rapidly spinning neutron stars, left behind after a large star dies) and quasars (very active supermassive black holes, bright enough to see in the very distant/early Universe).

        Quasars are mostly very distant, and their redshifts are difficult to square with a steady state theory.

        Most pulsars we can detect are relatively nearby (in our own Galaxy).

        • Machina ex Deus says:

          Yay! Science self-corrects, even on SSC!

          Yeah, I was definitely thinking huge-thing-that-radiates-as-much-as-thousands-of-galaxies, not little-spinny-thing-with-strange-signal. Second-grade me would be cringing at the misnomer.

          (I think it’s because a day earlier, I had read the Wikipedia article on quasars (no, I don’t remember why), so I was primed to see “quasar” when I saw the word “pulsar”.)

  3. JohnBuridan says:

    We can’t know much without others to help us, and scientific fields are actually pretty good at learning about the world.

    I’ve started emphasizing the “best tool we have” part of the sentence, and whispering the “isn’t perfect” part, rather than vice versa.

    Ah! I wonder if there is word for the warm sensation of hearing someone else voice your thoughts? Say it again: it’s the best we’ve got, but it isn’t perfect. Does this epistemic optimism reach to experts in other fields that are not strictly science? CIA analysis of foreign governments? Reviews on headphones? Police reports in court cases? Translations of literature?

    So much of our rationality and decision-making is dependent upon others, whether they be expert or not, that we essentially have little choice in the matter. Or is finding the right heuristic for determing who to listen to the important part? Who are the true experts anyway? The scientific community you speak of has it better than most, I should think.

    I have received my balance between epistemic pessimism and epistemic optimism from David Wallace’s essay Deciderization (which is pessismistic) and Peter Briscoe’s Reading the Map of Knowledge, which is giddy with the prospect of swallowing libraries and engaging in a lifetime of learning. I have a bibliotech-centric emphasis in the way I think about knowledge, thanks to Briscoe, and thanks to studying classical languages for too long.

    From Wallace, I guess it’s just sober, grim, near-despair that I’ll ever knowing anything. 🙂

    BTW, It’s fairly common in the arcane world of (Jewish, Anglican, Orthodox, RC) theologians (not mere religious people) to read the books of the Bible as an ever-evolving argument about who God is. So maybe you won’t anger as many theists as you think. I’ll cite some sources medieval and modern later, if anyone is interested.

    However, if what you say about expert opinion is correct, then we can predict that many non-expert theists would be upset, because they operate in the Bad Paradigm. Like Vox is to Scientific Concensus, the mechanisms which disseminate theology are flawed and may be much worse. The trick is to not treat Scientific Consensus as the target, but as a means to Truth. For believers, the trick is not to treat Scripture, ecclesiastical authority, or whatever as their target, but as their means to knowing and finding God.

    Does that make sense? It’s late.

    • Vermillion says:

      BTW, It’s fairly common in the arcane world of (Jewish, Anglican, Orthodox, RC) theologians (not mere religious people) to read the books of the Bible as an ever-evolving argument about who God is. So maybe you won’t anger as many theists as you think. I’ll cite some sources medieval and modern later, if anyone is interested.

      I’d be interested sure.

  4. srconstantin says:

    I think science is great.
    I also think that when science works it requires independence.

    There’s really *not* a lot of great scientific “mavericks” in the strong sense of being ridiculed by everyone and taken seriously by no one and having no connection to the scientific community. I agree with you on that.

    I also believe that wisdom-of-crowds is a real and significant effect, and that good discussion between well-informed people amplifies this effect. I’ve seen various forms of evidence that teams do better than individuals at problem-solving, or that “scenes” full of luminaries are responsible for spikes in scientific achievement. Most geniuses are not lone.

    I also believe the conventional scientific account of pretty much every topic outside the social sciences and philosophy. I don’t have firsthand evidence stronger than the books and papers I’ve read.

    Where I think there’s a little more nuance is that to the extent “Science” or “The scientific community” is doing stuff that works, they are doing things other than submitting to the authority of the scientific community. Science works precisely because scientists are not submissive!

    It is only thanks to scientists (credentialed and respected scientists, not outsiders) that we know about the replication crisis; on the other hand, public credulity of “studies show that” allowed unreplicable studies to proliferate and make the news and get funding.

    There are bubble-makers, bubble-growers, and bubble-poppers in the world of ideas, just as all institutions are governed by forces of creation, preservation, and destruction. An idea is a speculative commodity. Creators come up with ideas; some good, some bad. Preservers run with those ideas, attract attention to them, get funding for them, do follow-up work, etc. Destroyers debunk them. The problem with science, as with so many other present-day institutions, is that Preservers have lots and lots of leeway and power in our current society, and it’s out of balance relative to Creators and Destroyers. Preservers inflate bubbles in scientific fads. Preservers get the credit that’s due to Creators and dilute their ideas. Preservers write tons and tons of shitty studies.

    To use your examples:

    1. Ioannides and Simonsohn are Destroyers, clearly. They’re good guys in my book, and because science isn’t a total mess, researchers listened eventually.

    2. Taubes is a Creator, and like many Creators he’s flawed; that’s how it goes. The nutrition industry (and associated government agencies) have been extremely slow to update to the new research; they’re classic Preservers.

    3. The doubt cast on the IAT and stereotype threat is a great example of Destroyer virtue in action; but *people still believe in the IAT even when I tell them that it isn’t predictive and they nod along.* The Preservers are winning on sheer momentum.

    4. Judith Rich Harris is a Creator who won. (The many follow-up studies on behavioral heritability are due to Preservers following their proper function; though I expect that they will, or already have, overshoot their case and we’ll eventually have a new Not Everything Is Genetic debunking.)

    5. Nick Bostrom and Eliezer Yudkowsky are Creators whose work is being picked up (and significantly altered) by Preservers. I’m somewhat suspicious of where the Preservers are going with this.

    6. IQ: this is a case where the science is actually just fine IMO, the Preservers are doing their job of replicating very reliable findings.

    I suspect that because science used to be a much smaller endeavor (by # of people and total funding), the methods that evolved for developing scientific consensus were not adapted to the presence of swarms of Preservers. Lots of Preservers mean that fads can inflate for a long time before deflating, and influence public policy and industry quite a bit before being debunked.

    My ideology at the moment, in science as in other things, is “Bullish on Creators and Destroyers, bearish on Preservers.” This is quite different from the stereotypical misconception that “the scientific establishment” excludes maverick geniuses.

    • Ozy Frantz says:

      It’s totally legit to not believe in “scientific consensus” on philosophy, because the PhilPapers survey suggests that the only thing there’s really consensus on in philosophy is that the external world exists. (That’s if you define “consensus” as “more than 80% agree”; if you loosen it to “more than 70% agree”, you also get atheism, scientific realism, and the existence of a priori knowledge, all of which also seem pretty reasonable to me.)

      • Steve Sailer says:

        Philosophy exists to have good arguments. Things that become less arguable stop being philosophy over time and turn into a science.

        • onyomi says:

          Can you point to some historical examples of this process?

          • Steve Sailer says:

            Most of the pre-Socratic philosophers wouldn’t be considered philosophers today, but speculative physicists and cosmologists.

            The majority of what Aristotle wrote about wouldn’t be considered philosophy today, but instead biology, physics, political science, literary criticism, etc.

            Another example: Claude Shannon first heard about Boolean logic in a philosophy class. He then pointed out in 1937 that Boolean thinking could be applied to electronic circuits, with important consequences. There’s a good article in The Atlantic by venture capitalist Chris Dixon on how philosophy laid the groundwork for the computer:

            http://takimag.com/article/diversity_versus_debate_steve_sailer/print#axzz4eUb453nU

          • manwhoisthursday says:

            Actually, a lot of the concerns of the Pre-Socratics were purely philosophical. They were interested in questions like: Do things change? Does anything in the world have a stable identity? Should we primarily trust reason or sense experience? Is mind or matter the fundamental basis of reality?

          • onyomi says:

            @Steve Sailer

            Isn’t this just a more general phenomenon of specialization in learning? By this logic, literary criticism, which Aristotle also did, used to be philosophy.

        • Philosophisticat says:

          I wouldn’t put it quite this way. I think philosophical problems are those where how to think clearly about the topic is part of what’s at issue. Once there is a consensus reached on how to think clearly about something, then it tends to fall off the tree into the sciences.

        • wysinwygymmv says:

          Overly simplistic model. As one example, the list of interpretations of quantum mechanics has tended to get longer rather than shorter.

      • Eli says:

        I’ve never understood how they think a meat-brain that runs on approximate causal statistics is supposed to have a priori knowledge.

        • If you know how to make statistical inferences, thats knowledge. The hard-to-defend versiin of apriorism involves having certain descriptive knowledge. The easy-to-defend versions lower the bar and broaden the scope on what counts as knowledge.

    • srconstantin says:

      For a (minor) example of a situation where my personal experience directly conflicts with scientific consensus, the big studies about St. John’s Wort say it works about as well as other antidepressants, but the anecdotes I’ve heard are nearly unanimous in saying it doesn’t work at all. This makes me somewhat skeptical of the science, but I expect the more likely result is something that makes both the scientists and my acquaintances right, like “maybe the St. John’s Wort you get in US drugstores is very different from the kind prescribed in Germany, where the studies come from.”

      • Nornagest says:

        I have heard, but cannot verify, that the St. John’s Wort sold as a supplement often has very little actual St. John’s Wort in it, or much of any active ingredients. A controlled study presumably wouldn’t have that problem.

      • cuke says:

        For what it’s worth, anecdotally, we had a very anxious traumatized dog who responded dramatically to St. John’s Wort (dogs are often put on prozac for same symptoms). The effect though seemed to poop out after six months, which of course is common for SSRIs as well. I know two people personally who had similar positive initial responses to St. John’s Wort that also then petered out. At least in our dog’s case, it was unlikely to be placebo effect.

      • nydwracu says:

        I started taking SJW (lol) because my dad said it worked for him. It seems to have worked for me. I tried it once before with a different brand and got nothing out of it, so it’s possible that it’s confounded by the cartoon-villain market in herbal supplements.

  5. Sniffnoy says:

    This seems like less of a problem with expert consensus, and more of a problem of nobody else (including textbook writers!) listening to experts who are continually trying to beat reality into people’s heads.

    My understanding is that this sort of thing, textbook writers and all, is pretty common in general…

    • John Colanduoni says:

      And it doesn’t just happen when people have strong feelings about what they’re writing about. Physics textbooks can be pretty impressively awful, and I have trouble believing that the writers think things like “The so-called fact that the distance between Earth and some galaxies is increasing faster than the speed of light is a dirty, dirty lie and is why you don’t see many galaxies in physics programs!”

      • Deiseach says:

        Textbooks don’t get overhauled often enough, and I say this as someone who thinks the monopoly by certain publishers on textbooks for the set curriculum in Ireland is iniquitous in how they churn out ‘new version! you absolutely must have! no you can’t pass on that three year old textbook to your younger sibling!’ books every year.

        The trouble is the time it takes between ‘intriguing new paper gets published and discussed’ to ‘idea becomes accepted enough to be standard’ means textbooks will always lag behind and even worse, the information you learned in your school [whatever] class, and even your university [whatever] class, is going to be out of date before you even know it.

        • Yosarian2 says:

          This is one of the reasons Is think maintaining the strong connection between “higher education” and “the University as an institute of research and creating new knowledge” is valuable and worth keeping, even with the disadvantages (some researchers may not be great teachers, ect).

    • Brad says:

      Keep in mind that high schools in the US generally hold on to textbooks for six years. And they don’t always buy books that were just published the same year.

      It’s unreasonable to expect something that’s seven years into a ten year ridicule-to-consensus process among cutting edge scientists to show up in most in use high school textbooks. Even thirteen years in is pretty aggressive.

  6. doubleunplussed says:

    I’ve wondered why Pinker seems to get away with things that get others crucified. I’ve started to think it might just be that he’s eloquent, handsome, and doesn’t get defensive.

    Also, I’d love to see some updated surveys for interpretations of quantum mechanics. Preferably among physicists who work with open quantum systems or quantum computing (the rest have so little interaction with the measurement problem that it is possible for most of them to not really have thought about it).

    The latest I can find is that 18% of surveyed said many worlds in 2013:

    https://arxiv.org/abs/1301.1069

    (42% said Copenhagen, which as far as I’m concerned really means “I don’t know, haven’t thought about it, go away”).

    And 8% of surveyed said many worlds in 1997 (https://arxiv.org/abs/quant-ph/9709032).

    I feel like recent advances in experimental techniques are forcing people to think about the measurement problem more and more, and the only way I see many worlds not winning this fight is if actual new physics is discovered that rules it out.

    But if it wins out it will have been a long time coming. Everett wrote his thesis in 1957.

    • JohnBuridan says:

      Totally agree about what you say concerning Pinker: handsome, nondefensive, has a nice aura. He’s always seemed a little ‘too’ institutionally liked to me. But maybe it’s just because I didn’t think The Blank Slate was any good (it was the first thing I ever read by him, huge mistake). He writes other stuff which is more precise and more in keeping with his field and worthy of respect.

      • pedrodegiovanni says:

        I really liked The Blank Slate. Could you elaborate on why you think it was a mistake? I’m curious.

        • Enkidum says:

          So I haven’t read The Blank Slate but I’ve read various of his other works, work in an adjacent field, and have a fairly good knowledge of the overall issues, I would say. Here’s my basic problem with the line he often takes: he’s glib, misleading, and spends far too much of his time battling strawmen.

          Exhibit A: In The Language Instinct (which blew my mind when I first read it, but after learning something about linguistics I think it may be virtually entirely wrong), Pinker dismisses multi-layer neural nets as models of linguistic generation (now referred to under the fancy banner of Deep Learning that seems to have gotten everyone excited again) with a single example from one paper in which they produced a gibberish sentence. I really hope, in retrospect, he’s embarrassed about that.

          Exhibit B: A huge problem for the view (dominant among many of the commentators here) that the limp-wristed SJWs and social scientists are ignoring the obvious truths about genetic causes for behaviour, which has been filtered down through Pinker, is that the single most respected left-wing intellectual in the world (who happens to be Pinker’s graduate advisor) is also the advocate of a purely genetic theory of language origins (although he is, to be fair, completely ignorant of genetics), which was the dominant theory in linguistics for at least forty years! It’s not the dominant theory any more, probably for the simple reason that it has never had a very clear and specific articulation, and whenever anyone does articulate it, there are inevitably numerous counter-examples brought up. I haven’t met a Chomskian linguist under the age of 45, though I’m sure there are plenty out there (IANAL).

          Similarly, the dominant left-wing consensus since the late 80’s has been that sexual orientation (and, more recently, gender expression) is entirely genetic (or due to epigenetic effects in the womb). The fact that this doesn’t match up with the numerous population studies on the issue, which all show 10-40% of variance in sexual orientation being explained by simple genetic factors (which, for those of you paying attention, means that 60-90% of the variance is NOT explained by simple genetic factors), is irrelevant. (It doesn’t help that some of the authors of said studies are among the worst mis-representers of what they actually show.)

          Pinker pays lip service to the correct view, which is that virtually everything interesting about human beings is a complex interplay between genes, epigenetics, and learning. But I can’t think of a single case where he’s given the third factor its due.

          • Enkidum says:

            Just wanted to add that I am far too anti-Pinker in the above. He’s someone who has genuinely influenced me, he’s a great writer, has done some good science, and I respect him a lot (and I’m more or less on board with his account of, e.g., human progress). I just happen to think he has some fairly massive blind spots.

          • Deiseach says:

            Here’s my basic problem with the line he often takes: he’s glib, misleading, and spends far too much of his time battling strawmen.

            Sums up why I dislike him and can’t understand why he’s one of the experts always approvingly trotted out by articles like the one in Vox. Suppose that explains it, though!

          • Deiseach says:

            Similarly, the dominant left-wing consensus since the late 80’s has been that sexual orientation (and, more recently, gender expression) is entirely genetic (or due to epigenetic effects in the womb).

            I’m dubious as to how much of that is consensus about the science, and how much is “it is a politically useful tool to get the straights on board; it’s way easier to sell them ‘born this way’ and so can’t change it anymore than you can change your eye colour than ‘orientation is fluid and malleable’ which has them saying ‘so why can’t you hack yourself to be straight then?'”

            I’m seeing some signs of “yeah yeah, we had to say ‘born this way’ and ‘it’s in the genes’ before we got legal protection but now we’ve got gay marriage we can drop that angle and say ‘I choose this and I wouldn’t be straight even if I could be'”.

          • MugaSofer says:

            “Genetic” doesn’t mean genetic in this context, it means “makes discrimination somewhat justified/logical”.

            Limp-wristed liberal elites would have no problem if Chomskyan linguistics turned out to be true; it would just be an interesting fact about human neuroanatomy. But it would cause an uproar if it turned out that Europeans were genetically better-suited to speaking English.

        • JohnBuridan says:

          Hi pedrodegiovanni,

          Essentially, I found at the time that I read it that he always seemed to be refuting super lame arguments which I had never heard of before (strawmen, as Enkidum and Deiseach say) and he fails to be clear, concise, and, didn’t explain himself or cite well in places where I didn’t understand what he was trying to say, and overexplained things I thought were super basic, even to luddites and psych sceptics.

          Plus his writing style is boring.

          BUT… paging through the book seven years later, maybe I can understand where he is coming from better and give him a more sympathetic reading. On the other hand, his writing style in the Blank Slate still hasn’t improved, neither concise enough to be a tour de force, nor emotive enough to be a fun popularization. It’s like he forgot that he needs to be build tension in his presentation of ideas and research, keep building up both sides, then, at last, resolve it. Instead we get sentences like this one about society’s supposed rejection of psychological accounts of behavior.

          Something has gone terribly wrong. It is a confusion of explanation with exculpation. Contrary to what is implied by critics of biological and environmental theories of the cause of human behavior, to explain behavior is not to exonerate the behaver, (pp. 179).

          What’s with the italics, man? And further, since when do I (or anyone) get my ideas about determinism from Hilary Clinton and West Side Story?

          Okay, perhaps I’m being a bit disengenuous… but Pinker’s not my cup of psychological tea.

      • Scott Alexander says:

        Pinker is my Exhibit A in what I was saying last week about not abusing free speech. By very carefully picking his battles, saying inherently-taboo things in as measured and inoffensive-sounding a way as possible, and making sure he never sounds like he’s deliberately looking for a fight, he’s able to get away with doing basically whatever he wants.

        I think if other people copied his style, they could copy his success too, which makes me especially angry when they go more towards being as provocative as possible, especially since that makes life harder for Pinker-types.

        • Steve Sailer says:

          And don’t forget: And, if necessary, make public examples out of people like Malcolm Gladwell who try to get you Watsoned.

        • Steve Sailer says:

          But, Scott, in my experience, the public intellectual most like Steven Pinker is Charles Murray, whom you said last week was exactly the wrong person to invite to speak at Harvard.

          More generally, telling people to be more like Steven Pinker is a little like telling football players to be more like Tom Brady and win five Super Bowls. Don’t be like Peyton Manning and win only two, much less that loser Tony Romo who never won a Super Bowl. If you want to be a Trump supporter in Boston, for example, just be Tom Brady and you won’t get much guff.

          Pinker is superb, and thus he gets treated well. Very, very few of us could be like him no matter how hard we tried.

          • Steve Sailer says:

            Here’s an old piece I wrote in 2009 during the Pinker-Gladwell NFL QB draft controversy on the philosophical issues behind the appeal of the question “Who Would Win In a Fight: Tom Brady or Peyton Manning” using some ideas Pinker gave me on why we’re so attracted to arguments with no clear way to prove who is right:

            http://takimag.com/article/quibbling_rivalry/print#axzz4eUb453nU

            This is related to the topic of scientific consensus. Most things that there exists a scientific consensus upon isn’t argued over — e.g., leaf through a chemistry textbook — because it would not be interesting to do so. In contrast, football fans argued for 15 years over whether Peyton Manning (a #1 overall draft pick) or Tom Brady (a sixth round draft pick) were better because it was really fun to argue over because they were both so good, but in somewhat different ways.

            That, by the way, was tied into Gladwell’s mistake behind his notorious argument, which Pinker ripped him over in the NYT, that nobody can predict who would be a better NFL QB at the time of the NFL draft.

            I had gone and looked up a lot of NFL data that showed the QB draft prediction glass was part full and part empty. Higher draft picks did better in the NFL on average, but there were a lot of exceptions. Gladwell was remembering the much-celebrated sixth round picks like Brady who became superstars, but he was forgetting all the QBs rated as poor prospects coming out of college who turned out to be just as bad as expected.

            The reality is that there is a lot of boring stuff in the universe that is boring because it turns out just the way people expect.

          • wysinwygymmv says:

            But, Scott, in my experience, the public intellectual most like Steven Pinker is Charles Murray, whom you said last week was exactly the wrong person to invite to speak at Harvard.

            What if you read and acknowledged the actual argument Scott made instead of hammering on this straw man?

          • hypnosifl says:

            I don’t see any examples of Pinker inserting opinions into his work that are both a) not really supported by the evidence he’s presented and b) very likely to be seen as bigoted in the absence of such strong support, like Murray and Hernstein’s comment in The Bell Curve that “It seems highly likely to us that both genes and environment have something to do with racial differences.” The context makes it clear they are talking about racial differences in IQ, and it also seems reasonably clear that they are talking about a genetic contribution that’s large enough to play a significant role about our explanation of the observed IQ gap, rather than something negligible like the environment being responsible for almost all of the ~15 point gap observed in the U.S. while genetic differences contribute a difference of some fraction of a single IQ point. (Also note that if there is a tiny genetic difference that is swamped by the environmental difference, there’s not even any strong reason to assume the genetic difference would go in the same direction as the environmental one, i.e. blacks in the U.S. might have an genetic 0.5 point advantage over whites but the environment would subtract 15.5 points, which in Bayesian terms should be viewed as only marginally less likely than a scenario where blacks have a genetic 0.5 point deficit and then environment subtracts another 14.5 points.)

          • Desertopa says:

            But, Scott, in my experience, the public intellectual most like Steven Pinker is Charles Murray, whom you said last week was exactly the wrong person to invite to speak at Harvard.

            Scott was very emphatic, and I thought very clear, about this. It’s not that he was arguing that Murray was a bad person to invite to speak at Harvard, and indeed he explicitly stated that Murray could be a good person to invite to speak, given the right context. The problem was that they were inviting him to speak on the basis of his controversiality. The process by which he was invited is harmful to his ability to speak persuasively.

          • Steve Sailer says:

            And as I pointed out, among Scott’s many strengths, being a cunning Machiavellian public relations strategist probably isn’t one of them.

          • suntzuanime says:

            I don’t think he was especially clear. In one breath he says he has no problem with Murray per se, in the next he analogizes Murray to a hideous and deformed axe murderer. The article wasn’t very coherent, in part I think because Scott wants contradictory things

          • Baja Roki Thompson says:

            The question is, what is Pinker doing differently to Murray to not get listed as a Controversial speaker? Or is it just a matter of luck?

          • Conrad Honcho says:

            @Baja Roki Thompson:

            According to Scott, not speaking for the sake of being a controversial speaker.

            (For the record I disagree with Scott, but I believe that’s the answer to your question.)

          • Baja Roki Thompson says:

            It’s not like he only became controversial with this booking, he ddin’t conciously work to get such status. So it doesn’t quite answer it.

          • Steve Sailer says:

            I wouldn’t be surprised if before Pinker’s friend Larry Summers got in so much trouble in 2005 for expressing Pinkerian views on sex differences in the variance in the IQ bell curve, people who paid careful attention wondered how Summers got away with obviously not being a True Believer in the conventional wisdom. They probably attributed it to Summers deserving not to get in trouble because of various subtle reasons they dreamed up about how he followed the precise Rules.

            In reality, there is a lot of randomness in whom the Mob happens to focus their wrath upon. But we like to believe that the universe is just and the unlucky deserve their fates.

          • Hyzenthlay says:

            I don’t think he was especially clear. In one breath he says he has no problem with Murray per se, in the next he analogizes Murray to a hideous and deformed axe murderer.

            I understood what he was trying to say, the argument was just needlessly muddled by using Murray as an example, because his problem was with the motives (or assumed motives) of the people who chose him to speak rather than with Murray himself.

            He also mentioned Milo in the post; if he’d just used someone like Milo as an example to begin with, it would have been a lot clearer.

          • suntzuanime says:

            But that would have undercut his argument, because the people specifically declined to invite Milo, specifically on the grounds that he was too deformed and axe-murderous for them. He would have had to change his thesis to “good job, Harvard Free Speech Club, for your careful stewardship of our sacred principles”.

          • Conrad Honcho says:

            Which ultimately comes down to your opinion of the state of free exchange of ideas on US campuses. If you think it’s fine, then yes, being needlessly provocative for the sake of being provocative is poor form. However if you think speaking arguable truth on controversial issues will get an angry mob attacking you, then you’re not so much provoking as you are drawing the enemy out into the open.

            Scott, what do you think would happen if you took one of your essays on HBD or feminism and presented it on campus? Or if you turned one in as a homework assignment in a typical sociology class? If the answer is something other than “a fair hearing of the ideas and reasonable debate” then perhaps all is not well on US campuses.

            This is also my problem with Scott’s How The West Was Won. If a defining feature of the Universal Culture is lack of censorship, then the dominant culture of western education and media is not the Universal Culture. It may not be government doing the censoring by force, but the campus Code of Conduct, the HR department, the mobs, the media punish you to the point of exile from society for saying true but verboten things about race or sex.

            This is not a high entropy state. This is a low entropy state that requires enormous amounts of energy to be pumped into the system in order to enforce gender norms, diversity standards, and speech codes in all aspects of society.

        • Steve Sailer says:

          Here’s a 2015 tweet by Pinker:

          “Irony: Replicability crisis in psych DOESN’T apply to IQ: huge n’s, replicable results. But people hate the message. goo.gl/Vu03BV ”

          https://twitter.com/sapinker/status/645301814955388930?lang=en

          • Steve Sailer says:

            One interesting question is whether the conventional wisdom would be quite as hostile toward IQ if The Bell Curve’s co-author Richard Herrnstein hadn’t died of cancer about the time of publication, leaving Murray, a small town Midwesterner, as the sole face of the book. Having the big city boy Herrnstein around as well to promote The Bell Curve would likely have been more reassuring to opinion-molders. In contrast, Murray by himself somehow seems to bring out prejudices about peasants with pitchforks in urban intellectuals.

          • rlms says:

            The argument from my opponents are probably prejudiced is bad when the prejudice is racism, and equally bad when it’s ruralphobia.

          • hypnosifl says:

            Saying that IQ is predictive of a number of things including academic success, and that it’s highly consistent across multiple tests, is not really an opinion most would offensive even if they disagree with it (plenty of prominent people on the left would agree with it). It’s only when you start claiming that ethnic IQ differences have a genetic origin that you are likely to be seen as a bigot by many.

        • Ozy Frantz says:

          It seems to me that there are actually a very small number of issues that people get pissed off about (psychological racial differences, things that imply that rape is okay, saying that present-day infanticide is okay, trans people’s pronouns) and if you manage to avoid those issues you can have a really long and successful career being as anti-SJ as you please. Pinker seems to be really good at this.

          • wintermute92 says:

            This seems broadly true, but gender differences are the complicating issue. They’re some strange and delicate combination of widely acknowledged and a terrifying third rail, so Pinker’s success talking about those topics without taking excess flak seems interesting and much harder for others to replicate.

            (I’m not, and not qualified for, referring to any specific claim – I’ve just noticed that the topic lacks an easy distinction between “this is fine to say” and “this is incredibly inflammatory”.)

          • John Schilling says:

            It seems to me that there are actually a very small number of issues that people get pissed off about

            Perhaps, but it is an ever-changing small list, and one that seems to correspond with the leading edge of the Overton window’s leftward shift. If the idea is that everybody will be polite so long as we don’t dispute Team Social Justice’s claims this week to the territory they are fighting over this week, then no, not good enough. Those discussions may piss people off, but they absolutely are going to happen before that territory is ceded to your tribe.

          • Ozy Frantz says:

            John Schilling: That really doesn’t seem true to me. Sure, Murray gets protested, but evolutionary psychology is a thriving field with a lot of tenured academics, which often examines sex differences. Robert Trivers, Leda Cosmides, and David Buss do not seem to face the opprobrium Murray does. Similarly, William Julius Wilson (to pick one example) has gotten 45 honorary degrees in spite of his tendency to write books claiming that class is more important than race in explaining the outcomes of black people, and that culture too plays a role.

            I don’t mean to erase the suffering experienced by academics who would quite like to research racial IQ differences or argue that it is a good idea to kill babies, but honestly it does look to me like there’s a small set of consistent (if nonsensical) taboos.

          • vV_Vv says:

            @Ozy:

            I had never heard of the trans pronoun thing until ~5 years ago, and until maybe 2 years ago it was just a fringe stereotypical blue-haired snowflake thing.

            Now Canada is trying to criminalize not calling people “ze”, and Jordan Peterson gets screamed at and shut down for taking a stance against it, and Scott writes posts arguing that PetersEn is too extreme and controversial to be made a banner for free speech. Too extreme and controversial for saying what everybody believed until 5 years ago.

            The SJWs are definitely pushing the Overton window.

          • Urstoff says:

            To add another example to Ozy’s list: Geoffrey Miller seems like a dumb person’s caricature of an evolutionary psychologist, but his academic career seems just fine.

          • Steve Sailer says:

            Pinker’s friend Larry Summers got into massive trouble for saying Pinkerian things about sex differences in the IQ bell curve. Pinker bravely came to Summers’ defense.

            Part of the difference is that Harvard presidents aren’t tenured.

            But a lot of it is just random.

            Interestingly, Obama later gave Summers a great job, in part because Obama doesn’t have much if any intellectual respect for feminism. But almost nobody noticed that.

        • Deiseach says:

          I think if other people copied his style, they could copy his success too

          That does raise the question, what does his success consist of? If it’s “our favourite pet scientist”, that still leaves him hostage to needing to keep the good opinion of the public/society at large, and it still hobbles ‘inconvenient truths’ which will upset people if they express them.

          Pinker-style success could also be represented as “don’t rock the boat” and sometimes the boat needs to be rocked, else it’ll run aground.

          • Steve Sailer says:

            I think it’s a common assumption among the conventional minded that since Pinker is obviously enormously intelligent, he can’t possibly be on the Wrong Side.

            It’s like how Mike Judge is so brilliantly satirical that he can’t possibly be on the side of Hank Hill, even though he voiced him for a dozen years.

        • Steve Sailer says:

          Here’s a new tweet by Pinker:

          Steven Pinker‏Verified account @sapinker 11h11 hours ago
          More
          Embarrassing illiteracy, ignorance, & tyrannical arrogance of Claremont students, exposed by Heather MacDonald

          https://t.co/oGOpmRhijd

      • Hmm, I’d started The Blank Slate but put it down for having too low a ratio of fact to polemic. I’ll look at some of his other work.

    • Steve Sailer says:

      Pinker has the best combination for a heretic public intellectual: he seems nice and Pinker fights back devastatingly. Look at what happened to Malcolm Gladwell’s reputation after he tried to bring Pinker down in 2009 by complaining that Pinker had cited evil me’s research on NFL draft picks.

      • ADifferentAnonymous says:

        Do you have links to support this? As far as I can tell, most of the damage was done by Pinker’s initial review, which wasn’t retribution for anything.

    • pamape says:

      (42% said Copenhagen, which as far as I’m concerned really means “I don’t know, haven’t thought about it, go away”).

      This sounds very ignorant, simplistic and naive, and if you aren’t deeply familiar with QM that’s a very bold statement to make. I don’t know much about QM, but I’ve read a lot of pro-Many Worlds and anti-Many Worlds arguments and so far I haven’t seen anything that would undeniably support one interpretation over the other since it’s close to impossible to test those arguments (at least so far).

      Of course it has its problems, but one big point in favor of it would be the fact that most physicists find it the most useful to use when they are they trying to solve real research problems .

      • Harry Maurice Johnston says:

        I don’t know if this qualifies as “deeply familiar”, but speaking as someone with a PhD in Quantum Optics, that phrasing sounds pretty much spot on to me. (Caveat: said PhD was a long time ago and I am not a professional physicist.)

        … that’s kind of what makes it the “most useful to use”, actually, to the extent that the phrase makes any sense. Just doing the mathematics and not worrying about what it means is typically a very practical approach to getting your work done.

        • pamape says:

          Okay, thanks for clarifying it. But I still find it hard to believe that career physicists wouldn’t have thought about what their work means at any point during their careers.

          • doubleunplussed says:

            I think you’d be surprised. Most physicists aren’t affected by the measurement problem in their day to day work, and since Copenhagen is what’s taught in undergrad, they can be under the impression that there is no mystery there. Most physicists specialise a lot, and can go for long periods of time using specialised tools to make progress in their subfield without having to grasp the fundamentals of their own field. And there is even something of a taboo against thinking too much about foundations (people are fond of saying they adhere to the “shut up and calculate” interpretation).

            For the record, I’m an experimental physicist who works with open quantum systems. Most of my colleagues I don’t think have put much thought into the measurement problem. The best ones have, but most just go about their daily work without it affecting them.

          • smocc says:

            I’m a (early) career physicist in high energy / particle theory. My field is inherently quantum, everyone in my department uses quantum tools every day, and still none of us thinks hard about interpretation issues.

            The thing about interpretation questions are that, for a large part, it really doesn’t matter which one you take. No question I have ever asked in the course of my research has ever depended on whether MW or whatever else is correct. Since the question is hard, the predictions are ostensibly the same, and there’s no obvious way to make progress on it, but I can still use quantum mechanics to answer other questions, why would I think about it?

            For the question to become relevant you have to be either working directly on the problem of interpretation, or on one of a very few fields like doubleunplussed’s field of open quantum systems (and even then I imagine you can ignore the deepest questions if you want).

          • pamape says:

            doubleunplussed

            Sorry about that “ignorant, naive, and simplistic” part, I really didn’t know about your background.

            Although despite what you said, I’m still really puzzled.

          • doubleunplussed says:

            Oh, not at all, I am just some random person on the internet.

            If you’re puzzled because you didn’t expect physicists to think so little about their own field, well, that was a surprise to me too. But it really is not relevant most of the time. Formalisms get wrapped up in neat packages by the previous generation and we can just use the equations without needing to understand where they come from. I like to understand the foundations of as much as I can, but the time spent doing so trades off against doing actual new science, so I understand why people don’t bother.

            If you’re puzzled because of how dismissive I’m being of the Copenhagen interpretation, well, yeah, I really don’t think it’s much of an interpretation. It basically claims the macroscopic world and the microscopic world are different, and their interactions (“measurement”) require different laws of physics than what’s used to describe either of them separately. This may have been a good placeholder for when we didn’t understand those interactions well. It’s the sort of thing I’d describe as a “phenomenological” description of what’s going on. But I feel like it always should have been treated as a placeholder for the real description. Decoherence and many worlds explain why it looks like that interaction between macro and micro objects is special, even if it’s not. And whilst we haven’t ruled out that it actually is special, neither do we have any evidence that it is, now that we can explain how the illusion arises. So it looks like the same rules (quantum mechanics) can be used to describe all systems, big and small.

            I wouldn’t bet much on many worlds being actually *correct*. There is a large part of parameter space we have not explored, and new physics may be lurking there that could rule out the macroscopic superpositions that many worlds gets its name from. But the key is that that would be *new physics*. The current best supported theory predicts many worlds, we’ve shown that the patch that was added long ago in Copenhagen to make quantum mechanics agree with classical mechanics at large scales is no longer necessary, so keeping it around without any further evidence of it is not really very defensible. It also doesn’t matter, because positing that the collapse just happens at some scale larger than we’ve been able to check will always agree with many worlds. But it’s a god-of-the-gaps situation, and just because you can always adjust a theory so as to not let it be ruled out is no reason to actually believe the theory.

            Another analogy: general relativity predicted the existence of black holes long before we could observe them. People may have been tempted to “patch” general relativity to remove the black holes, since they were unobserved. But since not observing them was also totally consistent with them existing (the prior for observing them if they existed was low anyway), I’d be advocating for not adding the patch. I’d say “general relativity predicts black holes”, and people would label this an “interpretation” of general relativity and call their patch another “interpretation”. It would be a weird situation, but that’s the one I think we’re in in quantum mechanics. There’s a hard to observe thing the theory predicts, and people want to add a patch to make the theory rule it out. In the absence of a non-insignificant Bayes factor for this phenomenon existing or not, adding the patch just increases the complexity of the theory, and doesn’t make it any more right, because we can’t observe this thing (yet) anyway. Occam’s razor says don’t patch.

            So collapse might exist, but there is no evidence for it. So for the time being I choose to “interpret” quantum mechanics in a way that has no collapse. If we discovered evidence of collapse, that would be a new theory as far as I’m concerned, not a vindication of the old one, which there was never any reason to believe. It would be like discovering Thor and thinking it vindicated your belief in Zeus. Nope, if you were right by coincidence, you weren’t really right.

      • needed to listen to what real scientists thought. And again, when I looked into it, there was no consensus against the idea (of AI risk)

        How can there be when there is more than one idea? No one is going to dispute that automatic weapons systems are dangerous, for instance. Add an X to risk and its another story. …

    • leoboiko says:

      I feel obligate to point that Pinker is… not exactly respected (← wild understatement) in my corner (humanities/liberal) of the intellectual landscape (try searching for Pinker in the “bad philosophy” subreddit).

      • Urstoff says:

        He doesn’t seem to be persona non grata in philosophy, as far as I can tell. But I don’t swim in the political philosophy / ethics circles, only in the M&E type areas.

    • ti4 says:

      Just curious, in what sense can the interpretations be right or wrong? If it doesn’t matter for calculations, how can you decide which one is right?

      • Enkidum says:

        Also not a physicist, but I think most people believe that things are true or false independent of whether or not we have the ability to judge their truth or falsity. So we may not be (yet?) able to decide on an interpretation of QM using a clear-cut experiment. But that doesn’t mean that these interpretations are meaningless.

        • ti4 says:

          I guess my question is if this is only a philosophical discussion or if there is in principle possible to determine if one of these interpretation is wrong. To me this is an ocean of difference.

          • Enkidum says:

            Sorry, I’m confused by the clarification. I think saying it is in principle possible would mean that it is likely a purely philosophical issue – i.e. one which we do not have the appropriate practical ability to settle.

            But I think you’re asking if we have the actual, practical ability to settle it. To which I think (though again, I’m not a physicist) that the answer is presently “no” (although this may of course change).

          • ti4 says:

            Sorry, maybe I was not very clear. With a philosophical question I mean “not being able to decide ever” and with being able to decide in principle I mean “we could come up with some experiment in the future that decides this but we don’t know yet”.

        • smocc says:

          But it also means that scientific consensus isn’t very predictive because the scientists aren’t using the usual methods that makes scientific consensus so good.

          • Enkidum says:

            Eh, I think “lots of really smart people who are experts in an area giving their informed opinions” is a really important part of scientific consensus. Agreed, though, that in this particular case the opinions are probably less directly tied to evidence, which we generally want.

      • Blue Tribe Dissident says:

        Could be wrong, but it seems to me that the interpretations are attempts to understand theory in more normal conceptual terms rather than in rigorous maths. Therefore, it is the same theory and cannot have any differences in the predictions it makes. However, a given interpretation might make some empirical results seem more or less intuitively plausible. This intuition could be wrong, of course.

    • Jiro says:

      Eliezer and LessWrong are making much stronger claims, though. It isn’t just “Eliezer thinks many-worlds is correct”, it’s “Eliezer thinks that many-worlds is so correct that no knowledgeable person can reasonably believe anything else”.

      The survey may show that 18% prefer many-worlds, but that’s different from 18% not only preferring many-worlds, but thinking that scientists who believe anything else are incompetent.

    • drossbucket says:

      Re: quantum interpretations, I went along to a different conference in 2013 that did a poll, and many worlds came out quite strongly on top IIRC. But this was in Oxford where they are obsessed with many worlds. Valentini had also attracted a few Bohmians – I’d never met one before!

      I agree that people are finally thinking hard about this stuff, quantum foundations is a really exciting subject now!

  7. Jacob says:

    I wonder if a big driver of scientific consensus being a better than ever guide to truth are blogs.

    Like you mentioned, there used to be a strong incentive not to offend established science even if you thought they might be wrong, which made you less likely to realize that they were wrong in the first. But on all the controversial issues you mentioned, there’s a lively public discussion going on the internet that mostly doesn’t care about offending anyone. This ranges from blogs by actual scientists (Andrew Gelman or Datacolada) to the dozens of HBD blogs that will dissect any new paper on genes and IQ.

    As a scientist, people will call you an idiot online no matter what you do. But if you suspect that you may be wrong about something, the chorus of blogs calling you an idiot may actually make you change your mind faster.

    • wintermute92 says:

      This seems plausible, and I’d also note that the ability for scientists to run blogs seems important.

      There are a lot of ideas which exist in a gulf between “can’t possibly admit you think this” and “can’t get a publication or conference lecture on this topic”. Accepting specific point of genetics and developmental psychology isn’t going to trash your career, but there can still be too much inertia to get it discussed. Fisking every stupid forking-paths social psych result ever won’t get you published, but there’e still value in it.

      So on that level, I think we get a lot of good from academics writing blogs. Real-name blogs are more respectable than comments or anonymous letters, but they still create space for low-burden-of-proof discussion. It’s stuff that previously got casually mentioned in person, preserved indefinitely for mass consumption and further development.

      Fields from law (Volokh Conspriacy) to social psych (Andrew Gelman) to mathematics (Shinichi Mochizuki, abc conjecture) now have academics writing non-peer-reviewed content that enables much faster development.

  8. ilkarnal says:

    I don’t like how this is being framed at all.

    Scientific consensus is very seldom wrong, and it is essentially never wrong for what I would consider fairly reasonable definitions of ‘consensus’ and ‘wrong’ – that is, ‘wrong’ meaning making an error that could be detected with available information, and ‘consensus’ consisting of essentially all those who have seriously studied the subject.

    If someone comes up with a wild and wonderful new hypothesis and everyone who hears it thinks it is retarded, but ‘everyone who hears it’ is a pretty small number of people, he isn’t being contradicted by scientific consensus. You are only ever being contradicted by scientific consensus if your claim is specifically contradicted by something that is already in that edifice of nearly certain knowledge that fits the definitions I put forth earlier. Otherwise, your claim can be arbitrarily retarded but certainly not ‘going against scientific consensus.’ You’re not ‘fighting scientific consensus’ when you go off in a new direction where there is no scientific consensus to contradict.

    I don’t like that one of the greatest achievements of mankind, the great edifice of essentially certain knowledge that has been built up by science, and has produced such wonderful fruit, is having its reputation hijacked. I want a clear line to be drawn between what counts and what doesn’t count, and what we must respect and what we need not respect. What counts is what is CERTAIN – what we have consensus on, what can be and has been checked, with no significant dissent. What we must respect is that built-up knowledge of questions asked, answers found by a great consortium of minds and stored.

    We are perfectly free to ignore and disregard anything else, and to scorn anything and anyone else. No great rewards lie in wait for those who rush to consider mere majoritarian positions sacrosanct – or for those who consider the words of any man or small group of men unquestionable.

    You mentioned a couple of important matters on which there isn’t scientific consensus – diet, and IQ. You can take virtually any position you like – IQ matters, IQ doesn’t matter that much, sugar is terrible, sugar is fine, powdered sugar is all the body needs if you have a high enough IQ – and find some support. It is basically impossible to ‘go against scientific consensus’ within those subjects because of the deep disagreements that cut through those fields on very central questions.

    To put this briefly – even if these fellas in the examples all were scorned and ignored to their graves, it wouldn’t say a damn thing about the value of ‘scientific consensus.’ Their unfortunate victimhood would be a measure of the ability of scientists to fairly judge new ideas, not discriminate out bad ideas. The value of ‘scientific consensus’ is what it refuses to assimilate, much more than its batting avg for accepting good ideas! This is so so so important! A body of knowledge that has nothing wrong in it, and as a result of its high standards accepts 10% of correct proposals, is much more valuable than a body of knowledge that accepts 90% of correct proposals, has much much much more total correct information, but also accepts 30% of incorrect proposals and as a result is overflowing with nonsense!

    Now it is possible that there’s an inefficient filter that could let in more good ideas without bad ideas sneaking through. There are few more idiotic ways of trying to measure this than taking people who are recognized in the present day as having good ideas, but initially faced some turbulence, and seeing how many of them gained rightful recognition!

    Most of them were doubted for very short periods, continued to be respected in their fields for their other accomplishments even during those periods, or were part of medium-sized movements rather than being lone geniuses. After a few years – maybe an average of ten, very rarely as long as thirty – their contributions were recognized and they assumed their rightful place in the pantheon.

    That’s why you bloody well heard of them! You didn’t hear of those who were right but were never vindicated, because they wouldn’t end up on the article you linked. You have no idea whatsoever how many of those there are!

    • Scott Alexander says:

      The first part of your comment is good and important. The second part (below the blockquote) I think I disagree with – the relevant fact isn’t that science eventually righted itself (which as you point out I would never know if it didn’t). The relevant fact is how quickly it (usually) happened.

      I think there’s a big difference between an institution which takes a couple of years to sort out its disagreements, versus one that completely rejects something for fifty years only to later find out it was wrong.

      • ilkarnal says:

        Doesn’t it have to happen fairly quickly to happen at all? After a certain point, different subjects are in vogue, or the field has progressed sufficiently to make the contribution irrelevant, or someone else comes up with the same idea but has the right connections or puts it in a way that is catchier or easier to understand.

        These fellas in the examples – they are often incredibly accomplished, leaving entirely aside the subject where they had a little trouble, or they are not lone crusaders but as you point out part of a faction. They aren’t where you’d look to find a problem with good ideas eventually catching, because either they have shat so much gold that people would probe a genuine turd of theirs for years, or they have other people willing to shout themselves hoarse once they get tired.

        But also, can you think of ANY subject that is politically controversial that got ‘sorted out’ quickly? Diet isn’t sorted, isn’t close to sorted, even in terms of eliminating some very very basic competing hypotheses. IQ isn’t sorted. Racial differences, their relevance absence or irrelevance, not sorted. In all cases it has been more than fifty years, easily.

        We can give you answers, as long as you don’t care very much about them! It’s a genuine problem.

        Concretely, science matters because either we believe the knowledge will aid our engineering efforts, or our societal initiatives. In terms of engineering, chemistry and physics were enormously helpful for a while, but the usefulness of new discoveries as opposed to the built-up body of knowledge seems extremely low. It doesn’t seem to take long for a field’s practical application to be discovered and almost fully exploited. You go from the infancy of modern physics to nukes to thermonukes very quickly – and then that’s it, the advances since are essentially engineering refinements. You go from modern chemistry to efficient explosives and medicines and fuel very quickly, modernity is ushered in – and then we don’t get super-modernity, because we’ve gained the really fundamental understandings, discovered the really important mechanisms. ‘Quantum leaps’ are done, instead there’s incrementalist progress building on progress.

        In terms of societal initiatives – well. I think the reliability of scientific knowledge is a result of its failure mode. When you get a bunch of people and you tell each of them individually ‘your job is to figure things out,’ people’s contrarian nature and love of pet theories will result on them all agreeing on damn near nothing, and what they will agree on are things that they HAVE to agree on, that are proven. When the average intelligence or honesty drops, that doesn’t result in a flood of spurious results. rather no results. Factions screaming at each other until they are blue in the face, outsiders looking in, shrugging, and saying – well, that field is shrouded in controversy.

        Benign as that failure mode may be, the bloody point of the whole enterprise is getting answers. It’s not a good idea to shove more and more people in, necessarily lowering intellectual standards, and making the newcomers desperate for any source of support, necessarily lowering standards on integrity as well. If this has happened, it’s very important to scream bloody murder about it. That’s why I like Cochran’s trenchant attitude – ‘they are no damn good!’

        You can have immense respect for the great body of scientific knowledge and still think the world would be a better place if most social scientists were instead looking for their next meal in a dumpster. Valuing scientific consensus is not the same thing as valuing modern scientists and institutions.

        Meanwhile, I am left in the same perilous position for matters of race, IQ, fiscal and monetary policy, and my itchy ballsack. These matters of great importance that demand bold and prompt action shall be met resolutely, even if that means soldiering on without scientific consensus!

  9. dansimonicouldbewrong says:

    Eventually, science does seem to converge remarkably well on what the evidence supports. However, the convergence rate is very uneven. Sometimes it’s a lot slower than seems necessary, and sometimes it’s a lot quicker than the available evidence can justify (and then whipsaws back and forth for a while, until the evidence solidifies). These variations in convergence rates sometimes reflect biases in the scientific community, which is after all a community of humans, and therefore subject to the normal complement of human biases.

    So rather than simply assume that scientific consensus is authoritative, it might be worthwhile to understand what kinds of biases the scientific community is prone to, and take that understanding into account when assessing instances of consensus (or lack of it).

    A few examples:
    – Scientists are heavily biased against views that make their own work less relevant. Such views threaten their funding, their prestige and their self-worth. This applies to entire fields or subfields, not just to individuals.
    – Scientists tend to prefer complex, technical theories over simple, intuitive ones–up to a point, at least. Ideally, a hypothesis should be exactly complex enough that the scientific community can understand it–at least to the point of generating more research from it–but the population-at-large will have trouble without help from scientists.
    – Scientists abhor a vacuum, and are likely to converge on a single consensus more quickly in the absence of well-articulated alternatives–even when lacking sufficient evidence to justify confidence–than if two similarly attractive hypotheses have been proposed.
    – Scientists are members of a particular socioeconomic class, and will be naturally more inclined to converge on a consensus that is consistent with the dominant views of that class than on one that conflicts with those views.

    We can debate the truth or significance of these claims about scientific bias, of course. But my main point is that a better understanding of scientific biases can lead to more accurate assessments of instances of scientific consensus than mere “benefit of the doubt” credence.

    • Gazeboist says:

      Scientists tend to prefer complex, technical theories over simple, intuitive ones – up to a point, at least. Ideally, a hypothesis should be exactly complex enough that the scientific community can understand it – at least to the point of generating more research from it–but the population-at-large will have trouble without help from scientists.

      This seems wrong, or maybe misframed. You’re implying that scientists want to be this gatekeeping “brahmin” class, priests of some arcane truth. This is … not the case. Scientists want correct, explanatory theories. They start from an already-existing theoretical framework, pretty much all the time.* They build on that framework, either to expand it to a new area or to explain data that doesn’t (or doesn’t quite) fit. The result is that the theory gets more complicated. This keeps happening; the theory accumulates more caveats and correlates, and becomes harder to understand. After a while, someone might notice that some of the original parts of the theory aren’t necessary anymore, and prune things back, but it’s usually not possible to get back to something simple without losing explanatory power, correctness, or both.

      It’s important to note that scientists also want formal theories, which usually means a very careful, precise (ie complicated) statement of what was originally a pretty simple idea. This makes it harder to understand in general, just because it takes longer to think through the precise version of the theory than the simple version. That doesn’t mean the population at large can’t understand the theory, just that most people don’t take the time to work through the precise version, which might have non-intuitive consequences. That last bit, of course, is why we have scientists in the first place.

      And again, a lot of the complexity comes about because the theories are explaining a huge amount of data, all highly varied. Systems are generally connected; the world doesn’t divide well into discrete facts.

      * Sometimes results come out that fundamentally change the framework, such that it is effectively replaced by a new one. If you look carefully, though, you can usually see scientists proposing things as modified versions of existing theories, rather than a completely new framework. Someone else then generally realizes that the shift has/is happening, and reworks things from the ground up. This is what happened with phlogiston, it’s what happened with relativity (there’s a reason it’s not the “Einstein transform”), and it’s what happened with quantum mechanics (Planck did not believe in the reality of photons). I think it’s even what happened with classical mechanics (over a longer period of time) if I’m correctly remembering my reading on pre-momentum physical theories.

      • dansimonicouldbewrong says:

        If you replaced “scientists” with “scientists should” in your response, I’d agree with you completely. In reality, however, scientists work in a community in which peers evaluate the quality of each other’s work, where “quality” is defined by community standards that may or may not conform to the lofty scientific ideals you list. In practice, scientific research must seem “hard”–that is, difficult for mediocre researchers, let alone non-experts, to produce–to win respect. Occasionally, that means setting up unnecessary hurdles for research to overcome in order to be accepted, or downgrading work that appears to have taken too little effort to generate, even if it’s of high scientific value.

        (In case it’s not obvious, I’m writing from experience, not just external observation.)

  10. suntzuanime says:

    When a creationist says that we can’t trust science because it used to believe in phlogiston and now it believes in combustion, we correctly respond that this is exactly why we can trust science.

    I think you’re equivocating on “trust” here. Phlogiston theory really does prove that we can’t take scientific consensus as 100% unarguable truth, which is what the creationist is trying to say and is something that the I Fucking Love Science crowd really actually does. You’re making the point that elite scientists are capable of changing their mind in response to new correct argument, but I think the problem most people have with scientific consensus is not with elite scientists but with people who read Vox.

    • ashlael says:

      Thanks for this comment.

      Reading this whole article I was strongly reminded of that post Scott once wrote about how advice could be good or bad depending on who it was directed at. “Think more about other people” is great advice for many people, but terrible for some others, etc.

      I don’t meaningfully disagree with Scott at all, but I do feel like he is getting the emphasis wrong – if I had written this post the caveat at the end about the dangers of relying on consensus would have been the main body of the post and everything about it working pretty well a lot of the time would have been the caveat.

      I feel like the world at large needs less respect for “scientific consensus”, not less.

      • JohnBuridan says:

        Certainly depends on what part of the world you live in. Rationalists on the internet vs. Liberal Arts Homeschoolers vs. Gov’t Intelligence Agencies all have different attitudes towards scientific concensus.

        Your comment about who it is directed at is right. Some people could use more trust, others could use less.

        In fact, I bet you could take surveys of separate university department majors around the country and get meaningful statistical differences among attitudes about scientific concensus.

    • Patrick Merchant says:

      The I Fucking Love Science crowd are a good example of the self-appointed guardians of the consensus that Scott mentioned. There are definitely people who imagine that science is a collection of proven facts that only dumb people bother arguing with (rather than an interconnected lattice of constantly evolving explanatory frameworks), and who then consider their own belief in those “facts” as proof that they are smart. But, vitally, those people usually aren’t actual scientists. Speaking from (extremely embarrassing) personal experience, they’re usually high school students with an ax to grind.

      Maybe it would better to say that you shouldn’t blindly trust the scientific consensus, but you probably should trust the current scientific community, since they themselves don’t trust the scientific consensus and are constantly trying to improve it.

      Global warming might be a good example of this. Its self-appointed defenders will claim something like: “99% of climatologists say the world will definitely be underwater by 2100, and anyone who doesn’t accept this is a dumb science-denier.” Actual climatologists seem to broadly agree that anthropogenic climate change exists…and could possibly be dangerous. And that’s it. They know that confidently predicting the outcome of climate change means you have to accurately model the climate of an entire planet, which is pretty difficult. They know that scary graphs are just projections based on models, and do not actually constitute evidence in and of themselves.
      They know that they are not literally Nostradamus. That one funny clip from the Newsroom is, as it turns out, not an accurate representation of climate scientists, who tend to be more humble about their powers of prognostication.

      Overall, I think the real problem are the science fans, not the scientists.

      TL;DR:

      The scientific consensus =/= a collection of inarguable truths

      The scientific community = mostly honest truth-seekers who try to account for their own biases

      • Would there be any for science defenders if there were no scuence attackers?

      • wbeaty says:

        “Science fans.” Heh, the population of mindless science-worshippers.

        Note that this Ridiculed Mavericks list was originally posted as a response to all the science fans who angrily insist that such things cannot happen. Or more commonly, insist that Galileo was the only known victim. And perhaps Semmelweis.

        In my case, the 1990s community on SCI.SKEPTIC newsgroup were the self-appointed guardians. I imagined that I’d simply point out some soc-science websites and books showing all the evidence of intellectual suppression, mistaken ridicule, etc., but I quickly discovered that no such lists existed online, or even in print. So, I slapped together the version on amasci.com.

        The meme took off, and currently there are many similar articles online, several even including cut-pasted sections from my own!

        “Science is the worst method for understanding the world …except for all the others.” Paraphrasing Churchill quote about democratic government.

        See also: letter to a dissident scientist, B. Martin, website on suppression of dissent in science

  11. marxistbiology says:

    Undoubtedly listening to scientific consensus is necessary if you want to be able to have an intelligent conversation and make reasonable decisions.

    It’s not so clear to me that we should treat scientific consensus as truth. As non-expert citizens, trusting scientific consensus is better than making decisions based on our dumb opinions. But scientists I hope are coming to their conclusions about the scientific consensus not because other people in the community believe it to be true, but because they have challenged those ideas the best that they can and have been forced to accept those conclusions given the preponderance of evidence. The category a person falls under should hopefully determine how they respond.

    This is all complicated by unprecedented digital access (truism, tell me if I’m wrong) to scientific studies and information which people are able to actively critique and talk about in smaller academic-esque communities (e.g. rationalists). People increasingly feel the need to be scientists and critique studies and their methodologies to get closer to truth. As a result, people are increasingly encouraged to think independently of scientific consensus and examine it as a “scientist.”

    Sometimes this has a major impact. Examples like AI point to how non-experts can have a significant impact on what consensus is (I presume that blogs had a significant impact, I have no evidence). In these cases non-experts challenging consensus is valuable and leads to the promotion of truth.

    So while I’m persuaded that scientific consensus is very good, I’m not persuaded that challenging and disagreeing with scientific consensus is bad.

    • Scott Alexander says:

      I agree with this. I think we should distinguish between Outside View (scientific consensus is probably right) and Inside View (we seem to disagree with scientific consensus, and should pursue that disagreement in order to keep its error-correction mechanism intact, even though we are probably wrong)

  12. lambdaphagy says:

    Not to nitpick, but Watson and Crick were the first to discover the structure of DNA. DNA had previously been isolated and identified as the genetic material, and some of its qualitative biochemical properties were known. Watson and Crick (+/- Franklin) proposed the double helix structure.

    Incidentally, Linus Pauling, the best chemist in a good century for chemists, was in hot competition with Watson and Crick to solve the structure. In his haste to beat them, he proposed some kind of cringey, jacked-up inverted triple helix thing that would never have survived in solution. Watson later wrote that the error was so out of character for Pauling that he must have been under severe stress.

    I guess that one’s more like “eminent scientist publishes crackpot idea; mavericks quietly feel sorry for him.”

    • teageegeepea says:

      Franklin didn’t propose the double helix. She was a thorough-going empiricist, who didn’t believe the prior model Watson & Crick came up with (and she was right that it was wrong). When they proposed the double helix (after seeing some of her crystallography), she still wasn’t convinced and thought more empirical work was needed. The third person who won the Nobel along with Watson & Crick was Maurice Wilkins, but like Franklin he did not work directly on that model (though Watson & Crick relied on his & Franklin’s work).

  13. Douglas Knight says:

    1. If you want to make an Appeal to the Authority of the Scientific Consensus, you have to be able to figure out what that consensus is. What does it matter that nutritionists have a good consensus if you don’t know what it is, and instead believe the medical school curriculum? Similarly, your psychology textbook lied to you.

    2. It is very common, both in your economics example, and in the Nurture Assumption case, that there is an explicit consensus that method X is better than method Y, but people will just keep using method Y. It seems very fair to me to describe that as an implicit consensus that method Y is good enough. Moreover, it is common that the consensus accepts the aggregate results of the inferior method, just because of the volume of publications; and thus the explicit consensus on the object level is made by methods that violate the explicit consensus of methodology. (Also, most of the replication crisis details were discussed at length by leading psychologist Paul Meehl fifty years ago. Everyone abased themselves in shame and proceeded to change nothing. Is this time different?)

    3. You might say that (1) is a special case of (2). Everyone accepts that you should look to experts with good methods, but they don’t actually pay attention to nutritionists. I think that it is fair to call the exclusion of nutritionists “the scientific consensus.”

    • suntzuanime says:

      Yeah, if you have to administer anonymous surveys to get the truth and the textbooks/middlebrow news website science articles are full of lies, you don’t have in a social sense a consensus in favor of the truth. The top scientists may have private knowledge of the truth, but the truth is not public knowledge.

      • Fossegrimen says:

        Scott acknowledges this;

        […] actually a fiction peddled by biased industry or media sources slandering a scientific community which actually had a much more sophisticated picture.

        The problem of course is that a lot of people don’t ever get past the fiction.

        An interesting example from nutrition and exercise:
        There is an essential piece of information in exercise nutrition that you need to eat fat and protein fairly immediately after a workout (or during, or just before if it’s a short workout). This is because in order to build/repair muscle tissue, you need anabolic hormone. If you don’t have a surplus of protein and some fats (mostly BCAA), the hormone is not generated and the muscle tissue you have destroyed during the workout is not regenerated. Instead, whatever you eat (say a salad) is converted to fat.
        This makes future dieting and exercise harder because you now have less muscle to burn calories and more fat to burn off.

        The crucial thing here is that diet books never ever say anything about this, possibly because fat is lighter than muscle so that if dieters who try to lose weight actually did it right, they would gain weight at the start of the diet and possibly give up in disgust.

        The absolute worst way to do it will let you lose weight initially and then make it harder and harder until you give up and bounce back to more than your previous weight and blame it not on the diet but on your own willpower.
        (of course it’s more complicated than that, but it’s one tidbit that annoys me a lot)

        For the general point of the OP:
        The probability that you are being ridiculed given that you have found a bold and unconventional truth is vastly different from the probability that you have found a bold and unconventional truth given that you are being ridiculed.
        (I suspect I read that in a comment on this site long ago. If it’s your quote, consider yourself attributed.)

        • psmith says:

          There is an essential piece of information in exercise nutrition that you need to eat fat and protein fairly immediately after a workout (or during, or just before if it’s a short workout). This is because in order to build/repair muscle tissue, you need anabolic hormone.

          I’m gonna push back pretty hard against this.

          Most people find it helpful to eat a little bit around training. But the only mechanisms I’m aware of are increased rates of protein synthesis from high plasma BCAA levels (as per Norton and Wilson) and possibly replenishing glycogen stores in an endurance context (Noakes, but I think there may be some new science about this). And one quite often sees recommendations along these lines (i.e., “eating a little bit periworkout, preferably some fast-digesting protein and carbs and while keeping your total macro intake constant, may be helpful and is probably worth a try”) in diet books, see eg Lyle McDonald, Alan Aragon, Michael Israetel. And nutrient partitioning is improved postworkout more or less regardless of what you eat during or immediately after it, though the improvement is small in an absolute sense–none of this “whatever you eat is converted to fat” stuff. And fat being part of the periworkout meal may actually be harmful to the extent that it slows digestion. Also, acute changes in test/GH levels within normal physiological ranges have never been shown to have a meaningful effect on outcomes, AFAIK, and claims to the contrary are a pretty reliable indicator of bullshit hucksterism.

          Furthermore, lots of people have progressed perfectly well ignoring all of this completely and training fasted, or eating whatever they felt like before or after or for that matter during their training and letting heightened rates of muscle protein synthesis or whatever act on whatever they already had in their stomachs from lunch three hours ago without bothering to top it off. Digestion takes a while, and absent fasting the trainee need not eat periworkout to have elevated plasma BCAAs and/or reasonably full glycogen stores during and after the workout, though of course the relevant levels will be higher if he does. The consensus recommendation to eat a little bit, etc., may be helpful (it was for me), but it’s clearly not a universal requirement.

        • Matthias says:

          How does your factoid about nutrition square with anything in evolution?

          Ie it’s relatively easy to see how building more muscles when there’s overall enough food and mechanical strain is a good idea in terms of evolutionary fitness.

          Having a weird flaw were the organism has to time their food intake just right seems like something evolution would have strongly selected against. Wouldn’t it?

        • Cheese says:

          I’m not entirely sure whether this is meant to be a bit ironic given the theme of Scott’s post, but even as someone who is in a tangentially related field to nutrition science i’m aware that the particular theory you’re talking about (a short anabolic window) is on fairly shaky ground and not really widely accepted in nutrition/exercise science.

    • Scott Alexander says:

      Maybe this is kind of silly, but for me a lot of this is about how I (and the rationalist community) present myself. Saying “I think scientific consensus is often wrong” is a controversial position that leads to getting mocked and not taken seriously. Saying “scientific consensus is usually right” is more popular, but would be dishonest unless it was true. I think in this case we can have our cake and eat it too.

      I also think that if you know scientists agree with you, it’s sometimes an easier battle to prove that this is true than to prove that you’re right.

      Also, if you privately know that scientific consensus is right about most things, you should be more willing to trust scientific consensus on other things, either when there’s a major effort to figure out what it is (as in global warming) or insofar as you should emphasize the skill of determining scientific consensus more.

      • Douglas Knight says:

        If you want to be popular, say it, regardless of whether it is “dishonest.” It is better to lie to other people than to lie to yourself by gerrymandering categories so that you can lie yourself that you aren’t a liar.

      • suntzuanime says:

        I feel when you’re trying to support statements like “scientific consensus is usually right”, it’s not fair to start the clock from when scientific consensus has been proven wrong by the guy everybody eventually accepts and see how long it takes for him to be vindicated. You need to count from the beginning of the scientific consensus. The whole lifespan of the Bad Old Paradigm counts against scientific consensus, not just starting from the time you personally read a book deflating the Bad Old Paradigm.

        • Scott Alexander says:

          Maybe my claim is more that scientific consensus is efficient, in the same way as the stock market or prediction markets or something? That is, once there’s enough evidence that something is wrong, and the better theory exists, the scientific community is pretty quick to change its mind.

          Obviously scientific consensus isn’t “always right”, in the sense that the 17th-century Royal Society didn’t believe in Einsteinian relativity and that was a mistake on their part. But expecting them to believe in relativity seems like demanding the scientific consensus be omniscient, which it obviously isn’t.

          I’m not sure what it means to make a claim of “the scientific consensus is usually right” which is neither limited to a simple efficiency-type picture, nor demanding omniscience.

          • Fossegrimen says:

            I think that Isaac Asimov is very relevant to this discussion.

          • suntzuanime says:

            I think we’re arguing over terminology, so let’s ask: what question are we trying to answer when we ask “is scientific consensus usually right?” I can think of a few:

            1) How much should I penalize the likelihood of this idea I have because it disagrees with scientific consensus?
            2) How much should I undertake costly endeavors which rely on scientific consensus being correct?
            3) How much should I penalize the likelihood of this idea I heard from someone else because it disagrees with scientific consensus?
            4) How much should I penalize this other person’s status because they disagree with scientific consensus?

            For 1 and 2, it seems clear that what you’re interested in is how much scientific consensus approaches omniscience. Sorry if that seems demanding.

            For 3 and 4 an efficiency argument makes sense, but there are some conditions for where it applies. It doesn’t apply to novel theories, only ones that have been around long enough they should have been adopted. It doesn’t apply to minority theories within academia, because you count it as a win if people are not thrown out of their institutions for espousing the idea. It doesn’t apply if you haven’t taken an anonymous survey to see if science secretly actually agrees but just won’t say. It doesn’t apply if you’re not deeply familiar with the cutting edge of the field, if you’re just reading textbooks you may as well go home. So it’s not clear how useful it is even for answering these questions.

            If you were trying to answer a different question please let me know.

          • John Colanduoni says:

            Great catch Fossegrimen! A key quote:

            My answer to him was, “John, when people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together.”

          • Aapje says:

            @Scott Alexander

            The problem with arguing that scientific consensus is efficient, is that you are ignoring the bit where most of the problem is: that many scientists are using extremely shitty methodology. You are arguing that scientists are doing quite well at reading the tea leaves. My claim is that they should stop reading the tea leaves and use more efficient methods instead, even if those take more work and result in less ‘impact.’

          • Harry Maurice Johnston says:

            @suntzuanime, I don’t think (2) typically requires omniscience in practical terms, because if the scientific consensus is so far wrong as to be able to affect your endeavors the problem is likely to have been noticed already. For example, there’s no project anyone could conceivably have carried out in the 17th century that would have failed by virtue of not taking Einsteinian relativity into account.

            In that particular case, I’m fairly sure that relativity was worked out in theory long before a lack of understanding could have caused any practical project to fail. Obviously that isn’t always true, but the faster the scientific consensus catches up to the newly available information, the fewer costly endeavors depending on that consensus will fail. So the speed at which consensus changes really is relevant.

            (I think a similar argument could be made for (1) but that seems more complicated than I care to try to think about at this time of night.)

          • suntzuanime says:

            I’m thinking of things like trusting scientific consensus on learning styles when designing a new school curriculum. Or trusting scientific consensus on nuclear chain reactions when deciding to invest in developing a new weapons system. Or trusting scientific consensus on the effects of fat in the diet when deciding whether to start a new diet plan. Or trusting scientific consensus on atmospheric feedback loops when deciding whether to implement harsh regulations on carbon emissions. There are a lot of cases where you can’t just rely on previous practice and have to decide how much you trust the predictions of scientists who have a model.

            I will concede that where scientific consensus has been successfully put to practical use, scientific consensus is not so wrong as to preclude successfully putting it to practical use. But those aren’t the interesting cases, they’re not the cases where people cite “scientific consensus” as an argument. We argue over models that have yet to be vetted in the real world, or where their success or failure may be confounded and hard to measure.

          • Steve Sailer says:

            “I’m thinking of things like trusting scientific consensus on learning styles when designing a new school curriculum.”

            Is there much of a scientific consensus on education? In the 45 years I’ve been following the field, I’ve seen a whole bunch of charismatic entrepreneurs put forward fads. But I don’t see much consensus since fame tends to go to self-promoters who can’t afford to agree.

            But I haven’t seen much replicable evidence overturning the landmark Coleman Report of 1966, which was paid for by the 1964 Civil Rights Act, that what students bring to school matters most.

            The Coleman Report really did represent a paradigm shift since it was so much better funded and more careful than what had come before, and its findings were so extremely unwelcome.

          • Harry Maurice Johnston says:

            But those aren’t the interesting cases, they’re not the cases where people cite “scientific consensus” as an argument.

            In most cases that I can think of, “scientific consensus” is used as a counter-argument rather than an argument, and I think that’s a legitimate use. X being against the scientific consensus is at least moderately good reason to be skeptical of X. X might nonetheless be true, but you probably shouldn’t bet the farm on it.

            The only counter-example that comes to mind is the assertion that it was safe to operate the LHC. I’m not sure I can justify that one, except maybe by calling it a calculated risk.

          • manwhoisthursday says:

            Howard Gardner is not a scientist. He is averse to measuring his multiple intelligences. There is not and has never been a scientific consensus on multiple intelligences.

            Conventional wisdom (in a certain area) does not equal scientific consensus.

          • nydwracu says:

            Howard Gardner is not a scientist. He is averse to measuring his multiple intelligences. There is not and has never been a scientific consensus on multiple intelligences.

            That’s precisely the problem, isn’t it? When you say “scientific consensus”, people outside the sciences hear “academic consensus”, and they can cherry-pick whichever academic fiefdom they want.

            The Soviet Union didn’t need to crack down on natural selection and inheritance; it could’ve just created ‘biology’ departments to study the unpopular but accurate ideas of Darwin and Mendel, ‘agriculture’ departments to churn out jabberwocky on the wonders of Lysenko, and a few more departments to take potshots at the one doing the actual science. Historians agree, comrade — natural selection is bourgeois propaganda.

    • Glen Raphael says:

      @Douglas Knight:

      If you want to make an Appeal to the Authority of the Scientific Consensus, you have to be able to figure out what that consensus is

      Right. Climate might be the best example of this – both “alarmists” and “denialists” can claim their own views are consistent with “the scientific consensus” because nobody can agree on precisely what the “consensus” claim is much less how much support that claim has; what surveys there are tend to get done by people with an axe to grind. Or as Scott put it above, it is a field where these surveys have been done inaccurately or in a biased manner

      It’s been plausibly claimed that having to resort to “scientific consensus” to win an argument proves you haven’t got one. Nobody ever says “the overwhelming scientific consensus is that 2 + 2 = 4.” (eg, Michael Crichton)

      But if scientific consensus is a reliable guide to truth, then: what is a reliable guide to scientific consensus? How do we figure out what the consensus is and how much support it has? Self-selected email surveys carried out by a grad student at some random conference? Prediction markets? An “impartial and independent” professional polling agency? Or do we just trust the side that does a better job edit-warring their view onto the relevant wikipedia pages?

      • Scott Alexander says:

        I’ve never seen any credible claim by denialists to have consensus on their side – when I see them make the effort at all, it’s more “look, a few scientists here and there agree with us”. Likewise, the survey debate seems to be a matter of 90% of scientists/papers vs. 99% of scientists/papers. Am I missing something?

        • suntzuanime says:

          The distinction you yourself made between a “lone maverick being ridiculed by everyone else” and a “minority-but-totally-existing faction”?

        • Glen Raphael says:

          Let’s ignore papers for the moment and just look at “what scientists believe”. If you do a survey, you can find a very high certainty of a really weak claim among a tiny subgroup of scientists, or you can find a very low certainty of a strong claim among a larger group of scientists, but not both. Alarmists try to claim both via sleight of hand but if you look carefully at the relevant papers you’ll find the data only supports one or the other.

          So, how do you get the big numbers?

          99% agreement is Right Out – you can’t get that due to the Lizardman Constant. ~97% is possible. In order to get “97% agreement” into the headline finding, advocates do two things:
          (1) interpret “the consensus” in such mild, vague terms that most “denialists” would claim to support it too
          (2) then keep slicing the relevant subgroup smaller until you get the number you want.

          For instance, one definition used for “the consensus” in such studies is something like “the earth has warmed over the last 50 years and human activity is partly responsible for recent warming”. It’s weak because everybody (now) agrees there’s been SOME recent measured warming and everybody agrees there has been SOME human contribution to recent measured warming. You don’t even have to believe CO2 is a greenhouse gas to agree – you could say yes just on the basis of believing changing land-use patterns (paving parking lots, cutting down trees…) have a warming influence. You can agree with THAT consensus statement even if you don’t believe the warming is still going on or don’t think it’s likely to continue in the future or think warming is beneficial.

          So the first step in getting a high agreement percentage is to pick that sort of weak, bland statement of what the consensus IS, one that most skeptics would have no problem also agreeing with. But even that is apparently still not good enough to get even 90% agreement in a survey of earth scientists. 70%, sure. Maybe even 80%, depending how you word it. But for propaganda purposes you want to portray an overwhelming consensus, and “4 out of 5 scientists” doesn’t cut it. So the second step in assembling a “97% of scientists” result is to dramatically narrow the scope of whose votes count. The winning move in slicing that particular salami was to look at something like “earth scientists who identify as climate scientists and have recently published more than X climate papers in the most prestigious climate journals and have published more on climate than on other fields in the last Y years”.

          And if you just look at THOSE guys (throwing out all the other votes from scientists deemed less active or less prestigious or less connected to the center of the herd) then you can get 97% agreement…with a claim that the most prominent skeptics also agree with.

          One could of course construct a “consensus” statement that skeptics WOULD disagree with, but then you’d find lots of normal scientists disagree with it too. For instance, some advocates claim the consensus includes that human CO2 emissions are responsible for MORE THAN HALF of recent warming and that the RATE of warming is INCREASING over time and that anything over TWO DEGREES is a dangerous tipping point…but if you included all of those kinds of claims in a survey you might not even get 60% strong agreement with the intersection of those claims.

          The kind of consensus that you can get decent agreement on really IS “it’s gotten a bit warmer lately and humans are somewhat responsible” – and even that is maybe an 80% consensus among the relevant group of scientists. And it’s something lots of skeptics agree on.

          (There must be a better way to label this problem – it’s a bit motte-baileyish. A term like “the consensus” – much like the term “god” – needs to be better defined before it makes sense to argue over who does or doesn’t support that named entity.)

          • suntzuanime says:

            Much like the “race doesn’t exist” scientific consensus mentioned elsewhere, activists take a very weak claim to construct a scientific consensus, and then do their activism making a very strong claim with ostensible scientific consensus on their side.

          • Glen Raphael says:

            Followup: A paper constructed by the process I described above is Doran 2009 (pdf link)

            The questions they asked to establish agreement with the consensus were:

            1. When compared with pre-1800s levels, do you think that mean global temperatures have generally risen, fallen, or remained relatively constant?
            2. Do you think human activity is a significant contributing factor in changing mean global temperatures?

            I consider myself a skeptic on climate but would have no hesitation answering RISEN and YES to those two questions, so the survey would have counted me a “supporter of the consensus”. (I might have more trouble if the word “significant” were more explicitly defined.)

            The salami-slicing category they came up with was called “Climatologists who are active publishers on climate change”. (That category applied to 79 people out of ~3000 surveys returned from a list of earth scientists.)

          • gbdub says:

            I think this is important, because it’s exactly this sort of consensus abuse that leads people into the failure mode that Scott describes for himself back when he was more likely to question consensus: thinking he’s a bold maverick against a broken establishment when really he’s just caught the leading wave of a healthy paradigm shift, or maybe the dying ebb of an interesting but ultimately flawed minority objection to the consensus.

            Consider an intelligent amateur who’s not directly involved in climate science or a related field, first exposed to climate change research via popular media: “90% of scientists agree global warming is real and caused by humans! If we don’t vastly alter our carbon output in 20 years we are doomed!”

            So he digs in a little. “Wait a minute,” he says, “it looks like these climate models have a lot of uncertainty, and the lower ends aren’t that catastrophic at all – look at how much warming relies on feedback mechanisms that we don’t understand all that well… And there were like 15 years where it didn’t really warm up. Isn’t the time constant on these ice cores awful long… would they even show something like what we’ve seen in the last 30 years?”

            One reaction he might get is “DENIALIST!!! How dare you question The Consensus!?”

            Another might be “Those are good questions! So good in fact that we’ve put a lot of scientific effort into answering them… here’s what we’ve found so far.”

            Now the former response is mostly limited to the I Fucking Love Science crowd and the “science” writers at various left-biased media outlets. I’m pretty confident the actual scientific community is good at responding to skepticism with the latter response.

            But which response is our intelligent amateur more likely to encounter? And if he does get the “DENIALIST!!!” response, he’s likely to end up making the wrong but understandable conclusion that science is dogmatic, politicized, and broken.

            So trust consensus, but don’t weaponize it. I don’t think even “scientific consensus is usually right” is correct – it’s more like “scientific consensus almost always gets it right, eventually”. Heck, historically speaking, scientific consensus is usually nonexistent, wrong, or at least incomplete. But that doesn’t mean that, at any given point in time, it’s not the best we’ve got.

            EDIT: of course there’s the third response, exemplified by James Picone below, that makes some substantive arguments against skeptical questions, but delivers them with such aggressive condescension as to make attempting to prove him wrong the natural emotional response, whether he’s right or not.

          • John Schilling says:

            I’m pretty confident the actual scientific community is good at responding to skepticism with the latter response.

            Could you show me where they are doing this?

        • ksvanhorn says:

          This blog post by David Friedman picks apart the “consensus” claim:

          http://daviddfriedman.blogspot.com/2014/02/a-climate-falsehood-you-can-check-for.html

          • James Picone says:

            David doesn’t take into account that thinking humans have caused less than 50% of recent warming is an insane position. People who publish in climate science would know that; interpreting statements like “Greenhouse gases emitted by humans contribute to global warming” as potentially being consistent with <50% warming is the same kind of error as taking the entire contents of a friend's fridge, and when they ask you what the hell you're doing, responding "Well you told me to help myself…"

            Do the maths, if you can; try to come up with a set of physically-plausible climactic constants that allow <50% of the recent warming to be anthropogenic. Chances are you'll end up with natural variability that can runaway all on its own, or a combination large negative feedbacks (making ice ages impossible) and a gigantic positive forcing we don't know about (which is wildly implausible).

          • ksvanhorn says:

            James Picone: Friedman is not arguing for or against a particular position on how much global warming is manmade. He is discussing a claim of 97% consensus on the question.

          • James Picone says:

            James Picone: Friedman is not arguing for or against a particular position on how much global warming is manmade. He is discussing a claim of 97% consensus on the question.

            My point is that the literal-words game David is playing makes no sense in the context of the survey; the volunteers assessing papers wouldn’t have assessed them that way, and people writing papers in the field wouldn’t write papers that way. Because claiming that <50% warming is anthropogenic is making an extremely strong, wrong-on-the-face-of-it claim. If you were surveying scientists on what percentage of Earth's orbit was caused by various bodies, answers that specifically refer to the sun without quantifying are almost certainly going to say it's the dominant influence, by far, because people who think otherwise are going to make that very clear.

          • Radford Neal says:

            James Picone: How do you reconcile your great certainty that >50% of recent warming is anthropogenic with the fact that the warming in the second half of the 20th century was hardly any greater than the warming in the first half of the 20th century, despite the fact that CO2 would have had little effect in the first half of the 20th century (hadn’t increased enough to make much difference then). Are you supposing that non-CO2-related anthropogenic effects were significant? Do you think you know of natural forcing explaining the earlier warming? Are you really so certain about all that that you feel comfortable ridiculing anyone who isn’t so sure?

          • James Picone says:

            @Radford Neal:
            Because I can’t see any way to make the physics come out sensibly with anthro < 50%, nor can a rather large scientific edifice.

            Just to put some numbers on the trends you're talking about, these are using GISTEMP. Other datasets give fairly different values (1950 to 2000 goes as low as 0.85 or so, 1900 to 1950 goes to 1 at most), this is just the first one I tried:
            0.84c +/- 0.29 / century
            1.16c +/- 0.28 / century

            …despite the fact that CO2 would have had little effect in the first half of the 20th century (hadn’t increased enough to make much difference then).

            Preindustrial CO2 was 280 ppm, CO2 in 1958 is about 315 at the start of the Mauna Loa dataset. This is ~16% of the effect of a doubling, we’re now at ~51%, so there’s a fairly significant CO2 effect prior to 1950. Industrial revolution started in the 1800s, so there may well have been more time for climate to reach equilibrium, which would exaggerate the effect of the first bit of warming compared to the second.

            Secondly, there’s the aerosol forcing; particulates from smog and the like are an anthropogenic negative forcing of uncertain size that stopped mattering around 1970 but ramped up fairly fast in the 20th century; this potentially depresses temperatures around the 1950-to-1970 mark.

            Thirdly, there’s some questions about the temperature record around WW2. There was a very sudden shift in the way ocean temperatures were measured during the war, it didn’t persist, and it’s unclear how well it’s adjusted for. That potentially exaggerates the ‘spike’ visible in the 1940s in the temperature record.

          • Radford Neal says:

            James Picone: From your reply, I conclude that your confidence is based largely on “theory”.

            As your numbers confirm, the first half of the 20th century doesn’t look much different from the second half, even though the CO2 forcing was at most 1/3 as much. Actually, I think 1/3 is way too high. The CO2 number you give for 1958 would already be a fair bit higher than 1950, and CO2 earlier in the century would be much lower, since the economic growth driving consumption of fossil fuels (or vice versa) was exponential, with bigger increases at the end of the 1900-1950 period than at the beginning. Add in that much of the effect of CO2 is expected to be delayed, and it’s hard to see the effect of CO2 in 1900-1950 as being anything close to its effects in 1950-2000.

            Note that I put “theory” in quotes above, however. We’re not talking about any solid physical theory here, even though some solid physical theories are used. Climate modelling is far from having a solid basis in theory both because the real theory is too hard to compute and because the data needed to compute isn’t available. So it’s all an approximation, in which at hundreds of points it’s necessary to decide that although such-and-such might conceivably be an important effect, it probably isn’t, so it will be OK to leave it out of the model (seeing as it would be a lot of work to put it in). Even if this is true for any individual omission from the model, it could lead in the end to results that are very wrong. And the chances of very wrong results are even greater if the enthusiasm for investigating a possibly significant effect that would push warming up might be greater than the enthusiasm for investigating a possibly significant effect that would push warming down.

            So I don’t find “I can’t think of a way to get my model to produce <50% of recent warming being anthropogenic" to be very convincing, especially when the most direct empirical evidence doesn't show signs of this being correct. Similarly, regarding your comments below about high climate sensitivity being needed to explain ice ages, with the added reservation that you need to also be sure that it doesn't explain earlier ice ages that didn't actually exist (I don't know whether this is a real issue or not).

      • James Picone says:

        You’re forgetting the existence of the IPCC.

        AR5 WG1 was the work of 258 different scientists citing on the order 9000 papers.

        It concluded that it is ‘extremely likely’ (i.e. >95%) that anthropogenic causes are behind >50% of the observed warming, with a best estimate of ~110%, which you basically have to get without bending the physics to breaking point and making ice ages impossible. They give a sensitivity range of 1.5c to 4c for doubled CO2, with a best estimate of ~3c, because otherwise ice ages are impossible (and frankly good luck trying to make an ice age happen with climate sensitivity at 1.5c).

        You’ve personally claimed, /on this blog/, that there is a ‘pause’ in global warming (how’s that one working out for you nowadays? Or do you only pay attention to short, statistically-meaningless trends when they go the way you want them to?); you don’t agree with the Cook consensus.

        Like, jesus christ, can we please get a better class of climate ‘skeptic’? Do the goddamn maths. Try to make the recent warming <50% anthropogenic. Figure out the physically plausible parameters it takes to get that. It’s been tried. It’s /very/ hard to do.

        Hell, find a paper claiming that <50% of measured warming is anthropogenic. There aren't many of those, and precious few (possibly none?) outside of vanity publishers like E&E. If the consensus was as weak as you claim it is, there'd be piles of the damn things.

        EDIT: Oh, and as to climate ‘skeptics’ agreeing with very weakened forms of consensus statements, keep in mind that Judith Curry has platformed Miskolczi, as have the Friends of Science, a Freitas/Baliunas-associated industry group. Miskolczi, if you remember, thinks that he’s proven mathematically that climate sensitivity is roughly 0.1. Or something like that. Also has an interesting interpretation of Kirchoff’s law which makes radiative heat transfer impossible.

        Meanwhile, WUWT regularly has posts from Tim Ball (here’s a recent example). Tim Ball was one of the authors of “Slaying the Sky Dragon”. He’s also a fairly prominent Canadian ‘skeptic’.

        Consider the comments on this WUWT post by Eschenbach trying to convince the audience that the greenhouse effect doesn’t violate the second law of thermodynamics (why would he need to do that if commenters there didn’t think that?). How would you describe the response to that post? Notice a commenter claiming to be Monckton essentially disagreeing with the first-order response to a CO2 doubling, as well.

        • Glen Raphael says:

          @Picone:

          Miskolczi, if you remember, thinks that he’s proven mathematically that climate sensitivity is roughly 0.1. Or something like that.

          A fine example demonstrating my point – climate sensitivity of “roughly 0.1” is still at least statistically significant, hence Miskolczi should be counted as “part of the consensus” if consensus is defined as per Doran 2009.

          The comment-allegedly-from-Monckton is another example – he’s not saying sensitivity is negative or zero, he seems to be arguing that it might be a bit lower than other people think it is. Which still puts him in the category “part of the consensus” as per Doran 2009. Thus I hope we can both agree Doran’s statement of the consensus is far too weak to separate skeptics from activists. I further presume we can conclude (because: math) that consensus on any stronger statement would be much less than what Doran found.

          On to the IPCC: The IPCC report certainly reflects conclusions that the chapter authors believe, at least as filtered through what is politically palatable to say (since it’s a political document). But it doesn’t necessary reflect even what the other IPCC scientists think, much less what scientists as a whole think – they have lots of disagreements among themselves (the Chris Landsea IPCC resignation letter remains both relevant and a fine example of nominative determinism). If a tiny group of activists are personally so convinced that they want to say they think some result is 95% likely – which I grant they do – that doesn’t mean 95% of their peers agree with them. (Steve McIntyre was an IPCC reviewer. So was Pielke.)

          Consensus[Doran] makes weak claims which have ~76% support among earth scientists (including skeptics) and ~97% support among a tiny group of 79 people, according to their survey.

          Consensus[IPCC] makes much stronger claims on behalf of a teeny-tiny group which would like to claim it speaks for all the rest of Science, but we’d presumably need to do a survey to find the actual level of support among either scientists-at-large or earth scientists. How to actually do that – pick the right questions to ask and who to ask and how to aggregate the results in a nonpartisan manner – still seems like a Hard Problem.

          Nobody besides you wants to re-litigate claims that might have been made in the comments of an old WUWT post, so I will decline to engage on that point.

          On “the pause”, since you insist: “When my information changes, I alter my conclusions. What do you do, sir?”

          • James Picone says:

            A fine example demonstrating my point – climate sensitivity of “roughly 0.1” is still at least statistically significant, hence Miskolczi should be counted as “part of the consensus” if consensus is defined as per Doran 2009.

            Given that that would require ~22% of respondents to Doran 2009 to actively deny the laws of thermodynamics, or ~10% of publishing scientists who responded to Doran 2009 to actively deny the laws of thermodynamics, I’m reasonably confident that’s not how people interpreted the question.

            Verheggen 2014 has the explicit quantification question you want, although I’m sure you’re aware of it. Went out of its way to include non-scientists who are active in the ‘skeptic’ world.

            Out of all respondents, ~83% said that GHGs were a ‘strong’ or ‘moderate’ warming influence since preindustrial times and also didn’t say any of the other factors were ‘strong’ warming influences. ~66% agreed with an explicit >50% number, including people who said “I don’t know”, “It is unknown”, and “Other”. Excluding that last gruop, you get 86% and 84% respectively, which suggests the lower consensus value for the explicit quantification is scientists with no expertise in attribution declining to answer (although I wish they’d split out “it is unknown” from “I don’t know” and “other”). Consensus went up as you limited to people who publish more in the field (i.e. active climatologists).

            It would be surprising if no participants in the IPCC had issues with the process. It’s a lot of people. It’s even what you’d expect as the consensus strengthens – because some subset of people who disagree aren’t going to change their minds, and they’re going to end up sidelined and pissed off. Note that the 258 number I cited was authors, not ‘expert reviewers’, who are random volunteers. You, too, could be an IPCC expert reviewer!

            My point re: old WUWT post is that “Global warming violates the second law of thermodynamics” was (and still is) a viewpoint held by a sufficiently large group of people that it’s totally worth consensus-position people arguing about it.

            I am pleasantly surprised to find out you’ve changed your mind on pauses. If I’m parsing you correctly, anyway.

    • Aapje says:

      Also, most of the replication crisis details were discussed at length by leading psychologist Paul Meehl fifty years ago.

      Thanks for the mention. It led me to this interesting paper (and it’s also fun to read, the guy is a good writer):

      http://meehl.umn.edu/sites/g/files/pua1696/f/113theoreticalrisks.pdf

  14. siduri says:

    This was so beautiful it almost made me cry. Granted, I’ve been drinking G&Ts all night, but still.

    I’m slowly working my way through your archives. It’s gorgeous to watch your thought evolving and becoming more nuanced. You’re the most even-handed and truth-seeking nonfiction blogger I know (and your fiction is enlivened by both those qualities, plus a blaze of pure beauty).

    • maybe_slytherin says:

      This was so beautiful it almost made me cry. Granted, I’ll be drinking G&Ts later, but I haven’t started yet.

      In all seriousness, I love seeing new people come & find Scott’s writing. It can be an intimidating corpus, and I’m glad people still get through all of that and make inroads. Come join us (time/sanity permitting) in the comments, or meetups!

      • siduri says:

        That’s super sweet. Thank you.

      • abc says:

        How about people with no critical thinking ability, like siduri is demonstrating?

        • suntzuanime says:

          There’s no need to be a complete asshole, friend.

          • abc says:

            Yes, there is. We don’t want the comments on SSC to become dominated by these kinds of sycophentic yes men.

          • Aapje says:

            I bristled a bit upon reading it as well, but I’ve been giddy in the past upon discovering a new source of information. The honeymoon doesn’t last and reality sets in after a while.

            Give the new user some time to fit in.

          • Iain says:

            Given the choice between a comment section full of “sycophentic (sic) yes men” and a comment section full of toxic assholes, I’m going with the sycophants every time.

            Last time I checked, flowery compliments were not one of the signs of the apocalypse.

          • Whatever Happened To Anonymous says:

            Honestly, if the choice is 0 to 100, I’d go with the assholes. But it’s not, and I’d rather not have another asshole.

          • abc says:

            Given the choice between a comment section full of “sycophentic (sic) yes men” and a comment section full of toxic assholes, I’m going with the sycophants every time.

            In that case this rationality stuff may not be for you. How about a nice safe echo chamber.

          • Iain says:

            Well, siduri has sobered up and is making interesting contributions in the thread about childhood trauma. Meanwhile, you’re still here sneering about safe spaces.

            “Sycophents”: 1
            Assholes: 0

            If you would like to increase the quality of the comments here, I suggest spending more time writing interesting comments of your own, and less time calling other people idiots.

    • abc says:

      This was so beautiful it almost made me cry.

      Ok, who else has absolutely no critical thinking ability?

  15. IvanFyodorovich says:

    With regard to the replication crisis, when I started biology grad school in 2005, we took a class in which the entire point was to do critical readings of papers and rip them apart. That same year I learned the idiom “just because it’s in Cell doesn’t mean it’s wrong”. Biologists were well aware that many high profile papers were irreproducible, that people produced spurious results by finding reasons to exclude data-points they didn’t like, and that most outright fraud wasn’t (and isn’t) caught. We had to be aware, most of us quickly get experience trying to build projects off other peoples’ wrong stuff.

    Thus when I first encountered Ioannidis my reaction was “it’s cool that someone is systematically studying this” rather than one of shock or denial. Ionnidis crystallized a lot of pre-existing thinking very well, but he didn’t change our understanding. Was it different for the medical and psychiatric fields?

  16. manwhoisthursday says:

    RE: Social-Justice-Related Issues

    As a conservative, I’ve found lots of liberal academics to be quite honest and well informed in their area of speciality, especially in areas where there is rigorous measurement. This applies even in the most controversial areas, like IQ. “Everyone is a conservative about what he knows best.”

    However, the big question is whether any of this filters out beyond these specialized enclaves. IQ denial is rampant in education and social services policy, while stereotype threat is gospel. Denial of sex differences seems to me on the rise. Activists are pushing implicit association tests as a way to ferret out racists in government and education. The humanities are totally fucked. Crazy SJWism is more out there in government and the corporate arena than ever.

    But, yeah, liberal academia can’t be blamed for everything.

    —–

    The problem with Marco Del Giudice et al.’s study is that the facets of the Big 5 they used were picked out more or less randomly. These weren’t clumps picked out using factor analysis like IQ or the Big 5. Fortunately, personality scientists have started to refine things: there seem to be two sub-factors for each of the Big 5. Using these sub-factors together sex differences seem to be quite large.

    • Steve Sailer says:

      Right.

      Here’s an analogy from entertainment: Over the years I’ve read dozens of profiles of Mike Judge, the creator of “Silicon Valley,” “Idiocracy,” “King of the Hill,” “Office Space,” and “Beavis and Butt Head.” For example, here’s the latest in the NYT Magazine:

      https://www.nytimes.com/2017/04/13/magazine/mike-judge-the-bard-of-suck.html

      Almost none of these lengthy articles on Judge seem to notice that Judge is, obviously, a man of the right.

      Entertainment writers like to believe that everybody creative and artistic is on the left, like they are, so they automatically assume that Judge must be.

      Much the same happens with scientists. Everybody assumes that scientists must be discovering huge amounts of evidence for leftwing social assumptions, and almost nobody notices they aren’t.

      • herbert herberson says:

        Judge’s critiques of contemporary capitalism are too biting and consistent to say he is “obviously” on the right–to say nothing of the fact that he goes well out of his way to describe himself as centrist/apolitical whenever he is explicitly asked (which you could claim is just him being coy to avoid hurting his career, but that didn’t keep him from doing the half-hour interview with Alex Jones I’m listening to right now)

        • nydwracu says:

          Judge’s critiques of contemporary capitalism are too biting and consistent to say he is “obviously” on the right

          let me tell you about frogtwitter

    • Douglas Knight says:

      What do you mean “picked out randomly”? Do you deny that 16PF was created by factor analysis? Maybe in half a century people will deny that the models you like were created by factor analysis. A lot of claimed progress in psychology is by lying about the past.

      Even if they were “picked out randomly,” so what? If they are just trying to show that it is possible to write a test that detects sex differences, what does it matter where the test came from? Perhaps it is good to restrict to well-established tests to avoid various kinds of cheating, but 16PF satisfies that.

      • manwhoisthursday says:

        Fair enough, the facets of the 16PF were derived from the best factor analysis of 1949, which means they might as well have been picked at random. But you are technically correct and I am wrong.

  17. Steve Sailer says:

    I’ve been an aficionado of the social and human sciences since 1972. It’s a common myth that science generates a huge amount of research in support of SJW talking points. The opposite is much closer to reality, but not many on either the left nor the right ever notice.

    For example, the “Race Does Not Exist” chestnut has only grown more popular in journalistic discourse since it emerged as a talking point around the turn of the century. (My impression is that Bill Clinton’s Rose Garden ceremony honoring the Human Genome Project is the main source.) Since then, there has been a huge amount of genomic research showing that, yeah, actually race does exist (and the genes pretty much follow what Cavalli-Sforza was saying in 1994 and Coon in 1965). The New York Times published several dozen articles to that effect in its Science section over the first dozen years or so of this century.

    And yet, almost nobody noticed.

    • Scott Alexander says:

      Not exactly. There are a huge amount of scientific studies that prove SJ talking points. I just think most of them are wrong or exaggerated in their importance. I think this is an important distinction if we want to talk about how narratives get made.

      Likewise, “race doesn’t exist” isn’t really a false fact. It’s an isolated demand for rigor, basically “let’s define race so strictly that it can’t possibly exist, and if later we find out it exists anyway, we’ll tighten our definition until it doesn’t”. This isn’t a great way to engage with the world. But it’s also not literal science denialism.

      • Steve Sailer says:

        “There are a huge amount of scientific studies that prove SJ talking points.”

        I’d mostly disagree. You see a few areas with a lot of studies in support: stereotype threat and implicit attitude especially. I’ll grant you that.

        But in major fields like IQ or human genetic diversity, there are remarkably few studies supporting the conventional wisdom.

        You see a lot of studies showing that various deplorable stereotypes are true, and you get the Occam’s Butterknife assumption that the stereotypes must have socially constructed the reality. But you don’t see a lot of studies showing that politically incorrect stereotypes aren’t true.

        Personally, I feel like I’m capable of wielding Occam’s Razor myself, so I’m mostly amused rather than disgusted by all the researchers who protect their careers by invoking Occam’s Butterknife. I know how to read carefully and come to my own conclusions, so I’m happy to read research recounting facts even if its gets spun at the end like a 1952 Soviet physics paper that ends with a paragraph of praise of Comrade Stalin.

        • Due to the ease of publishing science methodologically weak science (small sample, left out potentially relevant models or outcome/outcome switching, endless method variations, etc.), and the number of scientists in social science interested in proving SJW theses (such as ‘social inequality is unfair’), you’d indeed expect to see lots of mostly weak papers in support popular SJW ideas. I think this is indeed what the scientific body of research shows. (Meta-analyses will then routinely show these to suffer from strong publication bias.)

          To pick a random example, consider how many hundreds of these audition studies you can find. Their setup goes roughly like this: 1) make up some portfolio of people with identical descriptions but varying in sex and race, 2) give these to some other group for judgment (e.g. students, n=40), 3) show that people tend to prefer white males. This will probably work due to omitted variable bias. If you get unlucky and cannot p-hack it to work (p<.10 for any relevant test), then you can always try again next semester with n=43 new students. Repeat this with a large number of left-wing academics working over many years under publish or perish pressure.

          It does not require any evil intent, just some systemic biases. The good thing is that people are wising up to these problems. Now people start demanding pre-registered, large studies for claims, and these are less easy to force into showing what ain't true. An exception is that outcome switching being a popular choice in the medicine literature, this only works because people are too lazy to check the pre-registration. And now we have people working on the concept of Registered Reports, which make this impossible due because they fix the introduction, methods etc. parts of an article and only allows authors to fill in results when the data comes in.

          • teageegeepea says:

            I don’t think I’ve ever actually heard anyone, including those on the right, suggest there’s a problem with audition studies, so I’d been operating on the assumption that they’re valid. It would be interesting to see a skeptic try to replicate them.

      • Christopher Hazell says:

        The big problem with “race” is that the legal and popular definitions of “race” don’t really come from very good scientific study or thinking, they come from other things. Fer example:

        (looking up the information on a medication I’m taking, one of the “tell your doctor if…” warnings is “if you are of Chinese descent” – presumably there are side-effects that are more likely or more dangerous for Chinese/SouthEast Asian people)

        The borders of China were definitely not drawn to isolate people who have inherited risk factors for that particular medicine. Obviously, some Chinese people will not have that, and some people living outside of China will. Probably a lot of people from right by the Chinese border will have that same risk factor, as will people whose ancestors came from the parts of China, etc. etc.

        “China” is a kind of useful shorthand that lay-people can understand, but it’s a rough proxy for what we actually want to measure, and it only works AS a proxy because of contingent, historical events that aren’t related to what we want to measure at all.

        If we lived in an alternate timeline where China had never become a unified country, and instead that same geographical area was filled with a hundred different countries, what would the warning on that medicine say? Would it list all one hundred of them? 75 of them? 50?

        There are a lot of ways to cut the human pie. We can look at the entire population as a whole; or, on the other hand, every single human is unique.

        One of the reasons it seems to make sense to slice the human pie along racial or national lines is that for the last four centuries or so those have been incredibly important legal constructs, and those legal constructs have done a lot to shape our lives.

        And since they have done so much to shape our lives, they aren’t irrelevant. Things like slavery and border control have done a lot to determine who is allowed to reproduce with whom.

        But they are legal constructs, and they weren’t built to answer the question of “What is the most sensible way to study and divide populations?” Given that, I think it makes sense in many situations to ask whether we are only interested in looking at populations by race or nationality because of a sort of inertia.

        I’ve said a few times here that I have a similar problem with the idea that “boys and girls have different learning styles.” In my experience a significant minority of boys will prefer the girls style of doing things, and vice versa, which makes me think sex and gender aren’t the best way of slicing that particular pie. It might well simultaneously be true that most boys prefer one learning style, and most girls another, yet we would be more successful introducing, say, a three learning style model that ignores sex/gender altogether.

        The fact that we CAN divide people into certain populations with broad similarities doesn’t mean that those divisions make the most sense for any given scientific or social engineering task.

        • Aapje says:

          I think that we have an issue in general with recognizing groups that we are not used to looking at separately.

          For example, there are more poor whites than poor blacks, but if you look by race, the former group is hidden by the whites that do well. So you get anti-poverty/pro-social mobility measures that are targeted at black people and that completely ignores the greater group of poor whites. An issue is that even if the group is recognized, the metrics that are used to determine success often exclude them, so then there is no incentive to help them.

          The same for genetically similar groups that do not match races that well, learning styles, etc.

          It might well simultaneously be true that most boys prefer one learning style, and most girls another, yet we would be more successful introducing, say, a three learning style model that ignores sex/gender altogether.

          It can also be true that there is a combination of socionormative issues (gender roles) as well as learning styles. In that case, you may need to address a gender role issue for all boys and also facilitate the proper learning style in a non-gender specific way. And you can also have a biological issue, for example, boys may benefit more from combining education with exercise.

          So it may be much more complex than how it is now presented.

    • Deiseach says:

      “Race” is such a tangled word that I don’t think we’ve any good working definition the majority of people will agree on. I’d agree mostly with “There’s no race but the human race” (because saying that various global populations of humans are not all part of the same species is wrong). On the other hand, there are definite huge changes in what is considered the correct approach to the racial problem: the colourblind idea has very much fallen out of favour, and now we’re getting to the point where simultaneously you must not entertain any idea of biological differences in populations and on the other hand drinking milk is racist because of lactose intolerance in sub-groups.

      “Race is just a social construct” is one of those “it’s both right and wrong” ideas: yes, there are noticeable differences in biology between sub-groups (looking up the information on a medication I’m taking, one of the “tell your doctor if…” warnings is “if you are of Chinese descent” – presumably there are side-effects that are more likely or more dangerous for Chinese/SouthEast Asian people) so the ‘invented construct’ idea is wrong, but that people are treated differently and that prejudices and biases exist and that there are inequalities on perceptions of ‘X people are more likely to be A, B, C’ is also undeniably correct, so the ‘social construct’ idea is right.

      There is a legitimate fear that saying “population A has in the average of all the populations a lower/higher intelligence than population B” will lead to prejudice and mistreatment of population A because this has happened in human history. So it’s not panic-mongering and denialism to try and control how that information is disseminated and used. On the other hand, if it is true, then it should be made clear that this is not the same as saying “any individual of population A is going to be stupider/smarter than any single individual of population B” and I don’t think it’s asking too much to show how you are going to prevent such conclusions being drawn or used abusively.

      If it’s true, it has to be taken into account for, say, approaches to education just as much as taking into account ‘giving milk to these students in a free school meals programme is going to make them sick so we need alternatives’. Screaming “that’s racism!” isn’t going to help and may hurt the very people you are trying to defend.

      • Steve Sailer says:

        The federal government puts a large amount of effort into collecting data on Americans by “race,” which it successfully uses in lawsuits. It long carefully distinguished race from ethnicity.

        Many years ago I put some time into researching a series of articles making fun of federal racial definitions. Eventually, however, I gave up because I concluded they were, on the whole, good enough for government work.

  18. skybrian says:

    “believe in the lipostat and think it’s important”: what does this mean? Is there a good article about this?

    “great and widely-accepted work on how intersex people take on the characteristics of their hormonal rather than their social gender”: where can I read more about this?

  19. Jiro says:

    I was expecting “scientists are usually right, so we should be more skeptical when AI danger is rejected by scientists”, not, as Scott has tried to claim, “scientists are usually right, and see, they are right about AI danger after all”. Even the ones who talk about AI danger mostly don’t seem to mean it in the way Eliezer means it.

    Also, pretty much no scientist thinks that the only sensible interpretation of quantum mechanics is many worlds. So Scott should reject the idea that the only sensible interpretation of quantum mechanics is many-worlds. (Many scientists do consider many worlds to be a sensible interpretatiom, but not the only sensible one like LW does.)

    • Harry Maurice Johnston says:

      It kind of depends on what you mean by “sensible” anyway. From a pragmatic point of view, if you’re just concerned about being able to build better transistors or whatever, it simply doesn’t matter whether the model of quantum mechanics you’re using makes sense, just so long as it comes up with the right numbers.

    • andhishorse says:

      Nobody would ever say there’s a scientific consensus in favor of Bostrom’s theories. But at this point I think it’s also indefensible to say there’s a consensus against.

      That does not seem to be what Scott is trying to say at all.

    • Gazeboist says:

      Does Scott even have an opinion on QM interpretation? I don’t think he’s ever brought it up.

      • Nornagest says:

        Scott doesn’t, but Eliezer has quite a strong one. Or maybe had; the article I read on it was from 2009 or something.

  20. John Colanduoni says:

    Doppler tested the effect himself by having someone play a trumpet on a train (really), someone else successfully tested it in 1845, and it was independently rediscovered in 1848.

    It’s important to note that the trumpet-on-a-train test is for the aucoustic Doppler effect, not the optical one. The conflict with the “Luminiferous Aether” doesn’t appear if you assume that the optical Doppler effect works in a similar manner to the acoustic, because the acoustic version references a privileged reference frame (that of whatever is carrying the sound waves) while the optical version cannot due to relativity.

    From what I can tell his initial treatise “Über das farbige Licht der Doppelsterne und einiger anderer Gestirne des Himmels”, while proposing that there is an optical Doppler effect in addition to the acoustic, provides only one set of formulas (the non-relativistic one that works for sound). I had always assumed he focused on the acoustic version since he was way too early to get the relativistic formula right, but it looks like he shot for the stars and missed. Not only does his theory not run into trouble with the Aether, but I think (my German isn’t great) he mentions the debate in the beginning and doesn’t pick a side.

  21. John Nerst says:

    Probably repeating things said above (it fills up so darn quickly), but here goes:

    Like with so much else, there appears to be a motte-and-bailey situation going on here, and this post says something a lot like “the bailey is not true, only the motte is”.

    The bailey is that whole scientific fields are often wrong and single non-experts can, while being mocked and shunned, find out the truth by themselves. This doesn’t seem to be happening to any significant extent.*

    The motte is that the popular perception of scientific consensus is often wrong and smart individuals can get it right. This doesn’t require that nonspecialists can take on whole fields, partly because whole-field consensus is rarer than we’re led to believe – a scientific field is full of lively debate and a lot of uncertainty and many apparently contrarian opinions one might have is represented by at least some experts (sometimes even a majority). If not, you’re probably a crackpot.

    Instead we tend to mistake conventional wisdom (supported partly by outdated, cherry picked or overinterpreted scientific findings) for scientific consensus. There’s a cute symmetry here: just like science is a slow and buggy way to imperfectly represent the nature of reality in human-readable form, conventional wisdom is a slow and buggy way to imperfectly represent scientific knowledge in a layman-readable form**. It’s not that hubristic to think you can do better than conventional wisdom by paying attention, read a lot and practice rationality. It’s a separate thing from thinking you can do better than science at what it does.

    There’s also another situation where the basis for a field rests on fundamental assumptions people can reasonably disagree about. I don’t think I’m out of line when disagreeing with theologians about the validity of their field without being a theologian, for example. Similarly with some branches of philosophy. There are softer examples as well, like when a field studies only one aspect of a phenomenon, making the implied statement that this is a decent approximation of the whole story. Practitioners of behavioral genetics or cultural anthropology might not literally be biological determinists or blank-slatists, respectively, but their bodies of knowledge are kind of (I could be wrong) built on such regulative assumptions and that makes them sort-of-rejectable by people who think that what they leave out is essential (and leaving it out makes someone wrong) – and what’s essential is a tricky philosophical issue that domain expertise doesn’t grant you authority on.

    *Although we stack the deck somewhat by only including typically “sciency” questions and exclude “seeing like a state”-like situations where local, situated knowledge is sometimes superior.

    **It’s probably the norm to encounter ideas originating in the academic sphere not in their original form but in simplified, popularized and often politicized versions, making them seem stupider than they are (e.g Scott’s economics example, social justice theory, evolutionary psychology etc.).

  22. Steve Sailer says:

    Most of what comprises “the scientific consensus” isn’t controversial. The periodic table, for example, isn’t much argued over. There’s nobody to argue with. Who is against the periodic table?

    The subjects that most make us want to argue are the ones that seem to have a pretty good case on either side: Who is better: Tom Brady or Peyton Manning? Who should be NBA MVP this year is getting argued over more than it was last year because there are multiple good but not flawless candidates while last year Steph Curry was a unanimous choice.

    Basically, there’s a selection effect involving what people want to argue over. Pinker says: “mental effort seems to be engaged most with the knife edge at which one finds extreme and radically different consequences with each outcome, but the considerations militating towards each one are close to equal. … “seems to explain a number of paradoxes, such as why the pleasure of sports comes from your team winning, but there would be no pleasure in it at all if your team was guaranteed to win every time like the Harlem Globetrotters versus the Washington Generals.”

    http://takimag.com/article/quibbling_rivalry/print#ixzz4eaWXCoVL

    I call this the Most Boring Insight of All Time because I think it explains a lot, but nobody thinks it’s interesting enough to argue with me about it.

  23. Steve Sailer says:

    There’s also a lot of incentive to create some new paradigm overturning scientific consensus so that you can go down in history as a peer of Copernicus, Darwin, and Einstein.

    And some people find it fun to argue an unlikely case against long odds.

    Plus, some scientific issues, like global warming, have lots of money on the line, while others, like evolution, have a lot of religious/ideological status on the line.

    And some are creative works. For example, on Twitter I’m trying to raise awareness of the old Hollow Earth theory because I find it piquant that America’s greatest Vice President, Dick Johnson, had sponsored legislation in the Senate to fund an American expedition to explore and conquer the inside of the Earth.

  24. Jack V says:

    I’ve been working on something like, holding in my head a “orthodox belief I used to hold” and “various alternatives I’m not convinced of but haven’t rejected out of hand”. I’m not very good at it yet. But I think I benefit from thinking of that way, that it’s actively worth having weird ideas around because they might be right. And favourite weird ideas which seem to have a lot going for them, but haven’t actually outraced the null hypothesis yet.

    I’m used to thinking more like, “this is right”, or “I don’t know” (or, something more baysean). But holding the alternatives in abeyance seems to help for many reasons, even if I have an idea which I think is right.

    • cuke says:

      This strikes me as a good description of psychological health. Also Buddhist non-attachment, and therefore path to less suffering.

  25. Aapje says:

    I think that you can look at it like a pyramid. At the bottom you have ‘truth’, which we presume exists, but is not something that we can perceive objectively or in its entirety. So it’s more something that we have to assume exists. Above it we have limited views on truth, called measurements. This is the part where mankind can get a glimpse of truth, by measuring a fraction of it. Then above that we have theories based on those measurements.

    The most important benefit of science is that it gave us methods that give fairly high confidence that we are measuring what we think we are measuring, as well as an understanding of the limits to our ability to measure many things accurately. However, these limitations are immense, especially when it comes to more complex questions. So there is a strong tendency to use sloppy methods and/or to claim that we measure more than we do. The amazing part is that this hasn’t totally corrupted science, as we somehow have been able to keep valuing a norm of demanding proper methods and making right claims, despite this norm probably being in the disinterest of individuals, even as it benefits the collective. However, something is still rotten in the state of Denmark, as many scientists are (unconsciously?) undermining the norm even as they profess to follow it.

    The checks and balances that ought to prevent this are insufficient.

    As such, we have a large quantity of science that is extremely inefficient as the measurements are poor. This directly undermines the top of the pyramid: drawing conclusions based on the measurements. It would be far more efficient to have fewer studies, but to do those well. Then you’d have high quality data to draw conclusions upon.

    The debate about whether or not scientists are efficient at digging through heaps of shitty studies misses the point: we shouldn’t have that heap in the first place.

    • Cheese says:

      “However, something is still rotten in the state of Denmark, as many scientists are (unconsciously?) undermining the norm even as they profess to follow it.

      The checks and balances that ought to prevent this are insufficient.

      As such, we have a large quantity of science that is extremely inefficient as the measurements are poor. This directly undermines the top of the pyramid: drawing conclusions based on the measurements. It would be far more efficient to have fewer studies, but to do those well. Then you’d have high quality data to draw conclusions upon.”

      While I totally agree with this, my practical response to the problem you outline is kind of ‘well i’d like a toilet made out of solid gold but we can’t have everything we want’. The weird tangled net of incentives that is modern scientific education, funding and research is really just a reflection of how society and people work (let’s not even talk about the extra complications that come when you’re doing something with potential direct medical applications).

      I’m not making a criticism of efforts to put more emphasis on methodological considerations. But I think a ‘we are doing science wrong and if only we did it properly* it would be great’ angle is talking about something that’s fundamentally unattainable.

  26. allspoilersallthetime says:

    For what it’s worth, Cordelia Fine denies that she’s claimed that there are no gender differences.

    “[W]e all believe, like Cahill, that sex matters; that is, that genetic and gonadal sex can influence brain development and function at every level, that useful information may arise from investigating such processes, and that this may be especially critical in understanding pathological development. Indeed, numerous explicit statements to this effect can be found in our work.”

    I don’t know if this is the same as ‘sounding really defensive’ but it’s probably all to the good if Fine is clarifying her position as being against bad science (eg. small sample sizes) and over-confident conclusions. It’s seems like she thinks of her position as skeptical rather than denialist.

    • Steve Sailer says:

      There are a lot of rewards from our culture for being an obscurantist about unpopular science. Stephen Jay Gould did very well for himself that way.

      • wysinwygymmv says:

        “When my side exaggerates it’s because they’re being provocative and you can’t hold it against them. When your carefully qualifies their statements, they’re being obscurantist and can be dismissed as liars.”

        You have a pretty serious double standard on a lot of this stuff.

        • suntzuanime says:

          That doesn’t sound like a double standard at all. Staking out bold positions is good, trying to avoid taking positions is bad, no matter which side is doing it. Perfectly consistent.

        • Steve Sailer says:

          Obscurantists and liars are rather different things, at least in the means they use.

    • mcduke says:

      I have not read Fine’s latest book, only “Delusions of Gender”, and while it definitely wears its bias on its sleeve, it did open my eyes to the important role cultural and social influences have on the development of gender identity and various things that usually get labeled as “sex differences” with the either explicit or implicit assumption that it’s biologically “hardwired”. My impression is that the whole field of gender differences oscillates on a nature-nurture axis over time, and that people like Pinker lend weight to the nature side, while Fine etc. give a healthy critical counter balance. And I think both sides are not free of occasional overconfidence and too much snark at the other side.

      Currently I’m reading “Brain Storm” by Rebecca M. Jordan-Young, subtitled “the flaws in the science of sex differences”, which is a good short description of the contents. It’s from 2010, so it might be a bit dated already, but it’s the most promising book I found to maybe get a balanced view of the important concepts, studies, issues, etc. of this subject. (It’s not easy to find something that’s neither 100% blank slate nor “cultural influences are just hippie SJW talk”.) Anyway, it seems to contain a fair overview of the history of the field of “brain organization theory” (basically, the influence of early hormone exposure to brain development), its important controversies and methodological issues. Like “Delusions of Gender” (which is way more social critique and witty opinion than this book), it contains an explicit disclaimer that it in no way tries to say that there are no sex differences, but that (then-)current studies were not in universal agreement on what the importance of hormonal influences etc. actually is, and that, basically, it’s complicated. Which I think is an important, healthy contribution. If anyone else here has read the book, I’d be interested in opinions. It’s easy to find either glowing adorations or aggressive shreddings of Fine’s book, but for Brain Storm, not so much. (Though it does have a positive blurb from Diane Halpern on the back cover, who I think wrote a textbook on cognitive sex differences, so there’s that.)

  27. Steve Sailer says:

    Scientists making major paradigm shifts post-Galileo have tended to be pretty popular pretty quickly. Newton was wildly celebrated almost immediately by the handful of people who could understand his big book. Darwin was a respected figure and was buried in Westminster Abbey. After the publication of his General Relativity, Einstein became a huge celebrity almost as soon as WWI was over.

    In general, our culture admires innovators.

    • sconn says:

      Right, and it seems like “overturning the entire consensus of the field” is exactly what every scientist secretly dreams of doing. Scientists are the opposite of, say, clergy — proving orthodoxy wrong will get them huge status boosts and so naturally it’s what they want to do. Of course all the other scientists want to resist the new paradigm — first because the new paradigm needs to be rigorously tested before you can be sure it’s right, and second because they don’t want the glory stolen by somebody else. But when adequate proof has been produced, everyone goes “wow, you won at science.”

  28. sconzey says:

    The important thing* is not what the scientific consensus is, but what people believe the scientific consensus is. If you’re advising people to trust the scientific consensus in absence of other evidence it is really important that you didn’t know what the scientific consensus was in fields in which you are unusually well read.

    *with regards to peoples’ decisionmaking and public policymaking

    And the scientific consensus really is only useful in absence of other evidence. If someone says ‘this theory is false for these reasons’ then ‘it is the scientific consensus’ is an unacceptable counterargument.

    • Steve Sailer says:

      It was pretty eye-opening to me to watch Nicholas Wade point out over and over again in the bully pulpit of the Science section of the New York Times that the new genomic data was disproving the current Race Does Not Exist conventional wisdom … with virtually nobody noticing.

      Zeitgeists are powerful.

  29. sconzey says:

    With regards to the politicisation of science, mis-sciencing is related to more than just politicisation. It’s a product of politicisation and testability:

    While the heritability of IQ is subject to more political pressure than global warming, smart parents still expect to have smart kids; heritability is incredibly testable and the truth or falsehood of the predictions it makes are obvious to everyone.

    Global warming on the other hand is a conclusion come to after extensive number crunching and analysis of huge datasets. The predictions made are about minute changes in an aggregate variable where the impact on things-we-care-about like extreme weather, ice melt and growing season length are themselves subjects of intense debate.

    • Steve Sailer says:

      Right. I mean I’m pretty good at noticing changes over time, but I couldn’t tell if there has been any climate change in my native Southern California over my lifetime. That certainly doesn’t mean it’s not happening, just that it would be quite a challenge to observe and feel confident that it’s happening globally. (I gather that hikers in the Alps have noticed a clear recession of glaciers over their lifetimes. But there aren’t any glaciers in Southern California.)

      On the other hand, the kind of topics I’m interested in are pretty obvious both from simple observation of daily life and from social science, which is why they are so unpopular to point out.

  30. educationrealist says:

    “Is there much of a scientific consensus on education?” (Steve Sailer)

    I was gonna say. No. There’s not. In fact, education is very much a Brady vs Manning area. Massive debate, intense scrutiny of absurdly unimportant or trivial differences, huge brouhaha over insignificant “significant” findings.

    And yet, oddly, the more I teach, the more I realize that a “pure” approach to education using IQ would be terrible in a racially diverse society. White folks defined IQ and its measurement. What Europeans think of as “smart” works for white people. But run into hundreds of otherwise ordinary Chinese or Koreans with perfect SAT scores who can’t explain the basis of the quadratic formula or any point about Macbeth that didn’t come from SPARCnotes, and you get skeptical. (Same thing for blacks the other way round, but my sample size is smaller.)

    • Ilya Shpitser says:

      Not a big fan of IQ (both the concept and its use).

      Lots of folks who are fans (with a not so subtle undercurrent of practically salivating at the thought of finding a bit of math to justify bigotry) couldn’t explain factor analysis, or what it does.

      Not a right-exclusive issue, either, my favorite left example is the “wage gap.”

      In the spirit of today’s post, I would be interested in a larger discussion of where the modern intersection of psychometrics and data science is (or ought to be!) The heavy emphasis on factor analysis in psychometrics is basically a historical accident, it does not imply there is some sort of deeper truth about human brains in the output of factor analysis.

      • sconzey says:

        a bit of math to justify bigotry

        Many people, on both sides of the discussion, seem to confuse intelligence with moral worth.

        • Ilya Shpitser says:

          Yeah, that’s fine. I am going to remain slightly cautious about the HBD-crowd’s understanding of “moral worth” of people.

          All I am saying is people who talk about intelligence differences are deeply confused about the limits of data-based psychometrics for measuring intelligence. They latched on to IQ like a drowning man to a raft.

      • HeelBearCub says:

        IQ itself seems to be a valid concept.

        But my sense is that the scientific consensus is way behind what the strongest proponents want to claim. For instance Scott seems to say that IQ tests aren’t culturally biased, but I believe that it’s more proper to say that IQ testers know that the tests can be culturally biased and compensate for it.

        And then there is the treatment of IQ as the only important factor. That success depends completely on IQ and nothing else. Scott even repeats this statement in this post, asserting that scientific researchers who have a novel and correct idea are necessarily smarter than all the other researchers.

        That is a sign of sloppy thinking on the matter.

        • Ilya Shpitser says:

          “a valid concept.” — what does this mean?

          • ignition says:

            I assume it means that it can consistently be used to make predictions significantly better than chance.

          • HeelBearCub says:

            @Ilya Shpitser:

            By “valid concept” I simply mean that we have good evidence that the results of an IQ test, within a cohort for which the test was designed, have a strong positive correlation with results of many other “intellectual” tasks.

          • Ilya Shpitser says:

            Sure, I agree, under that notion of validity.

        • Deiseach says:

          I’m coming around to the conclusion that what IQ tests measure, in a non-culturally biased way, is “who is good at maths and who isn’t” 🙂

          • Steve Sailer says:

            Nah. Strong verbal skills, especially verbal logic skills, also register on IQ tests. For example, Obama appears to have aced the LSAT test (he only applied to law school at Harvard, Yale, and Stanford, suggesting he didn’t feel the need for a safety school, despite undergrad GPA in the 3.5 range). He’s a high IQ guy, despite having no apparent math skills.

            Similarly, Lin-Manuel Miranda, the composer/lyricist of “Hamilton,” tweeted that he scored 750 on the SAT Verbal but only 500 on the SAT Math. I would suspect he would score decently on an IQ test.

          • sconzey says:

            UK Mensa give two IQ tests the Cattell III B and the Cattell Culture Fair test. The ‘Culture Fair’ test is the mathy one as it’s a series of progressive matrix problems. The III B test however is very much about verbal reasoning. It’s not considered Culture Fair as it requires a broad English-language vocabulary.

    • ksvanhorn says:

      >What Europeans think of as “smart” works for white people.

      Ah, that must be why Asians have higher average IQ scores than Europeans.

      • educationrealist says:

        Ooooh, you failed the reading quiz. Also the logic quiz, when you determine I’d need to be told that, given what I wrote.

      • ignition says:

        Sick burn, bro.
        Educationrealist is suggesting that the typical *ratio* between different IQ-associated abilities varies by race.
        This would mean that groups of IQ 100 individuals might have different average abilities at (remember digits, fix machine, prove theorem) depending on demographic. I do not know whether this is true, but it sounds totally possible.

  31. I was writing a comment, but it got too long and turned into this blog post instead.

  32. Robert Liguori says:

    One: I’m surprised that no one mentioned Semmelweis; he seems like a really clear-cut example of someone who should definitely make the list.

    Two: I definitely agree with the people pointing out that it’s the assumption of scientific consensus where there is none that causes the most harm. It’s a lot easier to laud a few scientists who agree with you and intimidate those who loudly disagree than it is to actually suppress a field of inquiry, and what matters isn’t what the scientific community believes, but what they are willing to say (loudly), and what people perceive them saying.

    • Steve Sailer says:

      I’ve always wondered about Semmelweis and Oliver Wendell Holmes Sr., the doctor and humor writer. Both advocated hand-washing for doctors to prevent transmission of puerperal fever and apparently were ignored. Poor Semmelweis went crazy and died. In contrast, Holmes Sr. lived a delightful life into old age as perhaps the most popular American intellectual.

      https://en.wikipedia.org/wiki/Oliver_Wendell_Holmes_Sr.

    • random832 says:

      As with my example of Wegener, he was also in the actual list but not in Scott’s writeup. His “top ten” are, in fact, the first ten in alphabetical order – A to halfway through D. The article does not make any attempt to rank its examples, and while alphabetical order might be a decent way to pick an unbiased sample (the “nothing up my sleeve” principle), calling it the “top ten” is misleading because it suggests they’re the strongest examples.

  33. Freddie deBoer says:

    You have to drop this Singularity shit, Scott. It is not science; it’s a cult. And it has every hallmark of social non-falsifiability I can imagine. It’s a real blindspot for you. It’s the Rapture for atheists – a concept dreamed up by unhappy people to explain how all of their problems will some day be vacuumed up by a powerful and benevolent force. It’s wishful thinking.

    • Ozy Frantz says:

      As a non-Singulatarian rationalist, I’ve heard “the Singularity is the Rapture for atheists” enough times that it makes me want to donate to MIRI out of spite. I don’t think it is likely to be more persuasive to people who actually think the Singularity is going to happen.

      (To point out only the most obvious of the non-parallels: I have never seen a Christian despair about a possible sign of the Rapture because that means there’s less time to complete their Friendly Rapture research program.)

      • Ilya Shpitser says:

        Ah, but Ozy, the Friendly Rapture Research Program is basically the central life aim of Christianity — “you know not the day nor the hour.”

        • Ozy Frantz says:

          That’s like if Eliezer responded to Roko’s Basilisk with “fuck yeah it will! Everyone give money to AI capabilities research! Friendliness has been solved!” as opposed to “JESUS FUCKING CHRIST NO.”

          • Ilya Shpitser says:

            That Eliezer’s modus operandi differs from a televangelist’s does not really invalidate the comparison. There are, in fact, serious Christians.

            MIRI still asks for money from people who don’t understand what they do at a more than a superficial level (unlike a standard grant, which is peer reviewed). I find that type of fund-raising deeply iffy.

          • Ozy Frantz says:

            Serious Christians are not, to the best of my knowledge, scared shitless of God and engaged in a desperate attempt to convince him to close Hell (although I admit that I’d respect them a lot more if they did).

            If you claim that people believe AI risk is important because it’s “a concept dreamed up by unhappy people to explain how all of their problems will some day be vacuumed up by a powerful and benevolent force”, you have to explain why the vast majority of discussion of AI risk is by terrified people who are very afraid that the AI is going to kill us all. I mean, it’s not that you can’t do Bulverism on it: you can say “they are finding a sense of meaning and purpose by saying they’re literally the most important people in human history, and isn’t it awfully convenient that saving the world involves math, computer science, and philosophy, which they like doing anyway.” But the “Rapture for Nerds” characterization smacks of lazily parroting some thinkpiece you read last week because actually noticing things that are happening in the world is too hard.

          • wintermute92 says:

            If you claim that people believe AI risk is important because it’s “a concept dreamed up by unhappy people to explain how all of their problems will some day be vacuumed up by a powerful and benevolent force”, you have to explain why the vast majority of discussion of AI risk is by terrified people who are very afraid that the AI is going to kill us all.

            Thanks for a nice clear statement of this. I was the kind of Singularitarian Freddie is mocking around age 17, reading Kurzweil and picturing STEM work ending most of the world’s evils.

            Many years of academic work later, I’m still open to the Singularity idea, but it’s in the hesitant, traditionalist, even frightened sense that a bunch of people are messing with something far more dangerous than they’ve planned for or even admitted. My view on over-eager strong AI developers is something like my view on the Soviet bioweapons program – interesting work, but dangerously focused on capability over safety or planning.

            If MIRI types are talking about the “nerd Rapture”, they’re the misotheists.

          • tkolbe says:

            Serious Christians are not, to the best of my knowledge, scared shitless of God and engaged in a desperate attempt to convince him to close Hell

            There is a great deal of fear of God and attempts to help people realize their salvation from Hell. For example, self-Mortification is an ancient Christian practice, and very important in modern Roman Catholic practice. Mortification is among the principle works that religious contemplatives do (e.g. monks/nuns that are cloistered) as an act of reparation to help non-Christians gain the graces to become Christian..

        • Deiseach says:

          That’s because the Rapture is a sub-set of a sub-set of Christianity – a certain strain among some denominations in American Protestant Christianity. As part of historic and global Christianity, it’s “The what?”

          I only heard of the Rapture via this movie which I went to see in the cinema under the impression that it was a good old SF apocalypse movie, and I wasn’t much enlightened otherwise that no, this was a doctrine held by some denominations until years later when I started getting into discussions with American Evangelicals online. When there was a Big Discussion and question to other denominations about “So what do you guys think of the Rapture?” I had to look it up on Wikipedia to find out that officially, us Catholics are amillennialist 🙂

          As to the “concept dreamed up by unhappy people”, the Christian version of that is “wretched urgency“; the fear that you are not engaged enough in soul-winning, that people will be doomed to Hell unless you personally go out and convert as many as you can in the time available, because if you have the chance to save a soul and let it slip, you are morally responsible for their damnation:

          “Witnessing” was the single and sharp focus of the Christian life in my church, and we were suspicious of those liberals who didn’t see the constant urgency of aggressive witnessing to the lost. We weren’t quite out on street corners preaching at the by-standers, but we would have admired that sort of fellow, and we would have probably been told we should aspire to such boldness. The maniac preacher-guy who railed at college girls and boys, calling them whores and hell-bound for make-up, movies and smoking, would have gotten a big love offering at my church. In all of this, it was a short walk to feeling badly about my own sorry and pitiful Christianity. I didn’t fit the mold.

          Feeling badly about things was a key part of the Christian life in my church. We called it being “burdened for the lost.” The ideal Christian lived in hours of weeping daily prayer, interceding and travailing for the lost. (Weeping was very important.) If we prayed adequately, the lost would be saved and revival would come. Every week. Our lack of prayer was always to blame for everything, as was our lack of support for door knocking and confrontational witnessing of every kind. If you weren’t willing to learn the “techniques” of soul winning, you were an example of Christians who didn’t love God or care if people went to hell. In other words, you were like me.

          You can see the fancied resemblance to if we don’t do something now, Unfriendly AI will happen and we’re all doomed! you are not taking this seriously! this is a real problem that has to be tackled right now and it may already be too late!

          • Brad says:

            Y’all don’t have the rapture per se, but you do have the Second Coming which involves judging the living and the dead and ushers in The World to Come, right?

          • tkolbe says:

            On the topic of wretched urgency consider this commentary by St. John Chrysostom, Doctor the Church, recognized by both Catholics and Orthodox (there are only 36 Doctors over 2000 years):

            I do not think there are many among Bishops that will be saved…. Do you not see what a number of qualifications the Bishop must have? To be apt to teach, patient, holding fast the faithful word in doctrine (see 1 Tim. iii. 2–9; Tit. i. 7–9). What trouble and pains does this require! And then, others do wrong, and he bears all the blame. To pass over every thing else: if one soul depart unbaptized, does not this subvert all his own prospect of salvation? The loss of one soul carries with it a penalty which no language can represent.

            Under Catholicism, people are supposed to be very afraid of God (work out their salvation in fear and trembling) and should be doing endless penances/reparations for the good of other souls.

      • ignition says:

        The parallel would be thinking my soul is geared towards hell at the moment and I worry I won’t have time to turn myself around before the world ends.

      • manwhoisthursday says:

        I am skeptical of the singularity, but it seems everyone is entitled to some irrationality, and this one seems benign enough.

        And, for the love of God, the guy who somehow believes we can somehow have “value everyone the same” socialism in a large scale society, yet who admits that the egalitarian structure needed to make that work couldn’t even function in his shitty little activist circle, is in absolutely no position to call other people out for irrational beliefs.

        Singultarianism, at the very worst, causes people to waste time and money, while economic egalitarianism has a history of royally screwing over entire countries, with Venezuela as the latest example du jour.

    • lifetilt says:

      Do you have any theoretic basis for discounting the possibility, or did you get to this opinion purely by social analysis?

    • Sniffnoy says:

      This is just, like… not an actual argument. Seriously, there’s nothing in here that actually attempts to bear on the issue. The closest it comes is A. Bulverism based on analogical reasoning and naïve verbal categories, and the slightly improved B. Bulverism without that additional mistake. Are you trying to actually argue or are you just trying to actually annoy people? This is the sort of thing that makes me want to say “Seriously, go learn what an actual argument looks like.”

      • Deiseach says:

        May I just remark that I very much appreciate an idea (Bulverism) originated by C.S. Lewis being used by rationalists and the rational-adjacent in this, and other, contexts? It’s positively ecumenical! 🙂

    • suntzuanime says:

      From a certain perspective, your socialism shit looks like a cult of unhappy people wishfully thinking about how all of their problems will some day be vacuumed up by a powerful and benevolent force. Would you drop it just because a singularitarian told you you had to?

      • Eli says:

        I don’t think that’s accurate. Socialism isn’t really supposed to be a “powerful and benevolent force”. It’s much more like, “Once I and others have more power over our own lives, we can solve our own damn problems.” Since so many problems are so eminently solvable except for the big ugly signs saying, “Don’t solve your problems with our stuff — the management”, this seems pretty sensible.

        • Nornagest says:

          That’s an awfully libertarian-sounding description of socialism.

          • Eli says:

            Well, if you read up on democratic socialism and libertarian socialism, you’ll see why. “Command and control” and “more government = more socialism” are… largely heterodox, Soviet-centered views of what socialism is.

          • ChetC3 says:

            In terms of their intellectual genealogies, libertarianism and socialism are fraternal twins.

    • Scott Alexander says:

      Oh, now we’re criticizing each other for wishful thinking / cultish beliefs that unhappy people use to explain how all their problems will one day be magically solved?

      Fine, I’ll drop the singularity stuff if you drop socialism.

      • Deiseach says:

        Now, now, children: play nicely! You’re both supposed to be joining together in dogpiling us theists about Sky Fairies and wishful thinking!

        But more seriously – as Ozy mentioned about the irritant effect of “the Rapture for nerds” analogy used as a knock-down argument, causing even those not originally sympathetic to the idea to go all-in supporting it, this is why similar ‘it’s the X for stupid people’ about Creationism makes those of us not Biblical Literalists about Noah’s Ark go “Well all right then, I’m going to go to the wall for the belief that there were two of every creature on a boat, so there!”

      • manwhoisthursday says:

        Fine, I’ll drop the singularity stuff if you drop socialism.

        A-fucking-men.

      • Steve Sailer says:

        Personally, I don’t worry about AI, but that’s because smart people like Scott and James Cameron are worrying about it for me.

      • Jiro says:

        Fine, I’ll drop the singularity stuff if you drop socialism.

        I would count this as a win/win.

      • herbert herberson says:

        Am I really the only one who is terrified of a Robert Hanson-style eternal precariat or of Goldman Sachs building an AI that eats us all in accordance with its fiduciary duties and accordingly think it is very important to think about both socialism and the Singularity?

        • Eli says:

          No, Herbert, you’re not the only one.

          • herbert herberson says:

            Do you have a blog or know of any other communities that take this nexus seriously? I’m mostly a dirtbag left kind of guy, but all the places I like are committed to this same “lol nerd rapture” take

            And, I’d add, it’s not like antagonism should be the only attitude of the left towards this stuff. “Fully automated space communism” doesn’t have to only be a punchline!

        • Ilya Shpitser says:

          You don’t need “AI” in that sentence. Unfriendly intelligence is already here, it’s just kind of slow.

      • wysinwygymmv says:

        deBoer has consistently argued that wishful thinking does not lead to socialism, but that hard work could. He also seems to think that solving problems required good ideas and hard work rather than magic.

        • suntzuanime says:

          Surely building an AI requires good ideas and hard work?

          • Eli says:

            Not according to deep learning hypesters!

            (I know deep learning has given us great results at tasks with nice, known cost functions. Consider this another round of the AI Effect: if you have to state an analytical cost function, it’s no longer AI.)

          • suntzuanime says:

            Surely deep learning has been worked hard on, and at least in the mind of its hypesters is a good idea?

        • manwhoisthursday says:

          Plenty of crazy religions have a “if only we work hard enough and get our heads right, we can bring about heaven” structure. Why should our glorious socialist future be any different?

    • manwhoisthursday says:

      Nah, the rapture for atheists is a functioning socialist economy.

    • herbert herberson says:

      As a socialist, I’m very concerned by powerful agents who are more-or-less divorced from human values and communities acting on the market, the state, and the environment.

      Right now, that mostly means I’m opposed to corporations and other accumulations of capital, as well as individual people who hold significant wealth while being atomized or sociopathic–and that’s where my focus should stay. But in what way does it follow that I should be totally dismissive of AI risk? Even if it’s very unlikely, its worth paying a little attention to–moreover, even if you don’t think we’ll ever have to deal with a hard-launching, self-improving, autonomous AI, we’ll undoubtedly have to deal with rich individuals and companies employing ever more powerful software for their own ends.

      At the end of the day, AI in whatever form is a form of capital, and even in its more likely and less threatening potential flavors it remains a particularly powerful form of capital. It routinely transforms industries, promises to accelerate that process in the immediate future, and almost always does so to the benefit of bosses and the detriment of workers. We shouldn’t be ignoring this shit.

    • Eli says:

      Sure, Freddie, but you’re not really thinking about the safety issue clearly. Does an AI have to be godlike and all-powerful to affect the world, for better or worse? Well, no. History contains plenty of examples of bloody-stupid human beings fucking us all over because they were in the wrong place at the wrong time.

      Why can’t a mildly subhuman AI fuck everything up because they were in the wrong place at the wrong time? Likewise, why can’t a mildly above-human (but still nowhere near all-powerful) AI in the right place at the right time be beneficial?

      Focusing on the extreme cases of “AI never works” and “AI becomes all-powerful” just keeps us from having to think about real life.

      I’ll give a couple of good metaphors I’ve heard from MIRI people for the kind of research they want to do (or see done):

      1) “Moon-rocket alignment”, but for AI. That is, the question is: if you’re going to build a rocket, how do you make it fly to the moon instead buzzing about the tree line, crashing into your neighbor’s house, and blowing up their cat? Or worse, how do you avoid having Apollo 13 spin off into space and drift once you’ve got it off the damn ground? How do you ensure that it goes to the moon, and not some other place your astronauts don’t want?

      2) Just formalize the ability to program, “Move this strawberry from one plate to another with this robot arm” into a reasonably sophisticated AI system. Long before you get to questions of machine ethics, you need to be able to program the machines to perform specific tasks without any deep philosophical content.

    • I would like to hear more about social falsifiability…i could make a guess…

    • Gazeboist says:

      As someone who would love to convince Scott that he’s wrong about AI … please stop making me look bad.

  34. Dan says:

    An anecdote: I know from a coworker that in Spring 2017 the Harvard Kennedy School class on inclusion and democracy begins with the IAT and Stereotype threat, with no reference that these ideas have been taken out to the woodshed by the reproducibility crisis.

    I suggested this was the case for the IAT to the student I know, providing metastudies, and merely vaguely suggested it was also the case of stereotype threat, because said person’s identity is very tied up in believing the things that stereotype threat implies and our relationship is professional. However, it appears to me the political utility to the current iteration of the progressive project to believe these things has led to propagating at the highest levels in the face of evidence to the contrary. I find this frightening.

    • ignition says:

      I think this sort of thing is a major contributor to the tendency for people who find out the truth to go a bit nutty about it.

    • Deiseach says:

      An anecdote: I know from a coworker that in Spring 2017 the Harvard Kennedy School class on inclusion and democracy begins with the IAT and Stereotype threat, with no reference that these ideas have been taken out to the woodshed by the reproducibility crisis.

      Yeah, the same way that Maslow’s Hierarchy of Needs gets uncritically trotted out in every bloody communications class I’ve have to take as part of work training as the infallible ‘and this is what motivates people’ explanation (and not just for the administrative side; it’s used in training childcare workers, nurses, care assistants and people going on to social work, etc).

    • wbrianhall says:

      In the same way that just before reading this I saw an academic Facebook friend linking to a Brown sociology professor’s attack on the Op-Ed “The futility of gender-neutral parenting” or another psychology prof friend bemoaning that David Buss cited her work in his latest paper.

  35. random832 says:

    After seeing the geology one, I searched the list for Alfred Wegener – who I think is a notable example because even people who basically concede he was right sometimes assert that plate tectonics and continental drift aren’t “really” the same thing, as if that is somehow a defense of the establishment idea that the continents had been static for all time.

    EDIT: Turns out he does show up on the actual list, I’d only searched your write-up before writing my original comment.

  36. phil says:

    “Once I realized that there was all this intelligent analysis going on that I’d never heard about, my claim to be boldly defying the scientific consensus evaporated.”

    Are there good ways to not be in this state? I assume Scott was already a relatively proficient googler, so that doesn’t seem like the right answer

    maybe something about knowing the right key words such that it picks up on how an issue is described by people doing the best work on it, but not crowded out by secondary sources that just recite the old conventional wisdom?

    idk, I feel like that is something I think fairly regularly, have a question, then wonder whether there already an extensive literature on this topic?

    ex. thinking about steroids in baseball, when people were just learning about the scope of the problem people were guessing what percentage of players took steroids. Was it 2%, 10%, 25%, 90%? I feel like I heard several different numbers thrown out in the generic media

    Seems like that might be something people have studied, or theorized about, the spread of taboo behaviors, or what sort of factors predict how quickly taboo behaviors spread, the epidemiology of taboos or something like that

    anyway, if there’s an extensive literature on that, I’m entirely oblivious to it

    it’d be interesting to think about ways to be less oblivious to that sort of thing

    • Steve Sailer says:

      I wrote a lot about steroids in sports going back to 1997.

      The real experts, like Bill James, however, didn’t write about steroids even though they were having a bigger impact on baseball statistics than anything in decades.

      Bill James, my hero, had virtually nothing to say about steroids until about 2009 when he wrote a ludicrous article saying that we shouldn’t overlook that Barry Bonds used a maple wood bat.

      The word “steroids’ only appears about once in Michael Lewis’s “Moneyball,” even though the Oakland A’s had been notorious as the home of steroids since Jose Canseco had arrived around 1986.

      Nate Silver, who was a professional baseball analyst before he was an election analyst, wrote a bizarre article about how he imagined we could all agree that Manny Ramirez was unlikely to be juicing.

      A friend of mine who is a baseball agent told me around 1993 after Jose Canseco had been traded to Texas and weird things started happening to Texas players in between hitting home runs: “Jose Canseco is the Typhoid Mary of steroids.” I knew that a dozen years before Canseco published his autobiography about it, but the sabermetricians didn’t seem to want to know.

      These guys were experts, but their incentives encouraged them not to think hard about steroids. So they didn’t.

      I don’t know if that’s generalizable, but it’s a pretty informative case. I’m a big fan of Bill James, Nate Silver, and Michael Lewis. I think they are good guys. But they really flunked the steroids tests. Incentives will do that to you.

      • J Mann says:

        I’m going to write a main level post about this, but there’s a big difference between expert scientific consensus (i.e., scientists who are expert in the specific question being asked) and popular consensus. As fare as I know, there aren’t really “baseball scientists” in the classic model – there are really smart people writing about baseball.

        A big part of your method is finding some popular idea you can be contrarian about, but since you typically base your contrarianism on the scientific evidence, I think it fits Scott’s model.

        • Steve Sailer says:

          I’m actually a counter-contrarian: many reasons are more obvious than we like to imagine they must be. For example, if baseball players are getting massive in the 1990s and hitting huge numbers of home runs, it’s probably less for subtle, complex, counter-intuitive reasons than for the same simple reason sprinters got massive and set world records in the 1980s: steroids.

          • educationrealist says:

            Yeah, I remember having a similar conversation about Lance Armstrong back in 2005. I said I’d never really cared about cycling, but any mild interest I had was killed by the obvious fact that Armstrong was juicing.

            You’d think I’d kicked a puppy or something.

          • Steve Sailer says:

            It’s appropriate that Bill James’ belated contrarian take on why a 40 year old Barry Bonds had broken the record book — is it the kind of wood he uses in his bat? — appeared in Slate.

          • Conrad Honcho says:

            Someone accuses Steve Sailer of being contrarian, so he corrects them that he’s actually counter-contrarian. That’s extremely contrarian.

      • Deiseach says:

        To be fair, there’s also the romantic element in sports about not wanting to think the purity of the game/sport is polluted by things like steroids (or the shenanigans in professional cycling – Paul Kimmage, an Irish former professional cyclist who went into sports journalism, had a personal crusade about doping in the sport and got a lot of grief about it until proven right about the likes of Lance Armstrong).

        We have a sort of hero worship of the sports stars we idolised as children and that hangs on as adults where we don’t want to believe that This Guy or Our Team is doing anything wrong, so you get people writing about maple wood bats and the rest of it. It really is a shattering of illusions and loss of idealism, and while that’s good in a way, it’s not an unalloyed good, because I think there’s room for some idealism in the world, that we need something to aspire to, and the alternative here (for instance) is “hell yeah, make it legal to dope, let athletes use all the resources of modern science, why not? it’s a professional business and all about money, and the money comes from people paying to see records broken, and to break records you can’t do it on talent and hard work alone”.

    • Ketil says:

      Someone on Facebook just posted that vaccines contain body parts of aborted fetuses. So I googled, but the top hit on Google for vaccine ingredients is this (Norwegian), which claims vaccines are ineffective, and contains harmful/toxic ingredients – and is the cause of “most cases of autism”. So much for Google – they prioritize information that is popular. Wikipedia prioritizes information that is sourced. Sometimes these criteria coincides with correctness, but for controversial topics, they often don’t.

      So the scientific consensus isn’t terribly relevant to the information non-scientists see and consume and believe in. You could go to Google Scholar, and get a more reasonable impression – but it takes skill to read and understand scientific publications, and you run the risk of cherry picking whatever supports your view. You could consult blogs by scientists, but since anybody can put up a blog, it boils down to whom you choose to trust.

  37. Walter says:

    “Huh, it looks like the author is writing something that can’t ever be wrong, since if anyone is against the consensus then they can just…” *reads to the part where exactly that is written*.

    Alright, so…like, if you agree that the stuff you are saying doesn’t mean anything (In the sense that even if I agree that the Scientific Consensus is amazeballs then, as you point out, I still have to do the part about deciding what it is and in what way it agrees with me, and if I ever see that it isn’t that just means it isn’t YET and even if it is actively mocking someone who dares to contradict it and embraces their work after hounding them to an early grave it retains my allegiance…), why say it?

    Like, pre this post SSC was like “I am a maverick who follows my own truth, and will bear up under Big Science’s opprobrium”, and after this post you are like “The wisdom of the consensus is vast and great, but always follow your Inner Voice because that is the mechanism we use to avoid everyone doubting consensus but no one being able to say so.” What’s changed? Is it just about emphasis? Like, the posture of Misunderstood Maverick and Orthodox Practitioner don’t seem to offer any particular advantages over one another, and you wouldn’t DO anything different if you were one or the other…

    • Scott Alexander says:

      I agree that this is about messaging and emphasis and not a major shift in how to do things, but messaging and emphasis are important.

      • Walter says:

        Fair enough.

        But you are a shrink, right? Like, presumably if you found yourself with a crackpot dissension vs. the authorities of your profession ethics would guide you, yeah? Like, how does change even come about in an arena where any attempt to do something different is gonna run afoul of “I don’t want to screw up this person’s life by deviating from THE PROCESS”?

    • Deiseach says:

      What I’m taking away from this post is that the story is always more complicated than the simple appealing tale of “Lone Maverick Genius takes on Hidebound Reactionaries and Is Proven Right”, with the conclusion drawn from that that therefore “Every time you hear of someone going up against the consensus, you should be on the side of Lone Maverick Genius”.

      Because sometimes they’re not Lone Maverick Geniuses, they’re crackpots. Or even if they are Lone Maverick Geniuses, they can still be wrong and this particular topic is a hobbyhorse. Sometimes The Establishment is right, a lot of the time The Establishment isn’t as monolithic as it’s presented and there is movement for change and reappraisal going on but this takes time.

      Fight The Power is something we find appealing since for the past fifty years or so we’ve been taught that Question Authority and standing up against the mass of opinion for what is right is the cool thing (and it is, but we’ve been conditioned to think that we or those we hear of who do this are the equivalent of the 60s Civil Rights Activist heroes, and there’s little questioning about maybe we’re not, we’re just being cranky and complaining about not getting our way).

      The story of Little Guy Stands Up To Big Power is a fairytale trope be it Jack The Giant-Killer or Mr Smith Goes To Washington (or Galileo vs The Inquisition and The Church) for a reason; we like the idea of the big bodies in charge of us or the forces that govern our world getting taken down by one of our own, it gives us a sense of control and vindication – plain common sense and hard work/lone genius and hard work win out over the authorities who are trying to tell us what’s what.

      But sometimes the authorities do know best, or are sincerely trying to find out what is true and best. They are not to be automatically distrusted or assumed to be too stuck in the mud and unchanging to have anything relevant to say to us in our modern new day. As a conservative, I approve of this message 🙂

  38. J Mann says:

    I think you’re right, Scott – when you dig down to the actual scientists working in a field, they usually have the best opinion available. The problem is more popular culture – other scientists who aren’t in that specific area signing letters and writing Slate pieces, science journalists, randos misrepresenting what the specific consensus is, etc. If we had a reliable process for determining consensus, that would be very helpful.

    On the other hand, there are plenty of people who argue with Andrew Gelman – do we just assume that the relevant consensus supports Gelman, so things are OK?

  39. James Picone says:

    Part of the problem here is that science is not a homogenous field. You’re combining criticisms of social sciences topics like stereotype threat, complicated biomedical fields like nutrition, statically-observed applied physics like global warming, observed semi-statistical biology/geology in the field creationism, etc. etc., and we’d expect different levels of rigour and consensus-accuracy in each. Broadly, the pattern I’d expect is that the harder the science the more you should trust consensus; physicists en-masse are almost certainly not wrong on details like relativity, quantum mechanics, etc.., because they’re very testable, very predictable phenomena. By the time you get to biology that gets less true, and by psychology or sociology it’s completely out the window.

    The second thing I want to point out is that when people do disagree with scientific consensus, it’s very, very often on far worse grounds than ‘the arguments the new guard used against the old guard’, and this matters. You might have a particular level of trust in scientific consensus derived from the outside view, but when the guy you’re talking to insists that gravity can’t be real because people who live in Australia would fall out of the world, your opinion on his correctness should probably be consensus-based. The problem with that is that identifying disagreements that stupid (or the related “But have they taken into account this simple issue I came up with in 5 minutes?” objection) sometimes requires an understanding of the field people on the whole don’t generally have, and cases where that problem is particularly bad (the field requires understanding complex mathematics or concepts) tends to be cases where terrible-arguments can sound more plausible (because the more abstracted and mathematical the field is the less people have a grasp on it).

    So on the outside view, the more complicated a field is, the more you should trust scientific consensus, because you have less ability to evaluate it and bullshit arguments against it are more likely to seem plausible to you. Although I guess being less sure you can evaluate what the consensus is probably matters here too?

  40. ksvanhorn says:

    Scott, you are poisoning the waters for debate when you refer to AGW skeptics as “global warming denialists.” This is a term of propaganda that implicitly associates skeptics with holocaust denialists. If you want to have an honest, productive discussion you don’t start out by smearing the other side.

    The term is also highly inaccurate in that many / most of the skeptics don’t “deny” that there is warming going on, that some portion of it is manmade, or that there is such a thing as a greenhouse effect. The disagreement is on the details: the accuracy of historical climate reconstructions, whether feedback is positive, negative, or neutral, the degree of manmade warming versus other causes, and the consequences of warming and increased CO2 (there are positives as well as negatives to consider).

    There’s nothing wrong with the term “skeptics.” Why not just use it?

    • Cerebral Paul Z. says:

      The connection with “Holocaust denialist” has always struck me as a stretch. “Denialist” is a useful description for the “it’s all a hoax” crowd; we just need to take more care to distinguish them from the true skeptics, that’s all.

      • ksvanhorn says:

        “Holocaust denialist” was the first connection that came to mind when I first heard the term “global warming denialist.”

        And I have never yet seen anyone who uses the term “global warming denialist” make a distinction between skeptics and the “it’s all a hoax” crowd.

      • ksvanhorn says:

        > The connection with “Holocaust denialist” has always struck me as a stretch.

        Doing a Google search on “denialism,” here are the first four entries:

        The Wikipedia entry for “Denialism” lists “Climate Change” just above “The Holocaust.”

        The introduction to “Denialism: How Irrational Thinking Hinders Scientific Progress, Harms the Planet, and Threatens Our Lives” discusses both “holocaust deniers” and “climate denialists.”

        Both “holocaust denial” and “global warming denialism” are mentioned on the front page of the Denialism Blog http://scienceblogs.com/denialism.

        The misnamed “RationalWiki” website lists “Global Warming Denialists” under the same “Denialism portal” heading as “Armenian genocide denial” and “Holocaust denial.”

        The propagandistic association of AGW skepticism with holocaust denial is pervasive, deliberate, and not even slightly subtle.

        • suntzuanime says:

          The clearest way to see that “denialist” is a propaganda term is to use it in unapproved contexts. Try talking about “pizzagate denialism” and see how far that gets you.

      • 1soru1 says:

        Denialist is a reasonably accurate term for anyone who:

        1. invents a fifth justification for the same conclusion after the first four are shot down.
        2. returns to the first justification as soon as the person doing the shooting down goes away.

        Like all useful terms, it sometimes does get misapplied. But if you lack such a concept in your mental toolbox, you have an open vulnerability to anyone applying that pattern of argument. Which means, sooner or later, someone is going to install a rootkit in your head.

        • John Schilling says:

          In the sense that “retard” is a reasonably accurate term for anyone who suffers a congenital intellectual disability, yes.

          The wisdom of using either of those terms, is another matter.

          • 1soru1 says:

            The history of forming scientific consensus consists largely of people being spectacularly rude to each other. That’s what respect, treating someone as an equal, looks like.

            Seems to work ok. But maybe you would prefer the relation of the courtier and the prince, the salesman and the customer, the grifter and the mark?

          • Nornagest says:

            Scientists are often blunt, rude, egotistical, etc. for the same reasons that any group of nerds is: because spending fourteen years obsessively studying one species of algae or something tends to funge against social skills. Most of the institutions they work in do a fair job of working around the problems this creates, but it’s not essential to the scientific process and it sure as hell isn’t some kind of disguised measure of respect.

          • John Schilling says:

            Seems to work ok. But maybe you would prefer the relation of the courtier and the prince, the salesman and the customer, the grifter and the mark?

            It may be that Rocket Science doesn’t count as Real Science in this regard, but I have often been part of a process that forms a consensus on e.g. how to build an exceedingly complex rocket that won’t blow up, that does not involve calling people “retard”, “denialist”, or any other such thing. Perhaps Real Science could maybe try it that way?

            But then, if Rocket Science isn’t Real Science it is sufficiently science-adjacent that I get to see unambiguous scientists at work fairly frequently, and they mostly seem to avoid the sort of blatant name-calling you are defending here. And if you ever want me to be part of your “consensus”, that’s the only sort that is going to “work OK”.

    • James Picone says:

      I object to the term ‘skeptic’ being smeared by being associated with people with an axe to grind with conventional science. That’s why I don’t use the term (or use it in quotes).

      The holocaust denial link is bullshit and you know it.

      If feedback is negative or neutral, ice ages are impossible. Ice ages happened.

      • ksvanhorn says:

        I think it is no accident that in the same post in which you support the use of the term “denialist” you also demonstrate your inability to engage in civil discourse.

        • James Picone says:

          From my point of view, you’re yet another person with approximately zero understanding of a very important topic, criticizing it in ways which make approximately no sense and which have been addressed over and over again.

          I see no reason to be civil with people who refuse to address fundamental issues with their claims while they also smear me by claiming I’m trying to tie them to white supremacists.

          Claiming that feedbacks are negative or neutral is deep into alternative physics or geology. Why should I be civil with somebody who hasn’t bothered engaging with the topic to the extent necessary to admit the existence of ice ages?

          • ksvanhorn says:

            You’re doing an excellent job of proving my point. Did it ever occur to you that you could just link to some references providing the details of your claim on feedback?

          • ashlael says:

            “I see no reason to be civil…”

            Here’s a reason. I’m a global warming sceptic. I’m open to being convinced otherwise. But I don’t put much stock in “it’s all so obvious and you’re all so stupid” arguments.

            If you want to expound on your “Ice Ages = positive feedbacks” theory on the other hand I would be very interested to hear it.

          • cuke says:

            I assume that people who call themselves global warming skeptics would be climate scientists of some kind, or at least scientist enough that they have engaged at depth with the research and still find themselves unconvinced by the evidence for reasons they can explain with reference to the evidence.

            There’s of course room for dissenters in climate science, but that dissent needs to be based on some mastery of the current evidence base and specific objections to it. Otherwise, it seems like it’s skepticism as a political stance. And then it seems like that’s more like saying “I don’t like what those global warming scientists are concluding because of how it’s getting talked about in the media.”

            I was teaching about global warming at university 30 years ago and have followed the research since then as an interested scientist who works in another field. I have to defer to the preponderance of evidence gathered by a majority of scientists in the field because I have no alternative set of credible evidence to stake my own scientific claim against it.

            Are those saying they are global warming skeptics here speaking as climate scientists who are qualified to assess the evidence at some depth? What degree of engagement with the data is your skepticism based on? More generally, what degree of engagement with a body of scientific research do you think a person outside the field in question needs to have in order to take a contrarian stance to that body of research?

          • The Nybbler says:

            @cuke

            No, a “climate scientist” is almost by definition someone on board with the AGW “consensus”.

          • James Picone says:

            You’re doing an excellent job of proving my point. Did it ever occur to you that you could just link to some references providing the details of your claim on feedback?

            Let’s just do the maths.

            Here‘s a NASA page claiming 4c to 7c warming since the last glaciation. Let’s take 4c; a smaller temperature variation means you need less of an energy differential is worse for my position.

            Climate warms by ~1k for every 3.7 W/m**2 forcing at the top of atmosphere – this corresponds to an equilibrium climate sensitivity of 1k for a doubling of CO2. Wikipedia covers some of this; Here is Roy Spencer, no friend of global warming, saying that “It has been calculated theoretically that, if there are no other changes in the climate system, a doubling of the atmospheric CO2 concentration would cause less than 1 deg C of surface warming (about 1 deg. F). This is NOT a controversial statement…it is well understood by climate scientists.”. Making that number larger hurts my case; because it means you get more temperature change for the same change in energy balance.

            So we need ~-4k * 3.7 W/m**2 of forcing to go from present-day conditions to ice-age conditions (and more than that to go back, because albedo increases dramatically). 14.8 W/m**2.

            That same Spencer blog post quotes 235 W/m**2 to 240 W/m**2 as downwards energy at top of atmosphere. That 14.8 W/m**2 of forcing corresponds to an energy balance change of ~6% of incoming solar radiation, which is goddamn huge. For example, that corresponds to octupling the CO2 in the atmosphere. Here‘s a plot of solar energy as viewed by Earth over the instrumental period (to convert to top-of-atmosphere, divide by 4 because Earth’s a sphere, then multiply by ~0.7 for albedo). Notice how the range of variation compares to 14.8 W/m**2.

            This is all linear, so if the ice age was actually 6k cooler you can multiply by 1.5 to find out you need -22.2 W/m**2 to get into it (and again, more to get out of it because Earth’s albedo gets smaller).

            I assume that people who call themselves global warming skeptics would be climate scientists of some kind, or at least scientist enough that they have engaged at depth with the research and still find themselves unconvinced by the evidence for reasons they can explain with reference to the evidence.

            I dunno about you, but I can count the scientist ‘skeptics’ I know of on two hands. Judith Curry, Roy Spencer, John Christy, Roger Pielke, Richard Lindzen. Nic Lewis I guess? Not professionally trained, but he’s published interesting and useful papers in the literature.

          • Adam says:

            Here’s a reason. I’m a global warming sceptic. I’m open to being convinced otherwise. But I don’t put much stock in “it’s all so obvious and you’re all so stupid” arguments.

            I sympathize tremendously with James, because he’s been making well-referenced, reasonable arguments citing the most current science for years, and I’m sure it grows frustrating to do that and still be confronted with people coming in and objecting to shit he’s been over hundreds of times at least. But you wouldn’t know unless you’ve been following comments here for a very long time, and all you see is the frustration.

    • Douglas Knight says:

      “Denialist” was invented in 1990. It was not about the Holocaust, but the Armenian genocide.

    • JonathanD says:

      I spend a lot of time in this neck of the woods, though I only occasionally comment. People of my political persuasion are mocked, constantly, as social justice warriors. Take it with a grain of salt and move on.

  41. ksvanhorn says:

    The replication crisis is not a good example (yet) of science righting itself. Yes, there are prominent people worried about it. There are also prominent people fighting reform, the problem is far from solved, and it’s not clear how to fix it. (Yes, there are technical fixes, but the problems here are institutional / social: perverse incentives to do the wrong thing that are very difficult to change.)

    I’m with you in general that science will generally right itself — but that can take decades.

  42. veeloxtrox says:

    Typo: In the 3rd paragraph of point 4. You say “onbe of the top journals in the field” I believe onbe should be one.

  43. Cerebral Paul Z. says:

    “I previously thought the scientific consensus was flawed because it failed to take the replication crisis seriously enough.”

    This seems like an unusual reaction. I’d think most people who are worried about the replication crisis would say the scientific consensus is flawed because it’s based on so many results which failed to replicate.

    • Scott Alexander says:

      I think I’m using “scientific consensus” to mean “the people doing science” here, and judging their moral fiber, rather than caring about the individual results at this stage.

  44. wysinwygymmv says:

    @SA:

    I realize I’m (ironically) risking making my narrative of scientific success unfalsifiable. Suppose someone wants to argue that scientific consensus is wrong. If they point to something it used to be wrong about, I can respond “Yes, but it self-corrected and it’s correct now, so that’s fine.” If they point to something where cutting-edge scientists say it’s wrong but nobody else agrees, I can respond “Yes, this is what the beginning of a paradigm shift looks like, so that’s fine”. And if they point to something where nobody in the field thinks it’s wrong, I can say “You’re a crackpot for going against all reputable scientists; the problem is with you.” And if later they turn out to be right, and everyone acknowledges it, I can say “Yes, but it self-corrected and it’s correct now, so that’s fine.”

    Maybe the better way to say “science is correct” would be “science generates models that help to understand specific aspects of the world subject to various assumptions, and then tests those models. The models are actually tested pretty rigorously, and obviously-misleading models are culled before they create too many problems.”

    The concept of big-T “Truth” is intuitively obvious but philosophically problematic, and we can never really know how well scientific knowledge reflects “Truth” (what would we use as a yardstick?). And there’s also Baudrillard’s point about the map vs. territory: that “Truth” is a map that is exactly as big and detailed as the territory that it is describing and is therefore useless as a map.

  45. spN44p8 says:

    About the replication crisis, you date the beginning to 2005, but Paul Meehl was writing about in the 1970’s. He was treated respectfully, but it didn’t make a dent in the way things were being done.

  46. Matt M says:

    “All my enemies are dead, so I have no one to gloat over.”

    Can’t tell if blessing or curse…

  47. VolumeWarrior says:

    A major way in which science is regressive is that they focus on problems that just don’t matter. There’s a myth that one day all the thousands of tiny papers will combine to support some giant breakthroughs. But most papers are just irrelevant, don’t get cited outside their tiny little niche, and ultimately disappear. Who cares if you found a different model to explain X, wrote some dumb paragraph about why it’s 15% better than other models, and then present your results? This is likely just noise. And as someone who has tried to replicate “famous landmark” models and failed, the majority of the field is utter garbage.

    Work backwards. Which problems really really matter? Mass agriculture to feed the world matters. Studying effective systems of governance matters. Life extension and cryonics matter. No. Please don’t research a really rare subtype of cancer because now you’re just going back to your esoteric little cave. Define the big problems with the modern human experience and solve them. This is what science should be doing but instead we worry about pulsars and quasars and whether dinosaurs had some intermediate blood temperature.

    Don’t get me wrong. The real science IS being done. It’s just that when apple comes out with the iPhone8, it isn’t called “science”, even if the smartphone revolution is really important. I’m sure pharmaceutical companies are working on wondrous eternal-youth elixirs they can patent. But the stuff called “science” in general parlance is utter garbage. Irrelevant, opaque, and usually “right” in so-narrow a sense as to be functionally wrong.

    • rlms says:

      Do you apply this attitude more widely, and decry everything that doesn’t “solve big problems with the modern human experience” (namely, everything that isn’t frantic work on the issues you have somehow determined matter)? If not, why not?

      • VolumeWarrior says:

        Yes and no. So take something superfluous like videogames or producing new music. There’s a sense in which those are “shallow” problems, and there’s a sense in which they’re “big” problems because we consume them every day.

        I’d certainly prefer it if we spent more time taking life-extension seriously and less time perfecting leisure goods. But both are very far away from mainstream “science”.

        It’s also a branding problem. I don’t have a problem with people who study esoteric topics, particularly if it gives them pleasure. But if you’re going to call yourself a “scientist”, the implication is that you are doing important work; that you’re on the frontier carving things out for the rest of us to use later one day.

    • Scott Alexander says:

      95% of papers could be replaced with the words “Look At Me, I Published A Paper So You Should Give Me Tenure” repeated for twenty pages, with no loss to the scientific enterprise, but I’m not sure how easy it is to know which ones those are beforehand, and I don’t think it’s as simple as a basic research/applied research dichotomy (if that’s what you’re claiming)

      • VolumeWarrior says:

        I am willing to throw the baby out with the bathwater on this one. Private researchers seem to do a reasonable job innovating. It’s not clear we get a good return on the “thousand monkeys at a typewriter” strategy.

        University “scientists” are called scientists because they’re in the public eye and very prolific in terms of sheer volume. They’re also propped up by “smart people” who like to debate using citations on the internet, and “peer reviewed publications” sound prestigious and are easily found.

        … whereas no one can hyperlink what walmart knows about consumer psychology. But I’ll bet you it’s pretty accurate.

    • Enkidum says:

      This is, with all due respect, utter garbage. All important innovations (almost by definition) are unpredictable and do not arise from directed study on a particular topic. Figuring out stuff in general, in as many directions as possible, without regard for utility has led to the most measurable improvements in the human condition in the shortest space of time in history. You’re so utterly and completely wrong about this that it’s hard to comprehend how to explain it to you.

      • VolumeWarrior says:

        I’m sure corporations can work out the optimal balance of direct-aiming vs. random cross pollination. But career scientists at public university aren’t producing ideas at random. They are producing very specific and narrow research, most of which hasn’t been done *exactly* because it is pointless and obscure.

        So you don’t wind up with a primordial soup of inventiveness. You just have a bunch of people not talking to each other and not reading each others’ work. There is very little cross-pollination.

        • Enkidum says:

          I’m having an awful lot of trouble pretending to respect this opinion. The vast majority of great scientific ideas since “science” was an acknowledged field have come out of career scientists at public universities. I suppose you can point to, e.g., Bell Labs and Xerox PARC as places with heavy corporate involvement that have produced true innovations, but (a) they were generally as much publicly-funded as privately, and (b) their output has been a drop in the bucket compared to work done at the university.

          Can you give any examples of what the hell you think you’re talking about? Other than that we shouldn’t research dinosaurs and astrophysics because it’s not immediately useful?

          • VolumeWarrior says:

            Computer processing. Pharmaceuticals. Aaaalll of the top websites like facebook, youtube, etc.

            Where do you think the technology for self driving cars is going to come from?

            You might object that these are too “pedestrian” or not “fundamental” enough. But they’re exactly analogous to scientific researchers trying to come up with better numerical algorithms or study rare forms of cancer. Except the people who hypothesized that online picture-sharing might be popular didn’t write a psychology paper about it.

          • Enkidum says:

            So fundamental scientific progress is… Facebook and Prozac? And this is more important than astrophysics, to the extent that we should abandon the latter in favour of the former? I… what? I have nothing left to add other than helpless flailing, sorry.

          • VolumeWarrior says:

            If you can find a way to do real useful astrophysical research, fine. My position is not that society ought to abandon fundamental work. But what’s being done now in astrophysics is mostly arcane. Researchers focus on topics very narrow, difficult to verify, and difficult to cross-apply.

            I would be lukewarm to a true astrophysics renaissance. But if we doubled the amount of astrophysical research, we’d merely double the amount of esoteric minutiae in the name of “science”.

            The really useful science isn’t being called “science” because walmart doesn’t write peer reviewed papers about their distribution networks.

          • Ketil says:

            To be fair, it seems a common sentiment (in politics, at least) that science should be more useful. Here, it tends to mean that publicly funded scientists should do corporations’ R&D for them, I get the impression that VolumeWarrior holds the more libertarian view that the public should simply spend less on science, and leave it to corporations. (Apologies if I’m misunderstanding). To me, the production of knowledge is a goal in itself, and usefulness is difficult to estimate in advance, and practically impossible to dictate from above. (See also: Lysenko)

            I guess you could make the same argument for art – 95% of what is made is easily forgettable, if noticed at all. But occasionally, there’s a Tolstoy, Mozart, or van Gogh. The question is, would we still get those if we slashed all funding of art, and left all art production to PR agencies – whose art is useful?

          • James Picone says:

            Computers, most modern computer processing, the Internet, and basically everything that makes the modern internet work were invented in universities by publicly-funded grad students.

            Facebook and Youtube are built on the backs of generations of people just doing some interesting fiddling at uni. Or if you’re looking at them over WiFi, government-owned-and-funded research organisations.

          • Paul Brinkley says:

            The question is, would we still get [great artists] if we slashed all funding of art, and left all art production to PR agencies – whose art is useful?

            The modern day has Hans Zimmer, Jimi Hendrix, Joss Whedon, Robert Heinlein, John Updike, Jeremy Soule, Marcin Przbylowicz, Mike Judge, Peter Gabriel, Yes!, The Beatles, Johnny Cash, Jim Henson, Metallica, Steven Spielberg, Meryl Streep, Philip Seymour Hoffman, Patrick Stewart, Gene Roddenberry, Skaven, Daft Punk, Depeche Mode, Ayn Rand, Eddie Murphy, Andy McKee, Soren Johnson, Waylon Jennings, Wynton Marsalis, CCR, Anne Geddes, Enya, Madonna, Michael Jackson, Carl Sagan, Neal Stephenson, Stan Lee, Will Smith, B. B. King, Stephen King, John Updike, Carol Burnett, Don Cheadle, Madeline Kahn, Frank Sinatra, Sia, Lin Manuel-Miranda, Prince, Penn and Teller, Martha Argerich…

          • Deiseach says:

            Aaaalll of the top websites like facebook, youtube, etc.

            If Facebook is going to be The Shape Of Things To Come, then Unfriendly AI that will turn us all into paperclips can’t come about fast enough to spare us all.

          • Machina ex Deus says:

            Computers, most modern computer processing, the Internet, and basically everything that makes the modern internet work were invented in universities by publicly-funded grad students.

            The Internet was invented at BB&N.

            Ethernet, which is almost certainly what your IP packets are traveling over, was invented at Xerox PARC. As were laser printing and WIMP user interfaces.

            The transistor, which you need a few billion of in your computer’s CPU and GPU, and somewhat smaller numbers in pretty much every other chip, was invented at Bell Labs.

            The first integrated circuit was at Texas Instruments, and the first silicon one was at Fairchild Semiconductor.

            The first microprocessor was produced by Intel.

            The U.S. Government gave us Ada.

          • Enkidum says:

            I don’t think it matters all that much for the argument, but (a) I have no idea what BB&N is, (b) I’ve always understood the internet to have two parents, one being the stuff Berners-Lee was doing at CERN and the other being ARPANET, neither of which were universities per se, but both of which were purely government-funded, and (c) Bell Labs and I believe several of the other examples you mention were heavily publicly-funded as well.

            I say it doesn’t matter much for the argument because the claim from the non-insane side was never that corporations don’t produce any research of utility, simply that most research of utility has come out of universities, and that investing a great deal into pure research with no applied goals is a major component of the best way to produce future research with applied utility. I consider at least the latter point so blatantly obvious from history that I’ll be unable to do much other than make fun of the people who disagree.

          • James Picone says:

            @Machina ex Deus:
            TCP, UDP, ARPANET, the WWW at CERN, the majority of UNIX, etc. etc.

          • The Nybbler says:

            Unix comes out of Bell Labs.

          • FacelessCraven says:

            @Ketl – “To me, the production of knowledge is a goal in itself, and usefulness is difficult to estimate in advance, and practically impossible to dictate from above.”

            Define knowledge. If I type random strings into a word doc each day, that is technically “data”, and if you memorize them that is technically “knowledge”. There are probably an infinite number of wrong papers it’s possible to write on the causes of, say, obesity. If we write them in sequence, each debunking the last while laying out new false conclusions to be debunked by the next, is that producing knowledge?

            From the outside, a lot of modern science and academia has come to resemble random bit-flipping. Where are we trying to get, and how much progress are we getting from the effort put in?

    • psmith says:

      the smartphone revolution is really important.

      wew lad

    • Deiseach says:

      Please don’t research a really rare subtype of cancer because now you’re just going back to your esoteric little cave.

      This matters to the people who contract that really rare sub-type of cancer (and their families). I agree that “who cares if 10,000 people out of a global population of 7 billion suffer from this, when tens of millions are suffering from not having food security?” is an argument that can licitly be made, but researching a rare cancer is not the intellectual equivalent of playing with your Legos, and it may have wider implications than merely alleviating pain and suffering for those 10,000 human beings.

      There’s no reason that both the small close problems and the big long-term problems can’t be worked on under the aegis of science.

      • VolumeWarrior says:

        My biased assumption is that we’ll solve rare-cancer when we solve aging. Hell, maybe we’ve already solved it and the answer is “cryonics until cell-repairing nanobots”.

        Your argument is good though. Where we disagree is that I actually do think having a small number of researchers try to cure a rare form of cancer is akin to playing with lego. In that their success rate is vanishingly small and their work will not be meaningfully peer reviewed.

        • Deiseach says:

          It’s entirely possible that the solution will be the other way around: by working out what the hell is going on with the cells in this rare cancer, discoveries applicable to the process of aging will be made.

          Look, the worth of practical and applied versus theoretical is an old argument that was going on before either of us were born and will be going on after we’re both in our graves. There’s much to be said for both, and certainly Science as much as anything else has to sing for its supper. But pure theoretical knowledge for the sake of knowledge lays the foundations of what the practical work then derives, and it may be a long time down the line before the use is seen because it took that time for the practical side to be able to catch up with putting the theory into practice.

          Stripping away all ‘useless if it doesn’t butter any parsnips’ research may turn out to be sawing off the branch on which we’re sitting when it comes to the new discoveries that the practical and market forces will put into use.

    • grrath says:

      This is the most wrong comment I’ve ever seen. Every single scientific breakthrough has been on the backs of relatively obscure information. It’s not like English farmers and laborers where waiting with bated breath on Newton’s formula’s. The way the world works is fairly mundane but it’s through that understanding that we manage to create all of the “important” work. There’s no way to separate the two.

  48. Rusty says:

    The big complaint made on Coyote Blog is not that the scientific consensus is wrong about the fact that greenhouse gases cause global warming. Rather his objection is that he doesn’t believe there is strong scientific consensus for the strength of the feedback loops which turn that warming into a runaway train.

    I have no idea if he is right about this but the way the consensus is presented (at least in the papers I read) never seem to address this issue. So is it possible for the consensus to be right but misrepresented?

    • Scott Alexander says:

      I’m not sure I’ve heard anyone claim there’s scientific consensus about the feedback loops. In fact, I thought most people said the strong version of that theory was kind of fringe. I thought IPCC predictions are pretty linearish.

      • crescentsmom says:

        Whoa! It looks like my thinking has been so strongly influenced by the Coyote Blog presentation (mentioned by Rusty) that I just assumed we were all talking about the same thing. So I’m kind of shocked to see this comment from Scott. If I may, here are the important propositions given in that presentation:

        1. Increased CO2 causes global temperatures to rise
        2. The rise in temperatures caused by CO2 leads to an additional rise in global temperature, as positive effects (e.g. ice melting causes more absorption of solar energy) dominate
        3. The total rise in temperature described in points 1 and 2 leads to large negative costs, on net, for humans

        Since seeing the argument presented this way, I can say that I think point 1 (CO2 -> warming) is true, I have some doubts about 2 (warming -> more warming), and I am very skeptical about 3 (warming -> global catastrophe).

        I always assumed Scott was saying skeptics should feel bad for not believing 3, since that is the scientific consensus. And I *did* feel bad about disagreeing with Scott and with science in general. But now are you saying, Scott, that I am already part of the mainstream consensus, and thus can worry slightly less about being a total crackpot?

        • rlms says:

          Define feedback effects and large negative costs. Feedback effects leading to temperature increases of tens of degrees and human-extinction-level disaster is fringe (or rather it is widely considered to be unlikely, which is not quite the same thing). Feedback effects changing a 1 degree rise to a ~3 degree one (with significant uncertainty) are mainstream, as far as I know (IANAC). I think large negative effects from that kind of rise, in the sense of tens of millions of deaths/refugees, are also mainstream (again AFAIK).

        • Nornagest says:

          My understanding of the consensus position is that greenhouse warming alone isn’t enough to explain the observed temperature changes, so there’s good evidence for positive feedback in the temperature ranges we’re used to. Positive feedback doesn’t necessarily imply a catastrophic runaway process, though, or the biosphere would have destroyed itself coming out of the last interglacial; it just means that the CO2 signal is amplified somewhat. (There is uncertainty over how much “somewhat” is.)

          One problem is that the popular media often takes bad-but-not-catastrophic mainstream predictions (displacement of coastal populations, hundreds of billions of dollars in damage over decades or centuries) and spins them as catastrophic, sometimes mixing them freely with much more speculative positions and getting away with it thanks to scope insensitivity.

        • James Picone says:

          Nornagest is correct.

          Consensus position as expressed by the IPCC is that feedbacks add between 0.5k and 3k to every 1k of direct warming, with the 0.5k to 1k range considered unlikely and 2k as the best estimate (don’t compound those; that’s already been taken into account). There is a thing called ‘runaway greenhouse effect‘; it basically occurs when temperatures get high enough that the oceans boil away. It’s almost certainly not possible on Earth; we’re too far away from the Sun.

          I think some people get mislead by the use of the term ‘positive feedback’; usually people who’ve studied control theory. The term is used differently in climate science; from a control theory perspective you should include the (negative) ‘planck feedback’ from the Stefan-Boltzmann law.

          EDIT: And I’m going to be a broken record and point out that if you don’t have that feedback mechanism it’s very, very difficult to come up with plausible mechanisms for ice ages, because of how large the temperature swings involved are.

    • James Picone says:

      Looked up the blog. Superficially looks like he’s talking about; actually doesn’t. For example, in this recent blog post, he appears to be (mistakenly) adding the equilibrium climate sensitivity values the IPCC quotes to the first-order warming effect of CO2 to get the IPCC warming estimate, and then he’s assuming that we’ll get the full ECS warming on a fast timescale, when we won’t get to it until the climate returns to equilibrium (because it’s the equilibrium climate sensitivity). The combination of those two makes him fuck up all his comparisons.

      The actual value to use is the transitional climate sensitivity, which the IPCC estimates at 1k to 2.8k in AR4

      0.8k gives a TCR of ~1.36, which with a standard-ish 0.7 TCR to ECS ratio gives an ECS of 1.9. Positive feedback, but pretty low on the IPCC’s ranges. That’d be because a) IPCC TCR is defined for a 1%/year increase in CO2, we’ve been going at well below that and b) natural and other anthro influences on temperature were negative over the last century or so, so the best estimate for how much of the current warming is anthropogenic is ~110%. That last one is easy to compensate for (just multiply by 1.1), the first one really requires a climate model.

  49. phil says:

    10 years seems like actually a long time to be in the professional wilderness

    if the average professional career is from ~25 – ~65

    that’s betting 1/4 of your professional life on something that might not have a payoff

    if we’re using the Kelly criterion to manage our portfolio of professional time, you need to be pretty certain of something before you invest 10 years of being professionally ridiculed

  50. deciusbrutus says:

    Suppose that I wanted to find what the distribution of scientific opinion on three climate-related questions were. Where would I, a mortal, find the opinions of scientists on:

    What global climate is most desirable?
    What level of at atmospheric greenhouse gases will lead to that global climate?
    What level of human production will lead to that level of those gases?

    If I can’t trust the Vox-quality media on social science, I can’t trust them on climate science. For all I know there’s a huge body of work that would support the implicit association bias as being important, but it’s not getting published because of political pressure.

    • Matt M says:

      What global climate is most desirable?

      I’d settle for just this one. Almost every time I’m in a debate regarding climate change, I ask it, and the other person refuses to answer.

      • Randy M says:

        To be fair, that question by itself would be misleading, if stopping an increase in temperature was much more difficult than starting one.

        • Matt M says:

          So answer the question, and then add that clarifier.

          • Randy M says:

            I agree with the implied point that some warming isn’t necessarily even harmful, but am just cautioning that this doesn’t mean all fears of warming are irrational.

          • Matt M says:

            That’s not helpful. I want a number. The ideal global average temperature. What is it?

          • Skivverus says:

            Warning: bullshit* ahead.

            I have no idea, and very little personal interest in finding out because I don’t actually simultaneously live on all points of the earth’s surface. That said, I do have some idea of the ideal local average temperature, and I imagine a global survey of opinions on ideal local average temperatures would be a decent approximation (I think 72F is the closest thing to “conventional wisdom” there is on this, but I could be wrong).
            For further refinement, that survey would ideally also have a question to the tune of “how much warmer/cooler is where you live than what you’d like?”; after summing up all those, and correcting for population density/understanding/lizardmen/etc, then I’d have a preliminary answer.

          • Erl137 says:

            Matt, your question is deeply misleading, impossible to answer for two reasons, and irrelevant for a third reason.

            1) The “ideal global average temperature” requires making tradeoffs that we are not equipped to make. Take apples and oranges; they flourish in different climates. If I like apples, and you like oranges, then you and I will have different preferences about the “ideal temperature”—I’d prefer a colder world, with abundant apples, and you’d prefer a warmer one, with abundant oranges.

            Adjudicating between two people with differing preferences is a thorny problem; it’s not clear what the “correct” response is. The two primary options we tend to employ—market and voting solutions—each have huge, glaring failures. (Market solutions overweight the rich; voting solutions equally weight the differently-motivated. For starters.)

            Adjudicating between ALL people with differing preferences on ALL POINTS (which is what setting a planetary temperature would require) is both unsolved and fundamentally unsolvable in an empirical sense—it requires vastly more data than our society can collect about itself, and it turns on philosophical questions that probably can’t be effectively resolved. It would be a stretch to ask this of a philosopher king; to require your interlocutors in a chat about climate to solve two unsolvable problems before you’ll listen to them is, frankly, quite silly.

            Climate policy in addition would require solving the question of future generations, changing preferences, and incommensurable economies between different worlds, all of which are hella intractable.

            2) The question is not of the global temperature, but of the global climate—a complicated system whose various linkages are poorly understood. Suppose for example we decided that a slightly warmer planet would be better, since so much of the earth’s landmass (i.e., Siberia and Northern Canada) is uncomfortably cold. So we kick the climate up a few degrees.

            Unfortunately, this collapses the Atlantic Gulf Stream current, and the temperature in the northern hemisphere plummets.

            The Gulf Stream is one of the most famous countervailing mechanisms in the climate. But there are huge numbers of them, many of which we don’t understand, and many of which are path-dependent. So even if we could somehow rank between different climate outcomes (as per 1, we can’t), we would struggle mightily and in vain to predict precisely which outcomes would result from a specified temperature.

            3) Neither of the above really matters, because all of the concern in global warming has to do with the state change costs, rather than an arbitrary ranking among states. Maybe a colder planet has better skiing and cheaper server farms. Maybe a warmer planet has better beaches and higher crop yields. So the fuck what? The process of getting there in mere centuries involves mass extinction, vast human misery, and untold expense—in relocation, mitigation, and other costs. THAT’s why climate experts neigh-universally prefer to slow climate change rather than target it—because even if you could choose a preferred target climate (you can’t) and even if you could determine the temperature target that would bring you to that desired climate (you can’t), the cost of getting there would be untold ecological and human damage.

            (Which is to say, something like what rlms said below—though with perhaps a more florid tone.)

          • Randy M says:

            That’s not helpful. I want a number. The ideal global average temperature. What is it?

            “Don’t talk to me about rate of change, tell me why I shouldn’t be on the ground floor!”

          • Controls Freak says:

            Neither of the above really matters, because all of the concern in global warming has to do with the state change costs … The process of getting there in mere centuries involves mass extinction, vast human misery, and untold expense—in relocation, mitigation, and other costs.

            Can you point to any justification for this that does not simply do timescales the wrong way ’round? Economic/political/technological/etc. timescales are fast compared to climate timescales. The simplified thought experiment that most people do to get “mass extinction, vast human misery…” holds all of the economic/political/technological/etc. dynamics constant, makes a big jump in the climate dynamics, and then evaluates the fast system. This is completely and obviously wrong to anyone who has done introductory dynamical systems at the graduate level.

          • Paul Brinkley says:

            @Erl137, everything you said here seems fairly agreeable. It also seems to support Matt M’s and deciusbrutus’ underlying point. If climate is as unknown and poorly understood as you portray, then the current claims of its trends, let alone the unrelated-to-climate claims about the costs to mankind, ought to be very weak.

          • Erl137 says:

            “Can you point to any justification for this [i.e., ‘mass extinction, vast human misery, and untold expense’] ” that does not simply do timescales the wrong way ’round?”

            Sure.

            1) Mass extinction is ongoing, at geologically unprecedented rates. I don’t know how to parse between climate-change-induced mass extinction, and human-civ induced mass extinction, but it doesn’t really matter that much—the timescale question you ask doesn’t arise.

            For 2 & 3—vast human misery & untold expense—my model of the dynamic relationship between human and climate systems suggests that human behavior will change in response to crises, rather than in a gradual, effective, ongoing fashion.

            So, for example: I expect Manhattan to build levees, rather than to relocate its population to Westchester; and I expect this trend to continue until an unusually northerly hurricane + higher sea levels (both climate change consequences) comes over the sea wall, kills a bunch of people, and destroys a great deal of property.

            I expect low-lying Bangladeshi cities to flood, badly, rather than to all move to the top of the hill.

            And so I expect relocation to occur not on geological timescales, and not on economic timescales, but on catastrophe timescales. The net impact may be several centuries of relocation to higher ground, but I expect it to occur piecewise via local catastrophes.

            Rather than for crops and global farming strategies to shift in an orderly and productive fashion, I expect to see wider variability in crop yields, leading to some years of famine.

            & so on.

            This seems consistent with the 1) climate change predictions of greater variability of climate outcomes, 2) behavior of previous societies in the face of climate catastrophes (c.f. the Late Bronze Age collapse) and 3) the present behavior of humans.

          • Aapje says:

            @Erl137

            Exactly, there is a huge amount of investment that people are just not going to walk away from until the cost of moving is less than the cost of staying.

            The current gap between the two is the potential for suffering that people will accept as the price for staying where they are.

          • Erl137 says:

            Also, geological timescales are real slow right until they really aren’t. The Bretz scabland story above is instructive: Bretz proposed that the scablands in Washington were the result of titanic ancient floods. The geologists laughed at him, and said that geological changes occurred slowly and consistently over geological timespans.

            But Bretz was right. What happened was that a huge amount of water was dammed up behind a large glacier. As the temperature rose, the glacier slowly, slowly melted. Everything was real gradual.

            And then one day the inhabitants of Eastern Washington looked up to find forty cubic kilometers of water per hour headed at them with the literal speed of a freight train.

          • James Picone says:

            @Paul Brinkley:
            Damage is nonlinear in temperature; uncertainty therefore increases the average damage estimate.

            Looked at another way, say I put 1% probability in the Clathrate Gun hypothesis, with associated very bad effects, because the data doesn’t quite exclude it and I’m trying to compensate for possible systematic bias.

            If I become more uncertain about how bad global warming will be, that pretty much guarantees I have to increase the probability I assigned to the clathrate gun, because I’ve just lost confidence in all the evidence that it won’t happen.

            That argument is even stronger for climate sensitivity, because it’s bounded very hard on the left.

            For a real-world example, remember that scientists didn’t predict the ozone hole. What’s climate science’s ozone hole? I don’t want to find out.

          • random832 says:

            The geologists laughed at him, and said that geological changes occurred slowly and consistently over geological timespans.

            Why was this position considered relevant for a change to surface terrain? Why did they consider this to be a “geological change”?

            (And “every 55 years for 2000 years” seems like a geological enough timescale for something that depends on ice activity.)

          • The Nybbler says:

            @Erl137

            Privileging the current temperature because adjusting to change results in vast human misery and untold expense doesn’t work, because attempting to hold the current temperature stable (by eliminating greenhouse gas emission) also results in vast human misery and untold expense. The global economy runs on energy, and energy means CO2 production (and given that global population is increasing as is energy use per capita, ever-increasing CO2 production). Putting a brake on that will require untold expense and result in vast human misery.

            As for Manhattan, sea level has been rising, but the island has been getting larger; humans seem well able to keep up. Hurricanes have been making to to that latitude every few years for at least 200 years; you’d be hard pressed to claim one as the result of global warming.

          • Controls Freak says:

            my model of the dynamic relationship between human and climate systems suggests that human behavior will change in response to crises, rather than in a gradual, effective, ongoing fashion.

            Are you aware that this model of human behavior directly contradicts standard economic theory? Does it bother you that when people try to integrate climate models with models in other domains, they tend to ignore the basic principles that govern those domains (and the dynamical systems properties that govern their interrelation)?

            I agree that some people will choose to put themselves in harm’s way. Guess what? They do so now. I’m not seeing any reasoning for why you think their ability to handle risk assessments is utterly disconnected from reality in this one special case. This also highlights that you haven’t done anything remotely like solve the timescale problem, because if true for climate dynamics, your claim would also be true for faster political/economic dynamics… and we’re right back where we started.

            Rather than for crops and global farming strategies to shift in an orderly and productive fashion, I expect to see wider variability in crop yields, leading to some years of famine.

            …because we’ve literally never figured out how to store reserves. I’m starting to get twitchy, because you’re really making me think that you just missed the last several thousand years of human history.

            Re: Bretz

            I totally agree that to the extent that you can actually predict a change in short timescale events due to climate change, it’s a real problem. That’s why I did actually care about the (kind of strangely defined) claims about hurricanes. However, to date, those claims have been mostly walked back. You can’t just appeal to, “But some other guy made short timescale claims from a long timescale system!” You have to actually make supportable short timescale claims from this long timescale system.

            Damage is nonlinear in temperature

            This is almost certainly true; it would be remarkable if it were linear. That being said, we have no bloody clue what the function actually looks like (written by a pro-mitigation person!), and anyone who appeals to terrible static estimates are almost certainly just blowing smoke. My favorite line from the cited paper:

            For the reasons cited [in the paper], not only do we not know the approximate magnitude of the net benefits or costs of mitigating climate change to any specific level of future global temperature increase over the next 50–100 years, but we also cannot even claim to know the sign of the mitigation impacts on GWP, or national GDPs, or any other economic metric commonly computed.

            When we don’t have even a justifiable estimate of the sign of the “damage” (can we at least call it “impact” rather than “damage”, because the latter implies that we know the sign), we should probably pause and step back.

          • James Picone says:

            @Control Freak:
            I feel pretty confident concluding that making the equator uninhabitable (at around 6c over preindustrial) will be pretty firmly negative.

          • lycotic says:

            @Erl137

            At the risk of forking this discussion horribly, do you have a pointer to evidence that the Late Bronze Age Collapse was due to climate rather than, say, the sea peoples?

          • random832 says:

            @lycotic

            At the risk of forking this discussion horribly, do you have a pointer to evidence that the Late Bronze Age Collapse was due to climate rather than, say, the sea peoples?

            Who were the sea peoples and why did they do whatever they did? To my understanding, one plausible theory is that they were in fact a wave of refugees from some (maybe climate, maybe other) catastrophe.

            Wikipedia backs up this understanding:

            Hypotheses regarding the origin of the various groups identified as Sea Peoples remains the source of much speculation.[12] These theories variously propose equating them with several Aegean tribes, raiders from central Europe, scattered soldiers who turned to piracy or who had become refugees, and links with natural disasters such as earthquakes or climatic shifts.[2][13]

          • Erl137 says:

            @ random832: “Why was this position considered relevant for a change to surface terrain? Why did they consider this to be a “geological change”?”

            I don’t know much about the subject; I’ve only read Gould’s chapter on Bertz in The Panda’s Thumb. My understanding is that geology took it as a general principle that all macro-scale land features were the result of gradual processes, and in particular proposals of giant floods tipped all the anti-Genesis-crackpot alarms. The scablands, for context, are thousands of cubic kilometers in size, and so their emergence seems to meet my intuitive criteria for “geological change” but perhaps I’m missing a nuance you’re working with.

            A good example of what we mean by geological time is “the Rockies emerged from millions of years before to millions of years after the dinosaurs”

            @Lycotic re: Late Bronze Age collapse

            Haven’t read the papers myself. This looks like a good start.

          • John Schilling says:

            To my understanding, one plausible theory is that they were in fact a wave of refugees from some (maybe climate, maybe other) catastrophe.

            Wikipedia backs up this understanding

            There is no understanding to back up. Nobody understands what caused the Bronze Age Collapse. There are a wide range of theories with inconclusive and sometimes contradictory bits of evidence for each of them, and any encyclopedic reference will elaborate to some extent on each of them without saying “this is obviously false”.

            If from this you say, “…and therefore wikipedia backs my explanation”, you lack understanding.

            If you go from there to “…and because the Bronze Age Collapse could happen again, you must support the political policies I claim are appropriate to preventing future collapses”, sensible people will lack the patience to deal with you further.

          • Erl137 says:

            And now for the shouty bits. Stop me if you’ve heard this before.

            —no, seriously, please actually do so. I popped into this thread because it seemed like Matt M was genuinely unclear about the (imputed) arguments of his interlocutors, and I hoped to give him a more useful take on their case, so that he wouldn’t have unproductive conversations where he demanded a particular commitment with little relevance to the discussion at hand.

            . . . and because I got a bug up my ass about the whole thing, I confess.

            But the replies I’ve garnered haven’t exposed me to (m)any new thoughts, and if my replies are similarly unoriginal to all of you—and I suspect they may be—then I’m not especially interested in continuing to churn them out. So I repeat, in deadly earnest: stop me if you’ve heard this before.

            @Paul Brinkley: “If climate is as unknown and poorly understood as you portray, then the current claims of its trends, let alone the unrelated-to-climate claims about the costs to mankind, ought to be very weak.”

            Well, we can make pretty robust predictions about the behavior of macrophysical processes, even if we can’t speak to the localized outcomes.

            Consider the following analogy. You and I are hiking through the woods of Japan when we come across a pagoda and enter it. I soliloquize briefly about pagodas—how differently they are constructed from Western edifices, how they avoid nails and screws and glue, how they distribute weight quite cleverly, how really we two know so little about them.

            Whereupon you take out an axe and begin to hack at various pillars and struts.

            Quoth I: “what the fuck are you doing?”
            Quoth you: “you said it’s not immediately clear which pillars are load-bearing!”

            Or suppose you had set fire to the pagoda—we don’t need to understand the structural feedback to know that you can burn the place to the ground.

            We don’t need to predict which precise ecological outcomes will result from climate change. All we need to understand is 1) the general category and 2) the general magnitude of the change. The category is “dramatic shifts away from presently stable systems” and the magnitude is “a bunch”.

            At least, that’s what the best science seems to say. (Recall, if you please, that we are literally in a thread OP of which is entitled “Learning to Love Scientific Consensus”.)

            @Nybbler:

            “Privileging the current temperature because adjusting to change results in vast human misery and untold expense doesn’t work, because attempting to hold the current temperature stable (by eliminating greenhouse gas emission) also results in vast human misery and untold expense.”

            A sensible argument, and I don’t know enough to weight them.

            I suspect that I’m more optimistic about the upside of decarbonization (“eliminating greenhouse gas emissions” is an implicitly exaggerated framing; since in the Glorious Carbon-Neutral City of Tomorrow I’m still going to be farting up a storm, such emissions will never be entirely eliminated) than you are, and less pessimistic about its downsides. In particular, I bet you a dollar that when the histories of decarbonization are written it proves to be net positive, i.e., that every dollar spent on decarbonization has more than one dollar in social return. But that’s really an underinformed hunch, based on generalizations from quite different cases.

            But I think we can agree that all else equal, we would prefer that the climate stay right where it is, which answers Matt’s question above rather handily.

            “As for Manhattan . . . Hurricanes have been making to to that latitude every few years for at least 200 years; you’d be hard pressed to claim one as the result of global warming.”

            This is a silly argument, and I suspect you know it’s silly. Of course no particular hurricane only exists because of global warming. But if global warming shifts the distribution in the direction of more frequent, more powerful, and more destructive storms, then reducing the magnitude of global warming will reduce our estimated damage from it.

            “As for Manhattan, sea level has been rising, but the island has been getting larger; humans seem well able to keep up.”

            Sometimes with major cities we beat sea level rise and subsistence; sometimes we fall behind. If the rate of sea level rise increases, more cities will fall into column B (compared to a scenario where the rate of sea level rise does not increase). Is Manhattan especially likely to fall into the second category? I don’t know; I’m not a climatologist. On the one hand, it has a great deal of low-lying waterfront areas and subterranean infrastructure which already flooded in Sandy; on the other it’s built on bedrock rather than a delta and it’s rich as shit, so it can afford to pursue resource-intensive strategies. But I don’t mention it as the first major city to go—I mention it as an emotionally salient example because I’ve lived there. So’s that’s all I mean.

            @ Controls Freak

            Hi, Controls. I feel like we’re not communicating effectively, and I think it’s because at least one and possibly both of us is making and/or perceiving more heroic claims than we really intended.

            So let me try to be clearer.

            Of course I expect that the economic system will price in different sea levels, the change between “this house is beachfront property” and “high tide comes into my kitchen.” And I believe people (with sufficient wealth in modern economies) will move accordingly.

            However, I also expect that global warming (i.e., sea level rise and increased frequency of serious storms) will increase the risk of storm-related catastrophe to waterfront property.

            (An aside: you can cancel the large storms entirely and climate change will still worsen large storm outcomes. One of the popular responses to rising sea levels is levees/seawalls. The more levees you have, and the further the city’s ground level is below sea level, the worse outcomes are available from similar-magnitude storms—because you can breach the seawall and flood downtown.)

            This increased risk may or may not be effectively priced in. (It’s empirically demonstrated to my satisfaction that even sophisticated financial consumers don’t always accurately price in the risk of large catastrophic events; c.f. the 2008 crisis, every financial panic ever.) Let’s suppose it IS priced in. Then the people who move into such property will be those with the lowest estimate of the likelihood of such a severe event (“there hasn’t been a hurricane here in fifty years!” “this house has never flooded!”) or those unable to pass up on the savings.

            As with the lottery paradox, the median estimate of the magnitude of the risk will be correct. Everyone exposed to the risk will be those who estimated it lowest (as they’ll underbid others). And so people will be exposed to outcomes far more severe than they themselves expected. And they will lose their property and their lives.

            Do you believe this model “directly contradicts standard economic theory”? If so, frankly, so much the worse for standard economic theory. We observe this behavior all the time. But I don’t think it does.

            “your claim would also be true for faster political/economic dynamics… and we’re right back where we started.”

            I am genuinely perplexed by what you mean here. I believe that global climate change increases the risk of catastrophic outcomes, and so it increases the amount of human suffering. I believe that political changes can increase the risk of catastrophe—war, violent revolution, etc.—and thereby increase the net expected amount of human suffering. I believe that economic systems can have higher risk of catastrophe—endemic overleveraging—and thereby increase the net expected amount of human suffering. I don’t see these claims as being in tension.

            I understood your initial claim to mean “because the economy is a dynamic system evolving more rapidly than the climate, it should efficiently price-in the changing climate so that its human impact is minimal.” Is that your claim, or have I misunderstood you?

            (You also write that “However, to date, those claims [i.e., about hurricanes] have been mostly walked back.”—can you explain what you mean by this, and cite it if necessary?)

            [in response to my suggestion of the risks of famine]: “because we’ve literally never figured out how to store reserves.”

            The existence of reserves has not prevented catastrophic famine over the past several thousand years. I agree, however, that the model I was sketching was insufficiently nuanced to capture the dynamics of (modern) famine.

            (Although, quick: how many years of disappointing grain yields would completely consume the U.S.’s present national strategic corn reserve? How bad would they have to get? I have no idea—but I suspect you don’t either.)

            In general, (modern?) famine is a combined resource/economic phenomenon. Rarely is food so scarce that nobody can eat; but in bad years, food prices rise sharply enough that the poor can’t afford to pay.

            At the same time, poor crop yields mean that farmers are unable to make their expected return. Prices rise, but not enough to compensate for falling outputs—and especially not if reserves are opened to control prices. (And today, 475 million small farms exist, meaning call it a billion people (maybe more!) dependent on either the food grown or the money derived from a small farm.)

            So in a year of bad harvests, food prices go up while economically marginal farmers get poorer. This is a perfect recipe for a modern famine, which we don’t know how to solve.

            Of course, in some parts of the world, climate changes and endemic famine are still directly linked.

            Re: Bretz:

            I wasn’t deriving a specific claim. I was rebutting your general claim that “Economic/political/technological/etc. timescales are fast compared to climate timescales.” This is often true. Sometimes it’s not.

            And finally, not to step between you and James, but you linked to an article where David Roberts opens by complaining that he finds two major academic publications which agree with each other to be silly, and then explicitly describing his quest to find some paper, any paper, that will confirm and express his pre-existing beliefs and rebut the academic consensus, but from someone with the appropriate credentials.

            In case readers think I’m joking:

            How much will it cost humanity to avoid dangerous climate change?

            Debates over this contentious topic inevitably focus on a set of economic models that seek to estimate those costs over the course of the next century. The latest such analysis is the 5th IPCC Assessment on Mitigation of Climate Change, which has people confidently asserting that properly addressing climate change will trim 0.06 percent a year off global GDP growth. Not 0.05 a year. Not 0.07. But 0.06. The much-ballyhooed New Climate Economy report, out earlier this year, draws many of the same conclusions.

            I’ve long had my own opinion on this, but I’m just a dirty hippie blogger. So I’ve been patiently waiting for somebody with credentials to express it for me, in a fancy peer-reviewed paper.

            Well, lucky me. I found the paper. (Thanks to the tweeter who flagged it for me — sorry I can’t remember who you are.) And it confirms my considered opinion on the matter

            It’s hard to find a more impressive example of the proverbial Man of One Study. Roberts explicitly says “I read two large reports which agreed pretty closely with each other, thought they were stupid, and waited until someone on Twitter sent me a paper that agreed with me.”

            Does that mean he’s wrong? No. But his procedure is entirely—yea, valiantly—ass backwards.

          • Controls Freak says:

            Gah. There are so many premises that I agree with, yet you’re just not quite there.

            I also expect that global warming (i.e., sea level rise and increased frequency of serious storms) will increase the risk of storm-related catastrophe to waterfront property.

            I assume that you mean it will increase the risk to property that is spatially-located differently (i.e., that is further inland). Ok. How is that different from the risk to property that is temporally-located differently? Suppose the counterfactual of no climate change. A catastrophic storm happens. People look at that location and say, “Yea, that was a freak storm. I’m going to build there again.” They do so according to the best estimate of risk. Later, another catastrophic storm might happen, because that’s the nature of risk.

            Now, here comes the kicker. Since the climate system is slow, fast-system economic choices will be building to a locally-constant risk profile.

            Let’s suppose it IS priced in. Then the people who move into such property will be those with the lowest estimate of the likelihood of such a severe event (“there hasn’t been a hurricane here in fifty years!” “this house has never flooded!”) or those unable to pass up on the savings.

            …or maybe the people who think, “It’ll still be cool to have this prime piece of beachfront property for a mere thirty years.”

            As with the lottery paradox, the median estimate of the magnitude of the risk will be correct. Everyone exposed to the risk will be those who estimated it lowest (as they’ll underbid others). And so people will be exposed to outcomes far more severe than they themselves expected. And they will lose their property and their lives.

            This is standard risk, and you should notice that climate change never came into it. Whether climate is constant or locally-constant, people will have different risk assessments, and depending on how they weight the benefits, there will certainly be winners/losers, yes. This actually brings up my favorite challenge (which mostly demonstrates what you’ve accepted by assumption – that risk is priced in). If you’re so sure that you’ve understood the modeling so well and quantified the risk so accurately, then simply go make your money. Go buy a bunch of not-quite-beachfront property, betting that you’ll be able to flip it as prime beachfront property only in the next thirty years or so. If you’re not quite convinced that such a plan will produce a suitable ROI… or maybe you even think, “Man, but what if I buy near Atlantic City instead of Myrtle Beach and totally unrelated economic considerations cause my city’s economy to die before I can cash in” (because those dynamics are much faster)… then you’re starting to get it.

            I believe that global climate change increases the risk of some catastrophic outcomes

            FTFY. The rest of your sentence is a non sequitur. The development of AI, financial systems, and bloody airplanes increase the risk of some catastrophic outcomes. They also produce gobs of utility for billions of people. I’m on board with trying to quantify the risk involved, but I’m not on board with stopping society until we actually have a defensible quantification.

            I believe that political changes can increase the risk of catastrophe—war, violent revolution, etc.

            This happens on timescales far faster than climate change.

            I believe that economic systems can have higher risk of catastrophe—endemic overleveraging—and thereby increase the net expected amount of human suffering.

            …are you about to suggest abolishing banks? Abolishing currency? I’m not sure what you’re going for here.

            I understood your initial claim to mean “because the economy is a dynamic system evolving more rapidly than the climate, it should efficiently price-in the changing climate so that its human impact is minimal.” Is that your claim, or have I misunderstood you?

            That is one way of characterizing it. My main claim is that every damage assessment paper I’ve read is just ludicrously stupid in light of basic dynamical systems theory.

            Re: Hurricanes – my understanding is that AR5 reduced the confidence in these assessments to the point of being “not much”.

            Re: Famine – you’ve stated “it can exist”. Agreed. Now, how does a slow climate system actually affect this? Famines are generally a very acute phenomenon.

            Re: David Roberts – dude who has been (and remains) pro-mitigation had sneaking suspicions about certain confidence in particular estimates. He turned out to be right. He’s still doesn’t comprehend how bad the problem is (…that’s what I’m trying to tell you, lol).

            Man of One Study

            He cited two papers by Ackerman et. al. and one by Rosen/Guenther. I’d definitely add Pindyck to the reading list of people going all the way through the publishing process just to say, “Maybe we should hold our horses a bit.” In academia, that’s a big step. In control theory, usually it just gets to the point of being a jerk during a conference presentation: “Would you fly on a plane controlled by your controller?”

            You’re literally reading a post about how scientists are generally on-point with criticizing possibly-wrong paradigms and then throwing away published, scientific criticisms of the past paradigm because you don’t like some guy’s smell reflex. That’s not even Man of One Study. That’s Man With One Guy Who Might Have Smelled Something And That Somehow Redeems The Rotted Meat.

            But seriously, if you have any technical reason why Ackerman, Rosen, Guenther, Pindyck, Barker, and, uh, me are all wrong on this, I’d love to hear it.

            And not to step too far into your other conversations, but:

            We don’t need to predict which precise ecological outcomes will result from climate change. All we need to understand is 1) the general category and 2) the general magnitude of the change. The category is “dramatic shifts away from presently stable systems” and the magnitude is “a bunch”.

            This is why I’ve emphasized that people are starting to point out that we don’t even know the sign of the potential impact.

          • random832 says:

            @John Schilling

            There is no understanding to back up. Nobody understands what caused the Bronze Age Collapse. There are a wide range of theories with inconclusive and sometimes contradictory bits of evidence for each of them, and any encyclopedic reference will elaborate to some extent on each of them without saying “this is obviously false”.

            If from this you say, “…and therefore wikipedia backs my explanation”, you lack understanding.

            You’ve misinterpreted what I said. What I was saying Wikipedia backs up is that this is a legitimate theory that exists and not something I just imagined from nothing, not that it’s necessarily the correct explanation. It certainly supports my claim that “was due to climate rather than, say, the sea peoples” is not necessarily even meaningful (since the sea peoples may have been ’caused’ in turn by climate events)

            If you go from there to “…and because the Bronze Age Collapse could happen again, you must support the political policies I claim are appropriate to preventing future collapses”, sensible people will lack the patience to deal with you further.

            I said no such thing.

            @Control Freak

            Now, here comes the kicker. Since the climate system is slow, fast-system economic choices will be building to a locally-constant risk profile.

            The core claim of climate change is that the climate system, in the current situation, is not that slow – that the timescale is centuries or less, which is the same as or shorter than the timescale it takes for the economic system to “evolve” its way to a consistent profile, rather than the conventional tens of millennia.

          • Paul Brinkley says:

            @Erl137

            I appreciate your efforts in devising an analogy. To me, that analogy looks like you left out the following:

            – I’m not simply hacking away at various pillars. I’m removing relatively small chunks of wood at a time and creating a vast array of other things with them – homes, books, pagoda-wood-fueled turbines, and so on.

            – I’m burning small bits of pagoda wood (again, in order to create other useful things), but not in a way that will clearly burn the place to the ground, esp. when the pagoda seems to be really big compared to the fires I’m setting, and even seems to be regenerating itself in the places I’m leaving alone.

            – Me setting fire to the entire pagoda would be more analogous to me managing to direct an asteroid into the earth, or figuring out how to set off all the volcanoes at once, or maybe just the Yellowstone caldera. But I’m very clearly (to me) not trying to do that.

            All in all, you seem to be trying to simultaneously claim that the climate works in mysterious ways, but doesn’t work mysteriously in this one way, that doesn’t look to anyone else to be any less mysterious. You’re insisting on the right to write the book on mystery, but writing it mysteriously – we can’t tell the difference between “mysterious mechanism” and “simple mechanism which only appears mysterious” apart from your saying so.

            I have, indeed, heard this idea before. But I don’t mind repeating the counter to it for now, as you seem to have either not heard it, or perhaps forgotten it temporarily.

            So again: if you want to get into new territory, entreating us to accept your prognostications on what is and is not mysterious will not suffice. Rather, you will need to share a way to distinguish the two mechanisms above, which in this case will require producing a method to predict climate, on timescales and at magnitudes relevant to much faster-moving economic and political systems (cf. some of the posts Controls Freak has been making here).

            While I’m at it: contrary to random832’s claim, it seems clear to me that economic systems are evolving much faster than climate ever will, short of hurling the earth into the Great Red Spot. The former are driven by cycles of as little as fourths of a year, as well as yearly and quadrennial (cf. all the attention to the economy every general US election).

          • Controls Freak says:

            The core claim of climate change is that the climate system, in the current situation, is not that slow – that the timescale is centuries or less, which is the same as or shorter than the timescale it takes for the economic system to “evolve” its way to a consistent profile

            It is? Really? I missed that part in the IPCC reports. IIRC, there was only one chart that even spoke to timescales (and it was my favorite; it gave the biggest cause for concern). I agree with you entirely that it’s not on the order of millennia, but have you seen economic systems? Economists generally don’t give a specific number to distinguish between short run/long run, but I am very comfortable claiming that the long run is still significantly faster than the relevant climate dynamics expressed by IPCC.

            Anyway, even if you really stretch and claim that the systems aren’t timescale-separated, you just made the problem worse. Now, we have to simulate both systems together! It was only by the grace of timescale-separation and weak-dependence that we could trust the climate models in the first place. If you’re killing that, then we probably have to go back to the drawing board for climate models.

      • rlms says:

        I think that comparing climates/temperatures statically is largely missing the point. The dangers of marginal climate change (as opposed to worst-case scenarios) isn’t really that other climates would be objectively worse, but that adapting to change would be very costly. There’s no fundamental reason why the population of Bangladesh should live there rather than somewhere else; the issue is that moving huge numbers of people is difficult.

        • ksvanhorn says:

          rlms, you sound a lot more reasonable than the people I seem to encounter on the “climate alarmism” side, who think that global warming is an extinction-level threat and smear anyone who suggests it might not be quite that cataclysmic as a “denier.”

          Anyway, if the point of being worried about global warming is that it may be costly to adapt, then it is economists we should be listening to, and the costs of reducing carbon emissions (including opportunity costs like reduced economic growth) become a very important factor in policy decisions. My impression is that the costs of actually reducing carbon emissions is much higher than the costs of adapting; part of that reason is that we’ll have the better part of a century to adapt. (David Friedman has addressed this issue on his blog.)

          • pontifex says:

            If the clathrate gun hypothesis is true, climate change could be an extinction-level event for humanity.

            If you argue that scientists don’t understand the climate well, you should also believe that we should stop radically changing the composition of the atmosphere, because we don’t know what we’re doing.

        • Controls Freak says:

          the issue is that moving huge numbers of people quickly is difficult.

          FTFY. But seriously, I make this objection over and over again, and I really am honestly waiting for a response. Moving large quantities of people on timescales of centuries is not particularly difficult. We have some examples of people large quantities of people due to economic/political/weather/climate/etc conditions. Sometimes, it’s done extremely rapidly, and that tends to have a pretty high cost (e.g., new borders are drawn, and if you don’t get on the other side now you’ll be killed along with anyone else in your ethnicity; entire regions see economic conditions change extremely rapidly – see Dust Bowl). Sometimes, it’s done much more slowly (e.g., Farmer Joe has a son John who sees that it’s becoming less profitable to farm, so he goes to school to become a mechanic, but that requires moving into a nearby down; Mechanic John has a son James who sees that it’s becoming less profitable to be a small-town mechanic, so he goes to school to become an engineer, but that requires moving cross-state to the city…).

          • Paul Brinkley says:

            Reminds me of an article someone once shared a few years back, exclaiming that Cape Canaveral would be underwater if the IPCC’s more drastic claims proved true. I noted that if those claims were true, it would take on the order of a century for the space center to be in peril, during which time it was likely to move several times for entirely unrelated, mundane reasons, such as Congressional pork drift.

            People aren’t sequoia groves. They’re constantly moving in and out of all areas, coastal or otherwise, for routine reasons like job opportunities or high school graduation. The cost of moving a lot of them is already being borne, all the time.

          • rlms says:

            I think the standard timescale for significant sea level rise is less than one century, not multiple (I might be wrong, I’m not a climatologist). But movement of people isn’t going to be evenly distributed across time, as people won’t leave their homes until it is unavoidable, and one of the causes of movement will be sudden events like storm surges (made worse by higher sea levels). If we started moving everyone who would be affected now, it would be OK (in the sense of being the same order of magnitude of refugees/year as the Syrian civil war by my estimate (climate refugees to 2100 between 10 million and 1 billion, 6 million Syrian refugees in 6 years)). But that isn’t happening, and would be very expensive to make happen as few people are being affected yet.

          • James Picone says:

            It’s well within physical possibility that we make the equator uninhabitable by humans within a century. Temperature increase of ~6c makes wet bulb temperature in the equator sometimes exceed 35c, at which point resting humans in the shade die of heatstroke.

            If that seems implausible to you, keep in mind that at 4c to 7c below preindustrial the Boston area was buried in a kilometre of ice. 6c average is a big number.

            For an outside view, consider how your approach would have handled CFCs, leaded petrol, smoking, or acid rain, and the implications for the modern day.

          • Controls Freak says:

            I think the standard timescale for significant sea level rise is less than one century, not multiple

            We’ll need to come up with a definition for a standard unit of ‘significant’, but this isn’t really important. It is >> (if not >>>) when compared to political/economic/other timescales. That’s the important bit. We don’t need to know precisely how much greater it is to know that we can perform timescale separation, which demands particular mathematical methods.

            movement of people isn’t going to be evenly distributed across time

            It isn’t already, and no one would ever claim such a thing.

            people won’t leave their homes until it is unavoidable, and one of the causes of movement will be sudden events like storm surges (made worse by higher sea levels)

            I’m kind of with you until the parenthetical. I’m only kind of with you, because people actually do consider risk assessments. They already consider estimates of things like whether they’re in a flood plain when making decisions. I jump off entirely at the parenthetical, because I don’t think it’s justified except by running the timescales the wrong way ’round. You have to let the fast system converge before you take a step in the slow system. At each time in the slow system, the fast system is going to consider the risk environment and converge to a solution. Therefore, we actually need to model all of the fast system in the future (probably impossible). I’d settle for at least a justifiable prediction that short timescale events will be worse given higher temperatures, but most people have walked back from those predictions.

            Frankly, any “climate refugees” prediction is almost certainly bupkis, because we have terrible fast-scale models, and these estimates are done by (again) running the timescales the wrong way ’round. I don’t know which ones you’re citing, but I would be willing to bet all the money in my pocket right now (ok, my pockets contain nothing but a turn of phrase) that they make the exact same error… because everybody does it. What else can you do?

            Temperature increase of ~6c

            …are not within any reasonable range predicted by IPCC.

          • James Picone says:

            @Controls Freak:
            RCP8.5 with climate sensitivity on the 5% high end would do it. Clathrate gun would do it (so that’s maybe 1% depending on how certain you are in the science that says it’s not a thing). How much of a betting man are you?

            EDIT: Actually RCP8.5 with slightly-below-best-estimate CS still gets to it, just not by the end of the century. RCP8.5 has CO2eq = ~1250 at 2100, ln(1250/270) / ln(2) = 2.2, 6 / 2.2 = 2.7, climate sensitivity of 2.7 gets you there with CO2eq as it is in 2100, probably by 2030 or so.

          • rlms says:

            @Controls Freak
            I think there are two problems with what you are saying. Firstly, what fast-scale models are you talking about here? I agree that the precise political and economic impacts of climate change, like the precise political and economic impacts of anything, are pretty much impossible to predict, because politics and economics are incredibly difficult to generally model. But people leaving their homes because they’re underwater or have been destroyed by flooding is not a complex issue. Modelling it only requires a basic understanding of human biology and psychology: people can’t breathe underwater, and like having roofs and walls.

            Secondly, you seem to be making a fully-general argument against all long-term political and economic predictions. But it is possible to make some perfectly accurate predictions of that sort: it is easy to predict that in 2100 the US will have more people who will view gay marriage more favourably and own faster computers than today.

          • random832 says:

            @rlms, I don’t agree with him, but it seems clear his objection is to the notion that they won’t move until their homes are destroyed (or some near (or nearby) catastrophe throws the possibility into sharp relief), rather than that they will move after.

          • rlms says:

            If you don’t move until your home is destroyed, surely you logically have to move after it is destroyed? Until implies you can’t move beforehand.

          • random832 says:

            @rlms

            Until implies you can’t move beforehand.

            I think you’re confused. Were you not aware that you made the claim “people won’t leave their homes until it is unavoidable”?

            I even agree with your claim, but it seems you’re not comprehending that this (or perhaps the implied “it will become unavoidable for a lot of people at once, rather than a manageable number of people one at a time”) is the claim Control Freak is objecting to.

          • Controls Freak says:

            @James

            Literally no portion of any of the recent IPCC reports get that high. I don’t care about the back of your envelope.

            @rlms

            Let’s start with the second first.

            you seem to be making a fully-general argument against all long-term political and economic predictions.

            No. I’m specifically calling out the interaction of fast political/economic models with slower climate models.

            people leaving their homes because they’re underwater or have been destroyed by flooding is not a complex issue. Modelling it only requires a basic understanding of human biology and psychology: people can’t breathe underwater, and like having roofs and walls.

            Agreed. That happens on an extremely short timescale. Climate change happens on a long timescale. When we have timescale-separated systems, we let the fast timescale system converge first, then step forward in the slow timescale system. Let’s see if you can agree to a non-politicized example, and then we can see how it plays out in the instant case. Consider an aircraft using fuel over the course of a flight. The change in weight and center of gravity due to fuel burn is a slow dynamical system; the altitude/orientation is a fast dynamical system. The correct way to evaluate this coupled system is to see that at time t1, the pilot will set a desired altitude near the most efficient altitude, and the orientation dynamics will converge to steady-state. Slowly, an appreciable amount of fuel will have been consumed, and we consider a time t2. The pilot will set a new desired altitude, and the orientation dynamics will converge to steady-state. Do you agree that this is the case?

            The wrong way to analyze this system is to imagine that a whole bunch of fuel suddenly disappears. Not only would there be a “damage function” which describes how much efficiency is lost from flying at the incorrect altitude, but some people might believe there is a CG Gun Problem that makes the orientation dynamics unstable! (I can reproduce this type of behavior in a flight simulator with a sudden change in fuel.) Do you agree that this is an incorrect analysis?

          • rlms says:

            @Controls Freak
            “No. I’m specifically calling out the interaction of fast political/economic models with slower climate models.”
            As I understand it, you are saying that we can’t predict the effects of climate change on politics and economics, because those systems have a fast timescale, predicting a fast-timescale system’s reaction to a slow-timescale system requires modelling the fast-timescale’s behaviour at each timestep, and doing so is impossible for politics and economics because those are very complicated (whereas in your plane example we could model the fast-timescale system, as it is fairly simple). I agree that politics and economics are too difficult to model as whole systems, and I have no reason to doubt you are correct about the general control theory. But if you take climate change out of the equation, you still conclude that politics and economics can’t be modelled (that’s one of your premises) and therefore we can’t make any accurate predictions about them. I don’t think that bit follows.

            I think you that’s because you are failing to make a distinction between modelling politics and economics as whole systems, and predicting specific variables. I agree that some variables in the system (the US president, the inflation-adjusted price of gold) are too difficult to predict, but in my previous comment I suggested other variables that seem perfectly predictable. That is the sense in which your argument is too general.

            I agree that some climate-change-relevant variables, such as the overall economic impact, are like the price of gold and can’t be predicted. But things, e.g. number of refugees seem more like the population size, as that mainly depends on a small number of factors (homes destroyed). Given a range of climate changes, a range of values for that can be predicted with reasonable accuracy.

          • Controls Freak says:

            if you take climate change out of the equation, you still conclude that politics and economics can’t be modelled (that’s one of your premises) and therefore we can’t make any accurate predictions about them. I don’t think that bit follows.

            I didn’t say that, because I didn’t take climate change out of the equation. My concern is purely the interaction of the two systems. If you hold the climate system constant, you might be able to say a few things about a faster system. For example, I would agree entirely with your prediction that we will have faster computers in 2060. Now, how fast are our computers in 2060 (holding climate constant) versus how fast our computers are in 2060 (with some estimate of climate change)? We simply don’t have the necessary resolution in the fast system.

            I agree that some climate-change-relevant variables, such as the overall economic impact, are like the price of gold and can’t be predicted.

            Read that sentence again. Think about it for a bit.

            But things, e.g. number of refugees seem more like the population size, as that mainly depends on a small number of factors (homes destroyed). Given a range of climate changes, a range of values for that can be predicted with reasonable accuracy.

            I don’t think you can. Let’s start with an estimate of the population distribution map in 2060 (holding climate constant). Go. (With error estimates.)

          • rlms says:

            OK, I see what you’re saying now. I agree that we can’t calculate the effect of the climate on things such as computer speed. But I don’t think that control theory comes into it; the reason for that is because climate change should have a negligible effect on that. In comparison, it will have a very large effect on mass migration from low-lying coastal areas.

            As I’m reading it, your comment about population is that you don’t think it’s predictable even without climate change. I agree that population distributions are relatively difficult to predict (although I think that with time and skill I don’t have they could be done — people predict future predictions quite frequently to e.g. work out government budgets). I also agree that population is climate change will have unpredictable effects on. But my example of an easily predictable variable was population *size*. Wikipedia says it will be about 9.9 billion in 2060, and it looks like the error could be at most a couple of hundred million (presuming no nuclear war etc.).

            With my new understanding of your argument, let me pose some different questions about the control theory side of things. Firstly, if you are saying that you can’t calculate the effect of a slow system on a fast one, does that also mean you can’t calculate the effect of a fast system on another? If so, aren’t we just back to basically being unable to predict anything, since pretty much anything interesting involves interacting systems? You seem to be saying that they are different, but I just want to check. Secondly, it seems like there are some slow systems that we can predict the effect of. For instance, take population growth in a sufficiently small group (a village, or maybe an extended family). This is a slow system, but it seems like it would be pretty easy to predict its impacts on things like the number of houses in a village, or cars in a family. Thirdly, what exactly are you saying about climate is a slow system?

          • Controls Freak says:

            if you are saying that you can’t calculate the effect of a slow system on a fast one, does that also mean you can’t calculate the effect of a fast system on another?

            If you have two systems with similar timescales, you must simulate them simultaneously. If you can’t predict particular states which might have large effects on other states, you’re probably out of luck.

            If so, aren’t we just back to basically being unable to predict anything, since pretty much anything interesting involves interacting systems?

            Well, not really. We’re actually better off if we have timescale-separated systems. The thing is that we have to work the right way ’round. For example, I am much more bullish on the idea that we can take a range of carbon scenarios for the fast system in order to simulate the slow system. The reason we can do this is because we’ve distilled the important information down to one variable (carbon) and are taking a wide range of scenarios. Of course, one of the sources of error we need to consider is that unknown dynamics could push us out of this range. (The aircraft analog is that we don’t have to actually simulate orientation dynamics. We can distill it down to one variable – altitude, and then consider a range of possibilities based on a reasonable set of assumptions concerning the pilot’s likely choices. It’s insane to go the opposite way.)

            it seems like there are some slow systems that we can predict the effect of. For instance, take population growth in a sufficiently small group (a village, or maybe an extended family). This is a slow system, but it seems like it would be pretty easy to predict its impacts on things like the number of houses in a village, or cars in a family.

            Right. I’m not saying that anything inherent in the fastness/slowness of a system causes it to be predictable/unpredictable. Instead, I’m saying that there is a correct way to model timescale-separated systems… and an incorrect way that is extremely common when people casually think about the effects of climate change.

            what exactly are you saying about climate is a slow system?

            The major mechanisms. Basically, the things that make climate “climate” rather than weather. We have a pretty good idea what the main inputs and physical mechanisms are. The whole premise that we’re operating under is that if we look at a large enough and slow enough scale, we can perform coarse modeling (and smooth over things like day-to-day variability that comes from smaller-scale details). This type of abstraction in modeling is how we’re able to view Jupiter’s Red Spot as being a large, stable entity rather than just a chaotic, unpredictable storm; it’s why we’re able to compute many basic electrical circuits without diving into quantum electron dynamics. All of these things come with caveats and special cases where the abstraction isn’t valid (for details, ask a climate scientist/meteorologist or an electrical engineer (particularly anyone who does processor component design); I’m a dynamics/control guy – I ignore the details and stay in abstraction land if I can get away with it).

          • James Picone says:

            @Control Freak:
            Oh, well, if we can just dismiss stuff by saying we don’t care about it, I don’t really care about your ‘fast system’ crap, which is effectively just saying “Maybe adaptation will be really easy and people will do it automatically!”.

            Sure, maybe it will, but maybe we should not do the thing with uncertain impacts ranging from “Only cost us more effort stormproofing, floodproofing, drought mitigating, moving, etc.” to “Oops, we figured out what caused the end-Permian mass extinction”

          • Controls Freak says:

            James, I’m sorry, but I’m not a climate scientist… so in the spirit of this post, I defer to the consensus on that matter – the IPCC. Funny how I’m likely lumped into the ‘denier’ category, when I’m the guy pushing to accept the IPCC consensus.

            On the other hand, I am a dynamicist, which is why I’m pointing out a common error that is regularly committed by non-dynamicists. You can call it “crap” if you like, but that won’t change the fact that it’s true. If you would like further reading on the subject, check out chapters 10-11 of Khalil’s book. I am on extremely solid ground here, so it would behoove you to not be so flippant in your bluster.

            Sure, maybe it will, but maybe we should not do the thing with uncertain impacts ranging from “Only cost us more effort stormproofing, floodproofing, drought mitigating, moving, etc.” to “Oops, we figured out what caused the end-Permian mass extinction”

            This is a different matter – how to deal with wide ranges of uncertainty. I haven’t addressed this complaint; I very particularly challenged one common error. If you acknowledge what I’ve actually said (and either agree that I am right or dispute that matter), we can move on to your other claims.

      • Iain says:

        The current one. There are a lot of ecosystems out there, human and otherwise, that are adapted to our current conditions. We don’t understand them nearly well enough to justify messing around with the global climate in the hopes that it turns out well.

        I find it baffling that this is somehow the “liberal” side of the argument, because this is a fundamentally conservative position.

        Edit to add: yeah, what rlms said.

        • gbdub says:

          Perhaps I’m taking a glib point too seriously, but I think it’s fairly obvious why the conservative/liberal split is as it is: the “avoid global warming” path apparently requires massive and immediate changes to our economy, probably requiring significant government intervention to force the issue. Whereas “ignore/deny/disprove global warming” lets us carry on as we have and get around to zero-carbon energy when the market makes it work.

          This is one of the frustrating parts of the debate – “Avoiding global warming will be expensive” and “Ignoring global warming would be expensive” are both true, but both are uncertain enough that everyone can just nudge the results in their preferred direction, if they even acknowledge the tradeoff at all.

          • Ketil says:

            “Avoiding global warming will be expensive”

            Not sure I believe this. Nuclear energy can be built as cheaply or cheaper than other kinds, and switching from cows and sheep to more climate friendly stocks shouldn’t really be all that difficult. Cutting petroleum subsidies would actually be economically beneficial. Shifting taxes from e.g. labor to pollution shouldn’t have dramatic adverse effects on the general economy.

            But instead, anybody who cares about the climate also seems hell-bent on totally ineffective or overly expensive remedies (like growing tomatoes on building walls, subsidizing electric luxury cars, or solar panels on top of buildings at high latitudes) – probably because this is about social posturing and indulgences, rather than fixing an actual technical problem. The poster child here is Germany, who has spent enormous amounts of euros on their Energiewende, yet remains one of the worst polluters in Europe.

            map of European electricity production

          • James Picone says:

            @Ketil

            I still mourn Australia’s carbon tax. 🙁

            EDIT:
            @gbdub:
            But Margaret Thatcher was one of the leaders who first brought the issue up in the political world, and it was nearly bipartisan until Bush junior.

          • gbdub says:

            I should have been more clear, I don’t think Iain’s phrasing of “stopping global warming is conservative” is nonsensical, just that I think the current split makes just as much sense.

            And I do love me some nukes. Incidentally, still looking for the equivalent of the Sierra Club without the dogmatic anti-nuke stance. Any ideas?

          • James Picone says:

            @gbdub:
            Afraid not. I’m not aware of any Australian groups with the combination sane views on nuclear and environmental policy, let alone American ones. Probably doesn’t help that nuclear is completely outside the Overton window over here.

          • The Nybbler says:

            Nuclear energy can be built as cheaply or cheaper than other kinds

            Maybe if China builds it. It isn’t going to happen in the West. And if China does build it, given their concern for safety and quality, expect another moderate-to-severe radiation-releasing event within 10 years or so.

            Cutting petroleum subsidies would actually be economically beneficial.

            Wouldn’t change anything. Petroleum isn’t used because it’s subsidized, it’s used because it’s useful. I haven’t seen any evidence that these “subsidies” are tipping the balance towards more petroleum use.

            rather than fixing an actual technical problem.

            Technology is the _enemy_. Technology is seen as the _cause_ of these problems, and to the CAGW advocates, trying to fix it with more technology is just a chain of escalation as silly as the one which ends with “the gorillas simply freeze to death”. The only acceptable (according to the prevailing wisdom) way to stop it is to use less energy, and less technology, so less-worthy primates freeze to death.

          • James Picone says:

            @The Nybbler:
            Pretty sure Bulverism is supposed to be looked down on here. I’m pretty confident that most of the scientists concerned about global warming don’t think we should dump technological civilisation.

          • The Nybbler says:

            I’m pretty confident that most of the scientists concerned about global warming don’t think we should dump technological civilisation.

            They certainly don’t say so in so many words. But you get stuff like this about halving CO2 every decade, which will have the same effect.

            Doesn’t matter what energy source you pick, a “consensus” will be against it. Fossil fuels cause CO2 and plus fracking contaminates water and causes earthquakes. Wind kills birds, the transmission lines and roads destroy habitat. Dams are environmental disasters. Solar damages the delicate desert environment and plus there’s those transmission lines again. Tidal power annoys whales and could cause catastrophic changes to ocean currents. Nuclear causes long-lived waste and plus Fukushima and Chernobyl. Biomass displaces food production (and is ridiculously inadequate in any case).

            It isn’t the same set of objectors for every energy source, of course. And while a few objectors answer the question “well, where do we get energy from?” by pointing to one of the others, most of them will say “Why don’t you just use less energy”.

            Nearly all will react with horror at the suggestion of direct technological mitigation of global warming.

            So, the vector sum is in the direction of “dump technological civilization”.

          • Aapje says:

            @The Nybbler

            Doesn’t matter what energy source you pick, a “consensus” will be against it.

            I am really wondering what you mean by consensus in this sentence. Given your scare quotes, I’m assuming that you don’t actually mean that a large majority of scientists will oppose every energy source. Or do you?

          • The Nybbler says:

            @Aapje

            I mean there will be enough scientists opposed to every energy source to make it appear that the “scientific consensus” is against each one of them, individually. This is partially an effect of those opposed making a lot more noise; no one makes studies designed to show that solar doesn’t kill birds, or wind power doesn’t kill fish, or whatever.

          • Aapje says:

            @Nybbler

            There is already overwhelming support for solar and wind by the American people. Given that scientists are far more progressive than the average American, I’d expect them to support it even more.

            IMHO, you are overestimating people’s contrarianism.

          • The Nybbler says:

            Yeah, there’s support for solar and wind among the American people. In the abstract. Any particular project, they’re opposed to.

          • James Picone says:

            @The Nybbler:
            My experience is that the people against wind and solar power are almost always people who also think global warming is nonsense or farmers who don’t want wind farms on their land. I haven’t seen environmentalist objections to it, let alone scientific ones. I haven’t even seen that for tidal power. I agree that environmentalists tend to be very against nuclear power, and I wish they weren’t, but in my experience scientists are less against nuclear power than the average environmentally concerned person.

            For reference the state I live in has a lot of wind power going. Like 34% of installed capacity is wind. I think I would have run into a lot of environmentalist complaints about them if it were a common thing.

            And saying “The vector sum of everybody’s opinions is less technological civilisation” is very different from saying “The only acceptable (according to the prevailing wisdom) way to stop it is to use less energy, and less technology, so less-worthy primates freeze to death.”.

          • Aapje says:

            @The Nybbler

            Only if it is in their back yard.

            NIMBYism is a completely different claim than “I mean there will be enough scientists opposed to every energy source to make it appear that the “scientific consensus” is against each one of them, individually.”

          • 1soru1 says:

            The only acceptable (according to the prevailing wisdom) way to stop it is to use less energy, and less technology, so less-worthy primates freeze to death.

            Quick reality check:

            – the Republican party controls the Presidency, Senate, and House. The quoted text does not describe the position of any Republican politician.

            – the Democrat party has many senators and representatives. The quoted text does not describe the position of any of them.

            – the Green Party has zero senators and representatives, but a plausible chance of winning one or two someday. The quoted text is not the policy of the Green Party, and no politician that did stand under it would be included in that plausible chance.

            Do you accept that you may be misrepresenting the prevailing wisdom slightly?

          • The Nybbler says:

            – the Republican party controls the Presidency, Senate, and House. The quoted text does not describe the position of any Republican politician.

            Non sequitur. I believe the view of the chief Republican is that there’s no need to stop it, because it’s a Chinese hoax. This is ridiculous of course; if it’s a hoax it’s Australian.

            – the Democrat party has many senators and representatives. The quoted text does not describe the position of any of them.

            No, Democratic politicians are not fools and realize that coming out and demanding austerity measures would lose them elections. So instead their suggestions tend to talk about replacing fossil fuels with “clean” energy, usually with some sort of cap-and-trade scheme for carbon. Thing is, once that carbon cap is in place, and the “clean” energy doesn’t materialize (either because it’s impractical in terms of engineering or gets blocked politically), it amounts to the same thing, and everyone involved should know this by now.

        • deciusbrutus says:

          Why not a colder one?

          I think that you’re taking the lack of a scientific consensus as a scientific consensus that the current climate is ideal.

          I will accept that the current climate is unlikely to result in catastrophic failures, which is a reason to hold a moderate preference for it in the ignorance of what other climates might bring.

          But widespread scientific agnosticism is an opposite of consensus. Preferring the current over the unknown is just being risk-averse, not optimizing, and indicates that more study is needed without demanding what results the more study has.

          And there are two more equally hard questions before we get to justifying cap and trade.

  51. sclmlw says:

    This conception of scientific consensus leaves out too much of the nuance of the actual scientific process. Take Watson & Crick, for example. Framing the issue as “who will get to discover the structure of DNA” imposes a modern understanding that the structure of DNA was obviously going to be important – something that wasn’t obvious until the after the structure was discovered. Despite the popular story about Chargaff’s rules and Avery’s experiments, until April 1953 the smart money was on protein as the genetic material. The best evidence at the time was that genes were very complex, with lots of information needed to be recorded, transmitted, and read. Protein, with 20 amino acids and lots of complex structures, looked very promising. DNA was obviously just a structural molecule and of little interest.

    “What about Avery’s experiments? He clearly showed DNA was the genetic material!” Avery’s experiment probably inspired a small community dedicated to the idea of DNA as the genetic component, but outside this community Avery’s paper didn’t get much play – and with good reason. It was poor science. Go back and read the paper and you’ll notice a surprising lack of controls. To the geneticists of the day protein was a more well-developed hypothesis, with more supporting evidence, than DNA as a candidate for the genetic material. Scientists were simply following the evidence where it lead them. Except the evidence in this case lead in the wrong direction – away from DNA as the genetic material.

    Watson & Crick – and even the eminent Linus Pauling – were part of an outsider movement inspired by a smaller body of DNA evidence and dismissive of the larger body of protein evidence. And often the DNA evidence did not inspire confidence in the theory. In Pauling’s paper on the structure of DNA, he created a horrible Franken-molecule that he openly admitted was unstable and had concerning van der Waal conflicts, but decided a triple helix with phosphate backbones smashed together and hydrophobic bases exposed to water was necessary because, as he said, the bases wouldn’t pair! A helix made sense, he said, because it matched the x-ray crystallography data. But this was the same data Rosalind Franklin understandably dismissed as an artifact since the only way you could get DNA to look like a helix was if you dehydrated it. But DNA is never dehydrated in an actual cell.

    Later, and across the pond, when a chemist happened upon Crick trying to pair the bases, he noticed Crick (and by extension the entire DNA community) was using the less stable enol form of the molecule, instead of the more stable keto form. Crick didn’t know anything about enol-keto tautomerization (doesn’t everyone?), but once he made the recommended changes he was easily able to do what Pauling couldn’t: pair the bases. He showed Watson and they made a model that afternoon. A major discovery born from a minor error correction. The structure suggested sequence and a possibility for replication, which provided a mechanism for DNA to act as the genetic code. That lead to a major shift in favor of DNA over protein as the genetic material.

    This episode is not unique, and shows a more nuanced understanding of scientific consensus – what is and how it functions in science. Scientists follow the evidence where it leads, sure. But sometimes the evidence suggests hypotheses that direct future research in the wrong direction. That’s not the norm, but it does happen from time to time. The important corrective is that a small number of “crackpot” researchers sometimes manage to pursue alternative hypotheses, which generates new evidence available to the broader scientific community. Therefore, it’s good to let a million flowers bloom. We don’t have to agree with them until they show us the data, but they can toil away on their crackpot theory and not offend scientific progress.

    While it may be true that the consensus doesn’t tend to ignore the best evidence in favor of a popular hypothesis. It is true that a popular hypothesis has a strong tendency to drive research focus and funding. In my field of immunology, I’ve seen many successive “fad” theories come and go. We follow the evidence, but that process feels more like Brownian motion sometimes than it does solid progress. I see evidence that the general trend is toward increased knowledge, but I wouldn’t put money on any single theory – however universally accepted today – not being overturned in the next 25 years. I’m skeptical of scientists from other fields who tell me this process doesn’t apply to their favorite theory.

  52. Ransom says:

    Hi Scott,
    Scientific consensus is what we learn in school, and most of us believe it without evidence. But the scientific community is very political (in the sense of herd/groupthink behavior, not necessarily aligned to the usual Left-Right axes). On un-politicized topics, in fields with mathematical theories that can make hard predictions so real testing is possible, the politics gets pushed out pretty quickly by the quality of the hard evidence. [This is what your examples seem to mainly show.] So textbook scientific consensus is good for physics, chemistry, molecular biology, etc.
    On the other hand, in fields where strong politics prevail and quantitative prediction/testing is hard or impossible (like most social sciences, and climate science at the moment, to pick prominent examples), scientific consensus can easily be mainly politics. Scientific consensus is MUCH less reliable there.
    A question: what has come out of the replication crisis stuff? I am very happy to see the issue raised so prominently, but there are deep institutional reasons why the standards of evidence were so low to begin with. Basically, for most say, psychological studies, it would cost millions and take decades to really make a good “measurement”–much more costly and time consuming than in, say, physics. So how does the average academic make a living except by publishing what he has? So I’m skeptical that there will be a real change.

  53. martinw says:

    If a field has become so politicized that the majority of the researchers working in that field will state for the record that they believe X, but when surveyed anonymously they admit to actually believe !X, that’s not an argument in favor of trusting scientific consensus. It’s an example of a horrible, spectacular, disastrous failure of the scientific method. Irrespective of whether or not X turns out to be true in the end.

    If the truth exists in a scientist’s head, but the scientist is so afraid of repercussions that she will only speak that truth off-the-record under conditions of strict anonymity, does it really exist?

    When the actual scientists in a given field are happily telling their honest beliefs to anybody who will listen, and it’s textbook writers or journalists who distort their views, then it’s fair to say that Science itself should not be blamed for what the textbook writers and journalists get wrong. But when even asking the scientists directly is not guaranteed to give you their true beliefs, can a consensus really be said to exist?

  54. Ruminist says:

    I think this is mostly right, but here’s another question to ponder: What if the dangerous heresy isn’t challenging scientific consensus? What if the danger comes when you try to turn scientific consensus into policy? For example:

    Lawrence Summers didn’t get into trouble because he talked about well-known research on sex differences in math skill. He got into trouble because he talked about it in a way that implied that Harvard should discontinue policies designed to increase the proportion of women in its STEM departments.

    Charles Murray didn’t get into trouble because he wrote about well-known research on racial differences in IQ. He got into trouble because he used that research to argue that affirmation action should be discontinued (in favor of income assistance).

    Peter Singer didn’t get into trouble because he explored a well-known mode of utilitarian thought. He got into trouble because he applied that mode to a real-world legal case.

    Meanwhile, millions of people ignore global warming, focus their weight loss efforts on counting calories, spend money on vitamin pills, etc. So yes, the scientific consensus is getting better and better. My worry is that it may have zero influence on anything else.

    • Urstoff says:

      Counterexample (maybe): E.O. Wilson’s Sociobiology was controversial for its content, not for any provocative policy proposals made by Wilson.

      • Ruminist says:

        That’s a good example, but I think it fits the pattern. The “Against ‘Sociobiology'” letter Steven Jay Gould et al. sent to the New York Review of Books say they object to the book because it provides “a genetic justification of the status quo and of existing privileges for certain groups according to class, race or sex.” Most of the arguments against it, and the attacks against Wilson, seem to flow along those lines.

        I get the sense that there’s a societal immune system that attacks anyone who tries to bring scientific findings to any audience that can shape policy.

  55. I appreciate this post. It’s important to respect the scientific consensus, as it actually operates, before you reject it.
    I studied econ throughout undergrad, then my masters, and still couldn’t nail down macro-economics. I didn’t take a strong ideological stance, like many of my peers.

    I figured I might as well see how it works in the belly of the beast, so I worked at the Fed for a few years. Turns out the economists within the Fed aren’t completely unaware of criticisms of their field. In fact, many of them are leading these criticisms, and we invited economist guest speakers who built their careers on disagreeing with the Fed.

    The Fed as an institution remains interesting, and it’s not crazy to imagine counterfactual US governments where due to historical accident the Fed as an institution ended up taking a much different approach towards economic philosophy of science.

    Still, these views that they are all unaware, whether it’s econ or any other field, tend to be weak.

    On the other hand, often a small minority of academics can build careers by offering up confirmation bias to all types of social activists. You might have gone a little too easy on the scientific consensus in this post by leaving out the phenomena of professors of literature, art, and sociology, building incredibly strong consensus on ideas of government or morality.

  56. eelcohoogendoorn says:

    Global warming skeptics talk about how maybe the scientific consensus on global warming is false because climatologists face political pressure to bias their results in favor of the theory. But scientists studying these areas face much more political pressure, and as long as you give the surveys anonymously they’re happy to express horrendously taboo opinions. This is about the strongest evidence in favor of the consensus on global warming – and scientific consensus in general – that I could imagine.

    This is not a bad analogy, but it has a serious limitation.

    Believing in the blank slate requires the willingness to flatly ignore an endless torrent of repeatable experiments. This does tend to take a lot of effort, even for the ideologically committed.

    Believing that AGW is a serious and probable threat or not is not constrained by intellectual honesty in the same matter, since our most sophisticated theories on the subject still have not surpassed the predictive power of the pirates-vs-global-temperature anticorrelation; let alone a repeatable experiment having been performed. People are about as free to choose their beliefs in this domain as in the matter of jesus versus mohammed; is there much more to it than tribal signalling?

    • eelcohoogendoorn says:

      To clarify: I dont think its a silly notion at all that CO2 emissions might influence climate, and that this might inform policy one way or the other. There are arguments to be made for that, but they are all extrapolations from (incomplete) theories; which havnt been experimentally confirmed. Sometimes (oftentimes) thats all you have to go by, but I am a little hung up on people pretending that ‘the science is settled’. That strikes me as woefully unaware of the huge attrition rate between ‘theoretically plausible and popular theory’ to ‘confirmed by repeatable experiment’. Whatever its merits or flaws, AGW theory is squarely in the former category, not in the latter.

      To grumpy skeptics such as myself, that missing piece of intellectual honesty has the same off-putting effect on me as Scott’s high school psychology textbook had on him. The falsehoods propagated in those textbooks went against repeatable experiments; im glad to see that those voices are losing ground, but they sure commanded a lot of mindspace for an awefully long time. With the current state of AGW research, I can easily see the the process of truth percolating to the surface taking an order of magnitude more time.

      I do trust ‘science’ to converge on truth eventually; but my rule of thumb is that if a ‘consensus’ hasn’t held up for a century or so, don’t be too surprised when it does come crashing down.

    • James Picone says:

      @Eelcohoogendoorn:
      Are you aware that the more CO2 -> warming prediction dates to 1896, when Arrhenius proposed a climate sensitivity on the high end of the current sensitivity value? That’s predictive.

      I assume you’re also unaware that stratospheric cooling is a prediction of CO2-driven warming that’s borne out by the current data (and distinguishes CO2-driven warming from solar warming), that Manabe predicted the increase in Antarctic sea ice early on under CO2-driven warming, and that the basic radiative physics that predicts a 1c increase for a doubling of CO2 (first-order, no feedback effect) is goddamn rock solid.

      If you think the evidence is purely correlational, you don’t know the first thing about the topic you’re criticising.

      @Glen Raphael:
      See what I mean? Zombies are real, and they’re here.

      • eelcohoogendoorn says:

        @James: indeed; the predictions of AGW theory are correct to first order, in a qualitative sense. That is, temperature has gone up as predicted; though our skill at predicting the rate has been pretty bad. Hence qualitative, not quantitative. Any higher-order features, such as the recent slowdown in warming, have caught us by surprise, and has been outside of the confidence intervals of the historical ‘consensus’. Hence the comparison with pirates-vs-global-temperatures is completely apt; that theory has comparable predictive power. Also qualitatively correct to first order. The point being: that is a terribly low bar to beat for any theory of any kind.

        Your assumption about my awareness of the other detailed implications of CO2 driven warning are incorrect. I am aware of the debate surrounding the many potential secondary implications of AGW; and not too impressed with the rate at which such tests come out positive. A meta-analysis of such claims would do a lot more to convince me of your position than to pick some examples in a snarky tone.

        Either way, we may differ in our degrees as to how persuasive we find the empirical evidence, but I dont want to dive into that to deeply right here, because its tangential to the point I am trying to raise. That is: none of the tests discussed here are repeatable experiments; and even in fields where this least debatable of all scientific instrument is widely available, finding consensus can be a long, winding and political process.

        • rlms says:

          We can’t repeatably experiment with evolution of multicellular organisms. Are you skeptical about that theory?

          • eelcohoogendoorn says:

            We can’t? Thats news to me.

            Either way, I am sure you can find other examples of theories that by their nature are (almost) impossible to test by repeatable experiment. And I may or may not put a lot of stock into those theories. I am not saying that repeatable experiments are the only epistemologically sound way of attaining knowledge. But it certainly is a convincing one to the average homo sapiens. One that the AGW debate will have to do without.

        • James Picone says:

          That is, temperature has gone up as predicted; though our skill at predicting the rate has been pretty bad. Hence qualitative, not quantitative.

          They Look pretty good to me. Aarhenius’ ECS is only slightly above the high end of the IPCC range, which is pretty good for worked-out-by-hand in the late 1800s.

          Any higher-order features, such as the recent slowdown in warming, have caught us by surprise, and has been outside of the confidence intervals of the historical ‘consensus’.

          What slowdown?

          Hence the comparison with pirates-vs-global-temperatures is completely apt; that theory has comparable predictive power. Also qualitatively correct to first order. The point being: that is a terribly low bar to beat for any theory of any kind.

          Also doesn’t have radiative physics behind it, which kinda matters. We have very good reasons to think CO2 has a certain first-order effect; we had those reasons before we observed the effect; we observed an effect with a magnitude well inside our error bars; paleoclimate provides comparable values for the effect when it undergoes transitions, etc. etc..

          The focus on ‘repeatable experiments’ is certainly an interesting example of convergent evolution.

          • eelcohoogendoorn says:

            The focus on ‘repeatable experiments’ is certainly an interesting example of convergent evolution.

            Not entirely sure what that means, but ill take it as having reached consensus on the point I was trying to make.

  57. Jameson Quinn says:

    I have experience with a mild paradigm shift that is going on inside social choice theory, the study of voting methods. This is complicated because this is a fringe enough area that there are many subgroups of “people who think about voting methods” and often little or no communication between them; and also because there are perfectly legitimate reasons to come at this area from a perspective of activism rather than pure inquiry. Quick US-centric and self-promoting timeline:

    1200s: Ramon Llull looks at existing elections, invents the methods we know as Condorcet and Approval. Within 100 years, his work is basically forgotten.

    1700s: Condorcet and Borda both realize voting is more complicated than people thought, invent “new” methods. Condorcet realizes cyclical preferences are a real problem, Borda gets huffy. Condorcet invents system that bears his name, but also the system we now know as Bucklin; the latter is what Geneva actually implements following (posthumously) his ideas, but Napoleon invades less than a year later so that doesn’t really matter.

    1800s: People start thinking about proportional representation and invent most of the basic ideas we have for that today. For instance, Dodgson (aka Lewis Carroll) invents what we now call “delegable proxy” or “asset” or “liquid democracy”

    1950: Arrow’s theorem simultaneously founds modern social choice theory on a foundation of ranked systems, and shows that no perfect system can be built on this foundation. Soon after, Gibbard and Satterthwaite show that Arrow’s theorem (even though it doesn’t directly apply to non-ranked systems) means that all systems, ranked/ordinal or not, admit strategy in some cases. Paradigmatic social choice theory is established as being about mathematical criteria and proofs.

    ~1980: Brams et al “invent” (that is, rediscover and name) approval voting; the first time a non-ordinal system is seriously considered from a theoretical standpoint.

    1990s: A series of non-majority outcomes in US presidential elections prompts increased US interest in single-winner reform. Activist organization FairVote is spawned from Citizens for Proportional Representation to promote IRV, a single-winner system which meets one criterion (later no harm) only by breaking many others (monotonicity, favorite betrayal, etc.) Given the understanding of voting theory at the time, this is somewhat excusable.

    2000: Prickly crackpot genius Warren Smith writes his paper “Range Voting”, in which he performs monte-carlo simulations (that is: a probabilistic, rather than a criteria-based approach) using simplified utility and strategy models to show that his non-ordinal method, which we now call score voting, seems clearly superior to almost all alternative system. Also, the US presidential election system fails spectacularly.

    2001-2015: FairVote continues to push IRV, and makes some progress, but also has setbacks, including a pathological (but still better than plurality!) election in Burlington 2009. Amateur voting enthusiasts on the internet make some headway convincing each other to consider non-ordinal systems like Range and Approval voting, but remain significantly weaker than FairVote. Most academics continue to work in the Arrovian paradigm, but there are more and more exceptions (such as Balinski and Laraki, with a median-based proposed method called Majority Judgment).

    2016: FairVote wins their first statewide victory in Maine. I do a new version of Smith’s monte carlo work, with significantly more-sophisticated and more-realistic models, under which score and approval voting are still good but no longer the runaway best methods. Methods like 3-2-1 voting and Score Runoff Voting seem best.

    ….

    My overall point is that over the past 15-20 years, I’ve seen significant movement away from the (decent but flawed) Arrovian paradigm in academia and from the (only barely better-than-plurality) IRV method in activism, but in both domains the ideas and outlooks I’d consider “correct” are still held by a minority, despite having been stated in a clear (if not entirely modern) form at least 15 years ago.

    • Urstoff says:

      How much worse is ranked choice voting than score voting?

      • Jameson Quinn says:

        “Ranked choice voting” is a rebranding of instant runoff voting (IRV), even though there are many other methods involving ranked choices. So I still call it IRV.

        When you ask “how much worse”, I have to choose a metric. I think that Voter Satisfaction Efficiency with a good (CrossCat-inspired) model is probably the best single metric; with things like strategic incentive worth looking at to add some context. Depending on strategies, Plurality’s VSE is 70-85%; IRV’s is 80-90%; score’s is 85-95%; and the best methods, like 3-2-1 voting or Score Runoff Voting, get 90-97%.

        As for strategic incentive/susceptibility, score is actually a bit worse than IRV in terms of incentives, but better in terms of severity of the favored strategy. 3-2-1 or SRV are better than either in this way too.

        Between 3-2-1 and SRV, it’s mostly a choice of ballot simplicity and slightly better strategy robustness (3-2-1) versus expressiveness and improved best-case VSE (SRV).

  58. Alex Zavoluk says:

    Maybe-related question: How do you know when someone is “qualified” (for lack of a better term) to disagree with the consensus? That is, for a layperson looking at a field they don’t have particular experience in, their best heuristic is to adopt the expert consensus. But as Scott pointed out, scientific consensus is meaningless if scientists agree with the consensus just because that’s the most likely outcome. So at what point are you responsible for/capable of adopting your own point of view, separate from the consensus?

    Side point: the fact that science is sometimes wrong, doesn’t make the crackpots more likely to be correct in general.

  59. afirebearer says:

    The ‘actual surveys’ link is not working.

  60. rahien.din says:

    All these points are subtle. And this is ground you’ve turned over so many times. Epistemic Learned Helplessness was always my favorite. There are even shades of it in the metis-vs-modernism comparison in your review of Seeing Like a State. But I can’t shake the feeling that you are conflating the scientific method with knowledge it produces.

    Science isn’t a body of knowledge. Science is a method, whereby we challenge paradigms with rigor. And the substrate of the scientific method is informed-conjecture-that-challenges-a-paradigm. In fact, the only reason a paradigm is accepted or useful is that it has survived challenge.

    I find it weird even to say “scientific consensus.” Consensus is the target of science just as much as it is the result of science.

    If we separate the scientific method from our current body of knowledge, the various permutations of…

    When is X group right/wrong about Y? | X=(contrarians, crackpots, non-consensus scientists, consensus scientists)

    ….all reduce down to “What ideas should we test via the scientific method?” That’s just a prioritization/allocation problem, rather than some knotty philosophical quandary.

    • rlms says:

      I don’t think the scientific method really comes into it. The post is about how science is practiced in the current world. In a totalitarian state where the government always kills you if you make scientific conjectures that challenge established belief, the scientific consensus will often be inaccurate and it makes sense to challenge it (see the USSR and Lysenkoism for a real example). At the other extreme, if the scientific culture is of pure apolitical devotion to truth and challenging previous theories is rewarded then challenging the scientific consensus will almost always be foolish. I read this post as Scott saying he thinks current scientific culture is closer to the latter than he previously thought.

      • rahien.din says:

        The practice of science does not necessarily involve the scientific method.

        Wait. What?

        If challenging scientific consensus will get you killed, it makes sense to challenge it. If challenging scientific consensus will be rewarded, it does not make sense to challenge it.

        Wait. What?

        I enjoy talking with you, but your reply has me genuinely baffled. Admittedly, neither of those paraphrasings is a worthy steelman… but I don’t even know how to steelman your claims. What are you even claiming?

        If you would explain, I would be happy to discuss further.

        ETA: it’s worth pointing out that our host thinks it’s really, really hard to politicize science. I strongly disagree that politicization is in any way the question at hand.

        • random832 says:

          In the second, I’m pretty sure “makes sense” is being used in a pure sense of whether the ideas one is challenging with are likely to be better than the current consensus, rather than whether it is wise for one’s personal safety.

        • rlms says:

          Yes, neither of those are what I meant. My point is that your comment seems to be focused on the scientific method (which is the same regardless of how it is used), whereas the post (as I interpret it) is about the current practice of science in the Western world. The scientific consensus on an issue is what the majority of scientists believe about it. Scott is saying that if you want to form correct beliefs, trusting the scientific consensus is generally the way to go. But (assuming he’s correct) this is only true in the specific present world; it isn’t a general statement about any conclusions the scientific method might produce.

          Regarding your first quote, I was saying that the post doesn’t involve the scientific method, not that the practice of science does. random832 is right about my meaning in the second quote. You should say you agree with Lysenko if you want to stay out of the gulag, but if you actually agree with him you won’t understand how plants’ characteristics are inherited. Likewise, in the truthphillic utopia no-one will harm you if you reject consensus, but you won’t get to true beliefs by doing so. Most of the time, our world is more like the utopia: long-standing consensuses (germ theory, Newtonian physics, heliocentrism) are invariably pretty much correct. Scott is further claiming that the consensus is actually generally accurate even when it’s less broad or more recent.

          Politicisation is one factor that affects how likely the consensus is to be true, but it is by no means the only one. In a modern context, scientific culture and incentive structures are probably more important, and things like the sheer number of modern scientists make modern consensuses more reliable than those from previous centuries.

          • rahien.din says:

            rlms,

            When you write this…

            My point is that your comment seems to be focused on the scientific method, whereas the post is about the current practice of science in the Western world.

            …and this…

            in the truthphillic utopia no-one will harm you if you reject consensus, but you won’t get to true beliefs by doing so.

            …you’re committing the very error I had been trying to point out.

            The practice of science is formulating a conjecture anti an established state of knowledge (whether consensus, presumption, or ignorance) and rigorously testing that conjecture.

            We get to more true beliefs by laying challenge to paradigms. If the paradigm survives, it is more likely to be true. If not, we modify it to be more true, or we reject it in favor of a more true paradigm.

          • rlms says:

            Let me clarify what I meant by “in the truthphillic utopia no-one will harm you if you reject consensus, but you won’t get to true beliefs by doing so”. The truthphillic utopia here is an unrealistic extreme where reputable scientists are constantly making new conjectures that challenge the status quo, and correct ones are instantly adopted as scientific consensus. You are correct in saying that if, as a scientist, you want to get to beliefs that are more true than the consensus (or you want to increase your confidence that the consensus is right) then you must make conjectures that challenge the consensus. But thanks to the constant churn of scientific ideas in this ideal world, the chance that a specific contrarian conjecture you make is correct is approximately zero (and this is even more true for a layperson).

            When I said you won’t get to true beliefs by rejecting the consensus, I meant that when you challenge the consensus you are more likely to increase a rational person’s confidence in it than to disprove it. Therefore, given the way science is currently practised, the best general way to form true beliefs on a scientific issue is to adopt the scientific consensus rather than come up with your own different theory. This method fails, but infrequently. In truthphilic utopia it approximately never fails. In the USSR, it fails more frequently. In a hypothetical dystopia where the government bans test tubes, numbers greater than five, looking at animals, and carrying out surveys, it fails most of the time.

          • random832 says:

            We get to more true beliefs by laying challenge to paradigms. If the paradigm survives, it is more likely to be true.

            I think the key difference is that in a repressive society, the “challenges” that are going unmade are easy-to-notice things like “maybe the crops failed because Lysenko’s science is wrong rather than because of kulaks”, so your challenge is more likely to be one of those rather than an obscure one which is as likely to be crackpot as not, compared to a more open society where there is no such low hanging fruit.

        • grrath says:

          I think this comes down to the difference between the practice of science and the institution of science which does not necessarily have to be related to the former. The USSR and current issues with pharmaceutical companies and the like influences researchers should show that it is possible to be a “scientist” and not perform proper “science.”

  61. Bugmaster says:

    Can you offer some examples of situations where you disagreed with the consensus, and were proven to be disastrously wrong several years later ? Otherwise, you are making it sound like “scientific consensus” is synonymous with “whatever Scott believes to be true”, or something. Or rather, you are making it sound like your reason for increasing your support for the consensus is that it confirms your beliefs all the time. I realize that you admonish against exactly this kind of attitude in your closing paragraphs, but without examples, the message rings a bit hollow, IMO.

  62. greghb says:

    A simple hypothesis for your changing your mind is that you’re just finding the people and ideas that make sense to you. Just as “rationalists” / “rationality” aren’t homogeneous, neither is “the scientific community / “science.” Part of developing as a thinker is learning whom and what to trust within the surprisingly broad range of people and ideas that fall under any given label.

    The “we’ve noticed the skulls” post sounded to me a bit like, “If you figure out who the smart and responsible rationalists are, and just transact with them and their ideas, you don’t see the kind of nonsense that some people ascribe to rationalists.” And this post sounds a bit like, “If you figure out who the smart and responsible scientists are, and just transact with them and their ideas, then you don’t see the kind of nonsense that some people ascribe to the scientific community.” This is all very nice, so congrats.

  63. pontifex says:

    I think the problem is not science but “scientism.” It’s very common to see people take some very tentative result from the social sciences, like a psychology study with n = 10 that was never replicated, and use it as ammunition for their side in the culture wars. The media trumpets this from the rooftops, and political leaders demand that we Do Something About the Problem. And then we end up razing cities and building giant equally spaced rectangular grids, because, you know, Science!

    Religiously-based societies don’t suffer from this problem as much. Religion tends to emphasize humility and the imperfection of mankind. It also has a well-entrenched class of clerics who are specifically there to give the orthodox interpretation to the laity. The end result is that religiously based societies tend to create fewer mountains of skulls than people following “scientism.” I think the history of the 20th century bears this out. Science is a dangerous tool in the hands of people without critical thinking skills and humility.

    • rlms says:

      “I think the history of the 20th century bears this out.”
      The histories of preceding centuries don’t, and Wikipedia’s list of countries by religiousness seems to me to show it anti-correlated with being a nice place to live.

      • pontifex says:

        I wasn’t trying to claim that religious societies were nicer places to live, just that they avoid some of the problems we have today around the misuse of science.

        People keep re-creating religion, over and over. Someone whose grandfather was a devout eastern rite Christian might become an eager convert to communism. Someone whose grandfather was a Presbyterian might become a social justice type. In both cases, I feel like the new belief systems are worse than the old ones. Belief systems that are “sciency” (but not actually scientific) like Marxism, fascism, etc. are very dangerous. And maybe we can even include whatever the hell we’re calling what Ray Kurweil believes in that list too…

        • rlms says:

          I’m not sure what distinction you are making. You give communism as an example of a “religion”, but Marxism as an example of a “sciencism”.

    • grrath says:

      You almost had it with your first paragraph and then you went off the rails by assuming religion wasn’t an even worse system. Unquestioned authority is always bad, at least science has a mechanism for questioning.

  64. John Schilling says:

    But scientists studying these areas face much more political pressure, and as long as you give the surveys anonymously they’re happy to express horrendously taboo opinions.

    But the anonymous surveys only gauge the present beliefs of scientists; they are not part of the process by which the scientific community can improve on its present beliefs. Were not part of the process by which the scientific community came to its present beliefs, consensus or otherwise, which has to cast some doubt on that “consensus”. One cannot hold an informed opinion without listening to what other people have to say, and so the scientist willing to express a taboo opinion under cover of anonymity – or a popular one in the light of day – is ill-informed due to the lack of full and candid discussion.

    This is about the strongest evidence in favor of the consensus on global warming – and scientific consensus in general – that I could imagine.

    You can’t imagine that the evidence would be stronger if the people willing to express dissenting views didn’t feel it necessary to do so anonymously?

    I think you may be inappropriately generalizing here, from relatively narrow scientific problems amenable to the Scott Alexander Deep Dive approach, to problems so broad that no person’s individual research or understanding can encompass more than a fraction of the relevant issues. If each scientist hears only the consensus-compliant opinions the others are willing to voice in public, then their private understanding of one small corner of the issue won’t change their belief that the consensus is in general correct – and in an environment where this then means their private understanding stays private, a great consensus may be found on shoddy but whitewashed foundation.

  65. spinystellate says:

    Scott notes that important things being on the line and yet scientists still usually getting things right is “is about the strongest evidence in favor of the consensus on global warming – and scientific consensus in general – that I could imagine.”

    I’d like to distinguish between things “being on the line” for the individual contrarian vs for the people making up the consensus they are opposing. Usually we think only about the former, e.g. being mocked, shunned, losing funding, etc. But sometimes the people of the consensus have something to lose to. This is especially true when, for external sociological reasons, the consensus must both be and appear *extremely* strong for the field to remain fundable.

    So even when someone might individually survive the blowback from being a contrarian (given fixed support for their field as a whole), they might choose not to be a contrarian if it would undermine support for the field via undermining the consensus that is usually a proxy for it. Concrete but hypothetical examples: one might really be into early childhood education, but to pursue a line of research openly predicated on the belief that many similar efforts had failed might threaten the entire broader enterprise, and thus might discourage peers from promoting or funding it and the researcher from pursuing it; one might have a new hypothesis for Alzheimer’s disease pathogenesis, predicted on recognizing that the rate of return on previous Alzheimer’s research has been incredibly small, etc.; one might have a new climate model, but which shows below average warming trends, etc.

    I think this might only be a concern for scientific fields that are very visible to the public, but that’s still quite a lot of important fields.

  66. jessriedel says:

    I find this to be a really unusual post. Scott’s take-home message seems ludicrous to me despite not disagreeing with any of his factual points. The scientific procress works “reasonably well” when it takes 10 to 30 years for big mistakes to be recognized? (New PhDs need to get a tensure track position within about five years.) Academia is OK because we must resort to anonymous surveys to measure scientists’s actual beliefs? (Just think of what that means for the distribution of research funding.) Medium-term human annihilation from AI risk is considered a likely outcome by a large minority of the scientists, but the number working on this full time is a few dozen?

    I guess Scott will just say “Sure, but the scientific consensus is the best mechanism we have”. But compared to what? Scholasticism? Color me unimpressed.

    I appreciate that Scott is admitting that, in the past, he was often overly confident in diagnosing all the ills with science and identifying The Truth. But rather than reflecting well on the academia, I would summarize it like this: “It turns out that a bunch of experts already knew a lot of the stuff I had to work for years to uncover, but they are either too intimidated or too lazy to make this information accessible, to the large expense of the public, generally, and the allocations of research resources, specifically.

    This post claims that if you survey and probe and wait and ask the right questions and ignore misleading indicators, the scientific consensus seems drift toward the right answer. But this is very slow, and has a greatly reduced real-world effect, because, with a few exceptions, the scientific consensus on any given topic isn’t common knowledge since it isn’t written down in a specific location. Scott blames on “the ability of people who don’t listen to the experts to get disproportionate social power and hide the existence of the expert consensus”. How, in the age of Google, could anyone hide the expert consensus if it was at all accessible? If we only can determine the consensus by enrolling in a medical residency program and investigating a topic for years, with lots and lots of room for subjectively selecting the indicators of consensus, in what sense does the consensus exist objectively?

    Let me offer a solution to work toward: We need much, much better mechanisms for identifying, distilling, and releasing the scientific consensus, and for documenting the disagreement when no consensus exists.

    • Ilya Shpitser says:

      “But compared to what?”

      Presumably compared to people ranting online about how busted academia is, and waxing wise about Bayes.

    • Christopher Hazell says:

      I had the EXACT same reaction, of being accidentally convinced that my previous good faith in “scientific consensus” was misplaced.

      The question of how quickly the scientific consensus changes is actually, honestly, kind of tangential to whether we should trust it. For example:

      Bakker did help produce the paradigm shift in paleontology from cold-blooded dinosaurs to warm-blooded dinosaurs.

      If we’re trying to decide whether it’s usually a good idea to trust the scientific consensus the relevant question isn’t how quickly it changes, it’s how inaccurate it was before it changed.

      In other words: Quick, it’s 1940 and I need to have an opinion on whether or not Dinosaurs were cold-blooded! Lives hang in the balance, and if I’m wrong about what kind of blood dinosaurs have, people will die.

      Luckily I can turn to the scientific consensus! So, before Bakker, I can see two different possibilities for what the scientific consensus was on dinosaurs:

      A) We don’t really know enough to have a guess yet about whether dinosaurs were cold or warm blooded.

      Or

      B) Dinosaurs were almost certainly cold-blooded.

      If the pre-Bakker consensus was B), then a lot of people will die when I confidently assert an incorrect fact.

      More frightentingly, Mr. Alexander explains that he often has difficulty assessing what the consensus is on issues that directly relate to his studies.

      So, if a psychology student can’t confidently assess what the consensus is on a given issue, how the hell is somebody like me supposed to assess the consensus on, say, Climate Change, when I do not study the issue and in fact have no relevant expertise whatsoever?

  67. AnonYEmous says:

    Scott, thoughts on the following scenario:

    “20 years later” Rationalist blogger: See guys, even though global warming was believed by consensus, it was eventually disproved. So science really does work, and you should trust the consensus!

    ———–

    Just saying that, as far as I can tell, this event would fit your model perfectly.

    Also wish that article had included some of the better consensus failures (tectonic plates, Pangaea) but that’s obviously on them.

    • random832 says:

      The “top ten” in Scott’s writeup were the first ten in the order the article presented (which was alphabetical), not the ten “best” examples by anyone’s measurement – Wegener naturally didn’t make that list. (I made the same mistake earlier)

      • AnonYEmous says:

        yeah i was talking badly about the article writers

        but looks like i should’ve actually read the article

        shrugs

  68. Ransom says:

    Most of this thread (and Scott’s post too) seem to conflate the hard sciences with the social sciences. Consensus in the hard sciences is (usually) reliable because the hard sciences are “easy” compared to the softer ones. The subjects of study are simple and can be literally torn to pieces in the lab, if that’s helpful. The questions being addressed are about regularities or laws, about repeatable phenomena.
    The social sciences only rarely even approximate this. Witness the comments on IQ above. I’m a physical scientist, so this is the one place that I know of where the data are really good enough to approximate something in physics or chemistry. But it took decades of data collection on many millions of people in institutionally-driven settings to acquire.

    Because the data are (usually) poor and the systems are complex, the scientific consensus in the softer fields is MUCH more a product of politics, both internal and external. I don’t see how anything useful about “trusting scientific consensus” can be said without making this distinction clearly.

    The situation is again very different in the historical sciences like astronomy, geology, evolution, where the models are “cases for X” rather than testable predictions.

    “Science” is not a single thing. The trustworthiness of consensus in the physical sciences is very different than in the social sciences, and different still in the historical sciences.

    • maybe_slytherin says:

      This piece in the LA Times is a pretty good popular article about the replication crisis. Relevant to much of this discussion, but particularly in pointing out the ways the harder sciences had these same issues:

      In the 18th and 19th centuries, scientists struggled even to determine the exact temperature at which water boils. It took many failures to learn that factors such as the material of the vessel or the presence of dust were crucial. Understanding that altitude was a critical variable (the higher the altitude the lower the boiling point) revealed the all-important relationship between temperature and pressure — one of the underpinnings of thermodynamics. But initially it just led to more than 100 years of puzzling replication failures.

      There’s also a longer story of a simple, clever experiment in heart physiology — that also failed to replicate for years.

      As for the content of your post, I think that it’s mostly valid but overstated. Physical sciences’ objects of study are often rare or the product of unusual conditions; perhaps it’s less political, but there is quite a lot of indirectness & opportunity for buggy equipment/sensors/code. Even granting that the term “historical sciences” might be useful even if not mainstream (created by creationists specifically to paint a target), “cases for X” do come with testable predictions. Say “these used to be the same continent, so in a certain period there should be similar fossils on them”. Finally, the acceptance and interpretation of results is always a political process — see Kuhn, and, in a new example for me personally, the de-chemistry-ification of the atomic bomb.

      • Ransom says:

        I heartily agree that the physical sciences have their own “replication” issues. My point is that they’re inherently easier (simpler systems, ability to get the whole system into a lab or a test tube, no ethical issues really, etc), less costly to get good data (however difficult still). And I don’t mean to denigrate the social sciences either; they just face a much tougher set of problems.

        But all of that is the reason why consensus in the physical sciences is much less susceptible to politics than in the social sciences. I think consensus in the physical sciences is a (fairly) reliable proxy for “it really does work”, especially for older fields or broad, general laws and models. I think consensus in the social/medical sciences is generally unreliable, especially for areas that suggest practical courses of action, like dieting or education policy, etc etc.

        So “scientific consensus” carries a lot of weight in physics, chemistry, large areas of biology, etc, but little weight in the softer areas.

  69. MawBTS says:

    A poll in Nature shows that 90% of scientists believe reproducibility issues constitute a “crisis”, compared to only 3% (!) who don’t.

    1. Talk’s cheap. Are they doing anything about it?

    2. I wonder if they’d say the same if the question was narrowed down to their field or their own area of research. US citizens typically give their congressional representative an approval rating of 50-60%…and give congress as a whole 24%. It’s easier to talk about dirty kitchens when it’s not your dirty kitchen.

    3. I’m wary of opinion polls where one side is obviously erring on the side of caution and respectability. Who’s going to be cavalier enough to claim that reproducibility isn’t any problem whatsoever?

    The relevant surveys I know trying to elicit scientific consensus (1, 2, 3) all find that, when asked anonymously, most scientists think these differences explain about 25% – 50% of variance.

    I hope I don’t get banned, but what happens when you phrase the question more bluntly? What happens when you use the word “race”? Here’s a November 2016 paper summarizing the views of anthropologists on race, ancestry, and genetics. N=3286

    73% strongly agree that race is “biologically meaningless”, and 86% strongly disagree that humans can be “subdivided into biological races”. Similar results for “race affects health” and a range of other things – no overwhelming consensus against, but still a fairly strong sense of denial.

    There are no questions about race and income or race and intelligence, but I don’t think the answers to those would have been greatly different.

    But 89% agree that some biological variability exists, and there you see the current state of the discussion. A token admission that differences exist (usually on something anodyne and apolitical like medicine), but on any topic with political implications the field still seems Gouldian in its outlook.

    To be fair, this was just a poll of anthropologists. I suspect geneticists or biologists would give different answers. Maybe. In 2014 Nicholas Wade put out a book mostly saying similar things to HBD blogs like HBDchick and Jayman’s. 143 population geneticists participated in a weird Amish-style shunning ritual, denouncing the book’s findings.

    I don’t see anyone claiming population differences are 0% genetic. I do see a lot of people defending positions like “racial differences are largely superficial” and “race is a made-up concept” and “g is a flawed metric”.

    This will change. We’re starting to find the plus-variants of some alleles associated with intelligence now, and as we find more it will get harder to pretend group differences are an inscrutable black box. But as of 2016 I don’t think the scientific community is exactly covering itself in glory here.

    • MawBTS says:

      Also, Greg Cochran on how hard it is to study Ashkenazi Jewish intelligence.

      We have talked to some people about doing such a study. A researcher in Israel was interested, but met massive lack of enthusiasm from the powers that be: try to guess why. It might not be what you would think. Someone else I know wanted to do it until his adviser pointed out certain likely side effects – I believe the phrase “unemployable pariah” was used. That, of course, if his results supported our work.

      I asked an associate of Plomin – one interested in our ideas – and was told there was no chance of it ever happening.

      Some of the events in our push toward publication might also clarify the picture: one editor who had expressed interest called up, crying, and explained that the dean had said he’d close down the journal if our article was published. In another case, an editor rejected on the grounds the the Ashkenazi Jews had been farmers in the Middle Ages: which mistaken idea I figured he had picked up from Fiddler on the Roof, which he admitted was the case. When I told him enough to correct thaat misapprehension, he went on to say that we might well be right (judging in part from hsi teaching experience), but that of course his journal would never publish anything so controversial. So why a six-month review process when the answer was foreordained? Ya got me.

      In another example, someone interviewing at the University of New Mexico was asked about sphingolipid mutations and intelligence (it was related to his thesis) and he said he couldn’t afford to think about it – he wasn’t Harpending.

      I doubt if any university IRB in the country would approve it. I doubt if the Feds would ever fund it – with the exception of one possible scenario, but probably not even then.

      Still, there are ways. It’ll happen.

  70. Slocum says:

    At the time I thought this was some sort of exciting failure of modern science that I alone had figured out. But this was well after sharp people like Ioannidis were talking about it, and only a few years before everyone was talking about it.

    But it seems to me the point isn’t that Science never corrects or never corrects quickly but rather that scientific fields can go off the rails and stay off for long periods of time before a correction even begins. How many years/decades did experimental psychologists confidently publish vast quantities of irreproducible results before Ioannidis started talking about the possible problems?

    Which leads to the question of which fields are now in a state comparable to Psychology before 2005 — running off in the weeds awaiting a correction? And is the existence of a consensus really any kind of guarantee that a field is in good shape? After all, wouldn’t the overwhelming pre-2005 consensus in Psych have been that most published, peer-reviewed papers used sound research methods and could be successfully reproduced?

  71. sconn says:

    Scott, I wonder if you realize that you duplicated Leah Libresco’s exact argument for choosing to trust the Catholic Church: “It’s right on most things I know about, and in the past, when I thought it was wrong, either I was wrong, or I was wrong in thinking it disagreed with me.”

    Now, I agree with your point and I don’t with Leah’s, because the sort of trust you are suggeseting people have in scientific consensus is a lot weaker than the sort of trust Leah has in the Catholic Church. With science you can say “I’ll give science the benefit of the doubt on topics I don’t know a lot about, and when I think it’s wrong, I’ll consider the possibility that it’s actually right.” Whereas the Catholic Church demands total unquestioning loyalty, and that’s kind of where Leah’s argument faiiled to convince me.

    But I have come to a similar conclusion as you have, in recent years: more and more, I’ve watched Experts and The Establishment be right a lot more than I expected them to be. Heck, even the dreaded Mainstream Media got a lot more facts right on a lot more stories than any of the fringier stuff that I and my friends read. I guess in my 20’s I thought it was hip and original of me to mistrust authority, but I’ve realized that if you mistrust the experts only to trust in some rando with a blog …. you’re not being skeptical, you’re just being dumb. Of course experts can be wrong, but they’re usually a lot less likely to be wrong than people who are not experts.

    And that’s why I took my kids down to the health department yesterday and finally got them their first vaccines. Not that I got any new facts, because I have been well aware of the arguments on both sides for a long time, but because I started treating the information available on the CDC website as the reasoned consensus of experts, rather than The Lies Told By The Man To Keep You Down.

    [To the curious: my NT kid is still NT and my autistic kid is exactly as autistic as he was before he got the shots. I would like to say I didn’t lie awake the night before worrying about this, but I’m not as rational as I’d like to be.]

    • Anonymous says:

      Whereas the Catholic Church demands total unquestioning loyalty

      No, it doesn’t. You may have it confused with the kinds of secret societies that the Catholic Church forbids you to be a part of.

      • Evan Þ says:

        Well… to some extent.

        With science, it’s totally fine to say “The scientific consensus is generally right, but on this one issue, I’ve looked into things and I’m pretty sure it’s wrong.” It doesn’t matter how basic the issue is; even if I’m arguing for a flat earth, I can do it as long as I can mount a good argument and propose experiments and listen to the results. With the Roman Catholic Church – sure, there’re some issues and some means where it’s fine to disagree, like with the late Justice Scalia’s support for capital punishment. But if I say “Actually, I’ve looked into things and I’m pretty sure there’s no special charisma for bishops” – it doesn’t matter how I support it; they’ll tell me to keep quiet about it or throw me out as a heretic.

        • sconn says:

          Thanks, you expressed that well. I admit my phrasing was sloppy, but I think the point stands: to say, “I’ve rationally examined this dogma and reject it” is grounds for damnation. Leah, for instance, doesn’t see what’s so bad about gay marriage, but she admits that since she’s a Catholic now she has to accept it’s wrong anyway. As for me, I could have stayed Catholic if the Church would have let me believe there were errors in the Bible, but I hunted and hunted for the slightest justification for this in vain. A number of people warned me (hoping, I think, to *save* my faith) that if I persisted in believing there were errors in the Bible, I was no longer a Catholic.

          And of course they had a point because if you believe a certain body has a direct line to Jesus and is preserved from error, that’s going to have to be in 100% of the revealed teachings and not 99%. If it’s 99% right, its claim to authority is false.

          Science, being a method and not a hierarchical institution, doesn’t expect any such adherence. Disagreeing with orthodoxy on a few details may in some cases be a sign that you’re a really serious scientist who has put some thought into things. Yet there is still a good argument for agreeing with scientific “orthodoxy” unless you have really good evidence for believing otherwise.

  72. Kyrus says:

    when asked anonymously, most scientists think these differences explain about 25% – 50% of variance.

    What about when asked non-anonymously? I would say that we exactly then have a problem, when there is a significant difference between those 2 cases.

  73. MB says:

    The author focuses too much on who is right and who is wrong. As soon as the scientific consensus ceases validating the received left-wing wisdom (about innate differences, vegetarianism, global warming, or anything whatsoever), then all of a sudden it will cease to matter. Whatever field of science is at fault will revert to being — instead of an infallible oracle — a culturally bound practice that reflects the inherent bias of the white males that practice it and a patriarchal tool of oppression. Unorthodox scientists will see their funding dry up, will be cast out of universities, etc..
    For hundreds of years, Theology was the most important faculty in European universities and probably still is at Al Azhar, for example (not to mention a myriad other places, some in the US). This is where most Western universities are headed too, but with Theology replaced by ethnic, colonial, and gender studies, Title 9 offices, bias response teams, and probably social science departments (as long as they espouse the right ideas, which they currently do). It’ll be just like in the Communist countries, where the Scientific Marxism professors hold a power completely out of proportion with the practical importance of their field.
    The ethos of free scientific inquiry is unnatural and cannot last. The other state of things is more natural — some political or religious authority setting the limits of scientific inquiry and of public debate. Probably some learned experts will still be allowed to make naughty allusions to forbidden topics, cloaked in forbidding jargon, and feel smug about it, but they certainly won’t be allowed to do it in public (this was Summers’s mistake; as for the other Harvard debate — nobody has heard of it and this is what makes it tolerable).
    Scientists have lost control over the universities. When they no longer show themselves useful to their masters, they’ll be shown who holds all the cards. With all the harassment suits, anonymous reports and complaints, with the diminishment of tenure, with the growth of the management ranks, however, they probably know it already.

  74. Bram Cohen says:

    Is there some popular writing source which reliably and accurately represents what the scientific consensus in various fields is? It was nice of you to do a survey of surveys, but it seems like investigating and reporting on that information accurately should be viewed by *some* journalistic entity as core to their mission.

  75. wiserd says:

    “Most people understand this intuitively about past paradigm shifts. When a creationist says that we can’t trust science because it used to believe in phlogiston and now it believes in combustion, we correctly respond that this is exactly why we can trust science.”

    Phlogiston is real. In the modern day, it’s called “Carbon Dioxide.” And if you saturate air with carbon dioxide you will, in fact, prevent combustion. The Phlogiston theory may not be complete due to it’s failure to consider the role of oxygen, but it is predictive to a fair extent. Asimov’s “Relativity of Wrong” applies well here. We’re quick to criticize past theories as 100% incorrect. But they’re sometimes just incomplete in some critical way, and not entirely fallacious.

    • Macrofauna says:

      Counterpoint: Phlogiston is oxygen with a sign error.

      • Paul Brinkley says:

        Neither of these is correct.

        Phlogiston (according to an account written by Asimov, ironically) was believed to be that substance which other substances had which permitted them to burn. That thing turns out to be any substance capable of bonding exothermically with oxygen, which carbon dioxide will not do.

        In that sense, “oxygen with a sign error” is closer, but it also implies that the less oxygen something has, the more burnable it is – which is also not true.

        • John Schilling says:

          In that sense, “oxygen with a sign error” is closer, but it also implies that the less oxygen something has, the more burnable it is – which is also not true.

          It’s pretty close – most of the common terrestrial materials that aren’t burnable, aren’t burnable because they are already fully oxidized (e.g. water, most rocks). If you add the insight that rusting is just burning done very slow, you get closer still.

          • Gazeboist says:

            And that insight was already there, I think. (I might be misinterpreting, though)

          • lycotic says:

            @Gazeboist

            Sure. Isn’t phlogiston “what makes metals shiny”? That’s how it was first described to me.

          • Paul Brinkley says:

            My first exposure to the concept of phlogiston was Asimov’s Adding a Dimension. It was (al)chemists trying to figure out what made some things burn and other things not. It had nothing significantly to do with metals being shiny. And I think rust was not yet understood as being in any way similar to burning.

            Rather, phlogiston was theorized as that substance which burnable things possessed, and lost as they burned. Then chemists quickly realized that things that burned ended up weighing more. For a while, they played with the bold notion that phlogiston had negative weight. Eventually, someone likened respiration to possession of phlogiston (I don’t remember how). Possibly lifeforms slowly using up their phlogiston over their lifespan.

            Meanwhile, there are substances which contain no appreciable oxygen, but will not react with oxygen either (e.g. helium). They won’t burn. But chemists didn’t know this at the time.

            What I recall ultimately gave away the game was when someone noticed wood, candles, and other burnable things would stop burning if you put them in a glass jar for a while. And mice would eventually die, too. And then someone figured out how to control this by replacing the ordinary air in the jar with some other gas – burnable stuff would refuse to burn immediately, and mice would likewise die right away. At that point, the best explanation was now that there was something in the air, as well as in the substance.

          • Douglas Knight says:

            No, shininess and rust really were part of the original phlogiston theory.

          • FacelessCraven says:

            @Paul Brinkley – “And I think rust was not yet understood as being in any way similar to burning.”

            Scale forming on red-hot iron in a forge would be a pretty good early indicator of the connection.

  76. Majuscule says:

    Right before I checked SSC today, I read this study, because a friend of mine works for a magazine that published what I suspect is a sensationalized interpretation of it:

    Earning Less Than Their Wives Makes U.S. Men More Partisan

    I’m not a scientist, so I’m wondering how statistically significant these numbers really are (e.g. is a +0.3 change on an 8-point scale actually worthy of note?), and how the author’s conclusions read on the speculation-o-meter. Anyone with quantitative chops care to take a look?

  77. Christian Kleineidam says:

    It’s interesting that this article concedes that textbooks might not show the academic consensus. Saying textbooks aren’t what the academic consensus is about, sounds to me like playing moat-and-bailey.

    • James Picone says:

      The academic consensus exists in the peer-reviewed literature; textbooks are the afterthought that accelerates students who have no idea what they’re doing into grad students who can read and understand the journals. Don’t expect textbooks to be accurate or good. They’re going to be better than most sources, and they’re certainly much more dense than trying to understand a field by jumping into journals, but they’re not up to date and they’re not an accurate reflection of what ‘everybody’ thinks.

      • Urstoff says:

        Then why wouldn’t textbooks for non-majors be an accurate reflection of the scientific consensus? The safe presumption there would be that the students are not going to pursue a career in that field.

        • James Picone says:

          It’s not that the textbooks are deliberately written to not match consensus views, it’s that you get more prestige for writing papers and doing research than for writing textbooks, and textbooks don’t have self-correction mechanisms, so they tend to be worse.

          They also tend to be out of date because whatever’s up-to-date is usually substantially more complicated, so you get the lies-to-children thing as well.

          • Urstoff says:

            The need to simplify seems a more likely explanation to me than the lack of prestige, given that often textbooks are authored by some of the biggest names in their fields (although I don’t think there’s much reason to believe that being at the top of your field in research means you’ll be a good textbook author).

          • smocc says:

            @Urstoff It was not true for me in my undergraduate physics degree that the textbooks were by the biggest names in the field. I just polled my officemate and neither of us used textbooks written by people famous for anything but writing a popular textbook. (Gradschool is a slightly different story).

            Maybe other fields are different? I have since used textbooks by big names, but they are usually very advanced, and would not have worked well for me as an undergrad.

          • Urstoff says:

            I guess it does vary by field; economics and psychology have textbooks that are often written by successful researchers, and insofar as philosophy has anything like textbooks, they’re generally written by well-respected philosophers.

          • sconn says:

            As I understand it, the big names on the front of textbooks are not really the authors. The headlining author does some general work — perhaps writing out the table of contents, the introduction, and a favorite chapter or two — but the main body of a textbook is often written by a committee. This is especially true in children’s textbooks but may also be so for college texts. The big-name author is there for the prestige, but writing a textbook is boring and tedious, so it’s done by much lowlier peons whose names aren’t included.

    • BBA says:

      Textbooks sometimes contain outright false, long-debunked “facts.” The map of the tongue, showing which parts have the taste buds for sweetness vs saltiness, etc., was based on a mistranslation and completely at odds with common sense. In fact, the buds for all tastes are present on all parts of the tongue. Yet the tongue map is still taught in schools, because the teachers and textbook writers learned it from their teachers and textbooks, etc.

  78. RH28 says:

    But I have a vague memory of having recently seen a survey (which I can’t find) that even experts in softer fields like sociology are generally in favor of IQ and admit that it has its uses

    I don’t know about IQ, bit I have memory of a recent study showing that sociologists deny sex differences, with only 36 percent or so saying, for example, that it’s plausible that men are more promiscuous due to differences in reproductive strategies.

    https://www.washingtonpost.com/news/wonk/wp/2014/10/28/liberals-deny-science-too/?utm_term=.200fe636ae7d

    And we’re only talking sociology here, not even getting into fields that are further divorced from reality.

  79. RH28 says:

    Regarding the plausibility of the Judith Rich Harris finding on parenting above, as someone who knows many people who grew up with immigrant parents, it seems very plausible to me that parents have no lasting impact on personality. Meet a Pakistani, Indian, or Chinese American kid and they’re just like all their friends, then you’ll see they’re parents and they’re so different you wonder how they communicate with them. Look at something as subconscious as accent. Makes evolutionary sense too, as you need to fit in with the larger society and find mates there, your parents are required to love you no matter what.

    • HeelBearCub says:

      they’re just like all their friends

      Proves to much, as this would imply that genetics has nothing to do with personality either.

      • RH28 says:

        Good point. I guess that’s why maybe we should rely on science more than intuition when it comes to human development, since intuition can push us in all directions, while the science of behavioral genetics has been telling the same story for decades.

        • HeelBearCub says:

          I mean genetics doesn’t actually tell us much of a different story than common knowledge, which is that people tend to take on an amalgam of the personalities of their relations, but that this is also affected by what happens to them.

          I’ve never really understood the “Nature or nurture?” question, as they answer is pretty clearly “Yes”. People keep seeming to point to every point in favor of the way one or the other affect the individual as if that becomes evidence that the other is irrelevant, which seems like balderdash to me.

          • IrishDude says:

            I’ve never really understood the “Nature or nurture?” question, as they answer is pretty clearly “Yes”.

            Knowing whether it’s 10% nature and 90% nurture or vice versa seems important for our understanding of the way the world works and how it can be influenced, no?

  80. Squirrel of Doom says:

    Let’s say we accept that Scientific Consensus is the best measure of knowledge there is.

    Then what fields do we consider “Scientific”? I don’t think it’s an easy question.

    Everyone would accept Physics, but many would deny Gender Studies. Yet both fields have university departments, professors and PhDs.

    How do we tell fields where we trust the consensus from bullshit fields that have taken on the visible attributes of science, but not the substance of it?

    • Ransom says:

      Those that have associated/derived technologies that work. Physics, chemistry, much of biology. Economics and medicine are at the border. Pretty much anything else, no.

      • Enkidum says:

        Does neuroscience count as biology? Because there’s plenty of technologies that work there. Does cognitive science more generally, the applied arm of which is called “marketing”, count as something there as well?

        • Ransom says:

          I don’t know what you mean by neuroscience (what is the technology in that case?) but for marketing, I’d say sure, though it’d be closer the economics or medicine than physics. I guess I’d count the expertise of the professional politicians too, though they’re not really different from marketers, I suppose.

          There’s not really a “scientific consensus” that’s publicly available in those fields though, so I’m not sure whether they fit with this thread. But I bet the marketers (especially) can pretty well predict what an ad campaign will do, maybe even predict pretty well how much sales will rise (or stop dropping).

          The ability to predict, and get data to see that it actually works, is what suppresses the politics, tribalism, and ideology, and makes the consensus reliable. Sales and votes are good data.

          • Enkidum says:

            In terms of neuroscience, things like TMS / DBS / BCI are very robust technologies, and more generally any form of neurological interventions, be they drugs, surgery, or whatever. I suppose that falls under medicine as well, but really it’s coming directly from the scientists in most cases.

            There’s actually plenty of scientific consensus in psychology and other fields of cog sci, much of which is directly applied to marketing. We know a great deal about, for example, what visual/auditory features attract attention, what makes things easiest to find, what makes things stick in your memory, etc. Many of these findings are at the level of having been replicated hundreds or thousands of times, so they’re simply part of the background of the field.

  81. Marshayne Lonehand says:

    Glen Raphael asserts (dubiously) “Nobody ever says ‘the overwhelming scientific consensus is that ‘2 + 2 = 4.'”

    Lol … and isn’t this the sort of unsupported assertion that hard-core rationalists never challenge? Uhhh … well … hardly ever, anyway! 🙂

    Mathematician Vladimir Arnold’s celebrated aphorisms come to mind:

    “Mathematics is a part of physics. Physics is an experimental science, a part of natural science. Mathematics is the part of physics where experiments are cheap.”

    In greater detail, Arnold explains:

    [The] development of mathematics resembles a fast revolution of a wheel: sprinkles of water are flying in all directions.

    Fashion — it is the stream that leaves the main trajectory in the tangential direction. These streams of epigone works attract the most attention, and they constitute the main mass, but they inevitably disappear after a while because they parted with the wheel.

    To remain on the wheel, one must apply the effort in the direction perpendicular to the main stream.

    Among the world’s citizen-scientists, the accumulated evidence for the reality of anthropogenic global warming has become as mathematically evident — in Arnold’s natural-science sense of “mathematically evident” — as the proposition “2 + 2 = 4”.

    More generally, for an overwhelming majority of the world’s STEAM-professionals, the world’s accumulating and soberingly Arnold-ian mathematical and scientific realities have become (in recent decades) as obvious as “2 + 2 = 4”.

    If humanity is to “to remain on the wheel” of a healthy planet (to borrow Arnold’s metaphor), the STEAM community’s creative cognitive efforts are best applied Arnold-style, “in directions perpendicular to the main stream” … that “main stream” being the stream of too-easy, too-self-serving, too-ideology-directed, too-math-denying, too-science-denying, too-far-left and too-far-right narratives.

    • rlms says:

      Using “STEAM” means you’re not even trying.

    • Nornagest says:

      John, please go away.

    • Urstoff says:

      To post in this subthread, you must google translate your post into the language of your choice and then back into English.

      To post to this subhilo must google translator post in the language of your choice and then back to English.

    • John Schilling says:

      Among the world’s citizen-scientists, the accumulated evidence for the reality of anthropogenic global warming has become as mathematically evident

      Your two embedded links point only to discussions of arctic sea ice volume without mention of cause. Even accepting the dubious mathematics of “ice volume change = global warming”, the anthropogenic part of your claim is wholly absent.

      You aren’t working at the 2 + 2 = 4 level here; you aren’t even working at the 2 + 2 = ‘D’ level. This is the “I fucking love science” level, and I’d fucking love it if you took it somewhere else.

      • Cerebral Paul Z. says:

        Look again: he’s working at the John “Hexapodia as the key insight” Sidles level.

        • Nornagest says:

          Hexapodia? I don’t remember that one.

          • Brad says:

            It’s a reference to Vernor Vinge’s A Fire Upon the Deep. There’s a usenet analog in there and there’s one poster that is some sort of weird gas cloud alien thing and his messages make no sense.

        • Ilya Shpitser says:

          Flawed analogy, that dude was completely right the entire time.

          • John Schilling says:

            If it turns out that John Sidles was right all along, I’m dialing Countermeasure up to eleven and pulling Earth all the way down into the Unthinking Depths. Consider it a mercy.

          • suntzuanime says:

            The guy was incomprehensible. So what if someone who already knows the answer can decode what he was saying into something meaningful? If people who don’t already know what you’re trying to say can’t work out what you’re trying to say, in a social sense you’re not right, you’re not even wrong, you’re incomprehensible.

          • Cerebral Paul Z. says:

            (counts legs carefully) Are you sure?

          • John Schilling says:

            (counts legs carefully) Are you sure?

            Hmm. A Skroderider agent in our mist, revealed by his arrogance and ignorance?

            Death to Vermin!

      • Marshayne Lonehand says:

        It is a pleasure to provide further references, John! After all, isn’t there a broad consensus among rationalists that the efficacy of rationality increases in utility in proportion to increasing appreciation of historical, thermodynamic, and mathematical knowledge? And isn’t there a converse consensus among rationalists too, that no amount of rationality can entirely compensate for scientific ignorance?

        Specifically in respect to increased factual appreciation of climate science, Spencer Weart’s professional-grade, scrupulously referenced, free-as-in-freedom, continually-updated, on-line resource The Discovery of Global Warming is an indispensable public resource.

        As Weart’s concluding essay Reflections on the Scientific Process, as Seen in Climate Studies states:

        Our human comprehension of climate goes beyond scientific reports into a wider realm of thinking.

        When I look at a snowless street in January I may see a natural weather variation, or I may see a human artifact caused by greenhouse gas emissions. Such perceptions are shaped not only by scientists, but by interest groups, politicians, and the media.

        With global warming the social influences run deeper still. Unlike, say, the orbits of planets, the future climate actually does depend in part on what we think about it. For what we think will determine what we do.

        Most rationalists will agree with Spencer Weart that “what we think will determine what we do” … and from this fact follows an obligation (both moral and pragmatic) of those who comment upon climate-change, to cultivate an appreciation of its historical, thermodynamic, and mathematical foundations.

        For working so effectively to cultivate a rationally-grounded public appreciation of the historical, thermodynamic, and mathematical foundations of specifically anthropogenic components of climate-change, the rationalist polity owes inestimable appreciation and thanks to all of the world’s “centripetally directed climate-science workers” (to borrow Vladimir Arnold’s marvelous “spinning wheel” metaphor) … and in particular Spencer Weart! 🙂

    • BBA says:

      You are John Sidles and I claim my five pounds.

  82. martinepstein says:

    The link now includes a good response to this post. The two simplest things you got wrong are that this is just the first 10 alphabetically and not the “top 10”, and also you mixed up the acoustic and optical Doppler effects.

    • Jiro says:

      Looking through that list, I see two more bad entries. Gaia theory is still nonsensical, but is just subject to motte/bailey where the defensible version is “there are feedback processes in nature”.

      And the idea that bacteria causing ulcers is a case of science rejecting a theory unfairly has been debunked.

      • Douglas Knight says:

        The scientific consensus is that CSICOP is always wrong. It took a century to accept that H pylori caused ulcers. Curing 30,000 people of ulcers was a crime.

        • Jiro says:

          https://sciencebasedmedicine.org/the-case-of-john-lykoudis-revisited-crank-or-visionary/

          The question is – what did Lykoudis do to convince the scientific community of his claims. Did he perform carefully controlled double-blind placebo-controlled trials? Did he attempt to enlist the help of a microbiologist to try to isolate the organism? Or did he just expect people to take his word for it?

          He was right, but if he didn’t know how to prove it, there’s no way to tell him apart from all the quacks except in hindsight.

          • Douglas Knight says:

            Those standards are anachronistic. He had more evidence than most treatments approved in the 60s, dramatically more than other ulcer treatments, as the post concedes. Yes, “science” is about process. He failed the process by being a country bumpkin. I can perfectly well condemn their process. Maybe I can also condemn Lykoudis for pissing people off — the next Galileo.

          • suntzuanime says:

            This article is an attempt to argue that correct ideas will be readily adopted by the scientific consensus, and that the narrative that scientific mavericks are later vindicated is an incorrect one. Given that, saying “he’s a scientific maverick so he doesn’t count” is completely missing the point.

  83. Todd K says:

    Larry Summers was not fired from Harvard but received a no confidence vote as president of Harvard. Steven Pinker, if the president, would have almost certainly received the same treatment even though more diplomatic.

    • The Nybbler says:

      not fired from Harvard but received a no confidence vote as president of Harvard

      Doesn’t it amount to the same thing? Or is Harvard like the UK Labour party where no-confidence votes are non-binding?

      • Todd K says:

        Summers took a sabbatical and then returned to teach at the Harvard Kennedy School where he is also teaching a course this semester.

  84. abc says:

    Appealing to “scietific consensus” sounds at best like a classic case of Goodharts law. Sceince is as the motto of the Royal Society states based on “taking no one’s word for it” or as Feynaman famously stated “the Belief in the Ignorance of Experts”. Of cource, if eveybody else is doing the verifying, you don’t have to. However, once people start basing their opinion on “scientific consensus”, they stop verifying and “consensus” stops becoming a useful guide.

  85. TheRadicalModerate says:

    It should be pointed out that the only time that scientific consensus is attacked publicly is when it has some implication on public policy. Beyond that, non-scientists don’t care, and only scientists with very flimsy ideas are going to risk the appeal to authority implied by invoking the consensus.

    In the public policy sphere, things are a lot more difficult, because somebody is wielding the consensus as a political weapon. Almost by definition, that means that it’s being used to erode the interests of one group in favor of the those of another. Since discrediting the consensus is a nice, simple way to disarm the weapon, worrying about whether the possessors of the gored oxen are going to fight dirty is pointless–they are.

    That seems to me to be an excellent argument for pulling the “scientific consensus” tool out of the political bag very judiciously, so as not to leave it all nicked-up and crufty. As a guideline for when it might be useful to use it as a political tool, I think these four cases pretty much fill the scenario space for when something like a consensus might have political value:

    1) The consensus has identified a problem and suggests a straightforward engineering solution to fix it. (Please, give us money to fix it.) Examples: Flood control in the Mississippi delta, infrastructure deficiencies, and a wide variety of public health no-brainers.

    2) The consensus has identified an issue and suggests that the issue should be studied. (Please, give us money to study it.) Examples: Studying how to avoid a fast-takeoff AI, or how to spot and mitigate asteroids on a collision course with Earth.

    3) The consensus has identified an area that’s important and/or useful enough to warrant broad-based public education, or a public behavior change. (Please, give us money to educate, and the authority to smite the wrong-thinkers and/or craven.) Examples: teaching evolution, public education about how vaccines work, or anti-smoking legislation and education campaigns.

    4) The consensus has identified an urgent problem, but doesn’t have an effective engineering solution to solve it. (Please, give us money to do… something?) Examples: obesity, climate change.

    Using scientific consensus in scenario #1 devolves–quite properly–into an argument about costs and benefits, which, since engineering can provide answers to those questions, probably gets solved semi-rationally. If you’ve truly got consensus in scenario #2, it should serve as a guide to avoiding spending public monies on crackpottery, although you have to be careful not to let it stifle things that seem a little out there.

    Using consensus in scenario #3 is highly problematic, because it implies that we’re going to do social engineering (to the extent that that isn’t an oxymoron) in service of something that some group will refuse to believe, or be disadvantaged by. As a general rule, I believe that the scientific community, as keepers of the consensus, would do best to be aware that there’s a fine line between a simple statement of facts and advocacy, and stay well on the side of the former.

    But at least in scenario #3 there’s something effective to be done, and once public debate has ground the issue to bits, the thing done will likely work.

    It’s the combination of urgency and ineffectiveness associated with scenario #4 that makes it so vastly unpleasant, and so fraught with peril in terms of scientific credibility. Simply shouting “do