back
114 comments
This isn’t a random crank we’re talking about. Robert Fisher is widely acknowledge as one of the foremost scientists in the last 100 years. His contributions to statistics alone would earn him that. https://en.wikipedia.org/wiki/Ronald_Fisher

He was wrong about smoking, but the more you read about him the less you’ll believe it had much to do with money. He was used to being right about so many things, and in this area he was blind.

The article also throws shade at him as a “eugenicist.” I looked it up, and again, the truth is more complex. He wrote this in the 50s:

“I am sorry that there should be propaganda in favour of miscegenation in North America as I am sure it can do nothing but harm. Is it beyond human endeavour to give and justly administer equal rights to all citizens without fooling ourselves that these are equivalent items?”

So first — even using the word “miscegenation” puts you in a bad camp, and there’s no defending his attitude against interracial marriage. OTOH he seemed honestly to believe in the “equal rights” part, too. Too much of the old British “white man’s burden” bullshit, I believe.

Modern connotations of the word "miscegenation aside" aside (the word's denotation means "mixing of races", which is exactly the correct terminology for the issue), and laying aside modern understanding of "race" vs "ethnicity", in Fisher's era racial separatism was considered a solution to racism, not only by White people but also by many leading Black rights activists, such as Malcom X and Muhammad Ali in the Nation of Islam. Ethnonationalism may well be a bad idea in theory or or practice, but it's not purely an oppressor's idea. (It's not obvious that there is any unalloyed good solution to racism and human nature.)

https://ccnmtl.columbia.edu/projects/mmt/mxp/speeches/mxt14....

https://www.npr.org/sections/codeswitch/2016/06/10/481414008...

>The article also throws shade at him as a “eugenicist.” I looked it up, and again, the truth is more complex.

You didn't look it up very well.

>In 1911, Fisher became founding Chairman of the University of Cambridge Eugenics Society, whose other founding members included John Maynard Keynes, R. C. Punnett, and Horace Darwin. After members of the Cambridge Society – including Fisher – stewarded the First International Eugenics Congress in London in summer 1912, a link was forged with the Eugenics Society (UK).[122] He saw eugenics as addressing pressing social and scientific issues that encompassed and drove his interest in both genetics and statistics. During World War I Fisher started writing book reviews for The Eugenics Review and volunteered to undertake all such reviews for the journal, being hired for a part-time position.

I think that if you:

1. are the founding chairman of the University of Cambridge Eugenics Society,

2. stewarded the First International Eugenics Congress in London in summer 1912,

3. saw eugenics as addressing pressing social and scientific issues, and

4. started writing book reviews for The Eugenics Review and volunteered to undertake all such reviews for the journal

..you are a eugenicist.

My research consisted of clicking on, and reading, the link you yourself posted.

This is my favorite passage from the article:

"Beyond this, people make mistakes. Brilliance represents an upper bound on the quality of your reasoning, but there is no lower bound. The most brilliant scientist in the world can take really dumb stances. Indeed, the success that often goes with brilliance can encourage a blind stubbornness. Not always—some top scientists are admirably skeptical of their own ideas—but sometimes. And if you want to be stubborn, again, there’s no lower bound on how wrong you can be. The best driver in the world can still decide to turn the steering wheel and crash into a tree."

It is one of those profound realizations that seems so obviously true it's irrelevant. But then ask if we evaluate the decisions and statements from smart people this way. Generally the answer is no.

While the brilliant person will have higher quality reasoning on average due to the stretching of the distribution... any individual belief or statement they come up with is being drawn from a distribution that still includes boneheadely wrong.

> While the brilliant person will have higher quality reasoning on average due to the stretching of the distribution... any individual belief or statement they come up with is being drawn from a distribution that still includes boneheadely wrong.

One is reminded of Linus Pauling's flubs on two counts: his obsession with Vitamin C and his denial of quasicrystals.

For some people at that time, smoking was a non-trivial part of their identity. Or even a significant part of what it meant to be a proper Englishman (that and tea). Fisher strikes me as that sort, just look at pictures of him.

The (fairly obvious) lesson here is that people lose their objectivity when it comes to fighting over stuff that involves their identity.

My first thought was that obviously he was a smoker.

I loved cigarettes. I haven't smoked in almost 15 years and I might say I still love cigarettes. There really is no replacement for the feeling of being a smoker, waking up in the morning and drinking coffee with a cigarette as the sun comes up. That is a large part of what makes the addiction so bad.

I say this even with the benefit of knowing how horrible they are for health. It is why I quit eventually.

It is really hard to be objective about your partner when you are in love. A highly abusive partner at that.

I suspect the only thing stronger in humans than love is denial. The combination can be especially deluding.

Fisher’s main scientific and statistical argument boils down to the possibility that gene variants that contribute to lung cancer are tightly linked with gene variants that make nicotine more or less addicting. By “linked with” he meant “close together on the same chromosome”. This kind of linkage can lead to strong a statistical association but without a mechanistic association (guilt by neighborhood).

Disproving this hypothesis is tricky, and Judea Pearl does a brilliant job of explaining the problem and its solution in his marvelous book: The Book of Why.

Fisher gets “assist points” for debilitating and killing millions, although full horrible credit goes to cigarette companies and their advertising co-conspirators.

Judea Pearl points out one cruel irony: The cholinergic receptor gene CHRNA5 that modulates risk of nicotine addiction also modulates lung cancer risk separately. To sort out the causality we now use Mendelian randomization.

Bottom line: smoking cigarettes does kill even when you tidy up the statistics and genetics.

To quote: "Ironically, Fisher was proven right, albeit in a very limited way: such genes [that increase both the tendency to smoke and the tendency for lung cancer] do exist."

The actual issue was not that (Cornfield wrote a paper in 1959 showing the effect was too small). It was that Fisher continued to repeat one finding in one study, despite that said finding had not been replicated in new studies (namely, lung cancer patients described themselves as inhalers less often than the controls), and continued to obstinately ignore all the other research coming out. But it was only 3 years between Cornfield and Fisher's death in 1962, so perhaps Fisher simply did not have time to change his mind.

Why would you expect anything else? Becoming "a top scientist" doesn't turn you into some non-human intelligent entity without flaws for the rest of your life, it just means you've done some notable scientific work in some areas. And that work doesn't even have to be correct! You could've been a top scientist with later top (or even low) scientists invalidating all your top science work
I agree with you but I think society has a certain bias that way — a saint or genius bias or myth. It extends into science and engineering itself, although I don't think it's limited to that domain either.

What's interesting to me is how some can recognize this in the past but not the present, or think it only applies in certain domains.

As far as I can tell the only reference to something that Fisher said was a clip from a newspaper article where he said that criticizing smoking was terrorism. The rest appears to just be other contemporary evidence that the cigarette execs knew it was bad but some scientists (not Fisher?) didn't believe that it was bad.

Am I missing something -- does this article spell out to what extent Fisher himself defended smoking?

Here is Fisher, in a letter to the editors of _Nature_ (perhaps the most prestigious journal imaginable) "Cancer and Smoking"[1]:

> If, for example, it were possible to infer that smoking cigarettes is a cause of this disease, it would equally be possible to infer on exactly similar grounds that inhaling cigarette smoke was a practice of considerable prophylactic value in preventing the disease, for the practice of inhaling is rarer among patients with cancer of the lung than with others. [...] There is nothing to stop those who greatly desire it from believing that lung cancer is caused by smoking cigarettes. They should also believe that inhaling cigarette smoke is a protection. To believe either is, however, to run the risk of failing to recognize, and therefore failing to prevent, other and more genuine causes.

As a geneticist, he of course took the position that it was a smoking gene confounding the causation.

[1]: https://www.nature.com/articles/182596a0

One aspect that modern finance has laid clear is the concept of monetization of various assets one possesses. In the past, it wasn't clear that one's brand was monetizable or how one could do that. Individuals would happen upon these now and then, even if it wasn't systematized. The most obvious answer is that R. A. Fisher happened upon a way to monetize his brand in a way that aligned with his politics: he was a good statistician, and therefore people believed him, but the value of "being a good statistician and known to be so" is much higher than just being a good statistician and this was one such thing where he could extract more value. The part about aligned politics is that it helps when you're trading reputation.

Today, most of this is well understood. MIT sells its brand under MIT Media Lab, something you can easily understand if you read the theses published by this division of the university. Other universities sell their brand under things like 30 day courses that grant a certificate named similarly to their graduate degrees. In some sense, they are internalizing the surplus generated by the brand. Interesting model.

See quotes in my profile.

If the world should blow itself up, the last audible voice would be that of an expert saying it can't be done. - Peter Ustinov

Sounds like Ted Turner needed to change the end of the world video on CNN.
I think Ancel Keys is in this category, and we are still suffering and dying from his opinions, which are a leading reason for the upside down food pyramid. I don't doubt his sincerity or intelligence at all. He's just an instance of how badly top scientists can get it wrong with the best of intentions, and how enormously expensive such failures can be.
You see this time and time again. A scientist/mathematician/technological leader who thinks because they are the "cat's pajamas" in one field that they are equipped to chime in on another. One example is John Clauser, Nobel winning physicist, making a downright embarrassing attempt to "debunk climate change." (see https://www.youtube.com/watch?v=_kGiCUiOMyQ) Another is Elon Musk, who seems to have an opinion on everything. Sometimes there is money or malice involved -often just hubris.
This was my brother's theory on why Steve Jobs didn't use traditional cancer treatments. The way he explained it to me was "Put yourself in his place: your whole life every time everyone said you were wrong about something and it was gonna destroy you you did it anyway and made a billion dollars. This time they were right and he was wrong but you can't blame him for playing the odds one last time."
Nobelists are so prone to it that Wikipedia has a page dedicated to it:

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

It's hardly limited to Nobel laureates. But it's certainly a strong marker that somebody had indeed tremendous scientific skills, and then failed utterly to apply that ability later.

Elon is only an authority in the topic of raising money and mostly not trashing the work of the better men whose work he buys when he correctly infers he doesn't know enough.

You see this wherein he has mostly avoids ruining spacex because he knows he isn't a rocket scientist but absolutely wrecks Twitter because he thinks he understands it.

See also his delusions about mars, desire to use a minisub in a cave rescue so tight divers had to remove tanks, stopping services needed for 2FA, and suggestion for relay satelites between earth and mars to communucate faster.

He's absolutely intellectually average with a way above average personality, ego, and wallet.

> You see this time and time again. A scientist/mathematician/technological leader who thinks because they are the "cat's pajamas" in one field that they are equipped to chime in on another.

You don't think one of the world's foremost statisticians should have felt that he was qualified to participate in a purely statistical argument?

This idea for me comes from a friend that does patent law. He said the hard thing about a patent is knowing what questions need to be answered. Not the answers. It's all about framing the problem, all the hard hard is really there.

We reward and give status to scientists that come up with answers and the people doing the hard framing work not so much. The two guys that used standard crystallography techniques to figure out DNA is a double helix got the Nobel prize. The lady that figured out how to crystalize DNA and get the films is completely ignored.

So yeah top scientists high on their own ego will totally biff it when dealing with some other field they know nothing about.

I have a friend who really got into studying people who buy into conspiracy theories.

His conclusion mostly is that cleverness does not shield you from believing falsehoods. These are 2 distincts properties of the mind. In fact, it's quite the opposite. Smart people are very good at finding causes that justify what they believe in.

The point is that there is a ton of things that we know that are in fact based on beliefs. Like: did I ever see an atom with my own eyes? Nope. Did I see a clock slow down because it flew in a rocket really fast? Nope. Did I ever check that the moon landing looked legit? Nope.

One of my favorites is the controversy about Q-tips, there are tons of people who say it's bad for your ears, and then there's a guy who did a study that concluded that no study ever proved that Q-tips were bad for your ears. I know Q-tips are probably bad for my ears, but they feel so good, so whenever my wife brings up that I should stop using them, I always refer her to that one guy who tried to prove that Q-tips weren't that bad.

Perhaps it's just user error, but if you compact the wax in your ear by shoving stuff into it then you'll be in for a world of pain when the doctor uses a watergun to uncompact it.
I think it's fascinating that lack of information can generate both belief and disbelief in equal amounts.

For example, it took a while for germ theory to gain acceptance. People cannot see germs, so it's hard to believe in then. But people also fervently believe in ghosts and bad luck, neither of which can be seen.

Both are ultimately "caused" by a lack of information. When the lack itself is enticing because it creates mystery, it invites fantastic theories and superstition, but if it's not, it generates doubt and resistance.

Conspiracy theories resonate because you can always form a picture that explains the mystery in an enticing way in the same primitive way that weather must be created by the gods and ghosts are responsible for strange sounds in the attic.

This kind of dichotomy between credulity and incredulity is seen everywhere where humans need to deal with the absence of clear information. What I find fascinating is that while science has given us tons of powerful tools to reveal the unseen, we are still so powerfully drawn to the opposite side.

“The first principle is that you must not fool yourself—and you are the easiest person to fool”.
This comes off as borderline hagiography, when the answer is pretty clear. People figured out that chronic smoking was bad for you a LONG time ago, the reason that understanding was undermined rebutted had nothing to do with science, and everything to do with money.
R. A. Fisher developed many of the foundational techniques of modern statistics in an attempt to support his odious beliefs in “racial hierarchy.”

There is much to learn from considering this reality, but most will dismiss it as irrelevant.

Can you provide evidence of this strong claim? It looks like you’re conflating his population genetics work with eugenics/race science (which in turn have significant differences that I won’t go into).
>There is much to learn from considering this reality

Like what?

It's kind of the same situation with alcohol now: a lot of people denouncing it, the alcohol industry throwing a lot of shade, and top scientists making confident pronouncements (on both sides).
> for whatever reason, cigarettes were a conservative cause

For.. the.. money?

Money is a very conservative cause. Anything that gets in the way of money is "terrorist" and "commonism" (sic).

It should be noted, tobacco was the cash crop that finally made England's American adventure profitsble. Without it, we euro-americans might be speaking french or Spanish across all of north america now...

Does it even need to be that analytical? Conservatism can also also be general skepticism of change. And I don't know if you've ever tried to quit nicotine before but it's very good at convincing you it's a relatively harmless habit that helps cope with all sorts of stress and anxiety. If you're already inclined to be generally skeptical of change, smoking would slot right in there.

That said I'm curious if we're projecting back strong moralistic or party/ideology-aligned takes on smoking before it was really prevalent. I don't think anyone is questioning the role money plays in broadly misleading the public about smoking itself, just that the ideological/partisan divide such as it exists today might not map cleanly to the past.

See also James Watson, Nobel winner for discovering DNA, who continues to make racist remarks despite scientific evidence that he's incorrect. This has unsurprisingly led to the complete destruction of his reputation and considerable loss of income. People don't need financial incentives to be wrong, apparently.
See also: Linus Pauling and vitamin C https://en.wikipedia.org/wiki/Linus_Pauling#Medical_research...
> But then this just pushes the question back one step: How could this brilliant statistician be so naive?

I would suggest that if they're taking money to spout bad science, they're not actually brilliant. So I would suggest this pushes the question back yet further, why do we (still?) think he was brilliant?

The guy seems like kinda a dick. But here's what Wikipedia says Fisher is known for https://en.wikipedia.org/wiki/Ronald_Fisher

Fisher's exact test; Fisher's inequality; Fisher's principle; Fisher's geometric model; Fisher's Iris data set; Fisher's linear discriminant; Fisher's equation; Fisher information; Fisher's method; Fisherian runaway; Fisher's fundamental theorem of natural selection; Fisher's noncentral hypergeometric distribution; Fisher's z-distribution; Fisher transformation; Fisher consistency; F-distribution; F-test; Fisher–Tippett distribution; Fisher–Tippett–Gnedenko theorem; Fisher–Yates shuffle; Fisher–Race blood group system; Behrens–Fisher problem; Cornish–Fisher expansion; von Mises–Fisher distribution; family allowance; Wright–Fisher model; Ancillary statistic; Fiducial inference; Intraclass correlation; Infinitesimal model; Inverse probability; Lady tasting tea; Null hypothesis; Maximum likelihood estimation; Neutral theory of molecular evolution; Particulate inheritance; p-value; Random effects model; Relative species abundance; Reproductive value; Sexy son hypothesis; Sufficient statistic; Analysis of variance; Variance

That's a pretty long list of achievements.

Why? Why do people insist that brilliance means some moral purity? What makes brilliance and greed impossible together?
Look him up, perhaps. If you’re a geek, nontrivial parts of your profession wouldn’t be possible without him.
> So I would suggest this pushes the question back yet further, why do we (still?) think he was brilliant?

Because he (mostly single-handedly) invented modern statistics.

Gelman comes down pretty hard on Fisher here but he doesn't question his brilliance.

See also Serge Lang on HIV/AIDS.
Money. Obviously.
Thought-terminating cliches prevent you from finding much more interesting causes of disreputable behavior.
I’m honestly baffled by how swiftly everyone demonizes cigarettes as the ultimate evil, as if that’s a done deal.

Whenever group-think is this loud, it’s a huge red flag we should crack open the raw data ourselves. Fisher wasn’t some mustache-twirling villain, just a stubborn contrarian pushing against the orthodoxy. And if Big Tobacco slipped him a check, that doesn’t automatically nuke his math.

Correlation hype is easy, real causation proof is hard, and I’d love to see all the data and methodology. We don’t push science forward by chanting from the same hymn book. We do it by asking hard, unpopular questions.