598 Comments
Comment deleted
Expand full comment
Comment deleted
Expand full comment

Interesting commentary. I’ve always believed Nostradamus was full of shit lol

Expand full comment

An excellent introduction into how to perform a tarot reading.

Expand full comment

" Fukuyama said some (no offense) kind of vapid stuff" -- I'm afraid that has to be filed under "Tell me you haven't read The End of History and The Last Man." It can be described in many ways, but "vapid" is not one of them.

Expand full comment

Well the biggest issue is that a lot of people predict a lot of things and so someone has probably predicted any given event. Probably lots of people predicted a peaceful fall of the Soviet Union. Many of them also predicted many other things but generally you only hear about successful predictions.

That book that predicted the peaceful fall of the Soviet Union was just some guy getting lucky. Predictions of the future is very much 1 million monkeys on typewriters.

Politics is a great example. Every cycle we get a dozen new guys who predicted shocking thing and a dozen guys from the previous cycle who predicted shocking thing but were wrong this time.

Maybe 10% of the population if not 5%, understands statistics and also applies it effectively in their actual life outside of academia. So predicting stuff is is always gonna be a shitshow and the media, both sincerely and cynically for profit is going to misunderstand predictions and take one lucky guess guys too seriously and guys with high calibration on important predictions not seriously at all.

Neither the average reporter or the average cashier is gonna check the calibration record of someone who predicts a high intensity issue wrong or right.

Expand full comment

I think making bets with knowledgeable and honest people is a good way to make predictions.

This way you can point to you winning the bet as evidence enough that you were right (or wrong).

The betting mechanism tends to nail down definitions and appoint arbiters etc. So that's not purely down to public opinion. (Of course, no one can force other people to defer to the judgement of your arbiter. But it's a strong Schelling point.)

Expand full comment

There’s three hard parts to predicting things:

1) hypothesis generation

2) calibration

3) phrasing things so that resolution is unambiguous

The problem with hypothesis generation being hard is that it can legitimately be a great prediction even if it’s only 10% likely to happen—if I predict 10 really surprising things 100 years out, and 1 of them comes true, that’s AMAZING, not a failure.

The problem with calibration being hard is that without stating actual probabilities and averaging over a long period of time, people will interpret all of your predictions as 100% confident. And even if you do, you’ll get bad press every time you make a 70% prediction and it’s wrong (c.f. five thirty eight in 2016)

The problem with ambiguous resolution is that companies write 100 page contracts and then spend 100 million in court trying to solve this problem and it’s still a shitshow.

Expand full comment

OK, not the topic, but... WTF is a "homo sapiens supremacist"??

Expand full comment

I predict that linking this at the top of predictions will have no discernible effect on the potshots Scott will get.

Expand full comment

You shouldn't be able to write about Fukuyama unless you've read The End of History and the Last Man, which it's clear you haven't. Fukuyama makes pretty much the opposite claim of what everyone says he does in that book.

Expand full comment

"There were a few guys in some caves doing terrorism, they got lucky once, the US got angry and invaded a few countries, and then everything continued as before."

Scott, I think you forgot the part where the US spent ~$2.3 trillion on Afghanistan and ~$3 trillion on Iraq, with enormous negative consequences to American's long-term well-being. Even if that money is funded by deficit spending that isn't directly paid, it gets amortized in inflation and other indirect ways.

Expand full comment

Re: Trump and minorities.

If you read the popular press now there is a lot of angst/hand wringing on the Democratic side due to Hispanic defections in the voting public to the GOP. Trump kicked that off when somehow he increased his share of the minority (and recent immigrant) vote from 2016 to 2020.

So Trump's political opponents may have been eager to brand him as the white supremacist president but in terms of how minorities actually view him to this day? I would say that Scott's prediction is actually looking pretty good. And if Trump is re-elected in 2024 because minority (especially Hispanic) working class voters continue to migrate to the Republicans?

Expand full comment

I dunno. Steven Pinker and Francis Fukuyama are both incredibly well-known public intellectuals whose ideas are endlessly discussed and debated, and Scott Alexander has done pretty well for himself in the "driving discourse" department, so the main thing I take away from this post is that it's probably a good idea to care less about people with dumb opinions on Twitter.

Expand full comment
Sep 28, 2022·edited Sep 28, 2022

To the contrary I think Francis Fukuyama has gotten a lot of mileage out of publishing his thesis with an inflammatory title. How many political scientists can you name that get namedropped whenever people discuss political events?

If I was trying to develop myself as a thought leader in a field, my first move would be to make a brazen and marginally thought provoking statement to tie my name to the field as tightly as possible. Imagine if Leibniz had published "The End of Calculus'; he'd probably be better known.

Expand full comment

I think this advice you’re giving seems to be something that might make sense for an individual but be terrible if everyone adopted it.

And aren’t Steven Pinker and Fukuyama still pretty famous and successful despite idiots on Twitter dunking on them? The more hysterical and pessimistic other intellectuals are, the more of a market it opens up for people who say something different. Twitter is where particularly dumb and lazy people congregate, it’s not the entire literate public.

Just tell the truth, be honest, and careful with your predictions. You might end up like Fukuyama, who wrote the most well known political science book of the last century.

Expand full comment

Glad to see you're thinking about writing stuff that may make people say you have bad takes again. Felt like this was trying to address part of https://astralcodexten.substack.com/p/why-do-i-suck where your interesting pieces had a habit of getting annoying responses from 0.5% of your audience, which stopped you writing them (totally understandable, but subtly disengaging for the silent majority).

Expand full comment
founding

I am very sympathetic to your desire to not be at the center of another "You are still crying wolf" mess. But a lot of your guidance seems to come down to, don't make the sort of prophecies that are most likely to be accurate and useful, because fools will jump all over you for them. Which, yes, they will, but it would be a shame to throw out volumes of accurate and useful prophecy over that.

Expand full comment

I'm pretty left-leaning, and I found "You Are Still Crying Wolf" quite comforting throughout the Trump administration, so thank you.

Expand full comment

Ok, well, I'm predicting that we will keep seeing small incremental improvements in AI (perhaps enough to finally allow self-driving cars into the mainstream); the power and influence of the US will continue to decline; and the climate will keep warming pretty much in line with the median projections. Does that make me Nostradamus or Fukuyama ?

Expand full comment

What's the verdict on Enoch Powell as a forecaster?

Expand full comment

> Amongst several transported to the isles,

> One to be born with two teeth in his mouth

> They will die of famine the trees stripped,

> For them a new King issues a new edict.

Well this one is obviously about Easter Island.

Expand full comment

I think I read the same weird almanac as a kid. Was it the People's Almanac?

Expand full comment

Great essay; I think there is a lot to what you say.

Expand full comment

Thanks, Scott. I appreciate the insight both into your own thinking and into the social dynamics that constrain public thinkers. I hope you will continue to explore the areas where you are comfortable pushing boundaries. Your "I suck" post a while back led me to reflect on the issues where wizened and experienced Scott has the advantage over young and brash Scott, and I'd love to see my own predictions confirmed, even if there's no one but myself to praise or blame me.

These insights aIso help to explain the changes in Internet culture c. 2007.

Expand full comment
Sep 28, 2022·edited Sep 28, 2022

By a strange coincidence, I have just picked up a copy of Nostradamus (Henry C. Roberts's 1947 edition updated for 1994 by his grandson Robert Lawrence) at a particularly jejune university library sale. Figured I'd flip through it for some vague post-rationalist thought-jog, or at least for a few smirks at compelling hindsight interpretation.

It really is the most horrendous rubbish, though. It's not just that the quatrains are vague and malleable, some of the interpretations (don't know if this problem is peculiar to Roberts and Lawrence) simply march over clearly contrary elements.

One example of many, Century V, #55:

"Out of the country of greater Arabia,

Shall be born a strong master of Mohammedan law,

Who shall vex Spain and conquer Grenada,

And by sea shall come to the Italian [the French text specifies Ligurian] nation."

This is interpreted as 'from a country of Mohammedan law (Morocco), shall come one who is a strong master of their law - the sword: Khomeini and the Moslem revolution in Iran.'

Khomeini did not come from Arabia and never visited Spain, Italy, or Morocco, to say nothing of conquering Grenada in even the most figurative sense. I really wonder how this stuff has managed to fascinate so many people for so long.

Expand full comment

Rules "first" and "fourth" (and maybe "second") can be summarized as a more general rule: predict the alternate media-reality, which is the reality through the lens of media.

Mentioned among "losers" Pinker wrote how media makes us to see reality in a distorted way. This picture can be seen as an alternate reality were the most of consumers of media live. All you need is to map your predictions from a "real" reality to an imagined media-reality, and then to think about how it can backfire.

I believe that this more general rule is better, because it makes you more prepared to new effects which you didn't identify yet. Though it probably needs more efforts to follow: you need to think of it explicitly, not just pattern-match.

Expand full comment

Interesting, I always considered You are still Crying Wolf, to be one of your best pieces, but noticed that you shied away from writing articles like it, since. I had no idea you had to deal with so much criticism after that article. I completely get why you avoid articles like that, but I do hope you will find some middle ground where you are able to predict the future accurately, without being harassed.

Expand full comment

Saying USSR has dissolved "without strife" is kinda painting over a lot of things. Without war - sure, but there was plenty of strife. Enough, in fact, that one of major propaganda themes used by Putin to consolidate his power was "you don't really want to go back to how it was in the 90s, do you?!"

Also, shouldn't London be considered the birthplace of communism (or, alternatively, Germany)?

Expand full comment

Bari Weiss has an interesting story about Trump. It seems that Black people are Jewish people weren't allowed to play golf in the big resorts in the northeastern states. That is until Trump bought the big resorts. Trump resorts were the first big resorts open to Black people and Jewish people ... but somehow Trump gets painted the racist.

Expand full comment

With the sorry state of legal prediction markets (PredictIt shutting down, Kalshi very limited and not liquid), this guidance only applies to a very limited set of people. Unless you're writing a book or already have a national pulpit as a writer or an "intellectual," nobody cares about your predictions and you won't make any money by being accurate. The only exceptions are business-type predictions that are reflected in the market but the number of public companies or other securities on which you can make clean bets is limited.

I predicted the 2020 homicide spike and gave reasons about why it will happen (https://twitter.com/LechMazur/status/1267863723013820416) but it was barely worth the minimal effort of a Twitter post. It's same situation as with betting on Metaculus or Manifold Markets - might be fun to think about things in a systematic way and keep track of your predictions but ultimately it'll always be meaningless for almost everybody.

Expand full comment

I must say the Climate Change thing looks perfectly optimized for this. Any weather happening (hot weather obviously, but cold weather too), any strong wind, and rain, or any absence of rain - all proves it. The press would never say "the weather this year wasn't substantially different from how it has been for the last 20 years" but will yell "This July 12th was the HOTTEST JULY 12TH ON RECORD!" It can not lose, really.

Expand full comment

A relevant example that I saw/lived through. Britain privatised its railways in the 1990s, and within the next ten years there were two headline-making derailments and crashes that killed people. Obviously it looked pretty bad for the private company running the railway infrastructure (Railtrack) and for the notion of privatisation in general. But a statistician went and ran the numbers and found, even including those big fatal crashes, there was an ongoing downward curve of railway accidents from the 1960s onwards, with a visible downward discontinuity at the time of privatisation - meaning that the railways got safer at that time. But to a casual observer, it really didn't look or feel that way. *On rereading the analysis now, I see that actually fatalities were higher post-privatisation because of one of the big accidents. But all other safety indicators were improved.

Link:

https://rss.onlinelibrary.wiley.com/doi/full/10.1111/j.1740-9713.2007.00213.x

Expand full comment

"It is very hard to make predictions especially about the future."

Yogi Berra*

(*bad attribution: https://quoteinvestigator.com/2013/10/20/no-predict/)

Expand full comment

I feel the idea that “some guys in some caves got lucky once then the United States bombed some countries” fails to take history seriously.

Al-Qaeda enunciated a coherent vision for a world order to compete with liberal democracy. Many called this doctrine “Islamofascism” at the time. This was a misnomer, as fascism is an inherently racist doctrine, always seeing the state as the highest embodiment of a racial group. Al-Qaeda’s vision, however, was anti-racist. Regardless of where you came from, you were not only able to convert to Islam, you were very warmly invited to convert to Al-Qaeda’s version of Islam.

After 9/11 it was not clear Al-Qaeda, or some similar group, lacked the ability to obtain nuclear weapons. After-all, Pakistani society’s compatibility with Al-Qaeda’s vision for the world was on the same order of magnitude with Pakistani society’s compatibility with the western liberal vision. The same could be said of Indonesia, Malaysia, and myriad other smaller muslim nations.

This interpretation of Al-Qaeda and its philosophical legitimacy is well described by Paul Berman in his 2003 book “Terror and Liberalism.”

Al-Qaeda then, represented a clearly articulated challenge to the western liberal world order. The Bush Administration recognized that challenge, and set about destroying it. Significantly, the Obama administration continued Bush’s policy to completion. Drone war was stepped up. Guantanamo was not closed. Edward Snowden was not feted, he was chased to exile.

In short, the Bush and Obama administrations both agreed the political philosophy Al-Qaeda expounded was a serious threat the needed to be met and defeated. We only dismiss the seriousness of that threat today because it was so resoundingly defeated.

I don’t know how one can point to Al-Qaeda in 2001 and categorically state they were less serious a threat than the Bolsheviks in November of 1917, or the Nazis in 1923.

Indeed, one feels that if, in 1923, Hitler had been dealt with “Obama Style” then the Nazis would be remembered as a comically ineffective right-wing militia.

Anyway, pet-peeve of mine. Just because western liberalism has been reliably winning for the last 70yrs does not mean those victories are guaranteed forever and automagically.

Expand full comment
founding

One of the most valuable talents for any public figure to develop if they want to stay sane is a thick skin.

Expand full comment

Another thing that can doom predictions is predicting the upside on things that take the stairs up and the elevator down. This is most notable when predicting market recessions - If anyone was dumb enough to predict there's no recession coming up, , they'd get buried in the news every time the market has a bad day (somehow people became aware of this, which is why people pretty much always predict a recession in the near future, at least since 2008).

This also applies to AI development: A month with nothing exciting happening should probably push your timelines back by about a month or two on average, but a single exciting event would make it jump down, which means that predicting "AI is coming faster than you think, no matter how fast you think" is great media - every time people adjust timelines down you get to be all smug, but adjusting timelines up is gradual and boring and no one talks about you when that happens.

(It feels like this is Eliezer's shtick this day, which is why people are right to be annoyed at him for refusing to make any concrete timeline predictions).

Expand full comment

This sounds like precisely the secret behind the success of QAnon. It helps to align yourself with a paranoid sect desperate for recognition.

Nostradamus codes as fairly harmless and fun. It's a bit more striking when you put your takeaway as wondering, "How can I be more like QAnon and less like Steven Pinker?"

Expand full comment

Problem is society expects people who r famous shouldnt change their predictions just like politicians shouldnt change their views or promises ( watch how the opposition usually comes with a statement of politician who made 30 years ago in contrary what he says today). When i was doing superforcasting i had to change and update my predictions often based on the new data i got or change of opinion. I face no repercussions if i am wrong (except maybe lose a spot in leaderboard). If u r public person you better be transparent about your predictions and “come clean” thats better way than your suggestion of not makig any predictions about certain events

Expand full comment

It's all about framing, as usual. If the book name was not "The End Of History And The Last Man" but something like "The Decline of Ideological Postmodernism and The Resurgence of Realpolitik" (just making something up), it would be far harder to throw cheap shots at. Not as catchy though.

Expand full comment

Prediction: AI will optimise for status and reputation.

Expand full comment

I've lived long enough to be able to recall many times I was wrong about the future, so I'm averse to forecasting.

For example, as a child of the 70s, I long assumed that inflation was a major threat that could come back at any moment. But, after the early 1980s, it usually didn't. So my worries about inflation usually turned out wrong. Until covid followed by Biden, when spending like a drunken sailor was in fashion. After all, inflation worriers had been wrong so many times over the last 35 years, so what's the worst that could happen? 4% inflation?

Anyway, rather than try to predict the future, I focus on noticing the present. For example, not only did homicides shoot upwards after George Floyd's death, but so did traffic fatalities. In fact, during the Great Awokening, homicide deaths and auto accident deaths have been closely correlated, probably because encouraging traffic stops discourages both bad driving and carrying an illegal handgun, while anti-police criminal justice reforms during the Ferguson Effect and the Floyd Effect encourage speeding and packing heat. That's an interesting current reality that nobody had noticed before, as far as I can tell.

The good side of my orientation is that I'm not wrong often. The bad side is that being cautiously right a lot hasn't made me terribly popular. I'd probably be much more popular if I were wrong more about current affairs.

Expand full comment

If your aim is your own peace of mind through reduced heckling, I can see why you'd want to do this - though for *you* Scott, surely this is bolting the door after the ship has sailed.

If the aim is to effect change through becoming a thought leader... The Prophet and Caesar's Wife comes to kind.

https://astralcodexten.substack.com/p/the-prophet-and-caesars-wife

Expand full comment

I suspect Fukuyama has had much more influence on public policy than has Nostradamus.

Expand full comment

So section IV lists things that you shouldn't do. Can we list some things that you should do?

If you could send a memo back to Scott when he was writing You're Still Crying Wolf with tips based on this post, how would YSCW look different?

Expand full comment

It's received wisdom in my circles that a significant long-term effect of 9/11 and 7/7 was the removal of many rights and civil liberties in the US and UK (and I believe other democracies though I'm less familiar with other jurisdictions). This does seem quite important to me and doesn't seem as though it was particularly likely to have happened without the excuse offered by terrorist attacks. Curious to check the vibe on this.

Expand full comment

I think the basic problem is that people don't have the time or will to independently evaluate factual matters, so they have to rely on the general vibes from authorities in their ingroup (such as the media or the political leaders and activists in theor ingroup).

Those vibes are generally not very accurate, so this prevents forecasters from being accurately evaluated, in particular in the ways you describe in the post. It is perhaps exacerbated because when the outgroup wants to push back against the ingroup's vibes, they will tend to link to articles that seemingly contradict them - I bet it's regularly happened that someone said that Trump was bad, and then someone else linked your Crying Wolf post to contradict them, even though your post admitted that Trump was bad.

This seems like a quite problematic overall situation, as it actively discourages accuracy among those we'd specifically like to judge for accuracy. I think the only solid way to solve it would be to somehow allow people to quickly evaluate accuracy, which presumably means setting up some sort of organization that does it for them. Of course then one has to ensure that that organization does not get corrupted.

I wonder if you could have done better with the Crying Wolf case if you had collected a bunch of objective predictions from people who *did* believe that Donald Trump was a white supremacist, and then shown how your predictions were objectively more accurate than theirs. But that's difficult...

Expand full comment

I've done annual predictions on my blog for 20 years now, and my top takeaway is that making good predictions is really, really hard. The world is a chaotic system, and even when one gets the underlying social or political circumstances suggesting the prediction basically right (in retrospect), random small events routinely end up having enormous short-to-medium-term impact--and of course long-term trends routinely shift circumstances significantly over time. Hence prognostications that seem highly likely to occur soon and practically inevitable within the next twenty years can be delayed for ten years by "random noise" events, at which point the conditions that made the prediction inevitable have shifted sufficiently that it's now quite unlikely after all. So while I continue to make annual predictions--more as an excuse for expressing my perception of the social or political circumstances that motivate my prediction--I no longer really expect them to hold up even remotely well over time.

Expand full comment

Here is a really impressive prediction:

In 1971, Talcott Parsons made three predictions about the Soviet Union: 1) It would fall, 2) it would fall peacefully, 3) it would be dismantled from inside the communist party.

The third one is the most impressive one. Parsons was able to predict how it would fall (dismantled from inside the ruling elite), not only that it would fall, and peacefully. This, and more, in the half-forgotten book "The structure of modern societies" (the Fukuyama-type book of its day). Still very readable. Perhaps worth a review in a future book contest?

....Parsons was the dominant figure in macro-sociology in the 1950s and 60s. He went out of fashion in the late 1970s, with neo-Marxism on the rise and all that, but has made a comeback in some circles today. If you are interested in how to locate societal mechanisms that may lead to predictions, his structural-functionalist paradigm for understanding the world & predicting future developments, offers some possible intellectual tools. There you'll find the reasoning behind his 1971 predictions of the future of USSR, and much more.

Expand full comment

'Discredited pastor' could easily be MLK.

Expand full comment

I think you're reading too much into the Twitter mob/ newspaper writers. You're never going to please those people. Being a thought leader isn't about making those people proud of you, all you have to do for that is pander to their prejudices, sound fancy, offer new bs evidence supporting their prejudices ("ivermectin 92% effective"). Its low hanging fruit. I know this is going to make me sound super elitist, but being a thought leader isn't about those people. Those people don't even want to think critically about their own opinions. Being a thought leader is about having a conversation with the people who actually really want to know what is going on/ are carefully trying to figure that out (even if the truth they find turns their whole life upside down). Talk to those people, move that conversation in a useful direction, then boom- you're a "thought leader", even though 80% of Twitter thinks you can be dismissed with a gotcha and a lol. The people on some ideological crusade/ just begging for likes are not the people you want to have a conversation with. They aren't even trying to find the truth (they already have it). I'm not saying these people are evil or anything, I'm saying that engaging with them rhetorically is kind of a trap/ a poor use of time. Don't focus on packaging predictions creatively/ sounding cool/ not having the Twitter mob hate you. That will just water down your genius, imo. Just say what you think is going on now, predict what you think is going to happen tomorrow, and know that we will forgive you when you aren't 100% right all the time. We know how this shit works.

Expand full comment

The End of History is a much subtler and vastly weirder book than the cereal box summary makes it seem. Would make for a good book review.

Expand full comment

>There was not a rising Islamofascism, there was not a clash of civilizations. There were a few guys in some caves doing terrorism, they got lucky once.

That's a very american-centric point of view. For many in western Europe, the clash of civilization is a very real concern, and if islamofascism isn't a word that really survived in the discourse, far right parties break records in one european country or another nearly every year. Even at the time, terrorists got lucky once in new york, then again in London, then again in Madrid, then after a lull (in the west, at least. The rest of the world got it's share. Remember Boko Haram?) until it came back with a vengeance in the mid-2010, and became a steady flow of shootings, trucks attacks & stabbings.

Expand full comment

A bit of a missed opportunity.

People would argue that Trump "kids in cages" fit the concentration camp prediction, but that is then complicated by the fact that the Biden administration, while ending family separation, is still using the same facilities.

Similarly, it's interesting how Drone Warfare is remembered as an Obama specific thing when it only increased after 2016 and even after leaving Afghanistan, the US is still using drones in West Africa.

Defining the terms of the prediction is crucial and no one gets points for "X will probably continue."

Expand full comment

"Pasteur is just French for “pastor”, and an honest translation would have just said “the pastor will be celebrated…”"

Maybe the pun is the prophecy, that it turned out to be a name all along... But yeah, we should only give this credence with a clear prior in favour of Nostradamus's work being prophetic, and that would be a silly prior to hold.

Expand full comment

Consider adding content from this to section IV: https://www.militantfuturist.com/rules-for-good-futurism/

Expand full comment

It's important not to get different epistemological frameworks mixed up. It's decades since I had much to do with Nostradamus, but I do remember that he saw himself as the heir to a tradition of receiving visions about the future which were not necessarily comprehensible even to him, and then trying to find words to express them. He was not "predicting the future" in any sense that we would understand it.

Fukuyama, on the other hand, has largely himself to blame for his reputation, because he chose a title for his original essay that was almost wilfully misleading. (In any case he was less forecasting the future than interpreting the present and saying it could continue). I was involved in international political events at the time, and if there was one thing that was obvious, it was that "history" (in the sense of world events) had never been more active. Indeed, the years from about 1988-93 were in many ways overwhelming. But he was right about the triumph of liberal-democratic ideas (not necessarily practices) to an extent that I don't think has been acknowledged. Effectively, since the end of the Cold War, Liberalism, previously a doctrine among many, has expanded to be almost the only political ideology that you can discuss in polite company, and the overwhelmingly preferred ideology of the international Professional and Managerial Class, and bodies like the EU.

Consider, when Fukuyama was writing, there was a Socialist government in France, and across the water, a decent prospect of a left-wing Labour Party returning to power. Today, the Socialist Party has effectively disappeared as a political force and the Labour Party is just another neoliberal technocratic managerialist collection of nonentities. At the time, Liberalism, had to contend with a whole set of Marxist and non-Marxist parties of the Left, as well as traditionalist parties of the Right, based around family, community and often religion. Now, right-wing parties have been taken over by neoliberalism, and left-wing parties mostly neutered.

The result -sometimes called the Davos Party - is an international ideological tendency built around extreme social and economic liberalism, which dominates political life and discussion, dominates the international media and international organisations. Its ideas are not actually very popular, but it has managed to acquire effective ideological hegemony because of the power and money it has at its disposal. Political leaders who stray from the path either to the left (like Corbyn) or the Right (like Orban) are marked for purging.

Thus, in true Hegelian fashion, Liberalism has won the war of ideas. But most people on the planet are not Liberals, and never have been: indeed, given that Liberalism is essentially an elite doctrine, it's hard to see how they could be. And they want action, not ideas. (I've covered the deficiencies of Liberalism in facing up to contemporary crises in a number of my Substack essays, so I won't go into that now). The problem is the lack of a coherent alternative ideology: Political Islam, for example, was supported out of desperation in a number of countries (Tunisia, Algeria, Egypt) simply because all of the other alternatives, generally imported from the West, had been tried and found wanting. But at least that was an ideology. In the West we have no alternative to Liberalism at all.

Expand full comment

<< If people were ranking threats to the world order now, Islam and terrorism wouldn’t make the top twenty.

But... ISIS?

Expand full comment

Shouldn't we wait more than two or three decades though? You had hundreds of years of feudalism in which to tell people that feudalism would never end and be proven correct. And then....

Complete agree about uncharitable readings of Fukyama. There's a decent chance that by the time the next wave rolls around nobody will remember his work, or any of the rest of us.

Expand full comment

The best thing about Nostradamus is that he inspired Al Stewart's great album "Past, Present and Future".

Expand full comment

Here’s the part where I defend Obama. My least favorite President ever.

I was on active duty doing an operational mission at GTMO during his term. He did not fail to close it.

In order to “close” it, all of those detainees had to go somewhere. Having seen most of their files, I can tell you that no matter what you think about the GWOT or the legal status of the detainees these are really scary people. They were caught on the battlefield chopping peoples heads off, digging mass graves, etc.

Each one (about 140 at the time) had to be negotiated for a place to go in some country somewhere. And each time we got close to sending one to some third world South American country, they would send a delegation to interview them and realize that they were psychopathic murderers and say “nope. Keep ‘em.”

Obama could not do anything about that.

Expand full comment

Been giggling over "beclowned" for a good while, thanks.

Expand full comment

FWIW, your Trump post was convincing to me and I thought it aged real well.

Expand full comment

Another delightful exploration.

I laughed because I had just made a casual prediction in the Comments section of Common Sense regarding the World Professional Association for Transgender Health's guidelines for gender-transition surgery. "The transition movement is beginning to wake up to the tsunami of lawsuits we will be seeing within the decade. ... The legal judgements are going to be staggering. A jury is going to listen to the stories of regretful young people, permanently deformed as minors by deep pockets institutions, and they are going to dole out financial remunerations of equally life-changing magnitude."

As an obscure non-thought-leader commenting to those with whom I already agree, I predict I will not experience the repercussions Scott describes.

I so love Astral Codex Ten. In real life, I have no one to talk with who understands my way of looking at things, and my circle is highly educated and accomplished. Agreement with conclusions is not enough. One wants to encounter others who have more deeply considered ideas than one's own, from whom one can learn and grow.

Expand full comment
Sep 28, 2022·edited Sep 28, 2022

It needs to be said: someone complaining about upheaval and instability in 21st century Britain is some real first world problems type stuff. The transition from Theresa May to Boris Johnson, the Brexit negotiations, and the Northern Ireland question? Uh...not exactly the Cromwellian Interregnum, was it? The death of Queen Elizabeth II? Not quite as impactful as the death of Queen Elizabeth I, I'll wager. If you squint hard enough, I say that tweet inadvertently vindicates both Fukuyama and Pinker. Don't @me, bros!

Expand full comment

Do they have to play this game? That depends on whether the people you disapprovingly quote matter. I think they don’t. What the Twitter mob tweets about Fukuyama, or you, is irrelevant for how the arguments he or you put forth are considered by those who matter. Because those who matter also know the Twitter mob is ignorant. Otherwise they wouldn’t matter.

Expand full comment

How much is the seemingly prevalent anti-Pinkerism mostly the work of the media/twitter Misery Kingdom? Most people I know act as if life will in fact continue as normal. Is pessimism actually popular, or is it just that public speech is dominated by Career Loud Minorities?

Expand full comment

TL;DR: People intentionally misinterpret the statements of people who have legitimately earned respect to borrow their clout and gain broader reach. Don't play their game.

Expand full comment

A little associated music. Are there any songs about Fukuyama?

Your songs are leaves upon the sea....

https://www.youtube.com/watch?v=hMR6qPn1nPI

What the hell was that? I was looking for Peter Gabriel (which was a mistake, should have been Al Stewart), and youtube didn't even offer him.

Something like reggae played fast and hard, maybe. A good version.

Now have some Al Stewart, the original was much more gentle music.

https://www.youtube.com/watch?v=qa89bt0GZvQ

Expand full comment

I don't have any interest in re-litigating your Trump stuff either, but I do feel like those predictions were harder to falsify than others you make. I think you're held to an absurd standard of rigor that other commentators aren't because you sometimes reject dominant narratives. That's a broader problem that you can't really solve. If people are prosecuting heretics and you commit heresy they'll find a palatable charge besides heresy to get you on.

But often your predictions are literally "this specific event with these well-defined terms will happen by this date" while occasionally your political predictions are more of the "This politician won't have racist policies" variety. And when people say "uh, this seemed pretty racist to me" and you say "It might have disparate impact based on race, but you can't prove it was racist" now you're in semantics hell, where nobody will ever admit you're right, and *you* will never admit you're wrong.

tl;dr: I think all of the above was kind of nonsense - it's true that some types of predictions are more fraught than others, but the real issue is making predictions with clear terms and falsifiable premises.

Expand full comment

Scott, I just want to say that while I disagree with your old article about Trump and think you're wrong not to have updated your opinion on the subject...I subscribe to you largely because of it. It was deeply useful to me to read a well reasoned opinion from someone to my right politically. I was, at the time, starting to feel deeply angry and resentful at anyone (including family members) who was willing to countenance the possibility of Donald Trump as president, and your arguments (while not convincing to me) showed me some ways it was possible to disagree on the topic without total war.

I get that it's been unpleasant for you, but I hope that you do write some more things I can really disagree with.

Expand full comment

"3. Prophet"

Bravo, bravo!

Expand full comment

So some predictions impress and win admiration, and others are magnets for abuse.

Is the difference just, saying that X

* will happen = easy win (Nostradamus predicting vague happenings, Taleb and something weird)

* won’t happen = epic fail (Fukuyama and liberal democracy won’t decline, Marcus and AI won’t do this and that)

Expand full comment

Posts like this are why I got hooked on SSC back in 2015.

Expand full comment

i know you really hated 'bingoing' people instead of arguing with them but you should have set a kind of counter when you made your prediction, with space for a copy of someone's post/tweet whatever and space underneath [maybe a fake dial/guage] for you to point out/mock how far of the mark it is/will be [no, the ridiculous orange man is not remotely hitler for being an idiot again/ the machines haven't risen up because alexa was rude to you etc]

Expand full comment

It's remarkably unpleasant to be publicly sneered and jeered at, and especially so if you experienced the kid version of it in school. And lots of people who have a quirky kind of smarts have endured that. I sure did. Wouldn't surprise me at all if Scott did too. And social media now makes it possible for thousands of people, rather than dozens, to unite in hating on somebody's ideas and attempting to discredit and even sort of dismember the self that produced them. I really don't know what to do about it. Something about the way social media works seems to neutralize any impulse to disagree in a limited and respectful way. You know how when you're driving and annoyed by the car ahead you look at the slow, badly-navigating butt of the car and it's almost as though you hate the car? If you think of the driver at all, they're just "the asshole driving that thing." Well social media fosters seeing the other speakers in the same impoverished way as we see the car's driver -- but simultaneously gives us access to some of their private, deeply held beliefs -- and also allows us to attack savagely in a way that has no analog on the highway. It's a bad set-up. I'm sure it must help some to have a setting where one's thoughts are admired and respected.

Expand full comment

Slightly analogous to the asymmetry between the reaction to predicting good things and the reaction to predicting bad things that you noted, there is a similar asymmetry in how easy it is to review something with praise and with criticism, and in how much debate each is likely to inspire. In light of that, I hope I can be excused, as someone who loves this blog and agrees overall with the ideas in this post, in jumping to a harsh criticism of one of its smaller points (and yes, this is in the assessment of the "still crying wolf" reaction, as you could probably guess).

> I expected his policies on race to be about the same as any other Republican president’s (rather than, say, putting minorities in concentration camps - which people were literally saying he might do!)

This is blatantly a weakman, and beneath what I've come to expect from the intellectual who introduced me to the term "weakman". Yes, some people were literally afraid that Trump would perpetrate a genocide, a relatively few extremely anti-Trump people (half of whom probably weren't being entirely honest with others or with themselves about what they truly expected). I'm not going to look up and reread the "You Are Still Crying Wolf" essay right now, but I pretty confidently remember that it didn't come across as addressed to the extremists saying things like that, that it was aimed at everyone who was saying that Trump was unusually racist for a president in an attempt to rebut this by demonstrating that nothing Trump said conveyed racist sentiments at all (which is a stronger claim than a prediction that Trump's *policies* would be no more racist than that of any other Republican). If the goal were simply to quiet fears that Trump would aim to be the next Hitler, this could have been accomplished by sympathizing or even endorsing the view that Trump's attitudes on race (or the attitudes of his supporters which he was winking to) were concerning and then explaining why Trump couldn't possibly have the inclination or gain the power to start wiping out minorities.

Expand full comment

Isn't what you're saying just a matter of picking the rhetorical explanations of the future that you like the most?

Expand full comment

Yeah, no, everyone got Nostradamus wrong because they completely misinterpreted him. An alter ego of mine set the record straight starting here: https://specgram.com/CLXXVII.1/11.claremont.nostradamus1.html

Expand full comment

At least I can thank Nostradamus for my favorite scene in the Sopranos

Expand full comment

How Not to Be Wrong: The Power of Mathematical Thinking, Jordan Ellenberg (2014).

Expand full comment

Broadly I agree with the thrust of this post. However you need a different example than Fukuyama, because he is an example of a different problem: people straw-manning a book from secondhand ignorance of it. His book was actually far more thoughtful and interesting than either its title (which I wonder whether he even picked) or the way it is summarized in media accounts and whatever. Honestly his book doesn't much resemble that title at all.

It's similar to how pretty much everyone thinks they know what Adam Smith wrote in "Wealth of Nations". I was one of those for a long time myself until finally getting around to actually reading the book. (The resulting realization was really mortifying, in some weird way I had a desire to go back in time and apologize to Smith.)

Expand full comment

Tentatively, this essay has a self-pitying tone that's unusual here. Does it seem self-pitying to anyone else? I'm sure I'm feeling this reaction, but not sure I'm right.

Take heart, almost no one gets a big reputation of any sort for forecasting, though I grant you have better odds than most.

Expand full comment

My personal opinion is that this post focuses on the wrong metric. I realize that Scott is not necessarily endorsing making vague, catastrophic predictions but is only saying that making such predictions is 'safer' and perhaps necessary for those who want to become thought leaders. My contention is that, yes, perhaps acting in this way could be narrowly beneficial to some, but I don't believe this would be good for people in this community nor would it attract a thoughtful audience like the one this blog has.

I think one of the best features of the rationalist community is the willingness to be honest about beliefs. It is funny because many rationalists are consequentialists, and I think this is strictly speaking somewhat of an irrational behavior from that perspective. A strict consequentialist would seek to have an internal model of the truth, but would also behave in a way that best serves their goals, which could entail deception or at least lack of transparency. There are consequentialist arguments for truth telling (e.g. others will trust you less in the future if they find out you lied, so in order to maximize your influence you ought to refrain from lying when this is a possibility) but likewise there are counterarguments, etc.

I thus think the great thing about rationalists is that the commitment to truth comes *before* consequentialism. This is what makes me favor this blog as a place to gain insights over traditional media. The latter takes the opposite tack-- it is ostensibly devoted to the pursuit of the truth but is heavily consequentialist in what it chooses to report. Readers who discover this become immensely disillusioned with the news, a major cause of the political polarization we see today. My worry is that if rationalists adopt these sorts of tactics, deciding what to say based on how they will be perceived, they will lose goodwill and essentially become no better than politicians and journalists. The wonderful community that has developed in these comments sections will disperse if and when people realize they are being (however slightly) manipulated. You may attract more people, but they will have worse epistemic hygiene and will not be able to contribute to your movement in any way other than blindly repeating your talking points and perhaps voting the way you ask them to.

I think the example of Nassim Taleb is actually a good case study. I haven't read his popular books, but I have spent some time seeing how he interacts with his followers on Twitter. When it comes to pure statistical theory, everything he says seems to be fine. However, when it comes to the application of his statistical ideas to real life, he seems to be very dogmatic about where his ideas do and don't apply, and goes so far as to insult those who come to different conclusions (he openly calls Steven Pinker, Phil Tetlock, and Nate Silver frauds). Even if he is often correct, the average reader has no way to verify who is correct in any particular case. The only way to be consistent with Taleb is thus to be bullied into accepting his opinions as ground truth, as there is no way to use his high-level insights on black swans and heavy tails independently in one's everyday life in a way that one could know he would sanction.

This is why I would prefer that rationalists stick to being as clear as they think is necessary to properly convey truths that receptive people will be able to absorb. Forget about what people with negative intentions will say, and focus on clear insights that willing people can apply in their lives rather than vague ones which have to constantly be clarified by some authority.

Expand full comment

At the end of the day, Pinker and Fukuyama are still highly respected academics on their field. They're household names among even the moderately educated, and there are plenty of college courses where their books are required reading. That's basically the pinnacle of success for academics! Why does this article treat them as though they're laughing stocks? Just because some random idiots on Twitter like to make fun of a highly inaccurate caricature of their ideas? Who cares?

Also, the Twitterati's myopic focus on current events isn't as one-sided as you say. Over the past two weeks, I've actually seen a ton of tweets saying "look, Russia's loss in Ukraine and the anti-government protests in Iran prove that Fukuyama was right all along, it's actually his critics who were the short-sighted morons!" Of course, it's based in the same shallow misunderstanding of his work as all the "9/11 proves Fukuyama wrong forever" comments, but it still goes to show that the effect you're talking about works both ways.

Expand full comment

I thought your post on why Trump wasn't literally Hitler was one of your stronger ones.

Expand full comment

I don't agree with you that history suggests Marcus must be wrong within our lifetimes, let alone ever, because technology always marches on. Rather the contrary. Certainly technology in general always marchs on, but it does not do so equally in all directions, and certainly in quite a number of areas it often becomes completely static. What's the last bold advance in pencils you've seen? Or shopping carts?

In my lifetime I've seen any number of straight-line extrapolations of exponentially advancing tech turn out to be ludicrous because exponentials never last, and it turns out by the time the tech is well known and amateurs are making bold predictions about it, it's actually already reached almost its peak. I mean, in 1970 the future was going to be interplanetary, and we could take a Pan Am shuttle to the space station and visit Saturn on a honeymoon at the latest by 2000. Oops. Enthusiastic predictions of talkign sentient computers go back even further, and many in the 70s or 80s would be a bit surprised how far we remain from them 50 years later. For that matter, aren't the confident predictions of self-driving cars from a mere 10 years ago rather a cautionary tale?

History unquestionably shows that breakthroughs happen in a regular way, and that once they do, we can go back and dig up someone who predicted they wouldn't and laugh at him. Unfortunately, what history also shows is that breakthroughs don't happen very often in areas where people fully expect breakthroughs. They are usually surprises, and happen in areas where people in general *aren't* expecting breakthroughs, or even much that's interesting.

That's why if in 1988 you'd asked people -- including many experts -- "What wil be the most amazing technology shift of the early 2000s?" nobody at all would've predicted the smartphone. And many of the things they *would* have predicted -- orbital hotels and commercial LEO space tourism, say -- turn out to have not come even close to being true, and look like long-term projects or dead ends now.

That's hardly surprising from a general social point of view: pretty much by definition, a breakthrough is an advance that is forseeable to only a very few individuals -- whichever lucky or brilliant people make them -- because that's what makes it a "breakthrough," and not just some plodding 100% forseeable engineering tweak that upper management can totally schedule on a Gantt chart and set a deadline on.

The only way I think you can be confident that Marcus will be proven wrong in a timescale less than a century is if you want to assert that no breakthroughs at all are needed, that in fact aware AIs, or whatever exactly it is you predict, are just at this point a series of 100% predictable and schedulable engineering advances -- that nobody needs to come up with any unexpected clever new insight, no new methods need to be invented, it's just a question of scraping up more Internet data or adding enough nodes to your neural net. I would gather this is exactly where Marcus disagrees with you: he thinks there *does* need to be one or more additional breakthroughs, and the timescale on which they will appear (if they appear at all) is as unknown as the timescale of breakthroughs usually is.

Expand full comment

The faker knew that Russian communism would collapse in the same way as the Russian communists knew that capitalism would collapse. It happens that the faker got lucky and the Russian communists didn't (so far, at least), that's all.

Expand full comment

Your "crying wolf" post about Trump 6 years ago is what got me reading your blog ever since. It's interesting reading stuff that goes against the grain—so long as it is intelligent of course.

Expand full comment
Sep 29, 2022·edited Sep 29, 2022

He's right: I do want to be a thought leader in land policy. I am completely surprised to have my ambition included in this list. Nailed that prediction.

Expand full comment

Scott, I won't presume that you are going to read this, but I hope you do. I was going to comment, but just made my own post. Take care.

https://ageofsubjectivity.substack.com/p/futurism-sounds-kind-of-fun-but

Expand full comment

Coincidentally, Dan Luu has recently done a survey of selected futurists and their predictions. (Duckduckgo for Dan Luu Futurist Predictions.)

The one who comes out best is the world-renowned expert in things future, Bryan Caplan. (Bill Gates and the old guard at Microsoft also do "uncannily" well, according to Luu.)

Who? Bryan what? Exactly, kind of. Is the point to be right and unknown, or is it to get onto the promotional circuit? (Well, perhaps not for Nostradamus. But posterity can be a motivator for some people.)

Expand full comment
Sep 29, 2022·edited Sep 29, 2022

I think Gary Marcus is in a pretty good spot, actually. No matter how hard people try to typecast him as the "AI won't have advances" guy, his actual position is more like "deep learning models are just big lookup tables".

Here's the thing, I've been observing the Stable Diffusion enthusiast community recently, and nobody is more keenly aware that the model is just a big lookup table than them. As soon as the models get impressive enough to produce something people want, people start using them, and then discover that the way to get it is not well-formed, clear sentences in natural language, but a seemingly random set of arcane keywords that force the generator on just the right path.

If Gary's right, and I do think he's right, then each crazy advance the current paradigm produces will necessarily be of the type that makes the models more capable while not changing that particular trait of theirs. So, all we'll get is more practical uses of AI, meaning more users of AI, meaning more people aware of how the AI should be approached, and this results in Gary's position becoming the widespread common sense. (And I don't think he's famous enough, or the current community big and loud enough, to shape his name into a symbol of something else entirely, Charles Murray-style.)

Expand full comment

I would argue Nostradamus and Fukuyama are largely used in the same way by now, as in journalists who are lazy and looking for an easy way to open their article refer to one's or the other's prediction, without really being interested in their work or trying to contribute to a discussion of it. This is not to say Fukuyama and Nostradamus had the same scholarly value, but just as even those who don't like the concept still have to mention "Orientalism" when writing about the West's relation with the East, you just cannot get around it.

Expand full comment

"With the benefit of hindsight, everything about 9/11 and the War On Terror was a random blip in history with no broader implications. There was not a rising Islamofascism, there was not a clash of civilizations. There were a few guys in some caves doing terrorism, they got lucky once, the US got angry and invaded a few countries, and then everything continued as before. If people were ranking threats to the world order now, Islam and terrorism wouldn’t make the top twenty. "

The problem with this take is that the West wasn't passive in the face of the prediction. The actions that were taken as a result of the prediction changed the outcome. It might have been true otherwise.

Expand full comment

One of the things I admire most about Matt Yglesias is that he seems to be utterly impervious to bad take criticism. You (Scott) should talk to him about this and adopt more of whatever he does to insulate himself. I’d love for you to be less wary. Some of the takes you are criticized for are some of your best/most useful takes. Your critics (the mostly sneering far left) want to break you as a person, not discredit your ideas.

Expand full comment

Prediction Markets Solve This.

Expand full comment
Sep 30, 2022·edited Sep 30, 2022

fwiw I found your blog through You Are Still Crying Wolf (from Scott Adams’ blog), got totally hooked, and have been a raving fanboy ever since.

(I had already been on LessWrong though years before and was into the Sequences etc, but had kinda forgotten about it, and finding your blog was like ‘coming home’)

Expand full comment

I mean if you are trying to be popular in this way why make predictions at all?

Do the Thomas Friedman thing and use language that is just vague enough that you can play word lawyer and never seem dumb. Or better yet, do the post scholar Jordan Peterson thing and make bombastic tweets designed to rile people up and then define them in an unusual way such that under this new way they are reasonable.

Constantly switch between scholar mode for when you want to have a decent conversation, and media pundit mode for when you wanna piss of the libs that axed your early career IQ funding and or harvest money from the republicans that really love having a renowned psychiatrist actually agree

with them for once.

I think Tetlock found a negative correlation with popularity and accuracy iirc, lol. Makes sense, how can you whoo an audience with constant allusions to probabilities and hedging?

It's a bit of a coordination problem -- I think ideally all the thought influencers would just say their most crazy opinions, and then nobody can be canceled if everyone is, but in practice, everybody just hides them except in the rare edge case where they think they can make a difference in a specific domain or something. I think the Overton Window should be as wide as possible, but it seems to eb and flow due to variables I can't suss out.

The name of the game in this media world is to never specifically talk about anything ever yet somehow be extremely interesting.

Expand full comment

And then people find out many of his quatrains are dated... like calling the death of the queen in 2022 from 462 years ago. And then the matter of King's coming abdication... I have found the vast majority of his misses are his interpreters.

Expand full comment

While I hope this thinking helps you to figure out ways to accomplish the same things while suffering fewer attacks, please don't let it dissuade you from saying what you think is important.

I don't mean to be insensitive, but I strongly believe that you have a responsibility to say these things and that ultimately they make the world a better place.

Similarly, we readers have a responsibility to send as much positive reinforcement your way as possible when it seems like the rest of the world has pitchforks out for statements you didn't make. If this blog can't be a beacon of rational authenticity, it's hard to imagine one that can be.

Expand full comment

9/11 had no wide implications? Maybe not in California. Completely changed European immigration thus demographics and its future, fundamental in Europe’s right wing turn. And for Iraq and Afghanistan? Also brought Islam as one of the major religions into world wide public consciousness for the West in general. And the mere symbolism of those two towers burning forever a wake-up call about US foreign policy and invincibility.

Expand full comment
Oct 10, 2022·edited Oct 10, 2022

Have not actually read Fukuyama, so I'll just respond to one of your claims about history relative to your characterization and not relative to whatever Fukuyama said:

> With the benefit of hindsight, everything about 9/11 and the War On Terror was a random blip in history with no broader implications.

Surely, this was a bit facetious? Setting aside the costly and protracted foreign interventions, the U.S. significantly ramped up its security state after 9/11. That event obviously paved the way for the Patriot Act, enhanced airport security, the morally dubious Guantanamo Bay, and perhaps "normalized" much of the domestic spying by media and tech companies that we all now take for granted. If the American government wasn't seriously threatened by 9/11, I think you *could* argue the American way of life was. And hasn't fully recovered since.

Sure, it doesn't feel like the world ended. But it does feel like our liberal democracy is a different kind of liberal democracy now than 30 years ago (perhaps less classically liberal, perhaps even less democratic). I would be surprised if 9/11 and the changes within American society that followed were not linked together by future historians and described as a major transitional period when our self-conception as citizens and individual sense of autonomy shifted towards the more paranoid and atomized, perhaps in ways we haven't fully comprehended and whose full consequences we have not yet experienced.

Expand full comment