The rationalists were stocking up weeks before the rest of us cottoned on. Credit: Naomi Baker/Getty Images
Imagine a hyper-intelligent, god-like alien called Omega arrives on Earth. It offers you two boxes. One of them is transparent, and contains £1,000. The other is opaque, and contains either £1 million, or nothing at all. Omega has now disappeared off to its own planet, and left you with a decision. You can choose either both boxes, or just the opaque one.
It may sound obvious that you should take both: you get the extra thousand in either case. But the twist is that, before Omega filled the boxes, it predicted what you would do. And if it predicted you would take just the opaque box, it put the million pounds in there. If it predicted you’d take both, it left the opaque box empty. It knows you know this. It’s done this to 100 other people already, and it’s successfully predicted their choice 99 times. What is the rational, optimal thing to do? Do you two-box or one-box?
This thought experiment is called Newcomb’s paradox, and it has been a source of heated philosophical debate for more than 50 years. The philosopher Robert Nozick, in a long and agonised discussion of the problem, wrote in 1969:
“I have put this problem to a large number of people, both friends and students in class. To almost everyone it is perfectly clear and obvious what should be done. The difficulty is that these people seem to divide almost evenly on the problem, with large numbers thinking that the opposing half is just being silly.”
The book Rationality: From AI to Zombies mentions this problem. Its author, Eliezer Yudkowsky, acknowledges the philosophical difficulties. But he has a robust and sensible approach. The rational course of action, he says, is the course of action that gives you a million pounds, rather than the one that doesn’t.
In the early 2000s, the young Yudkowsky had something of an epiphany. He was only 20 or so, but he was already quite a well-known figure in certain subcultures on the early modern internet: the transhumanist, technophile set who talked about cryonics and uploading our brains into the cloud.
This was before social media, so these subcultures existed as mailing lists. Great long nerdy conversations went on in people’s emails. Yudkowsky set up his own, SL4; it was about how to bring about the technological Singularity, where advances in computing and AI will render human life unrecognisable. He also founded what was then known as the Singularity Institute (now the Machine Intelligence Research Institute), with the same goal.
The idea was to bring about superintelligent AI as quickly as possible, because, he thought, being superintelligent, it would be clever enough to fix all the world’s problems: “I have had it,” he wrote, as a 17-year-old in 1996.
“I have had it with crack houses, dictatorships, torture chambers, disease, old age, spinal paralysis, and world hunger. I have had it with a planetary death rate of 150,000 sentient beings per day. I have had it with this planet. I have had it with mortality.
“None of this is necessary. The time has come to stop turning away from the mugging on the corner, the beggar on the street. It is no longer necessary to look nervously away, repeating the mantra: ‘I can’t solve all the problems of the world.’ We can. We can end this.”*
The reasoning went, very roughly: human intelligence has made human life much better, over the last few centuries; if we make something much more intelligent, it will fix the problems that remain.
But some time between 2000 and 2002, Yudkowsky had the aforementioned epiphany: just because an AI is intelligent, doesn’t mean it will do what we want it to do. Instead, it will do what we tell it to do. And the difference between those two statements is huge, and potentially disastrous. AI could, he said, ultimately destroy humanity, if we are not immensely careful about how we build it: not because it “goes rogue”, or “becomes sentient”, but simply because it follows its instructions to the letter, rather than in spirit.
Yudkowsky had, though, a devil of a time convincing people of this. If it’s intelligent, it would know what we wanted it to do, his critics argued. It would know the right thing to do.
So, in 2006, he started to write a blog, to explain why they were all wrong. He had to explain why artificial intelligence wouldn’t be like human intelligence. But, he realised, to explain that, he had to explain why human intelligence wasn’t exactly like rational thought — it is full of biases and shortcuts and systematic errors.
But to explain why human thought wasn’t like rational thought, he found, he had to explain what rational thought actually was — in the decision-theory sense, of making optimal decisions with the available information. And then, he found, he had to explain all the ways in which human thought differed from rational thought: he had to list all those biases and shortcuts and systematic errors.
And to explain those, he had to explain — almost everything. At one point, he wanted to write a relatively unimportant post using an analogy of AI utility functions. But in order to do that, he felt he ought to first explain a few other concepts. To explain those, he realised, he needed to explain a few more. But to explain those…
Anyway, long story short, over the following month he ended up writing about two dozen blog posts explaining evolution by natural selection. And then he could write the not-especially-central-to-his-thesis post about the utility functions.
The whole series of blog posts, which became known as the Sequences, sprawled out to a million words. It was edited down into an ebook version, Rationality, which is a more manageable 600,000 or so — though still rather longer than War and Peace, or than all three volumes of The Lord of the Rings put together.
The Sequences are the founding text for what would, later, become known as the “rationality community”, or the “rationalists”. I’ve written about them, and their AI concerns, a few times in the past; my own first book, The Rationalist’s Guide to the Galaxy, is about them. (It’s also just 80,000 words and gets most of the key ideas across, if you are pressed for time. All good bookshops!) They are a strange bunch: clever, thoughtful, sometimes paranoid; they get accused of being a cult, with the Sequences as their holy scripture and Yudkowsky as a sort of messianic figure (although I think that is not true.)
But the community is that rare thing, on the internet: a place where people can disagree and argue calmly and in good faith. And it sprang up around Yudkowsky’s great sprawling series of blog posts, because he explicitly encouraged that norm.
As well as — or as part of — trying to lay the foundations for AI safety, it tries to establish what it means to be rational, and to make us more so. In other words, the aim is to help us be “less wrong”, as the blog the Sequences are published on is called. Following decision theorists like Judea Pearl, he defined rationality in two forms.
First is epistemic rationality, establishing true beliefs about the world — maximising the extent to which my mental model of the universe, the “map” in my head, corresponds to the universe itself, the “territory” of reality. I believe that there is a door behind me, and that black holes emit Hawking radiation: those are facts about my brain. Whether there actually is a door behind me, or whether black holes really do emit Hawking radiation, are facts about reality.
Epistemic rationality is about trying to make the beliefs line up with the facts. “This correspondence between belief and reality is commonly called ‘truth,” says Yudkowsky, “and I’m happy to call it that.”
The second is instrumental rationality, choosing the course of action that is most likely, given what you know now, to bring about the things you want to achieve. If you want to achieve Goal X — wealth, world domination, world peace, preventing climate change, Oxford United winning the Champions League — what steps are most likely to achieve that? Rationality, he says, means winning. Winning at whatever you want to win at, whether it’s saving the world or making money. But winning.
I can’t begin to go into the whorls and sprawls of its reasoning: in its efforts to ground morality and rationality on Bayes’ theorem and utilitarian calculus, Rationality reminds me of one of those huge Enlightenment-era works of philosophy, by Spinoza or Leibniz or someone, that start out with some guy sitting in an armchair and end up trying to establish a Theory of Everything. But I do want to come back to Newcomb’s problem, because I think it is key.
In 1969, Nozick concluded, after much agonising, that you should take both boxes. Imagine the far side of the opaque box is transparent, he says, and that a friend of yours can see into both. Whatever Omega has done, whether the £1 million is there or not, she would be hoping that you take both: if it’s not there, you at least gain £1,000; if it is, you get a bonus £1,000 on top of your million. It’s already happened. The money is in the box, or it isn’t.
Yudkowsky, though, disagrees. The rational thing to do, he says, assuming that you would like a million pounds, is the thing that is most likely to make you a million pounds. Almost everyone who chose to one-box got a million pounds. Almost everyone who chose to two-box didn’t. Only a very clever person could convince themselves that the sensible thing to do is the option that will almost certainly cost you £1 million.
That’s why I like the book. It may address very serious, high-concept ideas — AI, transhumanism, morality, alien gods visiting the planet and playing strange parlour games with boxes — but it gets to them all in a robust, commonsense way. The moral thing to do is, usually, the thing that kills the fewest people, or makes the largest number happy. As he writes, in a wonderful section on how it seems obvious that consciousness has an effect on the world, because philosophers write books about consciousness, “You can argue clever reasons why this is not so, but you have to be clever.”
But following these robust, commonsense views to their logical conclusion often ends up in strange places. If you accept some fairly uncontroversial premises, you can end up with AIs destroying the world, brains uploaded to the cloud, galaxy-spanning civilisations of quasi-immortal demigods.
It’s been especially relevant this year. The rationalist community was well ahead of the public, and of many scientists, in spotting the dangers of Covid-19: they are used to data, to exponential curves, to reasoning under uncertainty. They, and the wider tech community, were using masks and stocking up on essential goods, even as others were saying to worry about the flu instead, or mocking tech workers for avoiding handshakes. And now, when we should be working out the risks to ourselves and our families of going home for Christmas, a bit of extra familiarity with Bayes’ theorem and probability theory would probably not be a bad idea.
In the 12 years or so since the Sequences ended, Yudkowsky himself has withdrawn somewhat, working at MIRI or on other projects, occasionally cropping up on Facebook or Twitter. But they remain as a fascinating piece of writing, pulling together a thousand strands of thought into one place. And, whether or not you agree with their conclusions on AI or rationality, they have created a small, safe harbour for peaceable disagreement on the internet. That alone is a legacy worth having.
*Precocious 17-year-olds can be quite annoying.
Join the discussion
Join like minded readers that support our journalism by becoming a paid subscriber
To join the discussion in the comments, become a paid subscriber.
Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.
SubscribeSorry, but that last piece of information wasn’t included in the original parameters of the problem. If you knew that almost everyone taking one box got a million pounds, you would obviously take the one box, but that wasn’t the question, it isn’t a paradox, and it doesn’t tell you anything interesting. Just like tech workers calling for everyone to work from home, like they already do themselves (even when most people can’t and are therefore going to incur huge costs), isn’t very interesting other than it tells you how stupid and self-centred supposedly “smart” people can be.
Rationality seems such an attractive idea because it seems so… well rational!
The terrible consequences of rationality become manifest at step one of the process of designing an overarching goal for AI systems to achieve. Any such system must be imposed on us, thus replacing democracy with the rationality of the nerds as the basis for our society.
However the real problem with AI systems is time as the genie is most definitely out of the bottle. There are many large scale AI systems already running and we, the people, have not been consulted on a single one of them.
The AI already in use by the Big Tech companies is designed to permanently change human behaviour to make addicts out of their nearly 3 billion customers for purely monetary gain without any meaningful regulation of their activities (watch The Social Dilemma on Netflix if you are in any doubt).
None of the politicians currently in office in the western world has a clue how any of these systems work, what their parameters are or the implications for the electorate.
The assessments of the Silicon Valley experts in the documentary I mentioned above of the long term impact of Big Tech’s current AI systems ranged from civil war in the USA to the total destruction of the human race. This is without even contemplating the development of AI for military purposes.
I think that piece of information
that almost everyone who chose one opaque box got a million IS included in the original parameters as “It’s done this to 100 other people already, and it’s successfully predicted their choice 99 times.”
So the problem with making a choice is this: if you chose an opaque box there is a 99% chance of you getting a million (an alien correctly predicted your choice) but there is a 1% chance of getting nothing as it is the probability of an alien making a mistake.
It seems to me that the critical piece of information that is missing is: what is “It’s” goal? Even if we did have that piece of information, the scenario leads to a game that, I suspect, may have multiple equilibrium outcomes. (Assuming of course that we can take the predictions of game theory as the “correct way” to understand rational behavior.)
I think, even if I do not know the alien’s goal, my goal, as a player is to maximise my expected monetary reward. Let’s calculate expected win for both choices.
If I choose the opaque box, my expected win is 1mln*0.99=990,000, because 99% is the probability of an alien prediction to be correct.
If I choose two boxes, my expected win is 1,000+1mln*1%=11,000,because 1% is the probability of the alien to make a wrong prediction of my choice.
Therefore, the strategy of choosing an opaque box should be a preferred choice of a player who wants to maximise expected win and thinks rationally/mathematically.
It’s not that straightforward. How desperate are you for money? How much would you value a gift of £1000? If you are reasonably well off and have no urgent need for £1000, then rationally you would choose just the opaque box. If you are starving with hunger because you don’t have any money, then rationally you would choose both boxes. How and why did the Omega choose you and the 100 others that preceded you? Perhaps Omega can tell how wealthy you are from your possessions or perhaps it knows how much money you have in the bank. That would explain why Omega has been able to predict correctly so often. So act rationally and you will most likely get what you expect.
I really like your comment.
Indeed, after making basic calculations, one should consider how much risk of not winning any money he is ready to accept based on personal circumstances.
I want to emphasise, you are not at risk of losing here at all as not getting a guaranteed 1000 is not a real loss.
After going through this exercise it is even more difficult for me to understand why, when it comes to Covid situation, where there is a real chance of losing your(your dear ones) life/ health with comparable chances as in the given example, so many have such a high risk preference.
If everyone gets a million pounds, how much would it be worth?
I’ve just watched a video of a woman walking round the Royal Gloucester hospital, currently in Tier 3. It is almost totally empty. Minor injuries is shut, there are perhaps 20 people in total in the whole film.
The Nightingale hospitals are being dismantled, because despite many thousands of retired doctors and nurses volunteering to go back to work, barely more than a handful have been taken on and so the excess capacity which is supposedly needed can’t be used.
These nerds whose praises you sing, may have forewarned against the spread of Covid, but if they are as marvellous as all that and if purely rational thought is of any use whatsoever in understanding human behaviour why, despite spending billions on the problem are we still faced with conundrums like this?
Politicians and bureaucrats are incompetent, money does not solve all problems and viruses do not kill in accordance with liberal, egalitarian principles.
You don’t have to read anyone’s attempt to explain how the world works to comprehend those basic facts, you just need to be a swivel eyed loon, a conspiracy theorist etc.
To attempt to bring things back on topic, in the writings of the Big Yudkowsky (PBUH) we read that evidence for a hypothesis is some observation which is more likely to be seen if the hypothesis is true than if it is false. (ref: What is evidence?, LessWrong)
Is it likely that PCRCantMeltSteelBeams4678278 #KBF 😊ðŸ´ó §ó ¢ó ¥ó ®ó §ó ¿ on Twitter would be able to wander in and film patients in a COVID ward? I hope not: infection control and patient confidentiality should both prevent this. Is it likely that people who are not seriously ill will avoid hospitals at the moment because they know they’re overloaded and because they risk infection? Might there be fewer PFO cases needing their injuries tended, because the pubs are shut? I think so. So filming empty walk-in areas does not provide much evidence for the hypothesis “hospitals are not overloaded with COVID patients”, since it is not more likely that we’d see it if the hospitals weren’t overloaded than if they were.
I have seen (on Twitter, naturally) ICU doctors appealing for journalists to cover the story and some journalists saying they’d found it difficult to get filming permission (for the reasons above).
I don’t know what’s going on with the Nightingales. The standard counter to the claim that shutting them proves hospitals aren’t overloaded is that they were a publicity stunt which could never be staffed properly. If so, this is a shame: I know a reservist who was called up to build them, and would hate to think he and his comrades wasted their time.
They didn’t waste their time. $100 million was spent on building them. Hopefully they were all tore down too? Maybe another “surge” will happen and they can build them again. Another $100 million can be spent and so on.
Hmmm, maybe I have missed the real point of the Nightingales. Did Matt Hancock’s favourite interior decorator get the contract after WhatsApping him, by any chance?
As always, to waste as much public money as is humanly possible.
In this particular case it must be considered an absolutely outstanding success.
At the time the Nightingales were built it was believed they might be needed to accommodate overspill patients from other hospitals – as a very last resort. Other measures were taken and capacity was rearranged to create more ITU beds and it was found that the type of facilities needed were different from the mass bedding facilities that the Nightingales were. Some have been repurposed for other health use. Creating them wasn’t a waste – it was a contingency measure that wasn’t needed.
How blithely you dismiss the squandering of about £90 million, on a “measure that wasn’t needed”.
Do you by any chance work for the NHS, if that is not too intrusive a question?
Of course it’s an intrusive question.
How spastic as ‘we’ used to say!
What are you hiding, may I ask?
If memory serves me correctly, you are about 54, had the misfortune to be brought up in ‘wee’ little Scotland, have teenage children, and are absolutely terrified by the thought of death?
If you are Christian or similar what is the problem? If an Agnostic or Atheist how can you possibly fear death, for death means, ‘nihil’, nothing, and how can one logically fear nothing?
Please stop this online stalking. It’s very creepy. Address my opinions not your obsession with who I am.
I couldn’t give a fig for who you are, you have already told ‘all’ of us, on this platform, about a month ago and I quote:
“I have an autoimmune disorder resulting in pulmonary/cardio damage. I’m 53, two teenage kids. Life expectancy? 15 to 20 years if I’m lucky with increased disability. Maybe a better outcome with a lung transplant. If I catch Covid? Odds not good at me making it to Spring. Forgive me for having a ‘terror attack’. I want to see my kids grow up, if at all possible”.
To use your own word “creepy” stuff.
You and I have diametrically opposed views on both C-19 and death, as such I find most of your opinions to be dubious to say the least.
Enquiring whether you are employed by the sainted NHS is hardly an intrusive question in these chaotic times is it?
I would have you thought would have been proud of the fact, or has paranoia set in?
A month ago you also stated that at that date 6,500 or I in 10, of C-19 deaths were under 65. That seems an extraordinary difference compared to Mr Embery’s figure of about 400?
Perhaps you can understand how some of us mere mortals are, to say the least baffled.
It is your opinion, regardless of whether I agree or disagree with it that counts and nothing else.
Embery’s 377, or 388 in other versions, was the number dying from Covid under 59 not recorded as having an underlying condition. The number I quoted was the total number under 65 dying from Covid (including those with underlying conditions). Quite simple.
Many thanks. Thus ‘with’ or ‘of’ are the crucial words here.
If I have the misfortune to be scythed down it will be of, and if you had a similar misfortune, it would with.Simple, as you say.
So it was better to send people back to care homes to die, or not even to admit them to hospital, as happened in the early stages of the pandemic?
I sit up an pay attention when ER doctors here in Ontario (and in many other jurisdictions around the world) start publicly questioning the ‘hospitals are crashing’ narrative. Among them is the former head of the Ontario Hospital Association. So, it isn’t just random Twitter accounts playing the contrarian role. I think everyone can agree that hospital and ICU capacity is an important metric, and there are some places (exceptions, not the norm) where health care is strained. But why is there so little transparency on this crucial data? Why so little contextual information? I agree with Alison’s observation that, if this is ‘rational’, something’s wrong with the picture. We’re dealing with a narrative that *has* to be true, among its supporters.
We need to learn how to do real, open-ended science during a pandemic. To a large majority, driven, I suspect, by fear, it has seemed perfectly ‘rational’ to stick to a handful of key assumptions about Covid (its lethality, PCR as the sole diagnostic tool, our immune response, asymptomatic spread, the efficacy of interventions, etc. etc.) even as mounting evidence in all these areas calls for a more nuanced interpretation, perhaps even (gasp!) a revision of previously held views. This would have been science. This would have been rational (not that I reduce rationality to the scientific method). But the dominant pandemic culture we’ve created doesn’t allow for such things. Inside of its worldview, everything might seem rational (and moral), but only at the cost of silencing or ignoring contrary voices.
I agree with everything you wrote. The “pandemic ” has not been spread evenly throughout the health system. That is the nature of the beast.
I would just add, as a fellow Ontarioian, my understanding is that the hospitals and ICU’s have been running at nearly 100% capacity for the last several years. COVID has forced the system into a more stark form of triage that is politically unacceptable. With no possible good outcomes and with liability and accountability foremost in the minds of our leaders (medical and political) they have been forced into “irrationality “.
Coming from a manufacturing background “running at nearly 100% capacity” is good. I don’t know how much slack you’d like to see built into the system. Some is needed but not to much.
I’m also old enough to remember the days when the NHS wasn’t constantly proclaiming it needed more (of everything) and that it was going to be overwhelmed “this winter” (whichever winter it was). I’m begining to wonder if the start of this was when the government of the day inserted “professional” managers in.
In manufacturing high capacity running is good if demand is stable. The problem with health systems is building in the flexibility to cope with variations in demand – heatwaves, bad flu years, cold spells, and seemingly random variations. If you run at near 100% of average demand and that demand increases for any reason you’re scuppered. The result isn’t an unmet order but patients on trolleys, cancelled cancer ops etc.
Thanks, Alison for pointing me in the direction of this video. Woman spends 11 minutes wandering around an outpatient department that would normally be closed on a Sunday and sees no patients. She calls two passers by who challenge her about her not wearing a mask ‘stupid old women’. (Personally I think I’d challenge an unmasked agitated person wandering around a hospital talking to herself). She then goes to A&E and films (without getting their permission) 7 people in the waiting room. She didn’t see any wards, ITU or A&E cubicles – where the patients are. Dangerous, disingenuous propaganda.
I saw exactly the same kind of video from a Chester hospital. We are all being taken for a ride over this covid bull shit.
The number of Covid-related deaths in England involving individuals under the age of 60 and free from a pre-existing condition is 377. This is for the entire period of the pandemic.
Source: Paul Embery Esq.
What is all the fuss about, or are we indulging in national suicide?
If so, why so?
I think you’ll find the fuss is that 70,000 people have died. This can be validated by looking at excess deaths.
Without the current controls the death rate would be extraordinary.
I’m not convinced by the ~400 c-19 deaths in the under 60’s, but even if that is true…
I don’t see the relevance in saying that younger people are less susceptible. It is also true that the c-19 death rate of under 5s is probably less than 10.
Have to disagree with you on the statement “Without the current controls the death rate would be extraordinary”. That is the last ditch unprovable defence of our political class, who need to avert any suggestion that they have made an absolute horlix of their pandemic management.
There are countries that have fared OK, and countries that have fared less well. The correlation with lockdown is pretty weak. https://twitter.com/boriqua…
Really? On what evidence? The first few minutes of the Cummins video makes clear that the lockdowns have not made even the SLIGHTEST beneficial difference. But you were too superior to need to look at it? You can see much the same data presented on pages 17-30 of this document http://www.pseudoexpertise[dot]com/clarke-covid.pdf See pages 17-30 and especially paragraphs 22 and 43 which show the utter charlatanism of the government and its pseudo-experts.
I’ll put the precise link in next comment. Cheers.
http://www.pseudoexpertise.com/cla... See pages 17-30 and especially paragraphs 22 and 43 which show the utter charlatanism of the government and its pseudo-experts.
Absence of evidence is NOT evidence of absence.
Leaving all data aside, anyone with some sense (and contrary to what this article implies, one needn’t be a nerd to grasp this) could see the rate at which hospitals in Italy (for example) were filling, and see that if nothing was done, things would spiral out of control very quickly. Don’t forget, the death rate of an illness is not fixed, it increases once hospitals are full.
Notice how every single government on the planet is having to put restrictions in place of some kind? Widen your perspective beyond the internal prattering of the UK government.
I don’t see the relevance in saying that younger people are less susceptible.
Maybe because it’s true? Maybe because it lends some context in an environment that conflates ‘cases’ with ‘deaths’?
From the start, it was clear that the most vulnerable would be the elderly AND those with other conditions, and that’s how it has played out.
It maybe rational to panic over 70,000 dead, but the government’s response cannot be described as reasonable by any measure whatsoever.
You infer that the figure of 377-400 deaths under 60 from C-19 are bogus. The evidence?
The relevance, with sadly alludes you, is that C-19 kills predominantly the old, and if I may lapse into the vernacular, the ‘knackered’, of whom I am proudly one.
That the rest of society should be brought to the brink of ruin over such a minor disaster, is a national disgrace.
I think you’ll find the deaths of the vast majority of that 70,000 were nothing to do with Covid. My suggestion would be to spend some time with the raw data and come up with your own conclusions. I did and it is clear 2020 is a marginally bad flu year. Nothing more
This is another disingenous statistic, since you know very well what the fuss is about. Embery got a well deserved pasting on Twitter and is now whining about it here (see his article published recently). I think, given his views, he’d be happier on Parler or Gab.
I think, given his views, he’d be happier on Parler or Gab.
this says more about you than Embery. His “pasting” was largely mindless personal attacks.
Mindless personal attacks and people pointing out that diabetes, or depression, aren’t usually fatal tbf…
Did I not admonish you a week ago about your ill mannered use of the word disingenuous, yet here you are again?
What precisely is wrong with Mr Embery’s figures, besides the obvious, fact that you don’t like them?
Forgive me, but what are ‘Parker’ and ‘Gab’? Even the infuriating marxist gremlin who hides within my Ipad didn’t recognise them!
Finally please don’t tell me that as an otherwise well educated chap, you really believe all this C-19 hysteria?
Disingenous is “not candid or sincere, typically by pretending that one knows less about something than one really does”.
People quoting the stat know that they are not just quoting the stat, they are inviting people to draw a conclusion. Embery later used the “just quoting without comment” defence, which is certainly disingenous. A good response to him I saw used the example of the statement that “I’m just saying Novichok is at Porton Down, and it’s much easier to get to Salisbury from there than Moscow”. This statement is true, but invites a conclusion which isn’t, and it is disingenuous to say “I’m just saying”, as if the speaker did not know that.
Embery invites the conclusion that COVID deaths are mainly people who were probably going to die soon anyway. This is a conclusion not in evidence from the statistic, since the list of things that count as a pre-existing condition is long and includes things which don’t typically kill people, and which a large proportion of the population suffer from.
[I think it’s also probably false, as David Spiegelhalter’s stats show that getting COVID19 roughly doubles your risk of dying in the year you got it (Use of “normal” risk to improve understanding of dangers of covid-19 in the BMJ).]
That’s better, well done.
However we shall have to differ over this.
Why did he deserve his pasting?
Why did he deserve his pasting?
Why did he deserve his pasting?
There seems to be a systematic flaw in this reasoning – that people don’t gain sneaky advantages over the experimentalist by simply outthinking them. The supposition is the experimenter controls everything relevant to the hypothesis or alternatively needs to know nothing about his subject except the specific behaviour researched. What if in, say, the Milgram experiment the ‘naive’ subjects went along with the experimenters’ designs because they were paid to participate in a patently absurd pretence and suspected the whole setup was as false and absurd as the experimenters’ silly white coats. No experimental psychologist could allow such mundane possibilities – they would be seeing themselves as ordinary people and it would be career ending. Instead, Milgram decided the outcome before he began the experimental design otherwise why do it in the first place. Dariusz Dolinski and colleagues in 2015 simply repeated the original error. Scientists see only what they want to see and hear only what they want to hear. If fusion research hadn’t demonstrated this, climate mania and Covid-19 has.
Personally, I would have told the courier to push off with his stupid boxes because it was surely a con – no-one gives money randomly to strangers unless they want a lot more back.
Great reply
“Scientists see only what they want to see and hear only what they want to hear.”
If someone is doing that they are not practicing science. A scientist proposes a hypothesis that is testable, runs the tests (properly), compiles the results, compares the results to the hypothesis, and adjusts or abandons the hypothesis on that basis. There are a lot of ways for this to go wrong – poor experimental design, poor measurement, background noise that obscures the effect, etc. The scientist’s biases among those.
A scientist must have the humility to accept that the hypothesis may be wrong. Recognizing and admitting that one is wrong may be a scientist’s highest achievement. It wouldn’t be bad for politicians and everyone else as well.
Want to return to Normal? PCR vs Rapid Antigen Animation
Ivor Cummins
163K subscribers
you tube watch?v=-u6fbHGWYuc
Short and sweet, please share – and start asking questions, before they destroy your society!
Yes, Ivor Cummins is the only person who seems to talk any sense on the subject of Covid.
Cummins is a crank. He’s a diet guru who’s found that contrianism about COVID19 is a good way to increase his social media profile.
He said in September that “around 80% are already de facto immune through cross-immunity, T-cells, prior coronaviruses” (source: Youtube Viral Issue Crucial Update Sept 8th: the Science, Logic and Data Explained!) 80% immunity would mean we had herd immunity already, which would mean rates of infected per day would only go down, not up. To put it mildly, this is not what has happened.
I am not familiar with his output pre Covid, and there is certain parts of his narrative which need further explanation given the current trajectories – that, I grant you, is true.
However, a lot of sceptics/contrarians are simply highlighting research which supports an alternative narrative – there is nothing wrong with the research (the prior immunity claim is backed up, just not – it seems – manifesting as implied or as universal).
In a debate without balance and where the absence of balance is causing considerable harm, I don’t think we should be indulging shrill denouncing of those who try and introduce it, regardless of what we perceive as their motives.
After all, if you insist people only do the right things for the right reasons, you’re likely to be perpetually disappointed.
Check out @profshanecrotty on Twitter if you aren’t already. His Nov 27 thread on this talks about a paper which says that the cross-reaction to SARS-nCov-2 from recent exposure to other (common cold) coronaviruses reduces disease severity (but not how likely you are to get it), but then another immunologist points to another paper which says the opposite. So the matter doesn’t seem to be settled.
I agree that things are not false just because Cummins says them (ad hom, genetic fallacy, etc.) but they are not true either. Rather, I’d frame the problem with the cranks as being about whose ideas to spend time considering. It isn’t rational to to spend time on people who are making it up as they go along. If they are ever right, it is by coincidence: a stopped clock isn’t contributing anything to the debate about what the time is.
Thanks, I’ll check those sources out.
80% pre-existing immunity would NOT mean we had herd immunity already.
This is because the R rate amongst the remaining 20% would be 5 times higher than that estimated from the general population.
On a related point, even 100% exposure to a virus in the population doesn’t provide for herd immunity if the virus mutates sufficiently that people can catch it a second time.
On the other hand I suspect the 80% estimate is probably wrong. If infections in healthy people are sufficiently short, it is very hard to detect COVID amongst people who may have had it and spread it and recovered over a say three day period. It is more likely that many people get COVID for such a short time that they get missed by the stats altogether. Until we have an equivalent of the ONS infection survey that samples a subset of the respondents every day, we won’t be able to estimate how many cases the infection survey is missing.
Why would it? As Adam Kucharski says, R is the product of four components: duration of infectiousness (i.e. how long an infected person is contagious); opportunities for transmission (i.e. contacts); transmission probability during each opportunity; and population susceptibility. (book ref: Kucharski’s The Rules of Contagion, he’s also on Twitter as @AdamJKucharski, his thread R on Oct 26 is the best summary I think).
Which of these increase for the remaining 20%? Population susceptibility has decreased in this situation, since only 20% remain susceptible. For R to be 5 times higher, you’d need one or more of the other things to go up by a factor of 25, so which ones are you claiming would go up?
True, in the simple SIR model I’m thinking of, we don’t consider reinfection. However, currently that does seem to be rare so I’m assuming it’s not a big worry in the short term. In the long term, dwindling immunity causes waves of re-infection, see What Happens Next?COVID-19 Futures, Explained With Playable Simulations by by Marcel Salathé (epidemiologist) & Nicky Case (art/code) for things you can play with to see what happens.
According to the “PCR can’t melt steel beams!” crew, PCR positives persist for ages after infection, so if you think that, you can’t also think the ONS survey is missing many short infections (if you’re not in the crew, fair enough, but a lot of people on here seem to be).
Kudos for having read “The Rules of Contagion”
Of the 4 DOTS factors:
durationof infectiousness
opportunities for transmission (i.e. contacts);
transmission probability;
and population susceptibility.
We only really have a handle on the first, and then that has been altered over time. We now assume it is 10 days plus incubation.
Given we don’t have much of a clue what the others are, how does anyone arrive at an R rate?
Simple, you divide the number of people infected one week by the number infected another (or some similar maths) and out pops an estimated R rate for a given population at a point in time.
Lets say, that for a given instant in time R is estimated at 2. Then if it turns out that only 20% of the population is susceptible, R is still 2 in the whole population, we got that empirically. And given that the first two are fixed under the assumption of 20% susceptible population then transmission probability must be assumed to be 5 times higher. If we are only considering the susceptible population then R is 5 times higher in that population.
If R0 is 2 for a given population, then R will be 1 when 50% of the susceptible population has been infected. If the susceptible population is 20% then R hits 1 when 10% of the entire population have been infected.
As for the likelihood of contracting COVID twice that was very rare I agree. Given the subsequent mutation of the virus will it remain rare? I’ve been hearing more and more anecdotes of second infection, including amongst people I know. And in each case the test was done by chance, nothing to do with symptoms. Checking out the government figures on deaths, hospitalisation and infections, something seems rather to have changed in the last month. My money is on second infections with a mutated virus, causing less effects to those infected a second time. I suspect the mutant strain is just as deadly for first timers as the original, but that is only based on vague reports on the BBC, not hard data.
But we do have this data … or a near equivalent. The US CDC has been running serology testing across numerous locations in the USA. The result ( last i looked) is that on infections are between 5x and 10x the number of recorded cases. This makes a reasonably large difference to the infection fatality rate .
This CDC data ..not some crackpot with a twitter account and too much time on their hands.
Dismissing tons of FACTS with mere cheap INSULTS is a great way to show off your intellectual level here. Your characterisation of what has happened more recently is ridiculous in the context of the proper understanding, with false testing and the seasonal flu factor coming into action. But carry on being “sceptical” of the “crank”, and believing The Truth from The Prophet Ferguson instead!
One thing you overlook is Cummins’ point that after some months some fresh old vulnerable people are available to add to the cases and deaths, and with the false testing they get mischaracterised as “covid” whether or not. Cheers.
I’m not the one positing capital letter screeds defending the honour of my diet guru hero, now, am I? Ferguson isn’t a prophet, he’s a scientist with relevant expertise. He’s also not relevant to this question, since I understand the maths well enough myself to know that Cummins’ 80% claim is wildly wrong. Cummins has Engineers Disease, which is unfortunately highly contagious on the Internet.
There is no evidence that PCR false positives rates are higher than about 0.04%, which we can see from data over the summer when there were many tests and few positives. See ONS’s Coronavirus (COVID-19) Infection Survey pilot: England, 17 July 2020 which says “For example, in our most recent six weeks of data, 50 of the 112,776 total samples tested positive. Even if all these positives were false, specificity would still be 99.96%” (and so the FPR 100% – 99.96% = 0.04%).
I think the amount of the guaranteed win needs to be considered in light of the person’s existing resources / expectations is important i.e. would I bet £1,000 on a 99% chance to win £1,000,000? Yes! Would I bet £10,000 to do so? Maybe. Would I bet £100,000? Definitely not! A £1,000 is very nice, but not life changing. £10,000 would pay of most of my car loan, but is not life changing. £100,000 would significantly change things for the better. Jeff Bezos, though, could take the last bet because a loss woule be utterly trivial. So, yes, I would go for the million but only because I wouldn’t bitterly resent losing a £1,000.
There is nothing rational about this belief system. Humans are animals and part of nature. We are part of the natural world. We behave like animals because we are animals. This is just a fact of life. The belief that we can transform is a utopian belief. It isn’t rational. That there will be no more disease, no more fighting, no more crime, etc. The technologists who pursue this utopia are proof of their own short comings. Seeking power, wealth, control, and domination over others. This is at the expense of the well being of the rest of society not the benefit. Society improves, mental health improves, poverty issues improve with access to resources and strong local communities. People working together to improve their local environments encourages social cohesion and positive results leads to more cooperation and less fighting and antisocial behaviors. This is what we built the positive aspects of our civilization on. This is what allowed us to pursue useful technological advancements. We seem to be in danger of losing that in our present times. The utopian beliefs is what drove the world war conflicts and pitted utopian belief systems against each other. Modern technology led to the largest human slaughters in human history. Advances in biowarfare and the ability to use viruses to manipulate the human gene may very well lead to a worse catastrophe than the world wars. Especially as the will to power and greed demonstrated by our so called “elites” doesn’t give a person a sense of confidence.
Wouldn’t all those boxes of cash mess up the M0 money supply? The rational decision might therefore be to convert all existing investments to inflation linked derivatives, and probably to buy canned produce and a gun… because £1 million isn’t going to worth much soon!
Secondary effects matter, are much harder to identify and manage, and can have a much greater long-term impact though with no control it is difficult to measure.
The crux of Newcomb’s paradox, for me, is the Nature of Omega’s prediction (and quite possibly reality). In the version of it I first read, Omega only predicted what you would choose with 90% accuracy. So — 90% of the people who picked the opaque box get a million pounds, but 10% get nothing. And 90% of the people who picked both only get £1000, but 10% get the £1,000,000 as well.
So what can you make of the success rate of Omega? One possibility is that Omega is a supernatural being (or in close communication with one) who is performing a miracle, i.e. something outside of the natural world and as such not bound by its laws. If this is the case than Newcomb’s paradox becomes a test of faith. The believers go home with a million pounds. The fervant unbelievers, unwilling to concede that miracles exist, even while repeatedly witnessing one, go home with a thousand.
But at this point, I have found something I want more than a million pounds. I think that I have been offered an unprecedented opportunity to test whether miracles exist or possibly how real randomness is. (to be continued as Unherd automoderates long replies as spam.)
Part 2 — we will see if part 1 gets called spam because of word counts —
Therefore I get a large number of like-minded people who are more interested in metaphysical reality hacking than money– and a real random number generator. Use atmospheric noise, or quantum decay in a an unstable radioactive isotope. And then pick ‘just opaque’ or ‘both’ depending on what the random number is, 50/50 split. If Omega keeps making correct predictions, then we have solid circumstantial evidence that Omega is performing a miracle. If Omega’s accuracy immediately plummets to ‘no better than chance’ then we have an entity who is operating within the natural world.
(Or, God doesn’t appreciate smart-arses like me who decide to test God, but then there is always that. 🙂 )
Well, I don’t have a degree in science, but I instinctively know when scientists are talking bullshit, so all those clever people who thought they were ahead of the game in “spotting the dangers of covid-19” were just plain wrong. There was no danger, there was no pandemic, and there still isn’t. I figured that out all by myself as far back as last March.
Yes, a lot of us did and do. Unfortunately, the scientific system of enquiry is no longer reliable in ways it once may have been because:
1. it sold its soul to vested agendas, Governments and corporations long ago.
2. modern science is a tool, and often a propaganda tool for vested agendas.
3. modern science is rarely if ever independent and therefore highly subjective.
4. modern science can only know what it can measure, and indeed, that is the foundation of the scientific system of enquiry. The problem today is that scientists are, by necessity, highly selective about what they choose to measure or indeed, are allowed to measure. If you can only know what you can measure then it becomes even more critical to have the independence to freely choose what it is you will measure.
5. modern science is a profit and power-driven industry and has lost any objectivity and integrity it may once have had. This does not mean individual scientists lack objectivity and integrity, but that the system, and systems drive behaviour, limits their ability to exercise such qualities.
Another problem is that if you educate a fool, you only end up with a well-educated fool. Actual wisdom comes through the doing of things, ideally in the companionship of other experienced practitioners of your art. If you need ‘wise craftsmen’ in your field, we truly haven’t come up with a better way of producing them than to apprentice the learners to the already wise. This is a slow process. And the success-rate is only so-so. It’s just the best thing we have come up with yet. Education, defined as ‘one-to-many knowledge dispersion via lectures and books’ cannot produce wise practitioners of any field. Of course, if an already generally-wise person gets educated, they may continue in wisdom. Or not. Corporate and University pressure to be ‘clever but not wise’ may ruin them as well. We can mass produce scientists-with-a-degree but not scientists-with-wisdom.
Instrumental rationality means “Winning at whatever you want to win at, whether it’s saving the world or making money”
Does this not lead to the most horrendous exploitation of the weak by the strong?
Utility functions have to be Turing-computable. Do we regard the class of computable functions to cover the subtlety of life and include the stuff we don’t know we don’t know? Plus there has to be a stochastic part to the calculus of utility.
This whole field has the prospect of restricting the richness of life and setting directions that would make the worst of dictators look relatively benign.
It’s not just that the AI people have no grounding in ethics, they have no concept of connection with their creator.
I thought Christians _wanted_ to win at saving the world, certainly the evangelicals seem to.
That is a big misunderstanding.
Sacrifice of the will is what distinguishes the Lord Jesus Christ as the head of a new way of living and being. The Lord’s Prayer is not Turing-computable.
This article appears to reflect the irrational and incoherent nature of what is called rationality. Humans are far more than their rational capacities. It is because we have forgotten that we see science and its increasingly ‘cult-like’ creation, allopathic medicine, lost in realms of dangerous delusion. Rationality without intuition is mere mechanics.
Indeed, the greatest scientists of the past and their discoveries rested in an initial intuition, sometimes in dreams, followed up by rational practices to prove,within the understanding of the times, their insights.
It is because modern science is trapped in a fantasy of materialist reductionism and mechanics that the insane Covid hysteria has come to pass. Government reactions are sourced in the results of modelling systems, which is the poorest form of science as some in the know have said, and yet, are seen as sources of the greatest wisdom and knowledge when they are not. Never more than in modelling systems do we see the results of garbage in- garbage out.
For those who manage to sift through the stupidity and hysteria some simple facts remain which indicate Covid is not what it is claimed to be. These include:
99% of people are at no threat from Covid. The risk group is clearly identified as very sick, with 2-3 co-morbidities, very old and generally in care. For the rest, even the healthy aged, there is little risk.
Of those who test positive, most have no symptoms or symptoms so minor they do not count. In other words, they are not sick.
Of those who do get sick most recover.
The PCR test is useless at diagnosing disease and what it calls Covid is often not Covid but a false positive. This it has been said is the case for 97% of those tested.
Beyond the spewing forth of data, lots of that and very little real intelligence, we also know most Covid deaths are not subject to proof via autopsy, if indeed that were possible and are ‘guess’ deaths where people are more likely to have died With Covid and not Of Covid.
But what about the people in hospital? Well, in many countries, probably most, even a lot of the Third World, hospitals are not full. And for those in hospital, unless we know what their previous health was; whether they had the Flu vaccine which predisposes to respiratory disease; whether or not they had been on various courses of antibiotics in the prior year which compromises immune function; what medications they were taking; what vaccines they had received and what treatment they got in hospital, it is impossible to know what part Covid may have played in their condition or any later death.
As an example for Australia where, some 909 have died, in that same group of very sick and very old, as Covid deaths and less than 29,000 have registered as Covid cases, out of a population of 25 million, and barely a couple of dozen ended up in hospital and ICU units remained empty – all this over a year of Covid, compared to the 2017 bad Flu season, with a smaller population, when, in half the time, over 2-5 months, more than 300,000 Australians, of all ages, sought medical help and some 1,100 died, of all ages.
Is it surprising that many conclude Covid, whatever it might be, is no more than a Flu season in risk and in most places, not even a bad one?
The likely reason that comparatively few people in Australia died from covid is that rather extraordinary measures have been taken to stop its spread. The internet tells me that there were about 200 influenza cases in Australia this year, and you cite 300 000 in 2017. So the anti-covid measures taken in Australia reduced the number of flu cases by a factor of roughly 1/1000. If we presume that the same anti-covid measures reduced the number of covid cases and covid deaths by about the same factor, we can deduce that the number of covid deaths without anti-covid measures would have been 900 000. Presumably if the same measure had been taken in 2017, the number of deaths would have been lower too.
I’m not saying that this is correct, but comparing a flu season with no serious public health measures to a covid season with public health measures and not taking the public health measures into account is even less sound reasoning than my argument.
I’d add to that luck. The Australians had an extra 6 months to prepare. Also climate does seem to have something to do with it.
Given their current situation, what the Australians are doing does seem fairly sensible.
I suspect your 900000 estimate might be a bit high mind, given that would have put Australia with a death rate 15 times higher than places like San Marino, which were overrun before they knew what was happening.
900,000 COVID deaths as a hypothetical for Australia is far too high. That’s about 3.6% of Australia’s population of ~25 million.
COVID IFR for a country with Australia’s general profile – 1st-world age pyramid, but also a 1st-world health system – is certainly not higher than 1% and may be as low as ~0.25%.
Looking at actual experience of countries most impacted by COVID – the Americas and Western European – these are largely converging on reported COVID deaths in a broad range of 500 to 2,000 reported COVID deaths per 1 million population. That’s with fairly wide differences in policy responses between countries (and between U.S. states), plus differences, demographics (age and co-morbidities), health system quality/capacity, climate, and timing of COVID arrival (with arrival of meaningful numbers of COVID infected people being unknown/unidentified in some cases).
Look at that experience, and it’s difficult to arrive at upper bound estimates of more than ~50,000 reported COVID deaths among Australia’s population of ~25 million people. That number could conceivably go higher – though not to anything like 900,000 – if we assumed a few years of COVID seasons without a fairly effective vaccine. It looks like such a scenario won’t be reality, however. And, in such a hypothetical reality of years-long delay to effective COVID vaccines, it’s also worth wondering whether the Australia/New Zealand policy of largely keeping COVID out of the country via border controls, with suppression to the extent that it does not get in past border controls, would be a viable or effective strategy.
The actual numbers for other country’s reported COVID deaths provide a far better guide for a counterfactual scenario than hypothesizing based on flu case count comparisons.
The conundrum of the two boxes assumes that every person coming to the two boxes to make a choice is similarly situated, and differs only in their level of rationality. But the situation of a hungry, homeless person may result in a different choice than the situation of a billionaire, based on the marginal utility to each of them about what might be in the boxes.
The homeless person might be entirely rational by choosing what to them appears to be the sure thing. And the billionaire might be entirely rational by tossing a coin, since the outcome would make no difference regardless of what was in the chosen box or boxes.
So, rationality in choice varies with the individual’s situation as well as their estimation or guess as to what is/might be in each box, and the nature of the game that Omega might be playing with them. That is why, unlike Rubik’s cube, there is no single correct solution that applies to everyone in every possible situation.
Huh? Sorry if I’m just being dim, but this article seems to be rambling nonsense.
I don’t know you so I don’t know if you’re dim or not but you’re right about this.
What utter nonsense. For example the link referenced about the ‘rationalist community’ being ‘ahead of the curve’ on Covid 19 (https://slatestarcodex.com/… is just another long rambling article saying essentially ‘I’m right and everyone else is wrong’ but doesn’t actually show any evidence. For example on masks it says:
“If you really want to understand what happened, don’t read any studies about face masks or pandemics. Read Smith & Pell (2003), Parachute Use To Prevent Death And Major Trauma Related To Gravitational Challenge: Systematic Review Of Randomized Controlled Trials. It’s an article in the British Journal Of Medicine pointing out that there have never been any good studies proving that parachutes are helpful when jumping out of a plane, so they fail to meet the normal standards of evidence-based medicine. “
This is such a ridiculous and false premise it’s amazing it ever got published. I think this Unherd article may be related to the fact the author wrote a book and is trying to shoehorn it into the current obsession.
I disagree with much of what he says but Nassim Nicholas Taleb coined a term, Intellectual Yet Idiot (IYI), that suits these people
Stocking up on goods wasn’t necessary, was it? Other than a few days when bog roll mysteriously vanished, supermarkets did an amazing job of keeping up during huge swings in demand. Really impressive, actually.
As for avoiding handshakes, wearing masks etc – the time series are all public and show no correlation with any of these things, not even lockdowns. If this is meant to be a success story for rationalism, it doesn’t look very good. Rationalists of all people should decide things based on genuine evidence. On those grounds they got all three wrong.
Very useful. The distinction between rational and reasonable thinking is very important. Thinking that is rational but not reasonable is very dangerous indeed.
For example survivalists are rational when they build their bunkers, but building bunkers in not a reasonable response to the level of risk.
There is a chapter on this subject in my book ‘Good Thinking and Bad’.
I consider it both reasonable and rational to ignore the “important” distinction you imagine to exist there. Cheers anyway!
How can I get a copy please?
I think you’ve made a mistake there.
The reward for building bunkers might not outweigh cost for you, but for someone else the cost will be lower. Since the level of risk is an estimate, and is in fact unknown to anyone on the planet, and given the reward of the vast majority of the human race being your progeny is pretty substantial, that survivalist might be very reasonable indeed.
Great article, particularly because of the Oxford United reference.
Essentially what Yudkowsky is saying had been said:-
It takes something more than intelligence to act intelligently.”
(Fyodor Dostoyevsky, Crime and Punishment )
It doesn’t surprise me that it was ‘tec-nerds’ who saw the dangers first. They are the one scanning the world exploring new horizons and datum.
It is really puzzling to me how unreliable the mainstream media have become – even when there is no political axe to grind. I knew coronavirus was going to be bad because I asked my friend who has family in Hong Kong to let me know what people were saying and experiencing there. Thinking back on it – I just assumed – correctly – that the the media would not be accurately reporting on the issue. That reaction of mine is a pretty damning reflection on the media. They have continued to not report clearly or honestly (mostly through omission) about Covid ever since. Ordinary people shouldn’t have to understand the technical issues – the role of media is to do that for them and explain it to us. I know media have very serious funding problems – but that doesn’t apply to the CBC (I live in Canada) – but they are no better and sometime worse.
For an interesting example of why it’s sometimes a bad idea to rely on common sense, look up The Monty Hall Problem.
Perhaps those Tech Nerds should peruse these Government figures. https://principia-scientifi…