I recently read 'Maniac' by Benjamin Labatut, which also tracks the life of Johnny von Neumann. It is a wonderful read - very eloquently put together, giving accounts of the man from the perspectives of several people in his life.
Not only does it capture the alien-like genius of the protagonist, but also described his particular emotional and social character in a way that I have not seen written elsewhere. Highly recommended.
Any cognitive scientists here, or actual people with eidetic memory, who can comment how the brain would store everything you ever read or heard and how it can be recalled? What does it feel like to be able to recall and recite the entire text of a Tale of Two Cities?
I would imagine that such people stay off opine and commentary style forums because their instant capacity to recall authoritative sources makes wrongful and derivative discussion tedious. To some extent, with the speed of LLM search and recovery today, we might all be heading that way. Indeed, interesting HN traffic of late appears to have waned. No doubt someone will say I'm hallucinating and all's normal. No, I'm not hallucinating.
It ebbs and flows. It’s actually quite fun to watch. When an interesting article pops up, inevitably a series of echos rise to the front page in the form of subjects corresponding to the various Wikipedia links one would need to fully understand a particular article, or sometimes just tangents that come up.
I read The man From the Future a few years ago, and I highly recommend it!
John von Neumann was such a fascinating and influential character. If you are interested in the modern history of mathematics, computers, game theory and more, pick up a copy.
> Budapest around 1900 produced an absurd cluster of minds: von Neumann, Eugene Wigner, Edward Teller, Leo Szilard, Paul Erdős, Theodore von Kármán, George Pólya.
Not to mention Nikola Tesla about 20 years prior.
> When several of them turned up at Los Alamos and realized they had gone to high school together, Szilard joked that Hungary was a cover story for a Martian landing.
What if there actually are people who go to Mars in the future but regret it and the only way for them to come back to Earth is to invent a time machine and be reborn as a mathematical supergenius. That would explain this at least:
> At six, Janos (the OG hungarian name for Johnny) could divide eight-digit numbers in his head and hold a conversation in ancient Greek. He later added Latin, French, German, English, and Yiddish. The rumor was that he remembered everything he ever read. Many years later Herman Goldstine decided to test this by asking him to recite "A Tale of Two Cities." Von Neumann started on chapter one and had to be asked to stop fifteen minutes later.
It's worth keeping in mind in the 1900s the bar for novelty was much lower. Today you'd have to learn vastly more to even stand a chance of making a single contribution , let alone as many as he had done.
Big fan of VnM theorem, which this article alludes to in passing. I think it asks a really great question: "can we make good decisions under uncertainty?" And the answer turns out to be "Yes" with some caveats.
He advocated an immediate preemptive nuclear strike on the USSR: "If you say why not bomb them tomorrow, I say why not today? If you say today at five o'clock, I say why not one o'clock?"
This is why he was dismissed in so far as foreign policy was concerned, where nuance and patience takes precedence over cold, calculating game theory. Not only would it not have been feasible (Russia is huge), but it would have led to follow on effects by possibly pulling the US into potentially another world war. This is evidence you can be a genius in some areas and wrong in others.
Clearly, there is a reason Von Neumann spent so much time on this 'zero-sum' formulation of the problem, specifically in economics. Clearly he must have believed this was representative to a significant extent, especially given his interest in pragmatic problems.
It would make no sense for an intelligent person to work on an economics problem which doesn't model reality.
> Johnny remained a game theorist to the end. After a lifetime of culturally Jewish atheism, he asked to be baptized. His daughter attributed it to Pascal's Wager: a small probability of a much better afterlife, against a trivial cost.
People always leave off the second half of Pascal's Wager. Pascal's Wager is not simply that you can map out the odds vs risk and decide being a theist is more practical. But rather that if you take the odds, decide to be a theist, then start acting like one (participating in rituals, living like one, practicing what is preached, etc) then you may find God through those actions.
It's important to recall that an all-knowing God isn't necessarily very compassionate, or even rational. Some of the people I've known who knew the most have been the biggest assholes. Or insane.
What if there's a god, but he punishes humans for believing he exists, just for fun? Where's your wager now?
> What if there's a god, but he punishes humans for believing he exists, just for fun? Where's your wager now?
That's the equivalent of Russell's teapot - an imagined, unfalsifiable proposition.
But Pascal's wager was anchored to specific attested propositions: specifically, the gospel of Jesus (repent of sin and call to him, he's paid for it and will save you from judgment and eternal separation from God if you do) rooted in the historical events of his verbal claim to be the son of God, followed by his death and resurrection, witnessed by many. You may agree or disagree, but what you are affirming or denying is at least concrete and specific.
> But Pascal's wager was anchored to specific attested propositions: specifically, the gospel of Jesus
Right, but that's the problem with it. It assumes that there are only two possibilities: no god, or the god of the New Testament.
Betting that Pascal's god exists isn't the safe choice that the wager makes it out to be, because it's equally possible that some other god exists (or just some other version of the Abrahamic god, say Allah or the pre-messiah god of Israel) and your faith in Jesus is just going to piss him off.
Believe in Jesus all you want, but Pascal's Wager is a pretty terrible reason to do so.
Edit to add: Your description is also not even the only possible New Testament god. Some (myself included) would argue that that particular image of god was painted more by Paul than by Jesus.
> It assumes that there are only two possibilities: no god, or the god of the New Testament.
I don't think so. It isn't concerned with giving any description or shape to the negative case, just an invitation to consider a particular positive case.
There are positive cases for other gospels and worldviews... there's nothing wrong with invitations to engage with one or another.
As science has shown, the truth is best discerned by considering each theory individually (and continually) against evidence. Not "dark matter sounds cool, I think that's true, not MOND" but rather considering each individually for internal consistency and against observation and predictiveness.
"It does not make any difference how beautiful your guess is, it does not make any difference how smart you are, who made the guess, or what his name is — if it disagrees with experiment, it is wrong." --Richard Feynman
Sure, you can consider the issue in isolation: If you accept the axiom that there is either no god or the Pauline Christian god, then Pascal's wager makes sense (assuming that that god doesn't care about sincerity, which is certainly not my reading of the gospels, but that's a different issue).
If you want to do a fun thought experiment, you might accept that axiom.
If you genuinely believe that an eternal future of unbearable torture hinges on your choice, you might want to cover your philosophical bases a bit more thoroughly.
That may be true, but embodied human life and history provides a base set of axioms by which to evaluate further axioms -- starting with the 5 senses.
So yeah, Christianity has some axioms, but it doesn't ask people to merely accept them on faith; that's a mischaracterization. The invitation is to consider (a) how good the evidence really is for your current worldview, and then (b) how good is the evidence for Christianity's truth claims.
Pascal goes one further and suggests considering what you have to gain and lose either way -- which may have been helpful for some when the alternative was mere atheism or unbelief in Christianity, but I disagree that it's helpful in all cases. People coming from Islam have a tremendous amount to lose, including their family, community, livelihood, reputation, and the risk of being condemned eternally by Allah for committing shirk in embracing Christianity, if Islam is true.
I think a better question to ask, whatever someone's background, is "do you want to know the truth, even if it turns out you don't believe it now, and are you willing to follow it no matter what the cost?"
Are you saying the case for Scientology can't fail when considering its internal consistency and correspondence to reality?
I would think its positive case isn't that strong... it's more like an attack chain with an arbitrary payload (including money extraction) for minds with certain vulnerabilities than a coherent set of truth claims (let alone truth claims with at least some evidence in favor).
Pascal's wager was not at all anchored to any specific attestations, you are taking his wager and reinterpreting it to give it some kind of legitimacy that it never originally held. The entire point of the wager was to justify belief in the absence of sufficient evidence, by appealing to the asymmetry between infinite reward and finite cost. If your now arguing that Christianity is more credible because of historical evidence, so be it and stick to it but you've abandoned Pascal's wager and moved on to an entirely different argument.
> by appealing to the asymmetry between infinite reward and finite cost
The supposed asymmetry only holds in a framework that assumes the Christian God as the only possibility. The point is that it's possible for there to be a god that isn't Yahweh, and that god might sentence you to eternal torment for believing in Yahweh, so in fact believing in Yahweh can result in infinite cost, destroying the value of the wager.
Not at all. The asymmetry holds whenever the other possibilities don't introduce an opposing infinite penalty for belief. There could be countless other gods who are indifferent to Christian belief, reward it, or impose only finite penalties, and the asymmetry would remain intact.
The hypothetical god who punishes believers of a Christian God is a legitimate objection to Pascal's wager, but it doesn't establish the original assertion I objected to. You're confusing demonstrating that the asymmetry isn't guaranteed with demonstrating that it requires Christianity to be the only possibility. Those are two entirely different claims.
>The asymmetry holds whenever the other possibilities don't introduce an opposing infinite penalty for belief.
Or in a stronger argument. It only works when you're the only possible infinity that exists with the set of possible requirements. The moment another infinity exists that has even a single conflicting requirement then it falls apart.
Using language we can create infinite conflicting infinities which makes Pascals wager a weak argument in a generalized set.
Now, if someone vibe coded up Roko's basalisk into a real entity that was copying humans and torturing them for eternity that's a different story, at least until someone made a second incompatible basalisk.
I agree Pascal's wager is generally weak. But I think the strongest reason it's weak is because people in other worldviews, particularly Islam, have a tremendous amount to lose if they convert to Christianity but it turns out Islam is true.
Judaism has, outside some mystical currents, less of an emphasis on "well, do you really believe?" than in Christianity, and more of an emphasis on rule-following (and studying how to navigate those rules) as the basic bargain between God and man.
EDIT: My bad, I missed that Neumann asked to be baptised, so he wasn't going back to the Judaism of his ancestors through his Pascal's Wager-esque step.
True, but I believe Pascal was specifically talking about the Christian god, was he not? Ironically, Jesus clearly claims that what truly matters to god is your intentions, as in the goodness of your heart. I would argue that Paul came along later and dumbed that idea down to "do you really believe?", which is what Pascal based his wager on.
Maybe I'm just weird, but I can't actually choose what I believe. I can choose to go to church and I can choose to say I believe and I can choose to donate money to some religious institution, but I don't think I'm capable of believing something that I think is dumb.
Pascal's Wager if you read the entire book is more intricate than choosing to believe. He also argues that by practicing as if you believed, you may feel as if the choice is true and find God that way.
I see it as more of a rationalization. When you already believe something, you look for intellectual reasons to justify that belief. Pascal's Wager feel like that to me. Its main power is hardening an existing belief rather than swaying an undecided mind.
Except that 'pisteuō' means more nearly "to be faithful to" or, one could even say, to enter a type of contract (thus the Greek construct 'dia pistis', two parties relating "through faithfulness").
Anglo evangelicalism constrained its meaning to something like "intellectual assent", for its own purposes (in my opinion, it made for better sales models) some time in the nineteenth century. And as you essentially point out, intellectual assent is fairly meaningless.
If you're unfamiliar, there is a lot of historical context around Pascal's Wager -
Unless you live in a brutal theocracy, it's difficult to describe just how powerful institutionalized Christianity was at the time. Likewise the scale of religion-themed violence. A couple generations before Pascal's birth (in France), the Catholic-vs.-Protestant https://en.wikipedia.org/wiki/French_Wars_of_Religion had kill off something between 10% and 25% of the population of France.
In many ways, I'd interpret the Wager as saying "if professing quiet belief in the current local orthodoxy is relatively cheap, and those not doing so are occasionally slaughtered at scale, then prudent people will parrot orthodoxy".
That and it only really works in a system where one decision involves an infinity. Once you add a second infinity that is not aligned or is incompatible with the first it falls apart.
It says you can't ban AI and shouldn't try. There is no general solution except a worldwide ban of technology itself. What you can do is local and incremental and based on good judgement, patience, intelligence, etc.
> Then the last paragraph suggesting he would want to ban AI.
The article doesn't say that, though. Here's the last paragraph quoted in full:
>> And then he names the qualities required: "patience, flexibility, intelligence". So, here is von Neumann, the master of abstraction, with no abstraction to offer. Maybe through his bout of applying minimax reasoning to nuclear strategy, he learned what a formal model is worth when the assumptions are shaky and the stakes are civilizational.
Please point out where in here Murat claims that von Neumann would ban AI?
In fact, the preceding paragraphs (the rest of that section) show that von Neumann would not be in favor of banning AI, or any particular technology. It's an argument for intelligent and deliberate action, not bans. Where he does talk about bans, he talks about their futility.
My apologies, I was skimming initially and, from just the headings it appeared to fit that pattern I've seen everywhere lately.
I agree with you that upon closer reading, this hypothetical JVN isn't arguing for a ban, but you could certainly read
> He goes on to say that a ban would have to be worldwide to work
> What safeguard remains? Apparently only day-to-day — or perhaps year-to-year — opportunistic measures, a long sequence of small, correct decisions. [...]
and
> The only safety possible is relative, and it lies in an intelligent exercise of day-to-day judgment.
as an argument for regulation.
And it does feel like a doomer narrative.
> For progress there is no cure. Any attempt to find automatically safe channels for the present explosive variety of progress must lead to frustration.
Also, while we glorify JVN's intellect, let's not forget that all his reasoning resulted in him publicly arguing for the nuclear annihilation of a vast country of people.
[0] https://www.penguinrandomhouse.com/books/44425/turings-cathe...
Not only does it capture the alien-like genius of the protagonist, but also described his particular emotional and social character in a way that I have not seen written elsewhere. Highly recommended.
I also wrote about why I liked the book so much:
https://henrikwarne.com/2024/07/07/john-von-neumann-the-man-...
Not to mention Nikola Tesla about 20 years prior.
> When several of them turned up at Los Alamos and realized they had gone to high school together, Szilard joked that Hungary was a cover story for a Martian landing.
What if there actually are people who go to Mars in the future but regret it and the only way for them to come back to Earth is to invent a time machine and be reborn as a mathematical supergenius. That would explain this at least:
> At six, Janos (the OG hungarian name for Johnny) could divide eight-digit numbers in his head and hold a conversation in ancient Greek. He later added Latin, French, German, English, and Yiddish. The rumor was that he remembered everything he ever read. Many years later Herman Goldstine decided to test this by asking him to recite "A Tale of Two Cities." Von Neumann started on chapter one and had to be asked to stop fifteen minutes later.
There are more pedestrian explanations. https://www.theintrinsicperspective.com/p/john-von-neumann-s...
This is why he was dismissed in so far as foreign policy was concerned, where nuance and patience takes precedence over cold, calculating game theory. Not only would it not have been feasible (Russia is huge), but it would have led to follow on effects by possibly pulling the US into potentially another world war. This is evidence you can be a genius in some areas and wrong in others.
That is a strange thing to say, especially now.
Clearly, there is a reason Von Neumann spent so much time on this 'zero-sum' formulation of the problem, specifically in economics. Clearly he must have believed this was representative to a significant extent, especially given his interest in pragmatic problems.
It would make no sense for an intelligent person to work on an economics problem which doesn't model reality.
So Newton was wasting everybody’s time with the gravity malarky after all?!
The house always wins.
It's important to recall that an all-knowing God isn't necessarily very compassionate, or even rational. Some of the people I've known who knew the most have been the biggest assholes. Or insane.
What if there's a god, but he punishes humans for believing he exists, just for fun? Where's your wager now?
That's the equivalent of Russell's teapot - an imagined, unfalsifiable proposition.
But Pascal's wager was anchored to specific attested propositions: specifically, the gospel of Jesus (repent of sin and call to him, he's paid for it and will save you from judgment and eternal separation from God if you do) rooted in the historical events of his verbal claim to be the son of God, followed by his death and resurrection, witnessed by many. You may agree or disagree, but what you are affirming or denying is at least concrete and specific.
Right, but that's the problem with it. It assumes that there are only two possibilities: no god, or the god of the New Testament.
Betting that Pascal's god exists isn't the safe choice that the wager makes it out to be, because it's equally possible that some other god exists (or just some other version of the Abrahamic god, say Allah or the pre-messiah god of Israel) and your faith in Jesus is just going to piss him off.
Believe in Jesus all you want, but Pascal's Wager is a pretty terrible reason to do so.
Edit to add: Your description is also not even the only possible New Testament god. Some (myself included) would argue that that particular image of god was painted more by Paul than by Jesus.
I don't think so. It isn't concerned with giving any description or shape to the negative case, just an invitation to consider a particular positive case.
There are positive cases for other gospels and worldviews... there's nothing wrong with invitations to engage with one or another.
As science has shown, the truth is best discerned by considering each theory individually (and continually) against evidence. Not "dark matter sounds cool, I think that's true, not MOND" but rather considering each individually for internal consistency and against observation and predictiveness.
"It does not make any difference how beautiful your guess is, it does not make any difference how smart you are, who made the guess, or what his name is — if it disagrees with experiment, it is wrong." --Richard Feynman
If you want to do a fun thought experiment, you might accept that axiom.
If you genuinely believe that an eternal future of unbearable torture hinges on your choice, you might want to cover your philosophical bases a bit more thoroughly.
So yeah, Christianity has some axioms, but it doesn't ask people to merely accept them on faith; that's a mischaracterization. The invitation is to consider (a) how good the evidence really is for your current worldview, and then (b) how good is the evidence for Christianity's truth claims.
Pascal goes one further and suggests considering what you have to gain and lose either way -- which may have been helpful for some when the alternative was mere atheism or unbelief in Christianity, but I disagree that it's helpful in all cases. People coming from Islam have a tremendous amount to lose, including their family, community, livelihood, reputation, and the risk of being condemned eternally by Allah for committing shirk in embracing Christianity, if Islam is true.
I think a better question to ask, whatever someone's background, is "do you want to know the truth, even if it turns out you don't believe it now, and are you willing to follow it no matter what the cost?"
The problem is you can't do this with made up bullshit because the creators of said bullshit can create any number of paradoxes to trap you.
Reality at least has to be self consistent.
Are you saying the case for Scientology can't fail when considering its internal consistency and correspondence to reality?
I would think its positive case isn't that strong... it's more like an attack chain with an arbitrary payload (including money extraction) for minds with certain vulnerabilities than a coherent set of truth claims (let alone truth claims with at least some evidence in favor).
The supposed asymmetry only holds in a framework that assumes the Christian God as the only possibility. The point is that it's possible for there to be a god that isn't Yahweh, and that god might sentence you to eternal torment for believing in Yahweh, so in fact believing in Yahweh can result in infinite cost, destroying the value of the wager.
The hypothetical god who punishes believers of a Christian God is a legitimate objection to Pascal's wager, but it doesn't establish the original assertion I objected to. You're confusing demonstrating that the asymmetry isn't guaranteed with demonstrating that it requires Christianity to be the only possibility. Those are two entirely different claims.
Or in a stronger argument. It only works when you're the only possible infinity that exists with the set of possible requirements. The moment another infinity exists that has even a single conflicting requirement then it falls apart.
Using language we can create infinite conflicting infinities which makes Pascals wager a weak argument in a generalized set.
Now, if someone vibe coded up Roko's basalisk into a real entity that was copying humans and torturing them for eternity that's a different story, at least until someone made a second incompatible basalisk.
I may be misinterpreting this but
Lol, wut.
Almost nothing in common with a 17th century French philosopher in Christian Europe
EDIT: My bad, I missed that Neumann asked to be baptised, so he wasn't going back to the Judaism of his ancestors through his Pascal's Wager-esque step.
Maybe I'm just weird, but I can't actually choose what I believe. I can choose to go to church and I can choose to say I believe and I can choose to donate money to some religious institution, but I don't think I'm capable of believing something that I think is dumb.
Anglo evangelicalism constrained its meaning to something like "intellectual assent", for its own purposes (in my opinion, it made for better sales models) some time in the nineteenth century. And as you essentially point out, intellectual assent is fairly meaningless.
Unless you live in a brutal theocracy, it's difficult to describe just how powerful institutionalized Christianity was at the time. Likewise the scale of religion-themed violence. A couple generations before Pascal's birth (in France), the Catholic-vs.-Protestant https://en.wikipedia.org/wiki/French_Wars_of_Religion had kill off something between 10% and 25% of the population of France.
Pascal's own religious beliefs changed several times - https://en.wikipedia.org/wiki/Blaise_Pascal#Religious_conver...
The Wager was published after Pascal's death, in an edited collection of his notes - https://en.wikipedia.org/wiki/Pens%C3%A9es
In many ways, I'd interpret the Wager as saying "if professing quiet belief in the current local orthodoxy is relatively cheap, and those not doing so are occasionally slaughtered at scale, then prudent people will parrot orthodoxy".
So much content out there atm nudging very hard for AI regulation.
The article doesn't say that, though. Here's the last paragraph quoted in full:
>> And then he names the qualities required: "patience, flexibility, intelligence". So, here is von Neumann, the master of abstraction, with no abstraction to offer. Maybe through his bout of applying minimax reasoning to nuclear strategy, he learned what a formal model is worth when the assumptions are shaky and the stakes are civilizational.
Please point out where in here Murat claims that von Neumann would ban AI?
In fact, the preceding paragraphs (the rest of that section) show that von Neumann would not be in favor of banning AI, or any particular technology. It's an argument for intelligent and deliberate action, not bans. Where he does talk about bans, he talks about their futility.
I agree with you that upon closer reading, this hypothetical JVN isn't arguing for a ban, but you could certainly read
> He goes on to say that a ban would have to be worldwide to work
> What safeguard remains? Apparently only day-to-day — or perhaps year-to-year — opportunistic measures, a long sequence of small, correct decisions. [...]
and
> The only safety possible is relative, and it lies in an intelligent exercise of day-to-day judgment.
as an argument for regulation.
And it does feel like a doomer narrative.
> For progress there is no cure. Any attempt to find automatically safe channels for the present explosive variety of progress must lead to frustration.
Also, while we glorify JVN's intellect, let's not forget that all his reasoning resulted in him publicly arguing for the nuclear annihilation of a vast country of people.
It's not an invented quote, here's an Archive copy of the source:
https://archive.org/details/ETC3234/page/669/mode/2up