Liberal democracy has historically flourished in periods of economic growth associate with trade/production. I think there's sort of a continuum from trader to king, based on whether you apply your skills more to bargaining or dominance. It seems natural that the latter selects for tit-for-tat, whereas the former selects for cooperation. Under normal conditions, traders rely on the dominance hierarchy to establish peace and make cooperation viable, but when there's a disequillibrium where trade is massively expanding (e.g. the industrial revolution), the trading population can grow more powerful than the dominance population, until circumstances have equillibrated. (In economics, this would be called "opportunity cost" - the effort you spend on developing weapons for dominance could instead have been spent on trade, which in a dreamtime disequilibrium is going to give you more value.)
Market conditions are not intrinsically positive-sum, and geopolitics are not intrinsically zero-sum. Undeveloped resources which requires lots of flexible human workers to collaborative develop change conditions to be positive-sum (for human workers). But eventually you reach an equillibrium where they are optimally developed and you've got excess labor, at which point they become zero-sum. This is especially true if you go out of your way to speed up the equillibration, e.g. with certain kinds of AI and free trade policies.
I mean. Some people are definitely right this second trying to conquer land, and small scale exploitation that kinda looks like slavery definitely exists, doesn't it?
Applied to the case of memetic war: if there's some subject that yields lots of valuable insights when studied, then truthseeking is selected for simply because it's a waste of time to fight.
The more you reach equillibrium (by finding the key information, distilling them into great insights, broadcasting them to people who can learn from them, and integrating them into applications), the less value you can get from pure cooperation.
This won't immediately cause defection, it will "merely" cause a collapse in that which produces collaboration, which may just lead to a collapse in the participation. But if participation *remains* high despite lack of selection for truth-seeking (e.g. due to external funding or entertainment value or community inertia), then the participants can be exploited via deception, leading to a growth in various strategies, and then as experiments show, this tends to lead to tit-for-tat.
Nope. Per the mandate of heaven, prestige and dominance are closely linked. The critically important piece is an expansion in the possibilities of excellence so there's a disequillibrium where you gain more from developing prestige while keeping dominance low than you gain from developing dominance. So more like prestige vacuum.
I think you're on the right lines when you ask who occupies the position where each strategy makes sense. If instead of traders and kings we think about the common people of Rome at that time, we get an answer to Scott's question.
Think of Rome as a very individualistic society (much like today) where people had a high degree of security from outside (no foreign enemies) and focused on their own lives - e.g. sports, romance, wealth. These things tend to be zero-sum and encourage tit-for-tat, since it's about fighting for positional goods. So we have a society where everyone is playing tit-for-tat all the time, and probably getting sick of it (because let's face it, it's brutal).
Then Christianity can be explained simply by the fact that under these conditions it's the funniest thing in the world to be nice instead. Everyone takes their petty zero-sum competitions so seriously, that not taking them seriously is like pissing over all them.
I hope this explanation is right and that our position is similar to the Rome of that time. This is normally taken pessimistically (probably because of the stereotypes about the fall of Rome leading to the dark ages), but it would be pretty great if something that funny happened again.
As a "sell everything to give alms" Christian, I feel it's important to make a distinction between the culture that won the memetic war and the Actual Christianity you are describing. I frequently use a similar definition of Christianity, but also say there are perhaps 100,000 Christians in America at the moment. I sometimes refer to the mass movement as Modern American Folk Christianity.
While I think your terms are pretty harsh, I do think you are on the right path. The version of Christianity (COOPERATE-BOT) that Scott is using is 90% strawman. There are simply practical limits to how much cooperation any nominal Christian group can do, and TIT-FOR-TAT is far from unknown among nominally Christian groups, as well.
To be fair, my threshold is close to "has read the gospels and strives to live by it", a standard so loose it includes Gandhi, it's just that neither Modern American Folk Christianity nor Moralistic Therapeutic Deism are at all open to "Jesus said you should do X" as a viable method of socially approved persuasion. I've visited dozens of churches, I feel my estimates for population rates are at least reasonable, if not as accurate as I like.
(Edit for fun fact: did you know Gandhi read the Sermon on the Mount twice a day for decades?)
I don't know much about Tolstoy, but from his Wikipedia, yes. In terms of forerunning, I mean, both him and I are a couple millennia late to the party, but yes, there is a cross generational convergence on the ideals here. I think there is more convergence than inherited tradition, by a large margin? Also, the more I read this wiki, post-Confession Tolstoy is unbelievably based?
I think a more "narrow brand" representation might be someone like Dietrich Bonhoeffer, although that is somewhat complicated by his involvement in a plot to assassinate Hitler, which is the most absurd corner case of "well if you really love your enemies" that it's sort of wild he actually faced the hypothetical irl.
Re: Dietrich Bonhoeffer. Kind of crazy that I'd just heard about him from my coworker today, and here I am reading his name on an ACT/SSC comment. I'm attempting to return to the (Catholic) Christian fold, so would you be able to expand on your thoughts regarding Bonhoeffer's seeming contradiction?
I don't have much to say, aside from, I've been spared similar choices. His books are very good, though, and are quite readable. Engrossing, really. Strong recommendation for The Cost of Discipleship.
Could you please explain what is going on in Christian churches in America so that congregants aren't meeting this low bar? To someone who isn't American and has never been to a mainline American church or an Evangelical church.
I don't know what causes it, I only observe it. They will say, you gotta be all in, trust Jesus 100%. Then you say, Jesus said, do radical good thing X, does anyone do that? And they say, no no that's too impractical and weird. It's not a skill issue, it's a preference. They don't want to. If they wanted to, and failed, that's classically in-bounds, but this is vanishingly rare. Pretty bizarre.
Replying to you here as well: I've heard it well put that Jesus's hardcore teachings may be seen as koan-like paradoxes. Meaning, they are practically nigh-impossible, but it is the living out of the path and the divine goal/target that is the way. It's funny how people think the OT is 'harsher' and the NT is 'kinder'...Jesus asks us to aim for almost insurmountable heights.
That's not *quite* the issue. You have no idea how many times I've met with people after church services end and said something like, "I really liked this bit of the service, when they said [quote]. Is that the type of thing people do here?" And then they're confused, ask what I mean, and I say, that things they said, it's a beautiful expression of gospel principles, do people do it? And they say no. Then I ask why not, and they have all these elaborate reasons why not even a single person does the thing they talk about or sing hymns about. I often have a hard time understanding what they get out of church.
I've been using the term Christian Nationalism. That describes the top end of malefaction, and there are degrees of shading for individuals depending on where they are in life and their state of mind.
I think that term mostly focuses on a political angle, where there *is* a philosophy and culture outside of politics I'm trying to point at. Like, volunteering for a VBS isn't anything to do with nationalism. I'm also not trying to speak unkindly about anyone or isolate merits on one end and demerits on the other. I'm just pointing out that (as someone who has volunteered for a VBS) it's a foreign cultural exchange in some ways.
I understand that, but the politics is willfully mixed into it. I wish the culturally dominant strain of Christianity in America wasn't a band of nationalist hobgoblins pushing women away from the pews. But, saying reality is what it isn't has not been that great a strategic choice so far. At best, these things are in tension and one side is growing to be the more popular one.
Maybe it wouldn't be the case if more people did foreign culture exchanges. Seeing becomes believing.
As I understand it, from the Bible, the idea that you can earn the title Christian is itself not Christian. Faith itself is a gift, that no one may boast, and nobody deserves grace, by definition. Nobody is good except one, that is God. If we believe Jesus is good how can we claim to follow Him when we are not? If we believe we are good we reject His teaching.
I would be careful about saying who deserves and doesn’t deserve the title Christian.
This is true inasmuch as I should not have used the word "earned", and I apologize for that.
However, Jesus does warn about false Christians several times, such as in Matthew 7:15-27, in which Christ states that the ones who does what he says are the ones in heaven. He elaborates on this in Matthew 25:31-46, stating that the ones who will be saved are the ones who fed the hungry, gave drink to the thirsty, take in strangers, clothe the naked, and visit those sick or imprisoned.
So, considering that Jesus warned us about false Christians, stated that you will know them by their fruits, and then gave a list of the sorts of things you can expect to see real Christians doing, I think that calling someone who never does those things and is, in fact, viscerally opposed to doing them, is probably not a Christian. Back to the subject at hand- if someone who calls themself a "Christian Nationalist" is harshly opposed to taking in strangers, it's a hard sell for me that they are a Christian.
Just to be clear: I'm not saying that they are going to be damned, necessarily. I certainly don't know who will be damned or not and, in my opinion, stating that someone is going to Hell is putting human limits on God's mercy, which seems to me very unwise.
What I am saying is that I disagree with the category name of "Christian nationalist" to describe people who are hardcore opponents of immigration, which is usually what it refers to. If instead we are talking about people who wish their nation to be more in line with what Jesus literally told us to do, then I don't object to the terminology.
I suppose it might describe myself, although the term "Christian Globalist" might be more appropriate, though it strikes me as rather redundant. To be truly Christian is to be universal, embracing all humanity without the confines of national borders. Thus, the most fitting label, if labels should be worn at all, would simply be "Christian."
Thank you for the thoughtful replies. One of the joys of Christianity is discussing / wrestling with / resolving the tension in various scriptures during Bible studies and with friends. I've typed and erased and typed erased. A comment section just doesn't work for me. I apologize for starting a discussion and not following through well. God bless you, brother!
I'm curious, from the "no borders/global" perspective how do you interpret John 14:2? “In My Father’s house are many dwelling places; if it were not so, I would have told you; for I go to prepare a place for you." - If the Father's house has many different dwelling places (and a place specifically prepared by Jesus for His followers) does that not imply that there can be virtue in boundaries?
What if someone does all of the things in Matthew 25:31-46, but only for strangers of a certain type (e.g. only Americans, or only Christians)? After all, Matthew 25:31-46 never said you need to let an infinite number of Muslims into your country so that they impose Sharia law and drive Christianity to extinction. If that were the likely outcome of immigration (and let's leave aside for the moment the question of whether it is), are Christians still required to support immigration?
I mean, Christians didn't take the "take in the stranger" far enough to be like "hey, centurions, welcome to our secret underground church", so I think the answer is in the past. Alfred wasn't very welcoming to Guthrum- for a while, anyway.
But, you know, The whole point of having deontological ethical injunctions is that arguments for violating them are often much less trustworthy than they look.
An obvious game-theory thought is that you are gifting all other Christians with a *reputation* for being COOPERATE-BOT, and this makes other bots with mixed strategies interact with them more favorably in general.
This is one explanation for how early Christians could have succeeded, by creating a reputation for being COOPERATE BOT while most of them are actually TFT.
They'll personally want to exploit you, but they'll also want to prevent their own enemies from exploiting you too much - so that you'll still be there to be exploited by them and their allies in the future. Norman Borlaug's dwarf wheat doesn't have a military policy above the microbial level, but it's cultivated by people who do.
Very interesting idea, so it is basically becoming such a useful and unique tool to be able to be integrated into a larger policy. Which for sure can work, but that simply outsources the original game theoretic problem onto the larger entity, practically making the problem disappear for you, but not for the larger entity.
Suppose two wolves and a sheep are voting on what to have for dinner. The sheep says "I propose that tonight we eat whichever of the two wolves is stronger, and then have mutton tomorrow."
The weaker wolf realizes, if this proposal doesn't pass, it certainly won't be the stronger wolf being eaten tomorrow, and so seconds the motion.
After resulting excitement is settled, sheep says to the surviving wolf "you're badly injured, and must be tired from doing most of the work in that fight, so I insist you accept the honor of eating a larger portion of the carcass."
Once that wolf is asleep with a full belly, sheep headbutts and tramples them to death, then trots off to eat grass.
in a war of all against all, with no credible alliances or limits of adjacency, default outcome is everybody ganging up on whoever looks the most threatening - that is, whoever they'd expect to face the most harm, or least benefit, from ending up stuck alone with.
Iterate that enough and whoever looks least threatening dies last. Thus, Mr. Rogers in a bloodstained sweater... or Putin, who was apparently the most blandly inoffensive man in the entire KGB, ending up on top by default.
Supposedly this is how Claudius became emperor of Rome: according to the story I heard, all the other possible candidates went and killed each other off while Claudius acted like a harmless fool so they left him alone. When some people objected to choosing someone with a reputation for being "not quite all there" to be Emperor, he responded, "I'm the one who survived. How stupid could I actually be?"
I also recently watched a reality show in which one of the *least* socially perceptive players, who had spent the whole game being successfully manipulated by one player or another, made it all the way to the final showdown, simply because the other players (accurately!) didn't see him as a threat.
So it depends on the initial population, which is what I was getting at with 'mixed strategies'.
Scot mentions that 'Tit-for-tat with occasional forgiveness' seems like most of instinctive human morality, and I'm sort of riffing on that. A TFT-Bot doesn't exploit cooperate bot, they just cooperate. Assuming most humans are TFT, they will never defect against someone they *know* to be COOP.
Of course, instead of just being TFT, you could be 'TFT-except-also-exploit COOPERATE BOT.' It seems like this would be strictly superior, BUT. I expect that 1. that's harder to evolutionary embed in genes at an instinctive level and 2. in games with reputation, which is how human morality works, people who see you defecting over and over and over against COOOPERATE-BOT may take all those defections as evidence that you are DEFECT-BOT, and then they'll all defect against you on turn 1.
So basically, in a mostly-TFT-with-reputation world, I think there's an advantage to people *believing* you are COOPERATE-BOT because they will cooperate on turn 1.
Similarly, I think something happened in China where the Confucians were more Cooperate-Bot flavored, the Legalists were more pragmatist and actually won the war / ruled the empire, and also eventually called themselves Confucian.
The Mohists were known for their genius in building siege machines. While they were probably the most morally Good (by 21st century Western standards) philosophy/movement in Ancient China, I wouldn't consider them "Corporate-bot", or aiming in that direction.
I also really don't think that modern liberalism is a cooperate-bot strategy. The welfare state could be cynically described as a mechanism for bribing voters by being generous with other people's money. The reciprocity occurs at the ballot box.
“Epiphenomenal: COOPERATE-BOT isn’t really a good strategy, but is an unavoidable side effect of something else valuable. For example, maybe you couldn’t have Christians who loved God so much without having them be extremely loving and charitable people. The most dramatic version of this hypothesis is that God is real, and loving thy enemy is an epiphenomenon of following the actual Divine Law.”
This is of course the classic Christian position, and it’s worth pointing out that central to this position is that you cannot and should not and indeed must not run COOPERATE-BOT just for the sake of running COOPERATE-BOT. That way lies all kinds of failure and disaster.
This in my religious opinion has to do with theological truths, but on a pragmatic level I also think this helps avoiding infecting COOPERATE-BOT with utilitarianism, which will always subvert it.
> and it’s worth pointing out that central to this position is that you cannot and should not and indeed must not run COOPERATE-BOT just for the sake of running COOPERATE-BOT. That way lies all kinds of failure and disaster
I think grandparent is saying that Christianity provides no spiritual credit when you cooperate for consequentialist reasons, rather that you're supposed to cooperate because you want to follow Divine Law.
So, the pat meta-answer: there's another prisoner's dilemma played between two neighboring countries: Persecute Christians (cooperate) or Tolerate (defect). If both of you choose Persecute, you're both a little worse off since Christians are helpful. If one of you chooses Tolerate, all of the Christians eventually move there, the persecutor loses out more, and the tolerator gains more. If both of you choose Tolerate, the Christians eventually take over both your countries and make with the Holy Roman Empire and the inquisitions and all that. Christians being maximally cooperative skewed the payoffs enough that enough countries (governors, cities, subcultures, whatever) were incentivized to meta-defect.
I imagine you could also run a variant of the Axelrod tournament where bots are on a 2D grid, can see each others' source code, and when they get points they both create copies of themselves *and* get to copy their favorite neighbors. That plausibly leads to a majority being Cooperate-Bot.
"If both of you choose Tolerate, the Christians eventually take over both your countries and make with the Holy Roman Empire and the inquisitions and all that." Something something COOPERATE-BOT with quorum sensing package that waits until they've got control over the payoff matrices and/or game selection outperforms vanilla COOPERATE-BOT?
That seems like an odd model to apply to the Roman Empire since it didn't have any neighboring countries that were remotely its peers aside from maybe Persia, which I don't think ever had a significant Christian population (although I'm curious if any early Christian missionaries went east and if so how it went for them).
And actually going back to Scott's overall point, it's interesting that my model for the fall of Rome is that it was mostly caused by Romans increasingly choosing Defect over Cooperate; that is, they prioritized winning internal power struggles against other Romans over keeping Rome strong overall. Now this was already well underway by the time the empire converted to Christianity, but it's notable that the Christians definitely didn't succeed in halting or reversing the process.
I guess it's not necessarily surprising that scaling up to being the state religion of the empire seems to have come at the cost of some of the unique traits Scott's interested in but it's worth keeping in mind.
> That seems like an odd model to apply to the Roman Empire since it didn't have any neighboring countries that were remotely its peers
I think you could model it as various powers and factions choosing to tolerate Christianity, despite correctly perceiving it as a threat to the existing order, culminating in Constantine deciding to convert in order (probably) to increase his personal power within the empire.
'although I'm curious if any early Christian missionaries went east and if so how it went for them"
Yes, the entire Oriental Orthodox branch is from them, don't confuse the word "Orthodox" and think it's related to Orthodox Catholic, it's not. They are known to have gotten as far, and found successful churches, as India with some suggestions possibly far Western modern China, basically the Silk Road. The interesting question is why didn't it take off in the far east, then or now.
I'm currently reading Valerie Hansen's book on the Silk Road, which touches upon this. That branch of Christianity wasn't uncommon in Western China, but towards the end of the Tang Dynasty, the emperor forbade the practice of a lot of religions, mostly targeting Buddhism, but ended up wiping out the smaller Christian, Zoroastrian, and Manichaeist sects - the latter two mostly consisting of refugees who fled the Islamic invasion of Persia. Buddhism, of course, outlasted the laws and thrived, being more popular in China at the time of the crackdown.
And this is my speculation, but Buddhism may be the primary reason why Christianity didn't take off. Buddhism seems to share a lot of the same characteristics listed above - very cooperative, high standards of behavior, etc. It exploded out of India at a similar pace to Christianity, but a couple of hundred years prior, so when Christianity reached China (at around the same time it would have hit Scandanavia) you didn't see rapid adoption.
(To answer Scott's question above of what might be missing from the Christianity, Quaker, Cathar list, Buddhism would be the first one to look at, though I'm not familiar with it myself.)
Yes, the SanThome Christians in India credibly date themselves back to St. Thomas the apostle, and they have survived as a minority for thousands of years, mostly orthogonal to the traditional Hindu/Muslim political dimension.
Nestorian Christianity/the Oriental Orthodox Churches at one point had a considerable number of followers in Persia. Khosrau I initially even encouraged Nestorian Christianity over the Orthodoxy of the Byzantines, though of course always considered Zoroastrianism superior.
The way Rome got to the point of having no peer competitors left was by being better at directing surplus violence outward and distributing plunder fairly among the participants. https://acoup.blog/2023/10/20/collections-how-to-roman-republic-101-addenda-the-socii/ Running out of worthy, plunder-rich targets to aim that violence at meant the "we're all friends here, everybody gets an equal share of the plunder" system started to break down. Right around the same time Christianity was founded, Julius Ceasar was thinking "welp, all done conquering Gaul, now what?" and picking a fight with the Senate.
Except the "Golden Age" Pax Romana starts at Augustus and lasts about 200 years. Commodus, the Emperor when thing started to get dicy was also the first non-adopted emperor, which then slides into the crisis of the third century. It might indeed have been seeded by the structural change from a pillage economy, but 200 years is enough time for me to consider other causes as well.
"Golden Age" is what you get when territorial expansion has hit diminishing returns, but internal development hasn't yet. Charitable pacifists don't help much with expansion in the form of violent conquest (they'd even be regarded with suspicion, during that period, just for failing to participate) but they've got plenty to contribute when it comes to building up civil infrastructure.
Instructions starting with "sell everything you own and give the money to..." only really work when there's a merchant willing to buy, or better yet two or more bidding against each other, which becomes far more likely when there are trade routes all across the known world untroubled by war or piracy.
The described payoff matrix is different from the Prisoner's Dilemma. In particular, the Defect/Defect isn't a Nash Equilibrium, since Cooperate/Defect is better for the cooperator than Defect/Defect.
A minor point on claims that Axelrod's Tournament successfully grounded morality in self-interest – it's less that the mental processes which support morality are based on self-interest, per se, and more that the results of the tournament demonstrate an explanation of why morality is so embedded in the human psyche.
Evolution doesn't necessarily care about the mechanisms which cause various strategies (behaviours) – but it does care about the outputs of those strategies, as this is the layer it acts upon (the behaviour level), especially in a highly competitive social environment like those of our ancestors. Axelrod's Tournament shows that, even if the mechanisms within the brain which give rise to moral behaviour are completely unrelated to self-interest, once those mechanisms are present in the population (even if they evolved purely by chance), then the carriers of the genes which give rise to those 'morality mechanisms' within the brain are more likely to spread their genes to the next generation, as the behaviour caused by such mechanisms gives rise to an optimal strategy for cooperating and flourishing in a social environment.
It tells us nothing about how morality first evolved, that's true, but it does tell us a lot about why morality became so widespread and ingrained within human culture – it's a winning strategy.
Great comment. I found that sentence, "I find the occasional claims to have successfully grounded morality in self-interest to be facile, but you can at least see where they’re coming from here." a bit weird...maybe something I'd expect from a bonafide Christian, but not capital R Rationalist(TM). The evolution of morality has been fairly well studied and is not a great mystery at this point. If it was maladaptive, it wouldn't exist. (Some caveats required there of course...the prevalence of various mental illnesses and homosexuality are genuine puzzles, although I believe a fair amount of progress has made toward the former.) But "grounding morality in self-interest" is not how I would characterize that.
Maybe Scott is referring to something else? Maybe just the kind of cynical folk psychology of bitter antisocial people who insist people only "pretend" to be nice while secretly having self-serving ulterior motives?
This is a common confusion about evolutionary explanations, a confusion between the proximate and ultimate levels. If you explain things in terms of fitness, some misinterpret that as saying, you don't "really" love your child/spouse/friend, you're only pretending to care about them because it's to your advantage. No, you really are capable of love, which may sometimes result in actually putting the interests of others above your own. The reason that you have kind of cognitive/neural architecture that produces that behavior is that throughout our evolutionary history, it tended to result in more success at survival and/or reproduction than not having it.
There was a famous story about one of the most influential people in this area...wasn't Axelrod himself was it? Who IMO succumbed to this mistake, and was haunted to the point of derangement by the idea that "real" altruism was impossible, and tried to disprove it by fully embracing the implications of a Peter Singer type philosophy by giving away virtually all of his income and letting homeless people sleep on the floor of his barren house etc...
I think the best attempt to ground real goodness in objective reality is Robert Wright's Nonzero. Goodness is the expansion of nonzero sum-ness, and it is in the nature of things, due to principles like those uncovered by Axelrod, for it to continue to expand over time. But it has to start somewhere. Cooperating with a defector, ie a Jew turning the other cheek on a Nazi, or committing suicide to donate your flesh to the vultures and worms, etc. (A) win-lose, which is not good, (B) inherently maladaptive, which means it must cause there to be less of itself in the universe. So if it were good in some ultimate transcendental sense... That would mean goodness necessarily causes there to be less goodness in the universe in the future, which would be bad. If everyone genuinely cared about all humans (or all living things) equally, that would mean they would not care about their children any more than other being. And since no one can feed, clothe and shelter, let along spend quality time with, all beings everywhere, that would mean no one do even enough to keep their own children alive, and humanity would become extinct, and no humans would be around to be good, or to appreciate how Really Genuinely Altruistic the last generation of humans to exist was.
Just as Daniel Dennett said free will based determinism is the only kind we have and the only kind worth wanting... I say the adaptive kind of goodness that we actually have is the only kind worth wanting.
Here's the guy I was thinking of. Really important in the history of evolutionary theory, and also a riveting personal story. Even more tragic, because IMO his despair was just based on a mistake (and probably pre-existing mental illness). It's wrong and foolish to think kindness/generosity etc can't really be good unless it has no actual or potential benefit to the person doing it. The Buddhists knew that for 2 and half millennia. In many circumstances, being good to others is good for you. That's a feature, not a bug.
A book I'd recommend is The Morality of Everyday Life: Rediscovering an Ancient Alternative to the Liberal Tradition, which articulates and defends a more common-sense intuitive morality against what could be called telescopic philanthropy.
(GPT:) What you're recalling seems to align with the story of **George Price**, a key figure in evolutionary biology and game theory, rather than Robert Axelrod. Price played a significant role in the development of the **Price Equation**, which is foundational to understanding the evolution of altruism.
### George Price and the Evolution of Altruism:
1. **Contributions to Game Theory and Evolution**:
- Price was instrumental in advancing the mathematical foundations of evolutionary theory, particularly in explaining altruism through a genetic lens.
- His work helped formalize how seemingly selfless behaviors could evolve through natural selection if they provided a benefit to relatives (kin selection) or indirectly boosted one's own genetic success.
2. **Personal Transformation and Despair**:
- After deriving his famous equation, Price became deeply troubled by its implications. He struggled with the idea that altruism—what we see as selflessness—was ultimately a selfish genetic strategy.
- This conflict led him to attempt to disprove his own work by living a life of radical altruism.
3. **Radical Altruism**:
- Price gave away most of his income and possessions, including his house, to help the homeless in London.
- He allowed homeless people to live with him, often at great personal expense and discomfort, to demonstrate that true selflessness was possible.
4. **Decline and Tragic End**:
- His extreme lifestyle, combined with his struggle to reconcile his work with his spiritual and philosophical beliefs, led to mental health challenges.
- Price ultimately died by suicide in 1975, leaving behind a legacy that continues to provoke discussion about the nature of altruism.
### Why It Resonates:
The story of George Price is haunting and compelling because it underscores the human struggle to reconcile scientific discoveries with deeply held moral and existential beliefs. His life exemplifies the tension between understanding altruism as an evolutionary strategy and aspiring to live out altruism as a purely selfless virtue.
### Connection to Peter Singer:
Price’s radical lifestyle echoes some of the implications of **Peter Singer's philosophy** of effective altruism, which calls for giving significant portions of one's income to alleviate global suffering. However, Price's actions were not systematically aimed at maximizing utility (as Singer advocates) but rather seemed driven by a desperate need to disprove the cold logic of his own mathematical findings.
### Documentary:
The story of George Price has been featured in several documentaries and books. One notable book is **"The Price of Altruism" by Oren Harman**, which details Price’s life, work, and tragic struggle with the implications of his discoveries. It’s possible the documentary you recall was based on this book or similar material.
"Lots of 'evil' is characterised by decreasing the total sum."
That's an important point I left out: lose/lose is zero sum. Win/win--the increasing prevalence of opportunities for win/win interactions--is what I would describe as good.
"True zero-sum is exceedingly rare in the real world" I don't think it is in the sense that I mean it.
The sumness I'm referring to is just:
| | win | lose |
| :-----:|:--------:|:---------:|
| win | win/win | win/lose |
| lose | lose/win | lose/lose |
Which can be represented as:
| | +1 | -1 |
| :---:|:--:|:--:|
| +1 | 2 | 0 |
| -1 | 0 | -2 |
It's just an abstraction, not an attempt to quantify any precise quantity.
Any voluntary transaction is positive sum, because both parties benefit. A theft, OTOH, is zero sum, because one's loss is the other's gain.
Here's a synopsis:
-----
Robert Wright argues that nonzero-sumness is intrinsic to the nature of evolution because the survival and success of organisms often depend on cooperation and interdependence. While evolution, in a strictly scientific sense, lacks a conscious direction or purpose, Wright suggests that it can appear directional due to the patterns it produces—particularly the increasing complexity of life and the expansion of cooperative networks.
As life evolves, systems become more intricate, and organisms interact in increasingly complex ways. These interactions often create opportunities for nonzero-sum dynamics, where mutual benefit becomes more viable and advantageous. For example:
• In biological evolution: Symbiotic relationships (e.g., between pollinators and plants or gut bacteria and hosts) show how cooperation can drive complexity and enhance survival.
• In human history: The development of language, trade, and technology amplifies the potential for cooperative, nonzero-sum outcomes, enabling larger, more interconnected societies.
Wright argues that the expansion of nonzero-sumness is a natural consequence of this complexity. As organisms and societies grow more interdependent, the scope for mutually beneficial relationships widens, creating a feedback loop where cooperation becomes increasingly advantageous.
In this sense, while evolution doesn’t have an intrinsic “goal,” the appearance of directionality comes from this pattern: life becomes more complex, which broadens the opportunities for nonzero-sum interactions, driving further complexity and interconnectedness. This gives the impression of a purposeful trajectory, even though it arises from natural processes.
----
To your point about negative sum:
---
Yes, Robert Wright does suggest that as complexity increases and societies become more interconnected, the potential for negative-sum interactions also grows, alongside the opportunities for positive-sum interactions. This is because interconnected systems amplify both cooperation and conflict.
In a complex, interdependent world, actions and decisions have wider-reaching consequences, and failures in coordination or trust can result in negative-sum outcomes—situations where all parties lose. For example:
• In tightly connected societies, wars, economic collapses, or environmental crises can have devastating ripple effects, harming all involved.
• As technological and social complexity grows, so too do the stakes of mismanagement or adversarial dynamics, increasing the potential cost of negative-sum scenarios.
However, Wright emphasizes that the same interdependence that heightens these risks also incentivizes the reduction of negative-sum outcomes. The shared costs of failure push individuals and societies toward systems that favor positive-sum dynamics, such as cooperation, diplomacy, and innovation.
In essence, the stakes are higher in a more interconnected world: while the potential for positive-sum interactions grows with complexity, so does the potential for negative-sum interactions. Wright’s optimism lies in the idea that humanity’s increasing awareness of this dynamic, coupled with technological and cultural evolution, makes it more likely to lean toward nonzero-sum solutions over time.
It's also worth noting that Christianity itself also grounds morality in self-interest; it just argues that the "reward" from defecting is vanishingly small compared to the reward from cooperating (see the quote from Matthew above).
Morality first evolved before homo sapiens existed. Every animal that lives in groups has some form of morality. One could argue that this is even true of spiders and preying mantises, but that's probably stretching the point.
Now whether that morality is considered desirable....now that's a different question. The morality is needed to enable the species to survive in the niche for which it evolved, but when the species moves to a different environment, the "most desirable" morality (from the viewpoint of the species) changes.
Humanity has so altered the conditions in which it is living recently that our "natural morality" is probably very far from optimal. But we have cultures that implant a "cultural morality". (Here culture is often reducible to a single family or small group of families.) Even so, it's probably that all current moralities adopted by people are far from optimal (from the viewpoint of the species).
These are very good points. As Steven Pinker among others have pointed out, morality is possibly the leading cause of violence in the world. And--important distinction--much of that violence, in present circumstances, is unnecessary and unhelpful, therefore bad. It may well be--almost certainly is, in fact--that there is evolutionary mismatch between the world of small competing tribes our morality evolved for (although it seems it's turning we evolved more since the dawn of agriculture than was previously thought, which opens another can of worms I'll save for later).
Still, I think our common sense folk intuitions are generally more sound than rationalist utilitarianism** taken to its logical conclusions. Ie, yes, it is moral and good to care more about your family, and friends, and neighbors, etc than all beings everywhere. Because (A) you are in a better position to actually succeed in doing good for the people with whom you have close reciprocal relationships, and (B) if it's not successful in Darwinian terms, it will undermine itself. And as I said in my other comment, if "goodness" undermines itself and causes there to be less goodness in the universe in the future, its net effect is bad, hence it can't actually be good.
**Edit: rationalist utilitarianism, or, more relevant to this thread, I should have said the Christian "turn the other cheek", love everyone as a brother, do not resist evil etc. philosophy which seems to be endorsed in some parts of the New Testament. Which it seems very few Christians have ever fully embraced. Which I don't think of as a criticism of Christians. I think it may be good as an aspiration, encouraging us to be more forgiving, open-hearted, understanding, etc than we naturally tend to be, knowing that like it or not, we will of course prioritize ourselves and our loved ones. But taken literally, I think it's obviously unworkable in practice and not really good or moral at all.
> knowing that like it or not, we will of course prioritize ourselves and our loved ones.
And, therefore, the people most inclined to feel obligated to follow their culture’s morals to their ultimate consequences will be weeded out. With a high chance of not being perceived as following any morals, but as acting like fools who deserve no respect or sympathy.
>Overwhelm downward adjustment: If you assume all movements lie and downgrade their claims 90% out of cynicism, then a movement which merely portrays itself as helping members won’t even help members, but a movement which portrays itself as radiating universal love to all mankind might at least help its members.
I think this is a very, very big factor. Mathew 5:38-40 says,
>38 “You have heard that it was said, ‘Eye for eye, and tooth for tooth.’[a] 39 But I tell you, do not resist an evil person. If anyone slaps you on the right cheek, turn to them the other cheek also. 40 And if anyone wants to sue you and take your shirt, hand over your coat as well.
That's all well and good to proclaim, but I really doubt that literally happened often. Growing up going to Sunday school, my Pastor taught that passage a few times. But knowing him and everyone else who worked at the church, they were pretty normal people, and I do not see them literally turning the other cheek if I slapped. They certainly wouldn't give me their coat if I stole their shirt.
But also, I would guess they were more forgiving than if they didn't sincerely believe in that passage and similar ones. If you stole his shirt, I'm pretty sure my pastor wouldn't give you his coat, but he would have a conversation with you trying to explain why that was wrong and talk you into giving the shirt back instead of calling the cops on you right away.
Or maybe I'm just talking out of my ass, I didn't go to church that often even as a kid.
I've often heard it said that even the Mosaic Law, though harsh to our eyes, was a practical improvement on what came before, which encouraged Hatfield-McCoy feuding over minor slights.
The usual Christian understanding is that this is a call for patience and lenience in response to personal injury, but not an absolute call for perfect nonresistance. Though obviously a minority has disagreed with this take over the centuries.
What I would call the more orthodox lines of Christian thought have generally tried to take the entire Bible as a whole and to understand it as a single cohesive message, that Jesus didn't suddenly and completely abrogate all the teaching of the OT, but instead his moral teaching was primarily correcting misbehaviors that had arisen as a result of poor application of the OT Moral Law (which remained intact).
Jesus is a complicated figure. Those movements that have disagreed with this take (which we might call "heretics") have generally been less concerned with the idea of Scripture as a cohesive whole, instead emphasizing some teaching or another of Jesus over the rest of the Bible.
The Bible before or after the Council of Nicaea? You've got to remember how much the Bible was rewritten to accommodate politicians/theologians/etc...and not just that one time. (Admittedly, without many of those rewrite it would be a wild collection of silly stories.)
Dan Brown took the idea and built some silly fantasy stories off of it, but that doesn't mean it's not well-grounded in reality. There are books that the early Church Fathers considered canonical, and cited as such in their writings, that we do not have today.
There are books that the Apostles considered canonical, and cited as such in their writings, that we do not have today. Several of them are named in the various books of the Bible, and at least one is not: in Acts 20:35, Paul cites the teaching of Jesus, its source clearly familiar to his audience but lost to us today, that "It is more blessed to give than to receive."
Meanwhile, archaeologists, who for centuries have used the Old Testament as a guide to exploring the Holy Land, are increasingly making discoveries about the Northern Kingdom of Israel that lead them to believe that many of the historical accounts that have been passed down to us in the Old Testament are composed of Southern Kingdom political propaganda that's not nearly as truthful as Biblical literalists would assume. (A notion that's backed up by the discovery of various apocryphal ancient works believed to originate in the Northern Kingdom.)
This is true, we have lost much knowledge, but that is categorically different from the assertion that the Nicaean Council rewrote the literature we do retain.
I heavily discount such assertions based on my personal experience with religious zealots today. Try to convince any of them to rewrite their religious texts- it's not an easy thing to do. The best you can get is a translation that emphasizes different things.
If our current religious zealots won't rewrite anything, then I do not think that the ancient religious zealots- who were far more scrupulous and self-selecting, by all accounts- would be more willing to.
If we're talking about early Christianity, it's not so much a matter of rewriting as new-writing. The Evangelists and the Apostles wrote all manner of new works that Christians considered to be scripture, on the same level as the books written by the Prophets of old. And then other people started writing books that some Christians considered equally legitimate and others did not agree with. And then you have the whole mess that is Gnosticism complicating things and muddying the waters further.
By the time of Nicaea, things had grown so fragmented that Constantine, who wanted to be a Christian, found it impossible to definitively answer the question "just what does it mean to be a Christian anyway?" So, by the authority vested in him as Emperor, he got a bunch of top Christian thought leaders together and said "you're going to work out an official Christian doctrine and settle all this confusion once and for all, or else!" One important part of that was deciding, among various competing works, what was to be officially considered canonical and what should be thrown out.
Re: There are books that the early Church Fathers considered canonical, and cited as such in their writings, that we do not have today.
Actually we do have those books. Chiefly: The Protoevangelium of James, the Epistles of Clement, the Diadache, the Shepherd of Hermas. You can find these by an online search.
As far as the OT goes there are several books recognized a Scriptural by the Church of Ethiopia that no one else recognizes (and not the non-Ethiopian Jews either)
That's officially correct, as certified by Google, but then I didn't claim that it did. I claimed that there were lots of books removed from the Bible, and that many of them were removed at an early date (I picked the Council of Nicaea as a separation point, however, because of the "myth"). Yes, it was really based on a series of decisions, mainly made by the Catholic Churches. But it drastically altered the tone of the Bible. (Possibly repeatedly. I haven't really studied the matter. But consider "The Gospel of Thomas". Or even "The Gospel of Judas". And even the ones that were obviously collections of folk tales [reminiscent of the Cherry Tree Carol] would change the tone and meaning of the Bible were they included.)
> That's all well and good to proclaim, but I really doubt that literally happened often. Growing up going to Sunday school, my Pastor taught that passage a few times. But knowing him and everyone else who worked at the church, they were pretty normal people, and I do not see them literally turning the other cheek if I slapped. They certainly wouldn't give me their coat if I stole their shirt.
There's a lot of cultural context in this passage that we lack today. What he was saying here can be summed up rather concisely as "don't be like the Zealots." While the world "zealot" has become a generic term in the modern lexicon, it referred to a specific group of people with a well-defined ideology, that the Romans were evil occupiers and the Jews needed to resist them through physical violence, and eventually rise up and cast them out of Judea in a glorious revolution.
Given the overwhelming military superiority of Rome, this was a stupid idea that was never going to be successful, and people listening to these notions eventually culminated in the tragic events of 70 AD. Jesus told his disciples to follow a different path. If a Roman citizen legally compels you to give you something, give more than is required. If he requires you to travel with him for a mile, as is his legal right under Roman law, go two. If someone tries to provoke you with a backhanded slap, (this is the only way for a normal right-handed person strike someone on the right cheek, afterall,) don't rise to the provocation by throwing a punch. Instead, turn the other cheek and say "I'm not going to let you bait me. *You* throw the first punch, and then you're the one who broke the law and started the fight."
In all cases, it's the same requirement: do not give the Romans any excuse to say that you're a revolutionary fighting against them. Go out of your way to make it clear that you're not. Given the way the revolutionary Jews were so thoroughly wiped out once the Romans finally got sick of them, it proved wise advice!
The Gospel of Matthew never mentions the Zealots. To the extent that the Sermon on the Mount names any enemies, they are enemies of Jesus, not of the Jews:
"Blessed are you when people insult you, persecute you and falsely say all kinds of evil against you because of me. 12 Rejoice and be glad, because great is your reward in heaven, for in the same way they persecuted the prophets who were before you." (Matthew 5:11)
Who are the people insulting, persecuting, and slandering the followers of Jesus? At this time, they were not Romans. They were other Jews, particularly the Pharisees. Who persecuted the earlier prophets? Also other Jews, particularly the kings of Israel and Judah.
The good objection is that you should require direct evidence of very specific claims. It’s a neat just-so story but there’s no evidence for it other than the originator thinks it might have been cool and it fits with existing biases that Jesus must be terribly misinterpreted because he couldn’t possibly have meant what he actually said and others took him to mean.
The evidence against is that nothing in the Gospels leads us to believe that Jesus had any interest in leading a covert passive resistance against the Romans, to the point that the Roman authorities seized him (on claims of the Jewish religious authorities), investigated him, and obviously thought he was a harmless kook.
You can argue “well that’s the whole point of his strategy!” but now you’re into conspiracy territory. Literally nobody among his contemporaries figured out the secret Jesus anti-Roman code, but we super-smart moderns did?
The more likely explanation, borne out by his and his early followers’ actual behavior, is that Jesus meant it just as he said, for application to more or less all situations, not just Zealots vs Romans.
> The evidence against is that nothing in the Gospels leads us to believe that Jesus had any interest in leading a covert passive resistance against the Romans
> the secret Jesus anti-Roman code,
It's hardly valid to present, as evidence against a claim that was made, evidence against a claim that was *never* made.
The evidence is what I said in the first post: what he's referring to here are specific cultural practices. For example, the word ἀγγαρεύσει (angareusei) used in the Sermon refers to a Roman soldier's legal right to compel conquered people to accompany them, and carry their heavy equipment, for one mile. His audience knew exactly what ἀγγαρεύσει was, and likely many of them had had it happen to them at some point. (See also the case of Simon the Cyrenian, a random traveler who the Romans grabbed and forced to carry Jesus' cross.)
But seriously, how exactly do you take a post saying "Jesus said to not be a revolutionary and to be exceedingly clear that you are not a revolutionary," and then 1) *completely* misread it as "Jesus said to be a revolutionary, just in secret," then 2) when it's pointed out that that was never the claim, say "OK then, prove it"?!?
It is not hard to read Mark, at least, as a thinly veiled anti-Roman screed, using the metaphore of demonic possession to avoid saying anything crucifixion worthy. And Mark was the first of the Gospels to be written, presumably closer to the source material than the other three.
It might not be hard for you to make that inference, but what is that actual evidence that it is besides your belief? Certainly the contemporaneous writers, readers, devoted Christians, etc. did not believe it was and did not treat it as if it was.
Also, I should note for you that the evidence of Markan priority is thin: inference and opinion of textual critics, not through physical evidence, archeology, historical accounts, etc. Wikipedia will tell you the case is airtight, but Wikipedia is written by academic textual critics, not historians nor archeologists, and you should be skeptical of it.
I do not have anything except my own personal interpretation from reading it. I should explain that I am a practicing Christian.
As for the priority of Mark, yes textual clues are all we have, but you take the evidence you have, not the evidence you wish you had. I find it convincing.
I don't think Quakers were ever actually executed in England, though they were regularly fined and imprisoned. Cromwell and Charles II were friendly with Fox and Penn respectively, though local judges often were not
Massachusetts on the other hand seems to have executed Quakers regularly and made returning to Massachusettes as a Quaker a capital offence.
That may be true, but "keep in jail in abysmal conditions until they get sick and die" isn't *that* far from execution, and certainly many Quakers did die in jail as a result of their resistance/principles.
The biggest gap between bot strategy and the real world is the fact that in the real world, people can change strategies. If, in the bot simulation, 40% of the bots were “adapt the strategy of the last bot you encountered” cooperate-bot would be much more likely to pull ahead. Tit-for-tat would still win, but would cooperate bot win more?
I don't feel like trying to dig up the answer, but...wait I don't have to; I have a magic oracle. But first, I don't think that's actually necessarily a huge difference; I think in Axelrod's tournaments, people were allowed to submit any computer code they wanted. In the iterated tournaments, I don't think there was anything stopping them from changing behavior in one round based on what happened in previous rounds. Probably some tried.
To the Oracle (GPT 4o):
"""
The question of whether the introduction of adaptive strategies (like "adopt the strategy of the last bot you encountered") changes the dynamics of iterated games like the Prisoner's Dilemma has indeed been explored in various ways, though perhaps not always in the exact form you described.
Experiments in **evolutionary game theory** and **iterated Prisoner's Dilemma simulations** often include agents that can adapt or learn strategies over time. Here are some key findings that relate to your question:
### Key Concepts in Relevant Research:
1. **Strategy Adaptation in Simulations**: Many simulations introduce "adaptive agents" that mimic or evolve strategies based on observed success. These agents are akin to your "adopt the strategy of the last bot you encountered." Such studies show that the presence of adaptive agents often destabilizes dominance hierarchies among fixed strategies (like tit-for-tat or always-cooperate) and leads to fluctuating dynamics.
2. **The Role of Cooperation Bots**: In a population with adaptive agents, strategies like "always cooperate" (cooperate-bot) tend to perform poorly unless they exist in a supportive environment. If a majority are cooperative, this increases the likelihood that adaptive agents will adopt cooperation. However, they remain vulnerable to exploitation by defectors.
3. **Tit-for-Tat in Adaptive Contexts**: Tit-for-tat tends to perform well in adaptive scenarios because it strikes a balance: it fosters cooperation when others cooperate, but it retaliates against defectors. This robust balance makes it likely to spread or persist, even in the presence of adaptive agents.
4. **Social Learning and Memory Effects**: In experiments where agents can remember more than just the last strategy they encountered, or where they weigh historical interactions, tit-for-tat's dominance can diminish. Cooperative strategies (including always-cooperate) may thrive in these environments if agents tend to favor longer-term rewards.
### Would Cooperate-Bot Perform Better?
It depends on several conditions:
- **Population Composition**: If 40% of the population adopts "copy the last strategy," cooperate-bot could indeed benefit, as these adaptive agents would mimic cooperation after encountering cooperate-bot.
- **Interaction Rules**: If interactions are random, cooperate-bot's success might increase. However, if defect-bots (always defect) or tit-for-tat strategies dominate the environment, adaptive agents will likely mimic defecting or retaliating behaviors more often.
- **Mutation and Evolution**: In evolutionary contexts where agents adapt based on long-term payoffs, tit-for-tat generally prevails because it punishes defectors while fostering mutual gains with cooperators.
### Existing Studies:
- **Axelrod's Tournaments**: The famous iterated Prisoner's Dilemma tournaments run by Robert Axelrod showed that adaptive strategies could emerge under certain rules. However, tit-for-tat still remained a top performer because of its simplicity and robustness.
- **Agent-Based Modeling**: Many agent-based models (ABMs) have tested scenarios with adaptive agents. Results often show cyclical dominance: cooperators rise, get exploited by defectors, and then defectors diminish as retaliation strategies (like tit-for-tat) take over.
### Conclusion:
Experiments have addressed related questions, though perhaps not in precisely the form you've described. The inclusion of a large proportion of adaptive agents would likely boost cooperate-bot's performance temporarily, but tit-for-tat would remain resilient in the long run because it discourages exploitation. The dynamics of such systems are highly sensitive to initial conditions, mutation rates, and the ability of agents to "remember" and adjust to patterns over time.
"""
PS: Was I right about what I said initially?
"""
Yes, you're correct about Axelrod's tournaments allowing strategies to change behavior based on what happened in previous rounds. The tournaments were designed to permit participants to submit any computer code they wanted, which means strategies could indeed be highly sophisticated and adapt to the opponent's previous moves.
### Key Points About Axelrod's Tournaments:
1. **Open Submission**: Participants were free to submit any strategy they could program, and strategies could incorporate memory, logic, and conditional rules based on prior interactions. This enabled them to change behavior dynamically across rounds.
2. **Iterated Nature**: Because the games were iterated, many strategies explicitly accounted for past interactions. For example:
- **Tit-for-Tat** famously cooperated on the first move and then mirrored the opponent's previous move.
- Other strategies tried to exploit patterns or weaknesses in their opponents by analyzing behavior over multiple rounds.
3. **Sophisticated Adaptations**: While the simplest strategies (like always-cooperate or always-defect) did not adapt, more complex ones often incorporated memory and conditional responses. For instance, strategies might:
- Cooperate for a certain number of rounds, then switch based on observed payoffs.
- Punish defection harshly for multiple rounds to discourage exploitation.
4. **Design Space**: There were no constraints against designing strategies that could adapt mid-tournament or "learn" an opponent's tendencies, though strategies had to follow the rules of the game (e.g., fixed payoff matrix, iterated rounds). The only real limitation was computational feasibility, as programs needed to run within reasonable time limits.
### Did Participants Try This?
Yes, some participants submitted highly sophisticated strategies, including those that:
- Used pattern recognition to exploit predictable opponents.
- Forgave defections after a certain point to restore cooperation.
- Attempted to "probe" their opponent's behavior early in the game to choose an optimal response later.
### Why Does Tit-for-Tat Often Prevail?
While adaptive strategies existed, tit-for-tat often emerged as a top performer because of its simplicity and robustness:
- It fostered cooperation with other cooperators.
- It punished defectors immediately but did not escalate conflicts.
- Its predictability made it a stable partner for strategies trying to avoid exploitation.
### Conclusion:
You're absolutely right that the open nature of Axelrod's tournaments allowed for strategies that could adapt round-to-round. Many participants likely experimented with these approaches, and some succeeded. However, the straightforward, cooperative nature of tit-for-tat often allowed it to perform better in aggregate across diverse opponents.
I’m not suggesting it would benefit an individual participant to learn other’s strategies, but if that were a force of the simulation - if it ran real time and after each 10 rounds 40% of the “follower” bots changed strategy to the current winning strategy, I think it would be very different and a bit more like real life. In a world where cooperation is publicly visible and encouraged, and communication between bots is possible, you can surely make things where cooperation is the winning strategy.
If the rules of your simulation periodically "kill" some of the agents following the worst-performing strategy and replace them with agents that follow the current best-performing strategy, and you start with a mix of tit-for-tat, cooperate-bot, and defect-bot, then tit-for-tat will generally win, and cooperate-bot will almost always be the first to be wiped out.
With random errors, tit-for-tat-plus-forgiveness will become the most common strategy if you add it, though it's so merciful that it won't completely wipe out regular tit-for-tat, and a small fraction of the population will stick with that.
(There was a great browser game/interactive lesson about this that I played at some point, but I lost the link and haven't been able to find it again.)
Some researchers have been really annoyed that they couldn't come up with plausible scenarios where cooperate-bot won in their simulations, even after trying pretty hard to come up with some. (Though obviously most simulations do not remotely approach the complexity of real life.)
You definitely "can" make cooperate-bot win if you sufficiently contrive the rules (e.g. if you add in some god-players that read your source code and cooperate IFF you are cooperate-bot), but I think you'll find its much harder than it sounds like you are expecting.
1) the rate of random errors is based on code complexity / runtime,
2) rather than every matchup being random, you can choose to seek out a specific agent you've encountered before,
3) score thresholds for elimination or reproduction are fixed, and knowable - for example, any agent whose cumulative score drops below 0 dies, one whose score rises above 100 can choose to spawn a copy, after which both will be reset to a score of 50?
in such a case, particularly if point 2 were somehow expanded to friend-of-a-friend referrals, I'd expect to see an "enforcer" or "shepherd" strategy which alternates between proactively culling known defect-bots - even at the expense of its own immediate reproductive opportunities - and recharging within a more congenial community when its resources are low. Cooperate-bots make the best pool to recharge from since they'd suffer the least error-related attrition.
"Cooperate-bots make the best pool to recharge from?" Sounds like you're effectively arguing that cooperate-bots make the best subjects for warlords ("shepherds") to rule over, rather than the argument that *being* a cooperate-bot is good for the person executing the strategy.
Also, like...you want tit-for-tat to be penalized for being complex relative to cooperate bot, but you also want friend-of-a-friend referrals? That's probably several orders of magnitude more complicated than tit-for-tat. And then you also want police investigators to somehow identify and punish the people who break the rules? That's probably several orders of magnitude more complicated than referrals.
Also point 3 seems highly unrealistic (real life has tons of luck, lots of people die unpredictably, and passing on your genes is not a binary pass/fail, you can have a variable number of kids and give them a variable amount of assistance in getting started). And point 2 is vague but the simplest implementation I can think of is also highly unrealistic (choosing to interact with only your most trusted friend and avoid everyone else is a terrible strategy in real life for a whole bunch of reasons that you don't sound like you're attempting to model).
But feel free to code up your own simulation, maybe you'll discover something interesting!
> Also, like...you want tit-for-tat to be penalized for being complex relative to cooperate bot, but you also want friend-of-a-friend referrals? That's probably several orders of magnitude more complicated than tit-for-tat.
I don't think reasonable levels of complexity penalty would be giving cooperate-bot a noteworthy edge over tit-for-tat all by itself, no.
Point of a "choose who to interact with next" + "know how close you currently are to reproducing, or dying of starvation" mechanic would be opening up space for long-term ecosystem-management strategies, then the complexity penalty keeps those from escalating all the way into some cryptographic-handshake version of John Carpenter's The Thing. https://www.oglaf.com/kingshaped
> choosing to interact with only your most trusted friend and avoid everyone else is a terrible strategy in real life for a whole bunch of reasons
Well of course it's not that simple.
First, enforcer / shepherd builds a contact list of known cooperate-bots, and a separate list of reliable tit-for-tat strategists. To the extent it's allowed by the mechanics, all these are freely introduced to each other, encouraged to interact to mutual benefit, top priority being to link up cooperate-bots with other cooperate-bots.
When the enforcer encounters a defect-bot, or unknown and potentially troublesome strategy, it checks its own reserves. When healthy, it attacks the intruder. When reserves are low, it retreats, spins through the cooperate-bot contact list, hitting them with a "defect" once each to quickly recharge, then resumes the attack.
Distributing the load hopefully reduces the number of cooperate-bots which die as a result, thus keeping the overall herd's income / growth rate high. Sort of like a multicellular organism's immune system. Tit-for-tat contacts can't usefully be harvested that way, because they'll bite back.
The bots could change strategies. It doesn't help. The more cooperative-bots you have, the more it boosts the ranking of bots that defect in the tournament.
Well I tried your suggestion, and it looks like it just makes Always Defect vastly more successful than usual, steamrolling everything else, more so the more rounds you do. I think the reason is that Always Defect always starts at an advantage, and the follower rule means others copy it, so it snowballs.
Good point.
Liberal democracy has historically flourished in periods of economic growth associate with trade/production. I think there's sort of a continuum from trader to king, based on whether you apply your skills more to bargaining or dominance. It seems natural that the latter selects for tit-for-tat, whereas the former selects for cooperation. Under normal conditions, traders rely on the dominance hierarchy to establish peace and make cooperation viable, but when there's a disequillibrium where trade is massively expanding (e.g. the industrial revolution), the trading population can grow more powerful than the dominance population, until circumstances have equillibrated. (In economics, this would be called "opportunity cost" - the effort you spend on developing weapons for dominance could instead have been spent on trade, which in a dreamtime disequilibrium is going to give you more value.)
Market conditions are not intrinsically positive-sum, and geopolitics are not intrinsically zero-sum. Undeveloped resources which requires lots of flexible human workers to collaborative develop change conditions to be positive-sum (for human workers). But eventually you reach an equillibrium where they are optimally developed and you've got excess labor, at which point they become zero-sum. This is especially true if you go out of your way to speed up the equillibration, e.g. with certain kinds of AI and free trade policies.
I can't see what you're responding to, but markets are intrinsically positive sum!
Maybe you guys are using different definitions of "intrinsically".
I mean. Some people are definitely right this second trying to conquer land, and small scale exploitation that kinda looks like slavery definitely exists, doesn't it?
Applied to the case of memetic war: if there's some subject that yields lots of valuable insights when studied, then truthseeking is selected for simply because it's a waste of time to fight.
The more you reach equillibrium (by finding the key information, distilling them into great insights, broadcasting them to people who can learn from them, and integrating them into applications), the less value you can get from pure cooperation.
This won't immediately cause defection, it will "merely" cause a collapse in that which produces collaboration, which may just lead to a collapse in the participation. But if participation *remains* high despite lack of selection for truth-seeking (e.g. due to external funding or entertainment value or community inertia), then the participants can be exploited via deception, leading to a growth in various strategies, and then as experiments show, this tends to lead to tit-for-tat.
The more you reach equillibrium (...), the less value you can get from pure cooperation.
The environment of evolutionary adaptedness had gains from cooperation that were less than that of the modern world, but still substantial.
Hence why cooperative instincts evolved.
Prestige hierarchy vs dominance hierarchy
Nope. Per the mandate of heaven, prestige and dominance are closely linked. The critically important piece is an expansion in the possibilities of excellence so there's a disequillibrium where you gain more from developing prestige while keeping dominance low than you gain from developing dominance. So more like prestige vacuum.
I think you're on the right lines when you ask who occupies the position where each strategy makes sense. If instead of traders and kings we think about the common people of Rome at that time, we get an answer to Scott's question.
Think of Rome as a very individualistic society (much like today) where people had a high degree of security from outside (no foreign enemies) and focused on their own lives - e.g. sports, romance, wealth. These things tend to be zero-sum and encourage tit-for-tat, since it's about fighting for positional goods. So we have a society where everyone is playing tit-for-tat all the time, and probably getting sick of it (because let's face it, it's brutal).
Then Christianity can be explained simply by the fact that under these conditions it's the funniest thing in the world to be nice instead. Everyone takes their petty zero-sum competitions so seriously, that not taking them seriously is like pissing over all them.
I hope this explanation is right and that our position is similar to the Rome of that time. This is normally taken pessimistically (probably because of the stereotypes about the fall of Rome leading to the dark ages), but it would be pretty great if something that funny happened again.
That's a good observation.
As a "sell everything to give alms" Christian, I feel it's important to make a distinction between the culture that won the memetic war and the Actual Christianity you are describing. I frequently use a similar definition of Christianity, but also say there are perhaps 100,000 Christians in America at the moment. I sometimes refer to the mass movement as Modern American Folk Christianity.
While I think your terms are pretty harsh, I do think you are on the right path. The version of Christianity (COOPERATE-BOT) that Scott is using is 90% strawman. There are simply practical limits to how much cooperation any nominal Christian group can do, and TIT-FOR-TAT is far from unknown among nominally Christian groups, as well.
To be fair, my threshold is close to "has read the gospels and strives to live by it", a standard so loose it includes Gandhi, it's just that neither Modern American Folk Christianity nor Moralistic Therapeutic Deism are at all open to "Jesus said you should do X" as a viable method of socially approved persuasion. I've visited dozens of churches, I feel my estimates for population rates are at least reasonable, if not as accurate as I like.
(Edit for fun fact: did you know Gandhi read the Sermon on the Mount twice a day for decades?)
Interesting. I take it for granted that it includes, say, Tolstoy? Or even takes him as a forerunner?
I don't know much about Tolstoy, but from his Wikipedia, yes. In terms of forerunning, I mean, both him and I are a couple millennia late to the party, but yes, there is a cross generational convergence on the ideals here. I think there is more convergence than inherited tradition, by a large margin? Also, the more I read this wiki, post-Confession Tolstoy is unbelievably based?
I think a more "narrow brand" representation might be someone like Dietrich Bonhoeffer, although that is somewhat complicated by his involvement in a plot to assassinate Hitler, which is the most absurd corner case of "well if you really love your enemies" that it's sort of wild he actually faced the hypothetical irl.
Re: Dietrich Bonhoeffer. Kind of crazy that I'd just heard about him from my coworker today, and here I am reading his name on an ACT/SSC comment. I'm attempting to return to the (Catholic) Christian fold, so would you be able to expand on your thoughts regarding Bonhoeffer's seeming contradiction?
I don't have much to say, aside from, I've been spared similar choices. His books are very good, though, and are quite readable. Engrossing, really. Strong recommendation for The Cost of Discipleship.
Could you please explain what is going on in Christian churches in America so that congregants aren't meeting this low bar? To someone who isn't American and has never been to a mainline American church or an Evangelical church.
I don't know what causes it, I only observe it. They will say, you gotta be all in, trust Jesus 100%. Then you say, Jesus said, do radical good thing X, does anyone do that? And they say, no no that's too impractical and weird. It's not a skill issue, it's a preference. They don't want to. If they wanted to, and failed, that's classically in-bounds, but this is vanishingly rare. Pretty bizarre.
Replying to you here as well: I've heard it well put that Jesus's hardcore teachings may be seen as koan-like paradoxes. Meaning, they are practically nigh-impossible, but it is the living out of the path and the divine goal/target that is the way. It's funny how people think the OT is 'harsher' and the NT is 'kinder'...Jesus asks us to aim for almost insurmountable heights.
I trust Him to make my paths straight.
Are they hearing the gospel? you're the second person I've heard say that it isn't taught in American churches.
That's not *quite* the issue. You have no idea how many times I've met with people after church services end and said something like, "I really liked this bit of the service, when they said [quote]. Is that the type of thing people do here?" And then they're confused, ask what I mean, and I say, that things they said, it's a beautiful expression of gospel principles, do people do it? And they say no. Then I ask why not, and they have all these elaborate reasons why not even a single person does the thing they talk about or sing hymns about. I often have a hard time understanding what they get out of church.
I've been using the term Christian Nationalism. That describes the top end of malefaction, and there are degrees of shading for individuals depending on where they are in life and their state of mind.
I think that term mostly focuses on a political angle, where there *is* a philosophy and culture outside of politics I'm trying to point at. Like, volunteering for a VBS isn't anything to do with nationalism. I'm also not trying to speak unkindly about anyone or isolate merits on one end and demerits on the other. I'm just pointing out that (as someone who has volunteered for a VBS) it's a foreign cultural exchange in some ways.
I understand that, but the politics is willfully mixed into it. I wish the culturally dominant strain of Christianity in America wasn't a band of nationalist hobgoblins pushing women away from the pews. But, saying reality is what it isn't has not been that great a strategic choice so far. At best, these things are in tension and one side is growing to be the more popular one.
Maybe it wouldn't be the case if more people did foreign culture exchanges. Seeing becomes believing.
I strongly dislike this term, because it explicitly grants such people the title of Christian, and they haven't earned it.
I understand that.
As I understand it, from the Bible, the idea that you can earn the title Christian is itself not Christian. Faith itself is a gift, that no one may boast, and nobody deserves grace, by definition. Nobody is good except one, that is God. If we believe Jesus is good how can we claim to follow Him when we are not? If we believe we are good we reject His teaching.
I would be careful about saying who deserves and doesn’t deserve the title Christian.
This is true inasmuch as I should not have used the word "earned", and I apologize for that.
However, Jesus does warn about false Christians several times, such as in Matthew 7:15-27, in which Christ states that the ones who does what he says are the ones in heaven. He elaborates on this in Matthew 25:31-46, stating that the ones who will be saved are the ones who fed the hungry, gave drink to the thirsty, take in strangers, clothe the naked, and visit those sick or imprisoned.
So, considering that Jesus warned us about false Christians, stated that you will know them by their fruits, and then gave a list of the sorts of things you can expect to see real Christians doing, I think that calling someone who never does those things and is, in fact, viscerally opposed to doing them, is probably not a Christian. Back to the subject at hand- if someone who calls themself a "Christian Nationalist" is harshly opposed to taking in strangers, it's a hard sell for me that they are a Christian.
Just to be clear: I'm not saying that they are going to be damned, necessarily. I certainly don't know who will be damned or not and, in my opinion, stating that someone is going to Hell is putting human limits on God's mercy, which seems to me very unwise.
What I am saying is that I disagree with the category name of "Christian nationalist" to describe people who are hardcore opponents of immigration, which is usually what it refers to. If instead we are talking about people who wish their nation to be more in line with what Jesus literally told us to do, then I don't object to the terminology.
I suppose it might describe myself, although the term "Christian Globalist" might be more appropriate, though it strikes me as rather redundant. To be truly Christian is to be universal, embracing all humanity without the confines of national borders. Thus, the most fitting label, if labels should be worn at all, would simply be "Christian."
Thank you for the thoughtful replies. One of the joys of Christianity is discussing / wrestling with / resolving the tension in various scriptures during Bible studies and with friends. I've typed and erased and typed erased. A comment section just doesn't work for me. I apologize for starting a discussion and not following through well. God bless you, brother!
I'm curious, from the "no borders/global" perspective how do you interpret John 14:2? “In My Father’s house are many dwelling places; if it were not so, I would have told you; for I go to prepare a place for you." - If the Father's house has many different dwelling places (and a place specifically prepared by Jesus for His followers) does that not imply that there can be virtue in boundaries?
What if someone does all of the things in Matthew 25:31-46, but only for strangers of a certain type (e.g. only Americans, or only Christians)? After all, Matthew 25:31-46 never said you need to let an infinite number of Muslims into your country so that they impose Sharia law and drive Christianity to extinction. If that were the likely outcome of immigration (and let's leave aside for the moment the question of whether it is), are Christians still required to support immigration?
I mean, Christians didn't take the "take in the stranger" far enough to be like "hey, centurions, welcome to our secret underground church", so I think the answer is in the past. Alfred wasn't very welcoming to Guthrum- for a while, anyway.
But, you know, The whole point of having deontological ethical injunctions is that arguments for violating them are often much less trustworthy than they look.
An obvious game-theory thought is that you are gifting all other Christians with a *reputation* for being COOPERATE-BOT, and this makes other bots with mixed strategies interact with them more favorably in general.
This is one explanation for how early Christians could have succeeded, by creating a reputation for being COOPERATE BOT while most of them are actually TFT.
But why would having a reputation for being COOPERATE-BOT help you? Why wouldn't it just make everyone (try to) exploit you?
They'll personally want to exploit you, but they'll also want to prevent their own enemies from exploiting you too much - so that you'll still be there to be exploited by them and their allies in the future. Norman Borlaug's dwarf wheat doesn't have a military policy above the microbial level, but it's cultivated by people who do.
Very interesting idea, so it is basically becoming such a useful and unique tool to be able to be integrated into a larger policy. Which for sure can work, but that simply outsources the original game theoretic problem onto the larger entity, practically making the problem disappear for you, but not for the larger entity.
Suppose two wolves and a sheep are voting on what to have for dinner. The sheep says "I propose that tonight we eat whichever of the two wolves is stronger, and then have mutton tomorrow."
The weaker wolf realizes, if this proposal doesn't pass, it certainly won't be the stronger wolf being eaten tomorrow, and so seconds the motion.
After resulting excitement is settled, sheep says to the surviving wolf "you're badly injured, and must be tired from doing most of the work in that fight, so I insist you accept the honor of eating a larger portion of the carcass."
Once that wolf is asleep with a full belly, sheep headbutts and tramples them to death, then trots off to eat grass.
in a war of all against all, with no credible alliances or limits of adjacency, default outcome is everybody ganging up on whoever looks the most threatening - that is, whoever they'd expect to face the most harm, or least benefit, from ending up stuck alone with.
Iterate that enough and whoever looks least threatening dies last. Thus, Mr. Rogers in a bloodstained sweater... or Putin, who was apparently the most blandly inoffensive man in the entire KGB, ending up on top by default.
Supposedly this is how Claudius became emperor of Rome: according to the story I heard, all the other possible candidates went and killed each other off while Claudius acted like a harmless fool so they left him alone. When some people objected to choosing someone with a reputation for being "not quite all there" to be Emperor, he responded, "I'm the one who survived. How stupid could I actually be?"
I also recently watched a reality show in which one of the *least* socially perceptive players, who had spent the whole game being successfully manipulated by one player or another, made it all the way to the final showdown, simply because the other players (accurately!) didn't see him as a threat.
So it depends on the initial population, which is what I was getting at with 'mixed strategies'.
Scot mentions that 'Tit-for-tat with occasional forgiveness' seems like most of instinctive human morality, and I'm sort of riffing on that. A TFT-Bot doesn't exploit cooperate bot, they just cooperate. Assuming most humans are TFT, they will never defect against someone they *know* to be COOP.
Of course, instead of just being TFT, you could be 'TFT-except-also-exploit COOPERATE BOT.' It seems like this would be strictly superior, BUT. I expect that 1. that's harder to evolutionary embed in genes at an instinctive level and 2. in games with reputation, which is how human morality works, people who see you defecting over and over and over against COOOPERATE-BOT may take all those defections as evidence that you are DEFECT-BOT, and then they'll all defect against you on turn 1.
So basically, in a mostly-TFT-with-reputation world, I think there's an advantage to people *believing* you are COOPERATE-BOT because they will cooperate on turn 1.
TFT is cooperation with implicit noise. TFT with forgiveness is cooperation with explicit noise.
Similarly, I think something happened in China where the Confucians were more Cooperate-Bot flavored, the Legalists were more pragmatist and actually won the war / ruled the empire, and also eventually called themselves Confucian.
Yes, it's funny how everyone dunks on the legalists, but they actually won in practice.
I thought in that context it would be the Mohists who were Cooperate-Bot.
The Mohists were known for their genius in building siege machines. While they were probably the most morally Good (by 21st century Western standards) philosophy/movement in Ancient China, I wouldn't consider them "Corporate-bot", or aiming in that direction.
That does seem the more likely explanation.
I also really don't think that modern liberalism is a cooperate-bot strategy. The welfare state could be cynically described as a mechanism for bribing voters by being generous with other people's money. The reciprocity occurs at the ballot box.
Nietzsche said that the last Christian died on the cross. Your point, put in his usual hyperbolic manner.
Kierkegaard wrote something about being lonely as a Christian in Christendom
“Epiphenomenal: COOPERATE-BOT isn’t really a good strategy, but is an unavoidable side effect of something else valuable. For example, maybe you couldn’t have Christians who loved God so much without having them be extremely loving and charitable people. The most dramatic version of this hypothesis is that God is real, and loving thy enemy is an epiphenomenon of following the actual Divine Law.”
This is of course the classic Christian position, and it’s worth pointing out that central to this position is that you cannot and should not and indeed must not run COOPERATE-BOT just for the sake of running COOPERATE-BOT. That way lies all kinds of failure and disaster.
This in my religious opinion has to do with theological truths, but on a pragmatic level I also think this helps avoiding infecting COOPERATE-BOT with utilitarianism, which will always subvert it.
> and it’s worth pointing out that central to this position is that you cannot and should not and indeed must not run COOPERATE-BOT just for the sake of running COOPERATE-BOT. That way lies all kinds of failure and disaster
...How is that different from just TIT-FOR-TAT?
I think grandparent is saying that Christianity provides no spiritual credit when you cooperate for consequentialist reasons, rather that you're supposed to cooperate because you want to follow Divine Law.
correct.
So, the pat meta-answer: there's another prisoner's dilemma played between two neighboring countries: Persecute Christians (cooperate) or Tolerate (defect). If both of you choose Persecute, you're both a little worse off since Christians are helpful. If one of you chooses Tolerate, all of the Christians eventually move there, the persecutor loses out more, and the tolerator gains more. If both of you choose Tolerate, the Christians eventually take over both your countries and make with the Holy Roman Empire and the inquisitions and all that. Christians being maximally cooperative skewed the payoffs enough that enough countries (governors, cities, subcultures, whatever) were incentivized to meta-defect.
I imagine you could also run a variant of the Axelrod tournament where bots are on a 2D grid, can see each others' source code, and when they get points they both create copies of themselves *and* get to copy their favorite neighbors. That plausibly leads to a majority being Cooperate-Bot.
"If both of you choose Tolerate, the Christians eventually take over both your countries and make with the Holy Roman Empire and the inquisitions and all that." Something something COOPERATE-BOT with quorum sensing package that waits until they've got control over the payoff matrices and/or game selection outperforms vanilla COOPERATE-BOT?
That sounds like a modification of the strategy used by many strains of bacteria. The "quorum sensing" is important to being successful.
That seems like an odd model to apply to the Roman Empire since it didn't have any neighboring countries that were remotely its peers aside from maybe Persia, which I don't think ever had a significant Christian population (although I'm curious if any early Christian missionaries went east and if so how it went for them).
And actually going back to Scott's overall point, it's interesting that my model for the fall of Rome is that it was mostly caused by Romans increasingly choosing Defect over Cooperate; that is, they prioritized winning internal power struggles against other Romans over keeping Rome strong overall. Now this was already well underway by the time the empire converted to Christianity, but it's notable that the Christians definitely didn't succeed in halting or reversing the process.
I guess it's not necessarily surprising that scaling up to being the state religion of the empire seems to have come at the cost of some of the unique traits Scott's interested in but it's worth keeping in mind.
> That seems like an odd model to apply to the Roman Empire since it didn't have any neighboring countries that were remotely its peers
I think you could model it as various powers and factions choosing to tolerate Christianity, despite correctly perceiving it as a threat to the existing order, culminating in Constantine deciding to convert in order (probably) to increase his personal power within the empire.
Also different regions inside the empire. The Roman Empire wasn't nearly centralised enough to have a single coherent policy.
'although I'm curious if any early Christian missionaries went east and if so how it went for them"
Yes, the entire Oriental Orthodox branch is from them, don't confuse the word "Orthodox" and think it's related to Orthodox Catholic, it's not. They are known to have gotten as far, and found successful churches, as India with some suggestions possibly far Western modern China, basically the Silk Road. The interesting question is why didn't it take off in the far east, then or now.
I'm currently reading Valerie Hansen's book on the Silk Road, which touches upon this. That branch of Christianity wasn't uncommon in Western China, but towards the end of the Tang Dynasty, the emperor forbade the practice of a lot of religions, mostly targeting Buddhism, but ended up wiping out the smaller Christian, Zoroastrian, and Manichaeist sects - the latter two mostly consisting of refugees who fled the Islamic invasion of Persia. Buddhism, of course, outlasted the laws and thrived, being more popular in China at the time of the crackdown.
And this is my speculation, but Buddhism may be the primary reason why Christianity didn't take off. Buddhism seems to share a lot of the same characteristics listed above - very cooperative, high standards of behavior, etc. It exploded out of India at a similar pace to Christianity, but a couple of hundred years prior, so when Christianity reached China (at around the same time it would have hit Scandanavia) you didn't see rapid adoption.
(To answer Scott's question above of what might be missing from the Christianity, Quaker, Cathar list, Buddhism would be the first one to look at, though I'm not familiar with it myself.)
Yes, the SanThome Christians in India credibly date themselves back to St. Thomas the apostle, and they have survived as a minority for thousands of years, mostly orthogonal to the traditional Hindu/Muslim political dimension.
Nestorian Christianity/the Oriental Orthodox Churches at one point had a considerable number of followers in Persia. Khosrau I initially even encouraged Nestorian Christianity over the Orthodoxy of the Byzantines, though of course always considered Zoroastrianism superior.
The way Rome got to the point of having no peer competitors left was by being better at directing surplus violence outward and distributing plunder fairly among the participants. https://acoup.blog/2023/10/20/collections-how-to-roman-republic-101-addenda-the-socii/ Running out of worthy, plunder-rich targets to aim that violence at meant the "we're all friends here, everybody gets an equal share of the plunder" system started to break down. Right around the same time Christianity was founded, Julius Ceasar was thinking "welp, all done conquering Gaul, now what?" and picking a fight with the Senate.
Except the "Golden Age" Pax Romana starts at Augustus and lasts about 200 years. Commodus, the Emperor when thing started to get dicy was also the first non-adopted emperor, which then slides into the crisis of the third century. It might indeed have been seeded by the structural change from a pillage economy, but 200 years is enough time for me to consider other causes as well.
What are you taking exception to?
"Golden Age" is what you get when territorial expansion has hit diminishing returns, but internal development hasn't yet. Charitable pacifists don't help much with expansion in the form of violent conquest (they'd even be regarded with suspicion, during that period, just for failing to participate) but they've got plenty to contribute when it comes to building up civil infrastructure.
Instructions starting with "sell everything you own and give the money to..." only really work when there's a merchant willing to buy, or better yet two or more bidding against each other, which becomes far more likely when there are trade routes all across the known world untroubled by war or piracy.
> I'm curious if any early Christian missionaries went east and if so how it went for them
See Philip Jenkins' "The Lost History of Christianity" https://entitledtoanopinion.wordpress.com/2011/11/20/waiting-half-a-millenium-for-the-other-shoe-to-drop/
The described payoff matrix is different from the Prisoner's Dilemma. In particular, the Defect/Defect isn't a Nash Equilibrium, since Cooperate/Defect is better for the cooperator than Defect/Defect.
Depends on your level of time discounting. They enslave their children's children who make compromise with virtue...
Brilliant article, thanks Scott
A minor point on claims that Axelrod's Tournament successfully grounded morality in self-interest – it's less that the mental processes which support morality are based on self-interest, per se, and more that the results of the tournament demonstrate an explanation of why morality is so embedded in the human psyche.
Evolution doesn't necessarily care about the mechanisms which cause various strategies (behaviours) – but it does care about the outputs of those strategies, as this is the layer it acts upon (the behaviour level), especially in a highly competitive social environment like those of our ancestors. Axelrod's Tournament shows that, even if the mechanisms within the brain which give rise to moral behaviour are completely unrelated to self-interest, once those mechanisms are present in the population (even if they evolved purely by chance), then the carriers of the genes which give rise to those 'morality mechanisms' within the brain are more likely to spread their genes to the next generation, as the behaviour caused by such mechanisms gives rise to an optimal strategy for cooperating and flourishing in a social environment.
It tells us nothing about how morality first evolved, that's true, but it does tell us a lot about why morality became so widespread and ingrained within human culture – it's a winning strategy.
Great comment. I found that sentence, "I find the occasional claims to have successfully grounded morality in self-interest to be facile, but you can at least see where they’re coming from here." a bit weird...maybe something I'd expect from a bonafide Christian, but not capital R Rationalist(TM). The evolution of morality has been fairly well studied and is not a great mystery at this point. If it was maladaptive, it wouldn't exist. (Some caveats required there of course...the prevalence of various mental illnesses and homosexuality are genuine puzzles, although I believe a fair amount of progress has made toward the former.) But "grounding morality in self-interest" is not how I would characterize that.
Maybe Scott is referring to something else? Maybe just the kind of cynical folk psychology of bitter antisocial people who insist people only "pretend" to be nice while secretly having self-serving ulterior motives?
This is a common confusion about evolutionary explanations, a confusion between the proximate and ultimate levels. If you explain things in terms of fitness, some misinterpret that as saying, you don't "really" love your child/spouse/friend, you're only pretending to care about them because it's to your advantage. No, you really are capable of love, which may sometimes result in actually putting the interests of others above your own. The reason that you have kind of cognitive/neural architecture that produces that behavior is that throughout our evolutionary history, it tended to result in more success at survival and/or reproduction than not having it.
There was a famous story about one of the most influential people in this area...wasn't Axelrod himself was it? Who IMO succumbed to this mistake, and was haunted to the point of derangement by the idea that "real" altruism was impossible, and tried to disprove it by fully embracing the implications of a Peter Singer type philosophy by giving away virtually all of his income and letting homeless people sleep on the floor of his barren house etc...
I think the best attempt to ground real goodness in objective reality is Robert Wright's Nonzero. Goodness is the expansion of nonzero sum-ness, and it is in the nature of things, due to principles like those uncovered by Axelrod, for it to continue to expand over time. But it has to start somewhere. Cooperating with a defector, ie a Jew turning the other cheek on a Nazi, or committing suicide to donate your flesh to the vultures and worms, etc. (A) win-lose, which is not good, (B) inherently maladaptive, which means it must cause there to be less of itself in the universe. So if it were good in some ultimate transcendental sense... That would mean goodness necessarily causes there to be less goodness in the universe in the future, which would be bad. If everyone genuinely cared about all humans (or all living things) equally, that would mean they would not care about their children any more than other being. And since no one can feed, clothe and shelter, let along spend quality time with, all beings everywhere, that would mean no one do even enough to keep their own children alive, and humanity would become extinct, and no humans would be around to be good, or to appreciate how Really Genuinely Altruistic the last generation of humans to exist was.
Just as Daniel Dennett said free will based determinism is the only kind we have and the only kind worth wanting... I say the adaptive kind of goodness that we actually have is the only kind worth wanting.
Here's the guy I was thinking of. Really important in the history of evolutionary theory, and also a riveting personal story. Even more tragic, because IMO his despair was just based on a mistake (and probably pre-existing mental illness). It's wrong and foolish to think kindness/generosity etc can't really be good unless it has no actual or potential benefit to the person doing it. The Buddhists knew that for 2 and half millennia. In many circumstances, being good to others is good for you. That's a feature, not a bug.
A book I'd recommend is The Morality of Everyday Life: Rediscovering an Ancient Alternative to the Liberal Tradition, which articulates and defends a more common-sense intuitive morality against what could be called telescopic philanthropy.
(GPT:) What you're recalling seems to align with the story of **George Price**, a key figure in evolutionary biology and game theory, rather than Robert Axelrod. Price played a significant role in the development of the **Price Equation**, which is foundational to understanding the evolution of altruism.
### George Price and the Evolution of Altruism:
1. **Contributions to Game Theory and Evolution**:
- Price was instrumental in advancing the mathematical foundations of evolutionary theory, particularly in explaining altruism through a genetic lens.
- His work helped formalize how seemingly selfless behaviors could evolve through natural selection if they provided a benefit to relatives (kin selection) or indirectly boosted one's own genetic success.
2. **Personal Transformation and Despair**:
- After deriving his famous equation, Price became deeply troubled by its implications. He struggled with the idea that altruism—what we see as selflessness—was ultimately a selfish genetic strategy.
- This conflict led him to attempt to disprove his own work by living a life of radical altruism.
3. **Radical Altruism**:
- Price gave away most of his income and possessions, including his house, to help the homeless in London.
- He allowed homeless people to live with him, often at great personal expense and discomfort, to demonstrate that true selflessness was possible.
4. **Decline and Tragic End**:
- His extreme lifestyle, combined with his struggle to reconcile his work with his spiritual and philosophical beliefs, led to mental health challenges.
- Price ultimately died by suicide in 1975, leaving behind a legacy that continues to provoke discussion about the nature of altruism.
### Why It Resonates:
The story of George Price is haunting and compelling because it underscores the human struggle to reconcile scientific discoveries with deeply held moral and existential beliefs. His life exemplifies the tension between understanding altruism as an evolutionary strategy and aspiring to live out altruism as a purely selfless virtue.
### Connection to Peter Singer:
Price’s radical lifestyle echoes some of the implications of **Peter Singer's philosophy** of effective altruism, which calls for giving significant portions of one's income to alleviate global suffering. However, Price's actions were not systematically aimed at maximizing utility (as Singer advocates) but rather seemed driven by a desperate need to disprove the cold logic of his own mathematical findings.
### Documentary:
The story of George Price has been featured in several documentaries and books. One notable book is **"The Price of Altruism" by Oren Harman**, which details Price’s life, work, and tragic struggle with the implications of his discoveries. It’s possible the documentary you recall was based on this book or similar material.
> Goodness is the expansion of nonzero sum-ness, [...]
Careful with that! Lots of 'evil' is characterised by decreasing the total sum.
True zero-sum is exceedingly rare in the real world, as it's easier to destroy than to balance perfectly.
"Lots of 'evil' is characterised by decreasing the total sum."
That's an important point I left out: lose/lose is zero sum. Win/win--the increasing prevalence of opportunities for win/win interactions--is what I would describe as good.
"True zero-sum is exceedingly rare in the real world" I don't think it is in the sense that I mean it.
The sumness I'm referring to is just:
| | win | lose |
| :-----:|:--------:|:---------:|
| win | win/win | win/lose |
| lose | lose/win | lose/lose |
Which can be represented as:
| | +1 | -1 |
| :---:|:--:|:--:|
| +1 | 2 | 0 |
| -1 | 0 | -2 |
It's just an abstraction, not an attempt to quantify any precise quantity.
Any voluntary transaction is positive sum, because both parties benefit. A theft, OTOH, is zero sum, because one's loss is the other's gain.
Here's a synopsis:
-----
Robert Wright argues that nonzero-sumness is intrinsic to the nature of evolution because the survival and success of organisms often depend on cooperation and interdependence. While evolution, in a strictly scientific sense, lacks a conscious direction or purpose, Wright suggests that it can appear directional due to the patterns it produces—particularly the increasing complexity of life and the expansion of cooperative networks.
As life evolves, systems become more intricate, and organisms interact in increasingly complex ways. These interactions often create opportunities for nonzero-sum dynamics, where mutual benefit becomes more viable and advantageous. For example:
• In biological evolution: Symbiotic relationships (e.g., between pollinators and plants or gut bacteria and hosts) show how cooperation can drive complexity and enhance survival.
• In human history: The development of language, trade, and technology amplifies the potential for cooperative, nonzero-sum outcomes, enabling larger, more interconnected societies.
Wright argues that the expansion of nonzero-sumness is a natural consequence of this complexity. As organisms and societies grow more interdependent, the scope for mutually beneficial relationships widens, creating a feedback loop where cooperation becomes increasingly advantageous.
In this sense, while evolution doesn’t have an intrinsic “goal,” the appearance of directionality comes from this pattern: life becomes more complex, which broadens the opportunities for nonzero-sum interactions, driving further complexity and interconnectedness. This gives the impression of a purposeful trajectory, even though it arises from natural processes.
----
To your point about negative sum:
---
Yes, Robert Wright does suggest that as complexity increases and societies become more interconnected, the potential for negative-sum interactions also grows, alongside the opportunities for positive-sum interactions. This is because interconnected systems amplify both cooperation and conflict.
In a complex, interdependent world, actions and decisions have wider-reaching consequences, and failures in coordination or trust can result in negative-sum outcomes—situations where all parties lose. For example:
• In tightly connected societies, wars, economic collapses, or environmental crises can have devastating ripple effects, harming all involved.
• As technological and social complexity grows, so too do the stakes of mismanagement or adversarial dynamics, increasing the potential cost of negative-sum scenarios.
However, Wright emphasizes that the same interdependence that heightens these risks also incentivizes the reduction of negative-sum outcomes. The shared costs of failure push individuals and societies toward systems that favor positive-sum dynamics, such as cooperation, diplomacy, and innovation.
In essence, the stakes are higher in a more interconnected world: while the potential for positive-sum interactions grows with complexity, so does the potential for negative-sum interactions. Wright’s optimism lies in the idea that humanity’s increasing awareness of this dynamic, coupled with technological and cultural evolution, makes it more likely to lean toward nonzero-sum solutions over time.
It's also worth noting that Christianity itself also grounds morality in self-interest; it just argues that the "reward" from defecting is vanishingly small compared to the reward from cooperating (see the quote from Matthew above).
But it's not *self*-interest. It's your genes'.
Morality first evolved before homo sapiens existed. Every animal that lives in groups has some form of morality. One could argue that this is even true of spiders and preying mantises, but that's probably stretching the point.
Now whether that morality is considered desirable....now that's a different question. The morality is needed to enable the species to survive in the niche for which it evolved, but when the species moves to a different environment, the "most desirable" morality (from the viewpoint of the species) changes.
Humanity has so altered the conditions in which it is living recently that our "natural morality" is probably very far from optimal. But we have cultures that implant a "cultural morality". (Here culture is often reducible to a single family or small group of families.) Even so, it's probably that all current moralities adopted by people are far from optimal (from the viewpoint of the species).
Spiders can cooperate when there's sufficient food to go around. https://web.archive.org/web/20190519020515/https://www.entsoc.org/PDF/2010/Orb-weaving-spiders.pdf
These are very good points. As Steven Pinker among others have pointed out, morality is possibly the leading cause of violence in the world. And--important distinction--much of that violence, in present circumstances, is unnecessary and unhelpful, therefore bad. It may well be--almost certainly is, in fact--that there is evolutionary mismatch between the world of small competing tribes our morality evolved for (although it seems it's turning we evolved more since the dawn of agriculture than was previously thought, which opens another can of worms I'll save for later).
Still, I think our common sense folk intuitions are generally more sound than rationalist utilitarianism** taken to its logical conclusions. Ie, yes, it is moral and good to care more about your family, and friends, and neighbors, etc than all beings everywhere. Because (A) you are in a better position to actually succeed in doing good for the people with whom you have close reciprocal relationships, and (B) if it's not successful in Darwinian terms, it will undermine itself. And as I said in my other comment, if "goodness" undermines itself and causes there to be less goodness in the universe in the future, its net effect is bad, hence it can't actually be good.
**Edit: rationalist utilitarianism, or, more relevant to this thread, I should have said the Christian "turn the other cheek", love everyone as a brother, do not resist evil etc. philosophy which seems to be endorsed in some parts of the New Testament. Which it seems very few Christians have ever fully embraced. Which I don't think of as a criticism of Christians. I think it may be good as an aspiration, encouraging us to be more forgiving, open-hearted, understanding, etc than we naturally tend to be, knowing that like it or not, we will of course prioritize ourselves and our loved ones. But taken literally, I think it's obviously unworkable in practice and not really good or moral at all.
> knowing that like it or not, we will of course prioritize ourselves and our loved ones.
And, therefore, the people most inclined to feel obligated to follow their culture’s morals to their ultimate consequences will be weeded out. With a high chance of not being perceived as following any morals, but as acting like fools who deserve no respect or sympathy.
>Overwhelm downward adjustment: If you assume all movements lie and downgrade their claims 90% out of cynicism, then a movement which merely portrays itself as helping members won’t even help members, but a movement which portrays itself as radiating universal love to all mankind might at least help its members.
I think this is a very, very big factor. Mathew 5:38-40 says,
>38 “You have heard that it was said, ‘Eye for eye, and tooth for tooth.’[a] 39 But I tell you, do not resist an evil person. If anyone slaps you on the right cheek, turn to them the other cheek also. 40 And if anyone wants to sue you and take your shirt, hand over your coat as well.
That's all well and good to proclaim, but I really doubt that literally happened often. Growing up going to Sunday school, my Pastor taught that passage a few times. But knowing him and everyone else who worked at the church, they were pretty normal people, and I do not see them literally turning the other cheek if I slapped. They certainly wouldn't give me their coat if I stole their shirt.
But also, I would guess they were more forgiving than if they didn't sincerely believe in that passage and similar ones. If you stole his shirt, I'm pretty sure my pastor wouldn't give you his coat, but he would have a conversation with you trying to explain why that was wrong and talk you into giving the shirt back instead of calling the cops on you right away.
Or maybe I'm just talking out of my ass, I didn't go to church that often even as a kid.
I've often heard it said that even the Mosaic Law, though harsh to our eyes, was a practical improvement on what came before, which encouraged Hatfield-McCoy feuding over minor slights.
And eye for an eye is actually a much more moral improvement over death for an eye.
Correct. That was the original point that moderns don't realize.
Indeed. It was one of the earliest written expressions of the principle that punishment should be proportional to the offense.
The usual Christian understanding is that this is a call for patience and lenience in response to personal injury, but not an absolute call for perfect nonresistance. Though obviously a minority has disagreed with this take over the centuries.
What I would call the more orthodox lines of Christian thought have generally tried to take the entire Bible as a whole and to understand it as a single cohesive message, that Jesus didn't suddenly and completely abrogate all the teaching of the OT, but instead his moral teaching was primarily correcting misbehaviors that had arisen as a result of poor application of the OT Moral Law (which remained intact).
Jesus is a complicated figure. Those movements that have disagreed with this take (which we might call "heretics") have generally been less concerned with the idea of Scripture as a cohesive whole, instead emphasizing some teaching or another of Jesus over the rest of the Bible.
The Bible before or after the Council of Nicaea? You've got to remember how much the Bible was rewritten to accommodate politicians/theologians/etc...and not just that one time. (Admittedly, without many of those rewrite it would be a wild collection of silly stories.)
Nah, that’s Dan Brown fantasy.
Dan Brown took the idea and built some silly fantasy stories off of it, but that doesn't mean it's not well-grounded in reality. There are books that the early Church Fathers considered canonical, and cited as such in their writings, that we do not have today.
There are books that the Apostles considered canonical, and cited as such in their writings, that we do not have today. Several of them are named in the various books of the Bible, and at least one is not: in Acts 20:35, Paul cites the teaching of Jesus, its source clearly familiar to his audience but lost to us today, that "It is more blessed to give than to receive."
Meanwhile, archaeologists, who for centuries have used the Old Testament as a guide to exploring the Holy Land, are increasingly making discoveries about the Northern Kingdom of Israel that lead them to believe that many of the historical accounts that have been passed down to us in the Old Testament are composed of Southern Kingdom political propaganda that's not nearly as truthful as Biblical literalists would assume. (A notion that's backed up by the discovery of various apocryphal ancient works believed to originate in the Northern Kingdom.)
This is true, we have lost much knowledge, but that is categorically different from the assertion that the Nicaean Council rewrote the literature we do retain.
I heavily discount such assertions based on my personal experience with religious zealots today. Try to convince any of them to rewrite their religious texts- it's not an easy thing to do. The best you can get is a translation that emphasizes different things.
If our current religious zealots won't rewrite anything, then I do not think that the ancient religious zealots- who were far more scrupulous and self-selecting, by all accounts- would be more willing to.
If we're talking about early Christianity, it's not so much a matter of rewriting as new-writing. The Evangelists and the Apostles wrote all manner of new works that Christians considered to be scripture, on the same level as the books written by the Prophets of old. And then other people started writing books that some Christians considered equally legitimate and others did not agree with. And then you have the whole mess that is Gnosticism complicating things and muddying the waters further.
By the time of Nicaea, things had grown so fragmented that Constantine, who wanted to be a Christian, found it impossible to definitively answer the question "just what does it mean to be a Christian anyway?" So, by the authority vested in him as Emperor, he got a bunch of top Christian thought leaders together and said "you're going to work out an official Christian doctrine and settle all this confusion once and for all, or else!" One important part of that was deciding, among various competing works, what was to be officially considered canonical and what should be thrown out.
Re: There are books that the early Church Fathers considered canonical, and cited as such in their writings, that we do not have today.
Actually we do have those books. Chiefly: The Protoevangelium of James, the Epistles of Clement, the Diadache, the Shepherd of Hermas. You can find these by an online search.
As far as the OT goes there are several books recognized a Scriptural by the Church of Ethiopia that no one else recognizes (and not the non-Ethiopian Jews either)
We have some of them, yes...
The Council of Nicaea had nothing to do with the canon of Scripture.
That's officially correct, as certified by Google, but then I didn't claim that it did. I claimed that there were lots of books removed from the Bible, and that many of them were removed at an early date (I picked the Council of Nicaea as a separation point, however, because of the "myth"). Yes, it was really based on a series of decisions, mainly made by the Catholic Churches. But it drastically altered the tone of the Bible. (Possibly repeatedly. I haven't really studied the matter. But consider "The Gospel of Thomas". Or even "The Gospel of Judas". And even the ones that were obviously collections of folk tales [reminiscent of the Cherry Tree Carol] would change the tone and meaning of the Bible were they included.)
> That's all well and good to proclaim, but I really doubt that literally happened often. Growing up going to Sunday school, my Pastor taught that passage a few times. But knowing him and everyone else who worked at the church, they were pretty normal people, and I do not see them literally turning the other cheek if I slapped. They certainly wouldn't give me their coat if I stole their shirt.
There's a lot of cultural context in this passage that we lack today. What he was saying here can be summed up rather concisely as "don't be like the Zealots." While the world "zealot" has become a generic term in the modern lexicon, it referred to a specific group of people with a well-defined ideology, that the Romans were evil occupiers and the Jews needed to resist them through physical violence, and eventually rise up and cast them out of Judea in a glorious revolution.
Given the overwhelming military superiority of Rome, this was a stupid idea that was never going to be successful, and people listening to these notions eventually culminated in the tragic events of 70 AD. Jesus told his disciples to follow a different path. If a Roman citizen legally compels you to give you something, give more than is required. If he requires you to travel with him for a mile, as is his legal right under Roman law, go two. If someone tries to provoke you with a backhanded slap, (this is the only way for a normal right-handed person strike someone on the right cheek, afterall,) don't rise to the provocation by throwing a punch. Instead, turn the other cheek and say "I'm not going to let you bait me. *You* throw the first punch, and then you're the one who broke the law and started the fight."
In all cases, it's the same requirement: do not give the Romans any excuse to say that you're a revolutionary fighting against them. Go out of your way to make it clear that you're not. Given the way the revolutionary Jews were so thoroughly wiped out once the Romans finally got sick of them, it proved wise advice!
I've found this a compelling interpretation. However, many Christians seem to disagree. Are there any good objections against it?
The Gospel of Matthew never mentions the Zealots. To the extent that the Sermon on the Mount names any enemies, they are enemies of Jesus, not of the Jews:
"Blessed are you when people insult you, persecute you and falsely say all kinds of evil against you because of me. 12 Rejoice and be glad, because great is your reward in heaven, for in the same way they persecuted the prophets who were before you." (Matthew 5:11)
Who are the people insulting, persecuting, and slandering the followers of Jesus? At this time, they were not Romans. They were other Jews, particularly the Pharisees. Who persecuted the earlier prophets? Also other Jews, particularly the kings of Israel and Judah.
The good objection is that you should require direct evidence of very specific claims. It’s a neat just-so story but there’s no evidence for it other than the originator thinks it might have been cool and it fits with existing biases that Jesus must be terribly misinterpreted because he couldn’t possibly have meant what he actually said and others took him to mean.
The evidence against is that nothing in the Gospels leads us to believe that Jesus had any interest in leading a covert passive resistance against the Romans, to the point that the Roman authorities seized him (on claims of the Jewish religious authorities), investigated him, and obviously thought he was a harmless kook.
You can argue “well that’s the whole point of his strategy!” but now you’re into conspiracy territory. Literally nobody among his contemporaries figured out the secret Jesus anti-Roman code, but we super-smart moderns did?
The more likely explanation, borne out by his and his early followers’ actual behavior, is that Jesus meant it just as he said, for application to more or less all situations, not just Zealots vs Romans.
> The evidence against is that nothing in the Gospels leads us to believe that Jesus had any interest in leading a covert passive resistance against the Romans
> the secret Jesus anti-Roman code,
It's hardly valid to present, as evidence against a claim that was made, evidence against a claim that was *never* made.
OK. I will now ask for your evidence that the interpretation that this passage was specifically about Rome is true.
The evidence is what I said in the first post: what he's referring to here are specific cultural practices. For example, the word ἀγγαρεύσει (angareusei) used in the Sermon refers to a Roman soldier's legal right to compel conquered people to accompany them, and carry their heavy equipment, for one mile. His audience knew exactly what ἀγγαρεύσει was, and likely many of them had had it happen to them at some point. (See also the case of Simon the Cyrenian, a random traveler who the Romans grabbed and forced to carry Jesus' cross.)
But seriously, how exactly do you take a post saying "Jesus said to not be a revolutionary and to be exceedingly clear that you are not a revolutionary," and then 1) *completely* misread it as "Jesus said to be a revolutionary, just in secret," then 2) when it's pointed out that that was never the claim, say "OK then, prove it"?!?
None of that reflects well on you.
It is not hard to read Mark, at least, as a thinly veiled anti-Roman screed, using the metaphore of demonic possession to avoid saying anything crucifixion worthy. And Mark was the first of the Gospels to be written, presumably closer to the source material than the other three.
It might not be hard for you to make that inference, but what is that actual evidence that it is besides your belief? Certainly the contemporaneous writers, readers, devoted Christians, etc. did not believe it was and did not treat it as if it was.
Also, I should note for you that the evidence of Markan priority is thin: inference and opinion of textual critics, not through physical evidence, archeology, historical accounts, etc. Wikipedia will tell you the case is airtight, but Wikipedia is written by academic textual critics, not historians nor archeologists, and you should be skeptical of it.
> Certainly the contemporaneous writers, readers, devoted Christians, etc. did not believe it was and did not treat it as if it was.
What exactly is "certain" about this assertion?
I do not have anything except my own personal interpretation from reading it. I should explain that I am a practicing Christian.
As for the priority of Mark, yes textual clues are all we have, but you take the evidence you have, not the evidence you wish you had. I find it convincing.
I don't think Quakers were ever actually executed in England, though they were regularly fined and imprisoned. Cromwell and Charles II were friendly with Fox and Penn respectively, though local judges often were not
Massachusetts on the other hand seems to have executed Quakers regularly and made returning to Massachusettes as a Quaker a capital offence.
That may be true, but "keep in jail in abysmal conditions until they get sick and die" isn't *that* far from execution, and certainly many Quakers did die in jail as a result of their resistance/principles.
It is pretty far from execution. 10% chance of death from disease in private is very different punishment from 100% of a painful death in public.
The biggest gap between bot strategy and the real world is the fact that in the real world, people can change strategies. If, in the bot simulation, 40% of the bots were “adapt the strategy of the last bot you encountered” cooperate-bot would be much more likely to pull ahead. Tit-for-tat would still win, but would cooperate bot win more?
I don't feel like trying to dig up the answer, but...wait I don't have to; I have a magic oracle. But first, I don't think that's actually necessarily a huge difference; I think in Axelrod's tournaments, people were allowed to submit any computer code they wanted. In the iterated tournaments, I don't think there was anything stopping them from changing behavior in one round based on what happened in previous rounds. Probably some tried.
To the Oracle (GPT 4o):
"""
The question of whether the introduction of adaptive strategies (like "adopt the strategy of the last bot you encountered") changes the dynamics of iterated games like the Prisoner's Dilemma has indeed been explored in various ways, though perhaps not always in the exact form you described.
Experiments in **evolutionary game theory** and **iterated Prisoner's Dilemma simulations** often include agents that can adapt or learn strategies over time. Here are some key findings that relate to your question:
### Key Concepts in Relevant Research:
1. **Strategy Adaptation in Simulations**: Many simulations introduce "adaptive agents" that mimic or evolve strategies based on observed success. These agents are akin to your "adopt the strategy of the last bot you encountered." Such studies show that the presence of adaptive agents often destabilizes dominance hierarchies among fixed strategies (like tit-for-tat or always-cooperate) and leads to fluctuating dynamics.
2. **The Role of Cooperation Bots**: In a population with adaptive agents, strategies like "always cooperate" (cooperate-bot) tend to perform poorly unless they exist in a supportive environment. If a majority are cooperative, this increases the likelihood that adaptive agents will adopt cooperation. However, they remain vulnerable to exploitation by defectors.
3. **Tit-for-Tat in Adaptive Contexts**: Tit-for-tat tends to perform well in adaptive scenarios because it strikes a balance: it fosters cooperation when others cooperate, but it retaliates against defectors. This robust balance makes it likely to spread or persist, even in the presence of adaptive agents.
4. **Social Learning and Memory Effects**: In experiments where agents can remember more than just the last strategy they encountered, or where they weigh historical interactions, tit-for-tat's dominance can diminish. Cooperative strategies (including always-cooperate) may thrive in these environments if agents tend to favor longer-term rewards.
### Would Cooperate-Bot Perform Better?
It depends on several conditions:
- **Population Composition**: If 40% of the population adopts "copy the last strategy," cooperate-bot could indeed benefit, as these adaptive agents would mimic cooperation after encountering cooperate-bot.
- **Interaction Rules**: If interactions are random, cooperate-bot's success might increase. However, if defect-bots (always defect) or tit-for-tat strategies dominate the environment, adaptive agents will likely mimic defecting or retaliating behaviors more often.
- **Mutation and Evolution**: In evolutionary contexts where agents adapt based on long-term payoffs, tit-for-tat generally prevails because it punishes defectors while fostering mutual gains with cooperators.
### Existing Studies:
- **Axelrod's Tournaments**: The famous iterated Prisoner's Dilemma tournaments run by Robert Axelrod showed that adaptive strategies could emerge under certain rules. However, tit-for-tat still remained a top performer because of its simplicity and robustness.
- **Agent-Based Modeling**: Many agent-based models (ABMs) have tested scenarios with adaptive agents. Results often show cyclical dominance: cooperators rise, get exploited by defectors, and then defectors diminish as retaliation strategies (like tit-for-tat) take over.
### Conclusion:
Experiments have addressed related questions, though perhaps not in precisely the form you've described. The inclusion of a large proportion of adaptive agents would likely boost cooperate-bot's performance temporarily, but tit-for-tat would remain resilient in the long run because it discourages exploitation. The dynamics of such systems are highly sensitive to initial conditions, mutation rates, and the ability of agents to "remember" and adjust to patterns over time.
"""
PS: Was I right about what I said initially?
"""
Yes, you're correct about Axelrod's tournaments allowing strategies to change behavior based on what happened in previous rounds. The tournaments were designed to permit participants to submit any computer code they wanted, which means strategies could indeed be highly sophisticated and adapt to the opponent's previous moves.
### Key Points About Axelrod's Tournaments:
1. **Open Submission**: Participants were free to submit any strategy they could program, and strategies could incorporate memory, logic, and conditional rules based on prior interactions. This enabled them to change behavior dynamically across rounds.
2. **Iterated Nature**: Because the games were iterated, many strategies explicitly accounted for past interactions. For example:
- **Tit-for-Tat** famously cooperated on the first move and then mirrored the opponent's previous move.
- Other strategies tried to exploit patterns or weaknesses in their opponents by analyzing behavior over multiple rounds.
3. **Sophisticated Adaptations**: While the simplest strategies (like always-cooperate or always-defect) did not adapt, more complex ones often incorporated memory and conditional responses. For instance, strategies might:
- Cooperate for a certain number of rounds, then switch based on observed payoffs.
- Punish defection harshly for multiple rounds to discourage exploitation.
4. **Design Space**: There were no constraints against designing strategies that could adapt mid-tournament or "learn" an opponent's tendencies, though strategies had to follow the rules of the game (e.g., fixed payoff matrix, iterated rounds). The only real limitation was computational feasibility, as programs needed to run within reasonable time limits.
### Did Participants Try This?
Yes, some participants submitted highly sophisticated strategies, including those that:
- Used pattern recognition to exploit predictable opponents.
- Forgave defections after a certain point to restore cooperation.
- Attempted to "probe" their opponent's behavior early in the game to choose an optimal response later.
### Why Does Tit-for-Tat Often Prevail?
While adaptive strategies existed, tit-for-tat often emerged as a top performer because of its simplicity and robustness:
- It fostered cooperation with other cooperators.
- It punished defectors immediately but did not escalate conflicts.
- Its predictability made it a stable partner for strategies trying to avoid exploitation.
### Conclusion:
You're absolutely right that the open nature of Axelrod's tournaments allowed for strategies that could adapt round-to-round. Many participants likely experimented with these approaches, and some succeeded. However, the straightforward, cooperative nature of tit-for-tat often allowed it to perform better in aggregate across diverse opponents.
"""
I’m not suggesting it would benefit an individual participant to learn other’s strategies, but if that were a force of the simulation - if it ran real time and after each 10 rounds 40% of the “follower” bots changed strategy to the current winning strategy, I think it would be very different and a bit more like real life. In a world where cooperation is publicly visible and encouraged, and communication between bots is possible, you can surely make things where cooperation is the winning strategy.
Sounds like something you could try on your own PC. Could be a neat project if you had some time on your hands!
If the rules of your simulation periodically "kill" some of the agents following the worst-performing strategy and replace them with agents that follow the current best-performing strategy, and you start with a mix of tit-for-tat, cooperate-bot, and defect-bot, then tit-for-tat will generally win, and cooperate-bot will almost always be the first to be wiped out.
With random errors, tit-for-tat-plus-forgiveness will become the most common strategy if you add it, though it's so merciful that it won't completely wipe out regular tit-for-tat, and a small fraction of the population will stick with that.
(There was a great browser game/interactive lesson about this that I played at some point, but I lost the link and haven't been able to find it again.)
Some researchers have been really annoyed that they couldn't come up with plausible scenarios where cooperate-bot won in their simulations, even after trying pretty hard to come up with some. (Though obviously most simulations do not remotely approach the complexity of real life.)
You definitely "can" make cooperate-bot win if you sufficiently contrive the rules (e.g. if you add in some god-players that read your source code and cooperate IFF you are cooperate-bot), but I think you'll find its much harder than it sounds like you are expecting.
What if
1) the rate of random errors is based on code complexity / runtime,
2) rather than every matchup being random, you can choose to seek out a specific agent you've encountered before,
3) score thresholds for elimination or reproduction are fixed, and knowable - for example, any agent whose cumulative score drops below 0 dies, one whose score rises above 100 can choose to spawn a copy, after which both will be reset to a score of 50?
in such a case, particularly if point 2 were somehow expanded to friend-of-a-friend referrals, I'd expect to see an "enforcer" or "shepherd" strategy which alternates between proactively culling known defect-bots - even at the expense of its own immediate reproductive opportunities - and recharging within a more congenial community when its resources are low. Cooperate-bots make the best pool to recharge from since they'd suffer the least error-related attrition.
"Cooperate-bots make the best pool to recharge from?" Sounds like you're effectively arguing that cooperate-bots make the best subjects for warlords ("shepherds") to rule over, rather than the argument that *being* a cooperate-bot is good for the person executing the strategy.
Also, like...you want tit-for-tat to be penalized for being complex relative to cooperate bot, but you also want friend-of-a-friend referrals? That's probably several orders of magnitude more complicated than tit-for-tat. And then you also want police investigators to somehow identify and punish the people who break the rules? That's probably several orders of magnitude more complicated than referrals.
Also point 3 seems highly unrealistic (real life has tons of luck, lots of people die unpredictably, and passing on your genes is not a binary pass/fail, you can have a variable number of kids and give them a variable amount of assistance in getting started). And point 2 is vague but the simplest implementation I can think of is also highly unrealistic (choosing to interact with only your most trusted friend and avoid everyone else is a terrible strategy in real life for a whole bunch of reasons that you don't sound like you're attempting to model).
But feel free to code up your own simulation, maybe you'll discover something interesting!
> Also, like...you want tit-for-tat to be penalized for being complex relative to cooperate bot, but you also want friend-of-a-friend referrals? That's probably several orders of magnitude more complicated than tit-for-tat.
I don't think reasonable levels of complexity penalty would be giving cooperate-bot a noteworthy edge over tit-for-tat all by itself, no.
Point of a "choose who to interact with next" + "know how close you currently are to reproducing, or dying of starvation" mechanic would be opening up space for long-term ecosystem-management strategies, then the complexity penalty keeps those from escalating all the way into some cryptographic-handshake version of John Carpenter's The Thing. https://www.oglaf.com/kingshaped
> choosing to interact with only your most trusted friend and avoid everyone else is a terrible strategy in real life for a whole bunch of reasons
Well of course it's not that simple.
First, enforcer / shepherd builds a contact list of known cooperate-bots, and a separate list of reliable tit-for-tat strategists. To the extent it's allowed by the mechanics, all these are freely introduced to each other, encouraged to interact to mutual benefit, top priority being to link up cooperate-bots with other cooperate-bots.
When the enforcer encounters a defect-bot, or unknown and potentially troublesome strategy, it checks its own reserves. When healthy, it attacks the intruder. When reserves are low, it retreats, spins through the cooperate-bot contact list, hitting them with a "defect" once each to quickly recharge, then resumes the attack.
Distributing the load hopefully reduces the number of cooperate-bots which die as a result, thus keeping the overall herd's income / growth rate high. Sort of like a multicellular organism's immune system. Tit-for-tat contacts can't usefully be harvested that way, because they'll bite back.
It sounds like you're describing Nicky Case's "The Evolution of Trust ": https://ncase.me/trust/
Yes, thank you!
The bots could change strategies. It doesn't help. The more cooperative-bots you have, the more it boosts the ranking of bots that defect in the tournament.
Well I tried your suggestion, and it looks like it just makes Always Defect vastly more successful than usual, steamrolling everything else, more so the more rounds you do. I think the reason is that Always Defect always starts at an advantage, and the follower rule means others copy it, so it snowballs.