Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Of course this would get upvoted on HN. A blog post by an impractical contrarian outsider who's seriously suggesting that the entire world replace advanced weaponry with stuff that got phased out hundreds of years ago, because it would be "low-tech and sustainable". He even suggests that it's unrealistic not because it's the opposite of the entire point of weapons (to kill more things faster and easier), but because it would require "global cooperation" and "uninventing things".

The Onion couldn't come up with this. It's so embarrassing. And yet, as I type this, dozens of people are taking this premise seriously and debating it on its merits. Humans really are doomed.



> The Onion couldn't come up with this. It's so embarrassing. And yet, as I type this, dozens of people are taking this premise seriously and debating it on its merits. Humans really are doomed.

Personally, I find discussion about hypothetical, albeit completely impractical, ways we could avoid killing each other a much better use of time than a lot of other topics, like discussing ways we could kill each other more efficiently.


A lot more people, as a percentage, died during medieval wars than do today in modern wars. Read about the crusades sometime. They would lay siege to cities and massacre the entire population. Not a single survivor.

They didn’t need advanced weaponry to kill tens of thousands of people. A few thousand soldiers would go through a city, house-by-house, and put every single man, woman, and child to the sword. They then looted what they wanted and burned the house to the ground.

Thousands more died during the siege, long before any soldiers reached them, due to disease and starvation from overcrowding and dwindling food supplies. It was absolutely horrific and it happened mainly because swords, spears, and bows were useless against city walls so long sieges were the rule.

The massacres that followed could be chalked up to deep resentment on the part of the attacking soldiers for the defenders holding out so long. I think another part of it is simply that the commanders of the day had less power over their soldiers, so looting and pillaging was part of the bargain to encourage them to fight.


Looting was in large part the payment for soldiering. Whomever raised the army might pay a bit regularly for maintenance, but all the real money was in the spoils, and everyone involved knew it. Many is the army that failed to destroy a defeated foe because the victorious soldiers mostly wanted to loot the enemy camp.


I'd be curious about data supporting this claim. Browsing through https://en.wikipedia.org/wiki/List_of_anthropogenic_disaster... seems to suggest otherwise, but its also not relative to population.


Erm, no, I don't think so. Can you give a source for that claim?

Medieval wars were in general shorter, more regional, used much smaller forces and affected the civilian population less than in later ages. Of course it is very difficult to give any numbers, but various infection diseases, childbirth and infant death were major causes of death. (E.g. women life expectancy overtook that of men only in the 19th century.) Also at least in the 13th and 14th century in Europe main reasons for a significant temporary population decline - during a time of general population growth - were black death and hunger, the latter being partially caused by limitations to agriculture technology and available land.


Medival wars in Europe were smaller and shorter because of weakness and fragmentation of European states at that time. The Mongol conquest cost the lives of 50 million people and left huge swathes of Europe and Asia uninhabited.


The Mongol siege and destruction of Baghdad. An estimation of 200k to 2M civilians killed in one week.

https://en.wikipedia.org/wiki/Siege_of_Baghdad_(1258)


a bit over the top, but the crusades could be quite bloody https://en.wikipedia.org/wiki/Massacre_at_Ayyadieh

as a general rule though this was also because it was Christians killing non-Christians etc.


>affected the civilian population less than in later ages

lol. lmao, even.

Just an army moving through a region - just walking, scouting and "foraging" (this doesn't mean bushes, this means looting from rural villages) - guarantees a famine for those involved. Only the invention of the railway prevented (no, lessened) this.


If you want to see some really brutal wars, you need to go back from the Middle Ages into Antiquity.

The ancient empires were able to gather armies much larger than any medieval king could, and, at the same time, people had few qualms about genocide. There weren't even religious reasons not to commit one, much less any chivalry codes etc.

The Romans in particular had the principle of "aries murum attigit", which meant that a city had time to surrender until the first battering ram touched its walls. After that, the captives would be put to the sword.


Agree, but then humanity spent a considerble effort since the dawn of time exactly on that: developing ways to kill things faster and safer for oneself.

It is, sadly, part of our nature. And because of that I prefer hypotheticals that actually have some basis in reality.

By the way so, during the time of bow, arrow, sword, spear and shield, we still managed to kill hundreds of thousands of people.


> part of our nature.

This comes up now and then. But it doesn't hold water, because if true then people would on average be killing people roughly at the same rate, regardless of epoch, geographical area or society. But nations and tribes that were constantly warring have over time become good, peaceful neighbors, while the opposite is the exception. In spite of the impression one might get from the daily news, person-on-person violence is at an historical low.

Why is this explanation still popular, despite the obvious falsity of it? Good question, not sure. Maybe for now suffice to say that it isn't true whenever it pops up. Eventually we might start scrutinizing the real reasons (or lack thereof) for going to war, and the mindless slaughter it necessarily entails.


It's part of the nature of a bomb for it to blow up, but you should note a bomb spends 99.999999% of it's life not blowing up, and many bombs never blow up before they are disassembled.

There is no obvious falisty here. Any intelligent entity has the ability to kill, it's part of the nature of intelligence. The question you're asking, which is different is "What drives more killing at particular times, and can it be avoided".

And here is the answer you don't want to hear.

No.

This is just game theory in practice. Every player has the option of cooperating or competing. If they decide to compete it is likely they will choose to escalate in order to win.


> This is just game theory in practice.

Maybe the model didn't take all parameters into account - a primary characteristic of all models? Such arguments may first seem like a plausible description of the world only to become arguments for a course of action.

In short I would not rely too much on a theory that says "we must go to war and kill thousands if not millions of civilians otherwise the opponent will" as a lame excuse to leave other options unexplored. We're getting close to that phase in history where other such options seem ever more realistic IMO.


You're seemingly shifting your point, which originally was that humans at times and for varied reasons kill each other. In which I was replying that this is innate and not a removeable part of the system. You may be able to quell said behavior, but you can never be sure your system is stable, or if your system is only quasistable. Any new variable added to your system at any point in the future can have a non-linear effect which is not predictable/non-computable before hand.


I think we can agree to disagree at this point. This is why I think so:

The argument that I originally objected to was that it is in "our nature" to develop ever more lethal means of warfare. Your objection to this is that (according to game theory) there is no way to stop this, so although it might not be in our genetic nature then it is in the "nature" of game theory (with rational players that might as well be robots). My objection to this stands: you can't know all the parameters of the game even if it probably is possible to retrofit the theory to past events so that it looks all encompassing. In systems as complex as total world history you simply can not pin down all the parameters sufficiently so that you can foresee all possible outcomes - no matter how much one might wish for such powers. It is a temptation that many, perhaps mostly mathematicians, have succumbed to: replacing the world with the model, or the terrain with the map if you will.

So, even this kind of "nature" is false, even if less obviously so. I don't expect you to agree, but I think we can agree that this is the point of contention, and since our beliefs are simply different (and I don't expect you to be swayed as a consequence of our little exchange here) there is no use arguing further.


But you're kind of doing the same thing...

You're simply submitting that a model exists where we can all get along, and handwaving why as too complex. Your model is as simply unpredictable as my model is. But with any models where there is a large amount of unpredictability, is you look at model probability. What is more probable? That everyone suddenly starts getting along with no defectors. Or that some groups play along waiting for the right time to defect?


I'm not predicting, or saying what is probable, I'm saying that something is not impossible, and pointing out a direction that might be possible and something to strive for. You can call that a model if you like. As a contrast models that use "nature" as a concept are deterministic: referring to human "nature" in terms of warfare is another way of saying that peace is impossible, and therefore meaningless to strive for. It is reflected in the language you used: "(...) can it be avoided [?] (...) And here is the answer you don't want to hear. No. ", i.e. very assertive, as if proven.

Now in your latest comment, you start talking about various probabilities instead, which speaks a different language than the one used previously. High probability based on historical and, as such different, circumstances is not same as predicting with absolute certainty what will happen, as I'm sure you agree.

Which brings me to reiterate what I said at the start of this thread: It is not part of our unchangeable nature to be ever more destructive towards fellow man. And, I might add: to insist that it is (in our nature) is not only false, it becomes an argument for not exploring the sorely needed avenues for peace that exist.


> developing ways to kill things faster and safer for oneself. It is, sadly, part of our nature.

If you really feel that way, please either (1) go vegan, or (2) start killing the animals you eat and wear with your bare hands.


case in point, I suppose.

Ok. Fine. I'm in. Let's discuss it.

Do you think we're better off using ai to decide the arms treaties, or should we hold off for quantum computing?


> Of course this would get upvoted on HN

I agree for less pessimistic reasons. The Hacker News crowd prefers intellectually stimulating conversation. Playing around with new ideas. Considering unexplored alternatives.

A huge pillar of this community is exploring ideas. Engaging in that doesn't mean that any of us think this could happen in the real world.


I've revised my own opinion of this. Its premise may be as absurd as you suggest, but I think that's just a framing device to compare the two weapons. Normally, they're incomparable in the same way that comparing a packmule to a container ship is. Both of those might be used to transport goods, but writing up a comparison to the two is difficult to do without some way to frame it in a way that could engage the reader.

And whatever other effect he might have on you, he engaged you enough to leave this comment. I think it worked.


A much simpler solution is to just ensure that politicians' children (and politicians themselves on rotation, perhaps) are the very first to the front lines in any given war they choose to conduct.


Well, considering that the ages when this was expected of political leaders were also far more violent and had far more endemic warfare than modern times, maybe this actually isn't a good idea.


the led the armies, but they also led the country. all of the risk and all of the reward.


Have you ever looked at how many American politicians are veterans? How many of them have children who are in the armed forces? It's certainly not universal, but they do have children who would be fighting their wars and they have fought in the last generation's wars.


I looked it up. 18% of congress is a veteran. 6.2% of the general population is.


These days...I'd put their leading financial backers (& children) on the front lines instead.


It is not that simple. For example, Hitler had no kids. And Stalin did, but he treated them indifferently and refused to save his son Yakov from German captivity.

Netanyahu's brother Yoni died in the Entebbe raid, didn't seem to make Binyamin a pacifist.


They won't. That's the problem, politicians and dictators don't see war the same way as civilians do.


Impractical ideas can be entertaining, even useful, thought exercises. Valid ideas and discussions can occur, even if the starting point is out of reach.


>Of course this would get upvoted on HN

Entire article begins with "What If" and then explores it seriously. What if questions are important even if the subsequent topic isn't fundamentally serious.


> Rather than being technically superior weapons, firearms took the skills and muscular effort out of killing someone from a distance

This, after detailed explanations that amount to why firearms are technically superior.

It is very silly.


Um...have you ever noticed just how many interesting-but-of-no-practical-use-whatever things are popular on HN? We are for-sure not setting public policy priorities here.


> Humans really are doomed.

I agree with you, except no.

I mean, the universe will eventually die in heat death - and long before that Earth will become inhospitable - and even before that - we'll probably get smashed by some giant meteor or something.

Smart people wasting braincells on pointless thought experiments does not make us doomed.

On the contrary, the extremely-rare crazy idea that turns out to be genius is probably worth all the pointless debate.


> Of course this would get upvoted on HN

I agree for less pessimistic reasons. The Hacker News crowd prefers intellectually stimulating conversation. Considering unexplored alternatives.

Exploring ideas is a huge pillar of this community. Engaging in that process doesn't mean that any of us think this could happen in the real world.


I think the point of the article was to show how much more lethal the average human has become over the years. The initial question, was clickbait.

>> Rather than being technically superior weapons, firearms took the skills and muscular effort out of killing someone from a distance. The main reason most European armies switched from bows to crossbows and then firearms was the short learning curves of these weapons. Crossbowmen and musketeers required little or no training, while it took many years of practice to build an archer skillful and strong enough to be of use in warfare. The crossbow and the firearm thus expanded the number of people in a given population that could become soldiers. That was great news for those in power because they could now build large armies quickly.


> the point of the article was to show how much more lethal the average human has become over the years. The initial question, was clickbait

That’s unfortunate then, since that point is verifiably false. Historically speaking people do considerably less killing today than they did in the past, despite having inferior weapons[1]. There is in fact a solid argument that readily available superior weaponry has a net zero effect on violence. For example controlling for demographics the USA’s murder rate is roughly on par with the Western European countries. And indeed, rural home invasions are virtually unheard of in the USA while they do occur in Europe.

[1] https://www.vrc.crim.cam.ac.uk/vrcresearch/london-medieval-m...


>> That’s unfortunate then, since that point is verifiably false.

You're thinking rate of violence day-to-day. I feel the author was thinking of how humans have increased their potential lethality, by analyzing the range, rate of fire, supply, required skill, accuracy of humans and their relationship with weaponry.

>> And indeed, rural home invasions are virtually unheard of in the USA

Wow, I need to live where you live! Home invasions are quite common in metro areas in the USA (edit: I see you said "rural" now). As a note the US has vast rural areas that see very little crime, so nationally speaking the US statistics versus Europe is horribly skewed. True comparison would be by cities, but every country and city reports crime differently!

Data for burglaries:

Europe: https://www.eupedia.com/europe/crime_maps_of_europe.shtml#bu...

USA: https://en.wikipedia.org/wiki/List_of_United_States_cities_b...


> For example controlling for demographics

bro just come out and say it


I did. Is there something you wish to add?


> an impractical contrarian outsider

I guess you didn't read much of lowtechmag, which is definitely just about the opposite of "impractical", being much centered on actual experimentation, detailed technical analysis etc. But then again "of course people on HN" would get offended by such slightly technocritical pov.

> suggesting that the entire world replace advanced weaponry with stuff that got phased out hundreds of years ago

Exactly what part of the title starting with "What if" did you not understand? This article -- like several other on the website, but for this one it's particularly obvious -- are thought experiment, designed very much to confront people with unusual ideas and hence make the current norm more obvious.


I read it as a stimulating thought experiment. It's supposed to get you thinking outside of those well worn ruts. Of course, those ruts are often well worn because they are tried and true, but it's nice to go off road once in a while.


The point of weapons is a bit complicated. One interpretation is that they create fear of engaging in violence in the first place. In that respect the killing ability may be secondary.


> dozens of people are taking this premise seriously and debating it on its merits.

Veteran's Day is being observed today in the US

not all of us but some are a bit bored today :)


Not to dismiss the entire HN community (theres some good discussion on here and it probably swings a bit older than most places) but the majority of people who even have the time to vote and comment on the internet tend to be younger and pretty naive. This is especially true on places like Reddit.


We really are doomed because some people like to think of ways that we could kill fewer of each other?


No, it's the ridiculous irony of thinking everyone would cooperate, but would still want to kill each other.


How about the fact that laws of war[0] exist?

[0] https://en.wikipedia.org/wiki/Law_of_war


Lots of these laws of war exist to limit the gruesomeness of war, but if you think it's reasonably necessary for the objective of the war, these rules (legally) go out the window (completely, but to whatever extent is necessary). For example <https://en.wikipedia.org/wiki/Proportionality_(law)>


Which unfortunately get ignored all the time by everyone.


Not all the time, occasionally the winners use them to take revenge on the losers.


Unless of course, the winner dod the same thing (unconditional submarine warfare, strategic bombing). But then I think people will have less decency when it comes to these things than back in 1945.


I think it’s just a thought experiment.


Reverting to bow and crossbows wouldn’t reduce the amount of people killed. Look up “tactical crossbow” (or assault crossbow) in your favorite search engine.


Honestly probably would increase it.

For the last 80 or so years the nuclear weapon has likely reduced the number of deaths in war more than anything else. Of course it's hard to prove a counterfactual. This said, the cost of outright war with nukes is so terrible that the large nations of the world have avoided going toe to toe and instead engaged in smaller proxy wars, that while terrible, have had relatively low population kill rate in relation to what occurred in world wars.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: