Arguing Without Warning
dynomight.net
dynomight.net
* "You vs Me" vs "Us vs the problem". In the latter, there's an objective Correct answer and each party would prefer to find it than to "win" the debate; in the former, the parties would rather "win" than discover new information that would change their mind. The article assumes that internet debates are of the latter variety, while in my experience they are 99% the former.
* With/without audience. Without an audience, you are trying to convince the person you're arguing with, which forces you to address their strongest points; with an audience, you (usually) have no real hope of budging the person you're talking to, and are only interested in persuading some hypothetical listener, so you are incentivized to ignore their strongest points and pounce on the weakest ones. Again, the essay is written as if internet debaters are speaking to each other, when in reality they are overwhelmingly talking past each other and hoping to persuade the audience.
Overall this reads like someone pontificating on whether it is politer in a bar fight to sucker-punch with the right or left hand. Perhaps of interest but not likely to affect any actual bar fights.
I think the blogger's missing forest is that it is nigh impossible to have an 'us vs the problem' debate unless 'you and me' both agree on the assumptions that frame the problem. If we do not first root out and address incongruous assumptions, we will forever talk past each other, no matter how many brass tacks we jump on, even if it seems to bystanders that we both speak the same language.
As an aside, I believe that a person's assumptions/beliefs are inseparable from who that person is (you are what you eat). For me, it is silly to critique an idea without much consideration about what person/people communicated that idea and why they believe it (assuming they do). I often find that the truth is not as cut and dry as I assumed. If your goal is 'internet glory', then sure, that is a waste of time. You are better off gaming engagement algorithms. But, if I may ride a high horse, mutual understanding is a sweeter nectar.
And straw man debate is a type of intellectual bullshit we have too much of already.
(So why am I engaging in this type of intellectual bullshit by writing this comment? Sometimes when I'm very tired and don't have energy to do more useful things like read a book I get a little kick out of trying to state arguments concisely and get little tokens of appreciation in the form of upvotes from fellow hackers.
These tokens are less likely in a drawn-out argument, even if it becomes intellectually more honest. So... I blame the system!)
Or, you criticize the thesis as honestly as you can with the information the author provided, the author responds with their steel man version, and you publicly go back and forth until clarity is achieved.
A public cycle of criticism based on a good-faith misunderstanding is itself useful, as it puts on display the disconnect between what an author thinks they are saying and what their words actually communicate to the public.
Sure. But conversations take time, and always require shared background understanding. A some point, you can't let every conversation devolve into rebuilding the foundations of mathematics from axioms.
Authors have a responsibility to explain their ideas to a reader with a "reasonable" background of shared knowledge. If a long dialog ensues that reveals that reasonable shared background is neither reasonable nor shared, then great. But if the dialog devolves into a back-and-forth with obstinate readers who both refuse to enter into shared background understanding (e.g. not putting in any work, by not doing suggested background research or flat out disagreeing on basic premises like matter is made of atoms, etc) and simultaneously have really strong opinions, then everything gets derailed.
So critics do have some responsibility to know what they are talking about as well.
From good-natured participants, the dueling blog posts format can be delightful to read.
[1] - I first noticed it in Scott Alexander's Slate Star Codex, but I've seen many others use this pattern since.
It's always one of the best parts of what he publishes.
This works even when the authors are using pseudonyms.
I've given a lot of thought to why it works so well, and I think it boils down to three things:
1. Scott really embodies the "strong opinions, loosely held" philosophy and leads by example.
2. The community is big and diverse enough that it doesn't allow a single dogma to dominate. Everyone is a heretic.
3. The moderation is heavy handed, but emphasizes "be kind to each other" rather than "conform to conventional narratives".
It's nice to visit a repository of user-generated content that isn't a cesspool.
- Be truthful
- Use pragmatic epistemology, don't over-intellectualise
- Expect and offer sincerity
- Aim for universalisation/inclusive and non-self-contradiction
- Always assume good faith
- Use concrete rather than abstract statements
- Good, careful humour, but don't soften hard truths
- Make contributions relevant and timely
- Beware self-deception and rationalisation
- Use the minimum facts necessary to make any point
I think if one keeps these in mind, it's always okay to offer arguments without warning, seeking permission, treading on eggshells or feeling the need to credentialise or excessively qualify a response.
This is probably wrong. There is tons of glory in checking someone's work and finding a mistake. Look at all the controversy and major media coverage over the the Reinhart and Rogoff paper, which was undermined by some majors data collection errors that were later pointed out. This was a major deal at the time.
Most people probably already expected Caplan's thesis to be true or took it for granted as true, so there was no compelling need to check. People will only check if the the conclusion goes against their assumptions or something commonly accepted to be true. So a paper which claims to refute the premise of Keynesianism is something that economists are probably going to be be more inclined to check than a conclusion that affirms the orthodoxy.
HAP made a ton of noise (in specialist circles) certainly. But was it a "major deal"? Was RR really "undermined"?
I see no major change in policy as a result, and popular wisdom among the intelligent-but-lagging public, seems to still that public debt drags on growth. (RR has been near the front of my mind this week as a CEO I know mentioned it and was totally unaware it had been so completely "undermined" a decade prior.)
I’ve seen colleagues waste weeks to get any kind of attention to clear, and when pointed out, quite obvious errors in scientific papers. The original authors have responded with insults, and the editors have had no interest in publishing any kind of criticism.
* You read it.
* You’re convinced that it’s true.
* You search for criticisms.
* They are devastating, so you no longer think the article is true.”
9/10 times on HN and I love you all for it. (Well maybe a bit less but I certainly have shared articles and read the HN comments afterwards and come to regret sharing)
I don't remember who -- maybe Scott Alexander -- brought this up in the context of saying that maybe intelligent people don't suffer enough from confirmation bias.
Intelligent people are so ready to drop their pre-existing beliefs in the face of good arguments that what determines their opinion in the end is not necessarily truth, but at which point they stopped reading criticisms of criticisms of criticisms of criticisms of the original thesis.
With a tiny bit more confirmation bias, intelligent people might be slightly more inclined to stick with popular opinion, and their accuracy might improve as a result.
(using a web archive link since the current page tries to force a login)
This is common on HN where authors are posted without being informed and it is not uncommon for someone who lacks the knowledge and ability to understand the sophisticated position to 'debunk' something that is nonetheless true. Then others like them pile on to the thing and everyone has concluded that the thing is false merely out of a lack of knowledge and intelligence.
An illustrated representation https://i.imgur.com/xyoCO9g.png
Orubyq nf V hfr gur zrzr abg nf vg vf zrnag, ohg vagragvbanyyl. Fhpu frys-ersrerapr.
2) altruism exists
3) if you briefly try to talk to the author you _may_ avoid misunderstanding their idea and thus less time is wasted.
3) being proven wrong is great. Being right is boring.
It would be nice if human communication had some sort of predicate whereby you could remove the second effect from your argument. But AFAIK it depends almost entirely on trusting the audience to decouple the argument from the arguer, which only a very few audiences can be trusted to do.
Instead, people who do this are just shoveling more confusion, reinforcing tribal lines in some cases, and frankly, muddying the waters.
I'll give an example. Relativity. There is absolutely no doubt in the physics community that Relativity is a.) a thing b.) well-understood mathematically c.) explains physical reality remarkably well, being supported by vast amounts of evidence. And yet general relativity is not yet unified with quantum theory.
Do we need more cranks who a.) do not understand the underlying mathematics and b.) do not understand what the fundamental shortcoming is, in detail, firing off articles "criticizing" it? Hells no. These types of things should be debated by physicists and mathematicians who do actually thoroughly understand these things. Physics journals won't suffer amateur hour, and they shouldn't. Physics journals, and all scientific publishing venues, subject articles to review that try to elevate the level of discourse by checking and re-checking claims.
But the internet is not journals, sure. It's a free-for-all. It has elevated so many random voices to a level of authority that it is not possible to weed through claims at speed; cranks make themselves look like experts. An experts understate their own confidence. The internet is a massive Dunning-Krueger melting pot.
I will absolutely push back on the idea that the problem we have right now is that the expectations are too high. You mean you have to understand what a person is saying, maybe even talking to the person first, before you launch a volley of utter nonsense at them? The humanity.
This usually happens when people make a snap judgment based off the title and maybe the first few sentences of an article without reading further.
That said, I'm not clear on why the author thinks that we're at a "turning point" in deciding whether it's socially acceptable to just publish an argument against something else that was published. Maybe in the scientific community that's a strict norm? It certainly doesn't seem to be in the blogging community.
Maybe the author is worried that good arguments are lost in the semi-private twitter scuffle with the original authors of papers and articles? Maybe they hope that people will respond to broadcasted arguments intended for the general public with counter-arguments that are also broadcasted, instead of directed? I don't know. It feels like I'm missing the article's "prompt".
0. https://kylesheldrick.blogspot.com/2022/03/evidence-of-fabri...
1. https://erikhoel.substack.com/p/why-we-stopped-making-einste...
2. https://astralcodexten.substack.com/p/contra-hoel-on-aristoc...
But you yourself recognize that the cranks have no qualms doing that, while the experts understate their expertise. The cranks are cranks; they're not going to take the effort to understand something thoroughly before responding to it, because then they wouldn't be cranks anymore. The only alternative is for everyone to shout as loud as they can about what an idiot Steve is for saying that stupid thing he did. Modesty has no place on the Internet. Fire on all cylinders and let Steve defend himself if he wants to.
> How much should these comments decrease your confidence in the original article? I claim: Not much, at least on average.
> First, if everything you read has devastating criticisms, you should expect them to exist.
Without explicitly commenting on the world in which we live, it seems to me that all the hypotheses are consistent with a world of low-quality discourse, in which many of the factual claims we encounter in popular articles are wrong (in possibly minor, but possibly major, ways). You should always evaluate your sources, to be sure—and the value of an anonymous comment on even a high-quality forum is fairly low—but a factual refutation of a factual claim should, I think, be given very high weight, even if most such claims have such refutations.
(The author seems further on to be pointing out perhaps the futility of refuting a 'headline point' when the headline will be fleshed out in the body. With that I can more easily agree. Of course, as is the way of such things, I was more eager to make my slightly contrary point than to finish reading the essay.)
> The mistake here is seeing a counter-argument as being against a person, rather than as against a particular artifact. What matters is what’s out there.
This is not how it works in most of real life. There are political impacts within a community or organization depending on how an argument goes down. Potentially permanent schisms.
> So, if the critiques of an article find only “moderately bad flaws”, that often increases my trust because my prior was that the flaws would be even worse. If the claim that some types of watermelon taste bad to certain people at some times of the year survives the skeptics, then that might be an above-average outcome.
This is empirically not how most people operate.
> We need to design our norms around the unfortunate reality that most people are not helpful when criticized.
No, design norms around getting the best possible world. That unfortunate reality is only one of many factors to take into account.
---
I'll hold judgement on the conclusion overall; I think it may make sense for arguments when nothing political is on the line and everyone is a psuedorationlist. Beyond that...unsure.
Whoa, whoa, hold your horses. Pause right there.
How does that "so" conjunction work? Given that you think they're wrong, why would you want to write a manifesto arguing that they're wrong? What's your motivation behind it?
There can be a plethora of possible motivations, all of them valid. Just to name a few:
- The original statement is in direct conflict with your world view. Because it's being read and shared, you want to write a counterpiece because you, too, wish to be heard.
- You believe that the original statement represents a belief that's factually incorrect, so you want to write a corrigendum because you believe that policies based on false premises are bound to lead astray.
- The original statement is written in a way that touches you emotionally. Perhaps it disparages a part of the world that's important to you. You want to respond in defense of that part.
- You don't find the matter particularly important, but you see an omission in the original statement's reasoning that you think you can amend. You write a supplementary response to improve your (and hopefully others') understanding of the problem at hand.
- You just have that irresistible urge originating $GOD knows where, that you have no better name for than "xkcd #386".
Depending on the circumstances, you may or may not want to follow the various norms that Dynomight mentions in their article. But I think it's of paramount importance to be self-aware of our own motivations. It also helps to be explicit about them.
There are two types of criticisms: criticism out of disagreement with the thesis, or a criticism that undermines the reasoning or conclusion of the thesis itself. The latter is much worse. Good journalists and writers try to keep the second to a minimum as it can come at a major cost to credibility.
> This is a compliment. If someone criticizes something you wrote, you should feel gratified that you were “right enough” to be worth arguing with.
This is the logic of internet trolls. I'm right, and if you disagree, that proves I'm right, and if you fully debunk my argument, that means you're desperate to keep people from believing me, which, again, proves I'm right.
I don't know if you've heard, but people will accept a lot of arguments that aren't anywhere near to right, because it appeals to their preconceptions, or it would be convenient to their goals if it were right, or their friend said it was good, or a million other reasons.
Sure but why? Lets focus on politics as it's at the centre of all this.
The usual or normal stake in society is virtually no discussion about politics. Once every few years you pick a representative and then you can ignore politics because you can't change anything. You ought to be talking about that local sports team or that new tv show. This changed in recent years, I'm not really familiar with this ever happening before. Politics is beyond represented in common discussion today but you CANNOT argue. Dont dare take a side or else you get labelled an *ist.
You will also notice that all thought commanders in the field of politics are either comedians or lawyers; while the actual politicians tend to be lawyers or celebrities. None of those 3 groups are allowed to speak their mind or argue.
The battlefield of thought is where this isn't true. You can find this battleground where the comedians and lawyers interview.
The prevailing theory also has the blame on the "STEM" movement in that highly qualified people left social sciences toward STEM leaving a huge talent gap in the social sciences. In addition to the reality of social media utterly destroying mainstream media. Just look at the fall of CNN to political activists.
What does matter is that we have a new hyper awareness of politics. The left wing right wing false dichotomy has collapsed. Why is it the universities used to be for free speech and are now certainly not?
It's also super important to notice how unsuccessful major protests have been. BLM, who literally represent something clearly broken, very clearly the USA should fix their objective racist and brutal police force. The USA absolutely needs to fix the multitude of problems along these lines. However they have received absolutely no fix. They have had major protests and riots all over and nobody at all in government is willing to fix it. You're not allowed to argue or even discuss this.
However, if you try to argue or point any of this out. You become a nazi or racist or whatever ist they decide upon.
Hopefully you can see why arguing has become impossible. It's not just politics, but politics is the center. Race, Gender, Drugs, environmentalism, guns, etc. Tons of things you are simply not allowed to argue about.
OP knows this as well which is why he discusses Watermelons. He actively goes out of his way to discuss something that nobody discusses because it's the only way to avoid hitting a subject you're not allowed to discuss. He could have discussed how everyone in Toronto lives in an igloo.
Why is this happening? One side of politics has decided that society is on the brink of collapse if not extinction. This is a matter of life or death.
Unfortunately, I made a huge mistake in this post. I argued.
I don't know at what point your depiction of "virtually no discussion about politics" ever "usual or normal". In my parents' lifetime we had political fights around interracial marriage, AIDS, and racial integration-- and that's just in the country I was born in. In the country they were born in, there was so much politics it became a massive civil conflict and they had to flee for their lives. In their grandparents' time there was already civil war and one of my grandparents was a POW, so I bet that was plenty of discussion too.
The average person does not talk politics in normal conversation. The conversation ought to be how terrible the toronto maple laughs are doing. Or book of boba fett or new obiwan mini series. So many different subjects, politics rarely become a subject let alone dominate like it is doing today.
>In my parents' lifetime we had political fights around interracial marriage, AIDS, and racial integration-- and that's just in the country I was born in.
Sure you are always going to have hot topics.
I'm curious, in normal conversation do you always talk about the political hot topics? You dont particularly care about $localsportsteam or whatever?
>In the country they were born in, there was so much politics it became a massive civil conflict and they had to flee for their lives.
Oh yes, that's precisely what's coming. Kind of the point I was alluding to.
> In their grandparents' time there was already civil war and one of my grandparents was a POW, so I bet that was plenty of discussion too.
Totally agree. I think we are in agreement here. Obviously I didn't spell it out to this extent.
Actually yeah, I keep track of my parents' immigration process, which is a political hot topic because they come from a non-european country and therefore their immigration status is always at risk here.
It's a terrible thing to watch as both people and issues are reduced down to binary options.
For example, I'd argue that the "Trumpist" white working class person in the Midwest has a lot in common with the black suburban working class person. These are now worlds apart, at times enemies even, but they share a lot of issues. Both groups, which I argue to be a single group, struggle to get a livable wage, access to education and healthcare, and so on. Somehow, society has sorted these groups into good and bad people, where good and bad change based on perspective.
Same for complicated divisive issues. Most people, yes even conservatives, would agree that racism is bad. But that doesn't necessarily mean that they support every method to combat it, like "defund the police", reparations, equity policy, and many other policies not at all related to the original problem. But you can't discuss any of the points or plans intelligently. You're all-in with every idea, including bad ones, or you're out, and thus a racist.
A system like this leaves only one option: silence.
In contrast, the political system in my country, the Netherlands. It's a coalition country where the ruling government consists of 3-4 parties to form a majority. So that means that the norm is that everything is multi-partisan. You are forced to work together or you will not rule. Likewise, you can't destroy or smear the other in a campaign because you'll meet them again at the formation table.
In a system like this, consensus is the norm, and extremism has no chance, they will never rule. This has its effect on people, as most people are centrists with a notch to the left or right.
My group of friends is politically all over the map. Nobody takes issue with deviations in opinions, there's no hard or clear line to sort people into. There's also no left-right consistency. I know progressive people that are conservative regarding immigration and I know conservative people that are progressive environmentalists.
Sorting people into left-right, good-bad, conservative-democrat is a crime. An insult to humanity. Because the really sad thing is most political needs are universal. The split is entirely artificial.
Oh for sure, they measure this objectively. https://www.visualcapitalist.com/charts-americas-political-d...
Put your finger on Republican line in 1994. You will see at first the republicans went left with the democrats but then went right back to where they have always been.
The polarization is entirely from the democrat camp.
>It has always had it due to the two party system but it's been getting exponentially worse. Or, alternatively, maybe the silent majority isn't necessarily more divided, instead the extreme ends are disproportionately promoted in media and social media, since outrage sells and reasonable people are boring. Reasonable people have nowhere to go, they have to pick one of the two types of crazy.
Social media is certainly at play but not in the way the media portrays. You can see the polarization started long before social media.
This isn't really the giant douche vs turdsandwich factor neither. They've been doing that since Lincoln's times. This isn't a key factor.
>For example, I'd argue that the "Trumpist" white working class person in the Midwest has a lot in common with the black suburban working class person.
This has far more to do with symptoms of the problem. The irony is that in terms of George Floyd or BLM. Trump is the only person who has done ANYTHING for them. BLM picked the wrong allies.
> These are now worlds apart, at times enemies even, but they share a lot of issues. Both groups, which I argue to be a single group, struggle to get a livable wage, access to education and healthcare, and so on. Somehow, society has sorted these groups into good and bad people, where good and bad change based on perspective.
Yet blacks overwhelming vote for the democrats. In fact, places like Detroit, Georgia, etc voted far more for Biden than they did for Obama. Biden is the most popular president in US history by his own words.
>Same for complicated divisive issues. Most people, yes even conservatives, would agree that racism is bad.
The USA certainly has a racism problem. Objectively proven over and over. However, labeling the republicans racist is hilariously wrong. In fact, it has gone way beyond reality. There's countless examples of the media labeling black republicans as white supremacists. Larry Elder or Winsome Sears for example. Which again, this isn't because of the republicans. This is coming from left-wing media and democrats.
>ut that doesn't necessarily mean that they support every method to combat it, like "defund the police"
Literally nobody supports defund the police. That's not a thing anywhere in the world.
> reparations,
Kamala Harris officially supports this. Perhaps the democrats will move on it? I highly highly doubt it.
>equity policy, and many other policies not at all related to the original problem. But you can't discuss any of the points or plans intelligently. You're all-in with every idea, including bad ones, or you're out, and thus a racist.
This isn't really about fixing issues.
>A system like this leaves only one option: silence.
They've run 2 parties since Lincoln. This polarization is much newer.
>In contrast, the political system in my country, the Netherlands. It's a coalition country where the ruling government consists of 3-4 parties to form a majority. So that means that the norm is that everything is multi-partisan. You are forced to work together or you will not rule. Likewise, you can't destroy or smear the other in a campaign because you'll meet them again at the formation table.
I'm a gigantic fan of proportional representation. Canada was about to do this, then Trudeau lied about supporting electoral reform and swung the polls in his favour. He got into power and never delivered because he wasn't planning to do proportional rep, he was planning to change to an electoral system that would basically keep in power forever. The provinces basically looked at his plan and told him to go away and wouldn't even consider talking to him.
>In a system like this, consensus is the norm, and extremism has no chance, they will never rule. This has its effect on people, as most people are centrists with a notch to the left or right.
I dont think extremism is the problem with the democrat's polarization. In fact, the extremists in the democrat party basically go unopposed. They dont necessarily achieve anything but they dont get called out.
>My group of friends is politically all over the map. Nobody takes issue with deviations in opinions, there's no hard or clear line to sort people into. There's also no left-right consistency. I know progressive people that are conservative regarding immigration and I know conservative people that are progressive environmentalists.
Left -right is a complete fabrication by the media. Take here in Canada.
We have the Green party which is really a single issue party about climate change and environmentalism. The media labels them far-left. According to the media there was a problem in the last election. Their leader who is black and jewish lost because the Green party was infested with nazis. They lost about 4% of the popular vote dropping to 2%.
So where did that 4% of support go? Well we know because no other parties really changed in popular vote. The far left vote went to the Peoples Party. Who jumped up to 5% of the popular vote. So who are the peoples party? According to the media. they are far-right white supremacists and nazis.
WTF? Canada has a serious Nazi problem eh. Or maybe is it that the media is lying and mislabeling these parties? That Canadians are good people and there's virtually no nazis. It's the media who are the problem.
>Sorting people into left-right, good-bad, conservative-democrat is a crime. An insult to humanity. Because the really sad thing is most political needs are universal. The split is entirely artificial.
It's literally only the media. Yellow journalism is back with a passion of calling everyone a nazi and racist.