= 27d567bc2d48d9f40e9ccc20ea370b15)
The noncentral fallacy - the worst argument in the world?
Related to: Leaky Generalizations, Replace the Symbol With The Substance, Sneaking In Connotations
David Stove once ran a contest to find the Worst Argument In The World, but he awarded the prize to his own entry, and one that shored up his politics to boot. It hardly seems like an objective process.
If he can unilaterally declare a Worst Argument, then so can I. I declare the Worst Argument In The World to be this: "X is in a category whose archetypal member gives us a certain emotional reaction. Therefore, we should apply that emotional reaction to X, even though it is not a central category member."
Call it the Noncentral Fallacy. It sounds dumb when you put it like that. Who even does that, anyway?
It sounds dumb only because we are talking soberly of categories and features. As soon as the argument gets framed in terms of words, it becomes so powerful that somewhere between many and most of the bad arguments in politics, philosophy and culture take some form of the noncentral fallacy. Before we get to those, let's look at a simpler example.
Suppose someone wants to build a statue honoring Martin Luther King Jr. for his nonviolent resistance to racism. An opponent of the statue objects: "But Martin Luther King was a criminal!"
Any historian can confirm this is correct. A criminal is technically someone who breaks the law, and King knowingly broke a law against peaceful anti-segregation protest - hence his famous Letter from Birmingham Jail.
But in this case calling Martin Luther King a criminal is the noncentral. The archetypal criminal is a mugger or bank robber. He is driven only by greed, preys on the innocent, and weakens the fabric of society. Since we don't like these things, calling someone a "criminal" naturally lowers our opinion of them.
The opponent is saying "Because you don't like criminals, and Martin Luther King is a criminal, you should stop liking Martin Luther King." But King doesn't share the important criminal features of being driven by greed, preying on the innocent, or weakening the fabric of society that made us dislike criminals in the first place. Therefore, even though he is a criminal, there is no reason to dislike King.
This all seems so nice and logical when it's presented in this format. Unfortunately, it's also one hundred percent contrary to instinct: the urge is to respond "Martin Luther King? A criminal? No he wasn't! You take that back!" This is why the noncentral is so successful. As soon as you do that you've fallen into their trap. Your argument is no longer about whether you should build a statue, it's about whether King was a criminal. Since he was, you have now lost the argument.
Ideally, you should just be able to say "Well, King was the good kind of criminal." But that seems pretty tough as a debating maneuver, and it may be even harder in some of the cases where the noncentral Fallacy is commonly used.
Now I want to list some of these cases. Many will be political1, for which I apologize, but it's hard to separate out a bad argument from its specific instantiations. None of these examples are meant to imply that the position they support is wrong (and in fact I myself hold some of them). They only show that certain particular arguments for the position are flawed, such as:
"Abortion is murder!" The archetypal murder is Charles Manson breaking into your house and shooting you. This sort of murder is bad for a number of reasons: you prefer not to die, you have various thoughts and hopes and dreams that would be snuffed out, your family and friends would be heartbroken, and the rest of society has to live in fear until Manson gets caught. If you define murder as "killing another human being", then abortion is technically murder. But it has none of the downsides of murder Charles Manson style. Although you can criticize abortion for many reasons, insofar as "abortion is murder" is an invitation to apply one's feelings in the Manson case directly to the abortion case, it ignores the latter's lack of the features that generated those intuitions in the first place2.
"Genetic engineering to cure diseases is eugenics!" Okay, you've got me there: since eugenics means "trying to improve the gene pool" that's clearly right. But what's wrong with eugenics? "What's wrong with eugenics? Hitler did eugenics! Those unethical scientists in the 1950s who sterilized black women without their consent did eugenics!" "And what was wrong with what Hitler and those unethical scientists did?" "What do you mean, what was wrong with them? Hitler killed millions of people! Those unethical scientists ruined people's lives." "And does using genetic engineering to cure diseases kill millions of people, or ruin anyone's life?" "Well...not really." "Then what's wrong with it?" "It's eugenics!"
"Evolutionary psychology is sexist!" If you define "sexist" as "believing in some kind of difference between the sexes", this is true of at least some evo psych. For example, Bateman's Principle states that in species where females invest more energy in producing offspring, mating behavior will involve males pursuing females; this posits a natural psychological difference between the sexes. "Right, so you admit it's sexist!" "And why exactly is sexism bad?" "Because sexism claims that men are better than women and that women should have fewer rights!" "Does Bateman's principle claim that men are better than women, or that women should have fewer rights?" "Well...not really." "Then what's wrong with it?" "It's sexist!"
A second, subtler use of the noncentral fallacy goes like this: "X is in a category whose archetypal member gives us an emotional reaction. Therefore, we should apply that same emotional reaction to X even if X gives some benefit that outweighs the harm."
"Capital punishment is murder!" Charles Manson-style murder is solely harmful. This kind of murder produces really strong negative feelings. The proponents of capital punishment believe that it might decrease crime, or have some other attending benefits. In other words, they believe it's "the good kind of murder"3, just like the introductory example concluded that Martin Luther King was "the good kind of criminal". But since normal murder is so taboo, it's really hard to take the phrase "the good kind of murder" seriously, and just mentioning the word "murder" can call up exactly the same amount of negative feelings we get from the textbook example.
"Affirmative action is racist!" True if you define racism as "favoring certain people based on their race", but once again, our immediate negative reaction to the archetypal example of racism (the Ku Klux Klan) cannot be generalized to an immediate negative reaction to affirmative action. Before we generalize it, we have to check first that the problems that make us hate the Ku Klux Klan (violence, humiliation, divisiveness, lack of a meritocratic society) are still there. Then, even if we do find that some of the problems persist (like disruption of meritocracy, for example) we have to prove that it doesn't produce benefits that outweigh these harms.
"Taxation is theft!" True if you define theft as "taking someone else's money regardless of their consent", but though the archetypal case of theft (breaking into someone's house and stealing their jewels) has nothing to recommend it, taxation (arguably) does. In the archetypal case, theft is both unjust and socially detrimental. Taxation keeps the first disadvantage, but arguably subverts the second disadvantage if you believe being able to fund a government has greater social value than leaving money in the hands of those who earned it. The question then hinges on the relative importance of these disadvantages. Therefore, you can't dismiss taxation without a second thought just because you have a natural disgust reaction to theft in general. You would also have to prove that the supposed benefits of this form of theft don't outweigh the costs.
Now, because most arguments are rapid-fire debate-club style, sometimes it's still useful to say "Taxation isn't theft!" At least it beats saying "Taxation is theft but nevertheless good", then having the other side say "Apparently my worthy opponent thinks that theft can be good; we here on this side would like to bravely take a stance against theft", and then having the moderator call time before you can explain yourself. If you're in a debate club, do what you have to do. But if you have the luxury of philosophical clarity, you would do better to forswear the Dark Arts and look a little deeper into what's going on.
Are there ever cases in which this argument pattern can be useful? Yes. For example, it may be a groping attempt to suggest a Schelling fence; for example, a principle that one must never commit theft even when it would be beneficial because that would make it harder to distinguish and oppose the really bad kinds of theft. Or it can be an attempt to spark conversation by pointing out a potential contradiction: for example "Have you noticed that taxation really does contain some of the features you dislike about more typical instances of theft? Maybe you never even thought about that before? Why do your moral intuitions differ in these two cases? Aren't you being kind of hypocritical?" But this usage seems pretty limited - once your interlocutor says "Yes, I considered that, but the two situations are different for reasons X, Y, and Z" the conversation needs to move on; there's not much point in continuing to insist "But it's theft!"
But in most cases, I think this is more of an emotional argument, or even an argument from "You would look silly saying that". You really can't say "Oh, he's the good kind of criminal", and so if you have a potentially judgmental audience and not much time to explain yourself, you're pretty trapped. You have been forced to round to the archetypal example of that word and subtract exactly the information that's most relevant.
But in all other cases, the proper response to being asked to subtract relevant information is "No, why should I?" - and that's why this is the worst argument in the world.
Footnotes
1: On advice from the community, I have deliberately included three mostly-liberal examples and three-mostly conservative examples, so save yourself the trouble of counting them up and trying to speculate on this article's biases.
2: This should be distinguished from deontology, the belief that there is some provable moral principle about how you can never murder. I don't think this is too important a point to make, because only a tiny fraction of the people who debate these issues have thought that far ahead, and also because my personal and admittedly controversial opinion is that much of deontology is just an attempt to formalize and justify this fallacy.
3: Some people "solve" this problem by saying that "murder" only refers to "non-lawful killing", which is exactly as creative a solution as redefining "criminal" to mean "person who breaks the law and is not Martin Luther King." Identifying the noncentral fallacy is a more complete solution: for example, it covers the related (mostly sarcastic) objection that "imprisonment is kidnapping".
4: EDIT 8/2013: I've edited this article a bit after getting some feedback and complaints. In particular I tried to remove some LW jargon which turned off some people who were being linked to this article but were unfamiliar with the rest of the site.
5: EDIT 8/2013: The other complaint I kept getting is that this is an uninteresting restatement of some other fallacy (no one can agree which, but poisoning the well comes up particularly often). The question doesn't seem too interesting to me - I never claimed particular originality, a lot of fallacies blend into each other, and the which-fallacy-is-which game isn't too exciting anyway - but for the record I don't think it is. Poisoning the well is a presentation of two different facts, such as "Martin Luther King was a plagiarist...oh, by the way, what do you think of Martin Luther King's civil rights policies?" It may have no relationship to categories, and it's usually something someone else does to you as a conscious rhetorical trick. Noncentral fallacy is presenting a single fact, but using category information to frame it in a misleading way - and it's often something people do to themselves. The above plagiarism example of poisoning the well is not noncentral fallacy. If you think this essay is about bog-standard poisoning the well, then either there is an alternative meaning to poisoning the well I'm not familiar with, or you are missing the point.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Comments (1742)
This post could also be titled "Why Kant was Stupid, Part One of One Thousand".
Even if your post remains voted down, I thought it was funny.
Voting up things purely because they're funny tends to be bad for serious discussion forums like this one. If encouraged, purely funny content has a tendency to take over and crowd out insightful content.
I like funny content, but not at the risk of insightful content.
My comment provided an insight insofar as it explicitly connected the analysis to Immanuel Kant's writing in order to criticize it.
I like it too.
I've banned all of eridu's recent comments (except a few voted above 0) as an interim workaround, since hiding-from-Recent-Comments and charge-fee-to-all-descendants is still in progress for preventing future threads like these.
I respectfully request that you all stop doing this, both eridu and those replying to him.
Bravo. I have no idea whether that was someone pretending to be ignorant and toxic for the purpose of discrediting a group he was impersonating or whether it was sincere (and ignorant and toxic). Fortunately I don't need to know and don't care either way. Good riddance!
Is it just me or do others also find that Eliezer coming of as a tad petulant with the way he is handling people systematically opposing and downvoting his proposal? Every time he got downvoted to oblivion he just came back with a new comment seemingly crafted to be more belligerent, whiny, condescending and cynical about the community than the last. (That's hyperbole---in actuality it peaked in the middle somewhere.) Now we just keep getting reminded about it at every opportunity as noise in unrelated threads.
I observe that wedifrid has taken advantage of this particular opportunity to remind everyone that he thinks I am belligerent, whiny, condescending, and cynical.
(So noted because I was a bit unhappy at how the conversation suddenly got steered there.)
I also observe that wedrifid's opinion of you doesn't appear to be steered with equal expected posterior probability in light of how you react versus his predictions of your reactions.
I'm curious as to whether I'm on to something there, or whether I just pulled something random and my intuitions are wrong.
If that bothers you, you may consider that whining that people find you whiny might not be the optimal strategy for making them change their mind.
I notice that my criticism was made specifically regarding the exhibition of those behaviors in the comments he has made about the subject he has brought up here. We can even see that I made specific links. Eliezer seems to be conflating this with a declaration that he has those features as part of his innate disposition.
By saying that wedrifid is reminding people that he (supposedly) believes Eliezer has those dispositions he also implies that wedrifid has said this previously. This is odd because I find myself to be fairly open with making criticisms of Eliezer whenever I feel them justified and from what I recall "belligerent, whiny, condescending, and cynical [about the lesswrong community]" isn't remotely like a list of weaknesses that I actually have described Eliezer as having in general or at any particular time that I recall.
Usually when people make this kind of muddled accusation I attribute it to a failure of epistemic rationality and luminosity. Many people just aren't able to separate in their minds a specific criticism of an action and belief about innate traits. Dismissing Eliezer as merely being incompetent at the very skills he is renowned for would seem more insulting than simply concluding that he is being deliberately disingenuous.
My suggestion is that Eliezer would be best served by not bringing the conversation here repeatedly. It sends all sorts of signals of incompetence. That 'unhappy' feeling is there to help him learn from his mistakes.
Mostly he's coming across to me as having lost patience with the community not being what he wants it to be, and having decided that he can fix that by changing the infrastructure, and not granting much importance to the fact that more people express disapproval of this than approval.
Keep in mind that it's not "more people" it's more "people who participate in meta threads on Less Wrong". I've observed a tremendous divergence between the latter set, and "what LWers seem to think during real-life conversations" (e.g. July Minicamp private discussions of LW which is where the anti-troll-thread ideas were discussed, asking what people thought about recent changes at Alicorn's most recent dinner party). I'm guessing there's some sort of effect where only people who disagree bother to keep looking at the thread, hence bother to comment.
Some "people" were claiming that we ought to fix things by moderation instead of making code changes, which does seem worth trying; so I've said to Alicorn to open fire with all weapons free, and am trying this myself while code work is indefinitely in progress. I confess I did anticipate that this would also be downvoted even though IIRC the request to do that was upvoted last time, because at this point I've formed the generalization "all moderator actions are downvoted", either because only some people participate in meta threads, and/or the much more horrifying hypothesis "everyone who doesn't like the status quo has already stopped regularly checking LessWrong".
I'm diligently continuing to accept feedback from RL contact and attending carefully to this non-filtered source of impressions and suggestions, but I'm afraid I've pretty much written-off trying to figure out what the community-as-a-whole wants by looking at "the set of people who vigorously participate in meta discussions on LW" because it's so much unlike the reactions I got when ideas for improving LW were being discussed at the July Minicamp, or the distribution of opinions at Alicorn's last dinner party, and I presume that any other unfiltered source of reactions would find this conversation similarly unrepresentative.
In case you need assurance from the online sector. I wholeheartedly welcome any increase in the prevalence of the banhammer, and the "pay 5 karma" thing seems good too.
During that Eridu fiasco, I kept hoping a moderator would do something like "this thread is locked until Eridu taboos all those nebulous affect-laden words."
Benevolent dictators who aren't afraid of dissent are a huge win, IMO.
I very much appreciate the attempts at greater moderation, including the troll penalty. Thank you.
Me too. Troll posts and really wrong people are too distracting without some form of intervention. Not sure the current solution is optimal (but this point has been extensively argued elsewhere), but I applaud the effort to actually stick one's neck out and try something.
Thank you both. Very much, and sincerely.
Accepting thanks with sincerity, while somewhat-flippantly mostly-disregarding complaints? ...I must be missing some hidden justification?
People who agree are more likely to keep quiet than people who disagree. Rewarding them for speaking up reduces that effect, which means comments get closer to accurately representing consensus.
Sometimes AKA the "Forum Whiners" effect, well known in the PC games domain:
When new PC games are released, almost inevitably the main forums for the game will become flooded with a large surge of complaints, negative reviews, rage, rants, and other negative stuff. This is fully expected and the absence of such is actually a bad sign. People that are happy with the product are playing the game, not wasting their time looking for forums and posting comments there - while people who have a problem or are really unhappy often look for an outlet or a solution to their issues (though the former in much greater numbers, usually). If no one is bothering to post on the forums, then that's evidence that no one cares about the game in the first place.
I see a lot of similarities here, so perhaps that's one thing worth looking into? I'd expect some people somewhere to have done the math already on this feedback (possibly by comparing to overall sales, survey results and propagation data), though I may be overestimating the mathematical propensity of the people involved.
Regarding the stop-watching-threads thing, I've noticed that I pretty much always stop paying attention to a thread once I've gotten the information I wanted out of it, and will only come back to it if someone directly replies to one of my comments (since it shows up in the inbox). This has probably been suggested before, but maybe a "watchlist" to mark some threads to show up new comments visibly somewhere and/or a way to have grandchildren comments to one of your own show up somehow could help? I often miss it when someone replies to a reply to my comment.
That's fair, and your strategy makes sense. I also agree with DaFranker, below, regarding meta-threads.
This said, however, at the time when I joined Less Wrong, my model of the site was something like, "a place where smart people hold well-reasoned discussions on a wide range of interesting topics" (*). TheOtherDave's comment, in conjunction with yours, paints a different picture of what you'd like Less Wrong to be; let's call it Less Wrong 2.0. It's something akin to, "a place where Eliezer and a few of his real-life friends give lectures on topics they think are important, with Q&A afterwards".
Both models have merit, IMO, but I probably wouldn't have joined Less Wrong 2.0. I don't mean that as any kind of an indictment; if I were in your shoes, I would definitely want to exclude people like this Bugmaster guy from Less Wrong 2.0, as well.
Still, hopefully this one data point was useful in some way; if not, please downvote me !
(*) It is possible this model was rather naive.
No; you're conflating "Eliezer considers he should have the last word on moderation policy" and "Eliezer considers LessWrong's content should be mostly about what he has to say".
The changes of policy Eliezer is pushing have no effect on the "main" content of the site, i.e. posts that are well-received, and upvoted. The only disagreement seems to be about sprawling threads and reactions to problem users. I don't know where you're getting "Eliezer and a few of his real-life friends give lectures on topics they think are important" out of that, it's not as if Eliezer has been posting many "lectures" recently.
EY has always seemed to me to want LW to be a mechanism for "raising the sanity waterline". To the extent that wide-ranging discussion leads to that, I'd expect him to endorse it; to the extent that wide-ranging discussion leads away from that, I'd expect him to reject it. This ought not be a surprise.
Nor ought it be surprising that much of the discussion here does not noticeably progress this goal.
That said, there does seem to be a certain amount of non-apple selling going on here; I don't think there's a cogent model of what activity on LW would raise the sanity waterline, so attention is focused instead on trying to eliminate the more blatant failures: troll-baiting, for example, or repetitive meta-threads.
Which is not a criticism; it is what it is. If I don't know the cause, that's no reason not to treat the symptoms.
You... know I don't optimize dinner parties as focus groups, right? The people who showed up that night were people who like chili (I had to swap in backup guests for some people who don't) and who hadn't been over too recently. A couple of the attendees from that party barely even post on LW.
That's kinda the point.
Those who actually don't care about such things what people think don't tend to convey this level of active provocation and defiance.
Sure. I can't speak for EY, clearly, but there are many things (including what other people think) that I find myself caring about, often a lot, but I don't think are important. This is inconsistent, I know, but I find it pretty common among humans.
It's not just you.
I'm starting to think there should be community-elected moderators or something, and Eliezer should stop being allowed to suggest things.
While the discussion arguably veered off-topic with respect to the original article, I don't think we actually have a rule against that. And I don't think eridu was actually trolling, though they do seem to have an overly-dismissive attitude towards the community. I do think there's a place for social constructivist / radical feminist views to be aired where they apply on this site, and I don't think eridu was doing a particularly bad job of it.
If we have a diversity of views, then people will disagree about fundamental sorts of things and we'll end up with people thinking each other are "not even wrong" about some issues, which certainly seems downvote-worthy at the time. But we do want a diversity of views (it's one of the primary benefits of having multiple people interacting in the first place), and so banning comments which are merely unpopular is not called-for, and will simply shunt out potential members of the community.
Of course, I'm basically guessing about your rationale in banning these comments, so if you'd like to provide some specific justification, that would be helpful.
I disagree. It was a perfect example of how the Worst Argument In The World (rather, an especially irritating subtype of the same) is often deployed in the field.
Right now that sounds like one of the most brutal criticisms you could have made of radical feminism.
Once again, please don't do that. (Hiding-from-Recent-Comments is totally okay, however.)
Hey, let's take a class of arguments and call them unilaterally bad because of a few bad archetypes!
I actually thought this argument was quite poor. There are lots of possible features in different cases of a type, and to claim some are vitally important seems to beg the question. Murdering a homeless loner estranged from any family or friends may lack many of the features mentioned, but there's little dispute it would qualify. And preventing the creation of a new life prevents the relationships that person would eventually develop. Pointing out that an example falls into a commonly understood category seems a pretty good starting point before delving into what features of that category are important (which isn't something universally agreed on or even consciously thought about). My preferred approach is what you said for eugenics: just admit that I'm alright with murder some of the time, as per the economically efficient amount of crime (such as theft!).
I also think it is a good thing there is a general norm against breaking laws (even stupid ones) and that it is highly questionable whether George Washington & other "patriot" actions did more good than harm, requiring actual justification in each case against an initial presumption.
This is like saying, "I'm really sorry about how I'm going to slap you". If you have need to apologize for something, just don't do it. If what you are doing isn't wrong, don't apologize.
But really how hard did you try to avoid politics? I doubt it was very hard.
This strikes me as a form of "logical rudeness". Since these are "just little examples", you get to make little jabs at people you disagree with by tarring them with a strawman argument. And if anyone responds to these rude little dismissals of enemy political ideas (and all of these "worst arguments" happen to be things most LWers oppose) , they look like idiots who are missing the point of the article.
If you want to avoid a comments derail (i.e. eridu's), avoid insulting people's political ideas. Frankly, this post should have been downvoted to oblivion for that alone. Congratulations though, you managed to hit the usual LW applause lights well enough to avoid that.
I think it is time to think critically about LW's "politics is the mind-killer" meme. As interesting as it is to discuss free will and AI metaethics (no sarcasm, I do think they're interesting), there are two main things for which LW really, genuinely has the potential to be quite useful: (1) instrumental rationality as applied to daily life (what job to take, what to invest in, personal ethics); (2) political issues.
Refusing to talk about the latter is missing out on a lot of low-hanging fruit.
Also, for what it's worth, all of these ARE instances of The Worst Argument. Yvain never implied that e.g., anti-abortion people are necessarily driven only by TWAITW, only that that particular argument "abortion is murder" is in fact TWAITW.
A fun special case of this is applying it to arguments:
Pointing out the that an author accepted money from interested parties is an ad-hominem argument.
I find this a pretty unconvincing analysis of the "X is murder" type arguments and "taxation is theft." My linguistic intuitions are pretty strongly on the side of all of those claims being false, or at least open to debate.
No one outside of extreme libertarians ever refers to taxation as theft, as far as I can tell.
And the reason we have a word "murder" that's separate from the word "killing" (or even "homocide") is that murder is taken to be, by definition, unjust. There doesn't seem to me to be anything too-clever about saying that, any more than there's anything too clever about saying that canines are a particular type of mammal.
Given the whole emphasis on rejecting a commonly accepted form of persuasion an even more proper response may leave out the clearly empty rhetorical demand.
First an admission: I did not read all the comments, there are too many for me, just the top 150 or so, so someone might have mentioned this before, if so never mind. This is for Yvain , an example of the worst argument I ever faced: The logic is as follows, since the sky is blue you are stupid. That is the end of the argument, since you SEE, you are stupid thus your argument is stupid. So what can you do then, except walk out? What can you do when one side is not only unreasonable, but irrational ?
I think most contemporary invocations of "That's racist!" are examples of the worst argument in the world so I'm not so sure about that.
Is it always illegitimate to bring up the facts that abortion ends a life, that taxes are not paid voluntarily, and that affirmative action benefits some races at the expense of others? If someone with right-wing views on these policies does weave one of these facts into his arguments on less wrong, will he be adoringly referred to this page as though it were a drop-dead refutation?
If in the course of an argument you become frustrated by the other party mentioning true things, perhaps he is making the worst argument in the world. And perhaps there is another explanation.
No. If you didn't get that, you should reread the post. The point is to discuss the relevant features of the subject in question, not say "murder" or "theft" and stop thinking.
That is the author's stated intention. But what he's created is an easily referenceable refutation of weak versions of strong arguments. You're concerned about someone crying "murder" or "theft" instead of thinking. I'm concerned about someone linking to this very popular article instead of thinking, or anyway instead of grappling with a strong argument.
There is no general way to make people think. Everything can be misused.
As Alex Mennen brought up on my blog, the problem you're worrying about is that someone will say "That's an example of the Worst Argument In The World, which is typically a weak argument", even though that particular version of the argument is actually quite strong.
Luckily, I hear there's a new post addressing exactly that error!
No and no (hopefully) and yes and yes.
The examples you give bring up specific points of those, the specific facts that are negative. The Worst Argument In The World is when you don't state the particular negative fact, but instead (truthfully) proclaims that X is part of larger set Y which notoriously also contains that one specific negative fact (which X really does have), but also many others which give Y a large net negative value, making X have a large net negative value (to uninformed audiences) by virtue of being part of Y.
Bring up the specific fact, not an arbitrary large subgroup which also contains both X and the specific fact and is known to have a massive net connotation.
In Lesswrong discussion, I've seen similar arguments made, and the most frequent response was "Taboo X and Y", followed closely by a more elaborate reduction.
Note, however that "X is in a category whose archetypal member has certain features", is strong evidence that X does in fact have those features. Thus the burden is on the person arguing otherwise to show that it doesn't.
Keep in mind that your brain's corrupted hardware is designed to fail in just this kind of "special pleading" situation. Or to put it another way there's a reason ethical injunctions exist.
Until some other bozo comes up with a different category. Then we get to play tennis.
This problem exists for all reasoning, e.g., Komogorov complexity falls apart when some bozo comes along with a different language.
I've got a brother-in-law who has used this argument often. We live in Australia, and unless you've been paying attention to the politics of refugees, immigrants and asylum in this country, this won't make much sense.
About 10 years ago, the Liberal Party (conservatives, ironically) put in place a policy (sending boat refugees to off-shore handling places to demotivate people to choose this route) and a directive (the navy to make sure boats never reached Australian shores, often by towing them out of Australian waters). Immigration by boat hence dropped dramatically, but the reason for that dropped was put on the introduction of the policy, treating the policy and the directive as the same category of "policy."
This lumping of various policies and directives into one encompassing category has had the unfortunate effect of showing all Australians that "the policy" was successful in demotivating people to hop into boats, when the reality was very, very different (and we don't know how many boats have sunk and how many people have died from this outside of Australian waters because of this; the Australian navy don't report on what happens outside their areas).
I'm not sure I follow where this ties into Yvain's "worst argument in the world".
Meta-note: Right now, as I check the top comments for today, all the top comments for today are replies to heavily downvoted comments. This is the behavior the downvoted-thread-killer was meant to prevent, but we don't yet have "troll-toll all descendants" feature. Noting this because multiple people asked for examples and how often something like it happened.
The eridu-generated threads show that the direct reply toll doesn't seem to work, or at least it didn't in this case. I still don't like the idea of the indiscriminate whole-thread toll, but I'm no longer expecting the current alternative to be effective.
I've thought of another option: maybe prohibit a user from posting anywhere in a subthread under any significantly-downvoted comments of their own? This is another feature of all bad threads that could be used to automatically recognize them: the user in a failure mode keeps coming back to the same thread, so if this single user is prohibited from doing so, this seems to be sufficient.
I was one of those who asked for examples. This is indeed a good example, and I take it to heart. I am still uncertain what the effect of the new and planned rules will be (troll feeding fee etc.). But it's now less a case of "what problem are you trying to solve?" and more "how should we solve this problem?"
In more detail: I missed this thread, but skimming the remaining comments, I think it would have been a waste of time to participate. But since many others did participate (while saying in many comments that eridu was quite irrational and/or wrong), it's possible I would have been drawn in if I had the opportunity. So I'm glad you stopped it.
At the time I make this reply, DanArmak's comment was downvoted (I voted it back up). Downvoting a comment like that above is the sort of reason why I am starting to distrust the behavior of meta-threads as a reliable signal of what the community thinks.
How else is one meant to categorise instances, other than by noting that they share features with training data?
Just because (e.g.) archetypal cases of theft have other things wrong with them doesn't mean that theft also isn't wrong qua theft. I think you're just choosing to favour some categories (e.g. benefit/harm) over others.
When you say,
I think you bed the question against those who oppose criminality qua criminality.
It's true that you should also consider the advantages of this specific case of theft. But individual exclaimations aren't meant to be complete arguments.
This implies that there is an intrinsic "wrongness" somewhere inside "theft" itself. Where, then, does the human hand reach into the vast void of existence to retrieve this wrongness to which theft is associated?
"Theft", the word, does not have any wrongness. Otherwise, we could use "Borbooka" instead. Let's do that. Does Borbooka have inherent wrongness? Well, what is Borbooka?
Borbooka is, apparently, when an item, which some animals apparently say verbally and apparently implicitly mutually agree is for the exclusive use of "one particular" animal, is moved from one point in spacetime to another point in spacetime such that another animal gains implicit exclusive use of this item without there being an apparent verbal exchange between animals that would apparently make them all understand that both animals "wanted" this item to be displaced thus.
Where, in the Borbooka defined above, is this mystical "wrongness" you insinuate? Are these not all simple conventions and agreements between said animals? Does Borbooka somehow create or destroy matter, or anything at all? If these conventions were not there and all the animals never had the implicit agreement that one item "belonged" to one animal, would Borbooka still be wrong? Would it still even exist?
I thought this was completely covered by a conjunction of the Metaethics and Guide to Words sequences.
Because there's an ethical injunction against it.
You may want to look at this post.
Given a general case, you should be able to argue about harms. Injunctions only come into play where you have some reason to rationalize a bad conclusion in an unusual-seeming case. As no society has so far collapsed due to lack of injunction against taxes, an injunction against all non-consensual-things-taking is unnecessary.
The point is that while rationalizing the conclusion doesn't seem bad from the inside.
This is very much debatable. If you look at actual collapsing societies throughout history, a large part of the problem is taxes strangling the economy.
I don't understand why you're making this point. It makes no difference to my point which metaethics you wrap it up in. Certainly, I had no intention of implying "that there is an intrinsic "wrongness" somewhere inside "theft" itself".
Then please answer my main question: Where does the "wrongness" come from?
How is a crime wrong simply by virtue of being a "crime"? If we defined crimes to be "every time you touch your own sexual organs", would it still be bad because it's a crime? Does the wrongness come from the fact that society agrees to punish crimes?
You're making a very strong claim. One that almost begs the question. "Theft is wrong because it is theft", one could put it. Where does the wrong come from?
If I have a device of unexplained origin that can supply food and unlimited electrical power to all of Earth's human population indefinitely, and am keeping it in a locked safe, with a notice saying "PRIVATE PROPERTY - DO NOT STEAL!", but I clearly have no intention of ever using or even acknowledging the existence of said device, and have stated publicly (with verified truthfulness) that I really don't get if my stuff gets stolen because I just steal other stuff in return... is it still wrong to "steal" it, because it is theft? If not, where is the line? Where does the wrongness come from?
I don't see how your argument could possibly be valid if theft and criminality do not have an intrinsic wrongness to them. If they don't have intrinsic wrongness, then something else which does not share all of their wrongness and has other goodness is clearly not to be lumped into the same point on the map.
Individual exclamations aren't meant to be complete arguments, but each argument is a soldier, and it is Dark Arts to make inefficient, biased human brains automatically associate two different things like this. The argument simply works in convincing people and "defeating" opponents, despite being "incomplete" and encouraging incorrect reasoning. Reduction is important here.
ETA: I'm not sure I still hold all the beliefs and agree with the content or form of this comment anymore, but removing it would make it difficult to understand the rest of the conversation. Take the statements above with the appropriate measure of critical thinking.
Many people who use the arguments mentioned would have different philosophical reasons for believing the claims. The Christian groups that claim abortion/euthansia is murder would appeal to it being wrong because only God has a right to kill, and the libertarians who argue that taxation is theft would appeal to the right to property. Both would refuse to argue on utilitarian grounds.
The libertarians I have read have all argued that people would in general be better off if libertarian principles were practised more. That is, they argue solely on utilitarian grounds.
This is a special case of Agree Denotatively, Object Connotatively.
Theft is however not solely harmful, obviously one party gains.
For most people I know, that is in the swedish libertarian community, theft is theft whether or not it has socially beneficial effects, because we use the definition that you gave; theft is taking from others without their consent. The implication is not that "As theft is always bad, it should be dismissed without a thought", because some libertarians do favor theft and are explicit about it, because they believe it's necessary. The moral breach of treating others as mere means to one's own goals can be (hypothetically for most) mended if it has other good consequences (or such). The point is that taxation is bad, which doesn't mean it should be dismissed out of hand, but it shouldn't be adopted out of hand! That is, taxation should be considered a bad, until it is proven necessary or otherwise positive.
This seems to presume that using others' property as a means to an end constitutes using others so, which seems dangerously close to question-begging the whole issue.
[citation needed]
It's bad when people use the dictionary to make political arguments, but it's worse when they write their own dictionary. For example:
Normal people define "selfishness" as "taking care of oneself, even if that means hurting other people." Objectivists define "selfishness" as "taking care of oneself, but never hurting other people." Hence, selfishness can never morally objectionable.
Normal people define "sexism" as "unfair treatment of a person based on their sex." Feminists define "sexism" as "unfair treatment of a person based on their sex, but it only counts if their sex has been historically disadvantaged." Hence, men can never be victims of sexism.
Normal people define "freedom" as "the ability to do a lot of stuff." Catholics define freedom as "the ability to do as God wishes." Hence, laws enforcing Catholic norms are pro-freedom.
That layer of indirection there is optional.
You think I could replace "unfair treatment of a person based on their sex, but it only counts if their sex has been historically disadvantaged" with "unfair treatment of a woman based on her sex" ? I don't think that would pass an ideological Turing test.
I am saying that the subset of feminists that are unsophisticated enough that they exclude unfair treatment of men from their definition of 'sexism' and yet sophisticated enough that the implicit definition in use is actually dependent on history is comparatively small.
I don't know what you're talking about. I'd wager 10-1 odds that women the "unsophisticated" feminists who ascribe to the "men can't be victims of sexism view" have more education than the general population. The historical dependency is taught in intro WS classes and feminism 101 blogs; they call it "power plus prejudice." Not all feminists agree with the redefinition, but more than a "comparatively small" number do.
Normal people define "true" as "good enough; not worth looking at too closely". Nerds define "true" as "irrefutable even by the highest-level nerd you are likely to encounter in this context." Hence more or less all of Western philosophy, theology, science, etc.; and hence normal people's acceptance that contradictory things can be "true" at the same time.
(Yes, I'm problematizing your contrast between various groups you dislike and "normal people".)
Not to mention that they define "hurting" as "damaging or destroying other's life, health or property by direct action" where normal people understand the word much more broadly.
Yvain, here is a challenge. Many of your examples are weak versions of strong right-wing arguments that you do not accept. (by your remark about Schelling fences, it seems you're aware of this). I challenge you to replace each of these examples with a weak version of a strong left-wing argument that you do accept. Since policy debates should not appear one-sided, there should be no shortage of weak arguments "on your side." And it would be an interesting kind of ideological Turing test.
Perhaps I'm wrong about "what side you're on" and you already accept the strong right-wing arguments. In which case you got me, well done!
This. The lack of examples from the left makes me uncomfortable sharing this article with people that will likely see it as an attack on their ideology. If they have some things to cheer for too, they are far more likely to accept it as a good post.
I have the impression that (1) when people post things in LW that are politically leftish, it's quite common for them to get a response along these lines -- complaining about leftward bias and suggesting that it should be addressed by a deliberate injection of rightward bias to compensate -- whereas (2) when people post things in LW that are politically rightish, they basically never receive such responses.
I have no statistics or anything to back this up, and it's not clear that there's any feasible way to get (or informatively fail to get) them, so I'd be interested in other opinions about whether this asymmetry is real.
If it is real, it seems to me quite interesting.
(One possible explanation, if it's real, would be that leftish views are much more common here than rightish ones, so that people with rightish views feel ill-treated and want the balance redressed. Except that I think I see distinctly more rightish than leftish political commentary here, and the rightish stuff more often gets large numbers of upvotes. I suppose it's possible that what we have here is a lot of slightly leftish people and a smaller number of rightish ones who feel more strongly. Again, this is probably hard to get a good handle on and I'd be interested in others' impressions.)
Off the top of my head:
Economic inequality is an unequal distribution of resources. The most salient example of this is an unequal distribution of resources that all have equal claim to, like a pie a parent bakes for their children. But [various convincing arguments in favor of at least some economic inequality.]
War is killing, which is bad because murder is bad. (Or eating meat, or capital punishment.)
Gay marriage is good because it's a right, and the most salient rights are good.
Welfare is good because it's a form of helping people, and helping people in ways that don't produce bad incentive effects and without taking from anyone else is good.
Processed food is bad because putting the most salient synthetic chemicals in food would be a really bad idea.
Note that "genetic engineering to cure diseases is eugenics" and "evolutionary psychology is sexist" are probably left-wing viewpoints, though not ones Yvain agrees with.
ISTM that categorizing many of those as "Left-wing viewpoints" or "Right-wing viewpoints" is a strong category error, one that we should attempt to reduce rather than redraw or blue boundaries. "Evolutionary psychology is sexist" is, afaict, a word error. It is not a position, but an implicit claim: "Because evolutionary psychology is sexist, it is bad, and thus evolutionary psychology is wrong!" - this is usually combined (in my experience) with an argument that the world is inherently good and that all humans are inherently equal and so on, which means that theories that posit "unfair" or "bad" circumstances are wrong; the world must be "good" and "fair". Stereotypicalism would call for a reference to religion here.
It may be a word error - I don't think it is, "Evolutionary psychology is riddled with false claims produced by sexist male scientists and rationalized by the scientists even though the claims are not at all well-supported compared to nonsexist alternatives" is a coherent and meaningful description of a way the universe could be but isn't, and is therefore false, not a word error - but if so, it's a word-error made by stereotypically left-wing people like Lewontin and Gould who were explicitly political in their criticism, not a word-error made by any right-wing scientists I can think of offhand.
In general, we should be careful about dismissing claims as meaningless or incoherent, when often only a very reasonable and realistic amount of charity is required to reinterpret the claim as meaningful and false - most people are trying to be meaningful most of the time, even when they're rationalizing a wrong position. Only people who've gotten in a lot more trouble than that are actively trying to avoid letting their arguments be meaningful. And meaningless claims can be dismissed immediately, without bringing forth evidence or counterobservations; meaningful false claims require more demonstration to show they're false. So when somebody brings a false claim, and you dismiss it as meaningless, you're actually being significantly logically rude to them - putting in less effort than they're investing - it takes more effort to bring forth a meaningful false claim than to call something 'meaningless'.
The sexism associated with evolutionary biology is typically the result of the perceived (or actual) claim that because sexual differentiation has a historical and evolutionary basis, it is morally correct to reinforce those differences today.
You can point out that that type of claim is not commonly made my evolutionary psychologists, but when lay people perceive that that claim is true and use it to justify sexist actions that they would not have taken in the absence of their perception of such a claim, then it is the case that evolutionary psychology contributes to behavior which unfairly discriminates on the basis of sex.
One of the key points is that "Evolutionary psychology is sexist" and "evolutionary psychology contributes to behavior which unfairly discriminates on the basis of sex" are very nearly the same statement, while "Evolutionary psychology is riddled with false claims produced by sexist male scientists" is a radically different statement.
Well, sure. But so do a million other things. After all, it would be much harder to discriminate unfairly on the basis of sex if we didn't have sensory organs capable of distinguishing an individual's sex, so the existence of such organs contributes to behavior which unfairly discriminates on the basis of sex. Unarguable.
Also uninteresting.
Surely a more important question is whether the study of evolutionary psychology differentially contributes to such discrimination? Which perhaps it does, but this takes more effort to demonstrate than simply pointing out that there exist lay people who use it to justify sexist actions.
Okay- I assert that there are almost zero people who seriously assert that 'Having sensory organs which can distinguish sex' justifies sexist actions, and that there are more than one hundred thousand Americans who demonstrably either claim, or allow the claim to stand, that EP justifies sexist actions that they themselves take.
I'm prepared to defend the second assertion if needed, which is why I choose a conservative number. The first assertion is trivial to falsify if you can find a significant number of people who believe that.
To be more logically complete, my unstated assumption: Lay people typically don't take actions which they believe to be unjustified.
I'd be interested in seeing this. Largely because I'm curious to see specific examples of what you consider unjustified sexism.
The most godawful example I've seen of EP being used as a cover for blatant sexism and misogyny is this NRO article, which basically says that as a rich boss with many male sons, Mitt Romney exudes alpha male power, and all women should fall in trance and vote for him.
Is your objection that the descriptive statement is false, or than it's sexist to say it even if its true?
Yes, how one's candidate appeals to voters' biases is not exactly something to brag about, but it's unfortunately a common occurrence in our political process.
Wait, are you asserting that sexism is ever justified? If so, we have a definition mismatch.
For a start, we have Forbes Magazine drawing a link from EP to why most women will never be CEOs (Never mind that most people will never be CEOs). I haven't yet demonstrated how many readers of Forbes allowed the claim that EP justifies the sexist treatment of executives, and also take sexist actions regarding executives; will you accept that 5% of board members of publicly traded companies make sexist decisions about executives, and that 80% of those people read Forbes and didn't object to that (4% of board members overall)? (again, I'm using numbers that I think are conservative, because direct measurements are hard.)
Since I specified unjustified sexism, you'll have to provide an argument for why said justification is incorrect.
I agree that many Americans assert that EP justifies sexist actions.
I agree that effectively nobody asserts that having sensory organs which can distinguish sex justify sexist actions. Nor did I claim anyone did.
My assertion was, and is, that the presence of those organs, much like the presence of those justifications, contributes to sexist actions that would not occur in their absence.
I assumed your objection was to the sexist actions, and the justifications were objectionable merely because they enabled those actions. In which case it seems that anything else that equally enabled those actions would be equally objectionable.
But, sure, if you're concerned specifically with asserted justifications rather than with the actions themselves, then I'm entirely beside your point.
Unrelatedly: lacking a justification for X != believing X to be unjustified.
I'm not sure what you mean by "reinforce", but it seems reasonable to take these differences into account when making decisions.
For example, suppose that evolutionary science has determined that is was pro-survival in the past for females to refrain from occupations which had high fatality rates.
Reinforcing that would be claiming that females should refrain from or be prohibited/discouraged from those occupations in the present and near future.
Also sexist is the line of thought "Females are statistically more/less likely to be X, therefore I require that it be a male/female who performs task Y.", when variation within each sex is great enough that there are a very large number of one sex who outperform a typical member of the other; a specific example would be "Females are less likely than males to complete a degree in mathematics; therefore it makes sense to award this scholarship to the equally qualified male instead of the female".
That's not the relevant comparison. In practice the comparison is between an above average members of each sex.
In your example, than depends on whether the first clause is still true after controlling for whatever qualifications are used in the second.
This is a worthy steel-manning when trying to reach an accurate conclusion about ev-psych, but I think you give the typical person who claims "ev-psych is sexist" too much credit here.
Natasha Walter makes the argument that Eliezer refers to in Living Dolls (not really about ev-psych, but about the idea of innate differences between genders in abilities), and I'm sure there are other examples (I haven't actually read all that much feminist writing). However, I have also encountered people who won't even discuss the issue with anyone who is pro-ev psych because they think that they're so morally appalling. Not sure how typical the people I'm encountered are though - I suspect they may be more extreme than most, and the most extreme people are the loudest.
Thanks for the link to Caplan's post, it's a very nice thought experiment. How about a thread where right-wing folks can give their strongest versions of left-wing arguments and vice versa, all the while quietly laughing about each other's misconceptions but not stepping in to correct? I could give it a try, as a right-winger imitating a left-winger, but I'd probably just embarrass myself.
I don't understand how you get from "policy debates should not appear one-sided" to "there should be no shortage of weak arguments 'on your side'". Especially if you replace the latter with "there should be no shortage of weak arguments of this sort on your side" -- which is necessary for the challenge to be appropriate -- since there could be correlations between a person's political position and which sorts of fallacies are most likely to infect their thinking.
In particular, I predict WAITW use to be correlated with explicit endorsement of sanctity-based rather than harm-based moral values, and we've recently been talking about how that might differ between political groups.
I think this is because of the way you're deconstructing the arguments. In each case, the features you identify which supposedly make us dislike the arcetypal cases are harm-based features. Someone who believed in sanctity instead might identify the category as a value in itself. Attempts to ascribe utilitarian-style values to them, which they supposedly miss the local inapplicability of, risks ignoring what they actually value.
If people genuinely do think murder is wrong simply because it is murder, rather than because it causes harm, then this is not a bad argument.
Absent any reason to do so, disliking all murders simply because they are murders makes no more sense than disliking all elephants simply because they are elephants. You can choose to do so without being logically inconsistent, but it seems like a weird choice to make for no reason. Did you just arbitrarily choose "murder" as a category worthy of dislike, whether or not it causes harm?
At the risk of committing the genetic fallacy, I would be very surprised if their choice of murder as a thing they dislike for its own sake (rather than, say, elephants) had nothing to do with murder being harmful. And although right now I am simply asserting this rather than arguing it, I think it's likely that even if they think they have a deductive proof for why murder is wrong regardless of harm, they started by unconsciously making the WAITW and then rationalizing it.
But I agree that if they do think they have this deductive proof, screaming "Worst argument in the world!" at them is useless and counterproductive; at that point you address the proof.
Ok, so replace "abortion is murder" with "abortion harms the fetus".
Absent any reason to do so, disliking instances of harm simply because they are instances of harm makes no more sense than disliking all elephants simply because they are elephants.
I don't want to assume any metaethical baggage here, but I'm not sure why "because it is an instance of harm" is an acceptable answer but "because it is an instance of theft" is not.
Excellent idea. It would be beneficial to how the community deals with politics, something that I've been very concerned about recently, to see this written out.
Almost 400 comments but not a word of discussion of the parsing Yvain provides for his seven examples! But if Yvain's parsing is wrong—as I think it is—then his analysis will serve to further bias our understanding of positions we disagree with and to forsake any charity in understanding these positions.
The question that is fairly asked of Yvain is what distinguishes his "worst argument" ("X is in a category whose archetypal member has certain features. Therefore, we should judge X as if it also had those features, even though it doesn't.") from any form of rule-governed reasoning in ethics (whether deontological or rule-utilitarian). When the examples are expanded and recast in those terms, they do not express Yvain's "worst argument"; they rather simply express moral premises subject to disagreement.
Taxation is theft. I'm no libertarian, but the argument isn't that taxation shares features with "archetypal" theft but that any taking of unearned property is wrong for the same basic reasons as "archetypal" theft is wrong, whether natural law or utilitarian calculus.
Abortion is murder. The claim almost always comes from a fundamentalist religious direction. Abortion is said to be murder not because it shares features with prototypical murders but because it is the same in the essential respect that it involves killing an innocent possessed of a soul. It's stupid enough as it stands; no reason to misrepresent it.
King is a criminal. If the psychologist-of-morality Kohlberg is believed, most adults in the U.S. identify morality with authority. Many believe it is immoral to break a law enacted through democratic procedures. Those who reason this way are indeed philistines, but their problem isn't with some formal fallacy in reasoning but with their premises.
Evolutionary psychology is sexist People who argue in these terms usually think it is wrong to "reinforce sexual stereotypes" even if they're true.
And so on. If you were arguing with someone defending their position in the ways Yvain summarizes, would you point out, say, that the archetypical murder is a lot different from abortion; or would you point out that souls don't really exist (or point to similar defective assumptions)? The first would miss the point. An answer to the antiabortion argument has a similar form to the "worst argument," but I think the answer is sound: Compelling women to remain pregnant is involuntary servitude. (More popularly, No forced labor.) The question is whether the essential features of abortions, relative to a well-chosen framework, are best captured by analogy to murder or slavery. We always reason by some sort of analogy; the question is whether the given analogy is adequate. Yvain's proscriptions consistently carried out would toss analogical reasoning in ethics.
You know who else made arguments? Hitler.
I purposefully use this version of worst argument in the world when talking about homosexuality/homophobia or atheism, etc.:
You know who else didn't like gay people / atheists / Communists? Hitler
Time for the meta:
You know who else uses versions of the worst argument in the world when talking about relevant topics? Dark Lords.
No, Hitler didn't make arguments, he made assertions; and you know what else was an assertion? Your comment!
All of the arguments are of the form A is an X, when A is not a typical example of X. Here are some arguments that are of that form.
-"Having sex with an passed out stranger is rape."
-"Sleep deprivation/sensory deprivation/stress positions is torture."
-"Writing and cashing bad checks is theft."
Are these all instances of the worst argument in the world? If they aren't examples of the worst argument in the world, why not?
If the main reason that these arguments are acceptable is our disapproval of A, then your worst argument in the world is not a valid. It is just a way to discount rhetoric you don't like.
Consider a X that is bad for reasons R1, R2 and R3. R1 and R2 are really strong, while R3 is quite minor.
Consider an atypical case of X, A, which has only the reasons R1 and R2. Saying "A is X" doesn't do much harm. The real reasons for which you reject X (R1 and R2) are present in A, so saying "A is X so A is wrong" is acceptable.
Now consider another atypical case of X, B, which only share R3. Saying "B is X so B is wrong" is using the emotional power of the horror of R1 and R2, which B doesn't have, against B, just because B can be said to be part of a cluster in which the typical elements have it. That's a really fishy argumentation. That's what Yvain called "the worst argument in the world", because it's wrong but convincing, and very hard to counter in a debate (it requires deconstructing "why is X is bad", extracting R1, R2, R3, showing that B only shares R3, so may be slightly bad, but not nearly as much a typical X).
Let's analyze the first one : "Having sex with an passed out stranger is rape."
Rape is very bad, I hope we all agree with that. Why is rape bad ? It's bad for many reasons. Some of the reasons (that it violates people's freedom of choice, that there are risks of pregnancy and STD, that it humiliates the victim, ...) hold for "having sex with a passed out stranger". Some other of the reasons against the typical rape (that it involves violence and threats, inflicts pain and fear on the victim, ...) may not apply for "having sex with a passed out stranger". Depending which of the two sets of reasons are you true rejection of rape, it's the WAitW or not.
But the main point is that "Having sex with an passed out stranger is rape." is not the reason why "having sex with a passed out stranger" is a bad thing to do. The reasons why it's a bad thing is because it doesn't respect the person's freedom of choice, because it risks exposing her (or him) to danger of STD/pregnancy without her consent, because it's likely to humiliate her, ... Those are the real justifications of why it's wrong.
If you state your position as "Having sex with an passed out stranger is rape." it's very to argue if you're right or wrong about saying it's bad thing to do. First we are stuck with the emotional weight of seeing a woman crying of fear and pain while her rapist rapes her violently a knife under her throat, which is not how "having sex with an passed out stranger" occurs, and then we'll end up engulfed into a debate about the definition of "rape" which is completely barren. If you give the real reasons then we can argue for each of them how bad they are, and we can end up knowing how bad "having sex with an passed out stranger" is.
"A is an X" is a valid argument for a lawyer in court, because law has to be written in words. But it's not a valid argument (and can lead to people making wrong conclusions) in a moral/political debate.
Reading through this and most of the comments, it occurs to me that virtually all the discussion is around MORAL argument. Do we ever succumb to the worst argument in the world in non-moral issues?
"How can we fix this light bulb?" "Use a hammer. A hammer is a tool and tool's fix things."
There it seems like an incomplete argument. And then somebody comes along and nails a few pieces of wood together with a hammer producing something stable enough to stand upon, and then proceeds to stand on it to reach the lightbulb, which just turned out to be loose so they tighten it, fixing it.
Staying out of the moral sphere, what is the worst argument in the world? Some of the first bad arguments that come to mind are 1) Because it's in the bible, 2) God said so, 3) My professor said so, (and to a much lesser extent) 4) Because Feynman said so.
I might vote for "Because X said so" as the worst argument in the world. For moral and non-moral questions, actually.
"Rapture of the nerds".
I'm confused about how to interpret your comment. Do you mean that people saying "Singularitarianism has some of the features of belief in a religious apocalypse, and believers in apocalypses are crazy, therefore Singularitarianism is crazy" is an instance of the worst argument in the world, or something else?
Yes.
Guilt by association, as has been mentioned before, is probably a better name.
I'm not sure "This specific example has something to recommend it" saves the example from being legitimately described in terms of the category. I'm minded of "Yeah, I killed him, but he deserved it" - that is, everybody thinks their example has something to recommend it, something that makes it distinct from the categorical description, that's why they support it/did it to begin with.
And George Washington was a traitor. ;)
I'm pretty sure the definition of 'traitor' includes "and lost" in there somewhere!
I'm not sure it's that tiny, especially once you're using the "steel man" version of the arguments; i.e. things like "Schelling fences" do not often appear in the reasons given for the disagreement, but that can still be what it boils down to.
People who object to abortion may be objecting to a weakening of the social stigma against the murder of innocents - that social stigma performs a useful function in society, so allowing anything that could be described as "murder of innocents" is perceived as bad, regardless of whether that thing is in itself bad.
In other words, even if words are hidden inferences with leaky generalizations etc. - social norms are still defined in terms of words, and so "pointless" debates over definitions still have their place in discussions of morality. Questions that shouldn't be morally relevant ("is abortion murder?") become so because of the instrumental value of social norms.
So yes, sometimes pulling out a dictionary in the middle of a moral argument may be justified. The discussion can then turn to something more useful, like "is it worse if the norm against murder is slightly weakened, or if women have to keep children they don't want?".
I want a Generalized Emergency Taboo button for just such cases; press the button and everyone is banned from using the word "murder" when talking about "abortion".
That way, in the future, we could talk about abortion using "abortion", and murder in general using "murder", whether abortion is murder or not, without weakening social norms in the process.
Or maybe Beisutsukai already have such a button? Perhaps they need a high enough level to unlock the skill? I had an idea of some third option we could use here to counter the social norm issue when I first read this, but got distracted and forgot it before I could follow up. Anyone else got any such ideas?
Even if that is true (and I stick to my guess that it's only a tiny fraction of the time) I still think deconstructing the argument is valuable. If people's true rejection of abortion is Schelling fences, then let's talk Schelling fences! I would ask why birth doesn't also work as a Schelling fence, and I would get to hear their response, and maybe one of us would change our mind.
But if their true rejection is based on Schelling fences, and instead they're just saying that abortion is murder, there's not much we can do except play Dueling Dictionaries. And the reason that has no chance of working ("Really? Merriam-Webster defines murder as killing a human after birth? Guess I'll go NARAL!") is directly related to it not being their real issue.
I agree that talking Schelling fences is usually more productive, and that it's probably not people's true rejection on abortion (norms around sexuality and fertility probably play a bigger role). Note also that unlike you, I never saw an "abortion is murder" sign in real life, and don't remember the topic ever coming up in real life.
Schelling fences probably play a bigger role for "justifiable killing" (like self-defense, the death penalty, euthanasia), where having a strong norm against killing in general discourages revenge killings (anti-abortion seem to be trying to hijack that norm to cover a case that doesn't fall under "killing" nearly as naturally). "Racism is bad" is another case where the norm is pretty valuable and useful in itself, and acknowledging that "non-bad cases of racism are not bad" would weaken it.
Eh, it probably depends of the reference class you're picking, and how charitable you're willing to be in interpreting people's reasons. when deconstructing a WAitW, it may be worth directing the discussion to one on Schelling Fences / norms etc., both as a way of raising the quality of the discussion, and of leaving a line of retreat.
There are real-world examples that could be described as getting the "dictionary" changed — for instance, the successful campaign to remove homosexuality from the Diagnostic and Statistical Manual of Mental Disorders, the American Psychiatric Association's "dictionary" (as it were) of mental illnesses.
Apple uses the WAITW when commenting on the Apple vs Samsung case:
"In a statement the firm [apple] thanked the jury for sending 'a loud and clear message that stealing isn’t right' "
Source: http://www.economist.com/blogs/schumpeter/2012/08/apple-versus-samsung?spc=scode&spv=xm&ah=9d7f7ab945510a56fa6d37c30b6f1709
Related to: List of public drafts on LessWrong
Draft of a critical response to this article
The worst argument in the world already has a different name. Philosophers call it the logical fallacy of Accident.
Calling out the worst argument in the world is not useful in practice. It is really hard to stop it from being a fully general counterargument against any high level abstract argument. The article seems to hold that for communication to work properly all statements must refer to “archetypes”, central members of a cluster in thing space. If so, this conflicts with the very idea of parsing reality into clusters-in-thingspace, which is inevitable. Every cluster, being a cluster and not a point, has more and less central members. If arbitrarily marginal members of clusters are invalid members, arbitrarily many things said by humans are The Worst Argument In The World. To banish statements that don’t locate one cluster-in-thingspace right into the centre of another cluster-in-thingspace is faulty, especially when the statements are slogans and the words highly abstract. To use it properly you have to come up with an argument that shows that either the rule or generalization you are attacking is wrong or the case considered is sufficiently exceptional that it no longer applies. I wouldn't trust myself to use that line of reasoning against an argument I already dislike to discount it. And if this really is a way to defend oneself from the dark arts as it presents itself doing, it should be good for precisely that! The article seems much more well made as a weapon to add to that arsenal but then it should be marked as such.
I see a (subtle but significant) difference between Aristotle version and Yvain version.
In Aristotle version, it goes like "doing A is X", "B does A" so "B is X". That's wrong, because (but Aristotle didn't know it) words are not precise definitions but fuzzy clusters. That's the main for which the "fallacy of accident" is a fallacy. And surgeons are not criminals.
The Yvain version is much more subtle. It acknowledges that words are fuzzy clusters, not fixed definitions. And that you can, without it being a fallacy (unlike in the first case) make claim like "abortion is murder" or "death penalty is murder". But that even if that claim can be make (even if we can consider them to be part of the fuzzy cluster) it's still a fallacy to use it as an argument, because while they are part of the cluster, they only share some of the problems that a typical member of the cluster has.
Now, if you consider my own point of view on those issues (but it could symmetric) : I'm pro-choice and against the death penalty. The WAitW idea is that I shouldn't argue for the right to abortion by trying to prove "abortion is not murder" and against the death penalty by trying to prove "death penalty is murder", being stuck in a definition match which is pointless, but that I should look deeper, dissolve what "murder" is and what it's assumed to be wrong, and show that most of what make us reject murder doesn't apply to abortion, and most of what make us reject it applies to death penalty. Or even completely discard the "murder" concept, and just look from a consequentialist point of view about the good and bad consequences of both.
The use of the Worst Argument in the World in practice is as a heuristic for tabooing words that don't fit very well (and hence leaking inapplicable/misleading connotations). You are not refuting arguments with it, you are drawing more attention to certain parts and calling for unpacking. A good argument should be unpackable to significant degree, but in practice it's too much work to unpack everything, so it's useful to have heuristics that would point where to start digging.
If the argument remains sensible after you unpack, then there is no problem. The bad thing is when an argument was relying on not being unpacked and crumbles once you look inside. So the refutation or the lack thereof depends on what happens after you unpack, the heuristic for deciding what to unpack doesn't itself perform any refutation.
Judging from the comments this is receiving on Hacker News, this post is a mindkiller. HN is an audience more friendly to LW ideas than most, so this is a bad sign. I liked it, but unfortunately it's probably unsuitable for general consumption.
I know we've debated the "no politics" norm on LW many times, but I think a distinction should be made when it comes to the target audience of a post. In posts aimed to make a contribution to "raising the sanity waterline", I think we're shooting ourselves in the foot by invoking politics.
Reading that HN thread, the problem appears to be a troll (who also showed up on Yvain's original blog post).
I've seen an even worse argument: Imagine the worst possible consequences of the other side's policies. Assert that the other side (or at least its leaders) intend those consequences.
I just registered http://worstargumentintheworld.com - it redirects to this post, and should be available shortly. Much easier to mention in conversation when other people use this argument, and don't believe it's a "real thing."
Great piece of work, Yvain - it's now on my list of all-time favorite LW posts.
"Real things" have their own domain. I registered this domain, therefore...