Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.
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.
"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.
"Affirmative action is racist!" True if you define racism4 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.
Now, because most arguments are rapid-fire debate-club style, sometimes it's still useful to say "Affirmative action isn't racist!" At least it beats saying "Affirmative action is racist but nevertheless good", then having the other side say "Apparently my worthy opponent thinks that racism can be good; we here on this side would like to bravely take a stance against racism", 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.
And what's going on should be no mystery to anyone who has read through the excellent Less Wrong Sequence On Words. Words are (jargon warning) hidden inferences, which form a leaky generalization over a set of cases that cluster along certain dimensions but may vary widely in their other characteristics. Because people feel like words are a single monolithic whole, arguments about the world tend to devolve into arguments about definitions of words (like "murder"), as if those definitions determined reality. To escape such arguments, the participants need to taboo that particular word and replace the symbol with the substance, which often means dissolving a term into its component inferences and reasoning about each one individually. The Sequence on Words mentions it's a huge warning sign if "you pull out a dictionary in the middle of an empirical or moral argument"; this post explains what that sign is warning you about.
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 racism even when it would be beneficial because that would make it harder to distinguish and oppose the really bad kinds of racism. Or it can be an attempt to spark conversation by pointing out a potential contradiction: for example "Have you noticed that affirmative action really does contain some of the features you dislike about more typical instances of racism? 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 racism!"
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.
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: Likewise, some people have attempted to "solve" this problem by saying that only privileged majority groups can commit "racism", and that it's literally impossible for a minority to be racist against anyone; if a minority did something we would call racist in a majority group member we would have to invent another name for it, like "unjustified ethnic hatred". This is a sufficiently bizarre piece of mental gymnastics that I think it forms part of the argument why proper recognition of the noncentral fallacy is necessary.