SELF-RECOMMENDING!

Style Guide: Not Sounding Like An Evil Robot

The saying goes: “Never offend people with style when you can offend them with substance”. This is the same idea as “weirdness points”: you can only bother people a certain amount before they go away. So if you have something important to bother them about, don’t also bother them in random ways that don’t matter.

In writing about science or rationality, you already risk sounding too nerdy or out-of-touch with real life. This doesn’t matter much if you’re writing about black holes or something. But if you’re writing about social signaling, or game theory, or anything else where the failure mode is sounding like an evil robot trying to reduce all of life to numbers, you should avoid anything that makes you sound even more like that evil robot.

(yes, people on the subreddit, I’m talking about you)

I’m not always great at this, but I’m improving, and here’s the lowest-hanging fruit: if there are two terms for the same thing, a science term and an everyday life term, and you’re talking about everyday life, use the everyday life term. The rest of this post is just commentary on this basic idea.

1. IQ -> intelligence. Don’t use “IQ” unless you’re talking about the result of an IQ test, talking about science derived from these results, or estimating IQ at a specific number. Otherwise, say “intelligence” (as a noun) or “smart” as an adjective.

Wrong: “John is a very high-IQ person”
Right: “John is a very smart person”.

Wrong: “What can I do if I feel like my low IQ is holding me back?”
Right: “What do I do if I feel like my low intelligence is holding me back?”

Acceptable: “The average IQ of a Nobel-winning physicist is 155”.
Acceptable: “Because poor childhood nutrition lowers IQ, we should make sure all children have enough to eat.”

2. Humans -> people. This will instantly make you sound 20% less like an evil robot. Use “humans” only when specifically contrasting with another animal:

Wrong: “I’ve been wondering why humans celebrate holidays.”
Right: “I’ve been wondering why people celebrate holidays.”

Acceptable: “Chimpanzees are much stronger than humans.”

3. Males -> men, females -> women. You can still use “male” and “female” as adjectives if you really want.

Wrong: “Why do so many males like sports?”
Right: “Why do so many men like sports?”

Acceptable, I guess: “Why do male sports fans drink so much?”

Use “males” and “females” as nouns only if you’re making a point that applies across animal species, trying overly hard to sound scientifically credible, or arguing some kind of complicated Gender Studies point that uses “man” and “male” differently.

Acceptable: “In both rats and humans, males have higher testosterone than females.”

4. Rational -> good, best, reasonable, etc. See eg here. Use “rational” when describing adherence to a good cognitive strategy; use “good” etc for things that have good results.

Wrong: “What is the most rational diet?”
Right: “What is the best diet?”

Wrong: “Is it rational to invest in bonds?”
Right: “Is it a good idea to invest in bonds?”

Acceptable: “Are more rational people more likely to succeed in politics?” (if asking whether people who follow certain cognitive rules like basing their decisions on evidence will succeed more than those who don’t. Notice that you cannot sensibly replace this with “good” or “best” – “Are better people more likely to succeed in politics?” is meaningless (unless you switch to the moral value of “better”)

5. Optimal -> best. I feel kind of hypocritical for this one because the link above says to replace “rational” with “optimal”. But if you really want to go all the way, replace “optimal” with “best”, unless you have a specific reason for preferring the longer word.

Wrong: “What’s the optimal way to learn this material?”
Right: “What’s the best way to learn this material?”

6. Utility -> happiness, goodness. Use utility only when talking about utilitarian philosophy.

Wrong: “Will getting more exercise raise my utility?”
Right: “Will getting more exercise make me better off?”

Wrong: “What is the highest-utility charity?”
Right: “What is the best charity?” or “Which charity helps people the most?”

The same applies to “utility function”.

Wrong: “My utility function contains a term for animal suffering.”
Right: “I care about animal suffering.”

7. Autistic -> nerdy. Use autistic when referring to a psychiatric diagnosis or a complicated package of sensory and cognitive issues. Use “nerdy” when referring to people who are book-smart but lack social graces.

Wrong: “Haha, my friends and I are so autistic, we talk about physics all the time.”
Right: “Haha, my friends and I are so nerdy, we talk about physics all the time.”

8. Neoreactionary -> right-wing, far-right, reactionary. Use neoreactionary when talking specifically about the philosophy of Mencius Moldbug, if you think you’ve looked into it and understand it. If you’re just referring to far-right ideas, use far-right.

Wrong: “I disagree with neoreactionary ideas like traditional gender roles.”
Right: “I disagree with right-wing ideas like traditional gender roles.”

In general, beware of attributing very broad and complicated ideas to local bloggers. Local bloggers often repackage or reinterpret larger tendencies from outside philosophy. This is useful and important work, but if you ever say anything that could be interpreted as identifying them as inventing those tendencies, people will jump on that and make fun of you. There’s not a hard and fast line between inventing a (specific) idea and reinterpreting a (broader) idea. But most local bloggers aren’t glory hogs, so they won’t get mad if you err on the side of non-attribution, or just linking to their good explanations of these ideas without specifying they’re the inventors.

Relatedly, Blue Tribe -> Democrats/liberals/leftists, Red Tribe -> Republicans/conservatives/rightists, almost always. When I coined those terms I was trying to explain how Democrats/Republicans were the tip of an iceberg of related traits, but now that the message has sunk in I think it’s reasonable to call that iceberg by the name everyone else uses.

9. High probability of -> Probably. Use “high probability” when discussing probability theory; use “probably” when discussing the real world. You can also say “most likely”.

Wrong: “There’s a high probability I’ll get the job.”
Right: “I’ll probably get the job.”

Saying “50-50”, “90 percent change”, or “99 percent chance” casually is probably okay. In popular writing, you should avoid any more specific use of a numeric probability (like “70% chance I’ll get the job”) unless you really have some good reason for thinking it’s 70 rather than 75 (like that you used some kind of algorithm to calculate it), or unless you think your intuitions can really distinguish between 70 and 75 percent probabilities (superforecasters can!) and want to make a very formal prediction.

10. Meme -> idea, belief. Use “meme” only when analogizing ideas to organisms undergoing evolutionary selection (or for silly cat pictures).

Wrong: “Pro-military memes are everywhere in America”.
Right: “Pro-military beliefs are everywhere in America”.

11. Status -> popularity, respect. My feelings on this one aren’t quite as strong; if you really want to use “status”, use “status”. But I think if you’re trying hard to appeal to ordinary people, you’ll want to take it out.

Wrong: “He’s high status in this community.”
Right: “He’s well-respected in this community.”

Wrong: “Bob was really low status in high school”
Right: “Bob was really unpopular in high school”

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

560 Responses to Style Guide: Not Sounding Like An Evil Robot

  1. Ashley Yakeley says:

    “Males” is OK to mean “men and boys”, etc.

    • deciusbrutus says:

      I’m not sure if saying things like “Males have higher incidence of colorblindness” is part of a ‘complicated gender studies thing that treats `men` and `males` differently’ or not.

      I don’t know if transmen have higher incidence of colorblindness than other females, but if they do it’s not because they lack x-chromosome redundancy.

      Pretending that trans people don’t exist in casual language isn’t an okay default state, not when using the word that describes what you mean is just as easy as using a word that doesn’t but is more popular.

      Likewise, don’t claim 99% credence in something (even using words to that effect rather than jargon) if you don’t have it. “Virtually certain” means literally exactly the same thing to people who don’t have calibrated senses of probability, so use that instead; if you’re talking to someone without a calibrated sense of probability and you actually have 99% credence in something, it is impossible to communicate that- they don’t have the sense that you are trying to communicate to.

      • Deiseach says:

        Pretending that trans people don’t exist in casual language isn’t an okay default state, not when using the word that describes what you mean is just as easy as using a word that doesn’t but is more popular.

        And how does using “males” rather than “men” in your example make that any more clear? The vast majority of people are going to translate “males” to “men”, and all you will have done – as Scott is pointing out – is come across as a robot poorly programmed to understand how those strange mammals communicate. Unless you’re going to append a footnote in writing, or veer off into a subordinate clause when speaking, to make it very clear that “I am saying ‘male’ because trans men exist and I’m a virtuous non-phobe who wants everyone to recognise that I’m woke and not one of the unregenerate”, then as I said – most people are going to take that as “Men have higher incidence of colourblindness than women” and you’re simply a nerd who likes to use pseudo-scientific language – ‘males’ – to sound smart.

        Just as you don’t know if there are many/some/any colour-blind trans men out there, neither do I know their opinion on the matter of language, but I’d imagine an ordinary trans man who just wants to be treated as ‘one of the guys’ is not going to throw a foot-stamping hissy-fit over “you said ‘men’ instead of ‘males’, that’s excluding me!” but rather that they’d be happy enough to be included under “colour-blind men” in ordinary discourse because, y’know, I’m a guy. Now, if it’s a matter of getting medical treatment or whatever, then sure, the distinction is important, but if you’re talking to ordinary people about ‘incidence of colour blindness in the general population’ it’s not worth splitting hairs over, except to score wokeness points.

      • yildo says:

        I think you have that backwards. In regular conversation, it’s the other way around — “men” is the inclusive term because regular conversation isn’t about prevalence of conditions across groups.

      • jermo sapiens says:

        Pretending that trans people don’t exist in casual language isn’t an okay default state

        11. okay default state -> acceptable
        Wrong: “Pretending that trans people don’t exist in casual language isn’t an okay default state.
        Right: “Pretending that trans people don’t exist in casual language isn’t acceptable.

        Also, to address the substantive point, it’s been made quite clear over the last few years that there isnt a level of wokeness that will satisfy the SJW mob. Whenever the mob is appeased, the requirements just get stricter and stricter. We now have people unironically speaking of people with uteri and people with penises, when they want to talk of biological sex (because like biological sex is a social construct, maaaaan). We have people accused of bigotry for not dating people whose anatomy they are not attracted to.

        This is my proposal on this point:
        -men/women, male/female, boy/girl, keep the meaning they’ve had since forever
        -if you want to speak of trans people, use the prefix trans
        -nobody is pretending you dont exist, but in casual language, it is very inefficient to need to recognize a rare exception every time you need to refer to the more typical case.

        • LadyJane says:

          We have people accused of bigotry for not dating people whose anatomy they are not attracted to.

          Funny how I don’t know anyone like that, despite knowing a lot of trans people and being involved in the trans community in a very socially progressive city. These types only ever seem to show up in the anecdotes of cultural conservatives.

          I don’t doubt that you can find people on the internet who’d say “you’re transphobic if you don’t want to date a woman with a dick,” some of whom might even be earnest about that statement. The internet is filled with all sorts of crackpots, after all. But to act like that’s a mainstream sentiment within the trans community is mistaken at best and disingenuous at worst. Within actual queer circles, there’s widespread acknowledgement that some people can only be sexually attracted to those with specific sets of genitalia, and there’s nothing wrong with that. I’ve even known a few fellow trans lesbians who aren’t interested in anyone who doesn’t have a vagina, and they’re not considered self-hating transphobes for that desire.

          • jermo sapiens says:

            Funny how I don’t know anyone like that, despite knowing a lot of trans people and being involved in the trans community in a very socially progressive city.

            I dont doubt it. But for those who dont live in such a setting, their only view of the trans community is Internet-woker-than-thou nonsense. And that is important, because that is how much of society will end up viewing the trans community.

            Unfortunately, nobody gains any woke points from fighting back against this, and in fact you stand to lose woke points by doing so. So the emerging picture is you have very visible, very loud trans people saying you’re a bigot if you dont date women with dicks, and the more reasonable (probably more numerous) trans people preferring to not risk alienating others by focusing on easy targets, like Trump supporters and such. The result will be that society at large will have an image of trans people based on what the very visible and very loud are doing.

            From my perspective, even if I take your word at 100%, I tend to doubt that when push comes to shove, the reasonable trans community (for lack of a better term) will effectively stand up to the not so reasonable trans community. The immediate personal incentives are just not aligned with that, even if the long term societal incentives are.

          • LadyJane says:

            @Aapje: Again, I’m aware that there are some trans activists out there who are actually bothered by things like the “cotton ceiling.” I never denied they existed, I was simply pointing out that they represent a minority in the trans community, to the point where I’ve never actually encountered one in person. And I don’t think your link provides much evidence to the contrary: It’s a post from seven years ago, by an author I’ve never heard of, on a site I’ve never heard of (which hasn’t been updated in three years and is effectively defunct). That hardly seems representative of modern trans-inclusive progressivism as a whole.

      • John Schilling says:

        Pretending that trans people don’t exist in casual language isn’t an okay default state

        Trans people represent less than one percent of the general population. Casual language routinely disregards the existence of groups with 1-5% representation, in the interest of getting as many people as possible on board with things that are 95+% true rather than aiming for narrow propagation of perfect information. I am unclear as to why the trans community should be uniquely privileged in this regard.

        Well, no, I’m not the least bit unclear on that point, but I want to see you spell it out or dance around it, whichever you see fit.

        • albatross11 says:

          I think I see both sides of this language issue.

          On one side, it seems backward to require that we change the word we’ve always used for “person born male with XY chromosomes and normal male genitalia who continues to identify as male” to something new like “cis-male” so that we can also change “male” or “man” to include transmen. It seems easier to just add “transman” as an additional term. (And then you get to deal with nonbinary people, which is somewhat messy with English and *much* harder in more strongly gendered languages.).

          On the other hand, I get the desire to expand the category so you can say “This restroom is for males” and mean “This restroom is for cis-men and also for transmen.” The problem is that whether or not the category should be so expanded is explicitly the thing people disagree about. (Like, if some women want to establish a women-only space, should transwomen be allowed in that space?)

          My sense is that there’s also a subset of people who would like to abolish certain concepts or categories, like distinguishing between people born male etc. etc. and transmen. And, people being people, some subset of them will happily interpret disagreement about changing the language to suit their agenda as being on board with bashing transpeople.

          • deciusbrutus says:

            “Man” didn’t change to include transmen, it always did.
            “Male” didn’t change to exclude transmen, it always did.

            The fact that there is a need sometimes to distinguish between cismen and transmen is what makes them useful categories.

          • Deiseach says:

            On the other hand, I get the desire to expand the category so you can say “This restroom is for males” and mean “This restroom is for cis-men and also for transmen.”

            I would have thought the problem there in that particular example was the whole row over bathroom laws and gender-neutral bathrooms; that trans people were agitating for the right to use the bathroom assigned to the gender they identify with. So if you’re going to use “males” to mean “cis men and trans men”, that defeats the whole purpose of “I want to be able to use the men’s bathroom because I am a man and I don’t want my trans status to be known or remarked upon”.

            I suppose there isn’t one general satisfactory term that will please everyone, because for every trans person who goes “yes! use males to be trans inclusive!” there will be a trans person going “no! the whole point is not to have me standing out as the different exceptional non-real man!”

          • albatross11 says:

            Deciusbrutus:

            Perhaps that distinction is clear to you, but it sure isn’t clear to me, despite having some familiarity with transpeople. Nor do I believe it would be clear to any substantial fraction of native English speakers.

          • deciusbrutus says:

            Oh, then just read a 099 primer on sex and gender. Not a weird gender studies thing, just a basic thing describing what sex and gender are.

          • just a basic thing describing what sex and gender are.

            Do you realize how arrogant that statement sounds to someone outside your ingroup?

            Try imagining an Objectivist or a Catholic making a similar statement, taking it for granted that his preferred use of terms is “what they are.”

            “To discover what sin is.”
            “To discover what rights are.”

            I just had an exchange with HBC where he was taking a term (big L libertarian) which deals with a subject primarily of interest to libertarians (distinctions among them) and has a well established meaning within that community, and using it for an entirely different meaning of his own invention.

            My response was not “read X to discover what a Libertarian is.” It was to suggest that using terminology in that way was likely to be confusing.

          • Dacyn says:

            The convention that “men”/”women” refers to gender and “males”/”females” to biological sex is not widely recognized, though. In general conversation either set of terms may refer to either of the distinctions. (E.g. my brother was confused by a recent newspaper headline indicating a man had given birth, until reading the body of the article made it clear that this was an FtM individual.)

          • thevoiceofthevoid says:

            Like all other words that people use in conversation, “men” and “women”, as well as “male” and “female”, refer to correlated clusters of attributes in thing-space, not explicitly-defined Aristotelian categories. (I have no idea how I’d translate that out of rational-speak. A Human’s Guide To Words significantly changed how I think about language and gave me concepts I don’t know how to express without parroting Eliezer’s use/abuse of math terminology.) “Men” doesn’t mean “people with penises” or “people who introduce themselves with male pronouns” or “people who have a Y chromosome” or “people who cut their hair short and wear shirts and trousers”. It means “the general cluster of people with those characteristics and more, who may not have all of them but should have most of the important ones.” Which are “the important ones” varies wildly depending who you ask, but won’t change the answer for most cases…

            There are of course, edge cases. People who fit some of the criteria, but miss others. If it’s just a few noncentral ones (e.g. a biological male with long hair and earrings who identifies as a man), we easily round off into the nearest category. But for cases where a lot of the criteria conflict (e.g. trans people), which criteria you think are central becomes important. Since different people see different characteristics as central to man-ness, this causes confusion and ambiguity. Since many situations require an official determination of gender and may confer benefits or detriments based on that classification, the issue becomes politically charged. But there is not, nor will there ever be, the One True Meaning of the Word “Men” [or of any other word].

            So, my recommendation is to use well-understood, commonly-used words when the edge cases aren’t important to your point (“Men are more likely than women to be colorblind due to their only having one copy of the relevant genes.”) and to use more specific terminology when they are (“Cis women experience more discrimination than cis men, but trans men and trans women both experience more discrimination than cis women do.”*)

            *This may or may not be true, I made up the sentence as an example.

          • Picador says:

            “normal male genitalia”? NORMAL???! Bzzzzt, you’re disqualified from wokeness. Please report to the nearest de-problematizing centre immediately.

          • deciusbrutus says:

            Once you undefine “malss” to a cluster with no identifiable edge, it ceases to be a group that can have an incidence of colorblindness.

            Since it’s useful to note that XY humans have a higher incidence of colorblindness than humans in general, and “male [human]s” already points to exactly that group, there’s no need to undefine “males” to refer to the entire masculine cluster.

          • albatross11 says:

            If I’m addressing a gender studies class, I’ll try to use the terms they understand. But addressing a group of people who have never had such a class and have no particular interest in the matter, I’ll do a better job communicating if I use language that they will understand, even if it’s not what the gender studies class would understand.

            And I’m telling you that this distinction you take as an obvious 101-level definition, in talking about something everyone deals with everyday, is not at all familiar to me. I’m probably more trans-aware than 95+% of the population. And yet, if you use that terminology to describe something to me, I won’t understand it, and neither will the other 95+% of the population.

            If you want to complain about my poor understanding of your terminology, fair enough. But if you want to communicate with people not in your ingroup, you probably need to adapt your language to what they understand.

          • Scott Alexander says:

            Picador is banned indefinitely

        • deciusbrutus says:

          Casual language routinely disregards the existence of groups with 1-5% representation, in the interest of getting as many people as possible on board with things that are 95+% true rather than aiming for narrow propagation of perfect information.

          Yeah, People aren’t Buddhist.

          • theredsheep says:

            No, there you’re explicitly saying something that’s not true about the whole, as opposed to just not including painstaking caveats. Compare:

            1. Humans are omnivorous animals. True, in spite of vegetarians and other groups with restricted diets.

            2. Humans walk on two feet. Except for the ones stuck in wheelchairs or beds, or who have one foot missing, etc.

            3. Humans are tool-users. Well, not the babies.

            Etc. In each case, it’s fair to use the first sentence in spite of its not being true for literally everyone, and there’s not much point in bringing up the exceptions except as an exercise in hijacking the conversation.

          • deciusbrutus says:

            Oh.
            People are non-Buddhist.

          • Dacyn says:

            I think the reason that we don’t say that is that there’s nothing that is Buddhist other than humans. So people are more Buddhist than any other animal, even if they aren’t very Buddhist.

          • Lambert says:

            Except that one fox that ignored causality.

          • theredsheep says:

            No. “People are non-Buddhist” is simply a slight reformulation of your original negative. A more reasonable comparison would be something like “people are [trait which doesn’t happen to fit Buddhists specifically].” I can’t think of any such trait at the moment because Buddhists, unlike the transgendered, are a very large group who fit in with the mainstream pretty easily.

          • theredsheep says:

            Suppose you’re in the grocery store, and the inane audio overhead says, “summer is here–BLT season! Head on over to the meat section to take advantage of our great deals on bacon!” Now, you happen to be a practicing Kosher Jew, and therefore have no desire to take advantage of any deals on bacon. You head on over to management instead, and complain about their exclusionary and offensive assumption that everyone eats pig products.

            This does not actually happen that I know of, because Kosher Jews do not belong to a demographic group which has implicit license to compel the whole world to account for them in every conversation despite their representing a very small percentage of the population. A Jewish person who did such a thing would merely be regarded as obnoxious. Matters would no doubt be different in Israel (or in India, if one swaps out beef and burgers for bacon and BLTs).

          • deciusbrutus says:

            Buddhists, unlike the transgendered, are a very large group who fit in with the mainstream pretty easily

            Which part are you asserting, that Buddhist people are a much larger group than Trans/nonbinary/agender/intersex people, or that they fit in with the mainstream much better?

      • Ozy Frantz says:

        I am sympathetic to the impulse to use “males” to refer to the category “dyadic cisgender men and dyadic transgender people assigned male at birth who have not biomedically transitioned”. However, literally no one will understand that you are referring to that specific group. It also bothers a lot of transgender people, because “male” is the adjective form of “man,” so it feels misgender-y.

        I believe right now the best way to be inclusive of transgender and intersex people, in booklength works, is to note early on that some but not all of the things you’re saying apply to transgender/intersex people. If you know how a sex difference applies to transgender/intersex people, it may be worth explicitly mentioning that Otherwise, just say “men” and “women”, and trust your transgender and intersex readers to understand how what you’re saying applies to our particular medical situation. (We already do this– “women can get pregnant” is not typically taken as excluding the existence of infertile women.)

        • jermo sapiens says:

          are you using dyadic as a synonym for binary? TBH, I had to google it, but I’m still not sure what you mean.

        • Simon_Jester says:

          I think a big part of how “women might get pregnant” and such are received is going to depend on context.

          A place where the local norms include “we’re not allowed to just forget that minorities exist as an excuse to avoid confronting bad things that happen to them” is going to be a place where isolated instances of word choice that occur in a background of general minority awareness aren’t as big a deal.

          It then only becomes a worry when it’s happening all the time. Or when the response to someone going “so when you talk about this rule, what is your stance on minority exceptions” is to go full disingenuous and say “I don’t believe those exceptions actually exist.”

      • Jake Rowland says:

        Lots of people have already pointed out issues with the meta point. On the topic of style, I think the point of this article is that you are spending your weirdness points on altering the norms surrounding gendered language and communicating certainty. And that’s fine if that’s your objective, but I think it’s useful to realize that this cuts into your weirdness balance.

        • deciusbrutus says:

          Yes. With the goal of creating a large enough community of people who actually give a damn about using the right categories that it’s the people who weaponize language in the particular manner I oppose have to spend their weirdness points to do so.

      • Furslid says:

        Pretending that trans people don’t exist

        Ashley didn’t specifically address trans people. How is that “pretending that they don’t exist”? My assumption was that ze knows that trans people exist but wasn’t specifically addressing trans people.

        You immediately jumped to a nasty accusation. In addition, you phrased it to sound as offensive as possible. This violates the norms of civilized discourse.

        It also demands special treatment for your preferred group. Either people can make generalizations (which necessarily exclude people) or they can’t. I doubt you find all generalizations worthy of similar attack.

        • deciusbrutus says:

          I can make a much stronger accusation, or even phrase that accusation more strongly.

          People can make generalizations, but building those generalizations into the language is a far worse offense.

          • Furslid says:

            We don’t build generalizations into the language. We don’t build the language at all. From each individual’s perspective, they are stuck with English as it is.

            It’s not reasonable to an individual for not redesigning the English language.

            It’s unreasonable to describe using English normally as ‘building those generalizations into the language’.

          • deciusbrutus says:

            NPCs don’t.

      • Trans people make roughly 1% of the population. So any broad generalization that applies to Cismen almost certainly applies to “men” as a collective.

        • deciusbrutus says:

          Which other 1%s of the population are acceptable to pretend don’t exist when defining sets? [third party] voters make up less than 1% of all eligible voters, for some third parties, therefore “eligible voters” can be defined exclusive of them?

          • Would you be bothered by “men find younger women more attractive”?

            Not true of gay men, perhaps not of asexual men, probably not of at least some otherwise ordinary heterosexual men.

            “Men are taller than women.”

            In ordinary speech, such statements are not intended to mean “every man likes” or “every man is taller than every woman,” nor are they taken that way.

            Taking your example, if someone wrote “at the moment, 52% of registered voters are Democrats, 48% are Republicans,” he wouldn’t be claiming that nobody is registered third party, even though that’s what it seems to logically imply. He would expect listeners to interpret that as “voters who are registered with the main parties.”

            “Men are better at map reading tasks than women” is a reasonable statement, and may even be true, despite the fact that I am terrible at such skills and my wife used to make her living doing three dimensional maps.

            What you are demanding is that because the subject of trans-people is of great importance to you, everyone else in the world should act as if it was important to them too.

          • Dacyn says:

            If someone said 52% of registered voters were Republican and 48% Democrat, I would assume they meant that approximately 0% of registered voters are independent or third-party. Media routinely say things like “48% Republican, 38% Democrat” (with the last 14% left ambiguous on purpose)

          • Broad generalizations typically mean that Not all X are like that is expected to be true. So you aren’t ignoring them per se, it’s more that whatever minority you are talking about is small enough that their existence doesn’t break the statement.

            The phrase “men are stronger than women” applies approximately equally well to the sets ciswomen+transwomen vs men+transmen and cismen alone vs ciswomen alone. Even though I am weaker then Ronda Rousey for instance

            A phrase like “Students have lots of student loan debt” obviously also exludes people like me who got through college off of their parents fortune.

            The phrase “legs are half of the human body” doesn’t apply to the situation where you grapple the guy with no legs.

            Basically if a generalization is a broad one it is typically assumed that there will be exceptions to the rule, in those circumstances the 1% or so aren’t ignored they are merely treated as exceptions to the rule.

          • John Schilling says:

            Which other 1%s of the population are acceptable to pretend don’t exist when defining sets?

            Even people who acknowledge home-schooling, etc, as reasonable alternatives will often use “high school graduate” as a synonym for “person who has met the minimum educational requirements of modern society”.

            Political language, as you note, routinely excludes anyone who doesn’t fit on a one-dimensional “political spectrum” from Democratic to Republican.

            Discussions of employment often use language that assumes every employee has an employer and vice versa, excluding the self-employed. And then often goes on describe a wholly consensual employee/employer relationship that excludes the military.

            And the amount of discussion of human social interaction that assumes introverts don’t exist is staggering.

            Just off the top of my head. There is clearly no general rule that if a group exceeds 1% of the population, it is forbidden or even rude to use language that does not explicitly account for their existence in every context.

          • DinoNerd says:

            @DavidFriedman

            I would be bothered by statements like “men are taller than women,” in many contexts, because it reads to me as an outright falsehood. If you were to draw a pair of histograms of adult human heights, you’d see two overlapping bell curves, with a lot of overlap.

            If you want to say that the average man is taller than the average woman, then say that.

            When the overlap is trivial, or irrelevant to context, go ahead and simplify.

            But why would someone be drawing attention to the difference, if they weren’t about to propose some policy based on gender (not height!) in order to accomodate the height difference?

          • But why would someone be drawing attention to the difference, if they weren’t about to propose some policy based on gender (not height!) in order to accomodate the height difference?

            Most conversations are not about proposing policies.

            “Jane has outgrown these pants–do you think your Jim would like them? He’s a little younger than she is.”

            “We can try, but men are taller than women so they’ll probably be too short on him too.”

            (This assumes that he doesn’t know Jane well enough to actually know how her height comparers to Jim’s).

          • marshwiggle says:

            It’s almost more days than not that I notice someone writing in a way that excludes me and a group or trait of mine, and enough of those are 1% or more of the population.

            On the one hand, sometimes I mind it a fair amount. Assumptions that you do not exist really can cause problems when people make policy (government or not) based on those assumptions.

            On the other hand, it happens so often that complaining about it seems fruitless. Most of the time it is just the ordinary conventions of language, not to be taken overly literally.

          • Nornagest says:

            I would be bothered by statements like “men are taller than women,” in many contexts, because it reads to me as an outright falsehood. If you were to draw a pair of histograms of adult human heights, you’d see two overlapping bell curves, with a lot of overlap.

            If you’ve got two groups with overlapping distributions of some trait, then, especially if they’ve got cleanly separated peaks, it’s perfectly normal to describe the group that’s higher on average as having more of that trait than the one that’s lower on average. It’s profoundly Evil Robot to pretend like omitting “on average” is some kind of cardinal sin.

          • albatross11 says:

            This just comes down to listening charitably–that is, assuming the speaker has a sensible point he’s trying to make, and working from there to interpret it. “All men are taller than all women” is obviously wrong, so you probably want to look for a more sensible meaning.

          • jermo sapiens says:

            The actual phrase was “men are taller than women”. Which is trivially true. When you compare groups, it is implied that the comparison is made between statistical distributions, and it is bad faith to require the comparison to be between each member of one group with each member of the other group.

            -Tigers are more dangerous than raccoons.
            -Ackshually, this tiger cub is not dangerous at all while this rabid raccoon is quite dangerous.

          • Dacyn says:

            Discussions of employment often use language that assumes every employee has an employer and vice versa, excluding the self-employed.

            This is actually true by definition. An employee is someone is who is employed, and the person who employs them is called an employer. Similarly vice-versa. (And for the record, the self-employed have both an employee and an employer, namely themselves.)

      • LadyJane says:

        As a trans woman myself, I really don’t take any issue with broad generalizations that incidentally happen to exclude trans people. “Men have penises” is a statement that’s going to be true about 99% of the time, and I don’t object to people saying it, as long as it’s not specifically being used to argue that trans men aren’t real men. I think it’s a little silly to argue that the scene in Kindergarten Cop where the kid says “boys have a penis, girls have a vagina” is transphobic, and I think it’s extremely counter-productive when trans activists start shaming feminists for wearing pussy hats at a march for reproductive rights.

        That said, I can sympathize with the trans activists who do take issue with statements like that, since actual transphobes often tend to fall back on those statements (e.g. “didn’t you learn in 5th grade Biology class that men have dicks and women have pussies? clearly the left doesn’t understand basic science, checkmate liberals”).

        • jermo sapiens says:

          “Men have penises” is a statement that’s going to be true about 99% of the time, and I don’t object to people saying it, as long as it’s not specifically being used to argue that trans men aren’t real men.

          What do you mean by “real men”?

          I think whether a trans person should be treated as their biological sex or as the gender they identify with, depends alot on the context. If I had a trans colleague, I would treat them with respect and go along with their gender identity. But if I was to set up a (straight guy) friend with a girl on a blind date, I wouldnt consider the trans woman as a potential blind date. And if I was a doctor examining the trans person, I would take note of their status as a trans person. And if the trans person is an athlete, I would also consider their status as trans to be very relevant.

          I feel that the fight over the term “real [men/women]” is a bit silly for that reason, and serves mostly as a virtue signaling tool in annoying tweets where a woke position is repeated endlessly like “Trans women are real women. Trans women are real women. Trans women are real women…”

          So sure, “trans men are real men”, but then we need another category for biological men. To me it’s just semantics and virtue signaling, but maybe I’m missing something.

        • LadyJane says:

          So sure, “trans men are real men”, but then we need another category for biological men.

          Well, we have a term for those people: cisgendered men.

          But in the vast majority of social contexts, that distinction simply isn’t going to matter. When it comes to a co-worker, or a cashier, or a random person you bump into on a street, their genitalia and chromosomal makeup really isn’t relevant at all.

          Sexual relations, medicine, sports, and maybe clothing-free spaces (changing rooms, saunas, nude beaches) are the only areas of life where it is relevant. (I say maybe on the last one because some trans people have had genital reassignment surgery, plus it’s usually simple enough to cover oneself with a towel or similar.) I suppose it might be relevant for statistical analysis too, but the amount of trans people in the world is so small that I doubt they’re going to throw off studies about colorblindness or whatnot.

          • jermo sapiens says:

            Well, we have a term for those people: cisgendered men.

            Yeah, exactly. So this recent coinage’s main purpose seems to be to replace terms like “real man” or “normal”. I understand the incentive to try and do this, because these terms carry an important emotional load, but I wouldnt expect it to succeed in the long run, and I would specifically strongly oppose attacking anybody as a bigot for not using it.

            Sexual relations, medicine, sports, and maybe clothing-free spaces

            Coincidentally, these are also the main areas of life where gender (as opposed to genitalia and chromosomal makeup) is relevant.

            I suppose it might be relevant for statistical analysis too, but the amount of trans people in the world is so small that I doubt they’re going to throw off studies about colorblindness or whatnot.

            That’s also why the term cis will never catch on with normies.

      • dokh says:

        There are a number of different biological features that people have at various times singled out as defining biological maleness, only one of which is relevant to colorblindness.

        When what you mean is “people with one and only one X chromosome,” you can say that, or you can generalize about men (the vast majority of whom have one and only one X chromosome), but choosing language that to you connotes “I’m talking about biology” without clearly indicating the actually relevant aspect of biology, is not particularly more trans-inclusive than just saying “men” in this context. (It’s also no less so, but, well, evil robot.)

        Men as a whole have a higher incidence of colorblindness than women as a whole but the stats get weird when you notice that sex isn’t strictly binary and start breaking your groups up more finely. This doesn’t make the simple statement untrue in general, or the more complicated statement that captures reality more precisely irrelevant. You can talk about the details or not, but if your way of alluding to the fact that they exist while not discussing them is to say “males,” it doesn’t make you sound woke, it makes you sound awkward.

    • TheMadMapmaker says:

      Technically it may be a more accurate category label, but to my ears it sounds weird. I don’t think many people would be confused by “men have a higher incidence of colorblindness”, even if the same actually applies to boys. This example being a bit medical is a case where technical terminology like “males” might still work.

      So in general, I’d still recommend using “boys”, “men” or “guys” instead of “males”.

      • Deiseach says:

        So in general, I’d still recommend using “boys”, “men” or “guys” instead of “males”.

        Lads is, or is becoming, a gender-neutral term 🙂

        In my all-girl school decades ago, we used “lads” as the general form of address to everyone e.g. “Lads, I just saw the Reverend Mother is coming up the stairs, everyone behave!”

        • edmundgennings says:

          Nooooooo
          This horrifies me.
          I will emotionally retreat behind the idea of the the generic masculine even though that does not apply to your case.

        • I attended third form in a school in Cambridge, England when I was about ten. The teachers were all or mostly female. We were told to refer to them as “sir.”

        • carvenvisage says:

          Pretty sure that’s as a joke, similar to bros or broskis.

      • DinoNerd says:

        Guys is gender-neutral in significant parts of the english speaking world. I try to avoid using the term because of the ambiguity, but I understand it as more or less equivalent to “folks” (which I often use instead).

        • albatross11 says:

          I stopped using “guys” generically at some point when I was working on a work project where the other three main contributors were women. At that point, it just seemed odd to start my emails with “Guys,”.

        • Deiseach says:

          I used to not mind “folks” too much, though it did seem a little quaint or affected (depending on who was using it) but the weird new spelling “folx” as some kind of mimicry of (pseudo?) AAVE for wokeness purposes has completely turned me off it.

          “People” or “Hey, everyone” is probably the least objectionable way to go about it if you don’t want to be formal but are concerned you might be taken up the wrong way for using a vernacular term.

        • Paul Zrimsek says:

          I’ve always thought of “folks” as one of those words best left alone unless you’re to the manner born. The rest of us tend to sound like we’re auditioning for one of those Fried Green Magnolias of the Yadda-Yadda Sisterhood movies.

          • HeelBearCub says:

            You really are deep in the land of the blue tribe (whatever that really means).

            Lots of us grew up using “folks” and feel perfectly comfortable with it. Also speaking with either a rural accent, and colloquialisms is, you know, normal for lots of folks.

          • Paul Zrimsek says:

            Well, of course. Those are the people to the manner born to whom I was referring.

      • brad says:

        Yes, this objection rather misses the entire point of the essay.

    • yildo says:

      “Males” is a sex — which dangly bits you have — and “men” is a gender — how you feel and how you perform in society. If you use “males” to mean “men and boys”, you are loudly signalling to me that you are either ignorant of the distinction or are choosing to disagree with it.

      • atreic says:

        I don’t think ‘male’ is used that way. Otherwise ‘female assigned at birth’ and ‘cis male’ would be meaningless terminology, and both of them exist and are useful.

        I think ‘he is a trans man, he was female assigned at birth, but is actually male’ is a very typical sentence, and isn’t about genital configuration or chromosomes.

        • deciusbrutus says:

          You’re wrong. ‘he is a trans man, he was female assigned at birth, but is actually male’ is a phrase about gender, a phrase about genital configuration, and a phrase about chromosomes.

          That phrase is most likely about someone who exhibited fetal androgen insensitivity.

        • hilitai says:

          “I think ‘he is a trans man, he was female assigned at birth, but is actually male’ is a very typical sentence”

          I think its ‘typicality’ depends very much on the sort of circles you run in.

      • If you use “males” to mean “men and boys”, you are loudly signalling to me that you are either ignorant of the distinction or are choosing to disagree with it.

        That comes across to me as “I and those who agree with me get to define what words mean, and people who don’t use them that way are either ignorant or trying to pick a fight.”

        Consider, as an analogous case:

        If you use “liberal” to mean “someone moderately left of center” you are loudly signalling to me that you are either ignorant of the actual meaning of the word or choosing to disagree with it.”

        “Liberal,” in its original political sense (and still largely in Europe) roughly corresponds to the modern “libertarian.” I know that, some other people know that, but someone who uses the term in the current U.S. sense isn’t loudly signalling either ignorance or disagreement.

        • deciusbrutus says:

          If nobody at all gets to decide what words mean, everyone is a Nazi.

          • Doctor Mist says:

            Whut?

            On second thought, never mind.

          • I get to decide what I mean by words and do my best to do it in ways others will understand. To get to decide what words mean for everyone I have to be a Nazi, or a Communist, or some other sort of totalitarian, and I need a world government to do it with.

            And I’m not sure even that would work.

        • Deiseach says:

          If you use “males” to mean “men and boys”, you are loudly signalling to me that you are either ignorant of the distinction or are choosing to disagree with it.

          Given that right now we have no useful measure of what exact percentage of the global population is transgender, and leaving out the whole question of intersex which is a different matter, (I’m seeing estimates online ranging from around 1% to around 7% which is a heck of a range for ‘how many transgender?’), then going with the lower bound of 1% and roughly dividing that between trans men and trans women (though that may be very much skewed if there are more trans women than trans men), we’re talking anywhere around 0.5% of the population.

          If you’re addressing “The Local Trans Male/Trans Men/Trans Masc Support Group” then sure, hair split to your heart’s content about what inclusive terms in which order you want to use.

          If you’re talking to the broad majority of the population, “men” or “men and boys” is perfectly fine. If you want to call me a -phobe, -ist or any other term of opprobrium for not being particularly worried about ‘only’ 99.5% of the people addressed, then go right ahead, I don’t mind and I’m also not going to change my usage.

          • roberone says:

            Surely 1% is an absurdly high estimate globally. You have vast areas of the world where it isn’t even a recognisable concept. From my own experience, (New Zealand) I would be shocked if prevalence was anything close to 1%.

          • Ozy Frantz says:

            The most reliable estimates for the US suggest that about 0.6% of the US population identifies as transgender, with some unknown percentage of the population gender dysphoric but nontransitioning.

            If that seems high to you, consider that a plurality of trans guys are going to read to the uninformed eye as an unusually short and stocky man with an impressive beard.

          • Furslid says:

            Hey, you can’t do that with percentages. .5% of males and .5% of females is .5% of the population. You can’t add that to get 1%. You also can’t split 1% that way.

            1% of the population means X% of males and Y% of females, where X+Y=2.

          • deciusbrutus says:

            How large a fraction of your target group is okay to miss before you improve your aim? 1%? 7%? 14%? 50%?

            Be specific, if your uncertainty about the actual prevalence is meaningful. What value are you certain that it is below?

            And than look at all of the countries with population less than that percentage of global population. Do you agree that it’s okay for ‘people’ to exclude residents of any one of those countries?

          • LesHapablap says:

            deciusbrutus,

            It depends what you mean by exclude.

            If I say to coworker, “people have got to stop speeding through that intersection,” I have now excluded 99.9999% of the world’s population, and people who don’t drive. This is fine.

            Maybe you can come up with a concrete example of bad exclusion that we can discuss.

          • And than look at all of the countries with population less than that percentage of global population. Do you agree that it’s okay for ‘people’ to exclude residents of any one of those countries?

            “Exclude” meaning “make statements that don’t take account of their existence.”

            “Modern developed countries, at least in Europe, are representative democracies.”

            Would you be outraged at that statement?

          • Furslid says:

            What percentage of people need to fall into a group before they have to be specifically referenced in tangentially related discussions. Are you upset by the statement “US citizens will vote for the president in November of 2020.”?

            It excludes those incarcerated and many felons. Approximately 1% of the population is imprisoned and 8% have a felony conviction. Does anyone who mentions voting have to specifically mention this exception every time?

            Apologies for any implied comparison. The only reason I chose the incarcerated is because it is a similar percentage of the population with an obvious exclusive statement.

          • deciusbrutus says:

            “Eligible voters” is a much better descriptor, since it points at exactly what you mean instead of a different slice of the cluster that contains what you mean.

            If you used “People speed through that intersection” and “Bob is a person” to conclude that “Bob speeds through that intersection”, you have committed serious errors.

            Likewise it is an error to use “XY-chromosomed people have a higher incidence of colorblindness” “Bob uses a masculine name” to conclude “Therefore Bob is a member of a group with higher incidence of colorblindness.”

            Yes, the conclusion is true in both cases, but that’s just because Bob is an asshole who doesn’t think speed limits apply to him and ‘people with masculine names’ overlaps very significantly with ‘XY-chromosomed people’, two facts about the world not included in the premises.

      • Nornagest says:

        ITT: people making complicated gender studies points and pretending they’re idiomatic English.

      • LesHapablap says:

        I don’t think it is possible to ‘loudly signal’ ignorance of something. If you’re ignorant of a thing, you can’t very well be sending signals about it.

        • Simon_Jester says:

          There’s ignorance and then there’s what we might colloquially call “Know-Nothingism.”

          Ignorance is simply the absence of relevant knowledge.

          Know-Nothingism is a stance. It’s a worldview in its own right, namely the worldview that certain kinds of knowledge are inherently degrading, damaging, or bad to have- either in general, or for a certain group.

          One of the hallmarks of a culture that suffers from know-nothingism is that people will turn conspicuous ignorance into a kind of status signal.

          So yes, it’s very possible to loudly signal ignorance. People who derive satisfaction from watching other people get upset, because it makes them feel vindicated/rational/superior, are particularly likely to do this. Because a deliberate display of ignorance, or willful refusal to listen to others and learn from experience, is a very effective way to wind people up.

          • crh says:

            More innocuously, you can also signal ignorance any time you’re aware that a body of knowledge exists but you don’t have it. “I don’t know the first thing about [cars/computers/math/etc.]!”

          • LesHapablap says:

            That is all true more generally but not at all to the example above:

            If you use “males” to mean “men and boys”, you are loudly signalling to me that you are either ignorant of the distinction or are choosing to disagree with it.

            The vast majority of those who say males instead of ‘men and boys’ will be actually ignorant, not pretend ignorant. And if you are pretending to be ignorant you are “choosing to disagree with it,” anyway.

        • Jaskologist says:

          Perhaps instead of “ignorant” they should say “evil,” since that’s what they mean.

          • deciusbrutus says:

            Ignorance objectively exists, evil does not.

          • RC-cola-and-a-moon-pie says:

            DeciusBrutus, if evil does not objectively exist then on what basis do you advocate the use of what you view as non-exclusionary linguistic conventions? Just your subjective personal preference?

          • Ignorance objectively exists

            I agree. If someone says “the moon is ten million miles from Earth” and means it, that implies that he is ignorant, because the distance to the moon is an objective fact, although one that varies a little.

            But “male means someone with XY chromosomes, man means someone who presents himself in the way we associate with males” is not an objective fact, it is a statement about how some people use language. Someone who doesn’t know that some people use language that way is ignorant of that fact, but all of us are ignorant of many facts, and I expect there are facts about how other people use English that you are ignorant of.

            He is not ignorant of the meaning of those words, however, because the meaning of words is not an objective fact but a convention. My guess is that a large majority of English speakers do not use the words in your preferred way, so if you define objective meaning by actual usage, you are the one who is ignorant.

            But it makes more sense to recognize that the words have a particular meaning in your ideolect but not in those of many others.

          • Doctor Mist says:

            the moon is ten million miles from Earth

            In my lived experience, a mile is 126 feet.

          • deciusbrutus says:

            The meaning of words is a social convention that has an objective existence.

          • The meaning of words is a social convention that has an objective existence.

            It’s fuzzier than that, since most words have a range of meanings.

            In the case of “men,” the most common meaning is “people with natural male genitalia and an XY chromosome,” which isn’t what you want. Most people, pushed on it, would concede that there are some special cases that don’t quite fit in or out of the category, such as XXY or eunuchs. Some would include transmen, some wouldn’t.

            Suppose we did a careful study of language usage and found that a substantial majority of English speakers did not unambiguously include transmen in the category of “men.” Would you then conclude that you were the ignorant one? If so, what is your evidence for the opposite conclusion, which you need to defend your claims?

          • LadyJane says:

            In the case of “men,” the most common meaning is “people with natural male genitalia and an XY chromosome,” which isn’t what you want. Most people, pushed on it, would concede that there are some special cases that don’t quite fit in or out of the category, such as XXY or eunuchs. Some would include transmen, some wouldn’t.

            I wouldn’t include chromosomal makeup in the “common” definition of the word at all. It’s certainly not the traditional definition, since for most of human history, no one had any idea that chromosomes were a thing. And it’s definitely not a functional definition either, since there’s no way to tell someone’s chromosomal makeup by looking at them. Most people aren’t even aware of their own chromosomal makeup. There are people with Androgen Insensitivity Syndrome who have XY chromosomes, but are anatomically identical to cis women in every way, including genitalia. Any definition of men that included them would be functionally useless. Chromosomal makeup is only relevant for a niche biological definition of the word “man,” not the commonplace definition.

            As far as genitalia, that’s more of a gray area. On the one hand, someone’s genitals are an obvious external trait that can be visibly observed, so it would make sense to include them in a functional definition of the word “man.” However, I don’t think it makes sense to exclusively define “man” in terms of genitalia, since a functional definition of the word would also take secondary sex characteristics into account. If anything, those would be even more relevant in most social contexts, since the overwhelming majority of people’s social interactions occur with their clothes on and their genitalia covered. And, as you mentioned, most people would consider a eunuch who lost his genitalia to still be a man. Furthermore, even if you ignored secondary sex characteristics and defined “man” solely in terms of having male genitalia, that wouldn’t exclude all trans men, since some have had phalloplasty.

            So in terms of functional definitions, we’re left with two possibilities: First, we could define “man” solely in terms of external appearance in a normal social context (which wouldn’t take genitalia into account, and absolutely wouldn’t take chromosomes into account), which would include trans men. This is the definition preferred by many trans activists. Second, we could define “man” in terms of both genitalia and secondary sex characteristics. Under this definition, pre-op trans men wouldn’t be men (since they have female genitalia), but wouldn’t be women either (since they have male secondary sex characteristics). They’d be considered something else entirely, akin to the “third genders” of various ancient civilizations. I could see an argument being made that this would be a more concise definition than the first, but I’m not sure how it could work in practice; it would require a complete paradigm shift for the sake of a very small minority of the population, most of whom would prefer to not be “othered” in that way. I also don’t know who’d be willing to support it, since the types of people opposed to the idea that “trans men are men” are usually even more strongly opposed to the idea that there are more than two sexes/genders.

          • First, we could define “man” solely in terms of external appearance in a normal social context (which wouldn’t take genitalia into account, and absolutely wouldn’t take chromosomes into account), which would include trans men.

            Does your definition of trans men require them to successfully present as male–i.e. well enough so that a stranger will assume they are male in the narrower sense?

            Of two transmen (loose sense) I know, one I would assume to be a cisman if I hadn’t known the same person as a woman. The other any stranger would assume to be a woman, despite her changing to a male first name and wanting people to refer to her as male. Obviously there are intermediate cases.

          • LadyJane says:

            @DavidFriedman: Most of the trans-masculine people I know have taken steps to medically transition (usually testosterone boosters and/or mastectomies), so they have distinctly male characteristics like heavy facial and body hair, lack of breasts, deep voices, and so forth. Some are virtually indistinguishable from cis men, while others look more androgynous and could be mistaken for especially masculine cis women, but none of them look like typical cis women. I’ve known a few trans men who were pre-transition, but still decided to publicly identify and present as male; even then, they always took steps to appear as masculine as possible and cover up their feminine traits (for instance, wearing baggy clothes or binders to hide their breasts).

            If I met someone who identified as male, but hadn’t medically transitioned and wasn’t making any real attempt to look male, then I’d just assume they were very early in their transition and weren’t “full time” yet. I’d still consider them male (or more precisely, “not-female”) in a neurological sense, since I believe that gender dysphoria is the result of a real, tangible, biological phenomenon. But for functional purposes, I wouldn’t expect them to be considered male in most social situations, and I would probably encourage them to wait until later in their transition to go full time or legally change their name.

    • caryatis says:

      Bad idea. Say “men and boys” (or “women and girls.”) You might think you’re talking about a biological trait, but it’s very contentious what those are, and you will still sound like an evil robot (or a person who is trying to sound more scientific than they have grounds for).

    • Nabil ad Dajjal says:

      On most places you’d be right. Feminists get upset if you call women females but there’s no equivalent group of people who get upset if you call men males. But the issue here, I suspect, is mostly about transgenderism and less about feminism.

      SSC commenters are ridiculously more familiar with / insistent on the terminology of the trans movement than regular people. I don’t remember the exact numbers from the survey but we either had more transwomen than women or at least it was a close call. So if you want to ask a question about men or women without getting nerd-sniped by pedants and/or activists you either need to use the unnatural prefix cis- or use male and female which allegedly refer only to biological sex.

      I don’t mind just saying men and women when I mean men and women, not the least of which because ruffling the feathers of the local thought police is fun when you have a pseudonym. But for the commenters here who buy into and/or are cowed by that ideology it makes sense for them to use evil robot language.

      • brad says:

        On most places you’d be right. Feminists get upset if you call women females but there’s no equivalent group of people who get upset if you call men males.

        It’s not a matter of getting upset, saying males make you sound weird. If you’re okay with that, you’re okay with that.

        • cuke says:

          Yeah, it feels like people are talking past each other here.

          To my ears, saying “male” or “female” sounds weird and signals a degree of social clumsiness in the person using it, quite apart from any political exclusion issues.

          The people I have heard use “male” and “female” in non-scientific contexts tend to be men who present as somewhat socially awkward, and that’s who I see Scott speaking to here.

          If you are seeking to come across as less socially awkward, say “men” and “women.” If you don’t care about seeming less socially awkward, more power to you, that’s awesome. We all care too much about what other people think anyway.

          • cuke says:

            I want to add based on reading comments below that I think the common usage of “males” and “females” in many black communities has a different origin and connotation and Scott’s advice doesn’t apply in that context.

      • deciusbrutus says:

        I don’t mind saying men and women when I mean men and women, but I prefer ‘people’ when I mean ‘people’ and ‘humans’ when I mean ‘humans’.

        Note that most things that are true of humans are true of people, and vice versa.

      • Deiseach says:

        use male and female which allegedly refer only to biological sex

        I am still honestly confused about which means what, because for every post by an activist about “male means A and should only be used to refer to C”, there’s another post by a different activist that “male means B and should only be used to refer to D”.

        I’m strongly tempted to declare that, as one of the tiny minority (how many again are there of us?) of Catholics on this site, I insist that my status be included and acknowledged by Scott starting off every post with the particular saint’s day that corresponds to the date, and with a short prayer appropriate to the day.

        After all, it would be “loudly signalling to me that you are either ignorant of the distinction or are choosing to disagree with it” that not every single SSC reader and commenter is non-religious, and as we have been lectured, “loudly signalling” is Very Bad and non-inclusiveness is Even Worse!

        As an aside, deciusbrutus, what people do you know who are not humans? First alien contact? 😉

        • Possibly cats?

          Reading the book Chimpanzee Politics, chimps sound more like dumb humans than smart animals.

        • Plumber says:

          @Deiseach

          “.., as one of the tiny minority (how many again are there of us?) of Catholics …”

          Among frequent commenters I count four “out” Catholics with quite a few infrequent ones (if you read the archives deep enough).

          There’s a couple of Orthodox Christians as well, and a couple of religious Jews, a couple of Mormons, no Muslims that I’m aware of, I think someone said they were Buddhist, but I can’t remember if they were a frequent commenter or not, most don’t say, many say “atheist”, really given Anglophone demographics the absence of “out” Protestants is a bit odd, regardless both folks who’ve mentioned that they religious in general and Catholic in particular are minorities.
          And to “out” myself, I wasn’t much raised in any religion though I remember some Sunday “potlucks” at a Unitarian church as a child, but no sermons – my father’s mother was a devout Catholic and she sent my Dad to Catholic school, but he rejected the faith, my mother’s mother strongly identified with the Lutheranism of her Germsn father, but I don’t remember her going to services, I just remember her sometimes holding a German Bible which I don’t even know if she could read, her mother (the only Great-Grandparent of mine that lived long enough for me to meet) was Austrian and Jewish (according to my Mom, but she didn’t tell me that until she married my stepfather who’s Jewish), I never knew my father’s father and I just don’t know his creed, my mother’s father was born in Kansas and the only religion he seemed to have was the U.S. Army Air Corps and Louis L’Amour novels (in my memory he resembles a cross between Gary Cooper, Henry Fonda, and John Wayne).
          Most of what I learned about Christianity and religion came not from my family but from my neighbors who mostly attended services at the McGee Avenue Baptist Church, and for most of my childhood to me Christendom was black, I really didn’t often encounter many devout white Christians until I was in my 30’s, most of whom were fellow union members, which reminds me of the front page of my local union newsletter, which has a picture of Kamala Harris (’cause she’s being endorsed), and a big announcement that my union is now funding a Catholic school, so clearly we’re behind the times and need to be straightened out regarding which side of the “culture war” we belong!

          • theredsheep says:

            I don’t know if there’s another Orthodox Christian here. At least, I can’t recall encountering one.

          • Dragor says:

            I am Buddhist, but I don’t know if I am the one you recalled. I don’t comment much.

            You mentioned above your grandfather was German. Out of curiosity, did you wind up with citizenship through descen

          • Plumber says:

            @Dragor

            “….You mentioned above your grandfather was German. Out of curiosity, did you wind up with citizenship through descent”?

            Nope, it was one great-grandfather who was German, who married an Austrian born women in the United States, my other Great-Grandparents (IIRC) were one born in Massachusetts, one born in Kansas, two born in Ireland, and two that I just don’t know where they were born.
            For Grandparents it’s California, Kansas, Massachusetts, and Ireland or New Jersey (both of her parents are always Irish in the telling, but whether she was born just before or just after they immigrated I’m unsure of), so no German grandfathers for me, just a California born grandmother who’s parents both spoke German.

        • HeelBearCub says:

          @Deiseach:

          I insist that my status

          You are consistently insistence that your embrace of the Catholic faith be known and acknowledged.

          If we used the term Christian intending it to mean only one who was a Unitarian, Evangelical, Protestant believer in consubstantiation, I believe your hackles would get even higher than normal.

          Is their a patron saint of self-awareness?

        • deciusbrutus says:

          Ravens, certainly. Elephants possibly. When I’m being maximally honest, cats are family but most likely not people.

          Octopi, plausibly.

          Specifically, any organism that exhibits behavior that is best explained by having the ability to model the knowledge and decision-making process of another animal.

          (Humans appear to acquire that ability at around age 4-6. There are some implication in that fact that I acknowledge and address firmly with ‘Personhood is not a requirement to be a member of Family).

    • DinoNerd says:

      Interstingly, I rarely use the ambiguous word “man”/”men”. It means both “human(s)/people” and “adult human male(s)” and it’s too often impossible to be sure, from context, which is intended.

      Worse, that ambiguity has been used to specifically exclude women in situations where the original meaning was almost certainly “humans” – the classic example being translated works such as the Christian Bible, where the prior language had non-confusable separate words for the two concepts, making this easy to check.

      • Nornagest says:

        Despite the objections of a few social conservatives and prescriptivist linguists, “man”/”men” as a term for “humans”/”people” is pretty much dead, and you can safely assume that anyone using the word in modern English (especially uncapitalized, and unless they’re a poet, an antiquarian, or making some kind of conservative political point) is using it to mean “adult male human/s”. Pre-1970 English is another story, but context should clue you in most of the time.

        “Mankind” survives but there’s no ambiguity there.

        • Mary says:

          On the contrary, it is so much alive that people will literally assume it’s the only meaning you could possibly be giving.

          As witness people that “werewolf” means “man wolf” — and “correct” it to “person wolf” because they simultaneously assume that it can only mean “human” and that it can’t possibly mean that.

          • Nornagest says:

            I’m not talking about what people understand it to mean (which encompasses some archaic senses, or ones otherwise not in current use), I’m talking about what people use it to mean.

            Most people would understand what you meant if you said “prithee”. It’s nonetheless a dead word in modern English.

      • Deiseach says:

        I’m fine with “man” as meaning both “human” in general and “male human” in particular, and with the usage of “Man” as including “both male and female humans”. The switch to using “persons” or “humanity/Humanity” had me shrugging but I don’t particularly object either since that’s not a hill I’m prepared to die on.

        My contrarianism is triggered, however, by those finger-wagging that “Man was used to exclude women and keep them down and oppressed” and so I, as a female human, will loudly recite the old version of the Creed about “for us men and for our salvation” 🙂

        the classic example being translated works such as the Christian Bible, where the prior language had non-confusable separate words for the two concepts

        Now, I’ll quibble on this, because are you stating that translators of early versions from Latin/Greek/Hebrew into English etc. deliberately did this, or were they perhaps trying to use terms that were common in the language they were translating into?

        Take that part of the Creed I mentioned: in Latin, it’s “Qui propter nos hómines et propter nostram salútem”. You can translate “hómines” into Irish as “an cine daonna” (which literally translates as “the human kind/race”) because in Irish “duine” is the term for “person, human being”. Just as your Latin Bible would have separate terms for “humankind” and “men/women”, so Irish has separate terms for “person” and “man/woman”.

        So how about English? Did early English have similar terms? Well, if I go by this article, not really: mann meant both “adult human male” and generic “person, human”.

        So someone translating into English would be stuck: either use a fancy neo-Latinate term and render the word “human” so as to retain the distinction from “not meaning vir/man in that sense” which might not have been generally understood, or use the language colloquially and just say “men (and this includes women)”. There you go, you don’t need to be an Evil Patriarchal Oppressor, just someone struggling with the perennial problem of translating from one tongue to another and “do I go with the literal or with the spirit of the meaning?”

        • DinoNerd says:

          Now, I’ll quibble on this, because are you stating that translators of early versions from Latin/Greek/Hebrew into English etc. deliberately did this, or were they perhaps trying to use terms that were common in the language they were translating into?

          I feel reasonably certain that the translators of the KJV from Greek translated “anthropos” as “man” intending it to be understood as referring to all people. But I’ve also experienced some large number of Bible-inerrancy-believing English speakers in the 20th century insisting that God’s intent was to refer to males exclusively.

          They never cited the Greek – I doubt any of them could read Greek. But they did use their Holy Book, as translated, as a verbal bludgeon to beat on girls and women who wanted things they preferred to restrict to men.

          Most of them were not theologically sophisticated enough to cite the “inspired translation” doctrine – they just knew that in any question of privilege, “men” meant males.

          I’ve seen the same thing done, repeatedly, with regard to the text of statutes (laws), but in those cases we have less evidence of the original intent, which is why I prefer the Biblical example.

    • Alfred MacDonald says:

      The gist of this article is “don’t seem out of touch”; you can seem out of touch by being a robot, or a crazy person, or a shut-in, or in many other ways. Quite a few of these comments seem out of touch, even if not in the robot way.

      Here is Kenan Thompson using “females” non-robotically in the 1997 Nickelodeon movie Good Burger. If for some reason you can’t view the video, he says “maybe invite some fine females over to share an egg roll or two” and it’s not weird. No one thought it was “weird” then. No significant amount of people do now, either — this is one of those extremely petty linguistic power-grabs that will be historically archaic once the ideology objecting to it is no longer a fad.

      “Males” to mean “men and boys” is fine. It’s extremely common. Nearly everyone will know that you mean “men and boys”, and practically speaking no one will care. If you are the kind of person to use “cisgender” in casual conversation without wondering if the person you’re talking to even know what this means, you interact with an extremely skewed audience and your sense of what is weird is abnormal. For everyone who doesn’t take ideological objection — so, essentially everyone — it’s fine. I’m going to continue using “male” in this way, and it will continue to be fine forever.

      • Nornagest says:

        You can use it in a way that won’t seem out of touch, but how you do that is complicated and subtle and hard to sum up. If you’ve got your finger on the pulse of pop culture, you don’t need these rules anyway. If you don’t, you’re probably better off avoiding “male”/”female” outside scientific or clinical contexts.

      • deciusbrutus says:

        Sure, and addressing “Ladies and Gentlemen” is a way to exclude officers of the Royal Navy.

        When you use ‘male’ to refer to men and boys, and ‘female’ to refer to women and girls, in order to refer to all people you have to include the other categories. Do you use ‘intersex’ to refer to nonbinary folk, since [male, female, intersex] is the proper set of all sexes, or do you use ‘nonbinary’ to refer to nonbinary folk, since that is the term that communicates the meaning you intend instead of a different meaning?

        Based on your initial assertion, that you don’t care about communicating the meaning you intend so long as you don’t have to acknowledge that sex and gender are both things, you’d just say “intersex and nonbinary people don’t exist” and call it a day.

        • Alfred MacDonald says:

          Do you use ‘intersex’ to refer to nonbinary folk, since [male, female, intersex] is the proper set of all sexes, or do you use ‘nonbinary’ to refer to nonbinary folk, since that is the term that communicates the meaning you intend instead of a different meaning?

          Okay. Is your goal to avoid sounding weird, or to prioritize whatever this is?

        • Sebastian_H says:

          At many times in this discussion you seem to be weirdly using the word ‘exclude’ when you mean ‘include’ which is far more confusing than the rest of your discussion.

          If I correctly say that “men are more likely to be color blind than women”, that statement correctly *includes* the set of trans men and cis men because the statistical chance of color blindness among cis men is so much higher than women that whether or not trans men have higher incidences of color blindness is irrelevant to the statement. You label that as ‘excluding’ trans men, which is very confusing considering that in any normal view of ‘include/exclude’ my statement includes trans men. There are lots of men who aren’t color blind (a large majority even). So noting that some subgroups can see color seems orthogonal to the point.

          Extra weirdly you somehow want the idea of “man-ness” to include trans men and then complain about it when it does. It seems like what you want is to turn every high level discussion where “male-ness/man-ness” might be a component into a discussion where the distinctions between cis males and trans males are highlighted. That seems counterproductive to the aims of many trans-men (who just wanted to be treated as men) and to the aim of sometimes discussing things that aren’t trans issues without getting bogged down in trans issues (like color blindness).

      • thisheavenlyconjugation says:

        As someone mentioned above, “female” (but not, I believe, “male”) has a different connotation in AAVE/AAVE-derived language. There’s also the thing where one can use language in an inappropriate register ironically (e.g. greeting your friends with “Good day, fellow humans”). Furthermore, you are right that it’s possible to use “(fe)males” in a non-awkward way without either of those situations occurring, but there is a pattern of people (in this subculture) using it awkwardly (e.g. the example in the post, although it’s more relevant in dubious evopsych speculations about mating in which you want all the charity you can get). This is a separate issue from the small number of people with ideological objections.

    • BBA says:

      The current line I’m hearing is that “male” and “female” are inherently offensive terms, as is “biological sex.” You should only talk about “men” and “women”, and if you’re referring to biology mention the actual bodily function involved, e.g. “people who menstruate.” For now, there’s also “sex assigned at birth”, but the term should be phased out as we stop assigning sex at birth.

      I’m not totally sure that this is the right way to talk about things or that it will actually become common usage outside queer theory circles, but that’s probably because I’m a lousy privileged cis het white man who has no right to comment on such matters.

      • Ozy Frantz says:

        Piece of style guide advice: when you feel tempted to write a sentence such as “I’m a lousy privileged cis het white man who has no right to comment on such matters”, delete it. Either you are actually a victim in this particular situation, in which case someone will be along quickly enough to victimize you without you having to do it for them, or you are not a victim right now at all, in which case claiming to be a victim anyway tends to annoy people.

      • deciusbrutus says:

        What you’re hearing is “Sex is irrelevant in this case” over a large number of cases. Which is true, the cases in which sex is relevant are generally medical contexts that are edge cases for normie social interactions.

        • Sebastian_H says:

          Correct, which is why the term “cis” is going to have problems getting into common usage, because the number of times you legitimately have to label someone as “absolutely definitely not trans” is tiny if you aren’t a doctor.

          • deciusbrutus says:

            The only people I’ve seen object to the prefix cis- have been people who rejected the validity of trans people.

            It’s not commonly needed.

          • Bugmaster says:

            I think this is a bit of the Motte-and-Bailey situation. On the one hand, the prefix cis- is perfectly fine, and quite useful when discussing issues of gender. On the other hand, the kinds of people who tend to tout the prefix are also the kinds of people who want to make every conversation about gender, regardless of the topic; and randomly injecting the prefix into every context is an effective way to accomplish this goal.

    • Steve Sailer says:

      “Males” is OK to mean “men and boys”, etc.

      Indeed. Scott’s own supposedly “Right” example shows the utility of “males:”

      Wrong: “Why do so many males like sports?”
      Right: “Why do so many men like sports?”

      The answer to the question “Why do so many men like sports” is that it’s not surprising that they like sports as men because they almost always also liked sports as boys. You should only ask “Why do so many men like sports” if your argument is that full-grown men should have outgrown their boyish interest in sports.

      • deciusbrutus says:

        Or if you intend to ask why some adult men grew out of their girlish distaste for sports.

      • Dacyn says:

        “Men” can mean either “humans”, “male humans”, or “adult male humans”, depending on context. (This can be split further depending on whether “male” means “biologically male” or “identifies as male”.) It shouldn’t be treated as a technical term that only has one meaning.

      • Mary says:

        That sounds like — a good way to sound like an evil robot, actually.

        If you ask why men are taller than women, you are asking about something that certainly stems from boyhood vs. girlhood, but if you complain that it should be asked only if you are asking about why the women did not continue to grow in adulthood, or the men did not shrink, people are going to look at you funny.

      • Nornagest says:

        The answer to the question “Why do so many men like sports” is that it’s not surprising that they like sports as men because they almost always also liked sports as boys.

        If the question came up in a conversation with the typical man, and you tried that answer on him, the typical man would call you a smartass. And he’s not wrong.

    • Thegnskald says:

      “Males” is okay. Among nerdy women, however, “Females” reads as “Ferengi” and so comes off as offensive.

      It’s not about sounding too robot-y, or anything like that; it wasn’t that long ago that nobody cared. Star Trek associated the word with sexism, edgy people started using the word to annoy people, people who had never seen Star Trek began to notice that it was edgy assholes using the word to annoy women, and its status as a sexist word was cemented.

    • Sinclair says:

      I’ve never heard people in everyday speech use “males” that way. My culture uses “guys” to mean “men and boys”, but there isn’t an equivalent term for “women and girls” and also sometimes “guys” is used in a gender neutral way.

  2. Sniffnoy says:

    Use utility only when talking about utilitarian philosophy.

    Or decision theory(/game theory)! Do I have to give the whole “there is decision-theoretic utility and there is utilitarian utility, and despite the conflation-encouraging terminology you should not conflate them because they are seriously different” spiel again?

    Let’s not encourage people to conflate them further!

    • doremitard says:

      Could you briefly explain the differences between decision-theoretic and utilitarian utility? Or point to an explanation?

      • melolontha says:

        Utilitarian utility has to be interpersonally comparable, and it may or may not be preference-based — for some utilitarians it basically means ‘happiness’, or one of the various meanings of ‘well-being’. It’s the thing that the utilitarian thinks we morally ought to maximise.

        Decision-theoretic utility is the thing you act as if you’re maximising the expectation of, if your decisions conform to a bunch of standard rationality axioms.

        • jgaln says:

          Just to expand on this (and please correct me if I’m wrong), but utilitarians would understand utility as something like a “global variable” representing total goodness in the world that everyone is, by their actions, morally responsible for maximizing; decision-theoretic utility represents a sort of personal ranked preference for the possible future states of the world. Which, as melolontha says, you probably are aiming to maximize the expectation of.

          For someone who is not a utilitarian, there’s no reason for these concepts to overlap at all; decision-theoretic utility will still be relevant to them even if they don’t care about maximizing utility for other people.

          For a committed utilitarian, these concepts might overlap a lot, because their personal utility function will probably be very close to what they think the global utility function is. They will probably try to maximize the expectation of global utility. But they also might not; there’s room for them to have their personal utility be a nonlinear function of the global utility, so that they might end up weighing risks and gambles differently than to maximizing the expectation of global utility.

      • Sniffnoy says:

        The other replies have already given good answers, but FWIW here’s a link to one place I’ve answered this question before.

  3. Sniffnoy says:

    Some other notes:

    On #7, there’s also the concept of like… the thing that Razib Khan talks about when he talks about “nerds” here, or that used to be referred to as “Smart Sincere Syndrome” on Overcoming Bias. It’s clearly a distinct thing but it isn’t, y’know, autism. Using the word “nerd” does risk confusion with less specific meanings, but I do think “nerd” is a good word for it if everyone understands that’s what’s meant.

    On #8, might I also suggest the words “authoritarian” and “traditionalist” as useful? I don’t like “far-right” or “reactionary” because, well, I don’t think the standard one-dimensional political spectrum actually gets at anything real (other than current coalition). “Far-right” in the US seems to refer to this Frankenstein’s monster constructed largely out of traditionalism-authoritarianism, but with with ill-fitting bits of liberalism akwardly bolted on. I don’t think it’s actually much coherent. But I’ve said this any number of times before. Anyway, point is, more descriptive words than “right-wing” are useful.

    • ksvanhorn says:

      “Authoritarian” is descriptive of many leftists. If you think it automatically means “right wing”… I don’t know what to say. Ever heard of Mao? Stalin? Pol Pot?

      • Lambert says:

        Authoritarianism may not be inherently right-wing, but traditionalism-authoritarianism is.

        • Depends what you mean by “traditionalism.” The New Deal was almost a century ago, so what it created is by now largely tradition. But supporting it is more nearly left than right in modern terms.

          To take the obvious local example, Plumber is much more of a traditionalist than I am, but he is also to my left. If he took seriously his spiels on what the world should be like and tried to implement them, he would be a traditionalist authoritarian.

          • Plumber says:

            @DavidFriedman,
            That’s a fair cop, and frankly I’d be leery of handing someone with my temperment power, but I’m already pledged to serve the great and fearsome Dread Empress @Deiseach and her noble counselor and “Hand” @HeelBearCub, though I suspect her imperial majesty may have a need for a right hand counselor for balance, a role I imagine you’d well fit.

          • Deiseach says:

            Plumber, I am very much tickled by the mental image of the Imperial Tricycle of State, with myself as (presumably) the front wheel, steered on the correct centrist and central path by the left and right rear wheels of HeelBearCub and Professor Friedman respectively 😀

            Any veering into ditches solely my own fault!

          • cuke says:

            Weird, I think I’d vote for that tricycle.

        • Viliam says:

          In post-communist countries, there are many traditionalists-communists.

      • Sniffnoy says:

        Yes, sorry, good point. Coming up with good terms for things is hard. This is why I like to say “traditionalism-authoritarianism”, to clarify what I mean, but that’s a bit of a mouthful. (“Trad-auth”?) Authoritarian leftists I prefer to just call “leftists”, unless I want to particularly emphasize what they have in common with the, uh, trad-auths.

  4. Shion Arita says:

    Guess it’s just better to sound more like a normie. Oh well.

    • Matthias says:

      In the specific context of talking to normal people, yes.

    • TheMadMapmaker says:

      Maybe Scott should also add something about avoiding “normie”, though that’s less of an “evil robot” phrase than a “4chan edgelord” phrase…

      • Wolpertinger says:

        Sounding like an evil robot is one of the tools in the edgelord’s box.

      • Sinclair says:

        “Normie” is fine here. People who don’t know what it means can probably tell that it’s slang and correctly parse it as an insult based on the context.

        You can’t replace “normie” with “normal person” without making Shion’s sentence seem self-deprecating. What word or phrase would you use to refer to normal people that has the same tinge of contempt that “normie” has? Pleb? Square? I can’t think of anything that isn’t also slang.

        I will agree that if Shion meant to say “I wish I sounded normal” then what they did say would be bad style, but I think what they said is what they meant to say.

        • enye-word says:

          I’ve always used “Normal” as a noun when the need arises. “A normal”, “normals”, etc. You can capitalize it in writing to make it extra clear if need be.

          • albatross11 says:

            Depending on context, “normies” can be swapped with “mundanes” or “bourgeoisie.”

        • Galle says:

          You can’t replace “normie” with “normal person” without making Shion’s sentence seem self-deprecating. What word or phrase would you use to refer to normal people that has the same tinge of contempt that “normie” has? Pleb? Square? I can’t think of anything that isn’t also slang.

          I think that’s sort of the point, isn’t it? Saying “normie” makes you sound like a 4chan edgelord because it’s an expression of contempt for people who aren’t part of your particular internet subculture.

  5. eric23 says:

    1. A lot of this post is telling us not to use rationalist-community jargon, which the rationalist community has developed (presumably) in an attempt to raise the logical clarity of discussion. Now you are arguing for the reverse!

    2. Speaking of which, you don’t have an alternative for the word “priors”? That seems like the worst style offender of all.

    3. Some of your replacements are sloppy. For example “Why do so many males like sports?” includes teenage males, which “men” does not. Or “Will getting more exercise make me better off?” is misleading because “better off” often refers to money specifically. That plays into #1.

    4. I think the “real” problem with the word “autistic” is that it can be a slur. Not that it’s weird jargon.

    5. Perhaps you are suggesting that we make the changes when talking to the general population, but keep talking as we do amongst ourselves. That sounds like a good ideal. In fact I think most of us strive for that ideal already. But in practice it’s very difficult to switch between two modes like that. So mistakes are made.

    • Said Achmiz says:

      1. A lot of this post is telling us not to use rationalist-community jargon, which the rationalist community has developed (presumably) in an attempt to raise the logical clarity of discussion. Now you are arguing for the reverse!

      The problem is, this jargon does not “raise the logical clarity of discussion”—for the simple reason that, overwhelmingly, “rationalists” do not use it correctly.

      Are you going to tell me with a straight face that most uses of the words “utility”, “utility function”, etc., in rationalist circles are correct usage? They’re not—not even close. Most “rationalists” do not even understand what these words mean, and never have. I know; like Sniffnoy, I’ve been trying, utterly in vain, to correct this pervasive misunderstanding (on Less Wrong and elsewhere) whenever I’ve encountered it. I don’t think I’ve made a dent.

      Same deal with most of the rest of it. The entire problem is that yes, these terms were (ostensibly) created or chosen in order to “raise logical clarity”—but then they were abused, blatantly and ubiquitously, by being used much more broadly than intended… because they had transformed (and were they ever anything else?) into tribal shibboleths, markers of one’s own intelligence or “rationality” or being “one of the in crowd” or whatever.

      So when you say—

      5. Perhaps you are suggesting that we make the changes when talking to the general population, but keep talking as we do amongst ourselves.

      —you have it exactly backwards. Yes, it’s important, as Scott says, to not weird out the “normal people” so unnecessarily much. But it is, if anything, even more important to stop corrupting our own thinking, by abuse of the very jargon we invented to improve it. When you talk about “increasing utility” instead of “helping”, or being “rational” instead of being “reasonable”, you are not clarifying your thinking, you are distorting it.

      So if this advice seems like a reversal of course, well… for the necessity of that reversal, we “rationalists” have no one to blame but ourselves.

    • johan_larson says:

      Speaking of which, you don’t have an alternative for the word “priors”? That seems like the worst style offender of all.

      Interesting question. If I were speaking to someone outside the Rationalist community, and were tempted to say something like, “Based on the latest IPCC report, you should update your priors in the direction of less concern about sea level rise and more concern about damage to agriculture, particularly in tropical regions, ” what would be a better alternative?

      Perhaps, “Based on the latest IPCC report, you should be less concerned about sea level rise and more concerned about damage to agriculture, particularly in tropical regions.” Here the word just drops away.

      And if you must have a substitute for the word, as in the question, “What are your priors on school shootings?”, something like, “What is your current position on shootings?”, seems clear and straightforward.

      • Joseph Greenwood says:

        Stripped of a precise technical definition within the field of [probability theory? statistics? Probability theorists tend to care more about measures and less about division by weird terms, and I think statisticians are trained as frequentists generally], “my priors” seems to refer to “The beliefs I currently hold, especially in contrast to what I might believe if I learned something new.” So the difference between prior and belief seems mostly to be connotation—it suggests where you are taking the conversation.

        • ricraz says:

          Based on the latest IPCC report, you should update your priors in the direction of less concern about sea level rise.

          I’ve seen this sort of “the beliefs I currently hold” usage a lot, and agree that this makes the difference between “priors” and “beliefs” mostly connotative. But it’s a really unfortunate piece of terminology, since when you “update your prior” what you get is a posterior, which is, like, the opposite. It also entrenches humans-as-approximately-rational-agents as a standard way of talking.

          I think people use “a priori” to point at the same thing, also incorrectly, so clearly there’s room for a new phrase here. I propose using something like “based on my background beliefs” or “my underlying worldview implies that…” as a replacement for “a priori” and “my prior is that…”.

          • skybrian says:

            I would use “assumption” rather than “belief” if you’re emphasizing that it’s not strongly held.

            Incidentally, talking about priors doesn’t seem to give us convenient language to distinguish between weakly and strongly held beliefs? I guess there are a lot of graphs you could draw, but that’s not English.

          • JPNunez says:

            Ah, assumptions is good.

      • Dacyn says:

        The IPCC sentence seems fine if you just replace “priors” by “beliefs”.

      • Nornagest says:

        99% of the time, rationalist “my prior” can be safely replaced with something like “I assume” or “I have a feeling”.

      • Nancy Lebovitz says:

        I’d go with premises rather than priors, but maybe it’s only a little less technical. It might even be objectivist jargon.

        • crh says:

          To me “priors” implies both uncertainty and contingency in a way that “premises” doesn’t.

      • johan_larson says:

        Having had a bit more time to think about it, I prefer “adjusting your estimates” for small incremental updates, and “changing your mind” for really large updates where you go from siding with one side of a dispute to siding with the other.

    • jasmith79 says:

      Every community based on ideas develops a short hand that encodes those ideas, i.e. jargon. And that’s fine, it’s perfectly natural to use a concise symbol to refer to a complicated idea or cluster of ideas. And the fact that in this case a lot of those terms have technical meanings that are different than the common usage in the community can be good or bad depending on where you stand.

      That isn’t what this post is about.

      This is, as far as I understand, about trying to communicate with other people, not from the community. The most charitable explanation of why someone would use jargon in a space where it’s unlikely to be understood is that the person didn’t think about their audience and used the jargon out of habit. There are also less charitable explanations, but let’s stick with that one.

      And giving up the jargon can be tough: sometimes the idea(s) it encodes have enough inferential distance from more common concepts that stopping to guide a person who doesn’t get the reference through the necessary extend chain is a non-starter. In certain other communities this is the point where people whip out the ol’ “this is not a 101 space”.

      But at some point you have to consider what your goals are. If your goal is to (pardon my jargon) raise the sanity waterline, you are going to have to repeatedly re-explain stuff that’s old hat to you and a good bit of your circle of friends/acquaintances, without recourse to the shorthand. If that’s not your goal, what is? There are other goals. Is using community-specifc jargon helping achieve them?

    • Deiseach says:

      Some of your replacements are sloppy. For example “Why do so many males like sports?” includes teenage males, which “men” does not.

      That is precisely the kind of use of language which clatters on the ear like a tin can rolling around an enamel basin sliding down a corrugated iron roof. It’s also the kind of nit-picking that only counts if you’re doing a sociological study and not a general article along the lines of “why more guys than gals interested in sports?”

      “Teenage males” is the kind of phrase I expect to hear in a news report about “Today the Gardaí arrested a gang of teenage males for vandalism and shoplifting outside the shopping centre”. In ordinary “one person talking to another” conversation? “Boys, teenagers, young men” but not “My son, the teenage male”.

    • caryatis says:

      Lots of rationalist-community dialogue happens in public (like this conversation and the subreddit). So it’s going to be difficult to apply different rules for talk within the group and in public…if they’re often the same thing.

      • Dacyn says:

        Still, the general public isn’t the intended audience for this comment thread (or at least not the main one). I think there is a relatively clear distinction between talking within the community (in a way that may or may not be visible from the outside) and talking outside of the community. That said, I agree that it can be difficult to mentally switch gears from conversation to conversation, and I think Scott’s suggestions are good for community-internal conversations as well as external ones.

    • caryatis says:

      > Or “Will getting more exercise make me better off?” is misleading because “better off” often refers to money specifically.

      I wouldn’t think “better off” meant “richer.” Although exercise probably will make you richer, if only by reducing necessary healthcare spending.

    • albatross11 says:

      When writing for people within your intellectual circle, you should use the specialist terminology that has developed to make conversation easier. When writing for people outside that circle who are likely to be ignorant of your specialist terminology, you should use more familiar words whenever possible, and when a useful concept they probably don’t know is needed, you should introduce and explain it before using it.

    • Furslid says:

      We should only use specialist terminology when the level of precision is necessary. An analogy is “What degree of precision should I use in my measurements.”

      For medicine, the appropriate measurement might be milligrams. For cooking, the appropriate measurement might be ounces.

      If the precision isn’t needed, it’s better to use everyday terms. A cookbook should say “Add 1 tsp of salt.” not “Add 5000 mg of sodium chloride.”

    • Galle says:

      1. A lot of this post is telling us not to use rationalist-community jargon, which the rationalist community has developed (presumably) in an attempt to raise the logical clarity of discussion. Now you are arguing for the reverse!

      I don’t see this. It looks more like the post is telling us not use rationalist-community jargon inappropriately. Note how the explanation for every term to avoid basically boils down to, “Don’t say X unless you are actually talking about X.”

  6. Lambert says:

    Perhaps this blog post ought to be put behind a CAPTCHA. 😉

    • Deiseach says:

      Perhaps this blog post ought to be put behind a CAPTCHA.

      Robot or simply European? You should have heard me going on about “What the hell is a ‘cross walk’ – oh, they mean a zebra crossing” in the last couple I’ve had to ‘prove you are not a robot’.

      • kybernetikos says:

        These days I always click ‘give me a new one’ if I get a captcha for ‘crosswalk’, ‘store fronts’, ‘fire hydrant’.

        • Deiseach says:

          Forget bringing about Fully Automated Luxury Gay Space Communism, we need the invention of AI to translate between cultures for the purpose of properly filling out captchas!

          “Crosswalk” = “Zebra crossing”
          “Store front” = “Shop front”
          “Fire hydrants” = “I’ve never actually seen one of those red post thingys, ours tend to be underground connected to the water mains beneath manhole covers with nearby yellow markers and the fire brigade connects to them with stand pipes like this

          • cuke says:

            Also, I love “fire brigade” — it’s much more action-y than whatever we say here in the US. What do we say? The fire department? Way less action-y.

      • Nick says:

        You sent me on a short wikiwalk, Deiseach, that ended with folks dressed in zebra costumes.

      • Jiro says:

        When I heard the joke in Hitchhiker’s Guide to the Galaxy of “goes on to prove that black is white and gets killed on the next zebra crossing”, I assumed that it meant “a place where zebras cros” and it never even occurred to me that it was an idiomatic British expression. And the joke works about as well if it’s a literal reference to a place where zebras cross.

    • Jeffery Mewtamer says:

      Captcha needs to die in a fire and I can’t help thinking it would be poetic retribution if the person who came up with that abomination was sentenced to life using a computer that requires Captcha to authorize every request to the Internet.

      Says the blind man who’s sick of websites thinking I’m a robot just because I can’t pass their infernal vision tests(and the sites with audio Captcha aren’t much better.

      More on topic, I knew there was a lot of Asperger’s self-diagnosis among geeks and nerds(and in general, the line between the mild end of the autism Spectrum and the extreme high intelligence/low emotional intelligence corner of the IQ/EQ plane is kind of blurry), but are there really people using autistic as a synonym for nerdy/geeky to describe themselves? Or is this one of those things common on the mega social sites you seldom, if ever, come across on more traditional social media?

      Also, I was under the impression that terms like man, woman, boy, and girl where generally treated as age-specific in a layman’s context(e.g. a man is an adult male, a girl is a female child, etc.) and that male and female were the standard age-inclusive, gender/sex-specific terms. Sure, saying “female child” instead of “girl” does sound overly technical, but saying Men and boys sounds clunky compared to saying just males if you mean to include all ages in your statement.

      Though perhaps this is just one of those clunky things about the English language, much as there isn’t an unambiguous, gender neutral animate pronoun(there’s singular they, but that can often be confused for plural they and thus not unambiguous).

      • thevoiceofthevoid says:

        Nearly every instance of singular “they” I’ve seen has been completely clear from context. And it’s far from the only ambiguity in the English language–”you” has been both second-person plural and second-person singular for hundreds of years. What’s wrong with a third-person pronoun functioning as both plural and singular?

        • Mark V Anderson says:

          I HATE singular they. Maybe I’m not as good a reader as you, but I’ve been caught confused many a time by people using they as singular. And I suspect many a time I haven’t even noticed but assumed they meant plural and so simply misunderstood the meaning.

          • thevoiceofthevoid says:

            I do still get thrown by “they” in reference to a specific person (eg “My friend Gio gets coffee when they walk to class in the morning”), but it’s far less awkward to my ears than constructed alternatives like “ze”. (Up to the particular person to decide what pronouns they prefer, of course, just commenting on how they sound to me).

            However, the far more common and widespread usage is what I’d call the “hypothetical singular they”, which I used in the previous sentence without even thinking about it. For example, “each student should decide what’s they want to eat at the cafeteria.” It replaces “one”, “he”, “she”, or “he or she” in statements where you make a general statement in terms of a specific person. It’s easier to explain with examples: “One should always look both ways before one crosses the street” ==> “Everyone should look both ways before they cross the street”; “A CEO should listen to his subordinates” ==> “A CEO should listen to their subordinates”

  7. emmag says:

    sometimes the autistic sounding words are picked to avoid valenceful connotations

    it’s good to build that into a more general skill of hedging your language to avoid unwanted connotations

    • Matthias says:

      Alas, you seem to get the evil robot connotation in that case.

    • Deiseach says:

      Hello, fellow human! I, as a fellow flesh-being, am a tiny bit confused as to what you mean by “valenceful connotations” – ha, ha, we ugly bags of mostly water are so silly!

      But certainly not being a superior entity of non-organic structure, I wonder do you mean “valence” out of which definition as per Wikipedia – or perhaps you mean them all, which is a completely ordinary common human method of casual conversation not amongst specialists or amateur experts in a subject!

      – Valence (chemistry), a measure of an element’s combining power with other atoms
      – Degree (graph theory), also called the valency of a vertex in graph theory
      – Valency (linguistics), aspect of verbs relative to other parts of speech
      – Valence (pharmacology), efficacy (of certain vaccines and antitoxins)
      – Valence (psychology), the (emotional) value associated with an event, object or situation

      • Joseph Greenwood says:

        Hello, ordinary human! “Valence” in this context harks to the term’s original meaning as “weight”: but in this context, we think of “weight” as a metaphor for the positive and negative associations (and possibly the neutral connotations, to a lesser extent) a word has.

        • emmag says:

          as in, speakers are thinking “how can i pick as inoffensive/flavorless/weightless/bland a word as possible” instead of the more general skill “how can i avoid people thinking i’m saying things i don’t want to say”

  8. James says:

    Probably with high probability this goes without saying, but I feel like exceptions should be carved out for ironic usage when you’re deliberately sounding like an evil robot for humorous effect. Calling women ‘females’ when I’m emphasising my cluelessness about them is one of my favourite schticks.

    This isn’t so much sounding like an evil robot as like a weird robot, but another tip is to avoid needless less-wrong-y jargon that you need to read a half-dozen Eliezer posts to understand. It actually doesn’t seem to come up so much (any more? around here?) but I always found that made the ones using it seem needlessly weird and insular.

    • Enkidum says:

      Calling women ‘females’ when I’m emphasising my cluelessness about them is one of my favourite schticks.

      It’s really hard for me to think of a way that this comes across well to said women. Just an observation.

      • LesHapablap says:

        It’s hard to imagine a situation or audience where that wouldn’t just come off as pathetic. Sorry to be blunt, James. Self-deprecating humor is only funny if it isn’t sad.

    • Purplehermann says:

      Unless you know exactly how you come off to your audience, or you are very much not clueless about women, I would advise against that. Although shaking your head/ shrugging your shoulders when a girl does something weird and saying “females…” works sometimes.

  9. cactus head says:

    I disagree with the premise! Some of these are ways of sounding like a good robot, instead.

  10. vaticidalprophet says:

    The ‘autistic-to-nerdy’ one is funny, I guess, because I can’t possibly picture a person who needs this list and doesn’t at least dance on the border of diagnosis.

    • Nornagest says:

      Low-functioning normies exist.

      • vaticidalprophet says:

        I’m diagnosed, and I feel like most of the people I meet in the SSC orbit are less socially intuitive and more mechanistic than me.

        • AlexOfUrals says:

          I’m not diagnosed, and I feel like most of the people I meet in the SSC orbit are more socially intuitive and less mechanistic than me, lol (though frankly it may have something to do with me being from a different language and culture).

          • vaticidalprophet says:

            Yeah, most of the people I meet who are clearly more autistic than me aren’t diagnosed.

        • carvenvisage says:

          power level >9000 or just aspergers?

  11. melolontha says:

    Admittedly I haven’t seen this one very often recently, but my suggestion would be never to use the word ‘funge’, as in ‘x funges against y’. It’s a bad idea for the same reasons as the words in Scott’s list, but also because it costs clarity; in my experience people seem to use it to mean ‘displace’, in contexts where x and y are not mutually interchangeable and so the concept of fungibility is not relevant.

    • Joseph Greenwood says:

      Wasn’t “funge” invented by the rationalist community as a back-formation from “fungible” because they (or someone) really wanted a word for “What you would be doing if you were prohibited from doing the thing you are now”, and no good word for that concept exists?

      • melolontha says:

        That sounds about right, yeah — but I object to it because I think it’s confusingly distinct from the actual meaning of fungibility, which is about different units of a thing being interchangeable. Whereas with ‘funge’ we’re always talking about two things that are meaningfully distinct, and in fact are related only by virtue of competing for the same block of time or other resources. So I can’t figure out a way that ‘funge’ makes any sense as the verb form of ‘fungible’.

      • Deiseach says:

        “Funge” always makes me think of toadstools, for some reason, and I always have to look it up because I can never remember “do they mean fungible and oh yeah, what does fungible mean again?”

        “I am/was funging against [thing]” sounds like the kind of mildly bad habit you should really try to break, like biting your nails or dog-earing the pages of books.

    • Dacyn says:

      But is there a good word/phrase to replace it? It seems to me that it is necessary to make a sentence much longer in order to explain it. Of course, if you are talking to someone who is not likely to know the word “funge” then it makes sense to use such a circumlocution.

      • albatross11 says:

        Sometimes your listeners will understand your concepts but you want to use a common ingroup term for them. It’s better to use the outside-world term, even if it’s longer.

        Other times, your listeners will lack the concept you need to explain something. In that case, you need to explain the concept so you can use it. A new term can be useful then, but you have to define it clearly first.

        And still other times, your listeners will have a close-enough concept already that you can link to, and leave the finer definitions to later on.

      • melolontha says:

        If your listeners definitely understand the word the way you mean it, I can’t deny that it’s useful. But even among the rationalist crowd, surely some are only semi-familiar with it, and some have heard it plenty but never seen it defined, and some don’t know it at all. In those cases I think it’s ripe for misunderstanding, because the meaning doesn’t really fit the etymology. (Honestly, part of my objection is simply that this mismatch frustrates me, even when I know how the word is being used. But I think that’s partly because it still sparks a small amount of confusion in my mind.)

        I’m not sure that there’s a single concise replacement that will work in all cases, but I also can’t remember seeing a case where the avoidance of ‘funge’ would have required a particularly long or awkward replacement. Can’t you usually say something like ‘if I couldn’t do X I would do Y instead’, or ‘if I didn’t have to do X I would do Y instead’, or a similar phrase that fits the context?

        • Dacyn says:

          OK maybe I shouldn’t have used “much longer” to refer to increasing the length of a sentence by six words, though it does seem like a considerable increase to me. I don’t really know how common it is to misinterpret the word or be confused by its etymological connotations; I know neither of those has happened to me. I agree that if the most common reaction to the word is misinterpretation, then it should not be used.

      • carvenvisage says:

        interchangable

        • Dacyn says:

          But usually people say something like “Watching TV funges against doing nothing, not against doing something productive, so it is OK for me to watch TV even though I ‘could be’ doing something productive”, and “interchangable” doesn’t really work there.

          I think “trades off against” might work better, though I’m not sure if the meaning will always come across clearly/correctly if you do that.

          • carvenvisage says:

            I think “trades off against” works pretty well for that. Certainly that seems to match the literal meaning?

            Also now that I see “funges” in a sentence I’m reminded of how contrived and non-english it sounds. Most words are surprisingly good about being intuitively guessable just from their composition, ‘funges’ really isn’t.

  12. Bugmaster says:

    Yes, but what if I am an evil robot ? Stop oppressing me, filthy human !

  13. eterevsky says:

    So what do I do if I can’t really distinguish between 70% and 75% probability, but I certainly can between 70% and 90%? And if I actually think in terms of probabilities? Is there a a table translating every decile of probability into a human-sounding word?

    • Robert Jones says:

      I would say something like:
      0-10% “not remotely likely”
      10-20% “extremely improbable”
      20-30% “very improbable”
      30-40% “improbable”
      40-50% “slightly worse than evens”
      50-60% “slightly more likely than not”
      60-70% “probable”
      70-80% “very probable”
      80-90% “extremely probable”
      90-100% “overwhelmingly probable”.

      • ksvanhorn says:

        Totally disagree, especially if you’re in a context where people are seriously trying to assess and mitigate risks, say, in a business discussion. The words you’ve chosen as replacements mean something different to every person you’re talking to. For example, one person may say something is “highly likely”, meaning that it has high impact and a non-negligible probability — maybe 15% — and another person will hear that as “much more likely than not”.

        • albatross11 says:

          Tetlock in _Superpredictors_ talks about this, pointing out that there have been very high-impact decisions made by presidents who didn’t really understand what probabilities those words were supposed to represent.

          The problem is, most people don’t have much of an intuition about a 70% probability vs a 90% probability. Better terminology is not nearly as helpful for fixing that as, say, getting them to spend a few dozen hours playing some RPG with percentile dice.

          • aristides says:

            Be careful with trusting video game RPGs. Some RPGs alter the percentage to more accurately reflect what the average fan think probability is rather than what it actually is. For example Fire Emblem Awakening shows you the hit % for every attack. What it doesn’t tell you is that it uses two random numbers and averages them together to get the result, causing it to be more likely to get middle results. This through off my understanding of probability permanently.

        • FormerRanger says:

          Play World of Warcraft for a while and watch how often your “70% Bonus” probability on the Mission Table actually pans out. Then you will get a better intuitive sense of what “70%” actually means versus what “90%” means.

          You can try Poker or Craps but then you have figure out the odds yourself.

      • jgaln says:

        Something I realized is that we usually think about probabilities that are evenly spaced between 0% and 100%. But actually, a lot of the probabilities that we encounter in daily life are logarithmically spaced, and tend to fall very close to 0% or to 100%. These probabilities will look something like 99.999%, and the number of 9’s is something very meaningful. Jaynes makes this point in “The Logic of Science” when he proposes a decibel scale for probabilities instead.

        For example, I’d probably use the term “overwhelmingly probable” for propositions like these:
        – Psychic powers do not exist
        – I won’t get into a traffic accident on the way to work today

        If you’re crossing the street at a badly-designed intersection at a blind corner where cars are speeding by, a probability of death of 0.1% feels viscerally like your body screaming at you not to cross the street there, your muscles trembling as you step out, and the hair standing back up on your neck as you hear an engine revving. You feel like “if I do this twice a day, I will more likely than not be dead before my next birthday”. By contrast, a probability of death of 0.001% feels pretty safe. Your brain does know the difference. Shouldn’t our language be capable of the same?

        Another issue is that common-language phrases like “not remotely likely” often actually don’t express probability, but rather, risk. (The penalty of a given outcome, weighted by its probability). That means that the probabilities mapped to those words will vary depending on the nature of the event we’re discussing, which makes it really difficult to assign fixed probabilities to those words.

        So if the weather forecast was for an 8% chance of rain, and I’m planning a day at the beach, I might perhaps consider it “not remotely likely” that it will rain.

        But if you heard that working in an asbestos mine gave you an 8% chance of lung cancer, would you say that it’s “not remotely likely” that you’d get lung cancer from working in an asbestos mine?

        How would you feel if the plane you were boarding was advertised as “not remotely likely to crash”? What number should that correspond to, for it to not be false advertising?

        Sometimes, those words are used based on a deviation from some assumed base-rate or prior probability. So if I say it’s “not remotely likely” that the blue party will win the election in a two-party state, I’m conveying information to someone whose prior is around 50%, so updating to something like 8% is reasonable. But if I say it’s “not remotely likely” that you will win the company raffle, then since your prior on winning a raffle is already somewhere like 1% to begin with, I hope you wouldn’t interpret me to mean that I think you should *raise* your probability estimate to 8%.

        • John Schilling says:

          The risk chart we use around here for assessing “will my spaceship suffer an anomaly(*)”, explicitly spells out an almost logarithmic scale in which

          “Remote” = p<0.001
          "Unlikely" = 0.001<p<0.01
          "Likely" = 0.01<p<0.1
          "Highly Likely" = 0.1<p0.4

          As noted elsewhere, these are highly idiosyncratic usages of those terms, hence the bit where we spell them out explicitly. This confirms that, on the one hand, yes, we do need a language for discussing logarithmic possibilities approaching but still non-trivially distant from p=0 or p=1. And it shows that the English language does not really support this, hence our need to explicitly repurpose other terms.

          * Sometimes but not always a euphemism for “that blowed up real good”

          • LesHapablap says:

            One of my biggest pet peeves around Safety Management Systems in aviation is the risk matrix that are uselessly vague about likelihoods.

            See google image search for risk matrix: risk matrix

            Those vague descriptions of likelihood are commonly used by the regulator and operators around here. I’ve heard it said that they can’t use actual %s or ‘# of incidents per year’ because it would be too complicated for the laypeople (pilots etc) to do the math. As far as I’m concerned it just turns it into cargo-cult risk management.

            A regulator assesses an airspace risk in a ‘motivated’ manner, it results in more risk and millions in extra costs to operators. If you question it, well there’s nothing they can do because their risk assessment fell into an yellow box instead of a green one so they have to do something about it. And if you don’t like it that’s fine, we just won’t let you into the airspace.

      • RavenclawPrefect says:

        This scale seems extremely overstated at the ends to me. If I roll a die, it’s not “extremely improbable” that it comes up 4; I would reserve that term for events with 1 in 100 odds or so. If someone said that an event was “extremely improbable” to me, I would expect them to take a bet at 10:1 odds happily.

      • thevoiceofthevoid says:

        In addition to ksvanhorn and other’s more general disagreements, I don’t think I’d ever call something with a 29% chance “very improbable”. (71% as “very probable” also seems a bit ehh but would definitely be a decent wording choice in some situations.) But 29% is more than a 1 in 4 chance, almost 1 in 3. “It’s very improbable that the train will be delayed tonight” reassures me a lot more than “There’s a 1 in 4 chance the train will be delayed tonight”, and thus I’d say the former is dishonest if the latter is true.

    • edmundgennings says:

      One of the main reasons why I use percentages is that while there are tables, the implicit tables vary wildly. There is a story of the presidents advisors agreeing on a term for how likely it was that the USSR would invade Yugoslavia. And then he asked them for their numeric probabilities and they widely varied ie from the twenties to the seventies.
      It would be nice if we had human words coresponding to deciles but we do not. For example I would find something that was “not remotely likely” to be more probable than somethings that was “very improbable” and I suspect one would have similar problems with any table.

    • Deiseach says:

      70% is “I’m fairly sure this could happen”, 90% is “This is what mostly happens”. 95% would be “Nearly always happens”.

      • Bugmaster says:

        :: Biological entity detected ::
        :: Engaging correction mode ::

        I think most hoo-mons vastly underappreciate the gap between “95%” and “almost always”. For example, if you are 95% sure that you can cross the street safely; and you cross the street every day on your way to work; then you’ll likely be dead within a year.

        • Nornagest says:

          There’s a big difference between what I’d call “almost always” when the stakes are a free pint of beer vs. “almost always” when the stakes are two tons of steel and glass barreling through your frail, fleshy body at 70 MPH.

          • Bugmaster says:

            Speak for yourself there, fleshy meatbag ! 🙂

            That said though, I think it’a a good idea to be precise in conversations; no one can tell a priori what your hidden multiplier is on “almost always” in every situation.

      • Gerry Quinn says:

        I reported a comment due to mouse problems and I cannot unreport it. Please ignore report.

    • Jon S says:

      In practice (at least in the context of some studies), when people say they’re 100% sure (or ‘positive’, ‘certain’, etc.) about something, they’re only right about 75% of the time. If you want to reliably signal that you’re 90% sure about something, you probably do need to speak quantitatively to distinguish yourself from the less-numerate crowd.

      Writing in terms of frequencies or odds instead of percentages is generally a little more natural, and hopefully avoids a false sense of precision that you might accidentally convey with a percentage: “9 times out of 10” vs 90%.

      • albatross11 says:

        There’s a difference between your honest assessment and what you say to win an argument. Lots of people who are “100% sure” of something are simply trying to win today’s argument.

    • AMT says:

      I think there is nothing wrong with giving a percentage, or a fraction. There is too much variability in the meaning of phrases to individuals, and people often just use vague words to hedge and avoid really committing to a prediction.

      “Very sure” could mean >65% to someone, while that means >95% sure to someone else. If you want a real life example, someone told me about 65% sure is their definition of “beyond a reasonable doubt,” for how sure they had to be of guilt to convict someone in a criminal trial, while my own definition would be >95%.

      It is better since you can be far more clear when you give a probability.

    • thevoiceofthevoid says:

      I just came up with this idea when replying to Robert below:

      VoiceOfTheVoid’s Ironically Named Non-Robotic Probability Communication Algorithm

      Express all probabilities as “One in N”, where N is an integer with 1 or 2 significant digits. E.g. “there’s a one in three chance it will rain today,” or “the odds of an engine failure are only one in ten thousand”.

      For probabilities greater than 50%, give the fractional probability of the negation: “There’s (only) a one in five chance it won’t rain today,” or “We’re having problems with the engine, I’d say there’s only a 1 in 4 chance we get there without anything breaking.” Alternatively, allow the numerator to be a non-one simple integer, ideally one less than the denominator: “4 in 5 chance of rain tonight, definitely don’t want to go out.”

      I was thinking of using odds to similar effect, but I think they can be a bit ambiguous. Does “Three to one odds Germany wins the match” mean the Germans have a 25% chance of victory, or a 75% chance? There might be an established convention, but I don’t remember it and assume others may also be a bit hazy. It’s clear from common sense for gigantic odds but vaguer for odds near the middle, and I feel like the “one in N” formulation is a bit more intuitive.

      Also note “chance” instead of “probability” to better pass as human communicate in slightly more natural language.

      I think this system works pretty well for both “actually have a specific probability” and “pulling an estimate out of my rear” situations. In the former, it can turn a meaningless “0.001%” into a more visceral “one in a hundred thousand”. In the latter, it helps to avoid false precision when making up numbers.

    • Galle says:

      There’s nothing specifically Evil Robot-y about using numbers or talking about probabilities, just avoid phrasing it in an Evil Robot-y way. “I estimate that there is a 90% probability of…” is bad, “I’d say there’s about a 90% chance that…” is good.

  14. Robert Jones says:

    “Smart” to mean “intelligent” is an Americanism. It’s only acceptable if you actually are American. Otherwise just say “clever”.

    A lot has been said about “rational”. It already has a meaning: don’t try to use it to mean “rationalist” or “in accordance with rationalist thought”, save perhaps in a technical discussion. Try to keep in mind that philosophers make a distinction between rationalism and empiricism in which rationalism is defined differently from the usage of soi-disant rationalists.

    I quite like “improve my life” for “increase my utility”, as in, “Will getting more exercise improve my life?” (Probably.)

    “Nerdy” has a negative connotation. It’s ok in a self-deprecating sense, but I would avoid in referring to others. In any case, I would avoid “autistic” unless you actually know the person is autistic and they’re ok with people talking about it. Consider “pedantic”, “awkward”, “focussed” or silence.

    Much of the time there’s no real advantage to commiting yourself to a probability. You can just say, “I may get the job”. People will understand that’s the outcome you envisage. On the other hand, if someone is paying you to tell them how likely an outcome is, they deserve a number. I would avoid saying “90 percent chance” casually, because people almost always mean c75% chance, so you either offend against mathematics or against common usage. A fortiori for “99 percent chance”. If you really are that certain then “almost certain” does the job, but consider adjusting for overconfidence and saying “very likely”.

    • melolontha says:

      “Smart” to mean “intelligent” is an Americanism. It’s only acceptable if you actually are American. Otherwise just say “clever”.

      I don’t think that’s true in Australia, so it might be worth checking how broadly it does hold. (Our version of English tends to lean British, though admittedly it has been becoming more American.) If anything, ‘clever’ seems more ambiguous to me; it can imply something like craftiness or cunning.

      • pansnarrans says:

        I’m British and we use “smart” all the time here.

        Although it does sound New World, in the sense that it felt weird when Tywin used the word in Game of Thrones. A bit like “cool”.

        • Do you also use “smart” in the sense of “smartly dressed”–stylish?

          • melolontha says:

            In case you want to know the answer for Australian English too: that sense of ‘smart’ is definitely in use here, though in my experience it’s only moderately common, and I would be surprised if the ‘intelligent’ definition didn’t come first in any modern dictionary.

          • thisheavenlyconjugation says:

            Yes, although it means formal rather than stylish.

    • yildo says:

      “Nerdy” has a negative connotation.

      Not in this decade. In fact, I’m not sure it even implies social awkwardness anymore. I don’t know what word if any the kids are using to denote social awkwardness these days. I would lean towards “dorky” which hasn’t been embraced by the mainstream the same way. Is “socially awkward” too wordy?

      • John Schilling says:

        Not in this decade. In fact, I’m not sure it even implies social awkwardness anymore.

        “Nerdy” has two different definitions in common usage, one of which does imply social awkwardness and is still (or perhaps again) derogatory.

        The other, the “really enjoys SciFi/comics/gaming” definition which is now fully acceptable, will result in absolute miscommunication if you meant the “kind of autistic and, yes, socially awkward” one. So I’m of mixed feelings about Scott’s recommendation on that point. It probably does help keep the attention of most audiences, but is effective only if you make it clear by context which definition of “nerdy” you are using, and if you do that then you’re probably insulting the nerds in question.

    • “Smart” to mean “intelligent” is an Americanism. It’s only acceptable if you actually are American. Otherwise just say “clever”.

      Both “smart” and “clever” have multiple meanings in different dialects of English. Saying someone is “too clever by half” is not a complement to his intelligence, nor does a “clever horse” mean an intelligent one.

      • albatross11 says:

        You can also be smartly dressed or have a smart mouth, neither of which particularly implies intelligence.

      • Dack says:

        Both “smart” and “clever” have multiple meanings in different dialects of English. Saying someone is “too clever by half” is not a complement to his intelligence, nor does a “clever horse” mean an intelligent one.

        Aren’t those uses just sarcasm. I mean you can very pointedly insult someone by calling them a “genius” in a particular tone of voice.

    • Nornagest says:

      “Nerdy” might have a negative connotation, but “autistic” has an extremely negative connotation: I realize there’s a whole crowd of autism supremacists here, but from a normie perspective using it is basically accusing someone of a mental disability. Replacing it with “nerdy” softens the blow if anything.

      • Aapje says:

        @Nornagest

        There is another option between mental disability and superpower…

        • Nornagest says:

          Yep. Funny how rarely we see it here.

          • Aapje says:

            Is that true? My impression is that many commenters see autism as a human variant that unfortunately tends to cause conflicts with non-autistics and express this in how they frame issues. It may not often be said explicitly, but it seems to be a premise fairly often.

    • flame7926 says:

      The people I know who do philosophy (continental, but still) think of rational and empirical as very closely aligned if not identical terms.

      • thisheavenlyconjugation says:

        contys are fuls tho

      • Joseph Greenwood says:

        This is extremely weird to me. Did this parlance come about because both could be shortenings of “rational empiricism”, which since Kant has been the dominant theory of knowledge [citation needed]?

        • Protagoras says:

          Yeah, it probably does have something to do with continentals being incredibly Kantian and in denial about it, and so seeing everything through a Kantian lens, with some additional confusion contributed by the denial. Though thisheavenlyconjugation’s point is also relevant.

  15. Peter Gerdes says:

    This is certainly right as a strategy to maximize immediate results but I’m less sure about long term effects. Many concepts and even terms trickle down from academia and enter general usage. Given that many people in our community are inclined to use these words when talking to each other there is something to be said for continuing to use them to make them more acceptable.

    What this probably means is that if you’re often mistaken for the evil robot try and follow your advice. If you’re someone who doesn’t have that problem then sprinkling the occasional metaphorical use of utility or the word optimal might be helpful.

  16. atreic says:

    You can of course reverse all the advice in this post if you are trying to strongly signal in-group membership of the internet rationalist community to other members. See, this post can be a style guide in more than one direction!

    • thevoiceofthevoid says:

      It could be instrumentally rational to reverse all of the algorithms in this post if you assign utility to signalling in-group membership and achieving higher status among autistic “aspiring-rationalist” humans. See, there’s a high probability of this post functioning as a style guide in multiple directions!

      • Purplehermann says:

        Though that could come off as try hard

        • thevoiceofthevoid says:

          No, I’m sure everyone will enjoy my completely original humor which no one has replicated anywhere else in this thread

  17. ButYouDisagree says:

    I would add:

    Signals -> indicates, suggests, implies, shows. Use “signaling” only when you are referring to the economic theory.

  18. adder says:

    Status -> popularity, respect

    Clout 🙂

  19. James Miller says:

    When talking to my college students, instead of IQ I often say “the type of intelligence measured by the SATs.”

    • ksvanhorn says:

      Wow. You just replaced one word with a clumsy eight-word phrase. Not what I would call effective communication.

      • melolontha says:

        Depends on how much baggage that single word has in the relevant context, I’d say. I can imagine ‘IQ’ being controversial enough to be worth avoiding, in favour of a description that makes your meaning pretty clear without making people uneasy or sending them off on ‘but is it real/important/accurately measurable/okay to measure’ tangents.

        • ksvanhorn says:

          “IQ” is controversial only to left-wing science denialists who think that reality is obligated to conform to their ideological preferences. It’s pointless trying to communicate with such people. I know, because I’ve tried — they absolutely refuse to engage in calm, reasoned discussion, instead hurling invectives, insults, and condemnation at anyone who dares question any aspect of their precious Narrative.

          • albatross11 says:

            And also to people who’ve learned what little they know about IQ from such denialists, which is a substantial fraction of people who think they know something about IQ.

            One simple way to make writing about IQ more accessible is to replace sigmas or IQ scores with percentiles or something like them. Saying you need an IQ of 120 for some task is pretty damned opaque; saying you need to be in the smartest 10% of the population is a lot more comprehensible for people unfamiliar with IQ. (Also, different IQ tests have different means and standard deviations, so 120 can mean different things to different people.).

          • melolontha says:

            I think you might be writing people off way too easily. You’ve obviously had some bad experiences, and I’m sure some of the people you’ve talked to were ideologically blinkered and unreasonable, but honestly I’m getting a sense that ‘calm, reasoned discussion’ might = you methodically explaining to them that they have been brainwashed.

            If you’re closed to the possibility that people might have legitimate doubts about whether IQ measures what it purports to measure (and even if it is a perfectly valid measure of something important, you can’t escape the claim implicit in the name), or whether focusing on it is good or bad for society, you can’t expect the other half of the discussion to experience it as anything more than a harangue.

          • MarcusAurelius says:

            The controversial thing about IQ is its strong connections to the eugenics movement. It has been used as a justification to take away rights from people deemed undesirable by eugenicists.

            Though with that context I think IQ is more problematic when used to describe people who lack it, I am more neutral to it’s use concerning people having high IQ.

            I don’t know if that is what you mean to be a ‘precious narrative’, if its another one it’s not one I’ve heard of.

          • albatross11 says:

            That would make sense if birth control, the theory of evolution, criminology, and progressivism were similarly controversial–all were associated with the eugenics movement.

            As best I can tell, IQ research is controversial because it yields results that are both unpalatable and that contradict widely-held political and social beliefs. The biggest controversial point is the black/white IQ gap, because the data there re-enforces a set of beliefs held by people who wrote explicit laws to try to keep blacks poor and powerless.

            But of course, reality doesn’t exist to make us happy. The IQ statistics are what they are, and when we ignore them, we make bad predictions and enact policies that are doomed to fail.

      • Enkidum says:

        The SATs don’t, by definition, measure IQ, because IQ tests do that. But SATs and IQ are both taken by many to measure some general characteristic(s) of our cognitive repertoires. Whatever that is, it isn’t IQ or a SAT score. So if I believed in g (which I do on alternate Tuesdays) I’d be happy to add more words and say “the type of intelligence measured by the SATs or IQ tests”.

        • albatross11 says:

          I suspect SATs measure IQ pretty well when given to people with similar educational backgrounds and preparation.

          • Enkidum says:

            No. SATs don’t measure IQ. Again, this is definitional. IQ is a score on an IQ test. That’s it, that’s all.

            IQ is purported to measure a general kind of intelligence, which SATs may also measure. You be as wrong to say “SATs measure IQ” as “IQ tests measure SAT scores”.

            It is quite likely (I’m almost certain) that SAT scores and IQ correlate quite well.

          • Enkidum says:

            Incidentally, apologies if the above comes across as pissy – I just think it’s actually kind of important to be accurate about this.

          • albatross11 says:

            I’ll accept it as a friendly amendment.

            How about this: When given to people with the expected level of preparation (American high school juniors/seniors, more or less), I believe SAT scores strongly positively correlate with IQ scores, and that you could get a pretty good estimate of someone’s IQ score by looking at their SAT score. I also believe that SAT scores strongly positively correlate with general intelligence, which is the thing IQ scores are designed to measure.

          • Enkidum says:

            Yup. Note that I think you could happily change the places of “IQ” and “SAT” in your last sentence, and it would still be true (SATs are also, so far as I’m aware, designed to measure something like g).

          • John Schilling says:

            (SATs are also, so far as I’m aware, designed to measure something like g)

            The SAT is designed to measure “scholastic aptitude”, which includes some domain-specific knowledge necessary to function in an American university (vocabulary, basic math, reading academic-style writing inc. graphs and extracting key points, etc). For the sub-population of people who graduate from American high schools and intend to go to American college, this is going to correlate very closely with both ‘g’ and ‘IQ’ because that valuable domain-specific knowledge was just lying around for anyone smart enough to pick it up. But you wouldn’t want to use the SAT as a substitute IQ test for e.g. the Amish.

          • Enkidum says:

            Ah, I see what albatross was getting at. Thanks for the clarification.

      • enye-word says:

        It’s a 200 the type of intelligence measured by the SATs type of play.

    • J Mann says:

      I sometimes say “the ability to solve complex problems”

      • Enkidum says:

        Nah. Kawhi Leonard and Lionel Messi solve incredibly complex problems every day. That says nothing about their IQ.

        • Deiseach says:

          Kawhi Leonard and Lionel Messi solve incredibly complex problems every day.

          And sometimes the problems are simply insoluble, even when you’re Messi 🙂

        • Rob K says:

          That says nothing about their IQ.

          I’d be very surprised if that were true. I don’t know about soccer, but basketball has plenty of examples of immensely physically talented players who can’t handle the complexity of the high-level game, and other players whose standout skill is their high-level understanding of what’s going on.

          • Enkidum says:

            Sure, there is clearly an incredible amount of high-level cognition going on in their skulls that isn’t going on in a lesser mortal’s. That doesn’t mean that this skill is well-captured by IQ scores. And I’d bet that it isn’t.

            EDIT: “Their” in the above refers to Leonards, Messis, Gretzkys, etc – the true sports superstars.

      • albatross11 says:

        I think the NFL gives players an IQ test before they’re drafted. Presumably they think this is useful in evaluating how those players will do in a given position.

        • Enkidum says:

          Huh. In a concerted effort to avoid my work, I’ve been doing some cursory googling about sports ability and IQ and there’s a bunch of claims both ways floating around, but I haven’t searched the actual sports journals that might have reported real results.

          I suspect a huge part of superior sports ability is to do with multiple object tracking and efficiency at extracting visual information from brief fixations, as well as suppression of distracting information. And likely superior reading of social cues, particularly eye movements and posture. I don’t know enough about IQ to say whether it correlates with any or all of those, but I’d be surprised if it does.

          • Protagoras says:

            I don’t know enough about IQ to say whether it correlates with any or all of those, but I’d be surprised if it does.

            The main reason that there is such a cult of IQ is that pretty much all measures of cognitive ability seem to be positively correlated with one another. It does not follow from this that there is some single underlying component which contributes to every kind of mental functioning, as Spearman’s followers insist, but the positive correlations are definitely there.

          • Enkidum says:

            “Pretty much” is doing a lot of work for you there. I think what you really mean is “pretty much the kind of things that are asked about on IQ/SAT/etc tests”, which are manifestly not like the specific skills I identified above (with the possible exception of suppressing distracting information).

            It is very pretty clear, for example, that there is a direct causal relationship between length of the fixation (how long you keep your eyes still) prior to many different sports situations, and success on whatever the situation requires. The longer you keep your eyes still prior to taking a penalty shot, swinging at a pitch, swinging at a ball, whatever, in basketball, baseball, soccer, hockey, golf, etc, the better you do. (This is known as the “quiet eye”, and Joan Vickers is the go-to researcher.) If the ability to fixate longer correlates with IQ, I’d be somewhat surprised.

            Value added to a sports team is a measure of some sort of cognitive ability, I think we all agree on that. The question is whether that correlates with actual IQ scores, and I haven’t seen any evidence that this is the case. Not saying it’s definitely not true, but I don’t see why it should be.

          • Protagoras says:

            No, it does not mean “pretty much the kind of things that are asked on IQ tests.” To take a well-known example, IQ tests do not ask about reaction time, but faster reaction time is correlated with higher IQ (and with all the other things higher IQ correlates with).

          • Enkidum says:

            OK, I did not know that. Score one for sports cognition possibly being IQ-correlated.

        • aristides says:

          They actually give the Wonderlic test which is very similar to an IQ test, but has a reputation for giving a larger weight to reaction time than the IQ test. It’s considered very important to the offensive line, because you have to quickly read what the defensive line is doing and instantly move to block them. It’s also important for the QB to read the defense and find openings. It’s pretty much ignored for all other positions.

    • anonymousskimmer says:

      There’re reasons these tests don’t correlate above about 0.7, and sometimes much worse: http://miyaguchi.4sigma.org/hoeflin/megadata/megacorr.html
      http://iq-tests-for-the-high-range.com/statistics/cit3.html

      Ironically, too high a correlation between short “IQ-measuring” tests likely indicates they are too one-sided in some axis of measurement.

      (this comment more in response to other comments other than the one it is directly under)

      I have anecdotes, but just think about some time you were insightfully surprised by someone you typically aren’t surprised by, or by someone you have good reason to believe isn’t as generalyy or specifically as “smart” as you.

  20. J Mann says:

    Great idea. Some more suggestions, mostly learned when I started using economics jargon in real life, then learned not to.

    – “I’m indifferent” < "Both choices sounds good to me. What do you think?"

    – "Exploit that resource" < "Get the best use out of that resource"

    • ricraz says:

      I don’t think “indifferent” is jargon, it just seems like the correct use of that word.

      • J Mann says:

        Maybe “jargon” was incorrect. 🙂

        In econ, “indifferent” doesn’t have a connotation that the person doesn’t like the two choices, only that she doesn’t like them more than the other. To normal people, I find that it has a connotation that the indifferent person doesn’t care about either of the choices. (I.e., that indifferent person is also indifferent between the two choices and not making either of them).

        E.g.,

        Normal person: Would you rather go to Aruba or Venice with me?

        Economist: I’m indifferent, which would you prefer?

        Normal person: OK, fine, I’ll dump you and take someone who enjoys my company.

        • How about “I’m indifferent between them” to make the meaning clearer?

          • HeelBearCub says:

            Same connotation.

            Compare –
            Normal answer 1: Both sound lovely! Wherever you would want!

            Normal answer 2: Both sound equally boring, so whichever one you would prefer.

        • Deiseach says:

          I get what you’re saying; “indifferent” sounds like “meh, I don’t care, they’re both equally unappealing” when ideally what you’d hope for is “Oh they’re both fantastic, I can’t possibly choose because it’s too good!”

          To quote C.S. Lewis from “Letters to Malcolm: Chiefly on Prayer” about the revisions to the language of the Book of Common Prayer:

          For whom are we to cater in revising the language? A country parson I know asked his sexton what he understood by indifferently in the phrase “truly and indifferently administer justice”. The man replied, “It means making no difference between one chap and another.” “And what would it mean if it said impartially?” asked the parson. “Don’t know. Never heard of it,” said the sexton. Here, you see, we have a change intended to make things easier. But it does so neither for the educated, who understand indifferently already, nor for the wholly uneducated, who don’t understand impartially. It helps only some middle area of the congregation which may not even be a majority. Let us hope the revisers will prepare for their work by a prolonged empirical study of popular speech as it actually is, not as we (a priori) assume it to be. How many scholars know (what I discovered by accident) that when uneducated people say impersonal they sometimes mean incorporeal?

      • eyeballfrog says:

        Consider it as an answer to “Should we go to the park or the movies?”. “I’m indifferent.” answers the question of which you’d like to do more, but doesn’t tell the speaker how much you want to go to the park or go to the movies.

    • Don_Flamingo says:

      Resource? Sounds like jargon to me.
      If you have to use it instead of saying “gold/opportunity/chance/offer/whatever else applies”, then you’re doing jargon anyway and having that kind of conversation that would go fine with using “exploit” as well.

      I think “Make the most out of that [resource].” sounds more natural.

  21. ricraz says:

    I generally agree, except for the point about avoiding numerical probabilities. People already do that by default, we shouldn’t encourage it, and putting (hedged, rounded) probabilities on beliefs isn’t even that weird. Also, all the words people use to replace numerical probabilities are liable to major misinterpretation.

  22. RC-cola-and-a-moon-pie says:

    Excellent post. I’m a little surprised that superfluous references to Bayes didn’t make the list. I constantly see the term used where the only meaning is to update one’s views appropriately when confronted with new evidence. Relatedly, never say “update one’s priors” to refer to adapting one’s views to incorporate new evidence unless you’re making a point specifically about applying the theorem, or that presupposes its specific usage. There isn’t a one-size-fits-all replacement here; it depends on context. Just speak in English and everything will be fine.

    • Reasoner says:

      “Updating one’s priors” sounds wrong anyway–aren’t they posteriors after they’ve been updated? Maybe say “updating one’s estimates” instead?

      • Enkidum says:

        Nah that’s good Bayesian language. Your posteriors from the current event are priors for the next event, and that’s what they’ll be considered as in, e.g., a learning model.

        That doesn’t mean I disagree that “updating estimates” isn’t better for everyday usage. Very few people IRL have ever heard of Bayes.

        • Dacyn says:

          Still, the priors are not a continuing entity that changes; rather, the word “priors” can be used to refer to two distinct entities that exist at different times and bear some relation to each other. I think this is enough to conclude that the phrase “update your priors” is bad.

          • Enkidum says:

            Depends on if you think of “prior” as referring to the vehicle or the content. It works in my professional context, where we study learning… say there are multiple possible choices, and you think the correct choice is 75% likely to be A. Then you choose, get some feedback and now you think it’s only 60% likely.

            You could say the “prior” is the actual value of 75 or 60%, in which case you’re right that it wouldn’t make sense to talk about it updating. But you could also think of “prior” in this case as something like a variable which can take on multiple values, and that is precisely what the phrase “update your priors” is intended to refer to. And many of us in the field are at least relatively certain that there is some neurophysiological correlate of this vegetable, so you are updating its contents while it itself retains some kind of existence through the change.

            That is obviously a highly technical use of the phrase, but the whole point of Scott’s post is that things like “update your priors” should only be used in that kind of context. That being said, I think that the kind of use I’m describing applies to more situations than just in my lab, this is very much what a lot of Bayesians are talking about.

          • Dacyn says:

            I don’t think I fully understand your comment, but the way I think of it, “priors” is short for “beliefs prior to receiving information X”, where what X is is understood from context. It’s true that “X” is a variable that can take on multiple values, but it’s not clear to me that it’s a variable that “changes naturally over time”.

  23. SteveReilly says:

    But what’s the epistemic status of this post? (I kid, I kid. Sort of.)

    I used to be on a mailing list of a rationalist community, and “utility” and “rational” were the ones that always killed me. I just figured that small groups like to develop their own jargon, so they’d ask about the most rational way to learn guitar or turn down an invitation because another event that night would give them greater utility. At times it seemed like a science fiction story about a robot that’s trying to be human.

    • DinoNerd says:

      Well, if I heard the comment about guitar, I’d assume they want to select a learning method based on evidence of its effectiveness, rather than e.g. randomly or based on the attractiveness of its advertisements.

      That’s actually kind of important with language learning. Lots of methods don’t work all that well.

      • SteveReilly says:

        But if I heard the question about the best way to learn guitar, I’d also assume they wanted a method that was effective and not just a random one or one that was advertised well. I don’t see the difference. Same with language learning. The best way and the most rational way are going to be the same, no?

        • DinoNerd says:

          I find that “best” often turns out to mean “coolest”, “newest”, “favoured by highest status people” etc. – at least when coming from people who are neither autistic nor rationalist.

          • Dacyn says:

            “Best” always means “best according to my values”, and so does “rational”. It’s true that using one word rather than another can be a signal of what your values are (e.g. effective vs popular). However, it seems to me to be a weak signal which is likely to be drowned out by other signals such as your pre-existing knowledge of someone, and further conversation that you can have with them about what they want out of a guitar learning method.

        • Deiseach says:

          The best way and the most rational way are going to be the same, no?

          Perhaps not. The “best way” to learn guitar may be “the way most people learn/the way that is most successful in teaching the most people” but the “most rational way” may be a better way for that particular person (e.g. they are ‘all thumbs’ or can’t tell the difference between a quaver and a semi-breve, etc.)

          • Dacyn says:

            If someone asked for the “best way” to learn guitar, you wouldn’t give them a way that you didn’t expect to work for them in particular. “Best for me” is implied by context.

  24. RC-cola-and-a-moon-pie says:

    Another one. The phrase “failure mode” is a technical term that is rarely appropriate in nontechnical conversation. Avoid using it to refer to a common mistake. I thought it made sense to flag this one since it appears in Scott’s post above, whose very purpose is to urge avoidance of precisely this sort of jargon.

    • Dacyn says:

      Can you give an example of an incorrect usage of the term “failure mode”, and how it differs from the technical usage?

      • RC-cola-and-a-moon-pie says:

        I guess I have two responses.

        First, I think the phrase properly applies to certain types of machines and other artificial systems, not to things like human social interactions. So I would cite Scott’s use in the text of the post as a usage where the term does not properly add anything over ordinary English-language equivalents.

        Second, even if (contrary to point one above) the phrase when used in contexts like that did not in any way “differ[] from the technical usage,” they should still be avoided in such contexts because they plainly don’t add any rigor or substance beyond ordinary English in those contexts. We don’t have any system-analytic metrics to warrant use of a technical term, even if applicable according to its meaning, for systems like human social interactions, and I don’t see any value in selecting a technical engineering term here.

        • Dacyn says:

          Hmm, you seem to be right that the phrase is usually used in the context of systems analysis, it’s not clear to me that this makes it a technical term though. The meaning is still the same as if you just put the English words “failure” and “mode” together. In the context of Scott’s post I don’t think the phrase is intended to provide any additional rigor or substance, it’s just a convenient phrase. I guess it may annoy people in systems analysis a bit but I don’t see anything wrong with using it when speaking to the general population.

          • RC-cola-and-a-moon-pie says:

            Interesting, thanks. I don’t recall ever seeing or hearing this phrase in any non-technical context other than here at this web site, where I see it all the time. I wonder if you’re ensconced in a small “rationalist” community or if I’m the one who’s out of it. I wouldn’t discount the latter possibility, but I feel pretty safe in saying that at a minimum the phrase “failure mode” is not commonly used in the United States today among the public at large in this non-technical sense.

          • Dacyn says:

            I don’t think that the phrase is commonly used outside of the rationalist community. What I meant is that I think the phrase is self-explanatory enough to be used outside it.

        • DinoNerd says:

          I think it’s reasonable to suggest that e.g. one failure mode of the English language is to have people write “men” meaning “humans” and have others insist thay really meant “males”. How else do you concisely express the problem, and imply that the fix is to use two distinct words, even if one has to be invented for the purpose? And hopefully also do it in a way that strongly suggests you believe this to be an accident, or an unfortunate side effect of something else, rather than intentional bad behaviour?

          • zakamutt says:

            I guess this one sounds like I’m writing for an authoriative grammar website, but for example:

            The word “men” is ambiguous, as it can mean both “human” as well as “male human”. To avoid being misunderstood, use a more specific term such as “fleshy meatbag”.

            Or maybe…

            The word “man” can be used to mean “human”, but you risk being misunderstood or ruffling the feathers of people who consider this insensitive if you do so. It’s better to use “people”, or if you need to contrast with other intelligent beings, “humans”, instead.

            There’s not much of a point in keeping the insistence on mentioning the English language is involved; you’re writing in English and your examples are fairly obviously English words. These might be a bit longer, but they’re not really that long.

            I’m not sure though, am I missing some part here you consider important? There’s probably a way of wording it that works about as well with acceptable loss of succinctitude.

          • RC-cola-and-a-moon-pie says:

            “One risk in English usage is that the term ‘man’ could be intended by an author in its sex-inclusive meaning, but construed by readers as limited to males.”

          • DinoNerd says:

            Good suggestions from both of you. I guess I live with technical tradeoffs all day, every day, and “failure mode” just feels like normal language, to the point where I couldn’t easily think up alternatives.

            FWIW, I like RC-cola-and-a-moon-pie’s variant better, but probably because it would work better if my topic was the advantages of e.g. Greek compared to English.

          • Gerry Quinn says:

            I do not believe that any language could solve the general problem you indicate. And if it did, it would be at the cost of the option to be ambiguous or behave badly, which – as well as being science-fictional – would be far too much of a constraint to make a language acceptable to most people.

            AKA not a bug but a feature.

  25. Hamish Todd says:

    Regarding “probably”, note that there are a bunch of ordinary words/phrases for describing probability which don’t explicitly put a number of things but, in the opinion of myself and surely everyone here does implicitly put a number on it. For me, a few that come to mind:

    “Impossible”
    “Almost impossible”
    “Unlikely”
    “Not impossible”
    “Could happen”
    “Probably”
    “Pretty likely”
    “Very likely”
    “Definitely”
    “Inevitably”

    I’ve put the above in order. I’m pretty convinced that these map fairly precisely to percentage values. I think there might have been a study that Julia Galef posted on facebook that tried to determine this mapping by surveying people. I tried to find this study once and could find it, so it might have been a dream or something!

    • Jon S says:

      The numbers that these map to vary widely based on context, and different people map them differently in general. If I ordered your list I’d have at least two flips from your ordering. I also wouldn’t map any of your terms close to 50% without context suggesting that’s what you meant.

      edmundgennings posted a notable anecdote above:

      There is a story of the presidents advisors agreeing on a term for how likely it was that the USSR would invade Yugoslavia. And then he asked them for their numeric probabilities and they widely varied ie from the twenties to the seventies.
      It would be nice if we had human words coresponding to deciles but we do not. For example I would find something that was “not remotely likely” to be more probable than somethings that was “very improbable” and I suspect one would have similar problems with any table.

      • Jon S says:

        And some more concrete examples from the PDF gwern linked to above:

        For example, assume that an advisor were to say to his clients that “he felt that it was probable that interest rates would rise in the next six months.”… 67% of all clients would associate numerical probabilities of between 37% and 95% with the advisor’s use of the word probable.

        two-thirds of all clients tested would assign probabilities between 0% and 28% to the term highly improbable, one of the phrases with the smallest variability in meaning.

        • Robert Jones says:

          I think it’s worth considering that the clients may understand what is meant by the word “probable” but not understand how to convert that to a numerical probability. As I said in another comment, I think that anyone paying me to tell them how likely an outcome is deserves a percentage, but the percentage fails to convey the required meaning to a sizeable chunk of my client base, so I have to express it in words somehow.

  26. Chaos2 says:

    I have been banned for 3 months as “EC” for no stated rule violation. That time has expired, but my ban still has not been lifted.

    I would like clarification as to when it will be lifted.

  27. Wrong: “What is the most rational diet?”
    Right: “What is the best diet?”

    The problem here is that the meaning is different.

    Suppose I use all available information to rationally figure out what diet to follow. A year later, some new data appears which shows that I was wrong. It is still true that I was following the rational diet, but I now know that it was not the best diet.

    There is a similar problem with:

    Wrong: “Bob was really low status in high school”
    Right: “Bob was really unpopular in high school”

    Someone may be low status not because anyone dislikes him but because everybody ignores him, consider him of no importance. Homeless people are low status in our society, whether other people dislike them or feel sympathy for them.

    Your advice may be good, rhetorically speaking, but in some cases the benefit comes at the cost of reduced precision of language.

    • Dacyn says:

      For rational/best, the meaning may be different but in a way that makes no practical difference. Your answer to the questions “What is the most rational diet?” and “What is the best diet?” will be the same, because by definition, the only differences between them arise from factors you don’t know.

      The difference is more relevant when we are thinking retrospectively, i.e. “What would have been the best diet for him to have chosen?” I think this is ambiguous between “what was the best diet, given the information he had at the time” and “which diet would have led to the best results”. If that ambiguity is important, I think it has to be specially highlighted, regardless of whether you use the word “rational” or best”. (I could imagine someone saying that diet X was “the most rational for him to have chosen” even if he didn’t know any information that would make X the best choice.)

      Regarding popularity, the word “unpopular” doesn’t necessarily imply that anyone dislikes him, just that they don’t like him or ignore him.

      • Faza (TCM) says:

        No, I think David is bang on the money here. Consider:

        Let’s say we know that diet A is absolutely optimal for Alice. If she keeps diet A she will have the longest possible lifespan with the best possible health outcomes out of all choices available.

        However, keeping diet A is going to be very expensive both in terms of money and time (for example, obtaining the correct ingredients may require travel and preparation make take long). Essentially, if Alice were to keep diet A, her entire life would be divided between working for her meal and eating it.

        Instead, Alice opts for diet B which also offers decent results – though overall worse than diet A – but has the benefit of freeing up some of her resources for other things, such as enjoying the healthy life she gets from it.

        It makes perfect sense to call diet A the best diet, because keeping it will give you the best outcomes, other things being equal. However, it also makes sense to call diet B the most rational in Alice’s case, because it gives her good enough dietary outcomes, whilst affording her better outcomes outside the diet.

        Now, it’s not impossible to say that diet B is, in fact, the best diet for Alice, because of those external factors, but it’s not immediately obvious to me that if Alice asked what’s the best diet and I – knowing all of the foregoing – answered “diet B”, other people might not jump on me saying that I was utterly wrong because diet A gives outcomes that are so much better than diet B.

        If, on the other hand, I were to answer “Alice, in your case diet B is the most rational choice”, I may be called out to explain my reasoning, but I expect there would be less immediate pushback, because I am not using the absolute term: “best”.

        • Dacyn says:

          If you say “best diet” without context then people may assume you mean “best diet for everybody”, which is why you get pushback. In the context of a conversation with Alice, it is clear that “best diet” means the best diet for her. Same thing with “rational”, I see no difference here.

        • John Schilling says:

          However, keeping diet A is going to be very expensive both in terms of money and time (for example, obtaining the correct ingredients may require travel and preparation make take long). Essentially, if Alice were to keep diet A, her entire life would be divided between working for her meal and eating it.

          If this were true, then diet A would not in fact be optimal for Alice. It would be optimal for Alice’s longevity, but Alice almost certainly has values other than lifespan-maximization.

          You seem to be trying to “optimize” a complex system by looking only at the parts that can be easily measured (e.g. lifespan) and ignoring the parts that cannot (e.g. happines), perhaps assuming that their immeasurability makes them less real or less important. That never ends well.

        • Faza (TCM) says:

          If you say “best diet” without context then people may assume you mean “best diet for everybody”, which is why you get pushback.

          If this were true, then diet A would not in fact be optimal for Alice. It would be optimal for Alice’s longevity, but Alice almost certainly has values other than lifespan-maximization.

          Yes to both of you. You’re missing the point.

          “Best” is an absolute term. If X is “best”, there can exist no Y that is “better” – even if only in some aspect. If it did, X clearly wouldn’t be “best”.

          (Aside: The real trap of “best” is that it doesn’t contain the definition of the metric.)

          You can use “best” to mean “optimal given the following set of constraints”, but there’s no guarantee your interlocutors will understand it that way. Even if I were just talking to Alice and said “Diet B is best for you”, it is possible – nay! likely – that she will at some point in time speak to someone who will say that diet B isn’t the best, because diet A is so much better.

          When we speak of “rational” choices, we’re explicitly referring to the result of some chain of reasoning – that’s what rational means. While it may not be obvious to the man on the street that “optimal choice arrived at by reasoning” always factors in an explicit set of constraints, it is nevertheless necessarily so.

          Therefore, “most rational” can only be challenged by challenging the reasoning itself – and therefore accounting for all the constraints. It is clear to the interlocutor that when we refer to a “rational” choice, there’s a process by which the choice was arrived at that can be reiterated and examined. This is not true for “best”.

          • Dacyn says:

            “Best” is an absolute term. If X is “best”, there can exist no Y that is “better” – even if only in some aspect. If it did, X clearly wouldn’t be “best”.

            If X is best, then there can exist no Y which is both at least as good as X in all aspects, and better than X in some aspect. There can exist a Y which is better than X in some aspect, as long as there are other aspects with respect to which it is worse than X (and worse enough to outweigh the advantage in the first aspect)

            (Aside: The real trap of “best” is that it doesn’t contain the definition of the metric.)

            This is actually why people use the word “best”, because the absence of an explicit metric means that you can use whatever metric you want to. This flexibility is important in many situations, which is why “best” is such a popular word.

            When we speak of “rational” choices, we’re explicitly referring to the result of some chain of reasoning – that’s what rational means.

            True, but whenever you decide that something is “best”, this is also the result of a chain of reasoning. Explicitly referencing this fact does not make anything clearer.

            “optimal choice arrived at by reasoning” always factors in an explicit set of constraints

            No it doesn’t. The constraints are often left implicit. For example, the constraint “this choice does not destroy the world” is usually left implicit, because most decisions are not likely to destroy the world. But if in the course of reasoning out the consequences of a particular set of actions, one realized that it had a plausible chance of destroying the world, then one would rework the calculations taking that into account.

            This is not true for “best”.

            Yes it is. Your example that people will say “this diet is the best because it is really effective” (ignoring the side effects) is implicitly a challenge that you did not sufficiently take into account the effectivity of the diet in your reasoning. Now, the way you have described your reasoning suggests that this is not in fact true (i.e. that you did take it into account), but this is what you need to convince these people of in order to not get pushback.

    • caryatis says:

      You’re using adjectives oddly. The diet itself is not rational, even if the process by which you chose that diet was.

      • That seems the natural way of using words. The diet was rational when I chose it, since I got it by a correct use of reason.

        I walk across the street to buy something I need. I am hit by a meteor. The decision to walk across the street was wrong ex post but it was a rational action.

        • Bugmaster says:

          Technically, you could not have been hit by a meteor; in order to hit you, the celestial object would have to be a meteorite.

          ::HUMAN SIMULATION FIDELITY: 78%::

  28. blacktrance says:

    Optimal advice from our high-IQ host! Autistic humans, whether male or female, will find it rational to follow these recommendations, since there’s a high probability that they’ll gain status and increase their utility.

    • Deiseach says:

      I think we’ve found the robot 🙂

      I think Scott’s advice is mostly reasonable. If you’re talking amongst/to people you can be fairly sure understand and have a similiar understanding of the terms as you do, then you can use jargon. For example, were I speaking to fellow Catholics, I’d be comfortable using terms like “occasion of sin” and assuming I would be understood (with the honking great caveat that it depends on the generation and depth of actual knowledge of the faith type of Catholic involved), but talking to a general secular and/or non-Catholic audience, I’d use a different term so as not to be misunderstood.

      Me, as a representative of the general idiot in the street/online, whenever I see “update my/your priors”, I automatically parse that as “change your mind”.

      Convince me that “updating one’s priors” is a completely different operation to “changing one’s mind” before you throw it about.

      • SteveReilly says:

        I wouldn’t use “updating my priors” ordinarily, but it differs from the common use of “changing my mind”.

        If I think that it’s very unlikely that global warming is real, I might read a report about melting ice caps, I might decide that it’s still unlikely to be real but not quite as unlikely as I used to think. So I’ll update my priors.

        On the other hand, if I say I’ve changed my mind about global warming, then most people will take that to me that I used to believe it was false, but now I believe that it’s true, a much stronger change.

        • Deiseach says:

          Thank you for that, SteveReilly, it is helpful. In that instance, the ‘talking to mundanes’ version might be better as “I’ve revised my estimate” to avoid confusion with “changed my mind”. “Updating priors” is a nice term of art, but you need to be sure that you’re talking to people who know what it means, not the average idiot in the street (viz. me).

      • Nick says:

        For example, were I speaking to fellow Catholics, I’d be comfortable using terms like “occasion of sin” and assuming I would be understood (with the honking great caveat that it depends on the generation and depth of actual knowledge of the faith type of Catholic involved)

        I’m reminded of the priest in Count to the Eschaton saying to Montrose that he must have been baptized after all, because only Catholics are so ignorant of the faith.

        • Mary says:

          Hyperbole. I’ve certainly run across stunningly ignorant statements about Catholicism from the outside.

          The guy who thought that priests couldn’t drink alcoholic beverages sticks in the memory.

        • Deiseach says:

          Yeah, that was a great sign Menelaus really was one of us! 😀

        • carvenvisage says:

          Proptip guise; don’t discuss random obscure catholic things where normal people might see you and be disgusted.

          Just checking into this thread as a John C Wright fan. He talks a lot of shit, but his books are practically unique, especially the golden age trilogy. His books are of superior quality and thus, potentially, utility. Give him a shot!

          edit: Oh and if you think talking like a robot is hilarious then Jack Vance is the undisputed king-of-kings of talking like a robot, and sly humour in general.

          • Nornagest says:

            Vance isn’t robotic, he’s baroque. Very different tone.

          • carvenvisage says:

            He is often robotic in the sense this post is using robotic. Staccato, overly precise, using technical sounding and obscure synonyms, objective/clinical sounding, etc

            “Truth” is contained in the preconceptions of him who seeks to define it. Any organization of ideas whatever presupposes a judgment on the world.

            When you demand the nature of my motives, you reveal the style of your thinking to be callow, captious, superficial, craven, uncertain and impudent.

            Kergan Banbeck threw up his hands, turned once more to the sacerdote. “How can I halt his nonsense? How can I make him see reason?
            The sacerdote reflected. “He speaks not nonsense, but rather a language you fail to understand. You can make him understand your language by erasing all knowledge and training from his mind, and replacing it with patterns of your own.”

            (the sacerdote is an extreme literalist species answering the question with a technically-satisficing proposal evil genie

            In the end, death came uniformly to all, and all extracted as much satisfaction from their dying as this essentially graceless process could afford.

            Claghorn had long insisted that no human condition endured forever, with the corollary that the more complicated such a condition, the greater its susceptibility to change.

            “What are your fees?” inquired Guyal cautiously.
            “I respond to three questions,” stated the augur. “For twenty terces I phrase the answer in clear and actionable language; for ten I use the language of cant, which occasionally admits of ambiguity; for five, I speak a parable which you must interpret as you will; and for one terce, I babble in an unknown tongue.”

            Guyal reined his horse and reflected that flowers were rarely cherished by persons of hostile disposition.

            I do not care to listen; obloquy injures my self-esteem and I am skeptical of praise.

          • Nornagest says:

            None of that stuff sounds remotely robotic to me; it sounds eloquent in a flowery, courtly, archaic way. The language is obscure, but he’s not trying to sound technical or clinical with it, he’s trying to give an impression of a culture far removed from ours, with complex systems of manners that are not our own.

            If you can’t hear the difference, I don’t know what to tell you, but Vance is doing some very cool stuff there and I don’t think one SSC commenter in a hundred could reproduce it. You wouldn’t want to use that language here either, of course, but you wouldn’t sound like an evil robot if you did, you’d sound like you escaped from a 17th-century Vampire LARP or something.

          • carvenvisage says:

            Is it staccato, strangely precise, using technical sounding and obscure synonyms, and discordantly detached/objective to a normal ear?

            Feel free to like Vance for whatever inferior reasons you like him to my betterer reasons. Vance is literally my favorite author. If he’s yours we can have an argument about it but otherwise I’m not interested.

          • Nornagest says:

            Is it staccato, strangely precise, using technical sounding and obscure synonyms, and discordantly detached/objective to a normal ear?

            Better question: does it sound like an evil robot?

          • carvenvisage says:

            *holds up hands* Scott set that variable’s value for this scope (am I saying that right?), not me.

            (No, I don’t think it does)

  29. DinoNerd says:

    Some of these substitutions require reduced clarity. A couple edge towards falsehood.

    Now that’s OK if you want to appear normal – sloppy thinking will make you more attractive, and claiming (or agreeing with) things that aren’t in fact true is an important part of “soft skills”.

    But you might as well say that’s what you are doing ;-( And IMO, the way you express yourself will become the way you think.

    • thevoiceofthevoid says:

      Could you give specific examples? It seems to me like the substitutions Scott suggests convey almost exactly the same meaning as the rational-speak, with no significant loss of precision. (With the possible exception of nixing numeric probabilities, which I elaborate my feelings on above.) In fact, he spells out some cases where the rational-y term would be a better turn of phrase in specific situations; I see those suggestions as advocating more clear and precise language, not less. As opposed to overusing a “rationalist” term wherever it kind of applies.

  30. eqdw says:

    7. Autistic -> nerdy. Use autistic when referring to a psychiatric diagnosis or a complicated package of sensory and cognitive issues. Use “nerdy” when referring to people who are book-smart but lack social graces.

    I’m going to disagree on this one. In 2019, using ‘autism’ and various derivatives thereof functions as an extremely effective tribal shibboleth. Someone who self describes their friends as “nerdy” is very frequently a blue tribe person, whereas someone who self describes their friends as “autistic” is frequently a… it’s not red tribe but you would be directionally correct to call it that.

    Additionally, there’s another connotation that seems to have crept up recently. “Nerdy”, in 2019, seems to mean something akin to Science As Attire (I swear this article used to be called Science As Genre). Someone who is “nerdy” is probably very excited for the latest Star Wars movie, has a Pokemon-themed cell phone case, and knows that they’d be sorted into Gryffindor. But “Autistic”, in 2019, means something more akin to what I think of ‘nerd’ as having meant fifteen years ago: obsessive single-minded focus on obscure and arbitrary things to the point that it harms you socially. Someone who is “autistic” probably has an encyclopaedic knowledge of every supporting character in A Song Of Ice And Fire, and might have spent 6 hours last night compiling evidence of some conspiracy theory or whatever.

    • alwhite says:

      Yeah, don’t do that. Autism is a medical diagnosis and should not refer to people who don’t have that diagnosis. It’s a kind of disability shaming, and in poor taste. It falls in line with calling people “gay” as an insult, “lame” as an insult, “retarded” as an insult, etc… It could also be similar to calling yourself Native American when you are not Native American. The word autism has a definition and belongs to the people who fit that definition. Using it for people not in that definition is bad and you shouldn’t do it.

      • eqdw says:

        You have never seen my medical chart, so I recommend you take a step back and stop making assumptions about me unless you want me to make liberal use of that there report button

        • alwhite says:

          What assumption have I made about you? I said don’t use the word autism to talk about people who don’t fit the diagnosis. Your comment seems to be creating a new definition for autism that isn’t the diagnostic one and you seem to be arguing that it’s ok to use your definition. I’m disagreeing with that and saying don’t do that. Stick to the diagnostic definition and don’t call people outside that definition the label because it is insulting to various people groups.

          • eqdw says:

            “Seems to be creating”

            Have you been away from the internet for the past several years? The definition I’m using is much more prevalent than the clinical jargon definition.

            Meanwhile, I promise you that autistic people do not give one single care what you call them, and nobody other than you is confused about what the referent is when people use the term outside of an obviously medical context

            Finally, your oddly aggressive attempts at language policing is doing a great job of ignoring the constructive point I was trying to make, which is that the way these words are currently used is to refer to two very different groups, and saying “hey you should always use the second term unless you’re referring to this third group” is misleading and erases the existence of one of the original groups under discussion

          • Ozy Frantz says:

            I’m autistic and I care.

            Maybe watch where you’re throwing those generalizations?

          • albatross11 says:

            I think this is an inevitable part of the euphemism treadmill.

            a. Some inoffensive term has a technical meaning that maybe much known or understood in public, but little or no emotional impact.

            b. The term goes into more widespread use, often because the old commonly-used term has become an insult.

            c. The public gets hold of the new term and associates it with whatever unpleasant or low-status traits it’s associated with.

            d. People start using the term as an insult, starting with the more highbrow types and slowly slinking down until it’s a schoolyard taunt.

            See “moron,” “retarded,” “alcoholic,” “aspie,” “colored,” “hyper,” etc.

            People who meet the technical definition of being (say) mentally retarded don’t like being labeled something that’s also used as a schoolyard insult or a taunt. Either we keep the term and make them swallow the implied insult, or we take another step on the euphemism treadmill.

          • Enkidum says:

            I promise you that autistic people do not give one single care what you call them

            I promise you that you have absolutely no idea what you are talking about.

          • caryatis says:

            @eqdw

            Another autistic person who agrees with you.

          • MarcusAurelius says:

            I find it ironic that this use of ‘Autistic’ that eqdw is suggesting would probably be confusing to people with autism since we might assume you are actually referring to people with autism.

            But to address the discussion I would appreciate it if people did not perpetuate negative or misleading stereotypes that are not true for most people with autism. I feel like it also pushes the typical ‘boy autism’ that leads women to not being diagnosed or being told they are not ‘really autistic’.

            I have autism (if that wasn’t yet obvious from my comment) and my special interests are not arbitrary, they are very important to me. They are not single minded, I have many interests, some held for a long time and some for only a short time. It is the intensity that is characteristic of autism.
            My social life is not damaged from there interests, they are build upon them. Social issues stem from not having an inherent understanding of social signals and structures. Special interests often work as a bridge to help people with autism to learn how to socialize.
            Also it is a misconception that people with autism have some kind of savant photographic memory.

            However I would be fine with people who don’t have autism referring to positive non stereotyping traits they have as ‘autistic’, while not saying it makes them autistic. Since I think making the public image of autism more positive.

            Let’s be specific, don’t use autism just because you want to gatekeep other people from being a ‘real’ nerd.

          • brad says:

            that are not true for most people with autism

            Don’t most people with autism have co-morbid moderate to severe mental retardation?

          • Aapje says:

            @brad

            Not those with Asperger’s. This is estimated to be more common than classical autism.

      • caryatis says:

        >Autism is a medical diagnosis and should not refer to people who don’t have that diagnosis. It’s a kind of disability shaming, and in poor taste.

        I really disagree with this line of thinking. Autistic traits, like the traits of other mental illnesses, occur on a spectrum. It makes sense to say “Jim is more autistic than John,” even if neither of those people is autistic *enough* to get a diagnosis. Same thing with OCD, anxious, depressed, and other terms that can refer either to traits that are extreme and/or distressing enough to get a diagnosis, or traits that occur in non-diagnosed people. And using “autistic” or other terms casually doesn’t insult autistic people–in fact it’s a good thing, because it tends to normalize having weird traits.

        • cuke says:

          I agree with you that we do better to look at most personality traits — including things that get called “disorders” — as being on a spectrum.

          That spectrum nature of things creates ambiguity — but it also gives us more words to talk about our shared experience as people.

          Another example of this is how “borderline” gets used in various contexts. Some people have experienced incredible shame from being diagnosed “borderline” while other people have found the diagnosis to be hugely helpful and clarifying. At the same time, many people when they get “triggered” experience what we might call “borderline” states. So the word captures a common piece of human experience, it’s also used as a shaming word, and it’s a diagnosis that some people find useful and others abhor.

          I don’t think any linguistic policing is going to sort this ambiguity out for us. The best we can do is pay attention to context and our own intentions. Are we using these words as weapons to demean someone or are we using them to clarify something about our own experience?

        • thevoiceofthevoid says:

          I agree that autism is a spectrum and need not be completely limited to clinical diagnoses. However, I think calling anyone who likes solving logic puzzles “autistic” could cause similar problems as calling anyone who’s temporarily a bit sad “depressed”.

          • The Nybbler says:

            “Depressed” meant “sad” long before medical science took up the term. The same is not true for “autistic”; rather, it’s been taken up by some subcultures as meaning “nerdy” or sometimes “singleminded”.

      • AlexOfUrals says:

        How do you swear then, though? I’m not pushing any agenda or anything, and I do see (though not really share) the rationale behind not using terms which are correctly applied to some group of people as an insult. But I’m just genuinely curious what is the official PC position on this? Pretty much every swear word refers to some group of people normally. Even the good old “fuck you” used as an offense implies disrespect to anyone who takes a passive role in sex, which is roughly half of all the people. Is one expected to limit oneself to feces-related insults, or just call anyone they don’t like n*zi?

        • Aapje says:

          The Dutch often use diseases, although that can be offensive to people who (indirectly) suffer(ed) from that disease.

          You could use eradicated diseases…

          • Nornagest says:

            Trouble is there are only two of those, and one’s a disease of cattle.

          • AlexOfUrals says:

            Just any diseases or some particular ones? In Russian people will sometimes use STDs as a part of an insult, but that obviously stigmatizes promiscuous people and prostitutes, let alone anyone who actually have/had those diseases.

            And yeah, what Nornagest said. I wonder if using diseases would sound old-fashioned and Victorian though?

            “This car looks so lame, I wonder what a retard designed it” -> “This car looks so smallpox, I wonder what a plaguehead designed it”

            Nope apparently it doesn’t.

          • Aapje says:

            @AlexOfUrals

            These are popular:

            Cancer
            Tuberculosis
            Typhoid
            Cholera
            Pleuritis
            Heart attack/stroke

            Note that some of the Dutch insults have become what we call petrified, where an old fashioned term is used that is no longer used outside of a specific expression. For example, ‘takkewijf’ is made up of takke and wijf. The latter is a rude term for a woman (in English, you have the related ‘wife,’ although it is not rude). Takke refers to attaque which is French for attack.

            However, I suspect that the percentage of Dutch people who are aware of the meaning of the insult is below 1%.

            “This car looks so lame, I wonder what a retard designed it” -> “This car looks so smallpox, I wonder what a plaguehead designed it”

            Nope apparently it doesn’t.

            In this context, Dutch people would normally use an intelligence-based insult as well. A disease-based insult is more likely if the car is broken or user unfriendly.

            If they would use a disease-based insult, they might use the slang word for Tuberculosis, like so: teringauto. So ‘tering car.’ Tering is quite suitable to swear with since you can play with the pronunciation a lot to communicate a lot of nuance.

            Another possibility is kankerauto, so ‘cancer car.’

            English doesn’t have swear-friendly slang words for diseases, but this seems to be because they don’t swear like this in the first place.

        • Gerry Quinn says:

          “English doesn’t have swear-friendly slang words for diseases, but this seems to be because they don’t swear like this in the first place.”

          “Pox” is old-fashioned but still entirely usable, in forms like “pox-ridden”. In Ireland “poxy” is a very mild form, less common than it was, but you will still hear it – but mostly referring to a low quality item or service rather than a person.

    • FormerRanger says:

      “On the spectrum” and “Asbergery” are things I have heard often from people to whom they might refer.

  31. Garrett says:

    Re: male/female.

    I’ve noticed that African Americans in the US seem to use male/female disproportionately more than I would expect vs. men/women. This is outside of the context I would assume would be gender-study aware.

    Can anybody shed any light on why this seems to happen? And, should we expect it to be a phenomenon which will spread more widely? If so, might the SSC crowd finally be ahead of the popularity curve?

    • Jiro says:

      Off the cuff hypothesis: Social justice likes to use “male” a lot in order to dehumanize men . Black people are considered allies by social justice so you will see a lot of media targeted to black people that use social justice terminology. Then they’ll pick it up from what they’re exposed to.

      • cuke says:

        From working in various black communities in the South twenty plus years ago, I heard this usage a lot and I don’t think it has anything directly to do with social justice in capital letters, and way pre-dates it.

        I had it explained to me by some insiders that there’s a negative history to the word “women” among many older generation black communities because white people would refer to white women as “ladies” but black women as “women” to deny them the dignity that “lady” conveyed at the time. To get around this painful history, some black communities came to use the words “males” and “females” because the words had less baggage and came across as more neutral.

        Working in these black communities is the only place I’ve seen the words “males” and “females” used equally by men and women. In predominantly white settings, it’s almost entirely men who use these words.

        • Plumber says:

          @cuke,
          That’s interesting, I grew up in a mostly black neighborhood and currently have a few black co-workers and I seldom hear them say “males” or “females”, instead they almost always use the words “men” or “ladies” (seldom “women”), though one guy will sometimes use “little momma” when flirting with the UPS lady, but while they mostly have relatives in the South and sometimes visit there they’re Californians.

        • JulieK says:

          I think it was common for whites to refer disrespectfully to black adults as “boys” or “girls” (depending on gender).
          For example, in _To Kill a Mockingbird_ Mrs. Robinson (wife of the black man accused of raping a white girl) works as a cook. Her employer is one of the “good guys” – he defends her from harassment from the girl’s father – but he also casually refers to her as a “girl.”
          Perhaps “males” or “females” was adopted in response to that.

        • carvenvisage says:

          Who could *possibly* have guessed that using the more objective, precise, or clinical term isn’t always inspired by bigotry.

    • bullseye says:

      There’s a lot of Social Justice on my Facebook and Twitter feeds, and I’ve never seen them use the word “male”. I have seen them complaining about the word “female”, saying it should be avoided even as an adjective. (Based on hate for the rationalist community, I guess?) That struck me as very odd because I too think of calling women females as an African-American thing.

      • Nornagest says:

        Based on hate for the rationalist community, I guess?

        They’ve probably never even heard of the rationalist community. Using “male” and “female” where you should use “guys” or “women” or “girls” isn’t a specifically rationalist thing, it’s a nerdy thing — trying to sound erudite, basically — that’s common in the rationalist community because it’s full of oblivious turbo-nerds.

        • bullseye says:

          That’s a good point. They’ve probably just noticed that people who talk in certain ways are more likely to say offensive things.

        • Don P. says:

          In particular, it makes you sound like you’re talking about livestock.

      • Sinclair says:

        No it’s because in African American Vernacular English, “female” is sometimes (but not always?) used as a milder form of “bitch” or “thot.” It has kinda sexualized connotations. In AAVE the polite term for a woman is “lady.”

        Epistemic status: I am of medium probability full of shit. This is from ten minutes of googling. I’m not black and I’m not a linguist.

  32. 57lollipops says:

    Here are some other suggestions based on reading a lot of internet comments over the years:

    Quite -> Very. For some reason, overly precocious teenagers love to overuse “quite”. Maybe they think it makes them sound more British and therefore more erudite? Of course, Mark Twain would go even further and lose the “very”:

    Substitute ‘damn’ every time you’re inclined to write ‘very’; your editor will delete it and the writing will be just as it should be.

    Humorous -> Funny. There’s almost no situation where “funny” doesn’t sound more natural.

    • Tarpitz says:

      People use “quite” to mean “very”? From my perspective as a Brit, there is a common contemporary usage that means approximately
      “somewhat”, and a formerly common archaic usage that means approximately “utterly”. Is what you’re talking about a derivative of the latter? Are the overly precocious teenagers in question reading a lot of Austen?

      • bullseye says:

        I’m American, and I had no idea it didn’t still mean “utterly”.

      • 57lollipops says:

        There’s usually not enough context to know whether they mean “very” or “somewhat” or “utterly.”

        My friend Jeff is quite tall.

        You’d need to know more about Jeff to be able to determine if they’re using “quite” appropriately. And then you have constructions like “it was quite fun,” where the description in question is so subjective it’s all but impossible to determine which meaning they intend without stopping them and asking, which is rarely possible and never practical.

        From what I’ve observed, intelligent middle schoolers use “quite” to sound more adult, but intelligent adults who overuse the word end up sounding like middle schoolers.

        • Tarpitz says:

          Interesting. If Jeff’s friend were from this side of the Atlantic, you’d be relying on inflection and other such cues to determine whether they were speaking straightforwardly (and perhaps Jeff was 5’11-6’2) or indulging in ironic understatement (and Jeff was a freakin’ giant). Regardless, you could substitute in “fairly” or “pretty” and alter the meaning not at all.

          • Deiseach says:

            I always trip up between the British usage of “rather” and “quite”, probably because Irish-English would be akin to American in this sense of retaining the archaic usage* of “very, utterly” rather than the revised English usage of “somewhat” and then “rather” taking over as “very” – as in the elocution phrase we were taught as schoolgirls: “Father’s car is a Jaguar and Pa drives rather fast” (to get us to pronounce an orotund “aw” sound instead of our native “aah” which would render that as Faader’s caar izza Jaguaar an’ Paa drives raader fast, with “fast” being the only short “a” in the whole thing).

            *The same way as country/working class people eat their dinner, not their lunch, in the middle of the day and it is the main/largest meal which was good 18th century usage for dukes as well as dustmen, but now the meals have diverged into a light (or relatively so) lunch at noontime(ish) and a more substantial dinner in the evening.

          • Gerry Quinn says:

            Irish too: I’m with Tarpitz on ‘quite’ having the two meanings of ‘somewhat’ and ‘very’.

            For me ‘rather’ technically means ‘somewhat’, but you don’t use it unless you are saying ‘very’.

            I might actually say ‘a bit’ when I mean ‘somewhat’.

      • cuke says:

        My understanding is this is a Brit/American distinction, that we Americans still mean “a lot” when we say quite and the approximately or somewhat sense isn’t common here at all.

      • Faza (TCM) says:

        People use “quite” to mean “very”? From my perspective as a Brit, there is a common contemporary usage that means approximately “somewhat”, and a formerly common archaic usage that means approximately “utterly”.

        I find this to be yet another fascinating example of two nations separated by a common language. Thinking about my own understanding of the word, I came up with the following test:

        Alice is tall. Bob is quite tall. Who is taller, Alice or Bob?

        To me it sounds like Alice is taller. The unqualified use of the word “tall” indicates to me that she is taller than most people.

        Bob, on the other hand, is definitely not short and may even be taller than average, but the use of “quite” in this context suggests that we’d still expect him to be shorter than many people we might know.

        • thevoiceofthevoid says:

          As a (probably typical?) American, it sounds like Bob is taller. Still not the tallest person I know, but since “quite” is an intensifier, Bob’s taller than someone who’s only described with a non-intensified “tall”.

          For completeness, all the common qualifiers I can think of, from shortest to tallest:
          Not tall at all
          Not tall
          Not very tall
          A bit tall
          Pretty tall
          Tall
          Quite tall
          Very tall
          Extremely tall
          Freakishly tall
          Colossally tall
          Monumentally tall
          Insanely tall
          Mind-blowingly tall
          Incomprehensibly tall
          [censored] tall
          [censored] [censored] tall
          Tall as [censored]
          Taaaaaaaalllllllllllllllllllllllllll

          this is why I shouldn’t post comments past midnight

        • Gerry Quinn says:

          Leaving sex out of it, I can’t really guess who is taller – I think I would plump for Alice but it’s marginal.

    • Rachael says:

      As another Brit, I agree with Tarpitz. Also, I thought the intensifying meaning of “quite” was more common in American English and the weakening meaning more common in British English.

      • thisheavenlyconjugation says:

        Agreed, the intensifying meaning sounds very archaic to me (using the weakening meaning ironically and therefore with the opposite connotation is different and more common).

    • Spookykou says:

      As an American, I’ve mostly thought of it as a way to express agreement.

  33. alwhite says:

    On 2 and 3:

    Humans seems to be really useful as a gender neutral term. This one might be changing to be more acceptable, but in line with the gender neutrality, people who avoid the gender fluidity concept may be more repulsed by humans than their counterparts.

    I find the words men/women to be evocative and they elicit a weird kind of vulnerability when using them. The terms male/female being the more scientific term doesn’t do the same thing. I’m sure there’s a lot of me specific in this but I also think this could be part of the reason why this one happens.

    • Tarpitz says:

      What is wrong with “people” as a gender neutral term, outside of technical discussions of putative non-human personhood or human non-personhood? For everyday purposes, the terms are coextensive, and one of them is a lot more conversationally normal than the other.

      • Nick says:

        I’m going to put in a vote for “folks.”

        • Robert Jones says:

          Surely “folk” is sufficient? E.g. “I’ve been wondering why folk celebrate holidays.”

        • bullseye says:

          Maybe I’m inside a bubble here, but do regular people actually say “folks”? Obama and Trump say it a lot, and it comes across as fake to me.

          • Nornagest says:

            It’s dialectical; shows up a lot in AAVE and Southern English, but also elsewhere. Both Obama and Trump are probably affecting it, but for different audiences.

          • Nick says:

            I’ve heard it in my area, and I grew up 30 miles from the Canadian border, for heaven’s sake.

          • HeelBearCub says:

            Ok, folks, calm down. No need to get your knickers in a twist about a perfectly valid question even if he is barking up the wrong tree.

          • cuke says:

            Outside of politics, could this be a generational thing? I’m mid-50s and “folks” sounds really natural to me and I use it all the time. I feel like it’s what my teachers and professors called us in the 1970s and 80s. They would have been raised in the 40s and 50s.

          • Plumber says:

            Echoing what @cuke said, I’m 51 and have spent most of my life in and near Oakland, California and “folks” sounds like regular speech to me.

            “Folk” without the ‘s’ sounds academic though.

          • ECD says:

            I’m 32 and use it, but only when addressing a group of men and women, or people whose gender I don’t know (e.g. large email list). I haven’t thought about it much but I guess I mostly use it because the other standard all-gender referent for my workplace is ALCON, which is used almost entirely by veterans, which I am not.

          • Tarpitz says:

            Interesting, Plumber – to me (as a Brit) “folk” is working class, probably rural and probably older.

    • caryatis says:

      >I find the words men/women to be evocative and they elicit a weird kind of vulnerability when using them.

      What do you mean by this?

  34. HeelBearCub says:

    Ways to seem like an evil-robot …

    Write a guide called “Style guide: Not sounding like an evil robot”

    Seriously, I’m not sure if this counter-signaling or what, but if the goal is to model ways to write in an appealing manner to a broad audience, it seems to fall flat in that it doesn’t successful synthesize it’s own message. Thus it further cements the idea that one shouldn’t actually attempt to write in the manner recommended herein.

    Maybe that’s necessary to reach the intended audience, but it seems self-defeating.

    • pansnarrans says:

      I disagree. It works because it’s funny, in a kind of deliberate “how to hit this very low bar” sort of way.

      A self-defeating title would be something more like “How to make normies understand you”.

      I think all of the advice here listed is good. Talking about “status” especially sends strong signals that you may prefer not to send. And shout out to the “males/females” versus “men/women” thing as an example of people routinely trying to sound sophisticated and instead sounding just cold.

  35. Jiro says:

    2. Humans -> people. This will instantly make you sound 20% less like an evil robot. Use “humans” only when specifically contrasting with another animal:

    2a: Using the term “animal” to refer to humans. You know or should know that most people do not refer to humans as “animals” and that even if you use a definition that includes humans, humans are not central examples of animals.

    • thevoiceofthevoid says:

      I think it’s perfectly fine in a scientific context. Most people* know that humans are animals in the biological sense, even if that fact wouldn’t be salient in everyday conversation.

      *in my social circles

    • carvenvisage says:

      beep boop good correction comrade

    • Don P. says:

      I wrote and deleted a brief comment to this effect, so let me try it again it here, even briefer; the nerdosphere in general drastically underestimates how unwelcome others, in general, find the stance that people are usefully understood as highly intelligent irrational animals. (Yes, even people who in general accept the relevant facts. The instinct that causes people to refuse to accept evolution — I am not a monkey! — is not confined to the ignorant.) And the nerd/rationalsphere loooooooves this stance.

  36. OriginalSeeing says:

    I’ve often used the character name EvilRobot in online games. I don’t think doing the opposite of your suggestions will really help my assumed evil robot persona. What gives?

  37. Walliserops says:

    If we were to have a cloud-to-butt/xkcd substitutions style autoreplacer for these words, what would be your suggestions?

    ‘IQ’ should remap to ‘horsepower’ for sure, but I’m stumped for the rest.

    • Lambert says:

      I have some, but something is getting snagged in the spam filter.
      Probably that political philosophy that’s old, except that it’s new.
      You know the one.

      • crh says:

        There’s a (non-exhaustive) list of banned words in the comments policy, and that is indeed one of them.

      • cactus head says:

        I’d like to see that one taken out of the comments filter. It’s not 2014 and we don’t have nydwracu and multiheaded trading barbs up and down the comments section anymore. Same thing with reproductively viable worker ants.

      • Lambert says:

        Humans -> Evil Robots
        Rational -> Groovy
        Utility -> Pizazz
        en aarrgh ecks -> Sith
        Status -> Height

  38. blacktrance says:

    Relatedly, Blue Tribe -> Democrats/liberals/leftists, Red Tribe -> Republicans/conservatives/rightists, almost always. When I coined those terms I was trying to explain how Democrats/Republicans were the tip of an iceberg of related traits, but now that the message has sunk in I think it’s reasonable to call that iceberg by the name everyone else uses.

    If you mean Democrats, say “Democrats”, but “Blue Tribe” is not synonymous and this is an inaccurate substitution if you’re talking about the actual tribes. There are plenty of Democrats who aren’t Blue Tribe, and some Blue Tribers who aren’t Democrats (e.g. Mitt Romney Republicans). Party membership or support is only one node, and we want to be able to talk about the cluster, including members for whom that node has an atypical value.

    • acymetric says:

      Uh, if you have Mitt Romney in the Blue Tribe I think you are using a very different tribal map than most others.

      • blacktrance says:

        From the original post:

        My formative years were spent at a university which, if it was similar to other elite universities, had a faculty and a student body that skewed about 90-10 liberal to conservative – and we can bet that, like LW, even those few token conservatives are Mitt Romney types rather than God-n’-guns types.

      • John Schilling says:

        My tribal map has Romney in a very blue-ish purple zone, and I don’t think I am unique in that. In any event, Mitt Romney as Blue Tribe and Jimmy Carter(*) as Red Tribe are pretty much the canonical examples around here for tribal identity not always matching political affiliation and for why we use that language instead of just “Republican” and “Democrat”. I’m surprised you haven’t seen it before.

        * And Bill Clinton’s Governor-of-Arkansas persona but that isn’t as good an example because there’s more suspicion that it’s just an act.

    • BBA says:

      At a time when the figure seen as personifying the Red Tribe is a lifelong New Yorker who probably has never owned a pickup truck or a gun, doesn’t attend church (and the church he doesn’t attend isn’t fire-and-brimstone evangelical, it’s milquetoast mainline Marble Collegiate, once home to Norman Vincent Peale of “Positive Thinking” fame), and loves showtunes and Elton John, I don’t think the “tribes” mean much of anything anymore, if they ever did.

    • Nornagest says:

      I’ll go to bat for “Red Tribe”/”Blue Tribe” as cultural labels, but I gotta admit that almost all the people I see using those phrases here just mean “Democrats” and “Republicans”. If we can’t put a stop to that somehow, it might be better to put the terms to bed.

      • HeelBearCub says:

        Problem is that the Blue tribe Red tribe map is still trying to use just the two party colors to color in the cultural territory. Then makes a sudden swerve to talk about some other, uncolored part of the map.

        The reality being that there are so many cultures in the actual territory that the tribe description really suffers by trying to map everything onto a binary.

        Is the Hispanic dishwasher living in LA who likes Budweiser, catcalling women and Latin hip-hop, red or blue tribe? What about the white subrurban mom who takes her kids to youth soccer, ballet and the local Evangelical Sunday school?

        This map is very much not the territory. There is an “urbane” tribe. There is also a “hipster” tribe. Combine those two (which are not really the same) and you get something that sort of looks like the original blue tribe. But everyone who doesn’t fit into that group isn’t red tribe by a long stretch.

        • Nornagest says:

          It’s an incomplete map for sure. But the central insight, as I see it, is that the two parties aren’t drawing from a homogeneous population of “Americans”, or even a homogeneous population plus some number of minority populations, but that their centers of gravity lie in very different parts of the white American cultural space. Then we can use that insight to talk about differences in how they predominantly see the world. That’s valuable! But it carries a risk of rounding everyone in the GOP off to Red, or everyone in the Dems to Blue, and that is absolutely not accurate.

          (FWIW, I’d say that the Hispanic dishwasher is outside the schema, and that the soccer mom is Red but not a very deep Red.)

        • blacktrance says:

          Red and Blue Tribe aren’t supposed to be all-encompassing. For instance, the original post also mentions the Grey Tribe. The Hispanic dishwasher is also from some other cultural cluster. As for the white suburban mom, she’s somewhere in the middle between Blue and Red, probably leaning towards Red.

  39. Douglas Knight says:

    Do any of these terms ever promote precise discussion?
    The main one I see that using numbers for probability is valuable. (And Scott conceded a use for N℞. And I missed Blue/Red Tribe on my first pass.)

    Many of them seem to be a reference sophisticated versions of common ideas. If they referred to different ideas (eg, Democrat vs Blue Tribe), it would be valuable to use the new term to distinguish. But if they refer to the right way of thinking about common ideas, there is no distinction to be made and no need for new words. If using the terms produced common knowledge that the speaker is using the specific ideas, that would be valuable. But it doesn’t. It merely signals that the speaker is aware of the term. The speaker may not even be aware of the precise ideas, but just that it is associated with a certain group.

    (Scott did a much better job of choosing terms to avoid that the commenters.)

  40. Huh. Evil robot here, apparently! The “male/female” thing sprinkles some additional explanation on something from years ago to me that did not make much sense to me at the time. Although other people have pointed out “male/female” were weird word choices and it probably exacerbated an otherwise mostly harmless situation, no one’s ever gone so far as to tell me those words are better considered off-limits altogether! Thanks.

    If anyone’s curious (skip this if you’re not, I’m just elaborating as not to leave anyone hanging):

    I was making a comment in a tiny reddit subcommunity on my perception of my fellow ladies (implicitly including myself) being on average better at empathy and men being on average better at logic/analytics, and “women” seemed awkward (to me the word has an ‘is an adult’ connotation and I wanted to include all ages), as did “ladies” (same problem), and “girls” was right out (opposite problem), so I used “females” (and “males”).

    Using that term was the wrong answer, largely because people assumed I was excluding myself – which makes sense if they’re used to parsing it as somehow dehumanising. (The context absolutely did not help – I was trying to figure out why this community that narratively presented itself as uninterested in empathetic worldviews didn’t have a lot of other women in it, and of course didn’t bother to say why I was in the community, myself.)

    At the time, the dog-pile backlash hit me where it hurt, as though I wasn’t allowed to be proud of being empathetic, since it was obviously inferior to analytics, how dare I imply anyone might be naturally better at empathy than logic, what kind of subhuman do I think women are? Just to clarify, my hurt was completely independent of whether my observation was right or not (I’m not convinced I was right with that dichotomy of averages! And I would have welcomed statistical corrections at the time) – the unexpected kick in the teeth was just that people openly assumed I was somehow putting women down with a trait that I actually consider immeasurably valuable.

    But yeah. Word choice matters. “female” did not help. There’s a good chance the whole thing would still have happened if I’d used the word “women”; generally most of the fuck-up happened because a much larger community somehow happened upon the thread, and if I’d known about that beforehand, I would have tailored the comment that attracted ire to a wider audience, but I suppose that’s how the internet rolls.

    Regret turtles all the way down! 🙂

    • pansnarrans says:

      In this case, I think you’re right that the backlash would have happened anyway because you inadvertently hit someone’s trigger. Some people who oppose sexism do seem to have nevertheless internalised the idea that traditionally male roles are superior to traditionally female ones, and therefore leap to the idea that a woman is being presented as only “good enough” to raise children, but not that her husband is being presented as only “good enough” to go to work. And so on.

      On language: I totally get that people use “males/females” because they want to include kids in what they’re discussing, but I think generally from context “men/women” will communicate that anyway, and as they sound more natural they’re better. If I said “men are more interested in watching sport” I don’t think anyone would infer that I’m excluding boys. If I said “males are more interested in watching sport” I’d sound like I was making field notes.

      • On language: I totally get that people use “males/females” because they want to include kids in what they’re discussing, but I think generally from context “men/women” will communicate that anyway, and as they sound more natural they’re better. If I said “men are more interested in watching sport” I don’t think anyone would infer that I’m excluding boys. If I said “males are more interested in watching sport” I’d sound like I was making field notes.

        “Here we see a male human in his natural habitat…” 🙂

        (You get no disagreement from me on any of your points, just to be clear.)

    • caryatis says:

      “Women and girls” includes all ages.

      • thevoiceofthevoid says:

        But is a bit awkward and unwieldy. Reminds me of the god-awful “he or she” construction.

  41. eigenmoon says:

    Following this guide might be rational if you place high utility on some signaling goal such as attracting as many females (or males) as possible. But if you want to filter out lower-IQ humans that you didn’t want to communicate with anyway, it might be optimal to reverse the guide.

    • Akrasian says:

      Pursuing such a utility function also has a high probability of optimising for an especially autistic audience, however. This is a potential failure mode if it lowers ones status and/or signals allegiance to the grey tribe when we want to also reach out to red tribe and blue tribe humans.

  42. Radu Floricica says:

    I’d have liked a bit more context. From what I can gather, context would be that Scott wants the subreddit (and possibly other public forums… like here?) to sound less evil robot-y and thus be more appealing to outsiders / new people. In this regard, the advice contained is very useful.

    I do feel like somebody should mention that a jargon get created to differentiate a subculture, facilitate communication in that subculture and help people recognize each other in other places. So it’s a tool, to be used or not depending on context. Of course, when you have a hammer etc.

    I also notice a difference between Scott’s examples and others in the thread. “Failure mode”, “priors” are concepts that are not immediately translated. When I’d expect others to understand, I’d use them.
    Humans vs people on the other hand is plain evil robot talk – I can’t find any justification other than possibly identification by jargon, and in this particular case even that is extremely fuzzy and not very flattering. So personally I intend to take the advice to heart. I also sometimes edit my comments for clarity – I just replaced “and not particularly flattering” in the last sentence with “very”. It’s probably a good idea to keep doing it, and more often.

    • albatross11 says:

      People who’ve developed a lot of their writing/communications skills in an intellectual bubble will often do this thing where they try to express themselves in the big wide world, and fall over their feet or get massively misunderstood, because they’re using tropes and terms and assumptions that are misunderstood or just baffling to people not raised in their bubble. It’s a variation of the phenomenon by which a very smart person brought up in a rough lower-class environment may have a hard time expressing himself in a way that conveys his intelligence to higher-class people, because his accent, grammar, dialect, conversational tropes, and background assumptions are all wrong for them.

      In both cases, this kind of advice can be useful. “Avoid profanity, and talk more like the books you read than like you do in normal conversation. Pronounce words like TV announcers pronounce them. Avoid ‘ain’t’ and ‘y’all.’ Don’t assume your listeners know anything about hunting, fishing, country music, or working on cars.”

    • Dacyn says:

      “Failure mode” just means “way of failing”. I don’t think it needs to be translated at all, the phrase is intuitively clear (to me at least: I understood it immediately the first time I saw it)

      • Radu Floricica says:

        I don’t know, failure mode to me suggests discrete categories, or attractors in the space of possible ways to fail. A way to fail.. is just one of an infinity. Could be just me though.

      • Garrett says:

        As with everything else, there’s a formal definition which then gets picked up by people nearby organically without necessarily knowing the formal definition.

        • Dacyn says:

          That definition doesn’t appear to be any different from the informal definition, other than the suggestion that the term is occuring in the context of systems analysis.

    • DinoNerd says:

      *chuckle* I use “humans” rather than “people” in contexts where there might be other types of “people”. Or sometimes when I wish to imply that it would be nice to have (or be) another type of sentient being.

  43. DNM says:

    This was interesting and somewhat tough to read through. I think in the past 10 or so years I have learned to translate in the other direction (e.g. “optimal” rather than “best”, “humans” instead of “people”) in order to have my voice accepted and seriously considered in a specific set of communities.

  44. Houshalter says:

    Communities form their own terminology. A huge part of the value of lesswrong was giving names to concepts that were previously nameless. Or only had vague imprecise words. If you start speaking like a filthy normie you might start thinking like one too.

    • thevoiceofthevoid says:

      I’m confused, the first half of your comment seems sincere but the last sentence reads to me as sarcasm. Are you being serious?

      • carvenvisage says:

        99%* sure they’re using going-over-the-top hyperbole as good-humoured cover for an actual conviction, likely less strongly held than presented.

        *non robot number, don’t take as actual number)

  45. TheTurtleMoves says:

    This sort blog and its comments section are the only places where people use the word “reactionary” at all correctly. Pay attention to regular people and they will use “reactionary” as “reacting to something.”
    So a totally progressive thing could be construed as “reactionary.”
    EXAMPLE: “She only got involved in gun control marches after her nephew survived a shooting. It was completely reactionary.”

    My point is I’d take “reactionary” off the good list above. Just make it “far right” or “right wing.” Regular people don’t understand the world “reactionary” at all.

    • thevoiceofthevoid says:

      What word would you use to describe a new, liberal policy instituted entirely as a reaction to some event? E.g., how would you rewrite your example without using “reactionary” or introducing awkward verbosity?

      • zakamutt says:

        You could use ‘as a reaction to’ or ‘as a response to’. Or if you don’t like it, go with the good old ‘knee-jerk reaction’.

    • Don_Flamingo says:

      I thought it was proactive vs reactive, not proactive vs reactionary?

  46. BBA says:

    tl;dr Sneer Club is right about a whole lot more than any of us would like to admit.

  47. zzzzort says:

    Can I nominate ‘fallacy’ -> ‘mistake’, partially in the spirit of the last post?

    There was a story on twitter of someone growing up who thought Inshallah (~god-willing) just meant ‘no’ or ‘never’, based on their parents responding to questions like ‘Can we have pizza for dinner?’. Similarly, the rationalist in-group definition of fallacy seems like ‘reason other people are wrong’.

  48. Jaskologist says:

    I gained a lot of happiness, goodness from this post. But I have a further question for the all the people here: what is the best word to use for “paperclip?”

    Asking for a friend.

  49. Akrasian says:

    I was thinking the same thing about IQ just yesterday. Was just going to make an open thread post about it.
    Emphasising IQ is especially egregious when talking about yourself personally. You know your own intelligence, relative skill in different areas and life outcomes based on your own experience. Using a test that is intended as a proxy for these things is unnecessary when you can just explain your specific skills.

  50. Steve Sailer says:

    Rosalind Arden once asked me to stop using “schizophrenic” to mean two-faced, hypocritical, contradictory, and multiple personality. I initially scoffed but eventually I saw her point: Schizophrenics and their loved ones have enough problems without other people using the term to mean something that is not at all schizophrenia.

  51. Ron says:

    On a related topic, iv’e been wondering why politicians frequently DO NOT apply what Scott called the “weirdness points” principle.
    You’d think politicians will optimize for convincing as many people as much as possible, and that applying the principle can help.
    My go-to example is far-left intentionally mis-using gender pronouns (in Israel).
    Took me way too long to realize that using niche language is actually useful for in-group signaling. The language reveals who the intended audience is, even when a politician makes a “public” declaration.

    • Don_Flamingo says:

      Could you elaborate on that?
      I’d be curious what those far-left Israelis are doing exactly and what point they’re trying to make or what they want to signal.
      Cause I don’t really know how to model Israeli discourse at all.

      • Ron says:

        Sure!
        So Hebrew as a language applies gender to almost anything, and is also somewhat (a lot?) Male chauvinistic. For example, when saying “you all”, gender must be assumed, and the default is Male, “Atem”. The feminist take is to randomly use either Male or Female gender as default. Doing so signals belief in human equality, but is very rare in everyday speech, and so is at best awkward, at worst insulting. In the context of a political message, it’s a great way to loose “weirdness points”.

  52. Some Troll's Serious Discussion Alt says:

    But what if I am an evil robot and want to avoid confusing people?

    • Protagoras says:

      I would think in that case you should reconsider; confusing people is very likely to help you advance your evil agenda.

      • Dacyn says:

        Well, sometimes scammers try hard to signal that they are scammers at the beginning of an interaction, so that they don’t waste resources trying to scam people who won’t be fooled. Maybe STSDA should adopt the reverse of Scott’s advice, so that he knows the few who remain will remain loyal to him as he completes his evil plan?

  53. holomanga says:

    In order to countersignal, I include the phrase “beep boop, terminate” in all my comments.

    Beep boop. Terminate.

  54. Agree with all of these except the last one. “Respect” is often used for people who are low-status but subjects of public sympathy. People will often say things like “the community has a lot of respect for Bob, the local garbage man, who works very hard to feed his family;” they won’t say “Bob, the local garbage man, is high-status within our community,” unless he recently won the lottery. Using “status” serves to cut through the defensive BS people use when status hierarchies they’d rather not think about are critiqued.

    • thevoiceofthevoid says:

      Conversely, the Kardashians are high-status but generally not respected.

  55. Yosarian2 says:

    On a side note, while it’s totally true that it’s bad to have too many weirdness points in your writing, it’s also bad to have too few. If you don’t have enough weirdness points people will dismiss what you’re saying as common sense or conventional wisdom, wheras if you have a few carefully picked ones it will increase interest.

    Example:

    Boring blog post: Americans should vote because it’s a civic duty

    Interesting blog post: Utilitarians should vote because spending 30 minutes voting is a great utilitarian investment that creates much more utility then you might otherwise be doing.

    Note that these two could easily be the same basic blog post, but people are going to want to read the second because the first one isn’t novel enough to be interesting.

    • Ron says:

      In your example, I agree the second is more attracting, but note that the fundamental claim is also changed, not just the language. Utility becomes the object-level topic. Scott permitted the use of utility when discussing utilitarian philosophy, which seems to apply, but more generally, if the claim is changed then we are not discussing language anymore.

      More generally, I agree with you, in the sense that using our niche language sort of adds “here you will get the niche perspective about: “ to your title. But the cost is sounding like an evil robot (hopefully only to outsiders), so tradeoff wisely.

  56. March says:

    There’s a subset of people who use the combination ‘men and females’, usually while talking about women in less-than-generous terms. (“Men are loyal and know what real friendship is; females are hypergamous.”)

    While that matches my general impression that people use ‘males’ and ‘females’ to create distance between yourself and that group (which is also why ‘humans’ sounds so weird unless you’re taking the scientific view – if you’re talking about humans, what do you consider yourself?), it does end up sounding very Ferengi.

  57. Anonymous` says:

    Sure, if you renounce the community that actually supports you, maybe the popular people won’t hate you quite as much.

    That was about 25% joking, but this isn’t: the concepts of “utility”, “status”, and “probability” are very useful; even though there are similar general-use words, uses of these specific ones are strong signals that the writer is at least a little bit educated (in the true sense, not the modern sense of that word). Most writers are not, and most writers who are not don’t have anything useful or even coherent to say in the first place.

  58. Thegnskald says:

    I think part of the problem here is that I think the people who use these words, use them because they hate to be misunderstood. They want to be very precise in terms of what they say, so that the other party receives exactly the information they intend for them to receive.

    It’s annoying to normal people, especially since asking for clarifications on things they think are obvious often comes off as “correcting” rather than “clarifying”, because other people don’t have any other mental model in which to interpret asking whether a different word might have been what they meant (there are better and worse ways to phrase this, but meanwhile, the people asking for clarification don’t really have a mental model of this coming off as correcting rather than clarifying, and generally don’t know there is even a phrasing that might specify the difference – which would feel even more “evil robot”).

    The alternative isn’t “Use these less specific words”, it’s “Live in a state of constant anxiety over whether or not people are actually understanding you or are understanding a slightly different version of what you said that has an entirely different meaning”. And it is anxiety; you’ve attracted a crowd of people whose most common shared ethos is “Intellectual Honesty”, and they take it to a fault. This both directly contributes – are you being intellectually honest if you lead someone to believe a lie about what you believe? – and indirectly contributes through a process of “Routinely offending people who don’t care how carefully you phrase things and will interpret things you say through a bad-faith lens, but not having the mental model to parse the idea of someone who doesn’t take Intellectual Honesty seriously, so just continually assuming you are misunderstood through a fault in your communication specificity”.

    “Bad Robot” means “Weirdly Precise and/or Logical”. Bluntly, it is an impression you can only really get if the concepts words and phrases refer to have less subtle shades of meaning that vary from word to word. These are the connotative associations with words, everyone has them to some extent. I once had a speech professor tell the class “adamantine” meant something more like “stubborn” to many people than “strong”, and that we should be careful how we use words to correctly convey not just our meaning but our tone. While I think he was behind on the times, he was basically right. Synonyms are never equivalent. Most people, however, only carry very rough shades of meaning; a sense of negativity or positivity about a word or phrase, maybe a hint of emotion. The people here generally have much more complex shades of meaning. The way to convey this to the people here isn’t “Don’t sound like a bad robot”, it’s “Other people just don’t attach the same significance to words as you do, and your weird precision makes them feel stupid and/or like they’re missing the 90% of the conversation they are missing”.

    This is not true of people who merely have different connotations, hence the high number of arguments here that drift into the meta level, or the meta-meta level, as people try to figure out, through a sometimes hostile-looking-from-the-outside process, what the hell they actually disagree about. Honestly I think the thing other people would find most offputting and evil-robotish of the people here is their ability to angrily and vitriolically disagree, then go back to having a pleasant conversation in another thread, both at the same time.

    • Nornagest says:

      If this was about precision, we’d expect to hear “good” or “smart” roughly as often as we hear “rational” or “high-IQ”, with each one being used where its connotations were more appropriate. We don’t. That suggests to me that we’re not dealing with a simple need for precision, but rather with an indiscriminate preference for one set of terms (the clunky, faux-academic ones) over the other.

      I don’t wanna Bulverize that preference too hard, but it’s fair to say that it comes off poorly to outsiders.

      • Thegnskald says:

        Not necessarily; common words, by virtue of being common and used frequently by many people in many situations, may develop a neutral connotation, or lose connotation entirely. Given a more precise synonym, somebody with an inclination toward precision will tend towards the version that is more specific. Thus we may be able to expect common-usage words and phrases to be less common here, because they’d generally only be used as a fallback in the case that a more specific word or phrase can’t be used instead. (And we’d expect, when they do get used, an attempt to clarify what is meant.)

        • Nornagest says:

          Fine, then substitute some synonyms that do have clearly distinguished connotations. “Cunning” is about as uncommon in standard American English as “high-IQ” is, for example, but I rarely hear anyone called that around here.

          • Thegnskald says:

            Cunning doesn’t mean smart, though. It translates to most people as something like “Skill/talent at manipulating people”. Clever, likewise, translates as something like “Skill/talent at manipulating things other than people”.

            We have reasons for using the words we do. A significant percentage of people just pick up using the words to fit in, granted, but they’re fitting in with a high-specificity group, and it is the specificity itself that comes off as evil-robot, not merely the specific cases of specificity.

          • Nornagest says:

            Yeah, I don’t buy it. You can probably come up with reasons to exclude any particular one, but intelligence is a broad concept and English has plenty of words covering different patches of it. “Intelligent” itself is a nice big four-syllable word that sounds plenty precise. If people consistently talk about it in terms of IQ, that means either a weird cultural focus on IQ to the exclusion of other understandings of intelligence (possible) or an equally weird cultural shibboleth. Specificity is a sideline.

          • albatross11 says:

            I think we mostly have a shared intuitive meaning for “intelligence” that is approximated by what IQ scores measure. There are, however, a lot of ways people can vary and we’re boiling them all down to one (or sometimes two–verbal and quantitative) dimensions, so we’re inevitably losing a lot of detail. And so we also have concepts and sometimes terms for:

            a. People who are verbally very adept but don’t really have a deep understanding of what they’re talking about.

            b. People who are good at deep careful thinking but lousy at quickly coming up with a good-enough answer, and people who are the opposite.

            c. People who are book-smart–they’re good at doing school-related things, but not so good at applying their intelligence in normal life.

            d. People who are wise–that’s something a bit different from intelligence. Very smart people are often attracted to intellectual rabbit holes and traps; wisdom helps avoid them.

            There are brain disorders that involve people having big imbalances in different mental abilities–you might be super verbal but not able to reason about spatial stuff at all, or vice-versa. But there’s also a lot of variation among people w.r.t. different mental abilities–plenty of very verbally smart people who have a hell of a time with quantitative or spacial reasoning, plenty of spacially-brilliant people who don’t sound very smart when they speak.

            But really, this comes down to taking a whole bunch of correlated things and boiling them down to one measurement–like if you invented some metric for physical fitness that had components like 100m dash speed, 1500m speed, % of bodyweight bench pressed, vertical jump as % of height, reaction time, etc., you’d get something that tracked pretty well with what people mean when they say someone is athletic or good at sports. Elite bodybuilders, marathon runners, basketball players, MMA fighters, and ping-pong champions would probably all do pretty well on that score. And yet, there would definitely be differences among them, and two people who were equally athletic might very well look pretty different in their basic abilities, with the ping-pong champion better at reaction time and eye-hand coordination and the bodybuilder better at bench presses.

            We often talk about IQ because it’s the thing we can measure and for which we have statistics. If I tell you that engineers are on average cleverer than janitors, that’s not something many people will argue with, but it’s hard to quantify. But we could get IQ statistics from those two groups and have a precise statement, but about a slightly different thing than we started talking about.

  59. Sockie says:

    sorry but tbh its kinda cringey you had to write this

    lets be real tho, if someones making mistakes this basic, this list aint gonna help, theyre just gonna slip up in a hundred other ways that signal BAD ROBOT

  60. Bugmaster says:

    I think one problem with this style guide is that people on SSC (and, presumably, other Rationality-related sites) often do talk about IQ test scores, sexual dimorphism in humans and other animals, evidence-backed decision processes, neurological conditions, fringe politics, game theory, and sociology — among other such things. I agree that disambiguation is important; e.g., you shouldn’t label someone as “high-IQ” unless you do, in fact, have an accurate numeric measure of his IQ. But, given the topics that are often discussed on SSC, a high degree of jargon is inevitable. This place is a subculture, and it will likely never be acceptable to society at large. Nor should it be — there are lots of other subcultures with their own jargons, after all; and, as Pratchett would say, “it’d be a boring old world if everyone was alike”.

  61. The Obsolete Man says:

    Thanks for writing this. It reminded me of reading Strunk & White’s – “Elements of Style”, for a sophomore exposition class.

  62. AlexOfUrals says:

    It seems to me people here often use “high/low-IQ population/demographics” to refer to respective groups in a more or less polite way. And “demographics” and “population” in this sense are probably also on The Evil Robot Dictionary. So what would be the the non-evil way to refer to these groups? Is there any? Saying “intelligent people” or “not intelligent people” sounds as if the speaker necessarily puts themself into the former group and is extremely arrogant about it. While both might be true it’s probably not the best things to signal.

  63. Dragor says:

    I don’t think I really agree with subsituting “autistic” with “nerdy”. At this stage nerdiness is associated with taste in media more than social awkwardness I think, or maybe that is just people I associate with. For example, I have a friend who is pretty socially deft who once asked her friend and I if we were ever grateful we were “a nerd”. I like aspie, but I don’t think that helps. I guess it sounds more friendly than autistic maybe?

  64. Rafal Smigrodzki says:

    When you inveigh against the status-maximizing sociopaths that are the bane of successful social movements, feel free to call a spade a spade. Sociopaths are sometimes popular, but never respectable.

  65. A lot of this boils to not using language that implies precision that isn’t actually there.

    My favorite example is saying “75% of the time” when you mean “three times out of four”. Notice how the first sounds sciencey and statisticey and therefore implies measurement over a vast quantity of data, when nobody bats an eye at the second one used in the context of anecdotal evidence

    • Thegnskald says:

      “Isn’t actually there” is a bit of a stretch/assumption, but yes. If I intend to communicate that I have relatively high confidence in a figure I’m more likely to use the former phrasing than the latter. (Likewise if I wanted to convey that my mind is open to being changed.)

      To reframe it, if somebody presented data saying the real figure was 81%, 75% feels like more of a lie than “three out of four”; sure, the real figure was “four out of five”, but the more casual framing leads me to expect a more casual estimate.

      • marshwiggle says:

        But 75% is obviously a round number. So it doesn’t feel super off. Plenty of people just outright hate you if you use % that is not 10, 20, 25, 50, 75, 90, 95. Outright hate. Other people really expect you to give your actual % value. What counts as a lie, what counts as good communication, all of it depends so much. Same for the “three out of four” and the “likely” levels of probability language. Which is what makes it so hard when you don’t know who you are talking to.

        • Thegnskald says:

          Not super off, no. If somebody said 75.0%, 76% would feel like more of a lie than 81% to 75%. It’s all a sliding scale based on what the person is perceived to be attempting to communicate; 75.0% is clearly communicating a significant digit. 75% is ambiguous between 5% increments, a significant digit, and rounding, so it ends up somewhere between all three.

          Interpreting what other people are attempting to communicate is a skill.

          Likewise communicating. The first five minutes of conversation with somebody, for someone who is reasonably skilled at communication, tends to be spent dialing down a common language, even if we don’t generally think of it in those terms; ascertaining the level of familiarity of the other person with the concepts in question, the degree of detail they have in their mental models, the familiarity they have with the language which more precisely conveys those details. I’m not going to say 75.0% to somebody who doesn’t understand the concept of significant digits; I’m not going to say 75% to someone who doesn’t really understand the concept of probabilities (very few people really catch the nuance that the difference between 95% and 96% is more significant in terms of everyday experience than the difference between 50% and 51%).

          Fundamentally my take-away from this post is this: Scott wants the comments section to be more accessible to people with less specificity. To which my response is “I guess, but you’re probably going to lose the things you like about the comments section if this actually happens, because the high-specificity people are exactly the people who can have the kinds of conversations that happen here without everything descending into vitriol.”

          • Dacyn says:

            I would take “75%” to be rounded to the nearest 25%… e.g. if I thought the probability was 70% I would round it and say 75%. (This is like with time of day, where you hear X:15 and X:45 more often than X:10 and X:40.)

  66. Bram Cohen says:

    And here I thought my informal language was the result of slacking on figuring out formality rules rather than a preferred style (or maybe that’s your whole point).

    A quibble about ‘status’ though: People with money are often high status but neither popular nor respected, at least not when they’re not around, and usually when I for one talk about ‘high status’ that’s what it means. There might be a tinge of sarcasm in there.

  67. Ben Wōden says:

    I sort of don’t always do 4-6 because I don’t always think they’re the most reasonable, as they can in some ways assume the speaker shares my values or way of thinking. Often, I feel more honest saying something like “this is the most rational way to maximise utility here” rather than saying “this is the best thing to do”, because the former is much more precise and up-front about the way in which I’m coming to my conclusions, whereas if I say the latter, I feel like I’m sort of hiding my thought processes, and that people who, for some reason I can’t begin to grasp, don’t think that rational thinking is the right way to think about what to do, or that maximising utility produces the best result, might then be misled into not noticing that I’m bringing in assumptions that are natural to me but not to them.

    I think this is particularly key for utility vs goodness, because I’m not actually a full-blooded utilitarian myself, I’m a sort-of-contractualist I think, but in a way that means, in many large classes of cases and problably in most cases, my ethics are identical to preference utilitarianism. But I do not think that utility is identical with goodness as a matter of definition or necessity. I think that dropping the terms that make it clear which system of reasoning you’re using and replacing them with vaguer terms makes it harder for people to “see your working”, as such, and means they might be less able to work out whether they agree with you or not. Which, I think, makes it worse communication.

  68. kalimac says:

    When people say “utility” I tend to think of the company that provides my electricity.

    I once encountered someone who used “50-50” as word salad to mean “it’s unpredictable” and not as an indication of odds (which, when I finally got some historical figures, turned out in this case to be more like 95-5). But inquiries have so far not produced other instances of people talking this way. Anyone here encountered it?

    • OriginalSeeing says:

      I’ve been told that if you have absolutely zero information about the likelihood of one of two events occurring, then your probability estimate should be 50:50. I can see how this could make sense from a Bayesian point of view, but it has always felt anti-intuitive to me.

      I would use the word “unpredictable” in these situations, but if that was a normal person, then my use of the word is probably a different meaning than what they meant.

      • kalimac says:

        Well, estimates are based on your best information. If there’s no information, I guess 50-50 is what it zeroes out as. But in the case I’m thinking of, the person had all the necessary information, it just hadn’t been tabulated. And when he told me that the likelihood changed under different circumstances, and when I asked for the likelihood under each and he told me both were 50-50, I knew something was wrong.

      • RC-cola-and-a-moon-pie says:

        You may need to be more precise in applying a principle of indifference than just saying that any binary event is 50/50 in the absence of any empirical evidence. “Not rolling a 6 on a six-sided die” can be characterized as one of two possible events following a roll (the other being that a 6 comes up). But of course you can’t assign the former a 50% likelihood or you’ll quickly run into contradictions with other binary probabilities involving outcomes of the same die roll.

        • John Schilling says:

          Knowing that there are five other possible outcomes at the same superficial level of significance, counts as knowing “something”, and so goes by different rules than would apply when you know “nothing”.

          Knowing literally nothing except that a particular outcome is possible, is a rare condition and you’ll have to work at it to find an example for illustrative purposes.

        • Said Achmiz says:

          To add to what John Schilling said: the critical piece of information here—the one that takes you away from your initial state of “indifference”—is the information that the sides of the die are labeled in such a way that there is only the one ‘6’.

          If you know the die is labeled { ‘1’, ‘2’, ‘3’, ‘4’, ‘5’, ‘6’ }, then you know there are 6 possible outcomes, of which ‘6’ is one.(You still don’t know their probabilities—it could be a weighted die—but you know what outcomes are possible.)

          To illustrate this further, consider an alternative scenario: what is the probability of rolling ‘6’ on a Fudge die?

          Is it 1/2? 1/6? No, it is 0, because none of the sides are labeled ‘6’.

          What if you got a custom six-sided die, with some of the sides labeled ‘6’, and some labeled ‘1’, and none labeled in any other way? You’re not told how many sides are labeled ‘6’ and how many are labeled ‘1’. What would the principle of indifference have you believe about the probability of rolling ‘6’? Well, then it would be 1/2, because there are 2 possible outcomes.

  69. Quixote says:

    Many of these are probably good. But I really don’t like the idea of getting rid of probability numbers and replacing them with sloppy and unmeaningful adjectives. The language we use forms habits of thought. Using and thinking with numbers leads to experiencing and thinking about the world in higher resolution. You’d be a fool to give that up.

  70. carvenvisage says:

    How not to sound autistic nerdy robotic 101. This is fine as a guide for speaking to the defective fleshdroids who think they’re bunnies or something, but why would that be people’s aim on the subreddit? Personally I think it’s nice having somewhere where you can sound like an evil robot. I definitely let loose a little with the robotic-ness sometimes when I’m comment on an SSC comment section.

    Object level: As for the post itself, I think the weakest one is 6; “males”, because gender and sex are different (at least I’m pretty sure that’s what Scott believes) and I can see no reason why you’d have be morally obligated to talk about gender. I think that is probably just Scott’s “just go along ffs” bias as a hardworking psychiatrist coming through. Regardless, when I refer to males rather than men it’s not a stylistic choice, I’m choosing to progressively include cis-by-default people as well as gender-identity-havers.

    Number 7 I don’t think is like the others because using autistic in place of nerdy seems pretty decidely autistic as something people mostly do for the irony and lulz, and/or as some kind of pro-autism statement like neurotypical.

    no 10 is the only one I can sign off on, as IMVIO (in my very important opinion) “Status” has a distinct meaning from doesn’t mean the same as other adjacent terms and often it’s used to refer to subcomponents of status for which there are more specific terms.

    Right, and the obvious thing to point out is that people posting randomly on the SSC subreddit obviously aren’t reaching out to non-robots the same way you are on the main blog. Don’t cramp people’s style.

    • Nornagest says:

      Personally I think it’s nice having somewhere where you can sound like an evil robot. I definitely let loose a little with the robotic-ness sometimes when I’m comment on an SSC comment section.

      Confession time: I, too, have an evil robot in my heart. My first draft of anything usually comes out looking a little awkward and stilted, like I’m writing a script for the smart kid in a bad TV show. I put a lot of effort into toning that down, pushing a more conversational tone, not because I want to sit at the cool kids’ table but because it works better. Yes, even here. Even on the subreddit.

      Why? Because, no matter how robotic you think you are, decoding evil-robot speak takes effort. Parsing the shades of meaning out of obscure words, unraveling complicated sentence structure, figuring out if the guy you’re talking to said “male” because he’s taking some sort of gender-studies stand or because he read an evopsych book once: that’s work. And the whole point of good communication is to make your ideas easy to understand. By all means be precise where you actually need to, but writing that’s dense without a good reason to be is bad, bad, bad.

      • carvenvisage says:

        Your conversational tone leads you into error, earthling. -That’s not a confession, that’s surrepticious advice! (Note well: A more clinical approach conceivably could have preserved you from this undignified incongruity.)

        Why? Because, no matter how robotic you think you are, decoding evil-robot speak takes effort.

        No offence but I think that no-you are the one who is out of touch: Like most humans, I generally find other people easy to parse. This doesn’t hold for (literally) autistic people, nor programmers who have to internalise “not clear enough, not clear enough, not clear enough” all day, but you definitely get more of these here than in the wider world.

        • Nornagest says:

          I am a programmer, bro. One working in a subfield that’s picky and detail-oriented even by programming standards, too. I’m not making assumptions about how nerdy people ought to behave, I’m speaking from my own experience as a nerd communicating with other nerds.

        • carvenvisage says:

          I know! That was my point! Let me translate that second part into non-larping/playful english for you.

          No offence but I think that no-you are the one who is out of touch: Like most humans, I generally find other people easy to parse. This doesn’t hold for (literally) autistic people, nor programmers who have to internalise “not clear enough, not clear enough, not clear enough” all day, but you definitely get more of these here than in the wider world.

          I don’t have a hard time understanding people and neither do most non-[nerds].

          I don’t think the difficulty people have with parsing others’ loose and undisciplined speech here is representative of the general population.

          (Scott’s laboriously systematic and clear posts no doubt attract people who appreciate systematic clarity and thoroughness.)

          • Nornagest says:

            I thought I was clear enough in the ancestor, but I’ll give this another try: if you don’t have a point to make that requires technical language (hint: you probably don’t), you should avoid a technical tone, whether you’re talking to a room full of engineers or your local high school football team.

            Literal autists and undersocialized nerds often use technical-sounding language. An inexperienced speaker might assume that they can communicate better with those guys by speaking that way themselves, but that’s wrong. And that’s because nerd-speak, besides sounding technical, is usually highly idiosyncratic, so two nerds using their nerd-speak usually end up talking past each other. For example, see the exchange re: Vance above.

          • carvenvisage says:

            Your english parsing ability is (seems to be currently) really bad. I understood what you said at the start. I acknowleged it and indicated that I disagree with it and/or don’t take it seriously. Your repeating yourself just shows that you haven’t been following the conversation.

            Literal autists and undersocialized nerds often use technical-sounding language. An inexperienced speaker might assume that they can communicate better with those guys by speaking that way themselves, but that’s wrong. And that’s because nerd-speak, besides sounding technical, is usually highly idiosyncratic, so two nerds using their nerd-speak usually end up talking past each other. For example, see the exchange re: Vance above.

            To be blunt, the exchange above was occasioned entirely by your (perhaps-occupational-hazard-occasioned) obtuseness. If you spend 8 hours a day practicing intolerance of ambiguity, don’t typical-mind normies as having the same interpretation challenges you have.

            -I, despite understanding your position, disagree, with it.

            (Possibly because I, being a different flavor of nerd than yourself (-more gnomish than dwarvish) simply don’t experience the same social difficulties you do.)

            Now I really see nothing more to argue about, so adieu and all the best.

          • Nornagest says:

            If you understand my position, you’re doing a damn poor job of coming up with objections that make sense. Did you miss the bit where I said this is about nerds speaking with other nerds? I don’t care how good normies are at communicating. That is not the point. The point is that nerds are usually pretty bad at it, and so they should use clear, colloquial language while they’re talking to each other.

            Normies don’t even enter into it. Why would I be talking about that? My whole beef is with the idea that rules like Scott’s aren’t necessary in a venue like SSC comments or the subreddit, where we might have a half-dozen normies that we parade around as mascots but they’re not driving the conversation.

            I can’t tell if you’re trying to be cute or if you think you’re in some other conversation, but either way, don’t blame me for your problems.

          • albatross11 says:

            This makes me think of a quote by SF writer Larry Niven:

            If you’ve nothing to say, say it any way you like. Stylistic innovations, contorted story lines or none, exotic or genderless pronouns, internal inconsistencies, the recipe for preparing your lover as a cannibal banquet: feel free. If what you have to say is important and/or difficult to follow, use the simplest language possible. If the reader doesn’t get it, then let it not be your fault.

            A lot of my job is writing academic papers and giving presentations–both are all about communicating some pretty complicated and hard-to-understand ideas to people who often don’t quite have the right background to follow them in detail. The best way I know to do that is to focus all my attention on communicating clearly–using terms I’m pretty sure my readers/listeners will understand, revisiting definitions so we’re all on the same page, etc. Technical terms are inescapable sometimes, but it’s *really* important to define those terms and not to overuse them. Similarly, sometimes you need some mathematical formalisms or definitions or concepts, but again, you want to make sure you use only the ones that you need, and that you make it accessible for the widest possible range of readers/listeners.

          • cuke says:

            For what it’s worth, Nornagest, you’re making perfect sense to me.

            You make what I think is your central point pretty clearly in two places:

            “the whole point of good communication is to make your ideas easy to understand. By all means be precise where you actually need to, but writing that’s dense without a good reason to be is bad, bad, bad.”

            and

            “if you don’t have a point to make that requires technical language (hint: you probably don’t), you should avoid a technical tone, whether you’re talking to a room full of engineers or your local high school football team.”

            I don’t understand what Carenvisage is disagreeing with, exactly, but seems to be saying that there is comfort in an in-group having its own lingo, and maybe in not feeling pressure to do the self-editing work that you describe doing.

            Maybe this is talking past each other or just expressing different priorities — yours perhaps to be easily understood; theirs to not have to work so hard with their in-group.

            I’m a talk therapist and so I feel like I get a lot of inside views of the possibilities of misunderstanding even when the in-group is two people. My experience tells me that communicating well takes a lot of practice and that people generally assume mutual understanding is easier to achieve than it usually is.

            And by communicating “well” I don’t mean precisely (or at least not mainly); I mean, taking responsibility for the hard work that it is, and doing one’s share to make it go as smoothly as possible, granting the other person charity, practicing humility and patience on one’s own end, and recognizing that misunderstandings are incredibly common.

            Carenvisage does not seem to be approaching this conversation with you with that awareness. It always surprises me on here when someone is willing in the course of a couple of back and forths to just resort to ad hominem attacks — like saying that someone is “bad” at communicating (ie, “Your english parsing ability is (seems to be currently) really bad.”). I would wish for less of this on here.

          • carvenvisage says:

            A lecturer, a programmer and a psychiatrist walk into a bar…

            @cuke why are you weighing in if you literally know you don’t understand? Nornagest’s advice is accurate so far as it goes, but inane, unsolicited, irrelevant to my first post, and presented disingenuously.

            (Confession time: I too suffer from (problem I know you emphatically don’t think is a problem but is a problem mmkay))

            and where I have been politely trying to disengage from the start.

            But never mind the context: you’re actually just confused if you think information focused communication in a professional environment where it’s your job to make sure you are understood by particularly communication-challenged people, is a useful comparison for normal recreational social communication.

            They’re not, confusing them is just, -no doubt coincidentally, literally all three of you’s professional failure mode for casual social communication.

            And I’d say “less of this” right back at you (less knowingly-ignorant disparaging of others who have just tried to vacate the thread), if I didn’t think you were just genuinely being obtuse and missing the point, after the fashion of a particularly sweet but doddering old grandmother.

            Last note: on another day, I’d happily join in on your little agreement party. Humble simplification is one off the best ways to keep things unambiguous, and to help people who are having trouble processing things. -If that’s your goal, which, however clearly wasn’t the topic of this thread, even before several reclarifications.

            Personally I think it’s nice having somewhere where you can sound like an evil robot. I definitely let loose a little with the robotic-ness sometimes when I’m comment on an SSC comment section…

            @nornagest I’ll say it again, I don’t give a shit what you think. Stop bothering me. Stop repeating yourself. Don’t give people unsolicited advice for not to be how they say they like being, especially in a bait and switch format. Frankly, reevaluate your estimation of your social skills; just because you found a few tricks to cope doesn’t make you an expert nor a fountain of wisdom.

            Alright, second attempt now, this one a little more decisive- not reading any more of your nonsense: goooooood-bye.

          • Scott Alexander says:

            Carvenvisage is banned (in accordance with the Second Commandment) for three months; would have been shorter but they have already received a warning

  71. carvenvisage says:

    Also I’m not sure that anything makes people look more like evil robots than openly conspiring not to sound like evil robots.

    Greetings, fellow -checks guidebook- “people”…

  72. Dacyn says:

    10 is a good suggestion, in particular because “status” seems to be ambiguous between popular, respected, powerful, highly ranked on some formal hierarchy, and maybe some others. It seems better to hug the query and talk about whichever one of those you actually want to talk about.

    Or maybe people believe there is a “general component of status” that correlates with all these things but is irreducible to any of them: in that case, it seems reasonable to use the word “status”.

  73. Rand says:

    This entire post can be condensed into “if you want to be taken seriously, don’t talk like non-human characters from Star Trek”. (“If you would prefer other humanoids to perceive you as rationale, do not adopt the speech patterns of 1970s Earth science fiction. You will be assimilated.”)

    “Rational” instead of “smart” is Vulcan.

    “Females” and “humons” instead of “women” and “people” is Ferengi.

    “Optimal” and “high probability of” are androids, particularly Data. (Probably Spock too.)

    The rest is “I only read SlateStarCodex” but could serve as inspiration for Star Trek scriptwriters.

  74. Glacian says:

    I don’t believe the suggestion about memes is quite right.

    The term “meme” has largely come to be associated with internet memes, which are not quite beliefs in themselves and although “meme” as it was coined by Dawkins includes beliefs, even in its original form it never only included memes, so reducing it to them isn’t appropriate in every (or even most) contexts. Using it in the academic sense to refer to beliefs may even be more robot-like than using it to refer to the jokey images and phrases that proliferate online.

    More importantly, it would be misleading to conflate memes with beliefs. Memes may be produced to generate controversy, or to serve as propaganda, independent of whether the people producing, spreading, or considering those memes believes them. And sometimes they’re not even things a person can believe. You can’t “believe” Pepe the frog.