Published in Pioneers

Gretchen McCulloch talks about why we talk the way we do online

By Devon Zuegel


70 min read

Gretchen McCulloch is an internet linguist — an analyst of the language of the internet, for the people of the internet. She's the author of the New York Times bestselling Because Internet: Understanding the New Rules of Language, a Resident Linguist at WIRED, and was formerly the Resident Linguist at The Toast. She also co-hosts a podcast called Lingthusiasm, a podcast that’s enthusiastic about linguistics.

DEVON: Hello, I'm Devon, and you're listening to Tools & Craft. Today, I'm talking to Gretchen McCulloch. Gretchen is an internet linguist, which means she studies the language and people of the internet. She's the author of The New York Times best-selling book, Because Internet: Understanding the New Rules of Language, which is one of my favorite books that I've read in the past few years. Gretchen is also the resident linguist at Wired, and the co-creator of Lingthusiasm, a podcast that's enthusiastic about linguistics.

Language and writing have always been important, but I'd argue that they're even more important today, because every year, more and more of us are spending our days typing into a computer, talking into cameras, and exchanging information. So, I'm really excited to talk to Gretchen about how language shapes the way we think, and how we are reshaping language. So Gretchen, thank you so much for taking the time to chat today.

GRETCHEN: Yeah, thank you for having me.

DEVON: So, I've noticed that in some companies, people use emojis so much in Slack, that it feels like if you don't include an emoji, you must be angry or something, and I refer to this as emoji inflation. First of all, does that resonate with you as something that you've seen elsewhere? And when this kind of inflation happens, how do people make up for it, so that information isn't lost?

GRETCHEN: This kind of emoji inflation reminds me of a linguistic phenomenon known as the euphemism treadmill, and the hyperbole treadmill. So, it's this idea that as a, for example, euphemism, gets used more and more for the properties of a thing that we feel gross about, or feel not great about. So if you have something like bathroom, which is, itself, a euphemism for maybe toilet, which is, itself, a euphemism for various other things, and as that gets used more and more, people start feeling even gross about the word bathroom, and they start saying, "Okay, well, maybe I'll use going to wash my hands, or going to the little boys' room," or something like that, as another euphemism for the same thing.

So the euphemism doesn't stay in one spot, it takes on the connotations of the thing that it was trying to replace. And the same thing goes with hyperbole. So words like very, and really used to be super hyperbolic. Very comes from... it's like, verily, as in true, and really, of course, comes from real, truly come from true. And so this is really, truly, verily the thing, except now you're just like, "Oh, yeah, I'm really tired."

We see the same thing going on in the modern day with a word like literally, which also has this meaning of real and true and physical. Literally actually comes from letters, like to the letter of the word. People use like, I'm literally starving. And okay, you're not starving by a clinical definition of starving, but you are using hyperbole in a way that's very familiar to how hyperbole gets used.

And then eventually, literally... This grinds a lot of people's gears, but it's been happening for centuries, that literally doesn't actually mean you're literally starving. So if you want to convey that you're actually, truly, literally, physically famished, you need to use other words. I think physically is beginning to take on some of this connotation, I think actually is taking on some of this connotation. It's a natural cycle that goes through, where the hyperbole gets inflated, or goes on this sort of treadmill, and it takes on this connotation, and to fill in the spot. So I think that emoji, and other types of ways of conveying emotion and intention online, can also become part of this sort of treadmill, or this sort of cycle of inflation, where you eventually need to pass on to something else.

DEVON: Okay. That makes me feel a lot better. Because I often found that... In press companies I've worked in, I almost found it hard to communicate sincerely, when I actually really liked something.

GRETCHEN: This happened was LOL, for example. So LOL is used... L-O-L is often used these days to indicate like, I'm just trying to indicate that I'm not being hostile, and it's... It sounds good, LOL, okay, fine. It's just a softening marker, it's just conveying this tone, and it's not literally conveying laugh out loud to a lot of people anymore. That's because it goes through this cycle of... At first, you're actually conveying laughing out loud, or full on laughter, and then you're using it to convey some sort of like aspirational laughter. Like, I wish I was laughing. I'm not actually laughing, but you're still my friend, and I want you to think that I'm laughing.

DEVON: Right.

GRETCHEN: Which is really nice of us, actually. If you think about it, it's very pro-social, it's like, oh, I want you to think that you're funny. I know, you're not. And then it gets used to pave over a moment of awkwardness, or to... Which we also use laughter for, to be fair. Like, you don't always laugh at someone just because they're a comedian, you laugh at someone because, oh, I feel a bit awkward about this. So LOL goes through this inflation, and now, these days, if I want to actually convey laughing out loud, I'm not using LOL.

Some people may still be doing so, but I think for a lot of people, if you want to convey genuine laughter, maybe you're doing an elongated version of LOL, like, LOL maybe you're doing LMAO, maybe you're doing hahahahahaha. A lot of cases, a really sincere laughter is somebody pausing to say, "Oh my God, I actually just spat coffee on my keyboard," or some really long description of what you're doing. And these just go through cycles.

DEVON: I've been trying to popularize L-L-O-L, for literally laughing out loud.

GRETCHEN: How's it going for you?

DEVON: So I don't have quite enough cache, I think, to make that happen.

GRETCHEN: I think it may also confuse people in comparison to the use of the elongated version.

DEVON: Yeah, yeah. I know. I'll have to work on another one. It seems like sometimes this inflation varies a lot from group to group. So, for example, I have some friends who grew up outside of the US, and they often point out that Americans always say that things are great or amazing, even when they're neutral or good, or like an Argentinian Spanish, they'll say, "re bueno," instead of just "bueno," it's like, very good, as opposed to just good, and they put it in front of every adjective you can imagine. And then other types of Spanish don't have that. What will predict if a culture or a group of people will inflate something faster than somebody else?

GRETCHEN: I think there's inflating faster, but there's also just inflating on different words or in different domains. So if great is being used in American English, or also Canadian English, I'm Canadian... For example, when I think of Germans using positive words for things, I think of them saying super a lot, which you wouldn't necessarily say in English, but is also an inflation, it's just a different word. And so you run into these sorts of communication things, where someone's using a word that feels very literal to you, and they're just using it... Or, British people will say brilliant a lot. And they're just saying, "Oh, yeah, brilliant, good." But, if you're not used to that, you're like, "Wow, everyone thinks I'm brilliant." So there's a lot of different domains that that can happen in, and running into one person using a word literally, and one person using a word hyperbolically, that's where you get some of these clashes of, "Wait, how am I suppose to interpret this thing?"

DEVON: That makes a lot of sense. And so it might feel like you go somewhere else, and they're hyperbole of all out everything, but it's actually fine. Because what they're doing... Not fine, that's putting too much judgment on it. But it's actually, they're just doing it in a different place than you are, so you notice it more.

GRETCHEN: It's a difference. I think that this is fairly common, at least in a lot of modern-day Western cultures, I wouldn't want to presume outside of that. But the idea that something being really good, and then making a hyperbolic claim about the goodness of something, or the greatness of something, is a desirable behavior to have. Any culture where that's likely to happen... If you have a culture where you don't want to praise things too heavily, because that might cause bad luck or something, which is also true in some places, then maybe you wouldn't see that type of hyperbole. But, you might see the same hyperbole in the other direction, where certain types of phrases become stock phrases that are "insults", but they're actually used in this positive way to avert bad luck.

DEVON: Can you give an example?

GRETCHEN: So I'm thinking of... And I wish I could actually remember the specific cultures that do this, but... because I don't. But I think there's a culture where you don't want to praise children, because that might attract bad luck, or might attract jealousy, or something like that. So instead of saying your child is like, good, or smart, or kind or something, you could say that they're bad, or they're dirty, or they're shorter, they're one-eyed, or something. There's a whole bunch of these stock phrases that are used to talk about children that sound very negative, but in the cultural context, they're positive, because you're trying to avoid them getting bad luck, or childhood diseases, or all these sorts of things.

So, in that type of environment, I would expect that a word like great, fabulous, good, brilliant, super, awesome, probably, wouldn't have that type of treadmill or in the in the positive direction, but you might have something like that in the negative direction, where something that means dirty or scrawny, or something, would start taking on like, oh, well, this is just what you call children, because that's the thing everybody knows you're doing.

DEVON: I've also noticed that in my own experience, I feel more pressure, internal pressure, to inflate the positive words that I'm using in situations where maybe I'm socially uncomfortable, or I want to make sure that the other person knows that I like them, and I want them to know that I'm on their side, or low context situations, where I really don't know the person very well at all, and I want to make sure that I come off as friendly. Let's say I have a taxi driver who's saying some crazy stuff that I really don't agree with at all, but I'm in their car, and I'm alone, I'll be really positive, and be like, "Wow, that sounds really interesting. That's really interesting-


DEVON: So basically it’s sort of a sense of social discomfort, pushing me to be more positive than I actually feel internally. I wonder if because the internet is often a solo context, do you see that occurring at a faster scale, or am I overreaching there?

GRETCHEN: What I have noticed is that there's a style that's the opposite of this super positive, super enthusiastic style, which is a type of graphical minimalism, where you don't use a lot of punctuation at all, you don't use a lot of capitalization, you maybe use just a line break or a period every so often, but you do have this minimalist style. I think of it as a sort of early 2010s Tumblr. Some people talk about it as a queer aesthetic, some people talk about it as a Gen Z aesthetic.

One of the things that's interesting about it is that it's not doing a lot of the authoritative signals that we have, like, I'm going to use standard capitalization and punctuation, because I want to show that I've mastered this set of norms that comes from a certain type of educational system or educational track. Using this, okay, I'm going to write everything in lowercase, I'm going to not put punctuation here, is saying, if you don't have the cultural context to understand the norms of this particular community, I'm not interested in having you talk to me.

So you can, sometimes, use the writing style itself to encourage or discourage people from engaging with you. Because if they find that hostile, if they find that weird, if they find that alienating, you're like, "Great, I didn't want to talk to you anyway." And you can attract certain people to which kinds of contexts you're trying to talk to, or not. I think that some of the styles get parodied, as well, the boomer style of all these dot dot dots, and all of these ROTFLOL emoji, and this sort of thing, and capitals in lots of places.

If you don't communicate in that style, if you think that style is not something that you're into, or something that you'd mock more than you would do, then that's a sign this isn't a social group that you particularly want to be a part of. So you can use it to draw boundaries, even though the internet makes it hard sometimes to draw physical types of boundaries.

DEVON: I feel like there are a lot of different things that exist in the physical world that do this, too, where... In verbal language, you might have slang, and in fashion, you might have some style, like goth makeup, for example. A lot of people find it really unappealing, and that's, I think, part of the appeal of doing it, is so that you separate yourself out, and you say, "Hey, I want to be with a totally different group."

GRETCHEN: Yeah. There's also the physical cues of like, if I were to try to go and hang out with the teenagers at my local high school, they'd look at me like, "You're an adult, you look like a creep." So you have that physical cue, whereas if I'm watching teens on TikTok doing their videos, they can't tell that I'm there, and they can't tell that I'm not their age. Especially when you have age-gated boundaries, there's a huge incentive for youth to engage in behavior and linguistic styles that older people are going to find off putting, because they want to put the older people off.

DEVON: Right. They don't want them involved in whatever they're doing.

GRETCHEN: Right. When your parents get Facebook, maybe you don't want to be on Facebook with your teenage friends anymore. That's really understandable.

DEVON: To stretch the inflation metaphor to its limits, if we have inflation of language, do we also see deflation of language ever?

GRETCHEN: We do, sometimes. I think a really illustrative example of this is with the exclamation mark. You may remember, back in like the '90s, the quintessential internet use of the exclamation mark was tons of exclamation marks, and then people would occasionally accidentally type the number one in the middle of their strings of exclamation marks. Remember this?

DEVON: Yeah.

GRETCHEN: And then that itself became parody behavior, where sometimes people would write out the word, O-N-E, eleven, E-L-E-V-E-N in the middle of the strings of exclamation marks.

DEVON: I don't remember that one. That's-

GRETCHEN: Okay. Well-

I'll start doing that again.

GRETCHEN: My friends  have been doing this. But you do still see it occasionally. So exclamation mark, exclamation mark, one, exclamation mark, exclamation mark, eleventy, exclamation mark, exclamation mark. This was this Internet behavior, and this was cool for, I don't know, a few years. And then it got to the point of parody, where now people were putting the one and the eleventy and stuff in to be like a hyperbolic, I'm not actually excited about this, I'm doing fake excitement.

And then, if you will, the exclamation mark stock crashed, and people stopped doing a lot with exclamation marks for several years in the 2000s. Like, really not a whole lot with exclamation marks. And then around... I think this was around 2015, 2016, or so, because it was while I was in the early stages of writing, Because Internet, which is how I mark everything in time. Now, there was an Atlantic article about people occasionally using three exclamation marks, and how this was this new thing, and I was like, "But people were also doing this in the '90s, it just like stock crashed and then came back."

In the meantime, this whole time, you did have a single exclamation mark, which is on a different path, and this is the sort of single explanation mark thanks! One exclamation mark just seems polite, and it seems cheery and breezy, and that one has continued this whole time. There's a paper by Erika Derrick's looking at it in business email, and how people will often use it to soften polite requests, or make things seem a bit more cheerful.

But the multiple exclamation mark use really had this decade or so of fallow period, and then came back by people who'd maybe forgotten, or maybe just like... you know how jeans go in and out of... how wide they are, had gone in and out of trend, and now the bell bottoms and the and the multiple exclamation mark are coming back. And then you see strings of long exclamation marks, but it might not stay there. Because eventually, it'll be like, okay, multiple exclamation marks, these are cheesy, we use this too much, are they actually sincere? Probably not. So maybe 2030 or something, it's like an exclamation mark stock crash again, or something like that.

I think we are also currently are in the early mid stages of a stock crash around the face with tears of joy emoji. So, this was the most popular emoji for years and years and years, 2015, 2016 2017, 2018. This was the most popular emoji. And then it starts going down, and especially younger people are saying, we're using different smiling emoji to convey strong emotion. Either we're using loudly sobbing face to convey like, this is so amazing, I'm overwhelmed, which isn't a smile, or they're using the smiling face one, with the eyes that look like the greater than less than signs, or some of the other ones, like the skull for like, I'm dead with laughter, various other ones that I probably don't even know.

And then you get this sort of panic of like, older people being like, wait, wait, but I was like... I thought we were cool with the face with tears of joy. Because when you get older, you stop maybe paying quite as much attention to micro trends, because you have a life and you have friends now, and you don't feel like you need to maintain your social standing quite as tightly, which I think is very healthy. It's okay to be an older person. I like not being a teenager.

People are like, "Oh, the teenagers are doing this. Should I stop doing this, so that I don't seem out of date?" And it's like, yeah, but, do you want to go back to being a teenager, when all you literally had to do was care about popularity, and you didn't have a hobby or friends who just likes you for who you are? Because you can just also just accept that you're an adult, and not be cool. I just want to defend not being cool.

DEVON: I am in my late 20s now, and so I'm starting to see my friends split between... Well, there's a split between people who just don't give a shit, and they just keep using whatever they were using. And then the ones who are like, "Wait, I'm not 30 yet, I'm still young," and they make an effort of keeping track of the TikTok and the slang that younger kids are using and emojis younger kids are using, and I think it's pretty... I will withhold my own judgment that I feel, but I also... It's entertaining to watch, to say the least.

GRETCHEN: Yeah, it can be disorienting to be, oh, yeah, like me and my friends used to be the arbiters of what was cool, and now we're not anymore, and we have to pay attention to these younger people, instead. And your constellation can be, it will definitely happen to these younger people, too, that in other 10 years, they will also not be the arbiters of what's cool. Like, you don't get to stay there for your entire life.

Also, I think, as somebody who was never really that cool, it's a great relief to me that my peers aren't, either. So there's this thing going on with the emoji, in terms of like, whether they're positive, whether they're laughing. Whenever there's a new face emoji, that one will often catch on pretty fast, because people are desperate for, oh, what's the new thing that we could do with emoji, because the ones that have been around for a long time have gotten tired.

I haven't yet seen this turn into an emoji backlash of, we're not going to use emoji at all anywhere, because emoji remind us of that 2010s, and they're so retro, and we're not doing them. But I wouldn't be surprised if it happened someday, because that's another place you could go. You could say, "Oh, we don't actually use emojis anymore, we only use plain text emoticons, because they're so retro, and they're so cool, and we love everything else about the '90s and the early 2000s." Bring back low rise pants, and also bring back text-based emoticons. I wouldn't be shocked if it happened. But obviously, you can ever actually predict something like this. At what point does it become a cycle?

DEVON: And if it does come back, it'll come back, but with its own evolution. It'll be done slightly differently, but with sort of a... It'll never be exactly the same as the thing in the past.

If you see it as, okay, language is always a moving target, it's a living thing that exists in the minds of living beings, and nothing about human life or human society, or human culture is exactly the same from one generation to the next, and language just comes along as part of that.

GRETCHEN: Right. And even if emoji do go out for another 10 years or something, then maybe they come back in, because now they're retro. So there's not just one thing, there is a whole bunch of potential places for these things to go. If you see it as, okay, language is always a moving target, it's a living thing that exists in the minds of living beings, and nothing about human life or human society, or human culture is exactly the same from one generation to the next, and language just comes along as part of that.

DEVON: So, interruptions are a lot more common on Zoom and, to some extent, phone call, but especially internet media conversations, sometimes just because the internet dies, or there's a lag. And so, in face-to-face conversation, I find interruptions to be just incredibly rude, and the person usually is doing it intentionally, or can't control the fact that they want to say something. Versus, on Zoom, I'll usually assume that it's just the internet connection, or something like that, and I try to adjust and realize they probably aren't trying to do it. Have you seen, especially in the last year and a half, with people spending a lot of their time on Zoom, have you seen interesting behaviors in how people handle interruptions? How has that evolved?

GRETCHEN: Well, so we're going to take a step back and talk about what we mean by interruptions, and how different... Because there were already different styles of the ways people deal with them. So the linguist, Deborah Tannen, talks about two different styles of behavior, which is definitely continuum. But we can talk about each end of the continuum first. This is a difference between high involvement conversation style, and high considerateness conversation style.

High involvement conversation style shows that you're engaged, and shows that you're paying attention, and shows that you're interested in what someone's saying by being very involved in what they're saying, and anticipating what they're saying, and reacting at the exact right point, even anticipating what they're saying, so close that you manage to finish their sentence with them in the exact way that they were doing it, and being very on the ball in terms of what someone's saying.

High considerateness conversation style is about leaving space between what people are saying, giving each thing that someone says sort of full air in terms of, okay, someone's going to take their turn, and then I'm going to give it a little pause, and then someone else is going to take their turn, and giving everything in there room to breathe. Neither of these are wrong. There's not a problem with either of these, and it is a continuum.

So you can find yourself, "Okay, most of the time, I'm at the considerateness end of the spectrum, but sometimes I end up talking to someone, where I'm like, 'Oh, my gosh, why are you not saying anything?'" And it's because they're waiting for a pause that's even longer. Or, most of the time, you can be at the high involvement end of the spectrum, and then sometimes you can find yourself talking to someone, and you're like, "I can't get a word in, edgewise. You're talking so much. You're interrupting me."

This can definitely interact with other things that are going on, but it's useful if you're talking with somebody who has a different conversational style from you to pay attention to that, and pay attention to if you're the only person talking, and the other person isn't getting a turn in, edgewise.

This can definitely interact with other things that are going on, but it's useful if you're talking with somebody who has a different conversational style from you to pay attention to that, and pay attention to if you're the only person talking, and the other person isn't getting a turn in, edgewise. Or, if the other person waiting for you to say something, and you can try saying it, even though they seem like they're still talking.

One of the stereotypes that Tannen mentions as far as helping people get a grasp on which groups are associated with which thing is that New Yorkers tend to be high involvement, and Californians tend to be high considerateness. Oh, you were saying, people interrupt so much in conversation, and it's bad. And I was like, yeah, well, not where I'm from. So, if we're coming to electronic types of communication, coming to it with a background, first, of like, okay, well, what happens offline? And what types of conversation styles can cause friction, even between different types of offline groups? And I'm going to give you some space to talk.

DEVON: No, I see what you mean. I guess what I'm talking about is interruptions in which someone cuts off a full sentence, and changes the direction, which feels a little different than the high involvement style you're talking about, which is like, you're saying a sentence, and I'm going to add on to the end, and kind of that thing. It's more of like, "Hey, I'm going to hijack this conversation and go somewhere else with it, because I don't like where you're going." Is that also more of a difference, though? And maybe that is less rude in high involvement or otherwise styles?

GRETCHEN: I think that one of the things that I find challenging about electronic communication, as a relatively high involvement person, is that because you potentially have like a second or so of lag, you don't have as many obvious cues to, is this person going to continue their thought, or have they finished, and is it my turn now? Especially if you're used to having those pauses be really tiny, you're sitting there being like, "Am I allowed to break in, or am I not? And is this person just continuing what they're saying, because they're waiting for me to break in, which is a very high involving thing to do, or is this person actually have a thought they're trying to continue?"

I think that one of the things that I find challenging about electronic communication, as a relatively high involvement person, is that because you potentially have like a second or so of lag, you don't have as many obvious cues to, is this person going to continue their thought, or have they finished, and is it my turn now?

So I think that, especially Zoom, in particular, as a platform, which tends to automatically mute or semi mute other people when one person is talking... I saw somebody complaining about this in terms of like, whoever designed this feature for Zoom has never met Jewish people, which is also a group that's stereotypically very high involvement. There are lots of groups of people that are high involvement. It's not only one group, but it's two types of behaviors that are found along a spectrum in various groups of people.

So I find that particular thing harder, and other platforms that don't automatically mute people while someone else is talking, I find it a little bit easier to break in, if I want to. Also, one of the ways that people regulate whose turn it is to talk is through things like gestures, and eye gaze, and things like that. So if you're sitting there with your hand outstretched, and you're trying to make eye contact with the speaker, that might indicate, oh, I'm trying to break in.

Yeah, you kind of have that in a video call. Some people use unmuting as a lightweight signal that you want to take a turn. So if everybody, except for the speaker, or main two speakers, are unmuted, and then I unmute myself, maybe that's a term that I'm trying to talk, as well, and I've seen some people notice those types of things and help facilitate conversations that way. But I think it is something that these really subtle delays that we don't really think we notice in a conversation can create this added layer of friction, where you're not sure if someone's trying to break in or not.

DEVON: Yeah. I've definitely done the flicker or the mute kind of thing to show that like, "Hey, I have something to say. But also, I don't need to cut you off right now, but just know that I have a thought that I want to add." Your mentioning it now makes me realize like, it's not actually clear to me that everyone sees that signal. Definitely, some people do, I've noticed it works often, but it also... it doesn't work all the time.

GRETCHEN: Not everyone sees the signal, not everyone is necessarily looking for the signal. I've noticed sort of status... I can't tell if its status differentiation, or age differentiation, because so often, that's correlated. But if I'm in a group of people... say, we're at a Zoom social at a virtual conference, or something like that, and there's like eight people or something, all of the students and junior people will be keeping themselves on mute the entire time, unless they have a specific thing to say. Whereas the older people, who are more also likely to have higher social standing in this situation, they have a certain amount of seniority, will just leave themselves off mute the whole time.

It's not clear to me if that's because the younger people have been attending Zoom school, and attending all of these sorts of meetings, where they're not expected to contribute, and so they're keeping themselves on mute a lot of times for that, or whether there's a sense of like, "This doesn't feel like my home turf, and I don't feel like I have the right to stay unmuted, because I don't feel like I have that much to contribute to this conversation in this social setting." I think you'd need to do a certain amount of virtual field work with younger people who are in a group conversation, where it's all 19-year-olds, and they're all friends with each other. And at that point, does everybody mute, or people take it in terms, depending on social status?

Because I do think if you had a group of like 40-50 something professors, who were all very senior, they probably... none of them would mute. But I don't know exactly what the emergent behavior is there among younger people who are used to muting a lot more proactively, maybe have a keyboard shortcut set up for muting. I'm not quite sure where that behavior's going, but I think it's interesting.

DEVON: That's fascinating. Because I'd actually always assumed that it was just people who have spent more time in meetings, in general, sort of... This is now me putting my own color on, and just how I've always thought of it, which is like, they know that muting when you're not talking is a good idea, because something could be making a noise in the background at any time, or your audio might be weird and creating some feedback that you don't realize. So it's just generally good to be muted, unless you're saying something. People who haven't spent as much time on Zoom, who I think correlate with older people, but it's not always true, wouldn't know that, and so that's why they don't, and then they have like a weird echo in the background, or something like that.

GRETCHEN: Right. But then, if they do come you, they're not as good at taking themselves off mute. So there's also the sort of, you're on mute, you're on mute thing. So it's what reflexes you're used to having. But often, if you're in a conversation... So if you're in a group of like eight people, and sort of... If you have somebody who's like chairing a meeting with eight people, they'll probably stay off mute for most of the time, because they're intervening after somebody else was talking. Maybe they're muting if somebody else is giving a report or something, but they might stay off mute the whole time, even if, in another context, they would mute themselves.

DEVON: Right.

GRETCHEN: So that's partly a social status thing, where if you have the status of chair of the meeting, you have the right to stay off mute for more time than somebody who's just delivering a few reports or asking a few questions here and there.

DEVON: Yeah.

GRETCHEN: I think it's something where the norms are still in flux. So I think it's a combination of these things, and in another five years or so, maybe we'll be able to disentangle them a little bit more.

DEVON: There's also other correlations, where if you're running a conference, let's say, you've probably organized your whole day and told your spouse and whatever to like, "Please don't come in the room, because I'm running this thing, it's important." Versus, if you're just a participant, you also might be like in your pajamas, cooking lunch, while you listen to the lecture, or whatever it is. So there's other things going on in there, too. But it's really very multifaceted, and I hadn't thought about the status aspect before.

GRETCHEN: There's also a status aspect in terms of, how likely are you to have a separate office space in your house, that you can guarantee is going to be relatively quiet? Whereas if you're a student, maybe you're living with your parents, maybe you have like three roommates, and you're like in a corner of your bedroom, or something, and you're in your kitchen, and there's other people in your house going to and fro. The more money you have, basically, the more potential you have to control your surroundings in an audio sense, in an audio meeting.

The more money you have, basically, the more potential you have to control your surroundings in an audio sense, in an audio meeting.

So, like putting yourself on mute, because you're worried about what your background sounds are going to be is also a factor of like, do you live in a tiny apartment with three roommates, who are going to potentially make a lot of noise, or do you live in a big house in the countryside, and you have a whole office by yourself and it's not going to be loud?

DEVON: Hmm. Yeah. Yeah. That is really interesting. Yeah, there's so there's so many dynamics that go in there. And because it's so multifaceted, it's also very easy to pull the wrong interpretation out. You could be like, "Oh, this 19-year-old kid has tons of sound in the background. They're not taking this conversation seriously." When no, they just don't have that much money, so they don't have... they maybe are still in college, they don't have any income, and so they don't have a place to do it, and this is actually the quietest part of their house, and they were really worried about this for 30 minutes before the call.

GRETCHEN: Right. Maybe they're living with their whole extended family, they've got kids or something going around. There's so many factors that can be in someone's life that taking it as a sign of like, "Oh, we could just do this seriously..." I think this comes into some of the discussion about like, going back to the office, and to what extent people like that, or don't like that. Yeah, if you have a whole home office space, that's a whole room that you can set up with an extra monitor, and all of this stuff, that's a very different prospects from like, "Oh, I don't want to head back to the office, I've got this nice office space," from, "Okay, well, I've been doing work from my bed for the whole pandemic, because that's literally the only space that's available." So what are people dealing with, in contrast to being able to work from somewhere that isn't their home?

DEVON: That's a great point, it seems like a common theme of a lot of your work is that, we should just be more generous when we interpret what people mean. Sometimes it's not well deserved, because sometimes people are actually just jerks. But most of the time, they're just doing the best they can. And just because they said something in a way that you interpreted one way... Try to think of the best possible interpretation.

GRETCHEN: I think part of the reason why I try to focus on that message, which, yes, people can be jerks, is also that there's so much discourse around language that talks about it in terms of, well, if this person doesn't do this one thing that I was taught to do in my grade 10 English class, then they're wrong, and they should be not listened to for the rest of their life. And it's like, really? Can we not think of better ways to interact with each other as humans?

If you're using language as an excuse to pin jerkish behavior on, just stop and think about what you're doing with your life, and think about your life choices, and what led you to that place, and what you could be doing differently. There are lots of different ways of interacting with the world. If someone is genuinely using language to be a jerk, they're also doing other things that are jerkish. They're not listening to people's boundaries, they're not listening to people's consent, they're overriding people's personal space, and lots of these types of things.

But there are lots of ways in which humans just grow up slightly differently, they have slightly different sets of norms, whether in terms of turn taking, or in terms of... I get a lot of people angsting, about emails, and how to do emails, and worrying about cliché, stock, social phrases that people use in emails, just circling back, or just bumping this onto your radar again, or just wanting to pick your brain, all of these sorts of stock phrases people use in emails, because they show that you're part of a social group.

I get a lot of people angsting, about emails, and how to do emails, and worrying about cliché, stock, social phrases that people use in emails, just circling back, or just bumping this onto your radar again, or just wanting to pick your brain, all of these sorts of stock phrases people use in emails, because they show that you're part of a social group.

I spend a lot of time trying to convince people that, especially if you're a professor, if you're a boss, if you're a hiring person, and you deal with a lot of cold emails from people who you don't know, who are significantly junior to you, it's really unfair to have this sort of hidden curriculum idea of, "I want everybody who emails me to address me as dear so-and-so. And if they say, hi, then they're being unpardonable rude." Well, how are they supposed to know that? Hi is really normal in a lot of other contexts.

Even hay, which gets some flack from people for being like, oh, this one's informal, this is pretty normal in a lot of contexts. There isn't a context in which dear is particularly normal in most other environments in 2021. This is this thing that we use in other social media platforms, you don't call people dear. I hear from a lot of younger people that they're resistant to the advice. They've seen the advice to address letters and emails as dear, but they're like, "Oh, no, this is uncomfortable. It's weirdly intimate. It will be inappropriate in the same way that addressing my boss, or my professor, or my manager as my darling. We're not on those terms. They're not my dear. Maybe you can say this to your grandma, but why would you call someone that in a work context? Wow, sexual harassment complaint, anybody. Seriously."

That's something that does not occur to people who are from the letter writing generations, when, of course, you'd just begin a letter with dear, because that's what you do. Even though when you pause and look at it, you're like, actually, maybe that is kind of weird.

DEVON: It reminds me of the Hamilton song, "Your Obedient Servant," which is all about us this, and-


DEVON: ... I think it's at the end of some letter, that Alexander Hamilton is writing to one of his other founding fathers or something, is... or to Aaron Burr, I think, and it's-

GRETCHEN: Yeah, to Aaron Burr. The song in the musical is... it's really great, because it's oh, your obedient servant A. Burr, A. Him, but they end this very obedient servant letter correspondence by telling each other to a duel. That's what obedient servanthood.

DEVON: Right. But, I imagine that was the normal way to write back then, even if you hated the person's guts.

GRETCHEN: Right. At the time, that was just a stock social phrase, like saying sincerely, or saying thanks, or best, even if you don't actually particularly want to thank them, or if you're not actually wishing them your best, you're just saying this thing because that's what people say. In the same way that modern day people, if they say bye, or goodbye, they're not actually thinking, God be with you, even though that's what it came from. They're like, "No, I just said bye. I'm not even religious. What is this?"

There are these phrases that get interpreted as phatic, P-H-A-T-I-C, which is this sense of like, the literal meaning has washed out through repeated use, and what's left with is the social meaning of, this is what people mean when they say this particular string of words, even if when you look at the particular string of words, you're like, "What in there actually means that?" It's the same thing with a phrase like, how do you do, where we know what that means at a phatic level. But if you look at the actual words, it's like, how do I do what? How am I doing anything?

DEVON: One difference that I've come across in my day-to-day life is that some cultures use voice messages a lot more than others. So, my Argentinian boyfriend's family uses WhatsApp voice messages all the time, it's the primary way they communicate. Whereas I feel like probably most of my friends and family have probably never used that feature, and maybe they don't even know that it exists, even though they use WhatsApp pretty often.

In some ways, I adore voice messages, they're really nice for getting across more emotional things, or stories, and things where you want to get a little bit more emotional color and tone to it. But they're really very frustrating when someone's sending you the address of their house or something, or trying to schedule dinner, and be like, "We're going to meet at this restaurant at this time." In these cases, I just want to be able to copy and be able to paste it into Google Maps, or whatever. In this case, I-


DEVON: I have a critique of the voice message, because it's like... it feels like it's getting in the way of the purpose of the communication. How would you approach a conversation and be like, "Hey, I love your voice messages, but when you send it so that we can go schedule an airplane flight, or something like that, it makes it a little bit harder." Is that an appropriate thing to do, or am I totally missing the point?

GRETCHEN: One of the things that I try to say on the flip side of encouraging people to be more generous in terms of what we can accept, and what we can tolerate, and what we can hear from other people is also saying that like, if you have a particular need, you can try to communicate that without making it about, this person is wrong, or this thing that someone is doing is wrong, or this thing is wrong. Without making it a sort of essentialist critique that there's a right way or wrong way of doing things, you can make it more about yourself, saying, "When we're talking about a restaurant to meet each other at, one of the things that I really like to do is copy paste the name of that restaurant into Google Maps, so that makes it really easier for me. And I'm always worried, when someone's telling me in audio, that maybe I'm going to misspell it or I'm going to miss hear it.

So it'd be really helpful to me, if in this one case, where I need to copy paste things, maybe you could send that in writing. All of the other cases... I love it when you tell me stories in audio messages." It's not like, "Why do you always do this, and this is wrong," it's, here's this one thing that's really useful for me, in the same way as if you were staying at someone's house, and you wanted to ask for an extra blanket because you were cold that night. You'd put a lot of politeness in that. You'd say, "Oh, do you happen to have another blanket?" Not like, "You keep your husband too cold, and I need another blanket, because you're wrong."

You could do that in a way that takes care of the relationship that you have with that person, while still advocating for a particular need that you have, which maybe as an extra blanket or an extra pillow, or something that would make you more comfortable, that is also not trying to do it in ways that are like... imply terrible things about their housekeeping practices.

DEVON: I like that a lot. That reminds me a lot of... I think it's a nonviolent communication tactic that's like, you should say, I feel, as opposed to, you are.


DEVON: And like, "I feel this way when this happens," as opposed to, "You are a jerk, because you do this all the time."

GRETCHEN: When we're doing things like negotiating how a particular thing is used, maybe they're also thinking, when you send them messages, and you tell them stories or something in text, they're like, "Oh, I wish you'd send it to us in a voice message, because then I would know how she was saying it." So maybe there is some way where you can reflect the strength of both mediums in this context of saying what's useful here. Maybe also look into some of the reasons why someone isn't sending something in text. People have different levels of comfort with typing, especially with touch typing, and typing on a smartphone. I know I have seen older people do like... I didn't think people actually seriously did audio searches. Because for me, I'd always just rather type in a Google search. But I've seen people just talk into their phones, and often, people who are less comfortable touch typing.

DEVON: I actually do a lot of voice to text typing, because I don't like to type with touch typing. And the result of this is that sometimes you'll get the weirdest typos from me. There's probably people who really are like, "Why does Devon do this? She is so inconsiderate. It's really selfish. She's making it easier for herself and harder for me, so rude."

GRETCHEN: There are lots of ways of interacting with people. There are other things that I do that are making things easier for me, rather than for somebody else. We're at a stage, which is an unfortunate stage, where I think a lot of people are realizing that like, hey, when you are criticizing people's language, they do take it as a personal attack, and there is so much baggage around language being used as a tool for elitism.

We're at a stage, which is an unfortunate stage, where I think a lot of people are realizing that like, hey, when you are criticizing people's language, they do take it as a personal attack, and there is so much baggage around language being used as a tool for elitism.

So even if you say something innocuous about someone's language... I really like observing people's languages, I'm like, "Oh, what an interesting word you have there, what an interesting vowel you said there," and people still take that as a criticism, because we're at a point and a society where so much of linguistic commentary has been criticism, and has been... Or if there are two or three pronunciations of a given word, people will be like, "Which one's right?" And it's like, that's just not a very interesting question.

It can be the case that one of them is more characteristic of one region, or another region, or one of them is more characteristic of one age group versus another age group, or one of them is more characteristic of people who learned words through reading, versus people who learn to read through speaking, which I think is pretty relatable to a lot of people. But that doesn't mean that there has to be a single right answer.

There's languages, this richly textured fabric, that has lots of different variation within it, it's not just this one, like, here's this one thread that everybody has to climb onto. So it's hard sometimes for people to hear messages around like, "Oh, I didn't understand you when you said that, can you say it again," without hearing that as criticism, and just hearing that as, "Sorry, it was loud here, and I didn't didn't catch that." Or, "Oh, how interesting. I'd say that word differently." But that doesn't mean that either of us has to be wrong.

DEVON: Yeah, that's very interesting, that history of the critique so often being based in judgment and class, resulting in future... not critiques, future comments, being perceived in that light, as well, even though they are not critiques at all, they're neutral, or maybe even positive.

GRETCHEN: It's like how there are a lot of topics where people have sensitivity, because they've gone through a whole life of hearing negative messages about what they're eating, or what they're wearing, or something like that. And if you hear like one more thing about your body or your culture, or something like that, that you've been criticized a lot for, even if that one thing feels innocuous, it's the straw on the camel's back, that you have this reaction to, that's not just based on the one thing, it's based on the whole lifetime of other things you've also heard in conjunction with it.

DEVON: Yeah, definitely. It reminds me a little bit of this thing I've heard of called the difference between ask culture and guess culture. Are you familiar with it?

GRETCHEN: Yeah, yeah, yeah.

DEVON: Yeah. So for people listening, ask culture is about people who are very comfortable just asking for something, and they say, "And." If you say no, that's fine. They just said it, and they asked it, and now you move on. And then guess culture is a little bit more of like, you would not ask for something, unless you think the answer will be yes. Classic clash of culture here would be that an ask person asks someone from a guess culture background, "Hey, I'm in town, can I stay at your house this weekend?" And the guess culture person's like, "Wow, that's so rude. They're barging in. They think that they can just stay at my house whenever they want. I don't even know this person very well."

DEVON: The ask person would be totally fine with just no, and they'd be like, "Okay, cool, I'll find a hotel," or whatever it is. Neither of them is entirely wrong in either way, it's just that the expectation is what matters, and what they're used to. Certain types of communication might be more effective for certain types of goals, but it's not like one is superior to the other in every way.

GRETCHEN: Yeah, exactly. It depends on what expectation someone's coming in with. Especially for guess culture, which often thrives in very densely networked communities, where people have been a part of the community for a long time... because you have to have a very sophisticated idea of what you could reasonably guess, which requires a dense amount of shared cultural context. Whereas in environments where a lot of people have arrived from multiple different cultural contexts, you end up veering more towards ask culture, because you can't assume that people are sharing the same set of guesses at all.

DEVON: People who identify as extremely online seem to really relish the idea that internet communication evolves quickly, and there's this implication that it evolves more quickly than offline language. To me, that feels pretty accurate, but I'm curious, is there research that compares the speed of evolution of online versus offline languages?

GRETCHEN: I have been looking for research on this topic for quite a few years now, and I don't... I haven't managed to find any that's that unambiguous. What I can say is that there is a sense in which particular words or particular expressions can go viral online, in the same way that a funny video or whatever, can go viral, and lots of people can see them, and lots people can can witness them. Now, whether that leads to durable language change, or just a, here's this thing everyone's talking about for this week, and then it's over, is a different question.

So on the one hand, individual words, or individual phrases, or something, can spread without the need of an intermediary. You don't need to get approval from the New York Times, in order for your TikTok to become viral and lots of people to see it. You do end up, often, getting news aggregator sites posting like, here's the Vox explainer of this word. So the media has this role, as well, of bringing it to yet another audience. The question of whether something is actually likely to get picked up by people, or if it's actually likely to become a durable part of people's vocabulary is a more complicated question, because people tend to get exposed to new items of vocabulary from people who are outside of their main social group.

So, you know this concept of strong ties versus weak ties? This is the idea from, I think, a sociologist called Granovetter, where you... some people that you know are strong ties, these are people that you know well, and that you often have lots of mutual friends in common with. Other people or weak ties, who you maybe don't know quite as well, and you also don't have a lot of mutual friends with.

So, if you are a lawyer, and you have lots of other lawyer friends, and they all know each other and come from the same circles, and you also have one friend who's an artist, that's your bridge across to the artist group. And then your artist friend, presumably, also knows lots of other artists. But for you, maybe they're the only artists you know, maybe you're the only lawyer they know, and you're making that weak tie bridge between social groups. The weak tie bridges are really interesting, and they're really important when it comes to diffusion of new ideas, and new words, and other types of new knowledge.

So, weak ties are more likely to lead to a new job, because your friends probably already know the same hiring opportunities that you're aware of, whereas a weak tie, somebody who's in a completely different social group, most of the time, but they know you, is more likely to have access to this whole other social groups, potential for maybe someone's looking for somebody in your exact position.

Weak ties and strong ties also have this really interesting role in language change. So, weak ties are more likely to introduce a new word to a community, because they have this exposure to whatever other social group is potentially using that word. But they're also... If you just hear a word from a weak tie, you're less likely to pick it up and actually use it, than if you also hear it from strong ties.

So, there's a study that was done on this on Twitter, actually, because on Twitter, you can see who follows who, which is a great gift for social networking researchers. Because trying to figure out who knows who was actually really hard before internet studies. I don't know if we realize how hard that was. You do these studies, where you'd have to get... You go into like a high school, and you'd ask each student to list five friends or 10 friends, and then you'd draw this map of who knew who, based on which friends people listed. But of course, you wouldn't be listening to all of your friends the way you can on a social media site, you'd just be picking like 10 people, which would be a feasible number of people to crunch, from a data perspective, before we had much computing power. It's really changed a lot.

So, in this Twitter study that I'm thinking of, they found that people were most... They could also see which... When a new word was being introduced to spreading, who you're most likely see a word from. So you're more likely to see a word first from somebody who's a weak tie, who had different social networks to you, primarily. But you're more likely to actually start using the word yourself if you'd seen it from a strong tie.

So you're more likely to see a word first from somebody who's a weak tie, who had different social networks to you, primarily. But you're more likely to actually start using the word yourself if you'd seen it from a strong tie.

So you're more likely to see a word first from somebody who's a weak tie, who had different social networks to you, primarily. But you're more likely to actually start using the word yourself if you'd seen it from a strong tie.

Just something becoming viral doesn't necessarily mean that I'm going to use this particular word with my friends, and less I also think my friends have also seen this viral video, and that we all have this shared context of, we've all seen it. And then the other thing was this effect of, have people seen this word or not on Twitter, and who they were following, and stuff like that. That only held for words that seem to take advantage of the online medium, somehow, or the written medium, somehow.

So things like acronyms, things like phonetic re-spellings of words. So this is stuff like gonna, spelled G-O-N-N-A, although gonna has achieved full saturation, though that one's not spreading. But I think the example they had in the study was suttin, S-U-T-T-I-N, for something, which hasn't achieved full saturation. So that can spread. Those are things that are very characteristic of the written medium, which is what you're seeing if you're on Twitter.

Whereas other words, words that are just as easily pronounceable out loud, versus written online, those didn't seem to have this direct correlation between whether or not people were seeing them on Twitter, which suggests that maybe we're missing like half the data on those words, because people were also getting exposed to those offline.

There's this interesting effect. I have a personal example on the, how do you get people to adopt a word perspective. Earlier this year, I ran an online conference about linguistics communications of people who do linguistic communication online, and I thought, "Well, we're still in a pandemic, people are attending online conferences. I'd like to do one that's more of a model for how you could do a really good online conference that has a lot of social interaction, and also brings together a group people that hasn't historically had their own conference." This was people who are doing linguistics communicators, and people have linguistics blogs, or YouTube channels, or podcasts, or all of these sorts of things. And there were about 100 of us.

I called this style lingcomm, which is based off scicomm, for science communication, you have linguistics communication. I've been using the word lingcomm for quite a while, for several years, since 2017, at least, in various contexts with other types of workshops I was running, and stuff like that. But it seemed to be that having everybody at this conference together was what made lingcomm escape from its initial creator, and now I just see people using it on Twitter, they're not even talking to me, they're not talking about me, they're just talking about their own thing, and they're talking about it as this existing phenomenon that everyone thinks exists.

I think it was partly like, not just having one person, who they knew, saying this word, but being at a conference, where everybody else that was at the conference was also using this word, and was also... knew what it meant. And so you didn't just have one person who knows what this word means, now they knew that 100 people knew what this word means, and that was enough to say, "Oh, I feel like if I adopt this word, everyone else is going to know what I mean at the same time." It's not just like one influencer trying... or me trying to be an influencer and trying to say, "Hey, this would be a good word for this thing."

DEVON: Right. Because words are only useful if people understand what they mean. So, you could hear a word and be like, "Oh, that's a great word, I should start using it." But then be like, "Well, people might not know what I mean, so I'm going to hold it back." Versus, if you have a group of 100 people, or whatever the number is, who are using it, then you're like, "Okay, I have confidence that this will be understood."

GRETCHEN: Exactly. And it's such an interesting example of how language can exist both in our heads, which it does... Even if you take away all of the other English speakers, I still speak English, and it's still there, it's still in me. But at the same time, it also exists in this community level, where... Part of the things that are... decisions that are behind my speaking English, or choosing which particular words to use, or choosing which new words to adopt are based on what I think other people are going to understand, or what I think other people are going to use around me.

DEVON: You are a resident linguist at Wired Magazine, and I think this is a pretty uncommon position. I don't think I've ever worked at a company that has a resident linguist. If you were the resident linguist at, say, Facebook, or Twitter, or any of these other social networking sites, how would you try to get the company to change the way the website is designed?

GRETCHEN: Oh, boy, that's a good question. Yeah. Well, so in the context of being a resident linguist at Wired, the first site that gave me the title, resident linguist, was The Toast. Which was... RIP. Was a very eclectic news/blogging site that had a really good run for a few years, and I wrote some very weird pieces for them, and they were like, "Yeah, this is Gretchen, she's our resident linguist." And so when I started talking with wired about writing that column, I said, "If we ask The Toast, if they don't mind if we borrow their resident linguist title. I think it would work well here." And everybody thought that was a good idea.

Talking about social media platforms is such a big question in terms of what you do. One thing that I think is interesting, that could be brought in from a more linguistic perspective is thinking about, what are some of the naive things that people believe about language, they haven't even really examined in their belief about language, but they just assume that that's how things happen, because that's how it works for them?

What are some of the naive things that people believe about language, they haven't even really examined in their belief about language, but they just assume that that's how things happen, because that's how it works for them?

So one of these is name policies. There are a bunch of trans people who are currently working on trying to get Google Scholar to let people change their names, which it doesn't currently let you do. There are lots of reasons why both trans people and sis people might want to change their name at some point in their life, and they want their publication record to keep following them. This is something that I think if you have a more nuanced understanding of names, and how they work, and that people do change their names, this is the kind of thing that... From my understanding, it's a very early initial coding decision to index people's records to their actual name, that has caused... made it really difficult to try to change that down the line, because it isn't linked to some sort of like a string of numbers, or something.

So, that kind of thing, or things in terms of... There are 7,000 languages in the world, and most tech platforms support a tiny, tiny subset of that. It's not as straightforward to say, "Okay, well, we'll just flip the switch, and we'll enable all of them," because for a lot of languages, maybe there isn't a dictionary, even, available that language, or if that dictionary exists, it was made in the '70s, and it's pretty incomplete.

How can you take a broader perspective that isn't limited to just this... here's this Anglo perspective that I've taken in my in my particular life, but how can you take a broader perspective to make tech platforms, in general, more welcoming for a variety of languages and a variety of experiences?

How can you take a broader perspective that isn't limited to just this... here's this Anglo perspective that I've taken in my in my particular life, but how can you take a broader perspective to make tech platforms, in general, more welcoming for a variety of languages and a variety of experiences?

DEVON: I actually had the interesting experience recently traveling to Bulgaria, where I was expecting a lot of people to not speak much English, and turning out that actually, they do speak a lot of English, and it felt almost more than the number of English speakers in Germany. I'm not actually sure if this is accurate, but anecdotally, it felt true. I asked them why, and they said, "Oh, it's because Bulgarian is a really small language. So, for us to like, be able to do most things, besides talk to our family and people who live in our town, we need to be able to speak English." Versus German, which is a very large language, and is supported by most platforms, and there's a lot of books and stuff written in that. That was a really interesting thing.

GRETCHEN: There are multiple German speaking countries. You can go to Austria, you can go to Switzerland, or something and you're like, "Oh, yeah, I could still be in German." I think this is true in Icelandic, as well. In Iceland, there's a lot of Icelandic youth that are like, "Oh, yeah, well, we'll use English, because that gives us access to all of the social social platforms and information and this sort of stuff." It's a tricky question. People should have the right to access stuff in their own language, and it's this position of us as English speakers, we're like, "Oh, yeah, everything's just available in my language. That's just how it is." And you're like, "Well, that's not true for everybody."

DEVON: Even for very large languages. My boyfriend's a software engineer, and he is from Argentina. But there's not that much software content in Spanish. It is still a lot more than probably almost any other language, besides English, but he still... If he wants to find a really technical document about some parse or whatever, he's going to read it in English, he's not going to read it in Spanish.

GRETCHEN: Yeah. There's an interesting sense in which most programming languages are still based on English, even if they're based on this highly stylized and formalized version of English. So if you take a language that has something like an if then statement, the word if and the word then are going to be in English. There are some languages... So I speak French, I live in Montreal, and I was thinking about, okay, so if, in French, is C, but there isn't actually, really, a particularly good equivalent to then in the way that you'd want to do an if then statement in French. You just use C on the one half of the sentence.

You can do this in English. Like, you can say, if it rains, I will bring my umbrella. You can also say, if it rains, then I will bring my umbrella. But you don't have to put the then in English. In French, you would almost always not put the then in, and you would just say like [French example phrase]. Like, you would not put another word in. You could put "donk" or something, which was the thing that some French friends and I came up with, is probably the closest word there, but there isn't... it's not as obvious a pair in another language.

Even one that's had a lot of contact with English throughout its history, mutual contacts throughout its history, there is an... even in something that's a fairly obvious bit of structure, doesn't necessarily always have a direct correlate. If you're learning a programming language, as an English speaker, you have all of these nice little easy mnemonics have an individual thing that begins with the same letter, there stands for variable, and you're like, "Oh, I know the word variable, that's a word I have." Whereas if you're trying to code, and you speak a different language, you just have to learn this arbitrary three-character string, and you don't have an obvious memory pack to put it on. You might even need to learn this string in a script that you don't know particularly well.

If you are more comfortable in Cyrillic, or in Japanese Hiragana, and Katakana, and Kanji, and so on, you might be learning a new script at the same time. If you say this to a lot of programmers, they get very defensive. I got more angry emails from that Wired article that I got from anything else I've written. And they're like, "Well, of course, you should... But that's because you should just learn English to learn how to code anyway, because you're going to need it to learn the documentation and the help files, and so on."

And I'm like, "Cool. So what you're saying is that the situation is actually even worse." But also, this is a lot like the arguments that people used in medieval Europe to justify why everyone needed to learn Latin. Because if you wanted to access the technology of writing... Writing as a technology. It's not natural to humans, it's something that got invented, and has gotten forgotten, and has gotten adapted.

Writing is a technology, and there was a period in which the vernacular languages in Europe weren't being written, or weren't being written very much. So if you wanted to access writing and written things, and writing things down, you first needed to learn, and at the same time, another language in which to use the technology of writing. Nowadays, people tend to... especially people, without a very historical view of education, tend to look back, like, "Why were they spending all this time learning Latin? Like, jeez, guys, you could have just done it in your own language, honestly."

But it's for very similar reasons as to why people around the world spend a lot of time learning English. Especially like, if you want to program you need to learn English, if you want to access this particular type of technology, it currently takes place in, primarily, this one language, in the same way that Latin was where a lot of the books and learning and historical materials and religious materials, and so on, existed, because that's what people knew how to write in. This idea that you could write in multiple languages was still something that was being developed in people's heads.

So that seemed like a very foreign perspective to us, if you look at it that way, because we're used to the idea that multiple languages can be written down now, but we're still not yet used to the idea that you could potentially code in multiple languages, or you could use multiple languages as a way of accessing those types of technological tools.

So that seemed like a very foreign perspective to us, if you look at it that way, because we're used to the idea that multiple languages can be written down now, but we're still not yet used to the idea that you could potentially code in multiple languages, or you could use multiple languages as a way of accessing those types of technological tools.

DEVON: Yeah. Something that makes a lot of this tricky is that there are really big returns to scale. So it will... Let's say there is a programming language written in Bulgarian, with the Cyrillic alphabet, and all of that. There's just not that many Bulgarians. They're going to end up with just a lot less content about whatever it is, which is a fundamental imbalance. That doesn't mean that we shouldn't we do it, but it ends up sort of... Over time, it just gets even more weighted in favor of the most dominant languages, because there's this network effect, where as the number of speakers and writers in that language gets bigger, the value of speaking that language increases, as well. So is that a fundamental problem, or can we get... Is that an oversimplification?

GRETCHEN: It's an interesting problem. There have been a handful of attempts to write programming languages based on other languages. I think it's Python in Chinese, that someone's done, but it might be Java in Chinese, I forget which one. So there have been various attempts to do this. There are also a lot of these very hilarious attempts to write programming languages based on other things. So there is a programming language that is entirely usable by Pikachus, where all of the units of the programming language like, P, Pika, Pikachu, and all you need is binary, ones and zeros, and you can ultimately represent everything.

So you have a few more bits than binary, because you have Pi, Ka, Chu, and Pika, and so on, and then you can just Make this extremely impractical and esoteric art lang that's not usable for a lot of programming purposes, but it has this stunt aspect to it. But, the fact that this is sort of at a similar level of feasibility to programming languages, in actual languages that are spoken by millions of humans, is this bizarre statement of priorities.

But I think there is this tension of like, the most useful thing of the programming language is that you can google it, and you can find snippets of code on Stack Exchange, you can copy and paste them into the thing you're trying to make. So how do you do that? I think one potential option is, because programming languages have a controlled vocabulary, so maybe they only have like 20 words or something, and they're all very controlled, this actually makes them potentially really easy targets for machine translation in a way that natural languages can never compare to.

Because if you're making a translation of a programming language, and, say, you're taking if and you're translating it to C. If always translates to C, it never translates to anything else, it never... There's no context dependency here, there's no like, well, in certain pragmatic situations, it would actually be rude, to, blah, blah, blah. No, the computer doesn't care. Computers are, in fact, good at not caring.

So, there have been a few programming languages, I think one of them is Scratch, and another one is Logo that had been designed for children, and they have like, "Oh, this programming language actually exists in 12 different localized versions in different languages." So the kids can learn to program in their own language. The idea is, eventually, they graduate to the serious programming languages in English, which is maybe also an interesting statement of priorities. But at the very least, it's giving them this sort of... it's a single programming language on the back end, but it's translatable in the user end to a bunch of different programming languages, so kids can learn whichever one they're more comfortable in.

You don't have to reinvent, "Oh, we'll create this new version of Java that does its own thing, and is usable for similar purposes, but it's not entirely the same." You could actually make an implementation of the same programming language that just uses different keywords for variables and things.

DEVON: Right, because you would just literally replace the letters with other letters, or other images, or whatever it is, and it's very straightforward.

GRETCHEN: It could be very straightforward. There are levels of complexity more than that. Because if you have a word that means... So say you have a word, like "si," which is actually a good example, which means if, in some languages, it also means yes, in other languages.

DEVON: Oh, no.

GRETCHEN: Right. There are some languages in which it's... it might even be a synonym, and there's some languages where it's synonymous for one, and "si" for the other and things like that. So you have to know like, is this the Spanish version, or is this the Italian version, or is this the French version? Because we need to know whether the "si" means if, or the "si" means yes. There is a potential for confusion. It's not quite as easy as just pop it all into Google Translate. But computers can do a lot of things and I actually don't think this is one of the things they would struggle with as much as many other things we're trying to get computers to do, if people thought it was a priority.

DEVON: It sounds relatively straightforward. Like if you said, let's translate all of Python's core library into Mandarin, as opposed to, let's translate this podcast into Mandarin, the former sounds probably much more straightforward than the latter, I would guess.

GRETCHEN: Right. Once you did it, you'll be done. It wouldn't-

DEVON: Right.

GRETCHEN: ... be as much of a, okay, we translate this podcast episode, and we've got to translate the next podcast episode, it'll be a completely different task. You'll be like, "This is a controlled vocabulary, it's a small number of words, we have to maybe have a few debates about how we're going to deal with some of the edge cases. But then once we've done it, it's done." So I don't think it's not doable, but I think it's... There's a question of will for who would do it, because the people who currently know how to program are people who are currently comfortable in English. But that's not true of the whole world.

There's a question of will for who would do it, because the people who currently know how to program are people who are currently comfortable in English. But that's not true of the whole world.

DEVON: So we're coming up on our time. So let's wrap things up with one last question. Why do I feel compelled to wave at the end of Zoom calls?

GRETCHEN: I think waving is actually a really elegant solution to this sort of coordination problem at the end of Zoom calls. Because Zoom call really dumps you out into the world really fast. You're there, and then... You're in the call, and then you're back in your home, at your desk, and there isn't really this sort of transitional moment, where in a physical meeting, you have this sort of, "Okay, well, let's pick our papers up, close our laptops, put all our stuff back in our bags, stand up from the table, finish that water glass, walk towards the door of the meeting room, go through the door, go in the hallway," and you have all these sorts of moments in which you can be like, "Oh, wait, just this one more thing," or adjust to the idea that you're leaving.

Whereas in a Zoom call, when you hit leave, you're out of it, and you're gone. So you want to make sure that you're on the same page as everybody else who's in the call, that we've actually all agreed that it's leaving time now. I think a really elegant way of doing that is you can look at the screen, and you can see everybody is waving, and you can be like, "Okay, we've come to consensus that leaving is what we're doing now, and so we're waving at the end."

DEVON: Right, right. That makes sense. So people don't feel like it's just dropped off.

GRETCHEN: Right, we do this in phone conversations, to some extent, too. Not with the waving, obviously, because you can't wave on the phone as much. Although, interestingly, people do gesture on the phone, even though no one can see it. And the idea is that gesture, in addition to being communicative, sometimes also seems to help with cognition, helps you get thoughts out. But that's neither here nor there.

On a phone call, what you'll see people do instead... And this has been well studied by linguists, unlike Zoom calls, which I'm sure there are a bunch of linguists who are currently working on Zoom calls, and I can't wait to read their papers in like two to three years, once they finally grind their way through the academic publishing system. So you have this process in a phone conversation of being like, "Okay, well, yep, all right. Well, great to talk to you." "Yeah, you too." "Okay." "Okay. Bye." And this is familiar, right?


GRETCHEN: And you have to have a bunch of these semantically vacuous exchanges, turns, where you're not actually conveying anything, where you're just coordinating around the idea that you want to be ending the call. So that it's not as disruptive when you just hang up at the end, whereas like in a TV show style phone call, where some of you are like, "The burgers are here," and then they'll just hang up, or you're like, "Nobody does this."

DEVON: Right. But it's also not very interesting to watch in a TV show-

GRETCHEN: Exactly. It's really not interesting to watch in TV show, it's very understandable, you cut it out. But if you listen to real life phone conversation, you do have these semantically vacuous turns that are just serving this function of coordinating the idea that we're going to be ending the call now, because... It's not that you don't see this in a physical environment. If you've been at someone's house, and you get out... Say, you're in the living room, and you're hanging out or whatever, and then you're like, "Oh, well, it's getting late," and so you have a few turns of like, "Yeah, better let you go, better let you get to bed," and you walk to the front door, you get your coat, you're standing there, maybe you're putting your shoes on, and you're having some of those conversations there.

Sometimes this leads into, oops, we're accidentally standing at the door, holding on to our coats, and we've actually had another half hour conversation. Sometimes this is just the sort of... or you start talking about the mundane logistics of, are you going to take the metro, or are you going to take the subway? How are you getting back? You can have the logistics conversation, but you do have those sort of transitional moments in other spaces, and I think it's really not surprising that we want a transitional moment like that in a video call, as well.

DEVON: I'm extremely tempted to just stop this call and jump off and see how are you react, but I'm not going to do that-

GRETCHEN: Extremely rude.

DEVON: ... to you. This has been a really fun conversation, Gretchen. Thank you for coming on Tools & Craft.

GRETCHEN: Yeah, thank you for having me. This has been fun.

Share this post

Try it now

Get going on web or desktop

We also have Mac & Windows apps to match.

We also have iOS & Android apps to match.

Web app

Desktop app