The Science Of Analysing Conversations - Elizabeth Stokoe

1. 0:00 – 0:23

   Intro

   1. ESElizabeth Stokoe0:00

      Um's and ah's are quite often thought of as errors that you want to get rid of. They're very commonly littered through our talk, and it's quite difficult to totally erase them. But what they're doing is quite a specific thing. So sometimes they're doing, "I'm thinking," "I'm searching for a word." Sometimes they are showing difficulty in putting this thing together. They're showing an orientation to delicacy. They're doing lots of different things in talk.

   2. CWChris Williamson0:22

      (wind blowing)

2. 0:23 – 6:46

   Why You Shouldn’t Ask People How They Are

   1. CWChris Williamson0:23

      What's your problem with asking people, "How are you?"

   2. ESElizabeth Stokoe0:27

      (laughs) I have a problem with people how they are. But it is a very interesting thing to examine as it unfolds in real conversations, in real time, in different settings.

   3. CWChris Williamson0:39

      Why?

   4. ESElizabeth Stokoe0:42

      Hmm. So, "Hi, how are you?" "Fine, how are you?" is a really common way in which conversations start. It's a kind of no problem, very mundane, routine way in which conversations typically start between people who know each other fairly well. Um, and those how are yous can sound like they're just fiddler talk, that they're not really doing anything very much. That people kind of lie in response. They say, "Fine, how are you?" Not, not, you know, "My life is horrendous." They don't do it at that point (laughs) in the conversation. And so, the fact is when you see loads of those conversations starting in that way, it can just look like nothing special is happening. But when- once you start to contrast those with other kinds of conversations, and the very start of them, you can start to see that that apparently pointless fiddler talk is actually telling you a lot about the kind of interaction this is, that's about to happen. So for example, how do you convey, "I'm in a rush," something's an emergency. You stop people doing those how are yous and you say immediately, "Oh, did, I just need to check, did you leave the oven on?" Um, I've got some nice examples where people immediately start a conversation, not with, "Hi, how are you?" But, "Chris, Liz, what, what's the deal?" And you can immediately see, hey, there's no how are yous, they're about to have a huge argument. (laughs) Um, and then I've got an, an amazing example where, um, a woman calls 999, um, and in order to convey to the, the dispatcher, somehow my life is in danger, but I need you to hear as, hear that I'm talking to a f- I need, I need the person who might be actually threatening my life in the house to sa- to h- to sort of hear, "Oh, they're talking to a friend." Then you do those, you, you do those how are yous and you hope that the person on the other end of the phone kind of catches on, which they do pretty quickly too. "Oh, this person is pretending to have a conversation with a friend by doing that thing that always happens at the start of a call," which is quite amazing to see. How are you

   5. CWChris Williamson2:29

      Most calls that go to 999 are not somebody asking, "How are you?" and taking their time with stuff.

   6. ESElizabeth Stokoe2:37

      Exactly, yeah. So, so I've got some amazing data in which people are phoning up, um, having to convey to the person that they're talking to, "I'm, I'm in danger," but they can't use those words because the person or the situation that they are in danger from is, is there as well. Um, so how does that happen without using any of the words? People say things like, "I'd like to order a pizza for delivery." How does the call taker hear that as a genuine request for help? Especially since actually one of the tasks that the set of police, um, call takers have to do is figure out, um, from all of, all of the potential nuisance calls that they actually do get, that do sound a bit like that, how do you figure that out? But they can figure it out pretty quickly.

   7. CWChris Williamson3:19

      What is a better question to ask somebody than, "Hi, how are you?"

   8. ESElizabeth Stokoe3:25

      It depends what you, what you want to do. So I, this, this, it's not that it's a bad question, it's just that if it's there at the start of a conversation, then probably it means that there's nothing urgent, nothing pressing, nothing, no conflict about to happen. (laughs) It could also be exploited by people. So I've also studied cold call sales, where you can see people saying things like, "Hi, how are you?" And, and because generally we can recognize someone's voice when they do that kind of thing, if you don't recognize their voice, you're immediately on your guard, 'cause you're like, "I don't recognize their voice even from the start of this conversation." And so then what happens is that you don't get, "Fine, how are you?" You might get, "Fine," and then nothing. (laughs) And if a salesperson's any good, they're hearing, "Oh, this person doesn't want to do small talk with me," and they just cut to the chase. Whereas I've got some really cringe-worthy calls where people still say, "Oh, and how's the weather?" (laughs) And you're like, this person's going, "Uh, listen to me, I don't really want to have this conversation with you." So the how, the presence or not of how are yous are just very interesting at the start, right from the opening sort of few seconds of a conversation you can tell a lot.

   9. CWChris Williamson4:30

      Have you ever looked at those pre-recorded sales calls where it's, "Hi, how are you?" Silence.

   10. ESElizabeth Stokoe4:36

       (laughs) That happens. I've- I've-

   11. CWChris Williamson4:38

       "Yes, it's duh, duh, duh in today," and that sort of, they've obviously tried to create a single-sided conversation with sufficient breaks in between as to hook someone in. I can't believe that that's ever worked.

   12. ESElizabeth Stokoe4:51

       I can't believe it either because people are really quick at figuring out that this isn't a, a human voice at the right pace, at the right pitch. Even voices that are, um, you know, generated through some kind of comp- algorithm or they, or they're just, or, or actors are used to kind of put into some kind of chatbot, you can tell pretty quickly. Um, I don't know if there's, there's a really nice, um, case called Lenny the chat, Lenny the chatbot. Um, and Lenny the chatbot, um, was developed by somebody who basically wanted to stop or, or just trap cold callers to his home, um, on, on a loop of their own kind of making. So, so if someone would phone the house and Lenny the chatbot would answer. And Lenny the chatbot didn't have any kind of algorithms or anything. It was just something like 16 phrases pre-recorded and then shuffled. And it sounded like the voice of an elderly Australian man. And so the person, the cold caller did think that they were talking to an elderly Australian man, and they tried to make sense of what this guy was saying to them. And so, you know, how long could Lenny the chatbot keep someone in a conversation? So that was, that was very nice and people can go and there's endless recordings on, on YouTube. You can go and see how Lenny manages to keep people on, on the line. And people are trying to get out, but they're like, "Oh, I can't really leave him now because he's, he sounds just like this old gentleman." (laughs) ... yeah. (laughs)

   13. CWChris Williamson6:09

       It seems like there's a particular satisfaction on the internet for watching scammers. Not saying that cold calling is the same as scamming, but some people put it into a similar bucket. There's, uh, entire YouTube channels dedicated to, um, really sophisticated computer white hat hackers that reverse the scammer cold calls, and they tell them stuff like, "I know the, the address of the, um, building that they're calling from, where they are," and they sometimes do other stuff to them, play around with their computer. So that's, it doesn't surprise me that Lenny the chatbot is, he's like the, uh, nerfed version.

   14. ESElizabeth Stokoe6:44

       (laughs)

   15. CWChris Williamson6:44

       He's like the kiddies version of that. So

3. 6:46 – 12:13

   How Humans Are Led by Language

   1. CWChris Williamson6:46

      given the fact that you do, uh, conversational analysis for a living, just how much-

   2. ESElizabeth Stokoe6:51

      Hmm.

   3. CWChris Williamson6:51

      ... are humans pushed and pulled around by language?

   4. ESElizabeth Stokoe6:56

      A lot. Um, which makes it sound bad in a way, like we're, we're kind of manipulating each other, uh, all, all the time. But, uh, everything gets done really through interaction. Um, even if we're not just talking about words, we're talking about gestures, and gaze, and, um, all the embodied things that go along with s- with words. And it's the thing that kind of drives everything. We, we ask, we invite, we offer, we get help. We, we do a lot of things in everyday social life and our workplace lives through social interaction. And without being aware sometimes, and, but also of course sometimes really being aware wha- w- if, if you f- have that bristle, bristly feeling that someone's doing something that you're finding difficult but you can't pin it down, sometimes they're the kinds of things that, for me as a conversation analyst, I can transcribe and try to unpack what it feels like when you're really pushing for service when you ... Like, for example, I've studied people calling their GP receptions. Um, and what does it feel like when you, when somehow you're not being offered things that is probably the, the person's job to offer you that thing? What, what does ... And if, because we all feel that sense of like, "God, I f- I feel like a burden. I feel like I'm really pushing you to, to, to give me some service." And you can sort of start to unpack that. So an, a really nice example is going into a cafe, uh, getting your coffee or your tea, and saying, "Do you have wifi?" "Yes, we do." "So, uh, i- can customers use it?" "Yes, they can." This is, this is real. This was a field note sort of scribbled down hastily after going into a cafe. I, um, and, and then eventually, the, the, the cafe very sort of, you know, cool, (laughs) cafe staff owner just sort of points at the code, which is painted in this very arty hipster way, right at the, it's like 12-foot high in the, on the wall, and you get the code. And so you really feel like, "God, I really had to drag that out of you," as though somehow you didn't understand that, "Do you have wifi?" wasn't a request for the code as well as ... It's not just a yes/no question. (laughs) Whereas, you know, the, the much nicer experience, which is also real, also scribbled down as a field note, was going into a cafe. As your tea comes across the desk, the cafe, uh, staff member says, "Oh, by the way, if you need the wifi, here's the code." And what's so nice about that is you feel really great about that person, the cafe, and it didn't involve things like, "Hey, what's your name?" I, uh, it didn't involve all of those more scripted-sounding things that s- seem to have found their way into some types of service encounter. It's actually about, "Maybe you need this thing, uh, and here it is. You don't even have to ask for it."

   5. CWChris Williamson9:32

      It's cutting off at the knees as well, the amount of back and forth that's needed between the person that's serving, presuming-

   6. ESElizabeth Stokoe9:38

      Yeah.

   7. CWChris Williamson9:38

      ... that they need to be efficient and not waste their time having a f- five-minute conversation about the wifi code. Just-

   8. ESElizabeth Stokoe9:43

      Exactly.

   9. CWChris Williamson9:43

      ... presuming that, "Look, if you need the code, it's up on, it's 12 feet on the wall, painted on a piece of driftwood or whatever," um, that, th- that helps.

   10. ESElizabeth Stokoe9:51

       (laughs) Yeah.

   11. CWChris Williamson9:52

       Yeah. Another thing as well I've noticed, it seems like brevity, or at least, um, appreciating people's time, especially in the modern world, because everyone's so distracted by their devices, that their, uh, time to actually do the things that they need to do now has become completely condensed down.

   12. ESElizabeth Stokoe10:07

       Yeah.

   13. CWChris Williamson10:07

       Um, I think that just trying to do things in the minimum, uh, amount of time. For instance, if, if you're doing outreach to people, my advice has always been that the message should be readable within the space of about 20 seconds, especially if the person that you're trying to reach out to is someone that's of high repute, right? Like-

   14. ESElizabeth Stokoe10:24

       Mm-hmm.

   15. CWChris Williamson10:24

       ... you, they are very, very busy. And if you send them a huge essay that they've got to go through, I think that your chances tend to decrease. And that's kind of the same for, you know, even my mum. My mum's a perfectly pleasant person, but she doesn't want to spend forever talking to the checkout cashier at the local shop.

   16. ESElizabeth Stokoe10:40

       Yeah. Yeah, uh, I think the cafe example that I just gave is a nice example of how quickly you can make someone feel a bit warm towards you whilst, and getting, and getting needs, potential needs met, but it's very, very short and very, very efficient. Um, uh, I've also done some work when people are phoning the vet. Um, so this is a nice example of actually listening for what people want in this moment as well, and whether or not you're able to do that. So Case One, someone phones up and says, "How much does it cost to get injections done for my new puppy?" And the vet receptionist, trying to build rapport or, I don't know, you know, just sound lovely on the phone is like, "Oh, what's your puppy called?" And there's th- there's this little delay. And in my world, uh, l- delays of like 0.7 seconds is quite large. And you might get, "Victor." (laughs) Now, that's someone who doesn't want to have a long conversation about the new puppy. But, but the call taker isn't really, she, she's just going for it anyway, "Oh, and, and is it your first puppy?" "Yeah." (laughs) It's like, you should be able to hear by now they just want the price. And in fact, in this particular call, it's almost like the sort of Star Wars credit that this, this sales, well, it's not even a salesperson. The receptionist goes on and on and on, "What we can do for you is ..." And, and in the end, it's, you know, it's going nowhere. Whereas actually, you know, maybe at that point where you say, "Oh, what's your puppy called?" And the person's like, "Oh, well, you know, we ..." Some people are dying to talk about their new puppy. So can you be nimble enough and listen enough to actually think, "Right, this is someone to do that with 'cause they're dying to do it. This is someone who just wants the information."

4. 12:13 – 19:41

   Judging Someone Based on Conversation

   1. CWChris Williamson12:13

      It's interesting thinking about the fact that people's words are one of the most, um, direct insights that we have into their nature as a person. So yeah, actions are important or whatever, you know, like someone saying, "Uh, I hope the funeral was good," whilst not being there, isn't quite the same. But...

   2. ESElizabeth Stokoe12:30

      (laughs)

   3. CWChris Williamson12:31

      ... someone's words and the way that they put them across to you, when we're talking about someone and saying that they're rude, or abrupt, or nice, or kind, or caring, or whatever, that-

   4. ESElizabeth Stokoe12:41

      Mm.

   5. CWChris Williamson12:42

      ... forms a massive amount of our interpretation of that person and their personality overall.

   6. ESElizabeth Stokoe12:48

      Yeah, absolutely. So I have this thing, it's not a serious thing, um, I should say, but, uh, called like the Conversation Ana- Analytic Personality Diagnostic, which is not a serious personality tool. But the point is, that when you're in a, an interaction, a conversation with somebody, you don't sort of give them a questionnaire, check what kind of personality they are on some scale, if, if you believe in that kind of thing. Um, and then on the basis of what they score, they're an extrovert, an introvert, whatever they are, you then take the next turn. You have to be much faster (laughs) than that all the time. And so exactly that, you're, you're kind of, on what basis would you say someone is rude or, or they seem warm, or they were help- you know, it- it's basically what they do. And of course, a lot of what you do is what you say. So we do have this, this sense of, you know, you can talk the talk but can you walk the walk? And actions speak louder than words. But in my world, again, you know, a lot of what we do is what we say, they're the same sorts of things. So, so what is being offensive? A lot of the time it- it's not, you know, you don't do being offensive by punching somebody. (laughs) I mean, that would be offensive, but you- you be offensive through your words. Um, so what you do, uh, is- is basically what a lot of people are using almost all the time to decide what kind of person you are. Yeah.

   7. CWChris Williamson13:59

      I like that reversal of the- of the age-old wisdom of it's not about what you say, it's about what you do. But yeah, you're right.

   8. ESElizabeth Stokoe14:06

      Mm-hmm.

   9. CWChris Williamson14:06

      For the most part, people aren't doing things, they're saying things.

   10. ESElizabeth Stokoe14:09

       Mm.

   11. CWChris Williamson14:09

       And if the broadest area under the curve is people just talking about stuff, making sure that you talk about things in an appropriate manner to get the kind of response that you're looking for and to have the sort of impact that you hope for-

   12. ESElizabeth Stokoe14:19

       Yeah.

   13. CWChris Williamson14:20

       ... is really, really important.

   14. ESElizabeth Stokoe14:21

       Yeah. Absolutely. Yeah. So- so, um, how do you... For example, I've got this- this, one of my categories is the idea of the first mover. Um, so parents can often be first movers, which- and by that I just mean that they take a problematic first turn (laughs) and- and then you can get trapped immediately in a conflict. So things like, "Oh, I don't really like that jumper." (laughs) It's like, wow, you know, if I said that to you, Mum, Dad, you'd be like, "Oh, I'm so offended." You know, and you, and if you call them out on it, they're like, "Oh, I was just joking." So- so, you know, people can be quite challenging in their opening gambits. And then somehow if you call them out on it, you're the problem by being offended rather than the person actually being offensive in the first, in the first instance. Um, so again, you know, lots of examples of that, in email as well, you know, people will write you an email that- that somehow is, you immediately feel annoyed, but you can't say anything because strangely enough, the sort of constraints around the way we interact with each other mean that it's actually quite hard to- to just maybe be as direct back. You have to let a lo- we- we let a lot of things slide, don't we? All, you know, because otherwise you- you'd just be angry all the time. (laughs)

   15. CWChris Williamson15:28

       Is there something about that, is there something about the convention that when somebody sort of is that first mover and maybe decides to take a left turn out of the conversation to say something that's a bit s- that's got a bit of snidey top spin-

   16. ESElizabeth Stokoe15:40

       Mm.

   17. CWChris Williamson15:41

       ... there is something that when you break the fourth wall and you go, "Hang on a second, that's- that's a bit- that's a bit unfair, isn't it?"

   18. ESElizabeth Stokoe15:47

       Mm-hmm.

   19. CWChris Williamson15:48

       And if- especially if you do it in a group and there's other people listening as well, it's like the record needle scratching, and everyone's like, "Ah."

   20. ESElizabeth Stokoe15:55

       Yeah. Yeah.

   21. CWChris Williamson15:55

       In a way that the first mover being a dick didn't.

   22. ESElizabeth Stokoe15:59

       Yeah.

   23. CWChris Williamson15:59

       That's so strange.

   24. ESElizabeth Stokoe16:01

       Yeah. I mean, people's- you know, some people's entitlement to take that first turn on the basis that quite often people won't- won't challenge them on it, um, wh- um, because it's difficult to, it's really difficult to. So everything from- um, so again, going back to the patients calling their doctors, those kinds of conversations, right? Sometimes at the end of those conversations, they- they end in a very orderly way. So in that, "Hi, how are you?" "Fine, how are you?" The ends of conversations, when somebody is calling up as a service user, the service provider provides a service, and then the service user, it's up to the service user to kind of say, "Right, this is over." And they say, "Thanks," or, "Thanks very much." And then the service provider will say, "Thanks," and then, "Bye-bye." So it goes, "Thanks." "Thanks." "Bye." "Bye." Really- really commonly. Now in my lots of analysis of all of these patients calling their GPS, occasionally what will happen is that the- the service provider, the receptionist will have, you know, said something or other and said, you know, "Your prescription will be here." And then they'll say, "Thanks." And they are trying to end the call, but it's not really their place to go first. And what happens very typically after that is then the patient has to sort of jump back into the conversation and say, "Oh, oh, oh. But so- so who am I seeing? So- so when?" And so you've got the burden on the patient to get maybe the confirmation that they were after, but also push past the receptionist trying to end the call. And basically what you see in- in those moments is just the- just the pain- the painfulness of withholding a thank you. So the next time you're unhappy on a telephone call, try not saying thank you at the end. Try- try to not say thank you first, and then if- when the receptionist or the service provider says thanks to end the call, try not saying thank you. You'll feel really horrible. So from- from- from that sense of like, it's very difficult to just not say thank you at the end of a call, all the way through to how do you- how do you be an ally? How do you speak up if someone is being sexist, racist, whatever the ism prejudice might be. And one of the- one of the interesting things, again, I've done a lot of work on what counts as an ism, which is different for different people, and then what do you do in response? And of course people will say what they would say in response to somebody being racist in a room, but actually there's a whole range of things that people do, and some of them are very tiny. So some of them are things like if you're in a meeting and someone says something...... that is problematic in whatever way. You may not say anything in that room, 'cause you, you're not the chair, you, you're, you're a junior. You can't, it's difficult. But what you might do is actually just look at, for a moment at someone, catch the eye of someone else in the room and know that they're seeing things in the same way that you are, so maybe you deal with it later. Um, so, so all the way through to doing a direct challenge and then seeing that ripple effect into the interaction. Um, and actually one of the, one of the most effective things that you can do when somebody is being problematic in lots of different ways is to just wait. Because actually people will quite often back off their own thing or they'll, they'll do what we call self-repair. They'll, they'll sort of notice that people haven't immediately come back with an mm-hmm, and they'll inspect their own thing and say, "Oh, but I didn't mean X," or, "Oh, but of course I didn't mean that other thing." So, and then again, you know, every t- every single time in conversation you're deciding to say something or not say something, say something or not say something. And it's running fast, you know, the conversation's running really quickly. So it's, it's one of the things that is really, I think, useful for people, and I quite often use these sorts of materials in, in workshops, is actually see these sequences unfold where someone says something, everyone in the room is saying, "Oh my God, I can't believe they said that, and this is what I would do." And then you see how it actually unfolds, and sometimes there's quite a gap between what e- people imagine they would do and what actually happens.

5. 19:41 – 30:09

   Insights on Silence in Conversation

   1. ESElizabeth Stokoe19:41

   2. CWChris Williamson19:41

      What are the interesting insights that you've learned around silences and stuff in conversations? That's something that on the podcast I had to get used to very early on that... Uh, there was a famous clip actually, I'm not sure whether you saw it, but you would, you'd be fascinated, you can go back and watch it. Uh, Elon Musk went on Lex Fridman's podcast about six months ago and a clip went round the internet and did about a million plays, and it was a 31-second pause after Lex asked a question and before Elon did. And after a while it becomes comical. They just sat in silence. But Elon, he's asked, "How long is it going to be until we get to Mars?" And Elon sits and thinks and thinks and thinks and thinks. But that ability to sit with silence is something that new podcasters really struggle with, and generally people overall do. So why is silence so uncomfortable and what are some of the interesting things that you've learned from studying silence in conversations?

   3. ESElizabeth Stokoe20:34

      (inhales deeply) So my field of conversation analysis is right from the start, it's, uh, sort of been, you know, r- running for 50 years of, so the thousands and thousands and thousands of conversations in all settings you can imagine have been recorded, transcribed. And we use a very technical system to transcribe which will, you know, pinpoint exactly how long any silence, any gap or pause is. And turns those, "Hi, how are you? Fine, how are you?" You know, you're looking at about 100 to 200 milliseconds, so that- that's quite a pace of interaction. Um, one of the founders of the field, um, said that a silence of one point naught, you know, like a complete, a whole second is- is starting to be quite a delay. And you can even see that when, um, y- you know, so as soon as you see a conversation where there's a two-second delay or three-second delay, they're quite big delays. Now of course depending on whether you are, you can see each other, you're co-present or you're on the phone, or you're writing, which is different, on a, on something like, you know, WhatsApp. But those silences can sometimes be clear about what they're doing. So silence isn't typically an inert thing. You can show someone that you're thinking. (laughs)

   4. CWChris Williamson21:41

      Like Elon did.

   5. ESElizabeth Stokoe21:42

      Yeah. And, but even so, you know, there's, there's a sort of s- there's a, there's a point in time where that just gets too long, and I would say, you know, 31 seconds or something is- is a massive outlier for interaction. So the- the longest silence that I have got in any of my data is police interview with a suspect. It's 11.2 seconds. The police have o- asked a question that's really difficult for the suspect to answer, 'cause almost anything is gonna be incriminating. (laughs) And in the end after 11.2 seconds the suspect says, "I really can't deal with that question actually." (laughs) And so they, they, you know, i- in the end it doesn't work, but... And it's interesting that you mention this because just, just today, um, I was talking with a colleague about another really famous clip that you can go and look online which was Charlie Stayt interviewing Lady Gaga about seven, six or seven years ago. And it's something like, um, he's asking her about Trump and she says she won't talk about it, and then there's this long silence and it's very, it's very aw- awkward to look at. And what's quite interesting is, uh, a- a year later I happened to be doing a talk and it happened to be, the event was sort of hosted by Charlie Stayt and he asked the same question about these silences. And then he s- he told this little story that that silence was actually 16 seconds long and before they broadcast it they cut it because it was so painful for viewers, they didn't want the viewers to have to sit through 16 seconds' worth. Mm-hmm.

   6. CWChris Williamson23:04

      So those long silences are a real breach and, and I suppose the fact that we see them as, we can immediately pick up on that being a breach of, of the sort of normal typical pace at which interaction occurs. You know, we wouldn't see it as a breach if we didn't sort of tacitly know somehow, you know, about a s- about a second is already getting quite long, so six is huge and, you know, 11 is massive and 31 is, you know, (laughs) that's a long, long painful silence.

   7. ESElizabeth Stokoe23:34

      Mm-hmm.

   8. CWChris Williamson23:34

      Why is it painful? What is it about the rhythm of conversation and the breaking of that rhythm that makes us feel like that, ah, t- nails on chalkboard thing?

   9. ESElizabeth Stokoe23:46

      I, I think it's, it's... I mean, I don't, the, the, the why is difficult but what I can say is that, w- w- that, what we see in interaction when we record it and transcribe it is simply that those lengths of delays don't happen, so they're not in our daily experience. So just for a, as a basic thing we're not used to regularly experiencing those length of delays because they just don't really happen. So y- it's like, "What's going on?" Immediately, this is really strange. It feels strange because it is strange (laughs) , quite simply, empirically. You know, when you look at the data it is just weird, yeah, and unusual. And the other thing is that we-... you know, we tend to, uh, want to fill silences as well. So, we- we- we're- we're constantly inspecting everything that everyone says to us as an analysis of what we did previously. So, if I ask you a question and you don't answer for 31 seconds, I'm think- I'm- you're starting to think, "Do I need to redesign the question? Was the question not clear?" Because typically, a silence, you know, it will- it will initiate what we call repair of some kind. So, it might be that I ask you a question, there's a silence, and I immediately redesign the question. Or I issue an invitation and there's a silence, and so I immediately say... Well, so a concrete example, you know, you ask someone, "Do you wanna come for dinner Friday?" There's a silence. So, we're so tuned into it that we immediately pick up that they're not gonna respond within those sort of split seconds and we say, "Or Saturday?" So, we're very attuned to how our interlocutors are responding to the actions that we are producing as they come out, and that's why, like, you're- you're nodding as I'm saying these things. So, you- you know, I almost don't have to finish the sent- you know where I'm going with this action, and that's how we're able to actually be so fast. And that s- you- you- you know, so you- we're preparing our responses, you know, so that they're precision timed within 1/10 of a second. How do we manage to do that? It- it's because of the s- sort of familiarity with a lot of actions that we're actually doing in our interactions, which makes anything a delay. And, you know, a classic, another one is, you know, saying, "I love you." You know? That, you've got to be like really rapidly coming back. (laughs) You know, any kind of delay or, you know, does my-

   10. CWChris Williamson25:55

       You're not allowed a 31-second delay on...

   11. ESElizabeth Stokoe25:57

       ... "does my bum look big in this?" That has to be quick.

   12. CWChris Williamson25:59

       (laughs)

   13. ESElizabeth Stokoe25:59

       You know? Two seconds is- is a problem. (laughs)

   14. CWChris Williamson26:03

       Yeah. Another thing that you see amongst, and this is me reflecting on me a long time ago, uh, new podcasters especially, is that they'll ask a question and then they'll give the guest a menu of options that they can choose from. "So, Liz, you know, why did you get into studying conversational analysis? Was this something that you did as a kid or was it," you know? Was it... And to me-

   15. ESElizabeth Stokoe26:27

       Yeah.

   16. CWChris Williamson26:27

       ... what I used to feel was sort of this visceral discomfort around asking a question and not having the confidence that the question was sufficiently interesting that I could just sit-

   17. ESElizabeth Stokoe26:38

       Yeah.

   18. CWChris Williamson26:38

       ... and just let the other person, uh, marinade in the question. 'Cause sometimes, it's a difficult question. Sometimes, it's maybe a question that they don't often think about or haven't heard before. But then by giving this sort of menu of options, I start to lead them toward... And it's a really dumb idea for podcasters to do, because what you're doing is you're- you're offering, especially if it's an open question, you're offering the guest the opportunity to pick from every single potential question on the planet. And then if you give them two options, they have to either pick one or the other, say why it's not the first or the second one, or somehow break the convention that you've already constrained over and go, "Well, actually it's neither." And very rarely will you ever hear someone say, "It's neither." Most of the time, the guest, unless they're super disagreeable, is just going to kind of take one of your suggested answers for them and then rework that into what they would have said in any case. But it's-

   19. ESElizabeth Stokoe27:30

       Yeah.

   20. CWChris Williamson27:31

       ... it's- it's mostly, in my experience, it's mostly about the comfort of question, stop, silence.

   21. ESElizabeth Stokoe27:37

       Yeah. Uh, uh, and it's interesting that you raise that because there are, you see that in particular settings where it's both surprising and really matters and is consequential. So, if you've got a list of questions that you're gonna ask somebody, in some settings, your list of questions is quite formal. It- it's a diagnostic instrument, for example, or it's a- it's a police interview, or it's somehow a regulated kinda conversation where there's guidance and some standard things, it's a standardized survey, you're meant to ask the same question in the same way over and over again. What researchers in my field show is that it's actually really difficult to, first of all, ask a question that is, even an open, you know, a W-H kinda open-ended sort of question or a- or a yes-no kind of question, um, which, again, a bit like, you know, "Do you have wifi?" That isn't a yes-no question. That requires a bit more. So, so open and cl- you know, this- this idea of open and closed is- is a little bit of a myth, but just sticking to this idea that you can just, let's say, read out a list of questions without embellishing them, without going "um" halfway through, without tweaking them slightly. And so, people are very surprised when they see what they imagine is a standardized instrument, standardized experimental instructions in a laboratory, standardized police interview, whatever it might be, and then you just imagine that- that that's how they op- that's how they work then. There is the law, there is a script, there is a list of questions. But it's, uh, remarkable how the reality of those interactions just doesn't look like that. And so, you can imagine situations where this is- becomes really problematic. So, if it's, for example, um, I've done some research on this with colleagues in Norway. We were looking at how teachers basically do oral communication assessments. And they're meant to ask a question, the same kinds of questions to all the pupils, but sometimes they do exactly what you say. They ask a multiple-unit question, and then the pupil's like, "Well, which bit of that shall I address?" Typically, you address the last bit first and then you may or may not go back to the first bit. But it- but it seems just quite difficult to just stick with this one question without doing a bit more work. It's the same in diagnoses, so, you know, a psychiatric consultation. There's meant to be a standardized question, it'll get a bit of embellishment. Not... For what- for what reason? It's- there's some interactional imperative that makes this- makes it very difficult to sound like a form. Even if you're a police officer in the situation where the law says, "Say it like this," it doesn't quite look like that. (laughs)

6. 30:09 – 38:14

   Uhm’s, Ah’s & ‘Like’

   1. CWChris Williamson30:09

      What are ums and uhs and, uh, you knows used for?

   2. ESElizabeth Stokoe30:16

      Ums and uhs and you knows and okays, they're all different sorts of things. So, they- y- they're not interchangeable, uh, in- in- in a conversation.Um's and ah's are quite often thought of as errors that you wanna get rid of, so, um, (laughs) like that one now-

   3. CWChris Williamson30:35

      (laughs)

   4. ESElizabeth Stokoe30:35

      ... they can, th- but they can, they can, they're very commonly littered through our, our talk, and it's quite difficult to totally erase them. And but what they're doing is quite specific things, so sometimes they're doing, "I'm thinking. I'm searching for a word." Sometimes they are, um, showing difficulty in putting this thing together. They're showing an orientation to delicacy. So, they're doing lots of different things in, in talk. People are critical of others who say um and ah. But when you see a transcript, when people see a, a conversation analyst transcript for the first time, they're quite shocked at how messy it all looks. But that is how we talk, um, so yeah (laughs) . But they're, but they're doing specific things. And I'll just say one thing about um's and ah's which is m- maybe interesting. I said, talked a bit earlier about people phoning the vet. So, I compared real pet owners telephoning the vet with mystery shoppers who don't really have a pet but are phoning up vet practices to report back on how good the service is. So, and, you know, these mystery shoppers, they have quite a lot of power in a way because what they're doing is going back to the organization to say how good people are at answering the phone. And so, my question is, how good is a mystery shopper at passing as an, as a, as a pet owner? Because no one's done that research, so we're just taking it for granted that this mystery shopper can, can pass as a, as a real pet owner. And the thing is, from a distance, they, they totally can. But when you start to look at lots of calls, you see how different they are. So, a, a very gross difference is real pet owners typically phone up to make an appointment. A mystery shopper will phone to say, "How much is the cost of the service?" And then when the receptionist tries to give them some help like, "Do you want me to book you in and sort of help you navigate calendars?" the mystery shopper's like, "Oh, uh, I might call back later," which is really weird. And you're thinking, "What are they, what can they possibly report back to the organization? They gave me a cost of something, that was it." But the um's and ah's are, are so interesting in these, in these calls. So, I- I'm gonna totally ha- make a mess of even saying this, but I'll try and get it right. So, the mystery shopper will typically phone up and say, "How much would it cost to get a dog vaccinated?" A real caller will say, "How much would it cost," if they're asking about cost, "How much would... Uh, I've got a new puppy. How much would it cost to get my dog vaccinated?" So, they don't say, "A dog," they say, "My dog," for a start. But the mystery shopper might say, "How much would it cost for a, a..." Sorry, "How much would it cost for a, um, dog (laughs) to be vaccinated?" Whereas the real caller is more likely to say, "How much would it cost for my dog to be, uh, vaccinated?" So, the uh's are in different places. And what's that about? I mean, I don't know. I don't have enough data to really address the question. But what it shows us is that the um's and ah's aren't randomly splattered everywhere. They are, when you look at enough of them, they're doing particular bits of work in particular ways, and it's quite fascinating if you're someone like me to figure out what that's doing.

   5. CWChris Williamson33:33

      I read a blog post not long ago that was talking about the relative interpretation of people who either used silences or um's and filler words in between on whether they were seen as more or less intelligent, and I can't remember what the outcomes of that study were. Do you, have you ever looked at this stuff?

   6. ESElizabeth Stokoe33:55

      I haven't seen that, but I, I, I can imagine that the outcome was that the people who say um and ah are judged to be less intelligent because that would be the stereotype, but... (laughs)

   7. CWChris Williamson34:03

      I can't remember. I feel like, I feel like it wasn't quite as simple as that. I feel like the results were s- m- more along the lines of what you do, which is pretty much everything's contextual. It's more to do with the flow of the conversation anyway, because somebody that's using silence is actually showing that they have confidence to be able to let that sit.

   8. ESElizabeth Stokoe34:21

      Hmm.

   9. CWChris Williamson34:21

      And that they maybe being more considered with what it is that they say, so I guess it's, it's much more colorful.

   10. ESElizabeth Stokoe34:27

       Yeah.

   11. CWChris Williamson34:27

       Speaking of filler words, why is everybody saying the word like?

   12. ESElizabeth Stokoe34:32

       (laughs) I don't know. It's the mor- it's

   13. CWChris Williamson34:35

       This is... Come on, Elizabeth.

   14. ESElizabeth Stokoe34:36

       Yeah, no, it's-

   15. CWChris Williamson34:36

       This is you. This is right in the middle of your wheelhouse.

   16. ESElizabeth Stokoe34:39

       (laughs)

   17. CWChris Williamson34:39

       Why, why are we cursed with this word?

   18. ESElizabeth Stokoe34:42

       I haven't studied it, but it is something that maybe I should start analyzing. I mean, obviously, people put it in rep- in place of say, um, or, or a, or a, an action, so "She was like," "He was like."

   19. CWChris Williamson34:54

       Hmm.

   20. ESElizabeth Stokoe34:54

       And it quite often comes before a quote, it, before reported speech. It can come before a reported action. So, it's definitely doing some kind of preface to describing the words, conduct, behavior of other people. Um, I don't know why it's so common. It's a generational thing. But, um, but like, like, lots of words, you know, I, I definitely say stuff that my mum doesn't. And, you know, I, I say stuff that my students don't. And we, we, you know, things just be- sort of come in and out of, of common usage, and that's just language. Language is alive. It grows. It changes. Uh, one of my favorite things to do is have a look at, at, uh, if you just Google any word, it will give you a dic- it will give you a dictionary def- definition. But it'll also give you use o- over time. So, for example, if you look at marvelous, you know, marvelous, I don't know which way this is gonna go on your screen (laughs) , but, you know, marvelous has dropped off a lot in usage, but awesome has increased, you know? And, and you can start to see the people who might be thinking, "Oh, we need to rescue marvelous. You know, that was, that was a great word." And, and, but then, you know, people are saying awesome, and that's hard to resist (laughs) . So, yeah, language changes all the time, and I think the important thing is to try not to put any kind of value judgment on this idea that saying like has something to do with that person's character. Because again, as soon as you start to look at transcripts of people actually talking, politicians, you know, everybody, you can show a lot of mess and repair error speech perturbations, um's and ah's in everything in, in pretty much anyone's conversation.

   21. CWChris Williamson36:29

       I saw a transcript that you'd done, I think, of Boris Johnson.... trying to give-

   22. ESElizabeth Stokoe36:33

       Hmm.

   23. CWChris Williamson36:33

       ... a p- pretty short speech and it just looks like someone's thrown alphabetic spaghetti down on a piece of paper. It's, it's not very slick at all. And you think-

   24. ESElizabeth Stokoe36:42

       Yeah.

   25. CWChris Williamson36:42

       ... this person has literally been elected to the highest position of power below-

   26. ESElizabeth Stokoe36:49

       Yeah.

   27. CWChris Williamson36:49

       ... the queen in the United Kingdom, and he-

   28. ESElizabeth Stokoe36:53

       Mm-hmm.

   29. CWChris Williamson36:53

       ... he basically can't speak when you read it out.

   30. ESElizabeth Stokoe36:56

       Yeah, absolutely. And I mean, there is, some of that is just kind of standard and normal. Like I was saying, we know we're- we're much messier it seems when you see it all written down and transcribed. Um, but yeah. Sometimes i- it can be interesting to really try to be, be precise about the mess, especially when, you know, if someone's judged to be a great orator, for example, or they sort of pride themselves on speaking Latin and dropping that into an interaction and then show actually it's quite messy. Yeah. But there's only so much of some people's voices you can take. (laughs)
7. 38:14 – 46:19

   Importance of Non-Verbal Communication

   1. CWChris Williamson38:14

   2. ESElizabeth Stokoe38:14

      Mm-hmm.

   3. CWChris Williamson38:14

      ... there's a common held belief that some huge percentage of our communication is non-verbal. Is that true?

   4. ESElizabeth Stokoe38:22

      (laughs) So if we... Yeah, this- this is a great communication myth, very easily busted. So, if you look, if you Google this, you'll see, uh, lots of examples of a pie chart, and the pie chart will tell you that communication is 93% non-verbal and 7% verbal. So, if I now turned off the audio, this podcast would make sense still for 93% of the time. We immediately know that can't be true. (laughs) But, and I, and- and, you know, this, it's very easy to- to sort of make that myth fall over. So for example, if- if the podcast, you're just listening to it on audio, how come radio is so popular if 93% of communication is non-verbal? How do we ever speak in the dark? How do you have a conversation when someone's downstairs and you're calling up and down? So obviously, this doesn't really make sense. If you go to a- another country and it's a different language, you know, like if you go to France and you don't speak French, you should be able to get by 93% of the time. So this, it just doesn't really make any sense. But it's so compelling, and people do love a good statistic and a nice pie chart, and they feel like, "Wow, I've really learned this thing," and it definitely makes sense. But one of the even nicer things about that particular communication myth is that the author of some of the original studies, Albert Mehrabian, himself has tried to bust the myth and sort of say how irritated he is by the sort of, he- he calls them something like self-styled communication consultant gurus who sort of stick this on pie charts, and it wasn't what he found, it was something else, it was more complicated, whatever. Um, but I've, I mean, I've been in lots of events, you know, where people are talking about this stuff, and they just trot it out and op- again, almost always, I just don't say anything 'cause it's just, it seems petty to say, "This is rubbish, but okay then." (laughs) Mm-hmm.

   5. CWChris Williamson40:06

      Or it, you just have to let it go. But it, but it, it's clearly wrong, and you only have to think about it for like two seconds to think, "Yeah, that can't be right."

   6. ESElizabeth Stokoe40:13

      (laughs)

   7. CWChris Williamson40:14

      Is body language stuff bollocks as well, then? If I look up into the right, if I cross my arms-

   8. ESElizabeth Stokoe40:20

      (laughs)

   9. CWChris Williamson40:20

      ... if I start mirroring your posture-

   10. ESElizabeth Stokoe40:22

       (laughs)

   11. CWChris Williamson40:22

       ... and stuff like that. Ho- how much truthfulness is in that?

   12. ESElizabeth Stokoe40:26

       So, obviously our bodies, our gestures, our gaze, and all of those things are all working in aggregate to get things done. So, if s- you know, if somebody is, you know, goes past, i- in front of your camera and they might do that, that might mean, "Do you want a drink?" And there's no words needed, we know what that means. But of course, you can't just do that. Like I can't do that to you now and like s- how, you know, how you're gonna get me that drink? It's not gonna happen. So, if you think about something like a request, depending on what you're asking for, who you're asking, your- your entitlement to ask for it, their obligation to fulfill the request, um, how important it all is, we will design our requests differently. And depending on the sort of in- the resources that we have to interact, we will deploy them as well. But of course, you know, doing things with- with our bodies and our heads and our eyes is really important. And to give you one example of- of this totally silent second-long interaction, I was coming back from Scotland on the train a few years ago, and I got on at Edinburgh, and the train was semi-empty then, I had a ticket. Got on, got my seat, and then gradually, you know, coming all the way back down to the East Midlands, it gets busier and busier and busier. Got rowdier, I think there'd been football, I'm not sure. But the train got really busy, really rowdy, very crowded. And bunch of blokes got on, they were obviously quite drunk and didn't know what their team had happened. But you know, it- it was rowdy and I had, I had that sense of like, not sure where this is going. And obviously, you know, for me sitting there, I'm just thinking, "Just read my book, not, don't make eye contact. Just hope, it'll all be fine." But there's a woman stood at the end of the carriage, so I'm sort of sat halfway down the carriage and there's a woman at the end of the carriage. And we just look at each other for a split second, well, you know, maybe a second. And in that moment, we tell each other, "We're- we're seeing the situation, we'll just keep an eye on each other." And you know, I've, I don't know who she is, never saw her again, nothing happened, it was all okay. But in that moment, you know some, you- you're doing something. And that's just a, a look.But of course, how do you turn a look into a l- into a look? You know, y- you have to even... We, we're even tacitly s- you know, we're not aware of it, but we're calibrating our gaze so carefully that we know what a look is. Versus when we're talking and we're, you know, looking to the right, lying, looking to the left, not lie, whatever it might be. But it, you know, those things I think generally don't have much evidence behind them at all, because it's actually really difficult to assess that kind of look to the right, look to the left. Is this lying? Is this flirt- You know, what, what are these things? Because they're almost never studied in natural settings, and that is the sort of the USP of my field. We look at real interaction where the stakes are real, whatever they are for people, in the wild, not simulated. We don't look at interaction in the laboratory. We don't ask people about their interactions. And that's very important to the kind of work that, that we do in conversation analysis.

   13. CWChris Williamson43:29

       I saw that a ton of psychological studies have just been put close to the chopping block. Uh, power poses, uh, nudging was in there as well. A bunch of stuff, um, that was really, really struggling to replicate. And I think that one of the big questions that was being asked there was around the fact that, look, in a naturalistic setting, there's just-

   14. ESElizabeth Stokoe43:50

       Yeah.

   15. CWChris Williamson43:51

       ... th- th- i- it's very, very difficult for us to get this. But, okay, so how are you going to study it if it, study if it's outside of the laboratory? And is this, you know, cultural conditioning masquerading as human nature?

   16. ESElizabeth Stokoe44:01

       Mm.

   17. CWChris Williamson44:01

       How do you separate out the cultural conditioning from something that's more inherent?

   18. ESElizabeth Stokoe44:04

       Mm-hmm.

   19. CWChris Williamson44:04

       I mean, that's... It's a, it, it is an interesting one. I remember seeing (laughs) this meme ages ago that was, um, the look that you give old people when you walk past them on the street as a teenager to let them know that you're not a chav. And it was the-

   20. ESElizabeth Stokoe44:20

       (laughs)

   21. CWChris Williamson44:21

       ... that face. It's not a smile. It's sort of down and to the sides, w- sort of thinning of the lips and widening of the face without opening the mouth up. And, uh, I haven't been able to stop thinking about that meme. I j- must have seen it two years ago, because it's completely true. I, I have no idea.

   22. ESElizabeth Stokoe44:38

       Yeah.

   23. CWChris Williamson44:38

       There's no name for what that greeting emotion is. It's not a hello. It's, it's actually something to do with, like, a, "I'm okay and you're okay," maybe. I, I don't know. Do you know the thing that I mean? Do you know the dynamic I'm talking about?

   24. ESElizabeth Stokoe44:55

       Yeah. Yeah, absolutely. I, I mean, I'm, I'm... S- probably phoneticians could give the answer to this, because they're very interested in-

   25. CWChris Williamson45:01

       What's that?

   26. ESElizabeth Stokoe45:02

       Somebody who studies the way speech is produced, including things like the diff- let's say the difference between going, "Yeah," and "Yep," (laughs) and something about the closing of the yep. You know, we all know when we've been yep-ed rather than yeah-ed. Uh, so I think the, there is something about a closed mouth smile. It's like the most neutral but friendly thing. "I'm not gonna open my mouth. I'm not gonna do anything. I'm keeping my mouth (laughs) closed." Yeah. I think that there, there is something in, in sh- d- showing, "I'm doing nothing," if you like.

   27. CWChris Williamson45:34

       Yes. Yeah.

   28. ESElizabeth Stokoe45:35

       And, and actually-

   29. CWChris Williamson45:35

       Very passive.

   30. ESElizabeth Stokoe45:36

       Yeah. So, so, uh, y- I have a colleague, and she gets her students to, as a kind of experiment, go on to, you know, go somewhere on campus and then do nothing. How do you do nothing in a way that doesn't attract attention and, of course, and you're not allowed to look at your phone? You know, how do you do it? And it's how do you, how do you stand at the bar and do, "I am waiting to be served," versus, "I'm at the bar, but I don't actually wanna be served yet, 'cause I'm waiting for someone." And we have really, you know, loads of ways of doing nothing in an innocent way. (laughs) How do you, how do you do nothing in an innocent way on the street? Well, we could do it. (laughs) So it's trying to... Uh, it's the same sort of thing. How do you show somebody that you're not threatening? And, you know, you've, you've picked out that meme, and I think that's, yeah, that's spot on.

8. 46:19 – 53:57

   First Date Do’s & Don’ts

   1. ESElizabeth Stokoe46:19

   2. CWChris Williamson46:19

      What about dating and first date dos and don'ts? What are some of the biggest things that people should and shouldn't do?

   3. ESElizabeth Stokoe46:28

      Um, (laughs) I, I did do some research 10 or so years ago on speed dating. It's my one and only foray into it, 'cause I thought, "Oh, God, I don't want to be the professor of dating," actually. (laughs) Uh, and the thing that I was particularly interested in was this o- this... Again, this was, uh, this was before things like First Dates on the television, which I've never seen. I keep thinking, "I should go look at those." But you never know what the raw data is like, so, mm. So I was looking at people on speed dates. They were in a particular age bracket, 30 to 45, and it was, you know, what do they, what seems to be the components of a date that runs smoothly? I d- didn't know what, what happened, you know, into the future of their lives. Um, and s- some of those things are, you know, c- quite obvious actually. Ask questions that aren't weird questions, that aren't too intrusive, that haven't been learned from a little book, you know, about... You know, ask them, like, "What is, what's the most dramatic thing that's ever happened to you?" is your first question, 'cause that would be a really interesting topic. Actually, some, m- you know, you need to do those mundane things first and just sort of, just, just turn by turn feel your way into the actio- into the interaction, and just take con- That's what we're doing in conversation. We're constantly taking our cues from the other person. So it's a bit like being that vet receptionist, the one who notices this one doesn't wanna talk about their dog versus this one definitely does. So all of those kinds of things are, are the same. I, I think having a topic, um, you know, finding a topic that you can expand for more than two or three turns is important. I've just started to do some work on first dates on Messenger and WhatsApp, you know, maybe the mo- the modern way to do it. So I'm looking at things like rapid, rapid, rapid, message, message, message, silence. (laughs) Uh, uh, you know, so you, so you can see... And I've got, I've got these people from early dating all the way through to, like, the development of when they're an item and then a bit more. And you can see some painful stuff like, yeah, message, message, message, message, you know, r- all going smoothly, and then there'll be a delay. And then the first person is like, "Mm, how long are they gonna wait before they try again to restart the encounter?" And I don't know. Even just reading these things that... I've never met the people. It's just text messages. But you sort of feel, feel the pain of the person who's like, "Mm, they haven't said anything. Shall I try again?" And eventually, you know-

   4. CWChris Williamson48:43

      Can't triple text. You can double text, but you can't triple text.

   5. ESElizabeth Stokoe48:45

      Is that right? (laughs) I, I also had a PhD student a few years ago who-... screenshotted, screen recorded people doing Facebook Messenger chats so we could see what they were writing before they sent it. And, you know, some of these conversations were hilarious because you'd see someone, the cursor would be blinking there on, on the screen, and they'd be typing X, X, X, blink a, you know, and then delete. (laughs) And then, like, maybe a sticky out tongue emoji, delete, and then like, what did they end up sending? And, you know, I'm thinking, "What would that look like in, in a real conversation if it was face to face?" (laughs) That would be quite a strange kind of like, are we going for like, cheek, cheek? What, what are we doing in this way? We're greeting each other or signing off. Uh, so yeah, it's g- but it, but what you see in though, in her particular, uh, work is the same kind of things that you see in, in spoken interaction, it's just that you can't hide the fact that you are starting it one way and then deleting it and doing something else. Whereas in fact, our conversation is full of things like, um, i- in, from my, from the dating stuff that I looked at. There's these moments that always happen, which is about whether or not the person's been married before. So this is this demographic 30 to 45. And so, um, one of the things that always comes up, either because people disclose it tacitly, like saying things like, "I was divorced," or, "I've got kids." So they give away some of their relationship history. But if they don't, then it becomes a bit of an elephant. Like, like, okay, who's... How, am I gonna ask you what your relationship history is? How am I gonna figure this out? And what I found was that women want, in these dates, this is, you know, 10 years ago as a particular cohort, women ask men, "So what's your relationship history then?" They wanna know, and men are accountable if they can't say at age, whatever, four, you know, 40, that they've had some kinda commitment in the past. Whereas, I have to say, in my small dataset, it didn't matter the other way around. (laughs) So men didn't ask the question and they didn't, you know, they weren't bothered if a woman hadn't had that kind of demonstrable commitment previously. But I've got this, this one case where the, the woman is, is saying, you know, "So have you got, so, so you've got no k- any kids? Like I, I probably can't say that you've got no kids on the base of everything you've told me so far, which sounds very sad." (laughs) And you see these little, little, little corrections as the question comes out. So if you were in, writing them down, you could, it would look like, "So no kids then?" Delete. "Got any kids?" (laughs)

   6. CWChris Williamson51:01

      Yes. Yeah, yeah, yeah. Well, I suppose it-

   7. ESElizabeth Stokoe51:03

      Of course, you see, you wouldn't see the presumption and then, "Oh, I suppose I should ask you this, even though I definitely know by now, I can tell," you know, whatever. (laughs)

   8. CWChris Williamson51:09

      When you're studying chats...

   9. ESElizabeth Stokoe51:11

      Mm-hmm.

   10. CWChris Williamson51:11

       ... you're only studying what people ended up pressing Send on.

   11. ESElizabeth Stokoe51:16

       Yeah.

   12. CWChris Williamson51:16

       But the number of times that you go back and forth, and then you have to think that there's one layer further up from that, which you need to actually speak to our mate Elon about to get Neuralink, which is, what were the things that arose in consciousness but didn't make it to your thumbs?

   13. ESElizabeth Stokoe51:30

       Yeah, right. (laughs)

   14. CWChris Williamson51:30

       You know, because there's a bunch of things that you could have said-

   15. ESElizabeth Stokoe51:33

       Yeah.

   16. CWChris Williamson51:33

       ... that this no, this no, this no, that no, still no, I wrote it, but no, and then-

   17. ESElizabeth Stokoe51:37

       (laughs) Yeah.

   18. CWChris Williamson51:37

       ... this is the one that I end up pressing Send on.

   19. ESElizabeth Stokoe51:39

       Yeah. Which, when you think about it, how on earth we actually do have these rapid conversations that are moving at such a pace, it's quite amazing. Yeah. So I mean, I don't know, you know, what's going up, g- what's going on upstream, because my, you know, again, my field is dedicated to what some colleagues have called the rich surface of interaction. So I don't know what's going on in your, under your school, you don't know what's going on under, under my school. We've only got what each of the says and does as evidence of anything. So that's, that's where conversation analysis focuses, which is different. I, I'm, I am a psychologist by background, but I'm not interested... I mean, I am interested as an ordinary human being, but I don't try to ascribe what someone really thought or whether they really are this thing, because what I'm looking at is how people are attending to, am I coming off this way or not? Or I didn't wanna come off this way, I need to attend to the way I'm coming across. Or if people say things like, "I don't know," is that... I'm not interested in whether they really know or not som- somewhere in their brain, and how you'll get to that. It's more like, why is someone saying I don't know at this point? Is it actually to try and avoid answering the question? Is it, is it to account for not being able to answer a question? What is that doing right now? 'Cause when you, again, when you look at these words and how they're placed in sequences like this, it's quite hard to just track them all back to some sense of what's going on cognitively. Because you can see that people are doing, doing things that it's not just about my brain to your brain, it's also about how am I coming across and, and, and fixing this, and am I saying things that make sense? Do I need to keep tweaking and qu- keep kind of turning the, the spanner a little bit until I've got the thing, and we have sort of, we have shared understanding of this thing that we're both doing together in this interaction?

   20. CWChris Williamson53:25

       Also, there's certain people whose personalities don't lend themselves to taking that feedback particularly well or at all, and y-

   21. ESElizabeth Stokoe53:32

       Mm-hmm.

   22. CWChris Williamson53:32

       ... everyone's got that friend that just continues to bloviate, and they're just plugging away and plugging away. And sometimes they're useful. If you're out on a, a night out or whatever and you just want someone... It's like background, it's wittering, you know?

   23. ESElizabeth Stokoe53:43

       (laughs) Yeah.

   24. CWChris Williamson53:43

       It's just the birds outside. You go, "Yeah, yeah, there, there goes Bruce. He's just continuing to churn on and on and on." A- and then you have other people who are, uh, overly receptive, right, to the awkwardness, to potential silences, to whatever. Um, you studied...

9. 53:57 – 1:09:25

   Studying Hostage Negotiations

   1. CWChris Williamson53:57

      Am I right in thinking that you looked at hostage negotiations and suicide hotlines and stuff like that as well? What'd you learn from that?

   2. ESElizabeth Stokoe54:02

      Yeah. Well, first of all, it, you know, it's, it's an amazing job that the crisis negotiators do. So I, I looked particularly at suicide crisis negotiation. So these are the recordings. So I, so I was given access to the recordings, uh, by the police negotiators who record at the scene as part of the job. So again, the data are kind of already there. They're like found materials because they're recorded and I'm not, you know, I'm not sure if anyone really ever looks at them again. But I, this is, this was the job. And, you know, the first time you start to listen to these recordings, you know, very typically, it's someone who's on a roof threatening to jump or threatening to asphyxiate themselves, do, do some serious harm to themselves. And I've obviously, it's f- you know, it's, it's the most difficult data that I've ever analyzed, and you also feel a real sense of responsibility because at the end of the research, I'm gonna go back to the negotiators and tell them what I...... what I found, um, and not, not, not do training, 'cause I don't like to think of it quite as training. But basically what I'm doing is showing them what they're doing that really is effective, and som- I'm showing them the things or hoping that they will see from the, the, the clips that I'm, will then choose to, to sort of play anonymized and play out for them, hoping that they will come to the conclusion that isn't an, isn't an effective strategy. But mostly it's about your colleague who is amazing at their job or is very experienced, what is it that they're actually doing? 'Cause one of the things that I, you know, as a total novice into this environment hadn't realized at the start was that a lot of these negotiations have a successful outcome in that the person comes down safely. So, I was, went into it thinking, you know, I had no idea what the statistics were and if you like-

   3. CWChris Williamson55:46

      You would be listening to a ton of people's last rites, basically.

   4. ESElizabeth Stokoe55:48

      Exactly, yeah. And that isn't how it is. So, so then what I'm interested in is, okay, along this two-hour, four-hour, nine-hour negotiation, what are the turning points? So that we can think about, what are the, what are the negotiators doing to get that person to, let's say, a more physically safe space on the roof so that they can really push the negotiation? So a lot of the first part is, how do we get someone from a precarious, very precarious situation where they might fall, even if they now don't intend to commit suicide, to somewhere much more physically stable so that we can then really start to, you know, support them, encourage them and so on? So, that was that. And, and a couple of things that I found that were very interesting, me and my colleague, um, Ryan Sickferland, we, we looked at basic requests for dialogue. So, any negotiation requires people to take turns in, in conversation. That's, that's what it is. It's nothing else. You don't know anything about the person typically, so you've got no background about, that you might think you're interested in a psychologist, like you, you don't know about their suicidal ideation, you don't know about their intent, you don't know anything about them. They're, they're a stranger. So you can't sort of assess their background and then on the basis of that develop a strategy. Your strategy has to evolve on the basis of the evidence of everything they say. So, what you want is to think, right, everything they say is staying alive. So every time the person says something, they're choosing to stay alive and not jump, so that's good. So, how do we keep them taking turns? But also, how do we keep them taking turns that are positive rather than negative? And that's what we were after. And so one of the first jobs is the negotiator arrives on the scene, you have to expect resistance 'cause if you do something so strong in a way or such a, take a, such a strong stance that you're on a roof and you're gonna jump, it's not the case that the negotiator comes and says, "Really want you to come down, buddy." "Oh, okay then." 'Cause people are s- in, in that situation are again surprisingly logical, very clued into how rational they appear, so they're not gonna just change their mind like that. There has to be a negotiation. It's going to take time. It has to, otherwise people don't look... You know, if you, if you take a stance, you don't just change your mind. People want to be seen as contis- consistent so you see that. So, how do you get someone talking? And the very, one of the very first things we found was that negotiators will typically say, "Can we talk about how you are?" And in this one nice example that I've got, the negotiator says, "Can we talk about how you are?" And when you just show that line and then show that the next line is naught point seven seconds, so you think, "Okay, that's not gonna be, uh, not gonna start talking. He gonna see resistance." And people think that they, they're gonna say, "I don't wanna talk about how I am." They focus on they're not gonna wanna talk about how they are to this stranger. But in fact what happens next is that the person in crisis says, "No, I don't wanna talk." So they don't attend to the how you are. They're just like, "I don't wanna talk. This thing, this ver- this particular thing that you've asked me to do." And then they put the phone down in this case, so that's, that's the... But now the negotiator's got to try again, so you just see these little natural experiments in which the negotiator can't give up, they got to try again. And in the trying again you see, well, what is it that's working? And what we found was that dialogue proposals that are built from the verb to talk get resistance. They get resistance like, "What's the point in talking? Talking doesn't do anything." And, and they actually orient to this notion that talking doesn't really do things. Whereas when negotiators say something like, "I wanna speak to you and let's sort things out," so speak gets traction. It doesn't get resistance. People don't say things like, "Actions speak louder than speak." We don't have that in our idiom. We don't have the same sense that speaking doesn't do anything, can, you can, you know, walk the walk-

   5. CWChris Williamson59:37

      Talk is cheap.

   6. ESElizabeth Stokoe59:37

      ... and just speak the speech. Yeah.

   7. CWChris Williamson59:38

      Speak is cheap.

   8. ESElizabeth Stokoe59:38

      Exactly. We don't, we don't say it that same way. And, and, and so we found... And this is all, generally my field is s- thought of as qualitative research where you look at, you know, in l- in a lot of detail at small number of cases, but because we were so surprised by this, we, uh, we went through and coded all the data, did some statist- statistical analysis as well and basically found that yeah, speak gets far less resistance and faster traction than asking to talk. And I suppose the nice coda to all of this is that we then went back and showed this to the negotiators in the UK and they were able to sort of get, start that, start it faster. It's also nice to show people things that is already part of somebody else's natural practice. We're not telling them to do something that's weird or not part of anybody's or, or experience and what they w- are just kind of doing naturally somehow. But it also shows you that verbs make a difference. So back to the point earlier about being pushed and pulled around by language we're not really aware. We also typically think that this person's gonna do it or not, it doesn't matter what words you say. But, but frankly if you don't look at the actual encounters you, you'd not, you can't say that. You, you can't say that words don't make a difference if you're not prepared to actually study those words in those real situations. And, and so that's what we found.

   9. CWChris Williamson1:00:55

      Didn't you have another example about...... bathroom towels in hotels?

   10. ESElizabeth Stokoe1:01:02

       (laughs) Yeah, that, well, that was, that was me thinking about this idea of nudge, 'cause of course, you know, it's a very, you know, very ... i- i- it's a pop psych kind of behavioral science kind of concept. It's not my type of thing, m- it's not my type of work. But I think it is interesting that in this particular study, I can't remember the, the authors now, but th- it was basically, you know, people recycle more, um, with the, with the wording of the sign in one direction rather than another. Um, I can't remember what that is now even. But, but really my point was let's look at the language much more than the s- the sort of overall message that it's pushing towards. So, one of them was about, I think it was something like, you know, "Please recycle your towels," versus something like, you know, "55% of people in this ... who stayed in this room, your room, uh, recycled." And it was something like that, a more personalized message drawing on social norms. And there's all sorts of theory you can put around that. But, but for me, what I'd be interested in is looking at the way we are pushed and pulled around by language in a not malevolent way. Sometimes it is malevolent, but, but actually words make a difference. So, asking someone to s- to speak rather than talk is doing something. It's, it's sort of tilting the, the outcome in some way. I- I've also shown that if you ask someone if they're willing to do something that they previously resisted doing a bit, then they're more likely to say yes than if you ask them if they're interested in doing it or if they would like to do it. There's a bit of a caveat here which is that willing is quite a heavy thing to ask someone if they're willing to do it, so you don't say to your partner, "Would you be willing to put the bins out?" 'Cause it kind of implies, well, what do you mean willing? Like, like, that's quite heavy. But, but, but actually willing gets these turnarounds in ways that don't ... other wo- other verbs don't get. So, you can start to see that words in turns make a difference to the very next thing that happens.

   11. CWChris Williamson1:02:51

       Mm.

   12. ESElizabeth Stokoe1:02:52

       Um, so, I, I, I, I mean, that, that isn't the kind of thing that, that nudge behavioral scientists typically look at, but I think it's very interesting the sort of natural way in which we are, you know, shaped by language all the time. So, going back to those how are yous and the, this, this, this, this recording of a conversation where it's two women on the phone and they go, "Debbie," "Shelly." "What are you, what are you ... You know, what's going on?" And, and you can see that they're just immediately not fi- they're not, they're finding it impossible to resist the, the, the conflict straight away. How do you not do that? It's very hard to immediately not engage in the frame that someone has set up. So, as soon as you accept that, you, you realize, God, we, we, you know, how do you not get involved in that argument? How do you resist that, you know, whatev- whatever it might be? Yeah.

   13. CWChris Williamson1:03:40

       I've got a friend, Alex Hormozi, and he did a video not long ago talking about how he always convinces his missus to go to Cheesecake Factory. Uh, and he frames the question in the same way that he used to do on sales calls. And instead of asking if she wants to go to Cheesecake Factory or saying that he wants to go, he says, "Would you be opposed to going to Cheesecake Factory?"

   14. ESElizabeth Stokoe1:04:01

       (laughs)

   15. CWChris Williamson1:04:01

       And obviously, like, well, I'm not opposed to going to Cheesecake Factory.

   16. ESElizabeth Stokoe1:04:05

       (laughs) Yeah. Yeah. Yeah.

   17. CWChris Williamson1:04:06

       I'd, I'd just rather go somewhere else. But that's pretty effortful in terms of a response to go to. And if you've already got, "Well, I, I would like to go to Cheesecake Factory," and you are not opposed, therefore if I have a preference and you have no preference, well, no, I do have a preference. I just haven't had my turn to say it yet. And-

   18. ESElizabeth Stokoe1:04:22

       Yeah.

   19. CWChris Williamson1:04:23

       ... yeah, I, that was really interesting. Another person, I've had Chris Voss on the show, who you may be familiar with, the ex-head negotiator of the FBI's anti-terrorism unit.

   20. ESElizabeth Stokoe1:04:32

       Mm-hmm.

   21. CWChris Williamson1:04:32

       Uh, and he has this really cool idea around getting a "that's right" from hostages. Uh, sorry, from hostage takers. Hostages can't say anything. And he was telling this story about, um, he was coaching one of his guys through a very long, uh, hostage negotiation that was happening maybe in the Philippines or in Vietnam or something like that, a decade ago, and this guy had had these people for ages and they knew that e- the statistics suggest that if it gets past a particular amount of time that people start being shot, this is not good, and they were just not getting anywhere. And Chris said to his understudy, "You need to get a 'that's right' from him today." And the "that's right" is about steelmanning the other person's position. It's about showing them that you understand precisely why they feel the way that they do. So, this guy went through a huge, big, long spiel of, "I understand that you feel like these imperial colonialist powers have come back and they've taken your country's sovereignty and that you're holding these tourists. I know that you feel this, and I know that that's," and da-da-da-da-da, this big sort of long, uh, very sort of, um, protracted diatribe about how this is how you feel and blah, blah, blah, blah, blah, blah. "Is that right?"

   22. ESElizabeth Stokoe1:05:43

       Mm-hmm.

   23. CWChris Williamson1:05:44

       And he got a "that's right" from him, and then within 24 hours, this guy had released all of the different hostages. But he fled, so they never ended up catching this dude.

   24. ESElizabeth Stokoe1:05:53

       Mm-hmm.

   25. CWChris Williamson1:05:53

       And then apparently about a month later, the hostage negotiator, Chris's understudy, received a phone call from a completely anonymous number, and it said, "I just wanted to let you know that if you hadn't spoken to me that day, every single one of those hostages would've been killed. I'm not sure what you did, but your bosses need to be very thankful for you for having done it. However, if I ever see you on the street, I'm going to kill you in cold blood." And I was like, oh my God. (laughs)

   26. ESElizabeth Stokoe1:06:23

       Yeah.

   27. CWChris Williamson1:06:23

       It's such an-

   28. ESElizabeth Stokoe1:06:24

       Wow.

   29. CWChris Williamson1:06:24

       ... intense story. Um, but yeah-

   30. ESElizabeth Stokoe1:06:26

       Yeah.
10. 1:09:25 – 1:15:44

    Reverse-Engineering Normal Responses

    1. ESElizabeth Stokoe1:09:25

    2. CWChris Williamson1:09:25

       Chris has another one where he says that a lot of the time in relationships, if you feel like there's something that's off with your partner, people's go-to is, "What's wrong? What's wrong? What's wrong?" And his advice, the, a, a better approach he suggests is to say, "It seems like there's something on your mind."

    3. ESElizabeth Stokoe1:09:43

       Mm-hmm. Mm-hmm.

    4. CWChris Williamson1:09:44

       And that, again, is ... It's so interesting, your work. I mean, you must find it incredibly fascinating, because it's at the intersection of a bunch of different things. It's not just linguistics. It's not just what the words mean.

    5. ESElizabeth Stokoe1:09:56

       Mm-hmm.

    6. CWChris Williamson1:09:56

       It's cult- culturally how is this word in the milieu, uh, beyond the definition of the word? Then how's it been delivered? Then how long's the wait been? Then what's the particular stage of relationship that you have between it?

    7. ESElizabeth Stokoe1:10:12

       Mm-hmm.

    8. CWChris Williamson1:10:12

       What, are you coming in with your priors? What are the expectations?

    9. ESElizabeth Stokoe1:10:15

       Yeah.

    10. CWChris Williamson1:10:15

        Yeah, I mean, it, it's a very multifaceted thing. And then to take that down and put it onto a piece of paper, and then, then to reverse engineer that back around, uh-

    11. ESElizabeth Stokoe1:10:24

        Mm-hmm.

    12. CWChris Williamson1:10:24

        ... it seems pretty complex. It seems like an incredibly complex, uh, science, art form?

    13. ESElizabeth Stokoe1:10:30

        Mm-hmm. I mean, it, it is complicated. But in a way, it's, what I'm doing is, is explaining and showing people things that people are often, you know, often because I'm working not just on dating, but, you know, the more sort of serious and institutional kinds of environments that I've mentioned, showing people things that they're not aware that they're doing, and how sometimes they're doing it incredibly well, and sometimes people like them, you know, i- uh, are doing this thing not so well at all. And it, it's a bit like proofreading, you know. You can't proofread yourself, but as soon as you give it to somebody else and look over their shoulder, you can see all your, all, all the errors. And, and so, it is, it is complex, but it's not like there's a huge conceptual gap between what I'm doing and the phenomenon. It's not like I'm studying black holes where it, like, they don't exist for people to understand them, they, whereas language, communication exists for people to live life and get everything done. And sometimes it can be a challenge, because we've all got our lifetime's experience and anecdata of how we talk. But what I do know, as a, as a conversation analyst is that how we think we talk isn't quite how we talk. And sometimes the things that we think are effective aren't quite right, or we can get a sense that that conversation didn't go very well, but I can show you how, exactly where. And then the challenge is to show that it's not just a one-off, it's a systematic thing, that people tend to do it this way, and it tends to immediately derail, create a friction, create a, something that you've got to fix. And you might get back, back on track, but you might not at the end of that, that conversation.

    14. CWChris Williamson1:12:01

        Have you looked at how easy it is for people to change their patterns, the speed pat- speech patterns, and the sort of words that they use, and things like that?

    15. ESElizabeth Stokoe1:12:12

        So I do a lot of training with the research findings, um, so some, and some of those findings are easier for people to see their way to doing than others. Um, so talk, and speak, and help, and sort, those words that negotiators use typically at the start, can, can be changed, if you like. People can start with, you know, "I wanna speak to you and get things sorted out," rather than, "I'm just here to talk and try to help." And you can change that fairly easily simply because they typically happen at the start of the encounter, where it is the negotiator doing the talking, and you ha-

Episode duration: 1:16:15