Episode 6 [Unedited]

This is episode is the full, unedited interview with Ted Gibson. If you haven’t listened to the fully produced episode yet, we strongly encourage you to do so before listening to this one. They’re shorter in length and much more refined.

EPISODE CREDITS

Guest Starring Ted Gibson, Professor of Cognitive Science at MIT’s TedLab

Produced & Hosted by Adam Greenfield

Executive Produced by Patrick Yurick, Instructional Designer – MIT OGE

Executive Produced by Heather Konar, Communication Director – MIT OGE

Special thanks to the following editors who provided us invaluable feedback that aided in the development of this show:

Christopher O’Keeffe, Co-Founder of Podcation

Kristy Bennet, Manager – MIT Women’s League

Jennifer Cherone, Phd Candidate – MIT Burge Laboratory

Erik Tillman, Phd, Formerly of the Kim Lab & Currently A Fellow at Vida Ventures, LLC

The Great Communicators Podcast is a part of Gradcommx. Gradcommx, targeted at enhancing research communication, is the first offering of Gradx – a professional development project created for the graduate student population at the Massachusetts Institute of Technology by the Office For Graduate Education.

MUSIC & SOUNDS

“Divider” by Chris Zabriskie is licensed under Attribution 4.0 International License (http://freemusicarchive.org)

EPISODE SCRIPT

ADAM GREENFIELD

Hello, Adam Greenfield here, host of The Great Communicators podcast series. What you’re about to hear is the full, unedited interview with one of the guests we spoke with. If you haven’t listened to the fully produced episode yet, I definitely encourage you to do so before listening to this one. They’re shorter in length and much more refined. You can find them all at gradx.mit.edu/podcasts.

The idea behind these longer, unedited conversation is to give you an opportunity to hear the entire talk, warts and all. This is not only a fun way to hear the full flow of the conversation but it also emphasizes the importance of the points made in the shorter, produced episodes, which again, can be found at gradx.mit.edu/podcasts.

Thanks for listening and enjoy the conversation.

Patrick Yurick:  So, we are going to start with, can you tell me your name and what you do here at MIT?

Ted Gibson:  My name is Ted Gibson, and I am a professor in Brain and Cognitive Sciences Department.  I am a professor of cognitive science here at MIT.

P:  Are there some things that you are working on?

T:  I work on human language, how people understand, produce, and represent human language.  I work on why language is the way it is. I’m looking at evolutionary models of why human language looks the way it is, but we look at how language is today to try to get it at that.  So, we look many languages corpuses, like some big texts of many, many languages to try and figure out why words look the way they do, why word order looks the way it does across as many languages as we can try to make inferences about what is natural about human language.  That is mostly what I do. That is kind of half of what I do, actually. The other half is, I work with remote cultures to try and also understand language and cognition, cognition more broadly because if you want to, I am telling you too much here, but, what I am interested in is why human language, why human cognition looks the way it does.  If you only look at industrialized nations, industrialized cultures, and industrialized languages, then you may be missing out on a huge portion of what cognition is, what language is. Because humans are such adaptive learners, we pick up whatever is useful in anyone else’s culture. So, it could be that language or number cognition or use of color words or something like that is picked up from interacting with some other group as opposed to being some part of innate human cognition.  So, you need to look at a very broad range of human, of the way that people live. I worked with tribes in the Amazon to try to look at some of the remotest human tribes to figure out what’s innately part of human intelligence and what is learned from other groups. So, that is another big part of what I do is working with remote groups.

P:  That sounds really big, it sounds like a really big task.

T:  It is a huge task but, I mean, I am not going to solve it.  I mean, all we can do is do our little piece, you know. So, if I want to, I want to know why human language looks the way it does.  So, I need to do more than just look at texts of industrialized languages. Anyone who wants to do this seriously is going to have to look at a broader range of the way human languages are.  I have to look at least a few. Lots of other people will be working, and we can collaboratively try to answer these questions. So, my group and I, we do a small part of that and, yes, I cannot solve this problem.  You are right.

P:  No, it is interesting because I feel like that ties really into what the course is about.  I mean you are studying, on some level you are studying communication but, languages about communication right?  I assume, but, I’m wondering. Well, I have two questions, I will start with what are some things you found that you found interesting about the work, like what is exciting you right now about the work?

T:  Well, it is kind of funny that you mention that, that you think it is kind of obvious that human language might be about communication.  I agree. Many people agree but, in fact, that is not the default assumption in much of literature. I do not know if you want to know this stuff, but Noam Chomsky, his default assumption is very different assumption.  He thinks language is not evolved for communication purposes but is evolved for complex thought, which actually is kind of hard to empirically test. It is kind of difficult to know what the evaluation would be, such a complex thought would be the reason that human language might exist.  The sort of most obvious view is the view you are raising, what we think is most plausible in some senses that human language might be evolved for communicative purposes. I find that hypothesis is really exciting, and so we look at dictionaries, lexicons of as many languages as we can, and we look to see how words change over time.  It looks like words get smaller. When they become more predictable in context, they get smaller, which is kind of what communication based ideas about language would expect. Then, maybe more interestingly is, I find that interesting on its own, I also find the structure of language as opposed to just the words themselves but how the words get put together.  For instance, the order of the words, that I find absolutely fascinating. The idea that there might be, languages that involve for easier communication in some ways. So, it may be that some word orders are actually more efficient for being robust to communication. So, a problem with communication all the time is I need to say something as a speaker. When I’m trying to get an idea across to you, I want to say something that

is most likely to get to you, so that I am worried about noise.  Noise at all kinds of levels, the way I might produce, the way you might understand, there might be background ambient noise or just sort of sound noise.  I need to say something to optimize the probability that you are going to get the meaning that I want you to get. It turns out that some word orders are more robust to noise, just an agent verb patient word order is more robust about the noise than an agent patient verb word order.  So, English has an agent verb patient, subject verb object word order. We say, “The boy kicks the ball,” but a Japanese or Korean or Turkish or Hindi have a verb final or SOV word order. It is very interesting to see how, so they would say, “Boy ball kicks,” that is the default way to say the word order.  Ignore, of course, the words are all different, but this is just the order and it looks like the SVO word order is more resistant to noise. The SOV word order, these other languages of many, it turns out those are the two most common word orders across the world’s languages, and it may be that the SVO is more resistant to noise.  It is more robust for communication purposes. So, what I am interested in, for instance, these SOV word orders tend to have endings on words to make them possibly more robust, as robust as they can. So, in English, we do not ever, when I say that the boy kicks the ball, I say the boy, or maybe I say the boy kisses the girl, okay?  I say the boy and the girl the same way. If it is the boy kisses the girl or the girl kisses the boy. In many languages, there is something called case-marking, which is an ending on the word. So, in Japanese you say boy ‘gah’ which means agent, girl ‘o’ patient, I actually do not know the words for girl and boy, but those are the endings that you say to let the hearer know who is the agent and who is the patient, who is the actor and who is the one receiving the action.  You find this across all the world’s languages wherever there is a verb final word, or where you always have this extra case-marking, this ending. When you have this verb medial like English you often do not, suggesting the idea is this case-marking is just like extra information to make that word order more robust to getting that information across. Anyways, so I am interested in why languages have those word orders and maybe they might be more or less robust to communicative purposes.

P:  Do you think they change their robustness, changes it all within the context?  I am thinking of, within this course we are talking about things like public speaking and we have actually done quite a bit of interviews around people who have been saying this kind of thing tends to work in a public speaking context, but it just strikes me that there is actually, I might be wrong, but the way you are looking at the language is granular, or not granular but like, I just wonder if there is contextual difference.  If you were having a conversation like we are, where I am interviewing you one-on-one, would the language have different effect than if I was in front of a group of a hundred people.

T:  Yeah, I mean, I think people talk to who they think their audience is, right?  So, I will talk to you based on what I think you know. I will talk to an audience based on what I, I mean, if I have a good audience design capabilities, and people will vary greatly on that, so depending on what I think my audience knows, I will say different things.  So, there is a lot of research on this is that, we actually do not know how people vary on this, how different individuals vary. But, there are definitely better ways to talk to people, depending on what they know. Depending on what the common ground is enough, I understand what our common ground is then I will describe things in at a very different level, depending on what I think my audience knows, right?  If I think they know all kinds of math about say, communication theories or information theories. This guy, Claude Shannon, worked out all of this stuff and if I think they know that then I’ll just start talking with technical terms like surprisal and entropy. If I do not, then I can explain all those. I will have to work through what the details of those things are, right, and explain. I would tend to do that anyway because in any audience sort of situation, I sort of think it is important to bring as many people in as you can.  There is always the danger that, any audience, there is always a danger that some people do not know the things that you may assume they know. So, that is a public speaking question, right? So, when you are doing a public speaking, I think it is safer to give broader talks than assume a lot of background knowledge. That is just, but different people do different things. So, some people, I guess, they know their audience then I might.

P:  Sure, no I think it is, this actually came up when we talked to another, Yang Shao-Horn, was talking to us about audience and developing communication to audience.  One of the things she was talking about was, I say, “Who is the audience for your work?” She is like, “Well, I do my work for the scientific method,” you know? I was like, “Woah that is interesting.”  So, there is communication that is being used in the published paper, and then she has to re-tool that for her classroom, when she teaches her students, and then she has to re-tool that also to large audiences when she is speaking.  My brain is lighting up about how that ends up happening. What do you keep what do you take away for those different audiences and does that change what is understood?

T:  You know, that is a very difficult question.  I think that is a hard question. The question of where you start and what you present to the audience is difficult.  I have a personal way of doing that where I like to start almost from basics all the time, almost no matter what and work my way up in case there are people in the audience that do not have the math they are suppose to have in that class that I am teaching, because it is not that hard to go through those background steps.  It does not take very long. It does not bore, and it should not bore the rest of the audience. So, that is my personal style is to attend to, but maybe, depending on what the material is, if there is so much math that that is impossible you know that you cannot start from square one and just teach people lineal algebra or something, so you have to make some assumptions I guess.  For the work that I do, the math is not that hard. I can sort of give the equations and explain all of those and in some detail to anyone, I think, so that anyone can get at least the flavor. I know that different people have trouble dealing with math, you know? Just being familiar with the notation basically, I think that is a lot of it, knowing what a lot of that notation is. It is confusing to some people.  I try to work that through initially. I give lectures to, you know, undergraduates, i guess then sort of general audience talks and then graduate classes. Those are really very different groups, each one of them. I always start with the same stuff and at the very basics and then work up to, depending on what the class is and how much time we are talking, of course we get into much more technical detail if there is a lot of time.  You know, an hour lecture, of course you cannot give that much of the technical detail other than describing what the formulas are and what they mean.

P:  Yeah, yeah that is really interesting because there has to be some kind of like a balance.  You cannot spend the whole entire talk, for an hour long lecture, explaining the thing that you are about to talk about, right?  There has to be like, how do I get to what I am talking about but also establish trust and understanding with my audience so that when they get there, they understand what I am saying.  Well, let’s back up for a little bit, the other thing I wanted to ask you about was how do you find this work, and what drew you to it?

T:  You know, probably everybody has interesting stories about that.  I do not know how interesting my story is. It is kind of accidental, I think, and maybe that is how many people’s stories go, but I was interested in math and computer science all the time.  So, that is what I did basically in school and high school. I double majored in computer science and math, and I ran into these problems of trying to understand how human language works, how we process and understand language in an artificial intelligence class in like, an undergraduate computer science class, so I just liked that.  It seemed to fit. It seemed like there was a lot of work there that people needed to do there, it was not well done at that point. People are progressing very fast over the last 20 years, a lot of stuff has happened. But, then there was so much stuff to be done, and I really liked the idea of working on language from a formal computational way.  So, I just started. I took one class in that, and then I decide to do a masters degree. Once I had done the master’s degree, then the PhD opened up, and I just kept doing this. There were always lots of opportunities, and there still are, you know? So, for all of my students graduating, I mean language has been, especial computational approaches to how language is structured, there is a lot of opportunity there to try and understand that.  All of our information, anything we know, is out there on the internet in human language, you know? That is how we communicate. So, if you want to get access to that information, there is like a lot about applied reasons to want to understand language. Say you want to do translation, that is an obvious application. So, there are just so many things that this is led to. I just work on why language looks the way it does and how we understand and produce it, like I am doing right now, how it is I am producing sentences and how it is whoever is listening to this understands what I am saying.  That is just a fun problem, right? That is a really interesting question, it seems so intuitively simple and obvious and it is really hard.

P:  Yeah, well it is very apparent.  I said to Adam yesterday, I was like, hey check this out, you know, the predictive text on my little phone.  I just clicked the center button like over and over and over again, and what is shocking to me is, I am not calling Google out our anything, but it just is really funny to me because it will not produce a sentence.  It just keeps going on and on and on and on, it just does this insanely long run on sentence, and that is Google’s predictive text on the thing. I just know that there are a lot of people trying to figure out basically what you are talking about on a computational level because, you know, these Siri and all the different devices that they are trying to help people get to be where they need to be in an automated fashion.

T:  Well, it really doesn’t have a message.  You have a message, something that you want to say.  So, you have to frame it, right? So, it is doing a different task, which is what is the preceding sequence that I have heard of words, and all I am going to do is try and guess the next two or three words, right?  So, that called end gram predictability. So, it’s just like from a couple of words, can I guess what the next word is or the word after that, which is very different from, “I’ve got a meeting that I want to convey”, like, I have an idea that I want to get across to you.  We are not doing that based on just a sequence of words. There is a structure to those words. It cannot do that with such little context. I guess, we are still a ways off before it has a good idea of what you mean, you know, and then makes guesses of how you might say that.  That’s hard. If that’s what you really want to do, but that’s hard. It’s doing some trick, right? It’s trying to just look at huge corpus of text, just giant, and then from all of those average texts, from the last two or three words, all of those averages, if you just said two or three words, it can make a guess about what the next one might be.  That is what it is doing. It is pretty good, right? It will often get the next word, but then after that, as you said, it is not going to guess a sentence, right? It is just going to guess sequences of these windows of three and four words, and that never turns into a sentence.

P:  It’s really interesting.  I am equally fascinated, well not equally, I’m not studying it or doing research on it, but I think this is part of the central problem of like communication.  There are some through lines through the different interviews we have done from very different departments around here with different faculty. You know, one of them that is really big has been adaptability in the context of trying to do it.  Like, sometimes you plan a talk, and then in the middle of it you’re like, this is not working. You have to read that moment of scrambling to reframe something. It seems like people who are good at communication can do that. I am interested in what you would think that would come from.

T:  I don’t know.  I mean, that just seems to me that if someone is good at doing that, it seems like someone who likes to perform and is a good actor in a way.  You know, someone who likes to be up on stage and can perform. That’s just hugely individually variable. I’m not like that. So, I would not be able to do that, the situation you just described of when a talk isn’t going well and it’s clear that people don’t understand, that’s a very hard thing to recover from, for me.  Maybe different people have. So, my approach to giving good talks is to over prepare. So, I am not a very natural performer. I think mostly what I do is just prepare, prepare, and prepare. So, I have had students who actually memorize whole giant spiels of what they’re going to say. I don’t do that. I do practice things over and over again.  I kind of just don’t like it when people are reading to me, which is what it feels like when something is completely prepared. So, I don’t like it. I prefer to work for message and just keep re-wording it over and over again to myself. I will just practice in a hotel room or wherever I am. I will practice the talk over and over and over again until it’s really easy for me to produce those ideas even though the particular sentences I use each time may be different.  I’m not good under that situation you’re describing. When things aren’t going well, I don’t know what to do. I’m someone who needs to practice things an awful lot to give a good talk. It looks like when I’m giving a good talk, it looks like I’m very natural, and I’m really not. It’s just all practice. It’s just tons and tons of practice of doing these things. I have done it the other way without practicing, and it’s awful. It’s just awful. I can’t do that. I know some people can, and I think they are great, but I’m not one of them.  I can’t do that.

Adam Greenfield:  When you practice, do you practice out loud?  Is that something that is cognitively better for preparing for that kind of thing?

T:  Yes.  So, when I prepare, I speak as much of it out loud.  I say all of it out loud, everything is out loud. So, I go through it  When I do that, then I find where my problem areas are very easily. It is very easy for me to know when it is not going well.  It’s amazing how much better it is the second time after how bad it was the first time, when you tried really hard to find those words, which really didn’t work.  The second time they would be much, much better. Then, the third, at some point, it’s smooth and then I won’t do it again. You get this impression that it’s very natural for me, and it really isn’t.  It’s just all, you just don’t see behind the scenes. It’s like not at all natural. It’s hard. Well, it’s not actually hard, it’s just time consuming. It takes up a lot of your time.

A:  I do poetry recitals, and I have noticed there is a big difference in when I practice it in my head and when I practice it out loud.  In my head, it sounds great. The first time I say it out loud, I’m like, “Well, maybe that didn’t sound as great as I wanted it to.” I’m just curious where that comes from?  Why that is with saying it out loud ?

T:  It is true, I don’t even know what it means to say something in my head, in fact, because I think it doesn’t really work.  It is very different from saying something out loud for real. You know, it doesn’t even seem real when I say it in my head.  I can fool myself. I have fooled myself, and it doesn’t go well when I don’t say things out loud and really practice everything.  So, I think people are very different on this. So, that’s what I teach my students. I teach a class on, it’s a lab class, but a big part of the lab class is they have to do a couple of experiments and then have to present their proposals orally.  They have to present their first set of results orally. They have to present the final projects orally. You can write it all out, but I have students who write everything out on your PowerPoint. You know, there’s like a presenter view, and you can just write everything down there.  If you want to read it, read it so that I don’t know you’re reading it. So, do it enough times so that I don’t notice that. Maybe, what probably generally happens when you write it out, if you really want to write it all out, is you get so practiced that you don’t have to read it at all because you spent the time writing it out and learned so well that you don’t have to.  I mean, people vary. So, this question about, can someone, what happens if something goes wrong? I don’t really have good advice there. It’s like, try not to show your panicking.

P:  I think what you’re say is valuable because we have talked to people, a couple people, that really adhere to that kind of thinking, but I would assume that is not every student who is going to be at some point a grad student here.  I am guessing there is a good deal of grad students at MIT who probably, like it’s good to hear somebody else who has done a lot of talks to be like that’s not how I give a talk because having multiple ways of approaching that problem is good.  If you can be a person who can recover in the middle of the thing, and that’s okay with you, that’s okay. But it’s interesting to hear about the refinement method you are talking about because that’s also come up. Like, the idea of, you know, when you read something that sounds perfect, it reads perfect, and it seems like it’s the first draft of that.  It is usually gone through like tons and tons of revisions before it has gotten there. So, yeah, my question. My question coming from that is, what have you noticed with that course? What is the most common thing that students are struggling with when they start trying to interact with taking that research they are doing and presenting it?

T:  I think the hardest thing for students who don’t have experience with public speaking is understanding how much they know about their topic and how they shouldn’t assume that we know that, the audience knows that.  So, the biggest problem I think people have is understanding how little the audience knows about the things that they know and that you have to give a lot of background introduction. If they assume that the material… anything that I know, everyone else must know and then just start to go into the technical details about the things they’ve done.  I think that is a mistake. People, I think, don’t understand how much they know and how, when you’re working at a particular project, not everyone else is also working on that project. So, you have to give that background, and it’s very important in any walk of life that anyone coming out somewhere like MIT is going to be. They are going to have to explain what they are doing to people who aren’t doing that.  So, you have to give a lot of background. That’s the biggest thing, I think, they don’t do well with introductory material for their topics initially. I warned them about that because I know that’s what they do, but that is still the hardest thing, what is the background? I guess I have the same problem. I think everyone has this problem, and anything you work on a lot, you can have this mistaken assumption that people you are talking with may also have that same experience in that same area.  So, of course you have to explain that. You may have to explain that to them also.

P:  Do you have any techniques or tactics that you tell them?  I mean, you talked about saying it out loud for yourself, but have you seen other things work for the students that you have said, try this out?  If anything comes to mind.

T:   No. So, the question is whether there are techniques of getting people to give good talks other than practicing a lot?  I actually don’t know of any other specific techniques. My observation is that there are people on a huge spectrum, from being very social and loquacious, so it’s not very hard for them to talk, and then there are people who aren’t.  Those are kind of the two extremes. The ones that aren’t, need the most work at being comfortable speaking, you know, just being comfortable. I don’t know because I actually find myself in that camp. I am very quiet in context of a talk.  When someone is giving a talk, I rarely ask questions. I rarely interact for example. So, I just tell them to practice and work on it. I guess I don’t have any other deep insights on that.

P:  I was wondering if you had any, when I was first learning to teach and give lectures and stuff, I started watching lectures that I was impressed by, ones that could, and that helped me.  I was wondering if looking at exemplars for you forms in any way how you construct a presentation personally. I mean, I don’t know if it does, but…

T:  I’m sure it does.  I’m sure it does. Watching other people talk gives me ideas about how I would like to present myself, but I think it’ll just a lot of each person has their own personality.  So, I like talks where you understand someone’s personality. You get not only the material, but you get a sense of who they are somehow or another. So, to that extent, it’s not very helpful for me to try and copy someone else’s presentation style because I have a different personality from them.  I think it’s best when a presentation is kind of true to that person somehow or another. I am just different from other people. I have a certain style where I like a little bit of humor, and some people don’t. I do like language, my research topic, it lends itself to allowing humor to get into almost any area because the topic area is language.  There are lots of examples of language that are funny in various ways. So, I try to bring that in, but for many people it doesn’t work. Some people just don’t like to say things, which are funny or too awkward. I have been told by faculty member and advisers here, mentors in my department, that you should never have any humor in your technical presentation.  So, this is like different presentation styles. I just ignore that because I think it should be not only, you should learn but it doesn’t hurt if there is something funny, which might motivate you in some way or another to remember things in some way. That’s my own personal style. I wouldn’t recommend, I have tried that. Other people will say to maybe insert joke here in this kind of thing from a student of mine.  I remember I did this ages ago. It really didn’t work for him. I was at his talk when he did this, and it fell really flat. He didn’t know what to do when people didn’t laugh at this joke, which was planted. I thought it was pretty funny later, but I was like, I was basically trying to get him to do something that I would do. It would probably work for me, and it just didn’t work for him somehow. I realize I shouldn’t be advising people on that level of how to do the talk.  They should do what feels most natural for them. The material has to be there and has to be super practiced, but the way you do it has to be what’s easiest for you, I think. It’s hard. That’s worked for me. I cannot say that there is a right answer to that question.

P:  I really, actually, appreciate that.  I think that what you are hitting on and what I was interested in what you said is that there is like this personality.  I practiced Aikito for a while. One of the principles in that martial art is no two people will ever practice that art the same way.  There is a way that you need to understand your body. In the beginning, you are doing very centered moves. Slowly, you’re suppose to learn your own body.  To become master, you have to be able to use your own body in that form the way it’s meant to be used and not the way other people use do Aikito because it is going to be very different.  I’m wondering, though, how did you figure out what yours was? Was it just doing it over and over again? Trying things?

T:  How did I figure out what my personal style for giving a talk is?  I mean, that’s just who I am. So, I sort of think that my personality in normal life is not very different from my personality in giving a talk.  I think that is the easiest thing for me. I think it works well if it’s the most, I’m trying to get this word. What is the word that there are always using today to describe a candidate who is running for president.  You know this word, what is the word? They have to be true to themselves. What’s the good word for that? When you have to be, there’s like a standard word. I’m just getting old and have trouble. There is a word that you guys know.  It is a very common word. It’s not an infrequent word. It’s where people have to be true to yourself. What is that? You guys are going to kick yourself when I. How stupid it is that I cannot remember this word.

P:  Solid.  Truthiness, that’s a Stephen Colbert word.  The truthiness factor is off the charts. Reliable? No.

A:  Is it mostly in politics?

T:  No.  I’m not getting the right…it doesn’t help.  On the internet, I’m not getting what I want here.

P:  You’re going to think of it later.

T:  It’s not interesting.  It’s a frequent word. Where was I?

P:  We were just talking about how you figured out for yourself.

T:  Yeah, how did I figure it out how I should give presentations.  How I should talk in public speaking forums. I want to be able to give presentations using my own personality.  So, that’s what I do.

P:  Yeah, it makes sense.

T:  Because that is easiest.  I think that is probably going to be the easiest for everyone, is to do what is natural for them.

P:  Yeah, yeah.  What was I thinking of?  I was listening to you. It always stinks when you’re interviewing and you think, I should’ve written down, I do have them all written down, but one of the things I was thinking of was, did you find yourself naturally bringing that presence from the first time you presented, or was it something that, I mean, were you really nervous the first time you presented?  I don’t know if you remember the first time you presented? How did that teach you about yourself? How did that process come for you?

T:  So, I went to graduate school at Carnegie Mellon in Pittsburg.  I was in computational linguistics, which was joining computer science and linguistics basically there.  Many fields in my experience back then anyway, did not have training in presentations. So, it was just, “Go out there and do it.”  You know? I had an advisor who gave me zero, absolutely zero advice about how to give a talk, absolutely no advice. So, I went up there with, back then it was a time for overhead transparencies, that was what we were using.  I made a bunch of overhead transparencies describing this particular topic I was working on, which was about how to understand human language. It was a computational algorithm to do some human language processing. I made these slides, which were unreadable.  They were 12 or 10 point font on these transparencies, and I realized when I put them up there, I could barely read them and there are all these people. There is like so much text in this tiny font, and I had not practiced. I just tried to work it through. It was awful, and so I learned from the very strong negative feedback that I got from myself, not from other people.  Other people were not so negative, but I thought it was just terrible. I think it was just sort of a property of the fields. This guy who was my advisor was, I guess, I guess he was a pretty natural speaker. He was pretty good. I don’t think he really had to work on giving presentations. He could just sort of stand up and knew something and just talk about, and I was not like this at all.  He also came from another field where people read stuff, philosophy, they would literally bring up a paper and just read it to you. They would sit down and read it, which I just found incredibly unpleasant. I mean, I can read a book for myself or read a paper, I don’t know why is that person there? I mean I want someone to explain to me in a conversational way, I guess, so I can learn things.  On the computer science side, there is also less motivation. There was not a real need to be really great at communication because, the jobs that people, there are so many jobs. It is so easy to get hired, and you are not really getting hired for your communication skills. I really wanted to be good at this, and there was no one training us. So, I mostly learned from watching myself fail terribly, and there was one that was just an awful talk.  My very first one that I gave. Then, I gave a lecture in a class, and that was also awful. Then, I learned that I, what happens if I just did it myself, practice something over and over and over again. I got feedback from my fellow students about what you know. I mostly just knew when something was awful. I do not know how much I need someone else to tell me when something is bad, when I I can tell if I am making all kinds of speech errors and I’m fumbling over the correct wording.  I know it’s awful. I don’t need someone, I don’t want to watch this I don’t want to annoy record this and watch this again, it’s horrible. I hide until it’s ready, and then people give me feedback. By that point, it’s usually pretty good, I know it is pretty good. So, then I guess I can definitely deal with organizational comments about how I might move things around to the point in a slightly cleaner way in some way or another. But there is an awful lot of back, again it’s always, so much work for me is doing stuff on my own.  I am too embarrassed to give a terrible talk in public, and I don’t want to do. It makes me feel too terrible, even for a practice group. I don’t want to give that impression that I know what I’m doing, which means a lot of practice for me.

P:  That’s really cool. I think it’s interesting because it does sound like over time, you develop this kind of method to prepare yourself to give the best talk that you can give.  I think that just going from that first presentation that you just described to where you’re at now, there’s a lot of critical analysis and looking to solve the problems that you might of had in that talk so that doesn’t happen again.  I think it’s interesting there is this bounce back and humility that you have to have about your own, you know, saying to yourself, wow that didn’t work. I know have to fix it and figure out how to fix it. It is really interesting. So, this is like totally out of left field, but this is my last line of questioning.  I wanted to get your thoughts on it because of the work you do. We have this week we’re going to be talking about digital communication and like how people network online and the way they connect with each other online. I didn’t know, I feel like it is just such a weird part, like we are seeing all this research come up about emoticons and like language development.  I have done some papers that are about language development rapidly changing because of texting, because it crosses, dialects used to not be able to expand as quickly or colloquialisms weren’t able to attend because of the Internet. I was wondering if you had any ideas about digital communication, how to think about that? Would, I am not going to say younger generation, but these PhD candidates going out into the world and having to think about how they’re talking to people through the internet.

T:  I don’t know that I have anything.

P:  It’s fine if you don’t.

T:  I am not sure how to apply things I do to.  So, what exactly is the topic? The topic is how students, say grad students, interact with the world through digital media.

P:  In communicating.

T:  So, via Facebook or Twitter or these kinds of.  So, what’s the?

P:  The question is, would it be the same as the way you would do it in person?

T:  I don’t know.  I’m old….in that sense.  In that sense, I don’t really know, I don’t really do Facebook.  I do Twitter, but Twitter is like a useful way to get a lot of information.  That’s why Twitter, I find that, I don’t know what to do with Facebook still.  I mean, I know why people use it, but I don’t use it for what they, I don’t know what that, I don’t know.  What I work on is like how language like to know how language might develop and change, but that’s different from how you should present yourself, right?  This is like how you present yourself to the outside world so other people can see you, so I don’t really know what to say. I mean I could tell you about how languages change.  How words develop. Words shorten over time as as they get, not only do they get shorter just because of their overall frequencies, but they get shorter because of their use in particular context.  As we know it, I guess texting is a nice demonstration of word shortening. We don’t want to type much. So, as words get more predictable, we type less because we have less effort for us as a speaker; as long as the receiver can understand it, then we will type less and less.  That is how we get all these shortenings of particular words that are used in context. So, that is how language words change. That is that the communication story about word development. So, we’ve got lots of evidence that that’s the case in every language that we can measure, but that’s not about how you present yourself.

P:  I think it is in a sense, one of the things that strikes me is what you were talking about audience earlier.  So like, if you are shortening….what I have seen, I work with teenagers, right? What I have seen is they shorten their language exponentially with each other.  Their establishing very quickly meaning around each other, but then they’re not conscious of how to invite other people into that.

T:  So, the question is that language serves multiple functions.  Language is a communication function, but it also serves the function of letting others know who you are, right?  So, it has social functions. So, you are communicating not only some predicate argument structure about some things that you’re going to be doing today and whatever these things, but you’re letting people know by the words you use some features of yourself.  Okay. So, what words you use in what context, you know, if you swear words or whatever, these kinds of things, or if you use slang terms, people will make generalizations about properties they think you may be in some way or another. Some people will certainly think negatively of you for some things that you may say in some context.  So, you have to be careful of that. I’m not saying that’s right, though. I am not advocating that. It’s hard. But, you should be aware that people make inferences based on what you write, what you say, how you talk, and how you write. If you spell things wrong, people don’t like it. I mean, not people, but some but some people. So, that’s always a problem.  It’s an issue. If you spell the word there wrong, a lot of people get so upset about it. So, the argument is, people who don’t like that, that is failing communication. But it is isn’t. It is actually completely predictable in that context. It’s failing because of some sort of social conventions where you’re telling me that when you don’t spell that correctly, that you don’t care enough about our conventions of how you should spell or how you should know it.  It’s not a problem for me knowing what you meant in that situation, but you’re actually conveying to me that you didn’t learn enough about what the spelling conventions are in English to get that right. So, then you are conveying to other people, which might be negative. I would tend to be conservative on this and try to figure out the fall those conventions in some ways, but I also don’t really like those conventions. I don’t think it’s so bad. Language changes.  People get very upset about the strangest things or people don’t like that we use nouns as verbs all the time. People who are in technical and business dialogue, they get so upset, you know, we talk about impacting, this work will impact something because impact used to be a noun. Now it is turning into a verb. Some people say, that’s not English. You can’t say impact as a verb. It is like really annoying. Every word started that way. You don’t realize that, but the word donate didn’t exist in the English language one-hundred years ago.  It actually started as a noun, a donation. Then, someone started using, like started generalizing and saying, hey, let’s talk about donating, and people were very upset about donating as a verb initially. This is true for many, many words. Like the word zoo. Zoo didn’t exist. So, these are words from one-hundred years ago that we had the same discussions about. So eventually, I don’t know eventually if it will happen, but it probably at least will always happen. Half of this stuff will probably always happen, and you’ll always get older people who don’t like language changed, who think that the way they talk is the right way and younger people are messing it up.  That’s the way it happens. I don’t think that will ever change. I would be surprised if older people always seem to like things the way they were, to think that was the right way. So, you should be aware of that at when you’re young. You should be aware that you’re going to be bothering older people who may be hiring you.

P:  There’s like a scalability of impact that has changed a lot, too.  When you used to, even texting is very different than posting something that you would text to your friend on your Facebook wall because the access to that communication can be from a lot of different, what the modern equivalent to standing on the corner on a busy street one-hundred years ago yelling something profane about the president is like exponentially seen because back then, it was only heard by like, what, fifty people in that town square.  Now, it’s being heard by anybody who cares to listen in the world, and it’s on some sort of record that that exists in a different way. We didn’t have, I just think it’s interesting because I’m like, this is tangential. I get interested in this because I do see that while students have the natural, just like I did in high school, have the natural impulses to create our own language and shorten the language quickly so we have inculture with each other, they’re not as aware of the impact of that on all of the audience they could have access to that language.

T:  I think that’s right. Older people don’t like it when things are changed.  They like things the way they are. So, you will always find that, that older people like things the way they were.  You will be potentially offending people. I was once asked about apostrophes and whether those are necessary. A guy from Slate.com wrote an article about this.  I’m trying to remember what his name is. Matt something. A guy with a british, some government officials of some kind, were trying to get rid of the apostrophes on their signs.  They wanted shorter signs, so they thought these apostrophes were unnecessary in some way or another. This guy is somewhere in the middle of Britain, north of London, started up a society to save their apostrophe because he thought the apostrophe was like dying in a written form of English communication, and he was making the claim that we couldn’t communicate with each other without the apostrophe.  So, the Slate guy wrote to me and asked me if I would comment on that. I said, “Well, that’s just not true.” There is lots of ambiguity in language, and we have no trouble understanding what’s meant. I’m not saying we should get rid of the apostrophe, but if we got rid of it, it would be zero problem for our writing communication system. There is like no, there’s no place where if you left it off, you wouldn’t be able to figure out what it meant.  Yes, it’s true that the word she’ll looks different from the word shell. It’s really hard to construct those examples. Try and find me a location where you put the word shell and you put pair words, she will, where you don’t know which one was intended. It is pretty hard to make up a situation such that you might be confused. So, language is like this all the time. We have words like to. The number two, too, and to. I say this in spoken language, but in written language, the word to actually has multiple meanings, there is an infinitive marker and there is a preposition.  Are you confused when I say I want to win or he went to the store? No. You read those things or whatever, and you say them the same way. You are not bothered. It wouldn’t be a problem either for she’ll and shell. You know, you would pick it up in the context, which one was meant. People like things the way they were. There is an older guy and he wanted to keep it. So, he likes it the way it is. It’s not going to be a problem for communication. It will just be a new communication system slightly tiny variant. So, this is exactly the same kind of question you’re talking about when people change the language, young people, whoever, shorten things, and if everyone doesn’t know about those things, they may not like those shortenings since they are used to their old things.  So, they may say those are wrong and they inhibit communication, they don’t inhibit communication. They wouldn’t, but they are different. There is learning involved. You have to learn those things. Maybe you don’t want to put time into learning these different ways to write these things because maybe you don’t use use those words in that way like these other groups do, right? You’re not texting. If you’re not texting, it doesn’t matter to you how many characters it is, right? If it’s three versus six characters, it’s not going to be a problem.

A:  Out of curiosity, is that the same with commas?

T:  Yeah, commas are pretty similar.

A:  I thought of the Oxford comma.

T:  Same argument.  It’s the exact same kind of argument.  People have very strong opinions about these things.  It’s funny. When you read about the Oxford comma and the apostrophe, and there is one that is so crazy with this writing style thing, you can get people going off on what you write if you’re a writer.  Some people put two spaces after a period, and some people put one space. There is one space and two spaces. They get so upset and it’s like look, this is a convention either way. It doesn’t make any difference, it’s just like why?  But, people really have strong opinions about this. This is kind of internet craziness. You need to go outside of the internet, everyone has….What do you like? Do you like one space?

A:  For me, it’s one space.  For whatever reason, two spaces, it’s just not.  My understanding of the history that the spaces are there for typewriters.

T:  That’s right.  That’s the history of it, but it’s like, what difference does it make either way really.

P:  The Oxford comma came up as a thing because of typesetting.  I mean it was like a way of saving space.

A:  But it can change.

P:  Well, it can.

T:   There are different meanings.  But in the context, almost always you will know what is intended.

P:  Yeah yeah yeah.

T:  There is very, this is the same kind of thing, like, you can set up situations, and people do on the internet, you can get situations where it’s hard to know which was intended.  But, those are rare. It’s like really a normal conversation, well writing in this case, it’s writing. It’s like, we know which was meant almost all the time.

A:  Does it matter as far as the speed of the cognition of the intended meaning, because if…

T:  It might.  That would depend on the reader probably, right?  So, if you really do learn that there’s a different meaning associated with having the comma in that conjunction or not, then you probably will be a little confused by one or the other of those things, but that’s, yeah, there are just two different conventions, right?  So, conventions means, there are conventions, there is a meaning associated with this form. You write a form, and if you have one of those, then it’s going to be weird for you to read the other. If the other meaning was intended and you don’t have that association, then it will be slow for you, for sure.  It will be confusing.

P:  It kind of brings up this thing of skill that I was talking about.  Like, there is an intended, if somebody cares about what you’re trying to say and they don’t understand the sentence, they’re going to try and find out what you’re trying to say even if they don’t get it when they’re reading it.  That’s actually grown, and I’ve seen that transaction grow in value. It’s actually really important for my audience to care enough to break through my communication barriers to get to the intended meaning. You see that with YouTube bloggers and and all these people who garner really big followings, they might try something and nobody understands what they’re saying, but the audience still is like invested in finding out.  Just going back to the digital communication thing, the thing I was also thinking about is what actually changed the conversation and shifted because we have begun to accept that when we put things online, a lot of people can see it. So, the conversation has shifted not like around judging it, well that’s actually decreased. I have seen a lot of people who will be like, “Well, the Facebook is a personal domain, so we can judge him less for the things he says there.”  So, those constructs are starting to happen, but then the worry about everybody seeing it has shifted over into an international/national conversation about privacy, which has little to do with communication, although it does impact how we are afraid or not afraid to communicate in different spaces. Anyway, that’s my interview. That was an awesome interview.

LISTEN TO MORE EPISODES