Skip to content

Natural Language: How We Speak and Who We Are

Share this article: Facebooktwitterredditlinkedinmail

Dr. Michael Housman and Dr. James Pennebaker talk about the power of words. What does natural language - the way we use pronouns, content words, exclusionary words, and prepositions - say about us?


Dr. Michael Housman: Hey everyone! Thanks for tuning in. My name is Mike Housman, and I am your host. I am very honored to be joined today by the very distinguished Dr. James Pennebaker who is the Centennial Liberal Arts Professor of Psychology at the University of Texas at Austin. He is a member of the Academy of distinguished teachers. He is an author of a variety of different books, most notably one of my favorites, The Secret Life of Pronouns: What Our Words Say About Us. So, before I begin, Dr. Pennebaker, thank you so much for joining us.

Dr. Pennebaker: It’s good to be here, thank you.

Dr. Michael Housman: Yeah absolutely. So, we want to start by talking just a little bit about words and the power of words. I mean this is really a focal point of your research and you’ve made it a point to teach us about the value of words. What are the words that are really interesting to you right now?

Dr. Pennebaker: Well, I come at words from a very different perspective than the average literary person or even a Computer Scientist. Language is made up of kind of two very broad categories: content words, which are the guts of our communication and tells us what we are talking about; but there are all these junk words that we don’t pay attention to. We call these function words. Pronouns like you, me, I, it; or prepositions like to, of, for; or auxiliary verbs like and, is, have, had. It is those words, those function words, that really tells us a lot about who people are, and how they’re connecting with their topic. So those are the words that I accidentally discovered are much more psychologically interesting. So, it’s not what people are talking about, but how they’re talking that’s much more interesting to me as a psychologist.

Dr. Michael Housman: Yeah, it’s fascinating and I think a lot of your research has really touched on how those words - those function words - can actually reveal something about someone’s personality. Maybe you could share a little bit more about that.

Dr. Pennebaker: I can tell you that probably the most diagnostic word in the English language is the word I. Now, everybody has a stereotype of the kind of person who uses the word I. Most of us think, well, everyone who uses I words they’re confident, arrogant, self-centered, et cetera. Turns out that’s not true at all. I words, and all pronouns, tell us where people are paying attention. So, if a person is really interested in a woman and they think about her all the time, they use the word “she” quite a bit, or “her”, but a person that uses the word I is naturally focusing on themselves very briefly and what we find is that the more a person uses I words, the more self-focused they are, and a person who’s self-focused tends to be self-conscious, self-aware, they tend to be more aware of their internal sensations. So, people who are depressed or who are physically sick actually use the word I much more than someone who is not depressed or sick. They also tend to be more self-effacing, insecure, and they’re more prone to depression. So, I words intrigue me because, in spoken language, the word I is the most commonly used word in the English language. It’s also this word that’s telling us not only if the person depression-prone, but also if they’re lower in status in an interaction. They are more likely to be honest and more personable. That’s an interesting example of this tiny word that packs a real punch psychologically.

Dr. Michael Housman: No absolutely, and I’ll be honest after I read your book, I was a lot more confident of the way I was speaking to people and my use of language. Do you - on a personal note - do you find that that’s true of yourself? That as you learn about the impact of words you’re a little bit more cognizant of your own language?

Dr. Pennebaker: You know, only a little bit. The problem is trying to pay attention to these kinds of words frankly is impossible but often when I’m doing some kind of writing I’ll be more careful than I used to be so a good example is the word we. We is a really deceptive word because it can signal both being more standoffish and distant. But it can also be inclusive but it depends on the context. So, if you are listening to a politician and that politician says we, try to figure out who is the we that politician is talking about and if you can’t figure it out, don’t trust that politician very much. So, you know when I have been in leadership positions, I’ve learned to say when I say, “we need to do so and so”, I will specify at the beginning and I’ll say, “everyone in this room and I all such and such,” and we, so on and so forth. So, they know that when I say we, I am specifically including them.

Dr. Michael Housman: That’s helpful and I guess on the subject of kind of politicians being honest or deceptive, do you find that there are words that you can use to tell if someone is telling the truth?

Dr. Pennebaker: Well, we can do better than chance. Never trust anybody who says they can tell with great accuracy if they’re telling the truth or not, because we can’t do it with language either. But there are group of words that tend to co-vary, that tend to go along with people being honest. So, a person who uses the word I tends to be more honest. Also, a group of words that are differentiators - words like except, exclude, but, these are words when you’re making a distinction between what is in a category and what is not in a category. People who use those words tend to be more honest as well. They’re able to say what I did but also what I didn’t do. But a person who’s lying, can make up what they may have been doing if even they didn’t but it’s almost impossible for them to say things that they didn’t do since they’re lying.

Dr. Michael Housman: It’s fascinating. I’m wondering, shifting from deceptive behaviors to kind of rapport building - obviously that’s very near and dear to our hearts here at RapportBoost - are there words that you find are helpful in building trust, building rapport?

Dr. Pennebaker: What we find with rapport, is rapport is a process. Word use itself – by one person - is not a great way to establish it. Instead, what we’re more interested in is how two people who are talking - once they have a rapport established - how do they talk? How do they match linguistically? And we find that people who are using these function words – pronouns, prepositions, and so forth - at similar rates are on the same page. There’s much more rapport between them. We can actually go through it to edit any text and analyze how similar the language is between the two of them and we can make some really audacious and accurate predictions. So, for example, on speed dates, the more two people are using words the same way the more likely they are to go on a date. The more that young dating couples use function words in comparable rates in their IMs back and forth, the more likely they’ll still be together several months later. In other words, it’s capturing the degree to which they are connecting. They’re on the same page.

Dr. Michael Housman: And it’s fascinating you can measure that. I’ve seen applications to coworkers in the office and how well they communicate with one another, and that being reflective of having a productive, collaborative relationship or them not getting along and not working well within the company. So all the applications I think are amazing.

Dr. Pennebaker: Yeah, it’s really promising.

Dr. Michael Housman: Yeah. So, switching to yours, I want to talk a little less about words and more about mediums because obviously, we live in a world where there are a lot of different ways of communication that are proliferating. You know, I’m wondering, does the medium impact the effectiveness of words? Are you finding differences in speech versus writing versus something like chat and social media?

Dr. Pennebaker: You know it’s been interesting. We find relatively similar patterns as a function of media. Now, it does vary some, so, for example in the deception world, you know I just told you that I words and these differentiation words predict telling the truth. Well, they do most of the time in face to face interpersonal situations but there are other contexts where the word is turned upside down. And one is where people are writing fake reviews for Yelp or TripAdvisor. There’s a great study by Jeff Hancock at Stanford finding that people who like fake reviews actually use more I words. It’s almost to say when they are on stage, they use I words in a way that will suggest honesty when in fact they’re being completely deceptive. So there is some context but whether it’s e-mail or a blog or twitter or whatever, it’s probably less important than the function of their lies and perhaps the audience of their lies.

Dr. Michael Housman: That’s interesting and I think speaking of electronic communication especially e-mail and correspondence, I believe I read somewhere that you apply the technique to some of your own e-mails and correspondence. Was there anything interesting that you learned there?

Dr. Pennebaker: Yeah, there’s it one of these situations indicative of the law of hammer, when you embed a new tool - a hammer – you discover the world is made up entirely of nails. And that’s also true with the development of my text analysis program, this is called Linguistic Inquiry Word Camp, LIWC. And the real program is just a very simple program. When I was writing a book, The Secret Life of Pronouns I started wondering if I could tell in my letters or recommendations for students how I really and truly feel about the students. Then as you my guest, when we write letters of recommendations, we’re really hesitant to say something bad about people but the reality is I like some people more than others and I’m old enough and I’ve written hundreds of letters. So I’ll pick through and analyze the history of my letters of recommendation and found they were certain aspects of my letters that predicted how much I truly liked them. And to my surprise one of the best predictors was use of positive emotions. Words like good, great and so forth. And you would think that would be associated with my liking the candidate more; in fact it’s just the opposite. The more I used emotional words, positive emotional words, the less I truly liked them and I went back to try to figure what this was. What I discovered was for a student whom I think is amazing I don’t say; “This is a wonderful, great student” and so forth. Instead, I will give examples, this and this, they were in this situation and they came up with this creative strategy that did so and so and so. In other words, If I’m really enthusiastic, I’m looking at them and what they accomplished and if I don’t have much to say about them, I just kind of wipe my hands of that and say, this good person, they’re great, they’re this and that. So, I think that’s the thing that surprised me the most through the analysis of my language.

Dr. Michael Housman: Well, that fascinating. It’s neat to be able to turn that lens toward yourself or maybe that hammer as you call but…

Dr. Pennebaker: I prefer to turn that lens toward myself…

Dr. Michael Housman: For what it’s worth, we’ve worked with receptivity which is based on Luke and I have analyzed many of my own e-mails and I’m astonished at how accurate it is. This could be that kind horoscope kind of thing but I truly believe, you know, it’s nailing me on extraversion and eroticism and all those different psychological factors. It’s fascinating.

Dr. Pennebaker: Yeah, it was interesting. When I started working with them, one of the guys who put that program together and I was blown away by it.

Dr. Michael Housman: Yeah. No it’s truly amazing. I encourage anyone listening to check it out because I think you can analyze just a random sample of language, and it’s fascinating to learn about yourself. So, I guess one last question. You know, we’re analyzing language from the context of trying to optimize communication specifically around chat. I’ve seen applications of tools that you help build towards analyzing various forms of language like corporate communications and a variety of others. I guess especially in an era where we’re seeing the proliferation of electronic communication. What do you think is the promise here? How do you think we can apply these insights to improve the lives of human beings?

Dr. Pennebaker: Well, that’s a pretty broad question but I think the answer is wherever there are words, they are analyzable and we can learn something about the person who’s speaking, the person’s relationship with the other person to whom they’re speaking, and also we can learn about that person’s relationship with the topic about which they are speaking. In any kind of situation where that’s important - on a very broad level - everything from sales and persuasion and political rhetoric to human relationships, dating sites etc. Any time that you’re interested in the connection between person A and person B or person A and robot B, then seeing the way in which they’re communicating, their style, is really critical to understanding much of what’s going on between the two of them.

Dr. Michael Housman: I completely agree and as an aside, you mentioned humans communicating with robots, we’re increasingly interested in understanding the nature of that interaction, right? We haven’t even really begun to scratch the surface of how those two are going to interact and how we’re going to interact with machines. You know, whether we’ll treat them like other human beings or whether I might curse at Alexa because she got the wrong order. So I’m fascinated by that and I think there’s going to be a whole bunch of data made available to study that and for us to understand that human-computer interaction.

Dr. Pennebaker: That’s right. And you can tell in talking to Alexa, it’s very seductive to make her real and you know the programming of it, we have to appreciate this is a very first foray into this world. The future of somebody like – I’m hesitant to use her name because she’s going to start talking to me in a second - with perspective is that I want her to, you know, she’ll say something and then I’ll say thank you and she’ll say, “No problem – any time Jamie.” So it’s a much more natural conversation and I think, OK, OK it is kind of creepy but it’s also kind of neat, and I anticipate some gigantic changes in that world in the next year or two.

Dr. Michael Housman: I completely agree. We’re definitely watching all those developments around the bots on messenger and Alexa. I think it’s an exciting time to be alive if you’re interested in language and psychology, it’s going to be fun to watch how these things evolve.

Dr. Pennebaker: That’s right.

Dr. Michael Housman: Dr. Pennebaker, thank you so much for joining us. This has been truly an illuminating conversation.

Dr. Pennebaker: Well, I’ve enjoyed it as well.

Dr. Michael Housman: Great. Thanks very much. We’ll check back in at some point and see how your relationship with Alexa is changing over time.

Dr. Pennebaker: Thanks a lot.

Learn more about live chat agent training from the team at RapportBoost.AI.

Follow us:Facebooktwitterlinkedinrssyoutube
Dr. Michael Housman

About Dr. Michael Housman

Michael has spent his entire career applying state-of-the-art statistical methodologies and econometric techniques to large data-sets in order to drive organizational decision-making and helping companies operate more effectively. Prior to founding RapportBoost.AI, he was the Chief Analytics Officer at Evolv (acquired by Cornerstone OnDemand for $42M in 2015) where he helped architect a machine learning platform capable of mining databases consisting of hundreds of millions of employee records. He was named a 2014 game changer by Workforce magazine for his work. Michael is currently an equity advisor for a half-dozen technology companies based out of the San Francisco bay area: hiQ Labs, Bakround, Interviewed, Performiture, Tenacity, Homebase, and States Title. He was on Tony’s advisory board at Boopsie from 2012 onward. Michael is a noted public speaker and has published his work in a variety of peer-reviewed journals and has had his research profiled by The New York Times, Wall Street Journal, The Economist, and The Atlantic. Dr. Housman received his A.M. and Ph.D. in Applied Economics and Managerial Science from The Wharton School of the University of Pennsylvania and his A.B. from Harvard University.

Leave a Comment

Scroll To Top