Skip to main content

The Globe and Mail

What our language choices reveal about our personalities

In his new book, The Secret Life of Pronouns: What Our Words Say About Us , social psychologist James W. Pennebaker, chairman of the department of psychology at the University of Texas at Austin, looks at the hidden psychological messages our language choices reveal.

In your book, you argue that the pronouns we use provide clues about our personalities and relationships. 'I' seems to stand out above the others, and you say it's especially good at exposing social hierarchies. Can you explain?

Almost immediately, two people in any given interaction subtly agree on who is the higher status person. Their language usually matches that, with the higher status person using 'I' less and the lower status person using 'I' more. In a relationship, the lower status person is more self-focused, anxious and self-conscious, while the higher status person is looking around surveying their world. So a lower status person might say, 'I think it's cold outside' compared with a higher status person who says, 'It's cold outside.'

Story continues below advertisement

You write how people use 'I' less when they lie. Is there a relationship between power and lying?

I think to some degree there is. Though we might not call it 'lying' in this case, but a head honcho who is deceptive, maybe to maintain confidentiality or to keep their cards close to the vest.

So if we meet someone and we think, 'Hmm this person doesn't use "I" a lot,' are we to think this is a high-status person or a deceptive one?

Here's the good news/bad news: It's almost impossible to tell if someone is using 'I' words at high rates – we can't hear it. A deceptive person might use the word 'I' 5 per cent of the time, a non-deceptive person 7 per cent. My computer jumps with joy over that, it appreciates that huge difference. But our ears can't pick it up.

Women use 'I' more than men, you write. But over the past 50 years as women's status has risen, has their 'I' usage gone down?

If they are working in the corporate world, yes. What's interesting about this is it's not as if you and I sit there and say, okay, I'm in this environment so don't use 'I' too much, or okay, use 'I' a lot. It just happens; it's automatic. It's almost magical how quickly we adapt.

Sometimes it's not us adapting to the environment, as you say, but to our life circumstances. You used the example of former New York mayor Rudy Giuliani.

Story continues below advertisement

Yes, Rudy Giuliani early in his administration was always viewed as cold, arrogant and detached. Then in a span of a few weeks in 2000 his life turned upside-down. He was diagnosed with cancer, his marriage fell apart, he dropped out of the senatorial race against Hillary Clinton. About a month later, The New York Times published an article saying his personality changed, that now he was a warm, nice guy. I looked at his press conferences and sure enough he had profoundly changed his use of language. Among other things, he was using 'I' at high rates and 'we' at much lower rates. 'We' among politicians is often a sign of being cold and detached. Upheavals like that do a lot to us.

War could be considered the ultimate upheaval for politicians. Do leaders facing the prospect of war display subtle language changes with their pronouns – fewer 'I's and more 'we's?

Yes, we found this with George W. Bush before he went into Iraq, Harry Truman before he dropped the atomic bomb, Adolf Hitler before he marched into Poland and others.

Opposite to war on the human relationship spectrum is love. Describe 'language style matching.'

The more people become engaged in each other, the more their language starts to match.

And this isn't just theoretical, you have a mathematical model to measure this?

Story continues below advertisement

Essentially we look at the degree to which two people use nine dimensions of function words, such as personal pronouns, conjunctions, auxiliary verbs, articles, prepositions etc. We have a simple measure of how similar two people are matching along each of these dimensions. It can go from we don't match at all, to we are perfectly matched. Mathematically, the number lies between 0 and 1.0.

So when people interact, if they start speaking like each other, they're more compatible. You've used this measure to predict, among other things, the success of speed dating.

I love this finding. We track their language while they talk to each other for four minutes and, using our style matching computer program, we can do a better job at predicting if they'll go on a subsequent date than they do. We also looked at young couples in love. They agreed to be in an experiment in which they gave us 10 days of their instant messages. We found we could predict if they'd be together three months later better than they could.

Did that shock you?

In a deliciously positive way.

People would pay a lot of money for that, any thought about a commercial application?

Oh, I'm sure there is. But at the website one of the exercises allows people to do that free. They can put in a text sample, like an e-mail, from themselves and from somebody else and see how closely the two are in sync. I was just at dinner last night with someone who used it and she was saying that her current lover matches her a lot better than her old lover did. I found that beautiful.

This interview has been condensed and edited.

Special to The Globe and Mail

Report an error
Comments are closed

We have closed comments on this story for legal reasons. For more information on our commenting policies and how our community-based moderation works, please read our Community Guidelines and our Terms and Conditions.