When Congress Speaks (and Posts and Tweets)
Sigmund Freud, who wrote about “slips of the tongue,” was one of the first psychologists to see language as a device for exploring the human mind. He was by no means the last. Throughout the 20th century psychologists developed many coding schemes to investigate what people say in order to better understand their mental lives.
Analyzing texts was once done only by hand, and it was back-breaking work for researchers who were willing to undertake it. Nowadays, we have computer technologies that can automatically analyze mountains of words in order to detect themes and patterns. One example is the popular Linguistic Inquiry and Word Count (LIWC) program developed by Jamie Pennebaker of the University of Texas. One of my former Ph.D. students, Joanna Sterling, and I used LIWC to quantify the ways in which liberals and conservatives use language differently.
First we looked at the use of language in nearly 25,000 English-speaking Twitter users and found that liberals and conservatives did indeed use language differently. Liberals used more benevolence language, including words and phrases such as improve, benefit, care for, assist, enhance, nurture, and lend a hand.
Conservatives, on the other hand, were more likely to use language emphasizing all of the following themes:
- Threat, such as hurt, warning, terror, loss, and risk
- Security, such as safety, defend, guard, protect, and shield
- Tradition, such as family, faith, religion, custom, and foundation
- Resistance to change, such as keep, normal, hold, continue, and prevent
- Certainty, such as all, every, always, sure, indecision, and never
- Power, such as big, up, God, and win
- Anger, such as kill, fight, hate, attack, and murder
- Anxiety, such as fear, doubt, worry, afraid, and stress
- Negative emotion in general, such as bad, miss, problem, and wrong
Many political scientists believe that ordinary citizens take all of their ideological cues from political elites, such as members of the U.S. Congress—in other words, believing what respected authorities tell them to believe about politics. This is not our view: we believe that even among people who are not very engaged in politics there is a meaningful connection between how people think (their psychology) and what they think (i.e., their ideology). Nevertheless, the question of whether liberal and conservative members of Congress use language differently from one another—and the extent to which their language use parallels liberal and conservative members of the public—is an inherently interesting one.
Therefore, we analyzed the language used by members of the U.S. Congress on Twitter (88,874 tweets), Facebook (15,636 posts), and the floor of Congress (6,159 speeches) over the same four-month period in 2014. Of course, speeches and other forms of communication issued by legislators are often written by staffers, but they tend to be quite ideologically similar to the politicians for whom they work.
Consistent with our earlier findings based on tweets sent by ordinary citizens, conservative legislators used more language pertaining to threat, risk, inhibition, power, religion, and—only on the floor of Congress—tradition and resistance to change. Liberal legislators, on the other hand, used more language pertaining to benevolence, affiliation, achievement, and—on the floor of Congress—universalism, stimulation, and hedonism.
Of course, there were also many categories of language use on which liberal and conservative elites did not differ, especially when they were on the floor of Congress. For example, there were no consistent differences in the use of long words, swear words, or language that signified anxiety, anger, self-direction, or future orientation. Overall, while there were a number of telling differences in the communication of liberal and conservative legislators, these differences were weaker and less extensive than what we found for ordinary citizens.
One day the kinds of tools that researchers have been developing to study patterns of mass and elite forms communication may help us to understand not only the ways in which leaders influence their followers (and vice versa) but also the dynamics of social movements as they are still taking shape. This is because language is one of the strongest cues to what we are thinking and feeling, as individuals, groups, and even entire societies.
For Further Reading
Barberá, P., Casas, A., Nagler, J., Egan, P. J., Bonneau, R., Jost, J. T., & Tucker, J. (2019). Who leads? Who follows? Measuring issue attention and agenda setting by legislators and the mass public using social media data. American Political Science Review, 113, 883-901. https://doi.org/10.1017/S0003055419000352
Jost, J. T., & Sterling, J. (2020). The language of politics: Ideological differences in congressional communication on social media and the floor of Congress. Social Influence, 15, 80-103. https://doi.org/10.1080/15534510.2020.1871403
Sterling, J., Jost, J. T., & Bonneau, R. (2020). Political psycholinguistics: A comprehensive analysis of the language habits of liberal and conservative social media users. Journal of Personality and Social Psychology, 118, 805–834. https://doi.org/10.1037/pspp0000275
John T. Jost is a Professor of Psychology, Politics, and Data Science at New York University. He is the author of A Theory of System Justification (Harvard University Press, 2020) and Left & Right: The Psychological Significance of a Political Distinction (Oxford University Press, 2021).
