Individuals holding strong political opinions on immigration can exert considerable impact by composing tweets rich with metaphors, influencing how others comprehend this intensely debated topic.

In a recent investigation, scholars from the University of Michigan and the University of Chicago examined information about immigration tweets from two political factions. They specifically employed a software application to analyze metaphors found in over 400,000 communications from U.S. citizens during 2018 and 2019.

The researchers chose seven themes for their examination: animal, vermin, parasite, physical strain, water, goods, and conflict. Each theme creates a unique rationale regarding the perceived dangers and viable solutions.

Metaphors associated with these themes can either be explicit or implied. For instance, one might clearly dehumanize immigrants by labeling them as “animals,” but could also subtly invoke animal imagery by discussing “swarms” of individuals “flocking” across the border.

Dehumanizing metaphors may be utilized by individuals demonstrating strong resistance towards immigrants, such as through the “water” metaphor when expressing concerns over immigrants “pouring” into the nation, as highlighted by the study. However, numerous metaphors have become so ingrained in our discourse about immigrants that they are regarded as normal and unbiased, as seen in references to “waves” of immigration throughout U.S. history.

“Utilization of dehumanizing metaphors can escalate discrimination and severe policies,” noted study co-author Ceren Budak, associate professor at the U-M School of Information.

Julia Mendelsohn, postdoctoral researcher at the University of Chicago and lead author, remarked that the influence of strong beliefs on metaphor usage varies and is somewhat intricate between the two political parties.

Individuals identifying as very conservative employ more metaphors than those who are more centrist. However, the situation is different for liberals, as those who lean significantly left do not utilize as many metaphors involving water or transactions, but do favor more metaphors related to living beings, like animals.

The outcome relates to the variance in retweets, Mendelsohn stated. When individuals incorporate creature metaphors in their tweets, these messages are generally retweeted more often, with this phenomenon being more pronounced among liberals than conservatives.

“Thus, the contrast between liberals’ tweets with and without creature metaphors is larger than the contrast between conservatives’ tweets with and without creature metaphors,” she explained.

The researchers indicated that they applied large language models, specifically ChatGPT, to pinpoint metaphors. Such models can serve as valuable instruments for comprehensive analysis of language in political discourse, but meticulous study design and human assessment of these models are essential, Mendelsohn stated.