By Kiri Scully
We use words to express our ideas, but the words don’t always dictate our meaning – two people could be speaking the same language, using the same set of words, and meaning entirely different things.
This happens every day across the world’s media, as political leaders us powerful words to further their causes. A new study has shed light on just how pervasive this is, by looking at the way US Presidential candidates used certain words in the 2016 election.
Through semantics, linguistics researchers study meaning – they look at the logic of words, phrases and texts to decipher how they are interpreted and how they form rhetoric, using language in a way that makes it more persuasive. In a recent study, a group of researchers from Penn State University analyzed the way Donald Trump and Hillary Clinton used language in the run-up to the 2016 election in the US, showing that words and their associated meanings changed depending on who said them, and when they were being said.
The researchers conducted a series of studies that focused on the language Clinton and Trump used in speeches. What they found was a stark difference between what the candidates said and what they meant, despite the clear crossover of key terms.
For example, Trump was fond of the word “deal” and used it in a vast range of contexts. He even managed to associate the business-related word with “family” and “education,” which are not traditionally linked to it. Clinton, on the other hand, used “education” when referring to “women” and “family,” associating the word more with issues of equality and access to education.
Image (CC) Pixabay
In a series of studies, the researchers used computer algorithms to track various politically charged words that the presidential candidates used in their speeches over a three-year period, such as “minority,” “spending” and “justice.” They tracked 213 single words and 397 phrases and looked at what words appeared together to determine associations. In doing so, they found that word associations changed over time.
“In a lot of ways, it’s worse than speaking two different languages,” said lead researcher Ping Li, professor of psychology and associate director of the Institute for CyberScience. “If, for example, I speak Chinese and you don’t, you have no idea what I’m saying. But, if we’re both speaking English and you think you know what I am saying, but don’t get what I actually mean, or worse, think it means something different, it can be really confusing.”
In a study that followed, the researchers focused on 324 participants who were eligible to vote in the elections and examined their word associations using machine-learning algorithms. Interestingly, they found that they could predict each participant’s political view and which political candidate they were more likely to vote for.
“We were able to predict voters’ reported political affiliations with a relatively high degree of accuracy simply based on the way they organized a list of 50 political concepts, that is, how they grouped these concepts,” said Prof. Li. “This also suggests the complex interdependence of language, speech and culture.”
Perhaps the most interesting point of the study is that this striking semantic divide – a split in the meaning of words – seems to be growing.
“What you see is that the parties have become farther and farther apart as time goes on,” added Prof. Li. “In other words, for the same word, people tend to associate different words for them and, hence, convey different meanings.”
What do you think this research could mean for the future of dialogue? Could we reach a point where we find it impossible to discuss politics and understand each other’s points of view? And what would happen if we did?