Lakoff argues that politicians and media outlets can influence the electorate’s thinking about particular issues through framing, or the repeated use of expressions that encourage a specific viewpoint. A prime contemporary example is the expression ‘fake news.’
Lakoff argues that right-leaning politicians and outlets are particularly adept at framing. If this is the case, then we would expect to see that right-leaning outlets more frequently repeat expressions like ‘fake news’ compared to center or left-leaning outlets.
In my Language & the Mind course, we examined all instances of the expression ‘fake news’ that appeared in online media discourse up to November 2018 available on the News on the Web (NOW) corpus. We then coupled all tokens with a media bias rating (From Left to Right), as provided by the website All Sides, resulting in just over 2,000 instances.
As the figure above shows, Lakoff seems to be right. Right-leaning media outlets more frequently use the expression ‘fake news’ compared to center and left-leaning outlets (as confirmed with log-likelihood ratio tests).
These results should be taken with a grain of salt, however: there are fewer right-leaning outlets than center and left-leaning outlets in the final dataset (though frequency counts are relative to bias rating).
But now comes the real question: If this difference is true, is there an effect on the electorate? In other words, has ‘fake news’ undermined people’s opinions of the media, and is this effect stronger for people who more frequently encounter the expression?