Anyone entering the field of framing quickly realizes that framing research is a fractured field. As early as 1993, Robert Entman, a luminary of framing research, wrote: “Despite the ubiquity of framing in the social sciences, one searches in vain for a theory that explains how frames embed and manifest themselves in texts or how framing affects thinking. More than 20 years later, the results are still similar. Thus, it can still be said that there is no coherent body of theory or network of theoretical statements in the literature. Rather, there are a number of very different papers that use the term framing but do not derive concrete predictions from framing theory. This makes framing research difficult to capture and describe scientifically.
However, there are a number of empirical studies that show the effects of different framing approaches. A classic among them is a 1981 study by Daniel Kahneman and Amos Tversky on people’s willingness to take risks in different linguistic contexts. The two psychologists presented the following story to their subjects and asked them to rate it:
Imagine the United States preparing for an outbreak of a disease that will kill 600 people if left untreated. There are two alternative programs to combat the deadly disease: Program A, which will definitely save 200 lives, and Program B, which has a 1/3 chance of saving all the people or a 2/3 chance of saving none of the 600 people. Which program do you prefer?
Given the choice, 72 percent of the participants chose the first program, and 28 percent chose the second. The researchers then repeated the experiment with a control group that was presented with the story from a different perspective:
Imagine the United States preparing for an outbreak of a disease that will kill 600 people if left untreated. There are two alternative programs to combat the deadly disease: Program A would definitely kill 400 people. With program B, there is a 1/3 probability that no one will die and a 2/3 probability that all people will die. How do you decide?
Under these conditions, only 22 percent of the participants in the control group chose Program A, while 78 percent chose Program B. How is this possible?
If you look closely, you will quickly notice that both texts describe exactly the same outcome. In the first program (A), 400 people die and 200 survive, and in the second program (B), all survive with a one-third probability or all die with a two-thirds probability. Whether participants prefer the safe option or take a risk depends on how the text is presented. Kahneman and Tvelsky call this the framing effect.
The study is still cited today in psychology and economics. Kahneman and Tversky’s work has become an important pillar of behavioral economics and psychological risk research. In 2002, Kahneman was even awarded the Alfred Nobel Memorial Prize in Economic Sciences for his contribution.
These two psychologists are also referred to in communication science, albeit in a modified form. Instead of different descriptions of the same situation, people are exposed to different situations, which inevitably leads to different results. Adam Simson and Jennifer Jerit conducted a well-known study on this in 2007, analyzing the terms used by journalists and politicians in the U.S. around the issue of banning abortion. They found that proponents and opponents each used an exclusive vocabulary to promote their position. For example, supporters of an abortion ban always spoke of “babies” instead of “fetuses,” while opponents used exactly the opposite.
In a subsequent experiment, Simson and Jerit showed that people who read media reports in which only one of the two words was used were more likely to identify with one or the other political position: against abortion bans when the word “fetus” was used, or for abortion bans when the word “baby” was used. Researchers from Stanford had already presented similar results in 2005 on a completely different question. They showed that meat consumption decreased in subjects who were more frequently exposed to the term “mad cow disease” in media reports. The use of the term “Creutzfeldt-Jakob disease” (a brain disease that can be contracted by eating contaminated beef) had no effect.
In this context, the terms “fetus” and “baby” have two different meanings. While “baby” describes a child with a face and hands, the term “fetus” conveys the image of a cluster of cells. And while “mad cow disease” refers to a specific observable behavior, “Creutzfeldt-Jakob disease” is the medical term for a complex pattern of disease that is largely unknown to the general population. Thus, in contrast to Kahneman and Tversky’s experiments, framing subjects in communication science tend to be confronted with very different stories, but this does not lead to differences in the overall results.
There is another school of thought that studies framing, called cognitive linguistics. It was founded in the early 1980s by linguist George Lakoff of the University of California at Berkeley. In his 1996 book Moral Politics, he argues that much of our thinking is unconscious and that our judgments (including moral judgments) depend on conceptual metaphors activated by language. He says that when we discuss moral issues, we use a variety of metaphors to frame moral issues- that is, to interpret them, to understand them, and to assess their consequences.
Lakoff based his theory of moral politics on this view. The core proposition is that American politics is shaped by two moral worldviews supported by family metaphors:
According to Lakoff’s theory, these models of education are mapped onto national politics through the metaphor of a “nation as a family. Accordingly, the use of language is consistent with the two models and influences citizens’ perceptions of political issues through framing.
But his school of thought goes further, for he says that our ordinary conceptual system, in which we both think and act, is fundamentally metaphorical in nature. That is, metaphor is not a matter of mere words, but a holistic construct in which our thought processes take place. To be sure, all thinking begins with some non-metaphorical impressions (i.e., sensations, actions, and emotions). However, at the latest when a current impression is connected with previous ones, such as when one conceives of one’s life as a journey that can be bumpy.
Lakoff and his followers belong to the so-called constructivists. For them, there is no such thing as reality, and thus no ultimate truth. Instead, their worldview is characterized by a struggle between competing opinions.
One of the most prominent critics of constructivism, or metaphorical thinking, is Harvard University psychologist Steven Pinker. He says: Lakoff’s thinking about politics is a disaster. In Lakoff’s world, political debates are pure contests between metaphors. According to this view, citizens are not rational and do not pay attention to facts unless the facts fit into a framework that is fixed in the brain’s neural structures by sheer repetition.
But how powerful is framing today? The fact is: Lakoff’s claims about the importance of metaphors and frames are untenable from either a philosophical or a psychological point of view. There is no evidence that people always perceive and process metaphors as such. Rather, in a study, my team and I had our subjects read the following short dialog:
Liza: “I feel like this relationship is on its last legs. How can we have a strong marriage if you continue to admire other women?”
Tom: Tom: “Your jealousy is the problem.”
We assumed that readers would process the phrase “This disease has infected you” more quickly after the conceptual disease metaphor “on its last legs” than after a story that begins with the words “Love is a challenge. However, this was not the case. Instead, subjects were surprised by the phrase “This disease has infected you” and took longer to read and understand this message. The situation was different when we presented the disease metaphor in an unusual way. In other words, linguistic frames have an effect especially when they are new and surprising.
Our observation has been replicated several times, including for political framing. In another study with Swiss voters, we were able to show that voters usually respond to framing by advocating their own (predetermined) political position even more strongly than before – regardless of whether the framing supports their own position or not.
A 2006 study shows that the effect of framing can also be highly individualized. Benedetto de Martino and his colleagues at University College London studied the brain activity of their participants while they made risk decisions. The tasks corresponded to the classic experiments of Kahneman and Tversky. The results showed that the subjects in the brain scanner were affected by the framing to different degrees – depending on how hard their prefrontal cortex, which is involved in action planning, was working. This suggests that people are not defenseless against the framing effect, as long as they think about what is presented to them.
So when politicians or media people are told that framing can have an effect on the viewer, listener, or reader, they’re absolutely right. Framing effects of various kinds have long been scientifically proven. But there are different views about the strength, direction, and limits of this effect. There is the camp around the linguist Lakoff, according to whom framing could change the world order. But this is total nonsense and only serves for self-promotion! Rather, framing can be used by an organization to polish its own image, but only if the frame fits reality. The Trump opponents (and not only them) realize this.