Wednesday 6 June 2018

The Lowbrow Status of Science Fiction

In trying to find out what makes a piece of fiction ‘literary’, Chris Gavaler and Dan Johnson of Washington and Lee University dove into papers from literary criticism and psychology that each proposed different definitions of ‘literariness’. Among the various definitions they came across, they found one that suggested that a text’s literariness might depend on how much it encouraged the reader to infer what characters are thinking or feeling (Kidd & Castano, 2013), an ability known as theory-of-mind. Gavaler and Johnson seized on this idea, hypothesizing that a literary text might contain the fewer explanations of a character’s state of mind, forcing the reader to make more inferences. These researchers were also interested in how this idea might interact with genre fiction, specifically the genre of science fiction. So they devised an experiment with four conditions. Participants would read a brief passage of science fiction or a realistic story, with or without explanations of the character’s state of mind. The authors would then measure readers’ ratings of literary merit, comprehension of the text, and inferencing effort. 

The researchers found that those who read a text with explanations of a character’s state of mind understood the text better and rated the text as having greater literary merit, compared to those who read texts without these explanations. This was regardless of whether the text was science fiction or not. Those who read a science fiction passage rated it as having less literary merit than did those who read a realistic passage. They also did not understand the text as well and made less of an effort to infer the character’s state of mind. This was the case whether or not the passage included explanations of the character’s state of mind. 

A subsequent experiment built on this research to examine the reader’s construction of a world’s physical and social rules, as well as their efforts to understand the plot. The results were similar to those of the first experiment. Interestingly, readers of science fiction exerted more effort in understanding the world of the narrative, yet did not understand the world as well as those who had read a text that was more realistic. Those who read science fiction also put in as much effort to understand the plot as did those who read a realistic passage, yet did not understand the plot as well. 

Long story short, it appears that the science fiction genre prompts a style of reading that is less attentive to characters, more attentive to the world of the narrative, and yet results in poorer overall comprehension. These experiments also show that the perception of literary merit seems to be tied to the inclusion of explanations of a character’s thoughts and feelings, and not the absence of such explanations, as the authors initially thought. 


Gavaler, C., & Johnson, D. R. (2017). The genre effect. A science fiction (vs. realism) 
manipulation decreases inference effort, reading comprehension, and perceptions of 
literary merit. Scientific Study of Literature, 7(1), 79-108. doi:10.1075/ssol.7.1.04gav

Kidd, D. C., & Castano, E. (2013). Reading literary fiction improves theory of mind. Science,
342, 377–380. doi: 10.1126/science.1239918

Post by Krithika Sukumar

* For a copy of the original article, please contact R. Mar (see profile for e-mail).

Bookmark and Share

1 comment:

Laura Vivanco said...

"The researchers found that those who read a text with explanations of a character’s state of mind understood the text better and rated the text as having greater literary merit"

That's really interesting, because popular romance fiction is generally not considered at all literary but it's full of "explanations of a character’s state of mind". This makes me think that the presence of "explanations of a character’s state of mind" can't be what determines whether or not a text's deemed to be literary or not.

One problem with assessing what's considered "literary" is that some of the works considered "literary" are hundreds or thousands of years older than others, and they come from many different societies with differing ideas about genre, characterisation etc. Many of them will need to be translated, and that creates a whole new set of issues. I strongly suspect that all this makes it really difficult to find some universal, objective feature in them which could be used to assess literariness.

On the other hand, if we're looking at contemporary readers, there may well be features common to those readers in terms of how they've been trained to recognise what is or isn't literary.

Pamela Regis ( ) has suggested that literary works are assessed according to "complexity, contemptus mundi, and social justice" but that complexity is very often in the eye of the beholder so if a reader has pre-categorised a work as "literary" they'll assume it's complex, while if they've decided it's junk they won't even bother looking for complexity.

Giving readers chunks of texts devoid of their paratext could get round the risk of prejudgment to some extent, though if the scene describes something which is stereotypically associated with a particular genre then that might also set off the assumptions about non-literariness. However, I have a feeling I've come across comparisons between, say, certain scenes from D. H. Lawrence and the equivalents from a romance novel and readers couldn't necessarily tell which of those was the "literary" text and which wasn't.

Related Posts Plugin for WordPress, Blogger...