Normal View Dyslexic View
Commentary
Art and civic values: the role of fiction-reading
Abstract
The quality of a civic life is to some extent dependent on its citizens’ capacity for empathy, imagination, and the appreciation of the varieties of experience that shape us. Many have argued that fictions of various kinds can enlarge these aspects of mind. Philosophers are among them, though they have rarely acknowledged that the claim needs serious empirical support. Psychologists, meanwhile, have been searching for the evidence. I reflect on a recent project across the disciplines of philosophy and psychology that sought to extend the evidence a bit, as well as providing a richer understanding of the explanatory options. At the end of our study we undertook a large-scale meta-analysis; I summarise our findings, commenting on their implications for bias, and their limitations. I address the unease people in the humanities sometimes express about this kind of empirical work. (This article is published in the thematic collection ‘The arts and humanities: rethinking value for today—views from Fellows of the British Academy’, edited by Isobel Armstrong.)
Keywords
fictionreadingempathyimaginationmindbiasThere are plenty of reasons for valuing the arts. Many have nothing to do with any contribution to civic life they might make. But perhaps the arts do also contribute to the quality of a community’s civility, its openness, its capacity for rational and humane decision-making. Having written for many years about the art of fiction, what it is, how it works and why it interests us, I recently came to see that, if works of fiction sometimes contribute these things, there ought to be evidence that goes beyond personal conviction. Philosophers have been particularly fond of claiming that this or that work or genre of fiction is a rich source of insight into the inner workings of the mind, or an instructive lesson in the complexities of moral choice, or a mechanism for expanding our empathic powers. Those views are regularly debated in the discipline. Rarely does it get asked whether there is robust evidence to support them; evidence, that is, for the existence of these effects and not merely for our believing in them. In philosophy-speak, the issue seems to have been treated as one for a priori reflection. The reflections of philosophers may be useful in this inquiry; they can help formulate a question or make a relevant distinction. But what we should be aiming for, beyond clarity, is evidence.
I’m pleased to say that the desire for evidence is now far from controversial within philosophy. Work on free-will, mind and action, reason, responsibility, and perception routinely draws on and elaborates empirical theories from the natural and social sciences. There is even such a thing as experimental philosophy, much of which is focused in testing the extent to which the ‘intuitions’ on which philosophical argument is often taken to depend are widely shared.
I soon discovered that the effect of fiction-reading is something empirical psychologists, the ones who do experiments and analyse the data that emerges, have been seeking evidence for in recent years. Had they found it? Not much of it, was my conclusion. They were going in the right direction but not very far or fast. They were often conducting small-scale experiments that involved asking people to read a brief fictional narrative and then to undergo tests designed to assess levels of empathy and ‘mind reading’, a common expression in the literature for our capacity to comprehend the beliefs, desires and intentions of others and ourselves. They also undertook correlation studies that related mind reading and other skills to lifetime histories of reading. These studies suffer the obvious disadvantage that they may be registering the fact that fiction-reading and mind reading have a common cause, or that people with better mind reading skills may be more attracted to fiction. Yet another concern was that a lot of the studies depended on self-report, with the cognitive effects of reading being measured by what people in the tests said the effects were; there is a lot of evidence to be sceptical of the reliability of such reports. Finally, many studies were so small in scale as to make it likely that genuine effects of fiction-reading would not be detected by them.
So here we have a disconnect between two projects, illustrative of the ways that disciplinary boundaries can impede thought. Philosophers rarely if ever brought forward more than anecdotal evidence for their claims about fiction’s capacity for moral and social enlargement. The psychologists often seemed to be operating with an implausibly mechanistic notion of fiction’s capacity for bringing about personal change. Who would pin their faith in fiction on its capacity to effect this sort of change after half an hour of fiction-reading?
Thinking it might be helpful if the philosophers and the psychologists joined forces, Stacie Friend (philosopher, Edinburgh), Heather Ferguson (psychologist, Kent), Lena Wimmer (psychologist, Freiburg) and I have worked on this for some years.1 In the wonderful jargon of contemporary life, we produced a number of outputs, including some experimental papers. These were designed to look a bit more deeply into the question of evidence, to use more substantial and varied exposure to fiction-reading than had been typical till now, to tease apart the cognitive effects of fiction and narrative (sometimes conflated in the psychological literature to that time), and to look at the way that different forms of narrative voice might be affecting readers. Covid hit just in time to limit our ambitions, and we had to move our studies from the lab to online, junking some projected tests. We did find some small positive effects; I suppose we made merely incremental progress, as the REF (Research Excellence Framework) sub-panel would crushingly say. Interestingly, we found no significant relationship between fiction-reading and what is called ‘moral cognition’ as measured by tests that range from self-report to implicit measures of proneness to prejudice. No one, I think, would claim that these tests probe the messy reality of moral thought and action in very sensitive ways.
Here I’ll say something about a part of our work with broader applications, a very recent large-scale assessment (a ‘meta-analysis’) of the experimental work (published and unpublished) which has been done on the cognitive effects of reading fiction, mostly over the last thirty years, including some of our own. We reanalysed data from 130 studies that together involved over thirty thousand participants in a variety of experiments. I believe this study, just now in print, is the best picture we have of the current evidential situation (Wimmer et al. 2024).
I will focus on a few high-level points, and then turn to the broader issue: what value should humanities scholars attribute to such studies? Our headline conclusion was that there is ‘robust evidence for a small-sized positive relationship between reading fiction and cognitive benefits’. More particularly, we concluded that ‘short fiction reading assignments cause small cognitive benefits and that lifetime exposure to written fiction is related to small cognitive enhancements’. Though small, the effects were shown to be larger than the effects of reading nonfiction, both expository and narrative.
This summary hides a lot of important qualifications. One I have mentioned: correlation is not cause, and the relation we noted between lifetime reading and cognitive effects may be only correlational. Secondly, as concerns experiments where people read fictional stories and then undertook tests, we acknowledge that the (small) positive effects on performance may be just transient priming responses. This sort of effect is commonly observed in psychology. One study (not concerning fiction) found evidence that ‘imagining a professor’ for a few moments can improve your performance on certain cognitive tests, perhaps because you are pushed unconsciously to imitate for a few moments the imagined thought processes of this fabulous figure: careful, thoughtful, attentive. Imagining a professor for a few moments won’t make you a more careful, thoughtful, attentive person in the longer run. Still, it is not worthless to find a positive correlation; if we had not found at least that, the case for fiction’s capacity to effect change in outlook would look very bleak. And transient effects are also not worthless, especially if they are refreshed by habits of regular reading.
Importantly, very few of the studies we analysed sought to assess the negative effects (if any) of reading fiction. While I suppose it will be widely agreed that certain particular fictions or kinds of fictions have a capacity for harm, a more contentious issue is whether there are harmful effects of reading fiction in general: a tendency perhaps to disengage from the world, to direct one’s emotions and one’s agency towards imagined beings, with correspondingly less left over for the real people who deserve your support and concern. Other worries include evidence that readers often fail to distinguish between what is true in a story and what is actually true, picking up false factual beliefs in this way with some ease. Empathy itself is said by some to be a compromised vehicle for moral thought, more easily deployed when thinking about people like ourselves than about those culturally more distant, and focusing us unrealistically on individual cases at the expense of the overall balance of welfare. I don’t rule out these possibilities, though we should recall that negative tendencies often go along with positive and countervailing ones. We might believe in the possibility of such effects while at the same time thinking that the overall impact of fiction tends to be positive. As it was, very few of the studies we examined looked at negative effects, and we could draw no real conclusions.
A tendency to look for positive rather than negative effects is one source of bias in these sorts of studies. Something else we looked at may be of interest for the topic of bias, given the replication crisis that has hit experimental science. Results cited over decades sometimes don’t hold up when the tests are repeated, generating concerns about ‘p-hacking’—using a range of strategies that manufacture significant-looking results from data that is in fact non-significant. We ran tests to detect p-hacking (yes, there are such tests and they are good at finding it) and did not find it across the studies we looked at. But we could not rule out a skewing of results due to the ‘file drawer problem’, the tendency of studies to remain unpublished when they don’t identify a significant effect.
Discussions I’ve had with colleagues in the humanities about the work we have been doing indicate occasional unease with this enterprise. People worry that failure to find measurable and beneficial effects of reading will be used against the arts by funders keen on seeing a practical return, especially where—as I have indicated—there is good reason to think that real and lasting effects are simply very difficult to measure. There is also a view according to which we misunderstand such categories as fiction, literature, and art in general if we suppose them to bear the kinds of systematic relations to reality necessary for them to be guides to good life and conduct. In particular, the creatures of fiction belong to the narrative realm, not to the world of biological and social beings. This last objection will take more careful responding that I can provide for it here. But briefly, my answer, so far as fiction is concerned, is that the imagination is an evolved capacity for planning that hooks into various cognitive competencies we use in dealing with the real world: inference, motor systems, mind reading among them. We can be prompted to exercise our agency towards the real world by narratives of imagined events that we construct—or which authors of fiction construct for us. No one is claiming that all fiction does that, or does it in ways that promote human flourishing. I am content to say at this stage only that there is no incoherence in the idea of imagined constructs that guide thought and action in effective ways.
For the rest, I agree that the empirical tests currently available are insufficiently sensitive to what we are probing for, but they won’t get better if we give up testing altogether. Nor will we win arguments with funders by refusing to consider the issue of evidence. Nor, finally, is an insistence on the search for evidence anti-humanistic or ‘scientistic’. What contributes to civic life and what does not will depend of course on how we characterise civic value. An important contribution the humanities can make to this issue is to articulate such a conception, though it is unrealistic to think we will all agree on one. While that debate is going on, we still do and should ask sensible, urgent, and factual questions about how well this or that public policy measure will reduce levels of mental ill health or improve levels of community cohesion. I suggest we look at the questions about the cognitive value of fiction in the same way.
I’m aware that this is one tiny corner of the area marked ‘the arts and civic values’. The work I have described tells us nothing about the role of music or the visual arts, or about narrative fictions in screen and drama (again, we did not have enough studies in those areas to get worthwhile results). We looked at a restricted range of cognitive capacities, using a range of tools agreed to be very imperfect in how they relate to psychological reality. And how changes in the psychological dispositions of individual agents affect and are affected by institutional structures is a whole other subject. But work across the disciplines is easier to admire than it is to implement, and I am glad to have done just a little by way of implementing it.
Acknowledgements
The thumbnail image that accompanies this article is by John Michael Thomson on Unsplash.
Reference
WimmerL., CurrieG., FriendS., Wittwer J. & FergusonH.J. (2024), ‘Cognitive effects and correlates of reading fiction: two preregistered multilevel meta-analyses’, Journal of Experimental Psychology: General, 153(6): 1464–1488. https://doi.org/10.1037/xge0001583
1
Funding in earlier years was provided by the Leverhulme Foundation (grant number RPG-2017-365).
Cite
DOI: 10.5871/jba/012.a37
Published on: 17 September 2024
Volume: 12
Issue: Issue 3
Article view count: 104
Article download count: 8
Copyright statement
© The author(s) 2024. This is an open access article licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International LicenseCite this article
Currie, G. (2024), ‘Art and civic values: the role of fiction-reading’, Journal of the British Academy, 12(3): a37 https://doi.org/10.5871/jba/012.a37Crossref
Dimensions
Altmetric
Article metrics
View count: 104
Download count: 8
No Data Found
No Data Found
No Data Found
Received
October 2022
Accepted
May 2023
Published
July 2023
The books that made me: Gary Younge Gary Younge, Amelia Horgan
From the British Academy's 2024 Summer Showcase, Professor of Sociology at the University of Manchester and Honorary Fellow, Gary Younge, discuss the literature that has influenced his life, intellectual development, and current work, with writer and assistant lecturer at the University of Essex, Amelia Horgan. From childhood memories of reading to present-day reflections, they delve into both fiction and non-fiction books that have shaped and inspired Gary’s thinking.
13 December 2024
The disregard of mothers’ knowledge and experiences in violent extremism discourse in Kenya Beatrice Kizi Nzovu, Fatuma Ahmed Ali
This article argues for the inclusion of women’s epistemology in discourse about violent extremism and approaches to tackling it in Kenya. It focuses on mothers of male recruits to violent extremist organisations, arguing that, although mothers have critical insights to offer, their knowledge and experiences remain unacknowledged and unheard in Kenyan responses to violent extremism. Although women, including mothers, are understood to be useful contributors to the fight against violent extremism, their voices remain peripheral in masculinised discourses and actions. This article uses an African feminist theoretical approach, informed by ‘Motherism’, and gendered peace – as well as security frameworks including UN Security Council Resolutions (UNSCR 1325 and 2242) on women, peace and security and women’s inclusion in efforts to address violent extremism – to argue that policy development and implementation processes in Kenya have failed to capture the meaningful contributions that recruits’ mothers can make to addressing violent extremism.
22 June 2023
No Data Found
