Why read a big book? Quantitative Relevance in the Attention Economy

In a 2016 essay in the Chronicle of Higher Education that functioned as a teaser for her book Making Literature Now, Amy Hungerford, Professor of English, boldly revealed that she refused to read David Foster Wallace’s Infinite Jest, the notorious thousand-page monster novel. Hungerford has her reasons. Among others, including misogyny and the undeserved hype created by the commercial publishing industry, she mentions the constraints on her reading time in defense of her choice of not allotting a month of her life to reading this doorstopper. She refers to Gabriel Zaid, author of So Many Books: Reading and Publishing in an Age of Abundance (2003), who “argues that excessively long books are a form of undemocratic dominance that impoverishes the public discourse by reducing the airtime shared among others” (Hungerford 2016). In her defense of not reading, Hungerford evokes the need for pragmatic resource allocation.

This makes sense. In our present-day information age, we are bombarded with unprecedented volumes of input from different channels. Especially in today’s ‘attention economy’ (Fairchild 2007), the enormous amounts of texts available, vying for our eyes and brains with other forms of information and entertainment, make the modulation and allocation of attention a daily-faced issue. The attention economy is a notion that originated in marketing, describing the principle where we assign value to something according to its capacity to attract views, clicks, likes, and shares—these are currency in a world saturated with media. Information is not scarce by any means: cognitive effort, energy, time and most importantly, attentional resources, are. By this logic, no reader in her right mind should spend a month immersed in Wallace, or Pynchon, or Vollmann.

Except that we do. These last decades have witnessed the publication of exceedingly large works, from Péter Nádas’s Parallel Stories (2005) to Eleanor Catton’s The Luminaries (2013). Also, under the influence of sophisticated forms like narratively complex TV-series, we are witnessing a revival of the serialized novel as an innovative form, shorn of its long-standing association with ‘low-brow’ culture. Authors like Vollmann and Mark Z. Danielewski have applied themselves to extended series of literary novels that take decades to write and that demand exceptional stamina from authors and readers alike.

Not only has the production of big books all but ceased with the information age, their reception is also going strong. Blurbs and reviews mention the imposing dimensions of novels like Karl Ove Knausård’s or Elna Ferrante’s as an asset, as if it were a mark of literary quality in itself, and bookish hipsters walk around with tote bags and mugs with the text ‘I Like Big Books and I Cannot Lie.’ Wherein lies the unique, and some might say uniquely anachronistic, appeal of big books in our highly competitive, fast-paced media landscape?

Mark O’Connell has expounded an interesting theory about readers’ complex relation to the long, difficult novel (he mentions William Gaddis’s The Recognitions). This relation he likens to the phenomenon of the Stockholm syndrome, the survival mechanism by which hostages start to feel a loving connection to their abductors. The victim starts to interpret every little abstention from violence as act of kindness or even love, constructing a good side to the perpetrator in order to cope with the terror of the situation. Admitting that the analogy might be a bit of a stretch, O’Connell points to the imperative posed by the monumental novel’s length, which is part command, and part challenge, a mix of pleasure and frustrations:

The thousand-pager is something you measure yourself against, something you psyche yourself up for and tell yourself you’re going to endure and/or conquer. And this does, I think, amount to a kind of captivity: once you’ve got to Everest base camp, you really don’t want to pack up your stuff and turn back. (2011)

This should then explain why ‘survivors’ of such novels tend to be more positive in their assessments than is perhaps strictly warranted by their literary merits. After you finished the reading, O’Connell believes, there is a true sense of accomplishment that stems as much from your own sense of achievement in having finished (‘conquered’) the work, as the author’s achievement in having written it. I think the reverse holds as well: quitting 500 pages in—like I did with Murakami’s 1Q84—is a bit like leaving your partner after fifteen years and two kids: you already put in so much work.

The idea that the costs associated with attention and behavior need to be counter-balanced or justified through compensatory payoffs is one that runs through the social sciences. You see it for instance in economic games, information theory approaches to, say, language processing; and in Relevance Theory, as developed by Dan Sperber and Deirdre Wilson (1985). Relevance theory deals with the allocation of cognitive resources in understanding written and oral communication. Cognitive effort can be affected by factors like logical or linguistic complexity, ease of retrieval from memory, perceptual salience, et cetera. Cognitive effects range from providing the answer to a question, strengthening a pre-existing belief, correcting a mistaken belief, to settling a doubt, or suggesting a hypothesis.

The descriptive principles of relevance theory include the so-called Communicative Principle, which holds that every “ostensive stimulus” (every linguistic utterance addressed to hearers or readers in particular) “conveys a presumption of its own optimal relevance,” whether or not this presumption turns out to be justified (Sperber & Wilson 2004, 612). Considerations of relevance help in the inferential process, as they guide the comprehension process towards an interpretation that fulfils expectations of relevance. Relevance is then the trade-off or balance between cognitive effort and cognitive effect; the higher the relevance, the more beneficial an attempt at interpretation is (Wilson & Sperber 1986; 2004). We expect what we are told or what we read to bring sufficient effect to be worthy of our attention, without causing us unnecessary effort of comprehension.

If we now apply these insights to the monumental novels under discussion, as has been done before with literary style (Cave & Wilson 20018) and modern art (McCallum et. al. 2019), we could say that here, a quantitative concept of relevance is at work. The material dimensions of these texts and their carriers are expected to metonymically point to what they ‘contain’. We often assume that big novels are large-scale both literally and sensorially, because they are in narrative scope and themes, implying a correlation between quantity and quality, which Bertram E. Jessup has called “aesthetic size.” Aesthetic size explains why we “speak with evaluative intent of a large canvas, a big building, a long poem, a major composition and a sustained performance” (1950, 31).

So, the sheer material dimensions weight, length, bulk, size, and number of pages steer our interpretations, even before the reading experience, towards a readiness to assume they will have something important to say that is not containable in books of normal size. We suspect them to be worth the trouble, to be not unnecessarily big and complicated. We expect them be of what Aristotle (1968) called ‘appropriate size’.

The cognitive effort, here, lies mostly in our scarce attentional resources, and the cognitive effect is that of an extraordinary reading experience—part pain, part pleasure. In an interview conducted by Kári Driscoll and me, Danielewski not only admitted to placing such high demands on his readers, but also suggested these readers should be warned:

You need a lot of imagination. You need a lot of skill. One of the things I’ve been toying with recently as I’m finishing volume five of The Familiar is to actually create a kind of reader rating system that somehow alerts readers, like skiers, that you are on a difficult trail. Because I feel the way that books are currently presented, everyone assumes, or in some degree feels entitled to be able to read everything that’s put out there. And I feel it’s a disservice to people who are good readers, who spend a great deal of time reading difficult books and can make their way through hard texts. It’s a lot to ask that of readers. (2018, 149)

That would be a clever ruse to carve out, and then appeal to, an elitist, skilled audience seeking to differentiate themselves from the Twilight or Hunger Games crowd. This high demand on readers’ skills makes his works an informative—albeit, in this case, failed[1]—experiment in to what extent an audience can be seduced to devote time to experimental serialized literature within an attention economy.

David Letzler (2017) has even entertained the intriguing (yet not empirically tested) hypothesis that ‘meganovels’ with a high degree of ‘useless’ text (or cruft, in his parlance), could offer a training in the allocation of attention. In his interpretation, novels like Infinite Jest or House of Leaves do cause unnecessary effort in terms of message, asking us to plow through pages and pages of irrelevant gibberish. But in doing so, they have the cognitive benefit of honing our skills in the allocation of attentional resources—skills that are of undeniable import today.

When I choose to devote my precious leisure time to high cost/high benefits stimuli like reading Bolaño’s 2666, this will obviously detract from consuming many low cost/low benefits stimuli, such as reading an article titled “24 Everyday Things You Never Even Knew Had A Purpose”, or binge-watching Crazy Ex-Girlfriend on Netflix. This comes to down strategy: the latter are less risky, whereas the former necessarily entails a gamble and hinges on a willingness to make more meaningful investment, with the risk of being let down (those hours with 1Q84 or Joshua Cohen’s Witz that I’ll never get back). This risk is especially present in the case of contemporary instead of canonical novels, as we probably trust that taking a chance on Crime and Punishment could not be a complete waste of time.

In conclusion, relevance theory helps answer the question why and how big books still enter in the competition for our attention, just like (and with!) our vacation pics on Instagram and cat videos on YouTube. If I do not, like Hungerford, say no to novels like Infinite Jest, it might just be that my investment of effort leads me to search for additional inferences that justify the effort. ‘I Like Big Books’ becomes a self-fulfilling prophecy.

References

Aristotle. On the Art of Fiction: “The Poetics.” Trans. L.J. Potts. Cambridge: Cambridge UP, 1968.

O’Connell, Mark. “The Stockholm Syndrome Theory of Long Novels.” The Millions, 16-05-2011.

Hungerford, Amy. “On Refusing to Read.” The Chronicle of Higher Education, 11-09-2016.

Cave, T., and D. Wilson. Reading Beyond the Code. Literature and Relevance Theory. Oxford: Oxford UP, 2018.

Driscoll, Kari, & Inge van de Ven “Book Presence and Feline Absence: A Conversation with Mark Z. Danielewski.” Book Presence in a Digital Age, eds. Kári Driscoll and Jessica Pressman. London: Bloomsbury.

Fairchild, Charles. 2007. “Building the Authentic Celebrity: The ‘Idol’ Phenomenon in the Attention Economy.” Popular Music and Society 30, no. 3: 355 – 75.

Jessup, Bertram E. “Aesthetic Size.” The Journal of Aesthetics and Art Criticism 9.1, 1950. 31-38.

Letzler, David. The Cruft of Fiction. Mega-novels and the Science of Paying Attention. Lincoln: University of Nebraska Press, 2017.

McCallum, Katie, et al. “Cognition, Emphasis, and the Viewer’s Experience of Fine Art.” OSF Preprints, 20 June 2018. Web.

Sperber, Dan, & Deirdre Wilson. Relevance: Communication and Cognition. Wiley-Blackwell, 1995.

Wilson, Deirdre, & Dan Sperber, On Defining Relevance. In Philosophical Grounds of Rationality. Oxford University Press, 1986.243–258.

Wilson, Deirdre & Dan Sperber. Relevance Theory. In The Handbook of Pragmatics. Oxford: Blackwell, 2004. 607–632.



[1] Announced as a 27-volume series with two publications per year, further publication of The Familiar was put on hold after the first five volumes in 2018, due to insufficient sales numbers.

1 Comment

  • comment-avatar
    Dan Sperber 6 May 2019 (17:54)

    A cognitive version of the sunk cost fallacy?
    Thank you for this thought-provoking post. I am glad, of course, that relevance theory helps address the question of why readers, far from shunning big books, often seem to assume that reading them might be particularly rewarding. As in the case of the “guru effect,” the phenomenon highlighted by Inge calls for an approach that draws not just on cognitive but also on the social sciences. She herself alludes to O’Connell comparison with the Stockholm syndrome and to Letzler idea that mega-novels serve to train attention. She doesn’t seem that convinced by these suggestions, however, and nor am I.
    Let’s try another social-science based conjecture. The so-called sunk-cost fallacy (famously studied by Kahneman and Tversky) consists in taking one’s past investment of resources (“sunk cost”) in some project as a reason to keep investing in it. A possible (non-standard) explanation of the fallacy is that investment made on such retrospective ground, while not justified from a narrowly focused economic point of view, may well make sense when reputational costs are factored in. Perseverance in one’s endeavours makes one in the eyes of others a more reliable and hence desirable partner for future undertakings. This hypothesis need not imply that people who commit the apparent fallacy do so because they take into account the potential reputational costs of an unexpected reallocation of resources. Seeing one’s own perseverance as a moral quality may be a proximate mechanism well-adapted to enhancing one’s reputation without even having to think about it.
    This conjecture about the sunk-cost fallacy suggests that the high investment of time and attention involved in reading a big book elicits a sense of moral worth that goes with perseverance seen as a virtue. This makes one reluctant to abandon the endeavour in mid-course, as if this would give gave evidence to oneself and possibly to others of moral weakness. Of course, becoming aware of the shaky morality involved should help us recognise the sunk cost for what it is and not compound it. In this light, Inge is to be commended for admitting that she gave up reading 1Q84 in the middle. I confess to not being ready yet for such a coming out.