Sick of Metaphors: Reading Shiller’s Narrative Economics
- Book Review of Narrative Economics: How Stories Go Viral and Drive Major Economic Events. by Robert J. Shiller.
- And if you wish to adorn, borrow the metaphor from something better in the same genus, if to denigrate, from something worse.
- —Aristotle, Rhetoric III, 1404b
It is an odd experience to be reading Robert J. Shiller’s new book, Narrative Economics, in the midst of the COVID-19 pandemic. Shiller’s argument is that economic ideas are spread by narratives and that the spread of narratives is most accurately understood by considering them as epidemics—like Ebola or AIDS. As a result, Shiller’s book is saturated in references to epidemics, viruses, contagions, and infection.
It seems as if Shiller’s focus on epidemiology as a way to understand the spread of economic ideas should make the book feel up to date in significant ways. Oddly, there is nothing like the frightening reality of an actual epidemic to make a metaphorized one feel out of date.
I am not an economist. I wrote a short piece a few years back criticizing the elitist stance of Shiller and Akerlof’s book Phishing for Phools. (Some of this elitism reappears at various points in Narrative Economics, such as when Shiller discusses fake news and its effects on economic outcomes.) In general, I prefer to leave extended discussion of Shiller’s economics to my more qualified colleagues. However, I do have a PhD in English literature, and my dissertation was on literary representations of illness, so I am more than willing to take on Shiller’s discussions of narrative and illness, and particularly his use of disease as a metaphor.
To begin with, I’m not certain that Shiller has much respect for narrative. For example, he notes casually that “In attempting to be vivid, storytellers often resort to fiction or fake news” (page 84)—an insulting parallel that would surely startle most novelists. He argues that we can “learn something about popular economic narratives by counting words and phrases in the digitized texts that are available”(276)—a reduction of narrative to a vocabulary word search that ignores the importance of such factors as context, genre, and tone. And when he presents narrative economics as “an adventure in consilience” (12) he leaves literary studies (and philosophy!) off his graph that charts the use of the term “narrative” in academic disciplines, and limits his discussion of “Literary Studies and Narrative” to two paragraphs on structuralism and one paragraph referring readers to Morson and Shapiro’s Cents and Sensibility, which has its own challenges in dealing responsibly with narrative.
Next, it is not at all clear to me what Shiller means by narrative. He defines it early on as not “synonymous with story” but rather as a specific kind of story that serves an “explanatory or justificatory” purpose (xi). In practice, however, Shiller sometimes uses “narrative” to mean a story with a didactic point. Sometimes he uses it to mean any story or sequence of events shared among people. Sometimes he uses it to refer to specific economic events in history. Sometimes he uses it to mean a symbol (for example: constellations and brand logos). That makes whatever Shiller means by “narrative” pretty hard to track.
For example, Shiller’s discussion of the song “Happy Birthday” begins by referring to the song as “another example” of an epidemic economic narrative. Shiller grants that “It is probably not an important economic narrative. Some might say it is not even a real narrative because the words of the song do not tell a story” (97). Indeed, anyone who attended to Shiller’s definition of narrative at the beginning of the book—that it is a specific kind of story that serves an explanatory or justificatory purpose—has good reason to feel that the goalposts have shifted. They shift even more when Shiller asserts that:
- … there is a story attached to the song in practically everyone’s consciousness… The story is this: Based on a long tradition that goes back generations, people have assembled to celebrate the birthday of a loved one. After someone announces that the ceremony is about to begin, a birthday cake is brought in with flaming little candles… The birthday person makes a wish and attempts to blow out all the candles with one breath in order to make the wish come true….Sometimes additional words are added to the song, such as “And many more to you”… The ceremony ends with applause. (97-8)
It is unclear to me that what Shiller calls a “story” here is a narrative and not a definition. Even if it is a narrative, and even if that narrative is “epidemic” in our culture, it turns out that this is not actually a narrative about the aspect of “Happy Birthday” that Shiller is interested in. He is interested in the details of the history of copyrighting the song—an interesting economic discussion that is only very tangentially related to things like modern Western traditions about birthday celebrations. It’s intriguing economic history. I’m not at all sure it’s a narrative. Even if it is, it’s not the narrative that Shiller began by outlining. And it’s not clear to me what Shiller hopes to gain by characterizing it as such.
Making Shiller’s use of the term “narrative” even more difficult for readers is his tendency to muddy his discussion of the epidemiology of economic narratives by using examples from the world of publishing and of medicine. On pages 60-61, for example, Shiller engages in a long discussion of the history of book jackets. Throughout, he treats book jackets as epidemic narratives, (which here seems to mean something like “advertising”) never noting the blurring lines between the narratives contained within the texts inside the book jackets, the narrative he is creating about the book jackets, and whatever narrative qualities may be possessed by the text and images on the book jackets. He runs into similar problems when he discusses events from the world of medicine—like people ignoring advice from their doctors—and blurs the lines distinguishing his metaphorical use of epidemics and medicine from his use of factual details about medicine.
“Shiller’s book blunders into a significant conversation among literary scholars that Susan Sontag began more than forty years ago, without referencing or, apparently, being aware of that discussion at all.”
But even if one is able to keep straight Shiller’s varied definitions of narrative and his overlapping use of metaphorical and literal examples from the same fields on the same page, the book has a larger and much more serious problem. Shiller’s book blunders into a significant conversation among literary scholars that Susan Sontag began more than forty years ago, without referencing or, apparently, being aware of that discussion at all.
Sontag’s book Illness as Metaphor (1977) and its sequel AIDS and Its Metaphors (1988) are crucial works for anyone who wants to—as Shiller does—use disease to mean something other than itself. Sontag cautions that reading an illness metaphorically is to read it morally. And with a moral reading comes a moral judgement:
- Nothing is more punitive than to give a disease a meaning—that meaning being invariably a moralistic one. Any important disease whose causality is murky, and for which treatment is ineffectual, tends to be awash in significance. First, the subjects of deepest dread are identified with the disease. The disease itself becomes a metaphor. Then in the name of the disease (that is, using it as a metaphor), the horror is imposed on other things. The disease becomes adjectival. Something is said to be disease-like, meaning that it is disgusting or ugly. (58)
When we make a disease a metaphor, we import all our feelings about it onto the thing that is metaphorized. And, with the possible exception of the expression “an infectious laugh” those feelings are not positive. We do not wish “a plague [of joy] on both their houses.” And we do not call a particular idea a “cancer” because we wish it would spread faster.
Sontag notes that epidemic diseases—Shiller’s key metaphor—are “a common figure for social disorder” because they are so widespread. They easily allow our feelings about social evils to be “projected onto a disease” and our feelings about the disease to be projected onto the parts of society we don’t like. (58)
Sontag even explicitly considers the use of diseases as metaphors for economic concepts. Speaking of two of the most metaphorized diseases and most feared diseases, Sontag observes the frequency with which they are described in terms that reflect the deepest economic fears of their times. She notes that:
- … the fantasies about TB which arose in the last century (and lasted well into ours) echo the attitudes of early capitalist accumulation. One has a limited amount of energy, which must be properly spent…. Energy, like savings, can be depleted, can run out or be used up, through reckless expenditure. The body will start “consuming” itself, the patient will “waste away.”
- The language used to describe cancer evokes a different economic catastrophe; that of unregulated, abnormal, incoherent growth (62)
What does Sontag mean for Shiller? Or what would she have meant for Shiller had he troubled himself to engage with her work?
Most importantly, I think Sontag’s work provides a strong caution against thinking that diseases can be used as neutral explanatory metaphors. Sontag’s work would have made it impossible for Shiller to assert that “If we are careful and polite, it should be possible to speak in a non-partisan way about epidemics of economic narratives.” Any reader of Illness as Metaphor or AIDS and Its Metaphors has learned that the word “epidemics” is not value neutral. To assert that it is, is to deny what we know about how language and metaphor work. At a minimum, reading Sontag could have prevented Shiller from making the shockingly tone deaf observation that, “AIDS has been a slow epidemic; developing over decades, even slower than the bimetallism and Bitcoin epidemics” (24)
What Shiller does, in fact, is import our negative feelings about epidemic diseases into his discussions of economic ideas, while pretending that doing so is a neutral and scientific approach. We are given, for example, a series of graphs showing the frequency of use of various economic terms and told that the frequency of occurrences of each term over time produces a “hump shaped graph,” shaped just like the graph of an epidemic infection. Thus, these terms constitute epidemic economic narratives. Aside from the difficulty that not everything that produces a hump-shaped graph is an epidemic, Shiller’s use of graphs and of epidemiological terminology veils his tenuous argument with a layer of certainty and objectivity that Sontag reminds us is not really there.
What is worse is that Shiller does not just engage in this kind of shell game. He cautions us against people who engage in it. “We have to avoid the ‘seductive allure’ of superficial arguments about the economy using scientific analogies to lend a sense of precision to a theory that in fact may be of little substance. We need to keep the true scientific method in mind even when trying to use an essentially humanistic approach” (272). Rather than doing both the scientific method and the humanistic approach well, Shiller has done them both very badly.
This means that the many non-economist readers of Shiller’s volume, who will be attracted to it by the promise of combining narrative with economics, will find themselves enormously skeptical of his whole project, regardless of their assessment of his economic facts. They will be aware that Shiller’s use of epidemiological vocabulary is not neutral. When he compares Bitcoin and bimetallism to AIDS, for example, he is inviting his readers to make connections that reach well beyond the speed of the spread of AIDS and the speed of the spread of those economic ideas. The economic ideas are judged, morally, by their connection to an illness that is, as Sontag famously notes, not just a disease, but a disease “linked to an imputation of guilt.”
In his TED Talk, “Metaphorically Speaking” James Geary neatly outlines the problems that arise when writers use loaded metaphors and pretend that they are neutral terms. He discusses an experiment where participants were given a series of sentences, some of which were true and some of which were false.
- The participants had to identify, as quickly as possible, the literally false sentences. They took longer to reject metaphors as false than they did to reject literally false sentences. Why? Because we cannot ignore the metaphorical meaning of words either.
- One of the sentences was, “Some jobs are jails.” Now, unless you’re a prison guard, the sentence “Some jobs are jails” is literally false. Sadly, it’s metaphorically true. And the metaphorical truth interferes with our ability to identify it as literally false.
Shiller’s irresponsible use of epidemics as a metaphor for economic ideas makes it impossible to consider them on their own merits. Who can think neutrally of Bitcoin or bimetallism once they have been compared to AIDS? Who can pretend that Shiller is presenting them, or any of the economic ideas he discusses here, neutrally?
Shiller’s book could have been a defense of the kind of economic history that has long been out of fashion. I think that might be what he intended it to be. One of the places where a real intellectual consilience is possible is in the overlaps among fields like economics, literature, and history as scholars work to tease out answers to questions about what people thought, when they thought it, and why. We need those answers. We need that defense. But it cannot be done by scholars who don’t take the time to understand the topics they are borrowing from other disciplines and to engage with the discussions those disciplines have already had. And it certainly cannot be done by scholars who respect neither the narratives that people use to understand the economic world nor the people who create them.