2.1.1. The Katz-Krzywinski-Cairo Debate: Storytelling has (no | a major) role in science
STEM scholars have a contentious relationship with storytelling (Bickmore & Grandy, 2014; Coen, 2019; Luna, 2015; Medawar, 1982) – tensions drawn between competing needs for epistemic precision and communicative resonance (Dahlstrom & Scheufele, 2018; Joubert et al., 2019; Wilson, 2017). Stories can either make science more compelling by “portraying science as a human-driven endeavor” or “can be used to misrepresent science” (Ettinger et al., 2021, p. 352; see also Blastland et al., 2020).
This tension is exemplified through a debate between Katz (2013), a tenured biologist at MIT, and Krzywinski and Cairo (2013a, 2013b), a staff scientist and Professor of Professional Practice at University of Miami, in a series of op-eds in Nature. Krzywinski & Cairo (2013a) equate effective presentations of data with effective storytelling. “Science is ‘full of vexing questions, conflict, dead ends, insights and the occasional thrilling leap’ and, as such, is ‘a story well told” (Krzywinski & Cairo, 2013a, p. 687). Olson (2015), a tenured marine biologist turned filmmaker complements this perspective by comparing the IMRAD[1] structure of scientific writing to a three-act beginning-middle-end structure of storytelling (see also ElShafie, 2018; Ni, 2024). Luna (2015, p. 391), on the other hand, provides a more tempered advocacy of storytelling in science. He argues that narrative tools contribute to more memorable scientific arguments but also requires “the same caution and precision as one would approach a scientific experiment.” However, Krzywinski and Cairo’s advocacy for storytelling rests not with the written word, but with story-friendly visualizations of data (Ma et al., 2012; Wong, 2011a, 2011b). Storytelling and communication within STEM are therefore driven by visuals, rather than the canonical text-driven model of storytelling discussed in humanities disciplines like applied linguistics[2] (see Section 3.)
“We have previously (Krzywinski, 2013]) encouraged the use of practical graphic design principles to inform the content and layout of figure panels. Now we propose that you apply the structural principles of storytelling to integrate multiple panels into a cohesive whole…Instead of ‘explain, not merely show,’ seek to ‘narrate, not merely explain’ (ibid.).
Katz’ critical response underscores latent tensions between the production and communication of science. “Great storytellers embellish and conceal information to evoke a response in their audience. A storyteller would plot the data in the way most persuasive rather than most informative or representative” (Katz, 2013, p. 1045; cf. ElShafie, 2018; Nat. Biomed Eng, 2018). Katz thereby problematizes storytelling on two grounds: (1) stories conceal the truth by aligning fitting facts to a narrative structure and therefore (2) omit counterevidence and nuance (Coen, 2019; cf. Knight, 2003). A great storyteller often begins with the end in mind and therefore constructs a narrative through motivated reasoning. Instead, “it would be better for both authors and readers if these [alternative interpretations] could be discussed openly rather than obfuscated” (Katz, 2013, p. 1045; though see Wulster-Radcliff et al., 2015 on the growth of public, open access peer review in STEM fields).
But Katz does acknowledge the value of storytelling in the process of discovery, endorsing Baker’s (2012) use of storytelling to discuss the serendipitous discovery of new functions for RNA granules. Bickmore and Grandy (2014) also write about how many scientists report thinking through scientific problems with non-scientific, borderline fabulist processes. Einstein often mused through scientific problems with the help of a God-like being called “the Old Man” while Belgian scientist Friedrich Kekulé “discovered” the ring-like structure of benzine after waking up from a dream where a snake tried to eat its own tail (Bickmore & Grandy, 2014). Such ruptures in logical processes of discovery underlie Kuhn’s (1970) historiography of science as one driven by revolutions and paradigm shifts rather than linear progressions, as well as Feyerabend’s (1974) critique of linear ideologies of scientific methods and production. Bickmore and Grandy go on to argue that science is storytelling through specific rules – rules rooted in (1) reproducible procedures, (2) logical explanation, (3) iteration, (4) natural phenomena, (5) uniformitarianism, (6) parsimony, and (7) respectful yet confrontational of prevailing explanations. Or as Bickmore and Grandy (2014, p. 52) explain, “Discovery is not a work of logical thought, even if the final product is bound in logical form…In the creative process, anything goes, so long as [the final product is] naturalistic and logical.”
Yet Katz (2013, p. 1045) concludes his column by saying, “these narratives are often deemed by scientists as irrelevant ‘fluff.’” Krzywinski and Cairo (2013b, p. 1046) concur with Katz in a response paper, writing that their previous column (2013a, p. 687) “was limited to the construction of multipanel figures, which summarize as much as they support detailed exposition of the text…[It] did not address how this text should be composed or the broad subject of motivation and design of scientific experiments.” Yet they also reiterate the centrality of visual storytelling in the communication of science (Wong 2011a, 2011b). “Readers often look to the abstract and then the figures to provide them with an initial impression and overview of the findings[3]” (ibid). In sum, the STEM scholarship sampled to this point seems to advocate for eye-catching story-driven visuals with rigorous, non-narrative descriptive text, and firm adherence to scientific methods of inquiry – the goal being to tell a complete story rather than a good story (National Science Foundation, 2015; Nat Biomed Eng, 2018).
[1] IMRAD stands for Introduction-Methods-Results-Analysis-Discussion. Olson (2015, p. 6) argues that this form of organization reflects a 100-year-old method of scientific storytelling as it is “essentially identical to the three-act the three-act structure that is at the heart of virtually every movie or play written today.”
[2] Though in the quote below, we also see how the “structural principles of storytelling” suggests Krzywinski and Cairo place storytelling as a textual medium and borrow it into a visual scientific mode.
[3] See also Montgomery [1996] for a discussion on the importance of multilingual abstracts and figures in the development of modern Japanese science
Ontogenetic Reflection
I can feel myself entering a hall of mirrors as I listen to this section read back to me. On the one hand, there’s a lot of references to STEM practitioners and their views on storytelling. This lends itself at least a smattering of emic credibility. And yet of course I as the comprehensive examinee, also hurriedly curated these quotes and assembled them in a particular order as I related to my own emerging story of how scientists seem to approach story. It reminds me of Karen Barad’s notion of diffraction, or the ways that measuring apparatuses can contribute to a scattering of particles — in this case, one could say an intra-active scattering of ideas or discourses. One can never measure the world without measuring with the world. This diffraction also occurs with you, the reader. As you make your way through this argument and ask yourself: Is what Ian is saying plausible? If you’re a STEM scholar yourself, you might have your own counterstories — the ways that my argument contravenes your own experience.
This is a fair point. Yet organization scholar Yiannis Gabriel also discusses the productive tensions that seem to exist between empirical goals to represent some kind of shared understanding of truth and the more poststructuralist or feminist methodological campaigns to elevate lived experience as valid sources of research data. As I continue to listen through this comprehensive exam and produce this ontogenetic log, I am in a way moving between these two positions. The ontogenetic log becomes a diffractive reading of my own comprehensive exam — my comprehensive exam being a diffractive reading of the literature of STEM scholars who have talked about notions of story and narrative. Two layers of experience coalescing (if not coagulating) into a petrifying narrative.
What really draws me to the notion of diffraction is I think it allows for an entanglement of researchers as loci of enunciation while also sidestepping the (over)indexing and (over)coding of social construction and language as representations of reality. And I wonder how this belief will hold up as I continue to diffractively listen to further sections of this lengthy paper.
But until then, I look forward to hearing your points of agreement or disagreement, or your own diffractive reading of my nested diffractive readings.
Until next time, cheers!
Leave a Reply