On the good and evil of scientific stories.

tl;dr: telling a good story is a vital tool in science communication, but it’s easy to go too far for a simple narrative.

If you’ve read this blog, attended a talk that I’ve given, or sat in on one of our lab meetings, you would know that one of my pet issues in science is communication. Scicomm, as it often goes by now, means more than explaining science to the public, though that is of course a large part of it. It’s also about how we communicate our science to other scientists, either in our field or ourside of it. Journal publications, conference talks, seminars, monographs, all of these things – and more – fall under science communication to me. And if you had found yourself as a fly on the wall when I was editing one of the Ph.D. students’ papers or critiquing conference slides, you would almost certainly hear me talk about story.

More precisely, you’d probably hear me say something like “what’s the story?” when I got through a rough draft of a manuscript, or after I watched a practice talk for an upcoming conference. When I say “story”, what I mean is the narrative and plot that ties together the work that you’ve done into a cohesive whole that the audience can follow and emphathise with. In the first chapter of his book Storycraft, Jack Hart cites this definition of story from Jon Franklin:

A story consists of a sequence of actions that occur when a sympathetic character encounters a complicating situation that he confronts and solves.

Story, as Hart says, consists of a recounting of a chronology of events (narrative), and the selection of arrangement of material so that a larger meaning can emerge (plot). Hart says:

For Eudora Welty “Plot is the ‘Why?’” Or, as the novelist E. M. Forster famously put it, the narrative is that “the king died and then the queen died.” The plot is that “the king died and the queen died of grief.”

I raise these issues because this is a problem that I’ve thought about at length when it comes to scientific communication. You might object that communicating science isn’t about a story, a narrative, or a plot, but I would strongly disagree. When you give a talk at a conference, you do exactly as Hart recounts: you construct a narrative and select material to form a plot (‘we identified some limit to our knowledge, we formulated some hypotheses, we did a test, we got some results, OMG science”), even if this looks nothing like what actually happened. You might be more familiar with this process in its rage form. Don’t fool yourself, this is story crafting. In its simplest form the scientist is the protagonist, the complicating situation is the unknown s/he is trying to banish as described in the introduction / methods, and the climax is wrapped up neatly in the results before the gentle falling action and dénouement in the discussion.

Story in formal scientific writing is often limited to the imposition of this narrative and plot structure, though stating it this way belies its importance; if you’ve ever reached the end of a journal paper and thought ‘what the hell was that paper about?’ (and we all have), chances are reasonably good that you’ve just experienced a failure of story. But when science is communicated to a wider audience, story begins to feature even more strongly. Whether written by scientists, science communicators, or journalists, it is easier to see this in action when the masters of the craft are in action. David Quammen, in his book Spillover structures his description of the hunt for Ebola and its reservoir around the story of the medical researchers who have tracked it through the jungles of Africa, winding in and out of their struggle to identify the source of the disease and the effects that it has on the people of Africa and elsewhere. It’s a detective story, which Quammen uses as a hook to lubricate the discussion of everything from molecular biology to mathematical epidemiology. But it’s the story that drives us through what would have otherwise been a textbook on epidemiology.

If I haven’t made it clear by now, I think that story’s important. Yet I also think that story has a dark side, one that we must be ever vigilant about as scientists, and it’s this: the push for a good story can obscure the truth. Science is messy, and full of complications and stumbles. There’s not always an answer, or a happy ending, and sometimes what we thought was right for a long time turned out to be incomplete, or even wrong. This fact is what makes writers like Quammen and science communicators like Carl Zimmer so valuable; they capture that messiness without letting it overwhelm the story, and in so doing make our science interesting to people. But if the push for a story goes too far, it can result in over-simplification and even simple and dangerous untruth.

I was reminded of this when I came across a post by one of my favourite writers on visual design, Garr Reynolds; Garr wrote the book Presentation Zen, and a series of other books like it, and I still recommend them to other scientists as a good way to get a handle on how to make your presentations suck less, visually. Recently, however, Garr wrote a post praising a video containing the work and narration of Paul Zak. The post, entitled “Neurochemistry, empathy & the power of story”, is itself curiously meta, as it disucsses work by Zak on neurochemical responses to the ‘dramatic arc’; in short, Zak claims that oxytocin and cortisol are part of the neurochemical suite that responds directly to the structure of a story, and can even be used in a predictive fashion (here, to predict the amount of donations that will be given when viewing a tearjearker story of father dealing with a young child dying of cancer versus the same father walking in the park with his son).

The irony of this, of course, is that Zak himself is an adept storyteller who has constructed a narrative around oxytocin as the ‘moral molecule’, reducing good and evil to the action of a single neurotransmitter. Here’s an excerpt from a Guardian article1 on Zak from last July:

What drives Zak’s hunger for human blood is his interest in the hormone oxytocin, about which he has become one of the world’s most prominent experts. Long known as a female reproductive hormone – it plays a central role in childbirth and breastfeeding – oxytocin emerges from Zak’s research as something much more all-embracing: the “moral molecule” behind all human virtue, trust, affection and love, “a social glue”, as he puts it, “that keeps society together”. The subtitle of his book, “the new science of what makes us good or evil”, gives a sense of the scale of his ambition, which involves nothing less than explaining whole swaths of philosophical and religious questions by reference to a single chemical in the bloodstream.

Here, we see the danger of story. In constructing a simple story with a compelling and digestible arc, Zak has swept the truth of this research under the rug, and the truth is that research on oxytocin is messy, contradictory, and provides few clear answers. As Ed Yong describes it, oxytocin can have distinctly contrasting effects depending on who receives it; some people may exhibit more social behaviour, while others in the same situation may exhibit more antisocial behaviour under the same dose of oxytocin. It can promote trust, or increase xenophobia. It may be that oxytocin is part of some motivator system: for example, people like James Goodson have worked to show that in birds like the zebra finch it2 is implicated in the ‘social behaviour network’ and may be instrumental in zebra finch flocking, though as in many other animals, this effect can be strongly sex-specific (usually to females).

All of this complication and mess is ignored in Zak’s story, which does a disservice to the reader who comes away with a simple view of the world that just doesn’t hold water. A friend of mine, a lawyer, asked me awhile ago if what he’d heard about this ‘cuddle chemical’ was true, and was visibly disappointed to learn that it was much more complicated than that. The problem here is that we are disposed to like a good, simple story; it has more emotional impact, which in turn makes it easier to remember and explain to others. Certainly, nobody wants to spend as much time reading journal articles and learning about nonapeptide hormones like oxytocin as I did for my PhD exam in order to tell a story at a party. This is why we have people like Ed, and Carl Zimmer, and Maryn McKenna, and all of the other great science communicators, writers, and science / scientist bloggers: they do the hard work of curating the facts and telling the story without losing the truth. Contrast Zak’s writing with Ed’s takedown of the oxytocin mess. It’s just as good a story, but it treats the truth with respect, and the truth is that we’re just not there yet. We have tantalizing ideas and scraps of evidence on how oxytocin affects us, but we can’t draw definitive conclusions. As Ed discusses, the hype around oxytocin has even led to people using it in an attempt to treat autism, with unknown and possibly harmful effects.

This isn’t an isolated problem. The TED talks have become a serious problem in this regard, and though I’ve seen some great TED talks over the years, they’ve grown to the point where the push for good stories has overwhelmed the ability of science to provide them. I saw the most recent example on Boing Boing when Maggie Koerth-Baker pointed to a problem in the widely-circulating story spun by 19-year old Boyan Slat on a plan to remove plastic from the oceans, namely, that it won’t work. Here again, we see the elements of story at work, this time surrounding Slat himself. A 19-year old phenom who rises to glory on the back of an award-winning school research paper, a hands-on problem-solver producing solutions and starting a foundation to implement them. It’s a feel-good story with a likeable protagonist who is tackling a problem that scares us all; it’s a shame that the scheme probably won’t work, and may even do more harm than good if ever implemented. The issue at hand, though, is that the story told by and about Slat is compelling but oversimplistic and potentially dangerous, just as the one told by Zak is3. As Maggie points out in her post:

Here’s a mantra to remember: TED Talks — interesting if true.

And the same is true with anything you read in the popular press about science. It’s interesting, if it’s true.

Now, I began this post by pointing out that I’m a big proponent of story in science, and I stand by that statement. Story is an important, and I would argue, necessary tool when we come to communicate the results of out work, for the same reasons that it can go badly wrong. A carefully crafted story draws the audience through the science, ties it together in a way that they can understand and remember, and adds punch to the work so that the audience cares enough to pay attention. Yet this process, while vital, needs to be kept in check by the demands of the search for the truth and the admission of messy detail and incomplete knowledge. The tension between story, which yearns to be complete, and science, where more research is always needed, must be respected and maintained lest you end up with bone-dry science or a compelling – but misleading – tale.

  1. or as Ed Yong puts it, ‘ad’
  2. under the name of mesotocin
  3. as an aside, I’d like to say that despite the problems inherent in Slat’s plan and how it ended up going viral, I hope that he keeps trying. He sounds like a smart guy, and failure is a great first step on the road to success.