The Importance of Storytelling in Science

“We owe it to each other to tell stories.”
– Neil Gaiman

Scientific writing is often belied as dry, stale; a block of indomitable, indecipherable text buttressed with vague language, passive voice, countless equations, and overly constructed, borderline-run-on sentences that seem to drag on and on and on. See what I did there? There are some papers and writers who amaze us with their style and their readability. Good science relies on good writing. If your prose resembles the start of this post, most people are likely to never read it and whatever amazing work you did in the lab or the field will be relegated to the shelf—sad and unread.

While we have all read our fair share of dense and unapproachable manuscripts, how much does narrative style affect the impact of our science? In the recent PLOS One work, “Narrative Style Influences Citation Frequency in Climate Change Science,” authors Ann Hillier, Ryan P. Kelly, and Terrie Klinger used a crowdsourcing platform to examine 732 scientific abstracts from 19 journals for narrative style. Using citation frequency as a proxy for impact, the authors restricted their look to climate change research and began with the hypothesis that the presence of a more narrative focused text, would result in a higher citation frequency. And as it turns out, the whole world loves a story.

“Research is formalized curiosity. It is poking and prying with a purpose.”
– Zora Neale Hurston

The authors define six factors that influence narrativity:

Setting – A successful narrative relies on a consideration of time and place. The reader wants to know where and when something is happening. Abstracts were gleaned for a mention of either time or place.

Narrative Perspective – The role of the narrator distinguishes stories from other forms of communication. A first-person narrator is a typically stronger narrative presence than a third-person narrator. In-text reference to the narrator through pronouns such as “I”, “we”, and “our” was assessed to quantify narrative perspective.

Sensory Language – Language that appeals to the senses or emotions can be used to create a connection with the reader to the work. As defined by the authors, this includes narrative expressions of “emotions, attitudes, beliefs, and interpretations.”

Conjunctions – The logical ordering of a narrative through the use of conjunctions to connect words and phrases results in prose with momentum towards a conclusion or completion. The authors used the presence of conjunctions to determine the “extent to which an abstract is logically ordered, based on the observation that a temporal or causal ordering of events is an essential, and distinguishing, characteristic of narratives.”

Connectivity – The use of words or phrases that create contextually explicit links within a narrative creates a sense of “connectivity” either through repetition or references to previous statements. The authors also note that “logical linkages” were also explicitly considered.

Appeal – There is the question, of course, of why do we care? What is the point of this work? A narrative must include some form of commentary, evaluation or “landscape of consciousness.” Does the work make an appeal or clear recommendation?

In addition to these narrative elements, the length of the abstract, number of authors, year of publication, journal identity, and journal impact factor were also considered as these are known factors that influence citation frequency.

“People think that stories are shaped by people. In fact, it’s the other way around.”
― Terry Pratchett

One unique method applied in this work, is the use of a crowdsourcing platform called CrowdFlower. The platform works by having multiple contributors who work for the site and who are paid small amounts of money for completing tasks—in this case tasks examining scientific abstracts based on the six defined criteria. Each abstract considered was examined by multiple independent contributors from the site. The idea is to have higher-quality contributors who offer a bit more technical expertise than just your average person off the street. While this method is costly, it does possibly alleviate some of the quality assurance and control issues of other crowdsourcing methods.

Four of the six narrative criteria were found to have a strong positive correlation with citation frequency:  sensory language, conjunctions, connectivity, and appeal. A “narrative index” for each abstract was created using the abstract’s scores on each of the six criteria. A strong positive relationship was evident with a higher narrative index correlating to a higher citation frequency. Each abstract’s narrative index correlated well with the impact factor of the journal of publication.

“There have been great societies that did not use the wheel, but there have been no societies that did not tell stories.”
― Ursula K. Le Guin

At first glance, this does seem to be quite interesting. However, it should be expected to a certain degree. Journals earn higher impact factors based on citation frequency. If citation frequency has a positive relationship with narrative index, then narrative index should also have a positive relationship with journal impact factor. The more intriguing question is about the spread here. Do journals with higher citation frequencies and impact factors editorially select for more writing that exhibits more narrative properties? Well, probably. The outliers here are Nature, Science, and Ecology Letters. Both Nature and Science are aimed at more inter-disciplinary audiences. Articles in those journals tend to be much shorter and require prose that is more direct and broadly appealing than other more narrowly focused journals. This likely bakes in a level of readability or narrativity. However, for the life of me, I doubt I will ever completely understand one of those articles on protein folding.

However, I have been told many, many times in the course of my education that good scientists do good science, but great scientists are also great writers. Are those publishing in higher impact journals simply higher caliber writers to begin with?

The answer is likely more complex than that. Often, more senior scientists are the ones publishing work in more high impact journals. They have likely been around a lot longer. One would hope they figured out some tricks of the trade along the way. There is a selection bias here, of sorts. There are also a lot of things that go into how highly cited any given paper is. Citations are only one means to measure scientific impact, but citations are scientific currency. Science requires communication and since the advent of the scientific method, writing has been at the core of how we communicate science. That said, writing is certainly a factor.  And honestly, whether it improves citation frequency or not, improving and clarifying scientific writing is a win-win to me and something we should all strive for.

There is the consideration of how well the readability of an abstract is connected to the readability of the entire work. But, in the honest application of science, the abstract is the first thing most scientists and academics read. If it is a bad abstract, we’re probably not going to read the rest of the paper. And if the work is paywalled, you can only get access to the abstract in the first place.

There is a distinctly, biology-centric view here to the journals selected, but the findings and indications are likely broadly applicable:

“Peer-reviewed scientific discourse is often viewed as a special form of communication, exempt from the qualities of narratives that humans inherently relate to. However, our findings support an alternative interpretation: scientists can engage readers and increase uptake by incorporating narrative attributes into their writing styles.” — from Hilllier et al. 2016.

Writing, like anything, is not a skill you are born with. It is one honed through countless hours of hacking and revising. Just keep pressing forward. Many people will tell you the most difficult part of completing an advanced degree, in science or any field for that matter, is actually writing the thesis or dissertation at the end. Writing is a process. Anyone who has put off writing that term paper to the night before can tell you that.

“A word after a word after a word is power.”
– Margaret Atwood

A really great mentor of mine always gives the same advice to people who are giving their first big presentation. They are often nervous and anxious. He reminds them of how enthusiastic they are about their work, how much time and effort and love they have put into their project, how much they care about it. A presentation, he tells them, is just you sharing that enthusiasm with others. If you’re excited, they’ll be excited. Just go out there and have a conversation. In a way, that applies to writing for me. You’ve done all this work, it’s time to tell people about it. That said, I have some writing to get done.