This and last fall, I was a teaching assistant for a class taught by Andrew Richardson called Writing Scientific Papers, in which we workshopped papers that graduate students were writing. The textbook that Andrew has used for this class since its inception in 2012 is Josh Schimel’s Writing Science: How to Write Papers That Get Cited and Proposals That Get Funded. But before this year’s class, we both read and really enjoyed Stephen Heard’s recently published The Scientist’s Guide to Writing: How to Write More Easily and Effectively Throughout your Scientific Career.

In a nutshell, the main difference between these two books is captured by their subtitles. While Schimel’s book is shorter, to the point, and very goal-oriented, Heard’s is longer, a bit more meandering, and more process-oriented. I think both books are worth reading if you work in science and write papers or grant proposals, and either could be a good textbook for the sort of class we were teaching. Below, I’ve listed some more comparisons between the two, with the caveat that I last read them back in September, and some points may be retrieved from a fuzzy memory:

• Neither book is prescriptive, and both recommend following principles over rules. This makes sense for a writing guide, in my opinion, as there are a million ways to write something well.
• Both books focus, at their core, on story telling. However, this emphasis is stronger and more present throughout Schimel than Heard. Schimel advocates for focussing on the story at every level of writing, from sentence to paragraph to whole paper. I personally find this a bit too much (I’m not sure I believe that every sentence has to tell a full story). However, it lends a coherence to the whole book that translates well into the classroom–whenever we got stuck in critiquing some writing in class, we could move forward by asking “what’s the story here?”
• The discussion of narrative arcs and funnel- or hourglass-shaped stories (start and end broad, narrow in the middle) is very similar in both books.
• While both books mix examples and principles, I learnt more from Schimel’s examples and Heard’s principles than the converse. I think this is an important way in which the two books complement one another.
• In particular, Heard carefully thinks through all possible ways of approaching particular writing challenges. For example, he has a section titled “Three properties of good paragraphs,” in which one subsection titled “Making paragraphs coherent” has a list of seven possible ways in which to organize a paragraph. The whole book is peppered with such examples of comprehensive thought, which I really enjoyed.
• Heard includes multiple chapters on the writing process, as well as the scientific authoring process. Schimel’s book doesn’t have equivalent chapters. We assigned chapters from Heard as supplemental reading, e.g. the chapters on outlines, tables and figures, and self-revision. With more planning, we would no doubt have added more, e.g. the chapters on coauthorship and review, both from colleagues and peer review.
• I think this approach of mixing the two books is ideal for the classroom. A course taught entirely off of Heard’s book may become frustrating if students don’t care about the process of writing at all. With just Schimel’s advice, I think you could hate writing but still get the job done. Heard, however, is probably a bit more likely to get you to enjoy writing.
• Whose advice you like better is going to be a matter of personal preference. I happen to like Heard’s choices and examples a  bit more than Schimel’s, because my style happens to match Heard’s a bit more closely. But again, there are a million ways to write something well, and both of these books are impeccable guides to writing. This is one of those choices where you can’t really go wrong.

If you’ve written a scientific paper recently, I’m willing to bet that, during the writing/editing process, someone told you to avoid writing in the passive voice (unless you’d already heard this advice before, and avoided it from the start!). This suggestion is usually mentioned in reference to the Methods section, which seems to be where the passive voice is most often employed.

I’ll admit that my first instinct is to write Methods sections wholly in the passive voice. I’m not sure why this is—most likely, it’s just a habit I’ve picked up from reading scientific papers, or perhaps I was taught to write this way in high school science lab reports. And I suspect that I get annoyed when commanded to write in the active voice partly because I’m stubborn. But for the longest time, I’ve found something about the advice to abandon the passive voice fishy. I recently realised that my doubts lie in the most popular reason I’ve heard for ditching the passive voice—that it’s incompatible with story-telling.

A recent piece from Slate summarizes this reason thus

“I hold that good writing is basically good storytelling. To tell a story well, we need to clearly identify our characters and then show the reader what those characters do. The passive voice makes storytelling more difficult because it hides the characters deep in the sentence—if it shows them at all.”

When I first read that the passive voice may hinder storytelling, I grew quite concerned with my unwillingness to abandon it, because I firmly believe in the power of stories. But my concern turned to skepticism when I realised that storytelling in a scientific paper is so different from other types of storytelling that this complaint becomes nearly meaningless. And I think this boils down to an unusual divide, in scientific writing, between the characters we care about (the organisms, the molecules, the ideas…) and the characters that act (the scientists).

The Methods section of a paper has a very definite purpose—to convey the series of actions taken to collect and analyse the data in the paper (or the equivalent series of actions in a theoretical/modelling paper). We don’t spend any time describing the characters taking these actions (the scientists), we don’t even stop to distinguish members of this cast of characters from one another. We don’t care about the characters’ emotional states, memories, or imaginations while they perform these actions, and we don’t care too much about the physical setting in which the actions take place. But these are exactly the sort of details that story-tellers use to write engaging prose in the active voice. To illustrate this point, here are some examples from Purple Hibiscus, a moving novel by Chimamanda Ngozi Adichie, which I happen to have just read twice through.

This novel is narrated in the first person by Kambili, a teenage girl with a rich and influential but oppressive and abusive father. A visit to her aunt’s home changes Kambili’s life, leading her to discover herself and happiness. One of the threads in this story is how Kambili learns to do household chores; at home, she lives according to a strict schedule, imposed by her father, that includes nothing but studies and prayer. In her aunt’s house, however, she is expected to help, and her inexperience only exacerbates an already-tense relationship between Kambili and her cousin, Amaka.

Aunty Ifeoma got one of the huge yams we had brought from home. Amaka spread newspaper sheets on the floor to slice the tuber; it was easier than picking it up and placing it on the counter. When Amaka put the yam slices in a plastic bowl, I offered to help peel them, and she silently handed me a knife.

“You are wasting yam, Kambili,” Amaka snapped. “Ah! Ah! Is that how you peel yam in your house?”

Later on, Kambili returns to her aunt’s home, to recover from a terrifying interaction with her father. Kambili has changed in many ways since her last visit.

I went to the verandah, still coughing. It was clear that I was unused to bleaching palm oil, that I was used to vegetable oil, which did not need bleaching. But there had been no resentment in Amaka’s eyes, no sneer, no turndown of her lips. I was grateful when she called me back later to ask that I help her cut the ugu for the soup. I did not just cut the ugu, I made the garri also. Without her eyes still bearing down on me, I did not pour in too much hot water, and the garri turned out firm and smooth.

Both these paragraphs are primarily composed of a series of actions, not too different from a Methods section. But the actions do not themselves make these paragraphs interesting—they’re also about the relationships between the characters taking the actions, and how the characters and their relationships evolve between one paragraph and the next. All of these details help make these paragraphs part of a compelling story. These details also contribute to making these paragraphs good writing. Having the option of incorporating these details into the series of actions allows the author her to vary the structure of her sentences. Variation in sentence structure—in length and rhythm and tone—is essential to lively writing.

Because most of these situation- and character-related details have no place in the Methods section of a scientific paper, the ways in which we can vary sentence structure are severely curtailed. Moreover, the characters central to our scientific stories are, in the Methods section, largely being acted upon by the minor characters (us, the scientists). From the perspective of both good writing and good story-telling, therefore, we should be allowed to use the passive voice in scientific writing. It offers us a tool to vary sentence structure, and allows us to focus attention on the important characters. Of course, a Methods section written entirely in the passive voice will likely be boring, because it will have limited variation in sentence structure. But the same can be said of a Methods section written entirely in the active voice. The solution, clearly, is to abandon all proscriptions and strive for variation.

Also, none of this is to say that we should never care about the scientists who collect or analyse data. Our identities matter, and influence the stories we tell. But this means we should be telling more and varied stories about science, in a wide variety of venues and targeting a range of different audiences. Some of these stories need to be focussed on the science. Others should make visible and celebrate the people involved in science, in ways that are more substantial than writing our Methods sections in the active voice.