Do scientific writers have “voice”?

Image: The Writer in the Window, by Mark Heybo via, CC BY 2.0

Scientific writing has a reputation for being dull (among other things).  By and large, it deserves that reputation (although there are admittedly exceptions).  I’ve become very interested in why our writing is dull.  I don’t have a definitive answer, but in today’s post I’ll begin to explore one possibility: that pressure for conformity has prevented most scientific writing from having voice.

What is “voice” in writing?  The concept is slippery, without a clean consensus definition, but what I mean by voice is the recognizable traits of text that makes a particular author’s text unique and that suggest the author’s attitude or personality.  Voice may arise from an author’s choice of sentence structures, patterns of punctuation, word choice, rhythm – all kinds of things that you may not notice consciously as you read, but that unite different passages by the same author while distinguishing those passages from the work of other authors.  It’s because authors write with voice that you can read a passage and recognize the author, even if you don’t recognize which book it comes from.  Consider these four snippets, from novels by four different American authors.  Can you guess who wrote each one?  (Tell us – be honest – how you did, in the Replies).

     “Come on down-stairs and have a drink.”
     “Aren’t you working?”
     “No,” I said. We went down the stairs to the café on the ground floor.  I had discovered that was the best way to get rid of friends.  Once you had a drink all you had to say was: “Well, I’ve got to get back and get off some cables”, and it was done.  It is very important to discover graceful exists like that in the newspaper business, where it is such an important part of the ethics that you should never seem to be working.  Anyway, we went down-stairs to the bar and had a whiskey and soda.  Cohn looked at the bottles in bins around the wall.  “This is a good place,” he said. (—Who wrote this?)

      Lou Ann walked past both of these establishments nearly every day.  Something about the Jesus Is Lord place reminded her of Kentucky, and she always meant to ask (if she only had the nerve) if the people there came from her part of the country.  Fanny Heaven she just tried to ignore. There was something innocent and primitive about the painting on the door, as though the leopard-bikini lady might have been painted by a schoolchild, except that she was positioned in such a way that the door handle, when a person pushed it, would sink into her crotch.  This door always gave Lou Ann the shivers, though she tried not to give it a second thought.  (—Who wrote this?)

     The gun made a ripping sound like the opening of the zipper on the fly of God Almighty.  The gun lapped up snow and vegetation with a blowtorch thirty feet long.  The flame left a black arrow on the ground, showing the Germans exactly where the gun was hidden.  The shot was a miss.
     What had been missed was a Tiger tank.  It swiveled its 88-millimeter snout around sniffingly, saw the arrow on the ground.  It fired.  It killed everybody on the gun crew but Weary.  So it goes.  (—Who wrote this?)

     “We got a lot to do and a lot to see,” said Towser.  “We got a lot to learn.  We’ll find things–-“
     Yes, they could find things.  Civilizations, perhaps.  Civilizations that would make the civilization of Man seem puny by comparison.  Beauty and, more important, an understanding of that beauty.  And a comradeship no one had ever known before – that no man, no dog had ever known before.
     And life.  The quickness of life after what seemed a drugged existence.
     “I can’t go back,” said Towser.
     “Nor I,”, said Fowler
     “They would turn be back into a dog,” said Towser.
     “And me,” said Fowler, “back into a man.”  (—Who wrote this?)

How’d you do?*

Now imagine that we did the same thing with scientific papers.  Do you think you could read a few sentences from one of my papers and identify me as the author?  Do you think I could work the trick on a few sentences from one of your papers?

I’m pretty confident that the answer is “no”.  Even if you had read every single one of my scientific papers**, I don’t think you could identify me as they author of a passage from my next manuscript***.  In other words, I don’t think my scientific writing has a recognizable voice, and I’ll wager a fair bit of money that yours doesn’t either.

This raises three interesting questions:

  • Is this a good thing or a bad one?
  • Why is it so?
  • If we wanted to give our writing voice, how could we do so?

The second and third questions I’ll leave for future posts.  But for now: if I’m right that you can’t tell my scientific-writing style from anyone else’s, should we regret that, or should we heave a sigh of relief?

The case for a sigh of relief – Voice is one reason that readers of fiction have authors they love, and authors they despise.  (It’s not the only one, of course.)  Some people love James Joyce, but others can’t stand him – finding his writing pretentiously unintelligible (and that’s a charitable way of putting it).  Some people love Leo Tolstoy, but others find him interminably long-winded.  Some people love Cormac McCarthy, but others find him – well, to be honest, I couldn’t even get far enough into Blood Meridian to decide why I hated it.  If my scientific writing had a strong voice, there might be readers who loved my papers and readers who hated them****.  I’m not sure it’s wise to turn off a set of readers stylistically – not when every paper I write inevitably competes for readers’ attention with the ocean of other papers they could read instead.  I can see a second problem with a voice-rich literature, too.  There’s a mental cost associated with switching, as a reader, from one voice or style to another.  Even if stronger and more individual voices didn’t make individual papers harder to read, they might make a literature harder to read.  I might find it interesting and fun to switch from a Raymond Chandlerish meta-analysis of host choice in plant-insect interactions to a Gabriel Garcia Marquezian report of a novel host shift in passionflower beetles – but I wouldn’t find it easy.  To a considerable extent, our literature has converged on formulaic structure (IMRaD, and much more) and homogeneous style because that eases the task readers face.  A single, consistent way of encoding and decoding information makes for efficient information retrieval.

The case for regretHomogeneity may be efficient, but it’s dull.  In converging on a style that’s consistent across our literature, we’ve converged on a style that’s plain and – more than that – grey. It’s important that our literature be clear, of course (that’s a major theme of The Scientist’s Guide to Writing); but I don’t think that means that our writing must be completely unornamented.  It’s possible for scientific writing to be clear, but also stylish (as Helen Sword argues in this very nice little book).  It’s possible for scientific writing to be clear, but also enjoyable.  Writing that has voice (well, as long as it’s not Joyce’s) is more enjoyable, more engaging, than writing that doesn’t.  Voice shows something of the personality of the writer, and we like to see the personality behind the science.  (Ignore those who argue that showing personality in writing robs our work of objectivity.  Objectivity, if it exists at all, is a feature of our methods, not of the rhetorical tricks we use to pretend to be objective.)  If I wrote papers with a distinctive voice, I might have readers who enjoy them; readers who tell their friends they’ve enjoyed them; and readers who return to my work because they get some pleasure along with their science.  I’d love to have such readers.  And if I lose a few readers who don’t like my particular voice, perhaps I’ll recruit enough new readers to outweigh the losses (in the Trolley Problem of Writing).

So: an argument against voice, and an argument for it; and of course the obvious possibility that the optimum is some voice, but not too much.   What do you think?  Would our literature be improved or ruined, if we each sounded a little bit like ourselves?

© Stephen Heard  April 9, 2019

Expect a few more posts exploring this issue of voice in scientific writing. Meg Duffy, Helen Fricker, and someone in the audience at my most recent talk (at Université Laval) asked me similar and very interesting questions about writing, collectively spurring me to begin writing up my not-very-coherent thoughts – thanks to each of them!

*^I kept these fairly short so they’d be a little bit challenging, although none is from an obscure work by its author.  Some of the four authors are more widely read than the others – and of course it’s completely expected that you wouldn’t recognize the voice of author X if you’ve never read their work!  If you don’t know the work of authors 2, 3, and 4, please check them out – each is wonderful.  If you don’t know the work of author 1, count yourself lucky.

**^I do not recommend doing this.

***^Other than possibly by topic, I mean.  It’s true that if you read a new paper on host- and habitat-associated genetic structure in leafmining flies on goldenrods, you could reasonably guess that Julia Mlynarek and I are (possibly among) the authors.  But you’d be doing so based on the flies and the goldenrods, not based on our inimitable style.

****^I heard that. Don’t make me come back there.



22 thoughts on “Do scientific writers have “voice”?

  1. sleather2012

    I’ve read all those but only recognized Hemingway – my excuse being that it is a hell of a long time since I read Vonnegut and Simak (although I really liked Simak back then). My apologies to Kingsolver (if she reads your blog) as she is a writer I really admire and should have recognised.

    Liked by 1 person

  2. Peter Apps

    The more authors there are the less voice there will be – even if one of them writes it the others will edit out anything distinctive.

    Liked by 1 person

  3. Markus Eichhorn

    Interesting thoughts. A number of years ago we had a discussion about academic writing in our research group. One of my PhD students moaned that all the prescriptions in science writing removed the opportunity to express a unique voice. My response was pretty much your first argument. Which of us was right? Despite being a naturally gifted and fluent writer, this student later laboured more than most in completing a PhD thesis and never published any journal articles. I don’t think science writing was the defining factor that led them to leave research; they also haven’t published any other forms of writing that I’m aware of. Nevertheless I do wonder whether some people decide that the things they want to say are best expressed outside the strictures of our stultifying conventions, or else struggle to conform to what they see as an unnecessary constraint.

    Liked by 1 person

  4. Alexandre V. Palaoro

    I think you could also add into that mix “why are you reading the paper?”. Whenever I am reading because I need something fast (e.g., an example for an argument), I skim through the paper and spend most of my time on the figures. And there are also times when I read the whole paper. Maybe voice matters only in the second scenario? But here is a plot twist: there are a few papers that I am unable to skim through because they are beautifully written. The prime example would be Hurlbert’s paper on pseudoreplication. There is just enough voice to make the paper such an enjoyable read that I can’t skim through it – I have to read a bit more.
    (I think) The point I am trying to make here is: maybe reading habits (e.g., skimming through papers) also select for voiceless texts. If I only skim papers to get a jist of it, writing voice may seem unimportant. But having a writing voice might make people break that habit and read papers more fully. Does that make sense?

    (A second point here is: make good figures, people a lot of attention on them).

    Liked by 1 person

    1. ScientistSeesSquirrel Post author

      Hurlbert’s paper is a great example, one of the exceptions I agree exists. That paper is long, technical, and could have been very, very dull. But it’s been cited >4500 times, and its voice (and in particular the phrase “demonic intrusion”) has to be one reason why.


  5. Gretchen Rasch

    As an editor at a scientific institute, I find I can recognise the voice of many of our authors in their reports for clients but less so in their publications. It has to do with sentence structure, choice of vocabulary and their ability to tell a scientific story. It can be important in unexpected ways – we had a High Court judge compliment one of our authors on his clarity, both written and oral.
    (I recognised Hemingway and Vonnegut (“so it goes”) but not the others.)


  6. Emilie Champagne

    I feel that our writing receive so much comments and revisions that it lost all its voice. There is one ‘scientific’ voice I might recognize: Samuel J. McNaughton. There’s multiple reasons for that: 1-He wrote many single-authored articles, 2-I read a lot of them, 3-There’s a lot of attitude (bordering on cockiness) in his writing. His ‘I’ is strongly assumed and present. At least it’s the way I felt when I first start reading his articles and it did made my reading more enjoyable.
    Also, I failed at your game, mostly cause it’s authors I haven’t read (yet), although I did read Hemingway in French. I thought there was a translation problem with Hemingway’s style… I’m glad to learn it’s also insufferable in English.

    Liked by 1 person

  7. kokedit

    I’m a medical editor who most often edits manuscripts written by researcher-authors who aren’t native speakers of English. Many of my authors do have a noticeable voice, and I work to preserve it when helping them with their medical English.

    I read lots of research reports in all sorts of specialty journals, and I’ve found that when an article is clearly written in plain English without dumping in unnecessary jargon, it gets a wider readership, which in turn helps get the authors’ findings out there to more professionals who can make use of them.

    Please, please, please, for the sake of sharing knowledge, stop writing the way you think your supervisor wants you to write. Sure, follow your target journal’s required format and topic organization. But explain things so readers from all medical disciplines can understand. Tell the story of your findings. Don’t show off by writing with a medical dictionary at your side.

    Liked by 1 person

  8. Jeremy Fox

    Off topic:

    Re: novelists’ voices, this gets into the fun topic of writers aping one another’s voices. You’d probably enjoy Thinks… by David Lodge.* Light academic comedy about a poet/creative writing teacher and a cognitive neuroscientist. I mention it here because the writing teacher gives her students exercises in which they have to ape the styles of famous writers. Which I assume was just Lodge giving himself an excuse to mimic Kingsley Amis and others. It seems like he had a lot of fun with it.

    We could also talk about bands mimicking each other, and not just when covering one another’s songs. They Might Be Giants’ “venue song” about Richards on Richards in Vancouver sounds more like The Cars than The Cars, except that of course John Linnell sounds nothing like Rick Ocasek. I assume TMBG were trying to sound like the Cars on that one, but I don’t know. And the guitar riff in Bohemian Like You sounds just like the Rolling Stones. And I think we probably underrate the skill of Weird Al Yankovic and all the session musicians he’s worked with to make his parodies sound like the originals.

    Ooh, and then you get into directors! Think of Brian de Palma doing entire sequences of mock-Hitchcock…

    So, maybe the worst thing about scientific papers mostly not being written in distinctive voices is that scientists don’t ever get to have the fun of mimicking each others’ voices.

    *Warning: Thinks…treats the sexual behavior of one of its characters as harmless, although its arguably not (and similar behavior in the real world definitely wouldn’t be). Thinks…might be off-putting to some readers for that reason.


    1. ScientistSeesSquirrel Post author

      Huh, your “off topic” is something I actually had in but cut only because the post was getting too long. Yes, “voice” is why pastiche works! Now I wish I had kept that bit… yet more evidence that we often thinking along disturbingly similar lines!


      1. Jeremy Fox

        I have been reading Matt Levine’s finance newsletter for months now and I really enjoy his style. In part because it’s well-married to the structure of his pieces. They’re basically smart, thoughtful, humorous explainers. I’ve been toying with the idea of trying to ape him in my blog posts, as a way of freshening them up. But I need to find suitable subject matter.

        Basically, the ideal subject is one that appears complicated and technical, and so not widely understood, but can be boiled down to an interesting, easy-to-understand essence. But it also has to be a subject that involves some human comedy. Finance is perfect for this because it has lots of highly-technical aspects, but at bottom it’s people doing things people have long done–make trades with one another in an attempt to make themselves better off. Finance also is good for this because Matt is mostly commenting on news items about people and organizations that are (i) rich and powerful, (ii) credibly accused of wrongdoing, or (iii) both. Thereby making them socially-acceptable butts of jokes. I am finding it hard to think of ecology- or academia-related subjects that would fit the bill for a Matt Levine-style treatment.


  9. janig717

    I’m all for voice. Hard to do though unless you have a genuine interest in the work you are writing about. I think it’s an aspect that is missing in a lot of studies today, but the system emphasizes the paper as the goal rather than good old-fashioned curiosity.


  10. Brian McGill

    I guessed Hemingway & Kingsolver (& kicked myself for not guessing Vonnegut & never read Simak) so pretty much proving your point about voice in fiction.

    I’m not quite as convinced that voice is absent in science writing (especially single author pieces). Also its hard to separate voice from choice of subject. Part of the tips for me were drinking for Hemingway, a young observant girl for Kingsolver, and (in hindsight) absurd war for Vonnegut. If you allow that level of subject as part of voice, I bet many ecology authors have a recognizable voice (again more noticeable in single author papers and probably squeezed out in Science or Nature papers).


    1. ScientistSeesSquirrel Post author

      About separating subject: I think I’d argue that you could still recognize Hemingway if he wrote about observant young girls or Kingsolver if she wrote about drinking. (What a weird counterfactual.) I see what you mean about subject being arguably part of a writer’s distinctive feel – Hemingway wouldn’t be Hemingway without the childish preoccupation with drinking and testosterone. But to me that’s what he wrote *about*, not how he wrote it, so I’m not going to knuckle under; for me, voice remains separate from subject. But of course: I’m no expert, the last English Literature course I took was in 1987!


  11. Pingback: (Limited) use-cases for “utilize” | Scientist Sees Squirrel

  12. Pingback: Reclaiming voice in scientific writing | Scientist Sees Squirrel

  13. Pingback: Is my worst writing habit also my writer’s voice? | Scientist Sees Squirrel

  14. Pingback: Writing for your students: some suggestions from Overconfident Greg (guest post) | Dynamic Ecology

  15. Pingback: Three reasons for the red pen | Scientist Sees Squirrel

Comment on this post:

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.