Nuancing brevity: less isn’t always more

Image: something I wrote recently. If you’re a sharp-eyed reader of the blog and think you know what it is, make a guess in the Replies. Although the only prize is my admiration.

Nearly every source of writing advice agrees on one thing: brevity is good.  My own book, The Scientist’s Guide to Writing, is no exception; I devote an entire chapter to brevity.  There are good reasons for this. Longer papers ask more of their readers’ limited time budgets and seem, likely as a direct result, to have less citation impact.  Journals have limited space and would rather publish more papers than longer ones*.  In general, shorter texts and simpler sentences are easier to understand.  And most writers need to shorten their first drafts – and most find this a challenge (as Blaise Pascal noted in his famous letter, “I’ve made this letter longer than usual; I haven’t had time to make it shorter**).

But just in the last year or two, I’ve backed off my fanaticism about brevity just a bit.  (As an editor, I’ve even recommended – twice – that authors lengthen their papers a little.  I expect they fell off their chairs.)  What led me to change my tune, even if only by a fraction of a semitone, was thinking about Edward Tufte’s pithy summary of ideal figure design: that a good figure gives the reader “the most ideas in the shortest time with the least ink in the smallest space”. I realized that I’d been paying fetishizing “least ink” and “smallest space” while losing track a bit of “shortest time” – which may be the most important thing of all.  Clear, easy communication (which is what makes “shortest time” possible) should always be our goal, and that doesn’t always mean the shortest text.

It’s easy to become mesmerized by “least ink”.  Scientists love data, and “least ink” (and “smallest space”) are easy to quantify.  For text, word processing software provides word count data – Microsoft Word even puts it right in the corner of the window where you can watch the count creep up as you type.  Journal length limits are usually expressed as word counts, too.  Response to Reviews letters nearly all need to signal compliance with a request to cut length, and word count data make that easy.  But focusing on word count incentivizes the wrong behaviour.  I’ve caught myself, more than once, replacing two short words with one long one, and feeling proud of myself***.

The speed and ease of reader understanding are, in contrast, very hard to quantify, or even to know about.  That’s partly because “understanding” and “ease” are inherently squishy.  It’s also because when we’re writing, we don’t have direct access to what our readers are going to think.  As writers we need two tools.  First is a bag of tricks for what I call “reader simulation”: forgetting all our insider knowledge to see the text as a reader will.  Second is willing and able reviewers whose reactions can stand in for those of our eventual readers.  We need to detect those cases where a few extra words would help rather than hinder.  After all, it takes much longer for a puzzled reader to read a short sentence twice than for a comfortable one to read a long sentence once.

I don’t want to swing the pendulum back too far.  It’s still the case that I see far more manuscripts needing shortening than I see needing lengthening (and this is especially true for my own manuscripts).  But when I revise I now find myself looking, amongst the opportunities to chop, for the occasional opportunity to improve by the judicious addition of an extra word or two in the right place.  And when I chop, I now try to keep only one eye on the word count, with the other reserved for whether removing a word will make the reader’s job easier or harder.  Easier, of course, is what it’s all about – whatever that takes.

© Stephen Heard ( March 13, 2017

*^This constraint is relaxed in online-only journals, but it doesn’t disappear.  There are still costs of copy-editing and typesetting that scale with paper length, even if actual storage capacity is too cheap to matter.

**^Je n’ai fait celle-ci plus longue que parce que je n’ai pas eu le loisir de la faire plus courte. No, Mark Twain didn’t say this.  Pascal seems to have been the first; the quote is from a 1656 letter, one of 18 he wrote addressed to the priests of the Jesuit order but intended to be widely read.

***^Several of the granting agencies I deal with have gotten wise to this, and use character counts instead.  Any day now, I’m going to start seeing proposals from Twitter-trained writers packed with emoticons.  I can’t wait 😉


9 thoughts on “Nuancing brevity: less isn’t always more

  1. sleather2012

    Just finished a paper for Oxford Bibliography which has a strict word limit and as a consequence was using compound sentences; on re-reading it found that although this saved a word or two, it made it a lot clumsier so have revised and made it slightly longer but hopefully more readable. (52 words)

    I have just finished a paper for the Oxford Bibliography series. The series has a strict word limit, both lower and upper. To comply with this I was using compound sentences. When I read the completed draft it struck me that although saving on words it felt rather clumsy. I have revised it using shorter sentences but more words :-). (59 words)

    Liked by 3 people

    1. Peter Apps

      The classic referee’s comments that used to drive me wild were ; “more detail it required on …….”, followed after some interval by “this paper is too long”.

      These days I referee many more more papers than I write and I find that the main cause of excess length is an apparent desire by the authors to demonstrate that they have read every possible piece of literature on the subject and considered every possible explanation for their results.. I suspect that this is because they are just cut and pasting from their M.Sc or Ph.D. theses.

      And, sorry, I can’t resist;

      I just finished a paper for Oxford Bibliography whose strict word limits led me to use compound sentences. Re-reading it showed that a word or two was saved, but it was much clumsier, so I revised it for readability at the cost of slightly increased length. 46 words.

      Liked by 2 people

    2. Pavel Dodonov

      I very much preferred the first version, with compound sentences. It has a better flow and is pleasant to read. Being pleasant, I found it easier to read as well. And notice that I’m not a native speaker 🙂 (But I did read The Silmarillion in English when I was fifteen, not that it really means anything…).
      And I liked Peter’s version even more!

      Liked by 1 person

  2. Elizabeth Moon

    Ease of reading is more a matter of prosody than word or sentence length. I once read a couple of books written as “easy readers” for children who had difficulty reading. Short sentences, short words = choppy unnatural rhythm and boring “story.” These books enforced “word by word” reading, because they did not conform to the natural rhythm of speech.

    I admire sleather2012 for bravely giving us two examples, but I’m going to pick at them anyway. Neither is “bad.” Both could be improved.

    I found that the first example (with long sentences) was easier to read than the second because it had better prosody: it “flowed” better, and thus the long sentences did not interfere with the mental representation of the sounds. The second example felt “chopped” and ploddy.

    If you read these aloud, the first example expects the skilled reader (and academics are writing for skilled readers) to add expression. Rising and falling pitch, changes in volume, tonal variation, and rate of speech mark phrases and clauses and clarify meaning. This creates the “natural speech” flow of good prosody. The second example has a repetitive rhythm and closes with a thump on three of the four sentences. Read aloud, it offers little opportunity for expressive reading. The third sentence, the longest, does offer the opportunity for more expression, but ends weakly with “rather clumsy” after an unnecessarily wordy “although saving on words” instead of “shorter.”

    When tightening a manuscript in revision–whether to cut to a definite length or simply to make it clearer–it’s essential to read the work aloud. In order to read aloud well, read others’ work aloud, particularly that of writers you admire. Notice which colleagues speak expressively; read their writing and notice whether their writing uses the same speech patterns. Unless the thought makes your toes curl, learn enough about poetry to notice and mark the basic rhythmic units (iambic, trochaic,etc.) and how they function in spoken language. Then read your own aloud, before and after any changes.

    In fairness to sleather2012, whose work I arrogantly analyzed without permission, I invite a critique of any of this comment.

    Liked by 1 person

  3. Aaron Dalton

    My preferred metric of clear communication is will the reader get your message *the first time they read it.* Obviously with data it takes time to absorb and understand, but the text itself should be understandable right away. I would rather read four 25-word sentences than a single 84-word whopper of a sentence, for example. As an editor I regularly cut pages of irrelevant material while at the same time encouraging the author to add another five pages in sections where the argument is less supported or less clear. Brevity for brevity’s sake is losing sight of the forest for the trees. But I agree that it is hard for the author to always know how the reader will react. This is why editors are so important! At the very least, have peers in different disciplines read your material.


  4. Pingback: A capital dilemma: “western Australia” or “Western Australia”? | 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.