If you’ve read pretty much anything I’ve written, you might have noticed how much I like the em-dash. I try to pay attention to writing styles, and like many people, I find the em-dash—which delivers a pause if used singly or a semi-parenthetical statement if used doubly—to be an essential component of my own writing style. I was therefore very excited when I read this piece in the New York Times last year, which extols the virtues of the em-dash. Author and English professor Ben Yagoda gets it exactly right when he says the following:
“Writers who deploy this mark comfortably and adeptly (rather than haphazardly) are conscious of the rhythm and dynamics of a sentence. A well-placed dash adds energy and voice.”
But then, just one sentence later, Yagoda says:
“[The em-dash] proposes a long pause—slightly longer than a parenthesis, significantly longer than a comma—that in a subtle way calls attention to itself; as the linguist Geoffrey Nunberg has remarked, dashes are primarily found in “genres that permit reference to be made to the act of composition, whether the break indicated by the dash is genuine or artful invention…” (In other words, be wary of using them in an international treaty or a scientific paper.)”
Oh no!
I was really quite dismayed to read that even a strong proponent of the em-dash felt it has no place in scientific writing. But then I realised that Yagoda isn’t a scientist, and perhaps scientists are actually okay with using em-dashes. So I googled “scientific writing em-dash”; distressingly, the two most relevant hits had this to say:
“But note that the em-dash is rather informal and rarely used in scientific writing.”
“Em dash (―) is an even longer dash [than a hyphen or en-dash] that is not used in scientific writing”
This wasn’t too surprising; after all, almost everyone I’ve asked to edit my scientific manuscripts suggests eliminating em-dashes. And these suggestions work—one of my publications has gone from five em-dashes in an intermediate draft to two in the final article. But I remained adamant that em-dashes can be used effectively even in a scientific paper, and set out to find an example.
My first stop was the work of a former labmate, who I knew was a fan of the em-dash as well. In reading Yagoda’s comment above, this labmate had said:
“I think his comment stems from a general misunderstanding of scientific writing – that it should be made intentionally bland. In fact, I think the two top reasons he outlined for using a dash – to be “conscious of the rhythm and dynamics of a sentence” and to add “energy and voice” – are exactly why we need them in scientific writing.”
However, a quick (and certainly incomplete) scan of some of his recent papers revealed no em-dashes, which really goes to show that it can be tough to incorporate them into scientific writing [update: I stand corrected. He did manage to sneak in a few in the papers I looked at, which I missed in my scans]. I then shelved this post for later, and began to hunt for a paper that illustrated the utility of these punctuation marks.
[two weeks later…]
Scott-Phillips et al. (2014) is a remarkable paper in many regards, foremost being their use of a novel format that they call an adversarial collaboration, resulting in a fairly balanced take on a controversial issue in evolutionary biology. But what’s relevant here is how well this paper is written—it’s one of the more readable scientific papers I’ve ever come across. And encouragingly, they have several great examples of em-dashes! This one is my favourite, using the em-dash to make a potentially involved sentence very clear:
“From a conventional perspective, beavers’ dams are ‘extended phenotypes’ (Dawkins, 1982), which evolve in essentially the same way as other aspects of the beaver phenotype—through the differential selection of dam-building alleles.”
And here’s another example. It might be the closest one can come to being funny in a scientific paper, and the punch of the sentence comes from an em-dash:
“For example, parasites sometimes compete with one another to extract as many resources as possible from hosts, and in doing so kill the host more quickly—an obvious reduction in the host’s fitness.”
For the sake of completeness, here’s a rather amusing piece illustrating how, like with anything, one can overuse the em-dash.
Ambika Kamath 
ugh…good writing is good writing and adherence to strict convention or standards in one genre versus another is ridiculous. If a writer understands the proper use of certain punctuation and uses it as they intend, then who cares (see for example, the passive voice).
Also, see this 2009 piece in the Chronicle of Higher Education about “stupid grammar advice”.
http://chronicle.com/article/50-Years-of-Stupid-Grammar/25497/
I agree, and since good writing is largely learnt, it’s unfortunate that people still make unwarranted declarations about style while teaching it. And thanks for sharing that piece, I love it! This was my favourite bit (“they” refers to the Strunk and White grammar and style guide):
“”Write with nouns and verbs, not with adjectives and adverbs,” they insist. (The motivation of this mysterious decree remains unclear to me.)”
Great post, Ambi! You’ll be happy to know that in development research, em-dashes are used effectively and frequently – in a fair few of my favourite articles, for example 🙂
Thanks for stopping by, Sareen! And yes, I am happy to learn that 🙂
I don’t think dashes have a place in scientific writing, and here’s why:
You say in the comments that “good writing is good writing.” I suppose that in a certain platonic sense this might be true, but it is not useful to the question at hand. Good writing is writing that achieves its purpose; to say what good writing is, we must investigate the purpose.
The purpose of scientific writing is to communicate, from one mind to another, observations of the world and ideas that explain them. Science depends on reproducibility, not just of data, but of thought. Imagine you wrote a paper proposing Ambika’s Hypothesis X-Y-Z, but wrote it in such a way that we all could interpret the hypothesis differently: our testing of the hypothesis would be meaningless, since everybody was not testing the same thing. For this reason, clarity of meaning is the sine qua non of science writing. Other sorts of writing are not so constrained: for instance, if you read a poem and come away with a perfect understanding of the thought of the poet, you just read a crappy poem.
And this explains why people (including me), don’t like dashes in scientific writing, although I share your fondness for them in my casual prose. You discuss two uses of dashes:
1) The single dash, which separates two clauses. The problem is that the dash (like my favorite other inappropriate punctuation mark, the semicolon) declares no logical connection between the two clauses. The dash permits authors to connect two statements without explicitly stating what the logical connection between them is. This undermines the clarity of thought that is the supreme virtue of scientific writing. This is clear in the second example you give: does “an obvious reduction in host fitness” describe “parasites compet[ing] with each other,” “extract[ing] as many resources as possible,” or “kill[ing] the host,” or some combination thereof? What precisely the author means varies depending on what you choose, especially with the first of the three.
The second example is less concerning, because the author includes a conjunction (sensu lato) right after the dash, so now we know what the relationship between the two clauses the author intends. But if a “great example” of a dash is one which can be replaced by a comma with no alteration in meaning, I don’t think that it speaks very well for dashes.
2) The double dash, which separates an appostive from the sentence it apposits. Like above, these dashes can usually be replaced with commas, except in cases where there are already too many commas floating about, and you don’t want to create confusion in the reader regarding which commas are functioning in pairs, and which are solitary creatures serving some other purpose. If these solitary commas are inside your appositive, I think that’s a sign that the appostive clause is doing too much grammatical work: you need to de-apposit it, and put it in a dependent clause with some sort of word that will clearly mark its logical relation to the rest of the sentence.
So, I suppose that–in science writing–I will allow for the use of paired dashes to mark off a brief appostive in an otherwise complex sentence. But this is all. I can’t think of any other acceptible use. Take your dashes and replace them with commas: if the sentence looks wrong, it’s probably because you need some sort of conjunction there, but that just means you needed the conjunction all along. I don’t think you need to worry about dulling your prose, either. If your prose had “energy and voice” before, it will still have them; if it doesn’t, no amount of unorthodox punctuation is going to help.
Jack, thanks for your thoughtful comment, and sorry for taking so long to reply! I mostly agree with you, and think the point of complete clarity is important. So yes, perhaps the second example above is a poor use of an em-dash. And yes, the first example’s em-dash can be replaced with a comma without altering the meaning. But it would change the energy and the voice of the sentence substantially, at least how I read it. It would shift the emphasis away from the last clause in a way that makes the sentence much more boring. I stick by my description of it as a great example of an em-dash in scientific writing!
We’ve actually come across a lot of manuscripts in which authors utilize the em dash. We feel it’s more appropriate in either an introduction or conclusion of a paper (or if/when they are responding to peer review comments).
From our perspective, if authors include such formatting within the body of the document, we often convert these parts of the sentences and encourage parentheses.