What if you ordered a bowl of soup in a restaurant and the waiter gave you a bowl of ingredients instead? “Here, make the soup yourself,” he might say.
A lot of scientific writing is like that bowl of ingredients. Scientific writers often string together concepts without connecting the words and phrases in ways that help readers understand the relationships between the concepts. These writers don’t do the work of effectively combining the “ingredients”—they leave that work to their readers.
George Gopen calls this kind of writing “splat prose,” because it’s splattered across the page without consideration of the reader’s needs.1 Readers, in order to understand the meaning of splat prose, have to understand the author’s intention.
But a reader shouldn’t have to guess the author’s intention in order to understand the relationship between the words and phrases in scientific writing. A reader who doesn’t know anything about a subject should be able to read a scientific paper and say, “I understand that A is being compared to B, and that C is being used for the purpose of D, even though I don’t know what any of those things are.”
If you read Lewis Carroll’s famous nonsense poem “Jabberwocky,” from Alice in Wonderland, you can easily understand the grammar, even though most of the words don’t make sense. Carroll organized the words so that their functions and relationships are apparent, even if their meaning isn’t.
The ability to write clearly like Carroll is a skill that takes work to acquire and is separate from subject-matter expertise. (In fact, subject-matter expertise can get in the way of the ability to communicate clearly, because of the “knowledge effect”: the tendency to overestimate how much your audience knows.)
Learning to write clearly is worth the effort for writers, especially in the sciences. Fields that aspire to precision and accuracy shouldn’t produce literature that routinely contains sentences that, grammatically, are incapable of meaning what they’re supposed to mean.
I didn’t survey a corpus of scientific literature especially for this article, but I read a lot of scientific writing as a reader and an editor, and I’ve noticed that certain words and phrases are so commonly misused in scientific writing that whenever you see them, you can almost guarantee that ambiguity and error lurk nearby. By looking out for the following problems in your own writing, you can write more precisely and serve your readers soup instead of ingredients.
What’s wrong with using?
The biggest problem with using is that it requires a user, but scientific writing is often written in the truncated passive voice, which omits the user.
- Active voice: The waiter gave me a bowl of ingredients.
- Passive voice: A bowl of ingredients was given to me by the waiter.
- Truncated passive voice: A bowl of ingredients was given to me.
When the word using appears in a sentence that is written in the truncated passive voice, something else in the sentence will usually be the user, grammatically speaking, and it will often be something ridiculous.
Look at this title of a journal article:
Surgical and patient outcomes using mechanical bowel preparation before laparoscopic gynecologic surgery: a randomized controlled trial2
In this headline, using is both incorrect and ambiguous.
The subject of the headline is surgical and patient outcomes, which can’t use anything. You could try to read using as part of a restrictive clause—outcomes using mechanical bowel preparation, as opposed to outcomes using public restrooms—but that reading is no better, because the outcomes are still using the preparation.
A headline like this one makes readers work harder, because the literal meaning of the headline doesn’t make sense. Literally, the headline says that outcomes use preparation before surgery. These authors have given their readers a bowl of ingredients.
A better title for this paper might have been “Mechanical bowel preparation before gynecologic surgery: surgical and patient outcomes from a randomized, controlled trial.”
If you search PubMed, you’ll find thousands of examples of the misuse of using. You won’t have to search very deeply, either, because many of the examples are in the titles of the articles.
(By the way, the problem with using also occurs with many other -ing verbs. If the thing that is performing the action of the -ing verb isn’t named in the sentence, then the sentence is probably unfinished, like that bowl of ingredients that I keep talking about.)
A movie can be based on a book, but you can’t watch a movie based on your eyes. Unless you watch a movie that is about your eyes, in which case you’re watching a weird movie.
Based on has to modify a noun, not a verb. Nevertheless, scientific writers constantly use based on to modify verbs. When you use based on to modify verbs, you create ambiguity and confusion, because your sentences also contain nouns that readers can mistakenly connect with based on.
Here’s an example:
The combination of these three preoperative conditions was identified as an independent parameter for early recurrence based on multivariate analysis.3
Many readers’ first interpretation of this sentence will be that the recurrence is based on multivariate analysis. But wait—that’s not right. So, the independent parameter is based on multivariate analysis? Or the combination of these … conditions is based on multivariate analysis? Those interpretations aren’t right either.
The authors meant that the identification was based on multivariate analysis, but they incorrectly used based on to modify a verb (identified) instead of a noun (identification), and placed based on too far away from what it’s supposed to modify. As a result, some of their readers either misunderstood the sentence or had to reread and reinterpret it. The first time that you read this sentence, you might have understood what it’s supposed to mean, but it doesn’t actually say what it’s supposed to mean.
Here’s an unambiguous revision:
Multivariate analysis identified the combination of these three preoperative conditions as an independent parameter for early recurrence.
Let’s look at one more example in which the authors misused based on:
The results revealed that SIPT had more efficacy than medication based on both scales.4
(SIPT, by the way, stands for spiritual integrated psychotherapy, and the two scales are qualitative and quantitative.)
Many people will first read medication based on both scales as a single phrase. Based on isn’t only ambiguous and incorrect in this sentence, it’s also unnecessary. None of the concepts in the sentence—results, efficacy, medication, revelation—is based on both scales. The determination-of-SIPT-having-efficacy-on-both-scales is based on both scales, but based on adds nothing but confusion to this sentence. The authors could have omitted based on and fixed the problem:
The results revealed that SIPT had more efficacy, on both scales, than medication.
I searched PubMed for the phrase based on and received 1,080,091 results. I looked at the first 20 results to see whether based on was used correctly or incorrectly, and it was used incorrectly 60% of the time.
The problem with due to is the same as with based on. The phrase due to is supposed to modify nouns, and because of is supposed to modify verbs. For example, the baseball game was canceled (a verb) because of rain, but the cancellation (a noun) was due to rain.
This is one of those picky, seemingly arbitrary rules of grammar that is rooted in the etymology of the phrases. In casual speech, people use due to and because of interchangeably, and few people distinguish between them or even know what the distinction between them might be. I don’t like rules like this one very much, because I’m a pragmatist, not a grammar policeman.
Nevertheless, even though few people know the difference between because of and due to, due to can confuse your reader whenever you use it to modify a verb but your reader can also read it as modifying a noun. Here’s an example from another journal article title:
The risk of biomaterial-associated infection after revision surgery due to an experimental primary implant infection5
Grammatically, this title says that the reason for the surgery was the infection (“surgery due to … infection”), but the authors intended for the title to say that one kind of infection created a risk of another kind of infection after surgery.
Here’s a revision that eliminates the ambiguity:
The risk of biomaterial-associated infection from an experimental primary-implant infection after revision surgery
I eliminated due to and replaced it with a simpler “risk of X from Y after Z” construction. I also hyphenated primary-implant infection, because biomaterial-associated infection is hyphenated, and parallelism helps your readers understand the relationships between concepts.
The phrases compared to and compared with are also like based on and due to: scientific writers attach them, vaguely, to verbs and nouns. As a result, readers can have difficulty interpreting sentences that contain compared with or compared to.
When we compare things, we compare things—we compare nouns to nouns. If you want to compare actions, like the verbs chew and swallow, then you first have to change those verbs to nouns: chewing and swallowing. Now you can compare the act of chewing to the act of swallowing, but you can’t compare chew to swallow. And you can’t chew faster compared to swallowing, either, because what is being compared to the verb chew in that construction? The noun swallowing? The speed of swallowing? The comparison is unclear.
Here’s an example from the literature:
Pathogen identification using Kiestra TLA combined with MS resulted in a 30.6 hr time gain per isolate compared to CM.6
What two things are being compared in this sentence? The two methods or the speed of their results? (Notice, too, that pathogen identification is “using” Kiestra TLA.)
The article from which I took this sentence compares the two methods, but this particular sentence compares the speed at which the two methods identified pathogens. The authors didn’t structure their comparison very well, though, because the sentence compares the speed of the results (a 30.6 hr time gain) to one of the methods, CM. They gave their reader a bowl of ingredients.
Here’s an unambiguous revision:
The combination of Kiestra TLA and MS identified pathogens 30.6 hours faster per isolate than CM.
In these examples, using, based on, due to, compared to, and compared with “dangle.” They are said to dangle because they’re not clearly attached to the words that they’re supposed to be attached to.
Scientific writers dangle their modifiers so often that doing so “has become standard usage in scientific English,” says Joseph M. Williams.7 Williams’ point isn’t that dangling modifiers are now acceptable in scientific writing, but that dangling modifiers are everywhere in scientific writing.
Williams blames the truncated passive voice. He says that if scientific conventions “both deprive their authors of a first-person subject and rule out dangling modifiers, they put their writers into a damned-if-you-do, damned-if-you-don’t predicament.”8
I disagree. Every communication problem has more than one solution, so we, as writers, can always find multiple ways to write coherent sentences in the truncated passive voice without dangling our modifiers. Just be aware of the problem and fix it wherever you can.
Reducing ambiguity isn’t the only benefit of writing clearly. Readers approach texts with finite energy, and you can quickly exhaust them with confusing and poorly structured sentences.9 When sentences clearly establish the relationships between concepts, “[r]eaders … process sentences swiftly and efficiently, and display better recall….”10 Clearly written sentences translate better and more easily, and enhance your arguments by demonstrating clear thinking. In regulated and highly technical industries, clearly written content can reduce risk. The FDA says that “poorly written documents … can impact or slow down the review process…,” so content that is clearly written might be processed and approved more quickly.11 If what we have to say is important, then it is important to our readers that we say it clearly!
- Gopen GD. Expectations: Teaching Writing from the Reader’s Perspective. New York, NY; Pearson Longman; 2004:137.
- Hayes JR, Bajzek D. Understanding and reducing the knowledge effect: implications for writers. Written Communication. 2008;25(1):104-118.
- Won H, Maley P, Salim S, Rao A, Campbell NT, Abbott JA. Surgical and patient outcomes using mechanical bowel preparation before laparoscopic gynecologic surgery: a randomized controlled trial. Obstet Gynecol. 2013;121(3):538-546.
- La Torre M, Nigri G, Lo Conte A, et al. Is a preoperative assessment of the early recurrence of pancreatic cancer possible after complete surgical resection? Gut Liver. 2014;8(1):102-108.
- Ebrahimi A, Neshatdoost HT, Mousavi SG, Asadollahi GA, Nasiri H. Controlled randomized clinical trial of spirituality integrated psychotherapy, cognitive-behavioral therapy and medication intervention on depressive symptoms and dysfunctional attitudes in patients with dysthymic disorder. Adv Biomed Res. 2013;2:5.
- Engelsman AF, Saldarriaga-Fernandez IC, Nejadnik MR, et al. The risk of biomaterial-associated infection after revision surgery due to an experimental primary implant surgery. Biofouling. 2010;26(7):761-767.
- Mutters NT, Hodiamont CJ, de Jong MD, Overmeijer HP, van den Boogaard M, Visser CE. Performance of Kiestra Total Laboratory Automation combined with MS in clinical microbiology practice. Ann Lab Med. 2014;34(2):111-117.
- Williams JM. Style: Toward Clarity and Grace. Chicago, Illinois; University of Chicago Press; 1990.
- Gopen GD. Expectations: Teaching Writing from the Reader’s Perspective. New York, NY; Pearson Longman; 2004:10-11.
- Douglas Y. How Plain Language fails to improve organizational communications. Warrington College of Business Administration Web site. http://warrington.ufl.edu/centers/cmc/docs/rp/TheProblemWithPlainLanguage.pdf. Published 2008. Accessed April 4, 2014.
- Wessling M, Seaman L. Strategies for working successfully with ESL authors. Presented at: American Medical Writers Association 72nd Annual Conference; October 6, 2012; Sacramento, CA.
[This article first appeared in the April 2014 newsletter of the American Medical Writers Association’s Indiana chapter.]