7 Keys to Clear Scientific Writing

I recently had the pleasure of giving presentations (on three consecutive days; phew!) to the postdocs at Caltech, the postdocs and graduate students at UCLA, and the graduate students at USC. Because so many of these people were scientists, I thought this would be a good time for a newsletter about how to write a clear scientific article.

Why Bother Writing Clearly?

You might be wondering why you should even bother to write well.  After all, as one person mentioned to me after my UCLA talk, the standards for writing are not as high in science as in, say, English.

You need to sell it or they won't buy it.  If you care about having your work accepted and admired, it's important how you present it.   Richard W. Hammer, a successful former Bell Labs scientist, once stated, that "it's not enough to do a job; you also must sell it."  Every time you write something for publication (or work on your dissertation), you are trying to convince your audience that your ideas are worthy of attention.  So it's worth your while to learn how to do that well.

You must reach your reader or you haven't done science.  In "The Science of Scientific Writing," George Gopen and Judith Swan* point out that science is accomplished through accurate communication.  No matter how lovely your data, you're not accomplishing adequate science if you don't do a good enough job of communicating it to others.

Clear writing helps you think better. I've hammered home the point in previous newsletters that you write in order to find out what you think.  Gopen and Swann make a further point -- they suggest that writing and thinking are so interconnected that if you improve one, you improve the other.  So it's worth writing better for the simple reason that you will accomplish better science.  This argument is another reason to "write before you're ready;" that is, don't wait until you have all your data perfectly analyzed before you start writing.  Writing often and writing more clearly will help improve the caliber of your thinking, and.thus your science.

Unclear writing annoys your reader.  If you make readers struggle to figure out what you mean you will annoy them and make them stop reading.  Enough said.

How to Write Clearly

Know what the reader expects. If you don't know how to meet people's expectations when you write, then you will confuse them and lose them.  Gopen and Swann offer some guidelines that will make your science writing easier to read, because they take reader expectations into account. Among other things, the reader expects:  

• to be reminded at the beginning of a phrase, sentence or paragraph what has come before,
• to be provided with context,
• to learn one thing new at a time, and  
• to be led into the new, exciting material.

I will offer a simplified version of their suggestions.  These will get you started writing more clearly, but to understand these concepts in depth, I suggest you read their article.

Use these guidelines judiciously.  While you are writing your first draft, don't check whether each sentence or phrase meets these guidelines.  Working this way will slow down your thought processes and interfere with your productivity.  Write freely (as I often urge you to do), and then when you revise your work, notice where the text seems murky or note where readers misunderstand your meaning.  Then use the following list as a way to tease apart how the offending writing has gone wrong.

Ease the reader into the sentence/phrase

1. Refer back to "old information" (what you've already addressed) at the beginning.  This provides linkage to what has come before and prepares them for the new idea to come.  (e.g., "These figures…")
2. Set the context or perspective for what is to follow early in the sentence.  (e.g., "these figures support")
3. Put the topic in the beginning of the sentence.  (e.g. "figures")

General rules for the sentence/phrase

1. The verb should follow the subject of the sentence with few intervening words. (e.g. "figures support")
2. Make only one point per sentence (e.g. "These figures support our prediction that x will lead to y.")

End the sentence/phrase with a bang

1. Put the exciting, new information at the end of the sentence (e.g. NOT "Our prediction that x will lead to y is supported.")
2. Put the emphasis at the end of the sentence

By reading these guidelines, you can see which of the following sentences is preferable:

• The idea that cold fusion is possible, which is the prediction that we had previously made, was supported by these data in a clear manner.
• The data clearly support our prediction that cold fusion is possible.

Sentence 2 is better, and not because it uses the active voice and is shorter.  It is better because it conforms to the guidelines proposed by Gopen and Swan, which take reader expectations into account.

Apply these guidelines to your previous writing.  Try looking over the last few pages that you have written and see if you can improve them by taking reader expectations into account.  I welcome all feedback, so let me know how it goes!

*Thanks to Åsa K Rennermalm, Ph.D, a postdoc in geology who attended my talk at UCLA, for letting me know about this article, and for her kind words about my talk!

Warmly,
Gina