Writing like a scientist (or not)

A charming essay by Adam Ruben on how to write like a scientist, whether you want to or not (beware: tongue planted firmly in cheek).
How to Write Like a Scientist, 23 March 2012, Science.

“Don’t, please, please, for God’s sake, don’t.”

Deirdre McCloskey is an economic historian and rhetoritician at the University of Illinois – Chicago. She has written extensively on the rhetoric of economics and social sciences. One of her gems is a short (of course!) book called Economical Writing (2nd ed., 2000). There are a number of helpful books that guide and advise scholars on writing; this is my favorite.
One of Deirdre’s lessons I’ve embraced almost as much as William Strunk’s “Form the possessive singular of nouns with ‘s” is her condemnation of the “table of contents paragraph” (known by others as the “roadmap” paragraph):

Still another peice of boilerplate, and one which kills the momentum of most papers in economics on the second page, is the table-of-contents paragraph: “The outline of this papers is as follows” Don’t, please, please, for God’s sake, don’t. Nine out of ten readers skip to the substance, if they can find it. The few who pause on the paragraph are wasting their time. They can’t understand the paragraph until, like they author, they have read the paper, at which point they don’t need it. Usually the table-of-contents paragraph has been written with no particular audience in mind, least of all the audience of first-time readers of the paper. Even when done well it lacks a purpose. You will practically never see it in good writing, unless inserted by an editor who doesn’t know good writing. Weak writers defend it as a “roadmap.” They got the idea from Miss Jones: “Tell the reader what you’re going to say. Say it. Say that you’ve said it.” It’s exceptionally bad advice, and the person who made up this memorable phrasing of it is burning right now in Hell.” (p. 37)

Plagiarism sucks

It has been my casual impression, as a teacher, that graduate students are plagiarizing more frequently than they use to (I haven’t taught undergrads since the earlier 90s, so I have no opinion about them). This view is widely held, I sense, though I don’t know if there is much careful evidence to support it. Certainly there is a sense that plagiarism is easier, now that there is so much text available on the Internet and cut-and-paste is so easy; the economist in me thinks that if the costs have gone down and the expected penalties are not much higher then the amount of plagiarism will increase.
(It is not immediately obvious that expected penalties have increased: it is also somewhat easier to catch Internet-based plagiarism, since search tools facilitate finding matching phrases. But as a teacher I think that the ease of catching has not increased as much as the ease of plagiarizing.)
In confronting students in the past few years, I also have the casual sense that there is less understanding of why plagiarism is unacceptable in academia. A typical response is “I did the research, I found the information, why should I have to rewrite it in my own words?”
Jack Shafer has published a column in Slate.com magazine entitled “Eight Reasons Plagiarism Sucks“. He focuses on plagiarizing in journalism, but his concerns for the most part translate to academia as well, and are worth reading by scholars. Most people seem to understand that plagiarism is a type of theft (but doubt that original authors lose much or care much if they aren’t cited in an unpublished term paper). But Shafer emphasizes harm to original authors is but one, and perhaps the least concern. His other concerns include “Journalism is about truth, not lies”, “It corrupts the craft”, “It promotes the dishonest”.
When I try to educate straying students, I, too, emphasize the importance of trust between author and reader. I point out that if a boss finds that an employee is cutting corners and plagiarizing, even if the ideas presented are good, the boss will stop trusting the employee, and that is the beginning of the end.
There is another concern for students: You don’t learn as much when you plagiarize. Learning is much more than learning to find what others have written on a topic. Even if you do nothing more than rewrite the ideas in your own words, cognitive psychologists have convincingly shown that the deeper processing and rehearsal that this entails greatly improves the writer’s learning of the material. (Which is not to say that rewriting someone else’s ideas is enough to avoid plagiarism: you need to provide attribution for the ideas, as well.) And there is also the not small matter of learning how to write in the first place!

LaTeX for new scholars

In my primary research fields (economics and computer science) most scholars use LaTeX for text processing, as in many other fields. It’s a fabulous tool for those who write scholarly text, especially if the work involves formulas, tables or figures.
It is not a WYSIWYG word processor, but a markup language backed by the world’s best automatic typesetting system (the underlying program, TeX, was written by the god of algorithms, Donald Knuth). On most platforms, however, there are integrator programs that do a nice job of providing an editor, a compiler and a previewer, which on today’s fast machines means that you can see the result of what you’re doing almost instantly. The advantages over word processors (like Word) is that the system is built soundly on scientific principles of typesetting so its default output is gorgeous, and it is infinitely configurable and extensible. (I produce essentially all of my papers in LaTeX; go download one of my recent papers to see an example of how it looks.) And the community of professionals who have written packages and extensions for it is huge and dedicated. The learning curve is a bit steep, though the defaults are so fine that to get up and started on basic term papers, etc., does not take long, especially for anyone with the skills and intelligence of a Ph.D. student.
There are a huge number of guides for starting to learn LaTeX. I still find that my first reference is often the original (though revised) by Leslie Lamport. A widely regarded online guide for learners is Nicola Talbot’s LaTeX for Complete Novices. One you get the basics, the two go-to guides I most often consult are by Mittelbach et al., and The Not So Short Guide to LaTeX 2&epsilon by Oetiker et al.
Nicola Talbot has written (still in process but current draft is available online) a new guide, specifically for Ph.D. students: Using LaTeX to Write a Ph.D. Thesis.

Time to write

Over the years, I have frequently seen advice from professional writers that to get writing done, it is best to:

  • set aside some time to write every day, and
  • force yourself to write something during that time.

One recent example was advice my colleague Yan Chen shared from a workshop she attended given by Jayne London, an “academic coach” who advises faculty members on writing. London advised that “Writing in short, regular sessions, e.g., 30-60 minutes every work day,leads to higher productivity than binge writing. Even ten minute sessions are better than binge writing.”
Another academic coach, Mary McKinney, recommends:

The Tolerable Ten
If you’ve been putting something off, it helps to start small. Begin working for just ten minutes on the daunting tasks of your life.
Almost any task, no matter how unpleasant, or anxiety provoking, can be tolerated for a short amount of time.
When you are having difficulty sitting down to work, set yourself the small but significant goal of working for just ten minutes on the project. After you’ve fulfilled that promise to yourself, you are free to either continue working or to stop.

For more of Mary’s thoughts, see “Overcoming Procrastination”.
Yet another coach, who focuses on advising college students, is Cal Newport, who writes the “Study Hacks” blog. In a 15 October 2007 article, he extracted writing advice from interviews (by others) of ten successful non-fiction writers.

  • 9 out of 10 write in the morning, 4 in the afternoon, 3 at night. Only one reported writing all three times.
  • Most set a specific starting time, and for most it is 8.30 am or earlier.