Saturday, December 5, 2009

Scientific Writing Secrets Revealed!

This post continues the series on writing, but focuses more specifically on scientific writing.  Before your eyes glaze over, let me hasten to add that this will not be your routine discussion of grammar, punctuation, and style.  I’m assuming that readers have a basic grasp of these skills.

Instead, I’ll be reviewing guidelines given in a paper by Gopen and Swan (1990) called “The Science of Scientific Writing”.  In this paper, they analyze what it is about scientific writing that makes it so difficult to read (and enjoy).  The authors don't stop there, but go on to develop clear rules for avoiding incomprehensible writing.  The underlying message of the paper can be summarized thus:

“If the reader is to grasp what the writer means, the writer must understand what the reader needs.”

Gopen and Swan go on to state that the rhetorical principles they outline produce “clarity in communication without oversimplifying scientific issues.”  They argue that the results [of applying these principles] are not cosmetic, but that “improving the quality of writing actually improves the quality of thought.” 

Amen.

If you are interested in reading their paper, the reference is given at the end of this post.  However, if you’d rather not read it, I’ll be giving the CliffsNotes version.  What I will do is take the same approach they did and use an example from a technical paper to illustrate their principles.  It’s something of an exercise for me, but you can follow along and see how well we both do in figuring out how to improve the writing in the selected excerpt. 

I selected an example at random from the internet.  I could have picked anything, but this was the first to pop up.  It is an abstract from a journal focused on optics, a topic I know absolutely nothing about but that appears to be fraught with convoluted, jargon-ridden language.  By the way, my example is not nearly as dense and full of jargon as the one Gopen and Swan use.  Here it is:

The discrete-dipole approximation (DDA) for scattering calculations, including the relationship between the DDA and other methods, is reviewed. Computational considerations, i.e., the use of complex-conjugate gradient algorithms and fast-Fourier-transform methods, are discussed. We test the accuracy of the DDA by using the DDA to compute scattering and absorption by isolated, homogeneous spheres as well as by targets consisting of two contiguous spheres. It is shown that, for dielectric materials (¦m¦ ≲ 2), the DDA permits calculations of scattering and absorption that are accurate to within a few percent.

Here is my translation of the abstract:  The authors reviewed the use of a computational method (DDA), tested its accuracy in computing light scattering and absorption by different types of spheres, and found the method to be highly accurate.

Most people would find this paragraph moderately difficult to understand, but not for the obvious reasons of technical jargon or lack of background in the field of study.  So what is the problem?
 
The first problem is subject-verb separation.  The first sentence places a string of words between the subject (“DDA”) and verb (“reviewed”).  Any words placed between the subject and its verb are viewed by the reader as interruptions and of lesser importance.  The reader must wait a long time to get to the verb and understand what the whole sentence is about.  In the meantime, the reader may be mentally skipping over key information in an attempt to close the gap between subject and verb.  Also, one wonders if DDA is used to calculate something or if its purpose is instead to scatter calculations around the room.  The second sentence similarly interjects a string of words between subject and verb.  These two sentences also state that something “is reviewed” and “is discussed”, a no-no in an abstract.

How would you go about fixing this subject-verb separation problem and write something that is more consistent with reader expectations?  We're just focusing for now on the first two sentences of the example.  Think about it, and I’ll provide my revision in the next post.

“Information is interpreted more easily and more uniformly if it is placed where most readers expect to find it.”

Gopen, G.D. and J.A. Swan. 1990. The science of scientific writing. American Scientist 78: 550-558.

1 comment:

Comrade PhysioProf said...

Here's how I would rewrite the abstract:

We review several aspects of the discrete-dipole approximation (DDA) for scattering calculations, including the relationship between the DDA and other methods. We discuss the computational considerations relating to DDA of complex-conjugate gradient algorithms and fast-Fourier-transform methods. We test the accuracy of the DDA by using the DDA to compute scattering and absorption by isolated, homogeneous spheres as well as by targets consisting of two contiguous spheres. Using these approaches, we demonstrate that for dielectric materials (¦m¦ ≲ 2), the DDA permits calculations of scattering and absorption that are accurate to within a few percent.