Chapter 4 of The Craft of Scientific Writing

Writing Guidelines
Writing Exercises
Writing Courses

Chapter 1
Chapter 2 (Excerpt)

Note: Until the textbook arrives in all the bookstores for the Fall 1999 semester, this chapter will remain posted so that students can keep up with reading assignments in their classes. Realize, though, that all materials at this web page still are the property of Springer-Verlag.

In scientific writing, precision is the most important goal of language. If your writing does not communicate exactly what you did, then you have changed the work. One important aspect of precision is choosing the right word. Another important aspect of being precise is choosing the appropriate level of accuracy. Just as you wouldn't assign the wrong number of significant digits to a numeral in a calculation, so shouldn't you assign an inappropriate level of accuracy to details in your writing. The appropriate level of detail depends upon your work and your audience.

Choosing the Right Word

As a scientist or engineer, you wouldn't choose the word "weight" when you meant "mass." Technical terms such as "weight" and "mass" have specific meanings. Many ordinary words have specific meanings as well. For instance, you shouldn't choose, as many writers do, the word "comprise" when you want the word "compose."
Water is comprised of hydrogen and oxygen.
Because "comprise" literally means to embrace or to include, the above sentence is imprecise. A precise way to write this sentence is
Water is composed of hydrogen and oxygen.
There are several word pairs that give difficulty not only to scientists and engineers, but to all writers. Listed below are some of these word pairs and the differences in their meanings.
affect/effect: Affect is a verb and means to influence (note that in psychology, affect has a special meaning as a noun). Effect is a noun and means a result (occasionally, people use effect as a verb meaning to bring about).
continual/continuous: Continual means repeatedly: "For two weeks, the sperm whales continually dived to great depths in search of food." Continuous means without interruption: "The spectrum of refracted light is continuous."
its/it's: Its is the possessive form of the pronoun it and is defined as "of it." It's is a contraction and is defined as "it is."
like/as: Like is a preposition and introduces a prepositional phrase. As is a conjunction and introduces a clause.
principal/principle: Principal can be either a noun or an adjective. As an adjective, principal means main or most important. Principle appears only as a noun and means a law ("Archimedes' Principle").
For a more detailed list see Appendix B.
Another precision problem, aside from that of selecting a word with an incorrect meaning, is grouping words such that they have no meaning:
The problem centers around the drying and aging of the bloodstain.
The phrase "centers around" makes no physical sense. What the writer wants is either "centers on" or "revolves around."
The problem revolves around the drying and aging of the bloodstain.
Many problems with word choice do not arise from confusion with other words or with lack of meaning. Rather, these problems arise because there is only one word that will do, and any substitute will cause imprecision. Despite what you may have learned, few words, if any, are exact synonyms. Some words have similar meanings, but are not interchangeable. Consider, for example, these two terms from the electrical breakdown of gases:
gas discharge: any one of the three steady states of the electrical breakdown process. These three steady states are the Townsend, glow, and arc.
spark: the transient irreversible event from one steady state of the electrical breakdown process to another (example: the transition from a glow discharge to an arc).
To use these two terms as synonyms is imprecise language. The first term is a steady state. The second term is a transition. However, in the first sentence of a recent journal article, a scientist tossed these terms around as if they were synonyms:
The last decade has seen a rapid development of new techniques for studying the enormously complex phenomena associated with the development of sparks and other gas discharges.
Because a "spark" is not a "gas discharge," this sentence is imprecise. More important, because this sentence is the first sentence of the article, this imprecision undercuts the article's authority. Why did the scientist make this mistake? Given that the article was a review article and that the journal had solicited the scientist to write it, the scientist most likely didn't make the mistake because he didn't understand the vocabulary. He probably made the mistake because he was concentrating on the rhythm of the sentence rather than on the meaning. In other words, he wrote as if being fluid in the sentence was more important than being precise--something appropriate for poetry perhaps, but not for scientific writing.
Many scientists and engineers hold the misconception that using synonyms is a mark of a good writer. These scientists and engineers write with a pen in one hand and a thesaurus in the other. Well, synonym writing is not strong writing. Even when you find true synonyms, using them often confuses your readers:
Mixed convection is a combination of natural and forced convection. Two dimensionless numbers in the correlations for mixed convection are the Grashoff number and the Reynolds number. The Grashoff number (for free convection) is a measure of the ratio of buoyant to viscous forces, and the Reynolds number (for forced convection) is a measure of the ratio of inertial to viscous forces.
Why did the engineer use "free convection" instead of "natural convection" in the third sentence? This synonym substitution added nothing to the discussion and only served to confuse readers unfamiliar with the vocabulary of heat transfer.
Another reason not to use synonyms is how inexact synonyms are. Consider how quickly these strings of synonyms, taken from a thesaurus, arrive at the antonyms of the first words:
Most professional writers don't hesitate to repeat a word if that word is the right word. Moreover, most professional writers don't rely on a thesaurus. Most professional writers rely on dictionaries because dictionaries tell you the differences between words. Dictionaries help you find the right word.
I once had an English teacher who made us write five-sentence paragraphs about different subjects--subjects such as whales. In the paragraph, whales had to be the subject of each sentence, but you couldn't use the word "whales" more than once. You couldn't use pronouns, either. Instead, you had to come up with four synonyms for "whales." Which students did she praise? The students who gave up and decided that anything but whales was imprecise? No. She fell all over herself for the freckle-faced boy in the front row who wrote "king and queen mammals of the sea" and "black ellipsoids of blubber." With an evil eye for the rest of us, she said, "Now, he's a writer."
I have news for her--"black ellipsoids of blubber" is no kind of writing. Not scientific writing, not journalism, not fiction. You should test any writing advice you've received against the writing of great writers. Did Flannery O'Connor use a string of synonyms in her writing? Did Churchill? Did Einstein? No. They used the right word and only the right word. As Mark Twain said, "The difference between the right word and the almost right word is the difference between 'lightning' and 'lightning bug.'"
Besides the dictionary meaning (or denotation) of words, you should also worry about the connotation of words. A word's connotation is its associated meanings. Many words, such as "adequate," conjure an associated meaning that works against the dictionary meaning. In the dictionary, "adequate" means enough for what is required. However, would you fly on an airplane with an "adequate" safety record? Probably not, because "adequate" has a negative connotation. For many people, the connotation of "adequate" is the opposite of its denotation. When something is described as "adequate," many people think of that something as being insufficient. Another word with a strong negative connotation is "cheap." Although "cheap" has the same denotation as "inexpensive," the connotation differs. The connotation of "cheap" is that the item so described will not work.
Using a word with a negative connotation when a neutral or positive connotation is wanted is weak writing. So is the opposite, using a word with a positive connotation when a negative or neutral connotation is desired:
The turbulence in the flow enhances the drag by more than 20 percent.
Because drag was an undesired quantity in this example, the engineer should have chosen a verb with a neutral or negative connotation ("enhance" has a positive connotation). Note that the denotation of "enhance" is also inappropriate here. To enhance something is to make an incremental change, but 20 percent is not incremental. A better word choice would have been either "increase" (neutral connotation) or "exacerbate" (negative connotation).
The turbulence in the flow increases the drag by more than 20 percent.
Finally, in choosing the right word, go in fear of absolutes, especially in fear of the adverbs "always" and "never." Whenever you use these words, you invite your audience to look for exceptions, and if exceptions do exist, your audience will find them.

Choosing the Right Level of Detail

The previous section discussed precision from the viewpoint of choosing the right word. Precision also involves choosing the right level of accuracy for the details in your sentences. In scientific writing, you should strive not for the highest degree of accuracy, but for the appropriate level of accuracy.
How do you attain the appropriate level of detail for your language? One way is to achieve a balance between general statements and specific details. Strong writing requires both general and specific statements. General statements establish the direction of thought, and specific statements give evidence to support that direction. Writing, however, that relies solely on general statements is empty. Consider this stand-alone entry from a progress report to the Department of Energy:
After recognizing the problems with the solar mirrors, we took subsequent corrective measures.
What were the problems with mirrors? What were the solutions? How many mirrors were damaged? This entry raises questions, but does not address them. Given that the field of solar mirrors cost over $40 million, this entry in the progress report did not satisfy the Department of Energy. A more precise entry to the report would have been as follows:
Our last progress report (March 1985) discussed the damage to ten solar mirrors during a February thunderstorm. The question arose whether high winds or hailstones had cracked those mirrors. Now, after finding that high winds had caused the cracks, we have begun stowing all solar mirrors in a horizontal, as opposed to vertical, position during storms.
Another reason to use specific details is that general statements, by themselves, will not leave much of an impression on your audience. By themselves, generalities are soon forgotten.
Our new process reduces emissions of nitrogen oxides from diesel engines and industrial furnaces.
Replacing this generality with a specific detail gives your audience something concrete to remember. Take a lesson from fiction writing. Good fiction writers rely on specific details to create scenes because good fiction writers know that specific details are what readers remember.
Our new process eliminates 99 percent of nitrogen oxide emissions from diesel engines and industrial furnaces.
Better yet, by grounding that detail (99 percent reduction), you insure that your audience understands its importance.
Our new process eliminates 99 percent of nitrogen oxide emissions from diesel engines and industrial furnaces. Previous processes have, at best, reduced nitrogen oxide emissions by only 70 percent.
Do not assume that general statements are inherently weak. In fact, sometimes general statements are desired. For instance, presenting specific details without general statements can be dangerous. Consider this example from an article on radon levels to a non-technical audience:
The average house in the area has a radon level of 0.4 pico-curies per liter.
Without a general qualifier, the audience is left with the question, How dangerous is 0.4 picocuries? That kind of writing is not only weak, but irresponsible. Revision gives
The average house in the area has a radon level of 0.4 pico-curies per liter, which is considered low by the EPA [Lafavore, 1987]. Levels between 20 and 200 picocuries per liter are considered high, and levels above 200 picocuries per liter are considered dangerous. For reference, the average radon level in outdoor air is about 0.2 picocuries per liter.
Sometimes specific details confuse because they give too much information. In other words, the specific details raise side issues for readers that the writer does not intend.
The number of particular hydrocarbon combinations in our study is enormous. For example, the number of possible C20H42 is 366,319 and the number of C40H82 is 62,491,178,805,831.
What was the purpose of including these two numbers? The chemist wanted to show the extent of his calculations. However, were all those digits necessary? Also, because the second number was so much larger than the first number, was the first number necessary? A more precise paragraph would have been as follows:
The number of hydrocarbon combinations in our study is enormous. For example, the number of possible C40H82 is over 60 trillion.
This revision achieves the desired result--showing the extent of the calculations--without making the audience wade through undesired numerals.
Consider another example, this one from a progress report about a solar power plant:
Operations at the plant stopped momentarily because the thermal storage charging system desuperheater attemperator valve was replaced.
The name "thermal storage charging system desuper-heater attemperator valve" is a problem. For one thing, when written as a single noun phrase, it is difficult to read. For another thing, the report's readers (plant managers) did not know exactly what this particular valve was. All the readers knew about this valve was that it was in the thermal storage system. For that reason, a more appropriate level of accuracy would have been as follows:
Operations at the plant stopped for 1.5 hours so that a valve in the thermal storage system could be replaced.
In this revision, the location of the valve was made less specific, while the time that the plant was down was made more specific. This report's readers cared more about how long the plant was down than about which particular valve was leaking.
Packing sentences with too many details also makes for tiresome reading:
A 1-mm diameter, 656-nm beam with uniform intensity across the beam was produced by using a wavelength/polarizer combination to split off part of the 532-nm output from the Nd:YAG laser to pump a second dye laser (Laser-Ray LRL-2, also operated with the DCM dye) with a side-pumped configuration for the final amplifier, and selecting the central portion of the collimated beam with an aperture.
There are just too many details in this one sentence. Are all these details necessary? Couldn't the physicist have spread these details over several sentences? Better still, couldn't the physicist have placed the secondary details (such as beam wavelength and manufacturer's name) in an illustration? Scientists and engineers sometimes worry so much about telling readers everything that they end up not informing readers of anything.
Being precise doesn't mean compiling details; being precise means selecting details. You should choose details that inform:
The fuel pellets used in inertial confinement fusion are tiny, the size of BBs, but they are potentially the most powerful devices mankind has ever known. If we can compress the fuel in the pellets to a plasma, the fuel's deuterium and tritium atoms can overcome their mutual electrical repulsion and fuse into helium atoms, giving off energy (E = mc2). The power needed to ignite fusion in the pellets is 100 trillion watts; however, the power released from the fusion is one hundred times that much.
This paragraph informs; it informs because the scientist selected the most important details about the pellets. Because scientific writing is compressed, you have room for only the most important details. Make them count.


Bennett, A., Literary Taste and How to Form It (London: George H. Doran Publishers, 1909).

LaFavore, M., Radon, the Invisible Threat (Pennsylvania: Rodale Press, 1987).

Back to Top

Last updated 1/99

© Springer-Verlag New York, 1996. All materials at this web page are property of Springer-Verlag. Please do not download, reproduce, or use elsewhere without written permission. If you are a teacher who is waiting on a reprint of the book, hypertext links may be made from your syllabi to this page as long as proper credit is given.