Kathy Underwood

“Rules may obviate faults, but can never confer beauties.”Samuel Johnson

Here’s a quiz for you grammar and language buffs: What grammatical principle (1) permits you to ignore syntax for the sake of sense and (2) gets you a bingo (50 extra points!) in Scrabble?

The answer is synesis (\ˈsi-nə-səs\), which Webster’s defines as follows:
a grammatical construction in which agreement or reference is according to sense rather than strict syntax (as anyone and them in “if anyone calls, tell them I am out”)

Synesis is sometimes calleda subcategory of notional agreement or notional concord (because of its following the notion of the meaning rather than the form or class of the word in question). The agreement in question is that Synesis concerns agreement between a verb and its subject or between a pronoun and its antecedent. While the practice is more common in British English, it turns up in American English more frequently than you might expect. A simple example of synesis would be “Eight years is a long time.” Here, we end up treating the notion “eight years” as a single unit, not a collection of units. In other words, we’re treating it just as we would the word “decade.”

But we more frequently follow synesis when we use certain “nouns of multitude,” such as amount, majority, bunch, group, a lot, percentage, and the like (Garner, 1998). The syntax in question most often consists of the noun of multitude plus the preposition of and a plural noun. An example would be “A small group of the swimmers were able to avoid the attacking sharks.” Conventional syntax (what Webster’s Concise Dictionary of English Usage calls “school-grammar agreement”) would tell us to match the verb to bunch, which is singular. But synesis—sense—tells us to match the verb to swimmers, even though it’s the object of a preposition.

I should note that not all usage commentators agree with the concept of synesis—even when they use it themselves. But the trend has persisted since around the tenth century (in Old English), so I think that we can count on its continuing influence. Bryan Garner has aptly characterized the phenomenon: “The problem lies just outside the realm of logic, in the genius of the language.”

But what, you may ask, do I do about the linguistic anguish caused by our having no generic third-person singular pronoun in English? When should synesis be preferred to “school-grammar”? Controversy continues over these topics. But for a thorough and helpful discussion of bias-free language and issues with pronouns, see the chapter entitled “Grammar and Usage” in The Chicago Manual of Style. I believe that it’s the closest we have to a consensus view at this time.

References

• The Chicago Manual of Style, 16th ed. 2010. Chicago: University of Chicago Press.
• Garner, Bryan. 1998. A Dictionary of Modern American Usage. New York: Oxford University Press.
• Johnson, Samuel. “Prudence.” The Idler, No. 57, Saturday, May 19, 1759.
• Merriam-Webster’s Concise Dictionary of English Usage: The Essentials of Clear Expression. 2002. Springfield, MA: Merriam-Webster, Incorporated.

Kathy Underwood

In 1986, Robert Connors and Andrea Lunsford started to research and document the most commonly found “errors” in the work of college composition students. They came up with their list of “Top 20 Errors,” which informed their work on the Second Edition of the St. Martin’s Handbook, published in 1992. While concern remains that focusing on specific surface errors can distract students from larger issues such as organization and quality of argument, for the weary teacher, it’s a great boon to be able to give students a specific list of taboos.

Similarly, the weary technical editor can opt to provide a list of common errors to an equally weary and beleaguered technical writer. Of course, a corporate “top 20” list is usually a mix of common usage errors as well as departures from company-specific practices. An example would be the use of the serial comma, which is one of the many issues about which usage pundits have no true consensus.

The Department of Technical Editing at SAS uses their own locally derived list that we call “Frequently Marked Editing Issues” (affectionately known as “feemies”) to prompt both writers and editors to recognize all-too-common surface errors. To encourage compliance, the Editing Department provides all writers and editors with a little ring-bound flip book (about 4 ¼ x 5 ½) called the “SAS Style Guide Feemie Flipbook” with each “feemie” printed on one page of card stock. Each card states the usage issue and gives illustrative examples. Where appropriate, further explanation is provided and links are given. On our online SAS Style Guide, these are, of course, live links. (Our style guide, unfortunately, cannot be accessed outside our intranet.) For example, we have a long table with the American English equivalent of certain Indian English and British English terms (we standardize to American English).

Here are a few of our other top “issues”:

To me, one of the most helpful items has to do with preposition choice. It seems like everyday there’s another flavor of dialog box or window. And the usual guidelines about how to decide whether something is “in” or “on” some other thing aren’t always that helpful in cyberspace. On our style guide Web site, we add this statement: “We emulate Microsoft in our choice of prepositions to use in conjunction with components of user interfaces. “In” generally refers to a position within limits, and “on” generally refers to a position in contact with something.”

Another very useful category to include on a “feemie” list is tagging. Should a given item be formatted as <code> or as <codeBlock> or have no tag at all? Particularly, if writers in your company use multiple authoring tools or are transitioning from one tool to another (SGML to XML, for example), everyone will welcome a “frequently confused” list to guide their tagging.

Do you and your colleagues maintain similar lists of “frequently encountered” errors or usage taboos in your workplace? I would be interested to hear about your experiences with this kind of document. Please write to me at kathy.underwood@sas.com.

Recommended Reading

Connors, Robert, and Andrea A. Lunsford. “Frequency of Formal Error in Current College Writing, or Ma and Pa Kettle Do Research.” College Composition and Communication, 39:4 (1988): 395-409.

Lunsford, Andrea A., and Karen J. Lunsford. “’Mistakes Are a Fact of Life’: A National Comparative Study.” College Composition and Communication, 59:4 (2008): 781-806. Available at http://bcs.bedfordstmartins.com/lunsford/PDF/Lunsford_article_Mistakes.pdf.

Williams, Joseph M. “The Phenomenology of Error.” College Composition and Communication, 32:2 (1981): 152-168.

Kathy Underwood

You know the kind of paragraph that seems to resist all your attempts at editing? Its sentences are usually filled with noun stacks (filled with abstract nouns), pronouns without clear antecedents, and passive voice—not to mention multiple independent and dependent clauses. You just want to write “awk” in the margin and move on, right?
The problem, of course, is that “awk” says no more than “This sentence makes me unhappy.” So what can you do to help both yourself and your writer to untangle a big sentence mess?
Over the years, I’ve adapted a technique to create a “diagramless sentence diagram” to help analyze problem sentences. Because sentence diagramming is not widely taught any more, I wanted to provide a visual representation of the sentence structure that could be used by writers who had no diagramming experience.

Let’s try this bit of bureaucratic reportage, which bears all the signs of a first draft:

Following a series of well-attended information sessions concerning the new multi-user features as well as other enhancements, user group members were invited to review and evaluate a project plan for the new users’ Web site to be established during a public forum sponsored by the Steering Team last Friday. However, the Steering Team was disappointed when only 2 of the 50 users were in attendance and because no questions were presented to the expert team assembled for the occasion.

“Awk” doesn’t begin to describe the problems in this paragraph. So let’s try to find something substantive by first identifying actors and actions as represented by the subjects and verbs.

sentence 1

• actor user group members
• action were invited to review and evaluate

sentence 2

• actor the Steering Team
• action was disappointed

At this point, we know that there are some users who’ve lost interest in whatever it is that the Steering Team is doing and that the Steering Team is disappointed. And if they all talk like these sentences are written, we know why. So let’s see what else we can dig out from those sentences.

sentence 1

• actor the Steering Team
• action were invited to review and evaluate
• when “Following a series of . . . sessions . . . concerning . . . features”
• why “to review and evaluate the project plan”

sentence 2

• actor the Steering Team
• action was disappointed
• when “when only 2 . . . were in attendance”
• why “because no questions were presented”

This far into the analysis, we should be seeing something meatier. What are we missing? The subject of the meeting—the actual topic about which there’s a lot of apathy, disappointment, and, one infers, smoldering resentment.

Note that as you identify additional key bits of content, your idea about the meaning of the text may shift significantly. Not infrequently, you’ll find that a writer will have buried a significant detail in a dependent clause or at the end of a prepositional chain. Also note that you might find that the explicit topic might be incidental to the real point of the sentence. It’s highly likely in such cases that you will need to query your writer.

So here’s a quick fix, admittedly done without the writer’s response to your carefully worded query and with a rash number of assumptions:

After the sessions on the new product release, the Steering Team invited user group members to ask questions and offer comments to an expert panel. The panel’s report will serve as the basis for services on the new product Web site.

The little set of labels (actor, action, when, and why) won’t rewrite the sentence for you. But they can at least help you tease out important details that seem trivial, thus enabling you to realize whether the writer might have hidden the real topic in the least significant parts of the structure.

I would like to add that this primitive analytical tool I’m using pales in its utility next to the real thing—sentence diagramming. For those of you who weren’t fortunate to be exposed to diagramming in school (and even if you were), I recommend Kitty Burns Florey’s Sister Bernadette’s Dog Barking: The Quirky History and Lost Art of Diagramming Sentences.

Kathy Underwood

In Act I, Scene 2, of Macbeth, a blood-soaked sergeant enters the presence of King Duncan, which prompts the king to ask the obvious question:

“What bloody man is that? He can report,
As seemeth by his plight, of the revolt
The newest state.”

What is the answer to this kingly question? That’s a usage pundit reporting some change in language usage that seems likely to provoke controversy. And that’s me after one of those interminable style discussions that characterize editors’ meetings. I sometimes—well, often—feel that I’ve just left the battlefield and that I need to find the nearest medic. It’s not that we’re all contentious. (In fact, very often these discussions are more like rowdy, amusing party games.) It’s just that the topics we address are fraught with complexities that make the plot of the Scottish play look like Dick and Jane. Worse yet, we frequently struggle with questions that few if any customers would ever recognize as questions.

Take the definite article. Please. The editors at SAS continue to struggle with the question of which SAS product names require the definite article and which require the zero article (linguist-speak for no article at all).

How do you make such a decision in the absence of clear guidelines from the usual usage manuals that we consult (Chicago Manual of Style, Harper’s Dictionary of Modern American Usage, etc.)? In the interest of precision, we have to make choices not unlike those parodied in the show Beyond the Fringe in the 1960s. In a sketch entitled “Portrait from Memory,” Jonathan Miller as Bertrand Russell meets the philosopher G.E. Moore, who, as it happens, is holding a basket of apples on his knees. Russell asks Moore if there are any apples in a basket. Moore says no. Then Russell asks if there are some apples in the basket. Moore says no. Finally, Russell asks Moore if he has apples in the basket. “’Yes,’ he replied. And from that day forth we remained the very closest of friends.” (Can you imagine what Moore would have been like in editors’ meetings?)

For our in-house style guide, my worthy colleague and fellow editor, Joel Byrd, created an article on the use of the definite article with product names. He reified our department’s lengthy and ongoing collaborative discussions and e-mails and produced a very useful table as well as a field test to help us make the best choice. Here is my paraphrased version of the guidelines:

• Use the definite article if the primary noun is countable (example: “the SAS Forecast Server”).
• Use the definite article if there is general industry agreement that the primary noun requires a definite article (example: “the SAS Intelligence Platform”).
• Use the zero article if the primary noun is uncountable (example: “SAS Business Intelligence”).
• Use the zero article if the primary noun is a metaphor “that does not have a defined, commonly accepted meaning in the software industry” (example: “SAS Management Console”).

The test devised to assist the editor is as follows:

• Write the primary noun in lowercase and then in uppercase (for example, “console” versus “Console” and “the console” versus “the Console”).
• Does the name of the object in isolation mean the same thing as it means in the product name? If so, use the.

Based on several examples of SAS product and component names, I decided to create a list of our extant primary nouns so that we could have an at-a-glance guide for new product names and whether or not they should take the definite article.

• Use the definite article with these primary nouns:
  • platform
  • portal
  • provider
  • server
  • software
• Use the zero article with these primary nouns:
  • analytics
  • console
  • intelligence
  • studio

This list should be satisfyingly clear. But consider the problem with console. Our basic rules say that if the primary noun is uncountable, we should use the zero article. But the word console is obviously countable. On the other hand, in the land of product names and software products, console is a metaphor and not countable. Very crazy-making. So every time I come across a product name with the word console in it, my editing hand wants to insert the definite article.

There’s an even bigger problem. With any usage guideline, there’s a “missing middle” (one or more ungrounded assumptions) in the syllogism. That is, when and how do we decide that a given agreement has reached a critical point that forces us to change our guidelines? Yes, you guessed it: We have to have another editors’ meeting.

Why is it that grammar and usage rules, which once mastered should provide comfort and confidence, can be so frustrating? Why are there so many exceptions to cope with? And why do all those usage pundits have so many different opinions and remarkably little consensus? I think that those questions are answered succinctly by Otto Jespersen in his introduction to The Philosophy of Grammar (found in the first sentence of Chapter 1):

“The essence of language is human activity—activity on the part of one individual to make himself understood by another, and activity on the part of that other to understand what was in the mind of the first.”

Even if we think of the effects on language of just a few of the more dramatic forms of human activity—revolutions, inventions, and catastrophes—we quickly realize that language is a quivering, changeful thing. From vowel shifts, to spelling, to syntax, language always reflects the psyche of any given group. (And which psyche is more shifting than that in the software industry? A new product or technology can completely change perspective.) That’s especially true when society is in flux. It’s that shared, shifting psyche that dictates what really happens with language. And we might argue that the shifting is where the sublime arises.

As editors, we often have to make many decisions that we know will ultimately prove ill-advised, if not completely wrong, and that any number of our fellow editors will find objectionable. But that’s what human activity, war, peace, and the dialogue of the psyche are all about.

Think of the changes from “on line” to “on-line” to “online” that have occurred over the past few decades. There was no legislation, there was no vote, and there was no universal epiphany. So why did “on line” and “on-line” drop out of use (mostly—they are still used in some settings)? In The Fight for English, the linguistic historian David Crystal says that the short answer to questions about disdained usage is “somebody thought it was wrong.” (See Crystal, page 110. Crystal’s book, by the way, is a very readable history of grammar practice and grammarians over the past few hundred years.)

All editors, and especially those of us who serve on in-house style teams, have to make arbitrary decisions all the time. We must be as autocratic as is essential to produce a cohesive body of documentation. But we can never lose sight of the world outside our doors. We have to listen to that bleeding sergeant even if we don’t want to or haven’t got the time to hear what he has to say.