|Oregon State Bar Bulletin JULY 2010
An earlier column (April 2007, “Word Choices: Pairs that Puzzle”) blithely gave advice on the difference between which and that. While the advice was sound, it merely scratched the surface of a raging debate. On one side of the debate are the curmudgeons, focused on precision. On the other side are those willing to sacrifice the clarity of written English on the altar of common — and often unclear — spoken English. I call those debaters the “conversationalists” because their grammar is based on usage in casual conversations.
As in most debates, authorities are lined up on both sides, though this curmudgeon has a clear favorite.
The raging debate is about how to tell readers which information in a sentence is necessary to its meaning and which information isn’t. We call the necessary information “restrictive,” so obviously the unnecessary information is “nonrestrictive.” The following examples demonstrate the difference:
The Smith files that were on the desk were stolen.
The Smith files, which were on the desk, were stolen.
In the first example, there were many files related to the Smith matter. Some were in file cabinets, some were in the copy room and some were on the desk. Only those on the desk were stolen.
In the second example, all of the Smith files were stolen. The files all happened to be on the desk just before they disappeared, but the clause “which were on the desk” isn’t needed to clarify the sentence. That information could have been omitted and the reader would still have known that all of the files related to the Smith matter were gone: The Smith files were stolen.
The Curmudgeon’s View
Two markers in the previous examples provide clarity by showing which information is restrictive and which information is not. Restrictive information is introduced by the word that and is not set off with commas. In contrast, nonrestrictive information is introduced by the word which and is set off with commas. The curmudgeon seeks clarity in the written word, and these two markers provide it.
The Conversationalist’s View
The conversationalist is guided by convenience, even at the expense of clarity. Because many of us would tell the story of the missing files using which rather than that, the conversationalist would find the following sentence acceptable.
The Smith files which were on the desk were stolen.
In the conversationalist’s defense, the absence of commas indicates that the clause “which were on the desk” is restrictive. I must confess that some well respected dictionaries and style guides agree with the conversationalists that which and that can both be used for restrictive clauses. But why sacrifice the absolute clarity that exists when that is restrictive and which is not? And how many current writers are so fastidious with their use of commas that you’re willing to bet your client’s goals on those little marks?
Fortunately, the curmudgeons and conversationalists agree on the following details.
The Redundant “That”
When a sentence gets long, the reader sometimes repeats that at the beginning and the end of the intervening clause. Just one that is sufficient in the example below:
The defendant argued thatif he had understood the charges against him that he would never have pleaded guilty.
The Necessary “That”
Sometimes that is not necessary, even though it’s not redundant. In the earlier example on stolen files, nothing would be lost if that and its verb were deleted. In both of the following sentences, the reader knows that just the Smith files on the desk disappeared, so the restriction remains.
The Smith files that were on the desk were stolen.
The Smith files on the desk were stolen.
But sometimes that avoids ambiguity. Consider this example:
The mayor later recognized those who left the council meeting dissatisfied (I start wondering where the mayor was when he recognized them — maybe at the gym? a restaurant?) might not vote for him in the next election.
Including the word that tells the reader not to focus on those who left but on their vote.
The mayor later recognized that those who left the council meeting dissatisfied might not vote for him in the next election.
Here’s another example:
The record contains substantial evidence for the judge’s finding the parties to the lawsuit (where did he find them?) were jointly and severally liable.
This sentence, too, leads the reader astray. Initially, I wondered whether the judge found “the parties” together at a pub after hours. Maybe my imagination is too active, but a simple that would have kept me focused on the judge’s legal finding.
The record contains substantial evidence for the judge’s finding that the parties to the lawsuit were jointly and severally liable.
Here’s a final example of ambiguity resulting from a missing that.
This action resulted in the type of prejudice of the decision maker prohibitions on ex parte contact are designed to avoid.
I’m left wondering what the “decision maker prohibitions” are, if in fact that is a term of art. Would a hyphen make “decision-maker prohibitions” clearer? Or perhaps a that is missing, which would separate the “decision maker” from the “prohibitions.”
This action resulted in the type of prejudice of the decision maker that prohibitions on ex parte contact are designed to avoid.
The Distant “That”
Ambiguity also can result when the pronoun that or which wanders too far from its antecedent. (Remember that the antecedent is the noun that the pronoun is replacing.) Some curmudgeons call this problem “remote relatives” because the relative pronoun is so remote from its antecedent. Here’s an example with a wayward that:
The two firms have different approaches to awarding bonuses thatconsider not only the hours billed but also the efficiency of the associates performing the work.
Technically, that refers to bonuses because that’s the noun appearing just before the pronoun. But the clause about hours and efficiency surely refers to the different approaches, not the bonuses that are awarded. One solution is to repeat the noun:
The two firms have different approaches to awarding bonuses; those approaches consider not only the hours billed but the efficiency of the associates performing the work.
Here’s another example, where the which has wandered too far afield.
In complex litigation, parties may have multiple claims against each other as well as against the state, which could be heard in either state or federal court.
The clause “which could be heard in either state or federal court” certainly refers to the claims, not to the state. Again, reworking the sentence promotes clarity. In the first revision below, the clause is moved closer to the claims. In the second revision, the noun is repeated.
In complex litigation, parties may have multiple claims, which could be heard in either state or federal court. These claims may be between the parties as well as against the state.
In complex litigation, parties may have multiple claims against each other as well as against the state. Some of those claims could be heard in either state or federal court.
The Bottom Line
The debaters agree on a few basic rules. I have summarized here the rules on which all my sources agree.
If you ever feel confused or queasy by the ongoing debate, refer back to this simple list.
The Chicago Manual of Style: The Essential Guide for Writers, Editors, and Publishers (15th ed. 2003).
Bryan A. Garner, A Dictionary of Modern Legal Usage (2d ed. 1995).
Merriam-Webster’s Collegiate Dictionary (11th ed. 2007).
The New Oxford American Dictionary (2d ed. 2005).
William Strunk, Jr., & E. B. White, The Elements of Style (4th ed. 1999).
ABOUT THE AUTHOR
Suzanne E. Rowe is an associate professor at the University of Oregon School of Law, where she directs the Legal Research and Writing Program. As the Luvaas Faculty Fellow for 2008-2010, she is grateful to the Luvaas Faculty Fellowship Endowment Fund for support of her articles in The Legal Writer. She appreciates the comments of Melissa Seifer Briggs on this article.
An archive of The Legal Writer articles is available here.
© 2010 Suzanne E. Rowe