The Technical Stylist: All That and Which Too

By Kathy Underwood

Do you think that the following phrase—from Roosevelt’s speech preceding the declaration of war in 1941—has a grammatical error?

a date which will live in infamy

This interesting and not at all simple question recently came up on the Word Lovers social network at my company (SAS Institute Inc.) The answer is “no—sort of” and “it depends.”

The grammatical content: In the famous phrase, the noun “date” is followed by the relative pronoun “which” and a predicate (“will live in infamy”).

The rhetorical occasion: an address to Congress and a radio broadcast by the President of the United States preceding a declaration of war. That is, the phrase is from an urgent oral communication, not a written one.

In the set of writing conventions that we call Standard Edited American English (AE) usage, we look for two signals to indicate clause restriction. Signal 1 is the presence or absence of a comma. Commas are used with nonrestrictive (that is, nonessential) clauses and not used with restrictive (essential) clauses. Signal 2 is the choice of relative pronoun. In Standard Edited AE, restrictive clauses use “that” and no comma. Nonrestrictive clauses use “which” and a comma.

Although most contemporary American English usage experts recommend using “that” (without a comma) to signal a restrictive clause, grammatically, there’s nothing wrong with “which.” It’s a relative pronoun that functions as a relative pronoun in the same way as its cousin “that.” “Which” is commonly used with restrictive clauses in British English. In terms of grammar, you cannot object to the use of “which.” In terms of usage in Standard Edited America English (AE), you can object to “which” because “that” is the preferred signal for a restrictive clause. (See for example, Copperud’s American Usage and Style: The Consensus and Garner’s Modern American Usage (Garner’s MAU.)

I focus on the distinction because I think it’s the watershed difference between style and usage now and style and usage fifty years ago. In the past, usage experts were assumed to be god-like beings whose birthright was to valorize usage. In that view, we tended to accept experts’ valorization schemes (style and usage guides) without pausing to analyze the linguistic validity. In other words, we tended to act more like proofreaders than like editors. Hence, the superstitions around the so-called “split infinitive” and other shibboleths so well documented in Miss Thistlebottom’s Hobgoblins (and in Garner’s MAU in the entry “Superstitions”).

I do think that usage experts are essential to our work as editors. I am, in fact, addicted to reading and collecting usage guides. But I become concerned when usage recommendations are perceived as elemental laws of nature and are assumed to govern any form of discourse. The reality is that usage guides are predominantly for written discourse—not oral discourse. (They don’t yet truly cover that notional territory in between that includes online, social network discourse.)

In our house style guide at SAS, we proscribe the use of “which” with restrictive clauses because (a) we try to follow the current usage-expert consensus on written discourse, (b) we’re in a technical environment where consistency is critical, (c) our work goes to translators, and (d) a large portion of our readership consists of nonnative speakers of English.

Garner’s Modern American Usage, as one would expect, has an extensive and illuminating article on restrictive clauses. (See entry A under the article on “that.”) Garner does look disparagingly at our British cousins’ use of “which” with restrictive clauses: “British writers have utterly bollixed the distinction between restrictive and nonrestrictive relative pronouns.” Garner, brilliant usage expert though he is, is also an attorney and something of a law-and-order guy when it comes to usage.

As to Roosevelt’s speech, because there’s no intervening comma between “date” and “which,” the clause (“which will live . . .”) is still being used restrictively. Of course, that’s assuming that the script for his speech doesn’t have an error. But given his rhetorical purpose (not to mention the speed with which the speech was written), the absence of the comma seems apt.

To access an image of the typewritten (and hand-edited) manuscript and the audio files of Roosevelt’s speech, see


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s