Rumpus Exclusive: “The Arrogance of Style”






1. Never use “comma-then” in literary writing.

When crafting complex sentences, avoid following a comma with the word then introducing a dependent or a relative clause, unless you want New Yorker regular and bestselling writer Jonathan Franzen to accuse you of “not listening to the English language when you’re writing.” Thus do not write,

I read a Jonathan Franzen polemic on the Farrar, Straus and Giroux website, then wrote a comment rebutting one of Franzen’s key claims.

If you believe that you do in fact listen to the English language, perhaps you are simply an unwitting product of MFA culture. According to Franzen, “No native speaker would utter [a comma-then phrase], except in a creative-writing class.” Thus eschew,

Franzen begins his diatribe invoking the worn cliché about “so many ____, so little time” (in this case, it’s “so much to read”), then snides into specifics with “one of the best reasons a writer can give me [to put a book down] is to use the word then as a conjunction without a subject following it.”

2. Except if you follow “comma-then”
with an independent clause(?).

Franzen allows for this seemingly arbitrary exception, explaining that before independent clauses, then functions as an adverb, not as a conjunction. Native speakers, he claims, are okay with this, too. Thus you might continue your narrative with:

My first encounter with the comma-then “rule” came during the copyediting of my second book, during which Jeremy Hall noted the “problem” in the following sentence: “Peter stopped and explained, ‘You collect the spit in the front of your mouth, then you tighten your lips, clench your teeth, and push your tongue forward so the spit squeezes through the space between your front teeth.’” He said “Adverbial in nature, then should not technically operate as a conjunction in this fashion.”

As alluded to above, this particular example includes a subject following then (“then you tighten your lips”); thus Franzen would allow it.

But Jeremy Hall would not, at least initially. Basing his argument on the fact that then is not one of the FANBOYS (a mnemonic to remember all viable coordinating conjunctions: for, and, nor, buy, or yet, so), and that it can be moved around in a sentence without losing meaning (“you then tighten your lips,” “you tighten, then, your lips,” “you tighten your lips then”) while true coordinating conjunctions cannot (go ahead; try it), he deemed my sentence ungrammatical. Along with many others in the book.

3. Well, if you’re by nature more of a descriptive
grammarian than a prescriptivist, then push back.

Perhaps you believe, either passionately or vaguely, that the English language has developed the way it has over long years as a kind of negotiation among millions of speakers from myriad backgrounds and societies and statuses, some of whom are educated in the niceties of grammar, most of whom are not, and that even the prescriptivists are not tapped into the mind of God on the matter. They began with an already evolved language and attempted to exert influence on it, to shape it and standardize it, and they’ve certainly had some success, but the language keeps on making its own sense via unruly speakers and writers who care not a whit whether their conjunctive adverb is a FANBOY or whatever.

To his credit, Jeremy had also noted the “smug/arrogant tone” of the Franzen diatribe, which he had consulted independently, and when pressed on the issue, he agreed to a compromise, with me stetting about half of my comma-thens, “not killing [my] darlings at the expense of [my] voice, but perhaps keeping them in check with measured usage,” as he said.

That example doesn’t even include a comma-then or any other “clever” dig at the “rules.” So are we done here?

4. For the sake of your mental health, just let it go.

Otherwise, you might end up spending an inordinate amount of time researching such a minuscule, insignificant, easily resolved issue, bothering your more-grammatical friends, asking of them time and insights far beyond the merits of your one-way feud with a faraway famous writer whose tone has rankled you. Thus please do not report:

I asked my old editor boss, Andrew Olsen, what he thought. He replied that he often used comma-then, “which tells how I feel about it,” and he “asked a colleague, Jake Frandsen, to see what the Corpus of Contemporary American English (COCA) shows about the construction. He found that [, then] had 68,595 hits (mostly in fiction), and [, and then] had only 15, 446 hits. So the construction that Jonathan Franzen objects to is preferred 82% of the time in that corpus.”

My colleague Brian Jackson astutely pointed out that Franze’s okay example “I sang a couple of songs, then Katie got up and sang a few herself,” would be Franzen’s own argument by a comma splice (given that both phrases include subject and verb, and then is not a coordinating conjunction), yet Franzen allows it for “propulsive effect”?).

My recently retired colleague Kristine Hansen went above and beyond in her response to my query, researching widely but primarily in the Longman Grammar of Spoken and Written English (London: Pearson, 1999) and writing me a two-thousand-word letter full of detailed examples and explanations. Based on Biber et al.’s analysis of a forty-million-word corpus of American and British texts, Hansen discovered that “some of the adverbial uses of then are very much like the uses of conjunctions.” She points out that then is the most common time adverbial, which we already kind of know instinctively, and we can understand comma-then in this sense/function, but then is also a “linking adverbial,” which, according to the Longman Grammar, is used to “make semantic connections between spans of discourse of varying length,” those connections including “enumeration and addition, summation, apposition, result/inference, contrast/concession, and transition.” Any number of these functions could be seen as justification for a comma-then clause. While linking adverbials differ from coordinating conjunctions in their flexible positioning (as demonstrated above: they can be moved without losing meaning), they nevertheless can perform in the position where one might usually supply an and, and in fact they “don’t have to be used with a conjunction.” Hansen concludes that “As it conveys a sequence of events, then not only expresses a meaning of time, but also serves a cohesive function.” Frankly, the more I learn about then, the more impressed I get.

This all seems a little technical for descriptivist grammar, but thank you for bothering your colleagues in the pursuit of truth. I believe we can all rest easy now, knowing that the question is settled once and for all. As your colleague Kristine Hansen said, “If there is one true statement that can be made about language, it is that it is always changing. What we consider correct is purely a matter of convention and sometimes a matter of prejudice.”

5. So… perhaps we might say: Don’t make
absolute claims about other writers’ practices
without even bothering to check if you’re right.

For instance, if you happen to respect Charles Dickens and the Brontë sisters, and/or think that they represent a bygone era of quality writing, and/or think that your readers will recognize your name-check and accept your premise without question based on the ethos of these canonical novelists, and/or think who’s got time to reread all those long novels anyway, please remember that all of these texts are in the public domain and are easily searchable. Thus beware:

I read Franzen’s claim that “Dickens and the Brontës got along fine without comma-then,” then went straight to Project Gutenberg to find out, then wrote a comment on the FSG site, saying “Just for fun, I just took about 5 minutes to do the most basic of searches on to see if Dickens and the Brontës ever employed comma-then in just the way Franzen forbids. Oh yes they did. Quite a lot. [‘Uriah kept it up a little while, then sent it back to Mrs. Heep.’ / ‘I listened doubtingly an instant; detected the disturber, then turned and dozed.’]”

6. Fascinating.

So Franzen tried to bolster his argument by claiming that Dickens and the Brontës were on his side, but they’re actually not?


And it’s untrue that comma-then is “an irritating, lazy mannerism… that occurs almost exclusively in ‘literary’ writing of the past few decades”?


7. What happened next?

Did Franzen respond? Anyone from FSG?

My comment got dozens of upvotes or thumbs up or likes or whatever, and several other commenters similarly argued convincingly against the comma-then prohibition, then FSG removed all comments from their website.

Again, that’s one of those Franzen-approved “propulsive effect” sentences, no? Do you think the decision to remove commenting from the site was a direct result of your comment?

Who can say?

8. Well, we certainly can say go ahead and
use comma-then in literary writing.

Dickens and the Brontës did it; so can you! Do you have more examples?

Do I!? I may have to add an appendix to hold them, and I’m not even convinced that I’ve gotten them all (though I’ve been quite thorough). It’s worth admitting, for the sake of fairness and full disclosure, that Anne Brontë alone among our four forebears seems to agree with Franzen. She never wrote comma-then in The Tenant of Wildfell Hall or Agnes Gray, though she used comma-and-then over a hundred times.

And yet?

And yet…

The Pickwick Papers (1836):
“Like a gas-lamp in the street, with the wind in the pipe, he had exhibited for a moment an unnatural brilliancy, then sank so low as to be scarcely discernible; after a short interval, he had burst out again, to enlighten for a moment; then flickered with an uncertain, staggering sort of light, and then gone out altogether.”

Oliver Twist (1837):
“Oliver involuntarily shrunk back, and then laughed at himself for being so foolish, then cried, then laughed again…”

Nicholas Nickleby (1838):
“His hopeful friend and pupil drew a chair to the breakfast-table, and essayed to eat; but, finding that impossible, lounged to the window, then loitered up and down the room with his hand to his fevered head…”

The Old Curiosity Shop (1840):
“This done, she begged them in a kind of deep despair to drink; then laughed, then cried, then took a little sip herself, then laughed and cried again, and took a little more.”

Jane Eyre (1847):
“I took that dear hand, held it a moment to my lips, then let it pass round my shoulder.”

Wuthering Heights (1848):
“He laid them on the table, looked eagerly towards the window, then rose and went out.”

David Copperfield (1849):
“Dora left off shaking her curls, and laid her trembling little hand upon my shoulder, and first looked scared and anxious, then began to cry.”

Bleak House (1852):
“Jo searches the floor for some time longer, then looks up for a moment, and then down again.”

Great Expectations (1860):
“Joe looked at her in a helpless way, then took a helpless bite…”

…and those are just one example each from only about half the books.

Okay! Okay! Uncle! I’m convinced. And grateful for the lesson. But we seem to have strayed from our original purpose to instruct and edify writers. How about we end with an old, inoffensive classic? Such as…

9. Omit needless words.

Gotcha. Like the whole of Jonathan Franzen’s fulmination? And maybe this entire rejoinder?



Rumpus original art by Max Winter.


Excerpted from Disparates by Patrick Madden. Copyright © 2020 by Patrick Madden. Reprinted by permission, courtesy of University of Nebraska Press.

Patrick Madden is a professor at Brigham Young University. He is the author of the award-winning Sublime Physick: Essays (Nebraska, 2016) and Quotidiana: Essays (Nebraska, 2010), and coeditor, with David Lazar, of After Montaigne: Contemporary Essayists Cover the Essays. His essays have appeared in a variety of periodicals as well as in The Best Creative Nonfiction and The Best American Spiritual Writing anthologies. Visit Madden’s website at More from this author →