Q. “He wanted to see his friend John Smith before going out”? Or “He wanted to see his friend, John Smith, before going out”? I know the rule for spouses and siblings (as in CMOS 6.31), but what about friends?

A. Because people usually have more than one friend, the first version, the one without commas, will almost always be correct. Even those who might claim to have just one friend could make another one at any time. Besides, friend is an elastic concept that can apply to people who aren’t exactly “friends.”

All this to say that, without additional context, the word friend by itself isn’t enough to tell us who you’re talking about in an example like yours (Which friend?), so the name (John Smith) is essential (or restrictive) and shouldn’t be set off by commas.

By contrast, people generally have only one spouse at a time, making commas the default when a name follows a word like husband or wife: “She wanted to see her husband, John Smith, before going out.” In that sentence, the word husband alone narrows things down to one person; the name gives extra information that isn’t essential (it’s nonrestrictive) and is thus set off by commas. (They’re called spousal commas and not friendship commas for a reason.)

For more on commas like these (and why even spousal commas can be omitted sometimes), see “Your Dog[,] Smurf: Understanding Commas with Appositives,” at CMOS Shop Talk.

Q. Do we refer to fictional characters by their first or last name?

A. Either, neither, or both. In fiction, people are known by the names they’re called, typically by the narrator but sometimes by others (as in dialogue). In other words, whether you’re referring to Huck Finn or Jo March or Edward Casaubon or Heathcliff or Juliet or Romeo or Elinor Dashwood or Don Quixote or Miss Marple (or Miss Jane Marple if you prefer, at least on first mention), you can usually take your cue from the novel or story or play or other such work that features them.

You can do the same for any mentions after the first, as in Huck or Jo or Casaubon or Heathcliff or Juliet or Romeo or Elinor (not to be confused with Mrs. Dashwood) or Don Quixote or Miss Marple. When more than one form could be used, shorter may be preferred over longer, as with Jo (who is rarely “Josephine” in Little Women) and Huck (as he’s usually called in Huckleberry Finn even when the last name is used, in spite of the title) and Casaubon (even though, in Middlemarch, he’s called “Mr. Casaubon” five times as often as “Casaubon”).

Speaking of George Eliot’s quixotic, tragicomic scholar (Casaubon, that is), you could likewise shorten “Don Quixote” to “Quixote,” even though that more famous character is rarely mentioned without his honorific in the four-hundred-year-old novel that brought him to life.

Though all the characters mentioned above are from older classics, the principle for newer works is the same—as it is when you’re writing the characters yourself—though there will always be some exceptions.

Q. Correct to hyphenate “four-hundredth birthday”? Seems right, but I can’t find a source. Thank you!

A. Though the hyphen doesn’t do any harm in your example, you can normally leave it out. See the hyphenation guide at CMOS 7.96, section 1, under “numbers, spelled out,” which says this (the rule is stated on the right and the examples are on the left):

The last example—“five hundred fifty-second contestant” (some writers would add “and” before “fifty-second”)—is analogous to “four hundredth contestant” (or “four hundredth birthday”); “fifty-second” is hyphenated in the example in CMOS because “fifty-two” is normally hyphenated.

Note that some of the numbers above would normally be written with numerals (e.g., 552nd). And though four hundred is usually spelled out according to Chicago’s general rule for numbers (which says to spell out the whole numbers zero through one hundred and extends to multiples of a hundred), you’d usually write 400th if you were following Chicago’s alternative rule (see CMOS 9.2, 9.3, and 9.4).

It should also be noted that a number like four hundred is hyphenated sometimes—for example, when it forms part of a compound modifier (“four-hundred-acre farm”); see section 1 of the hyphenation guide, under “number + noun.”

Q. I haven’t been able to find anything in CMOS about how to punctuate a sentence that begins with a dependent clause that consists of two independent clauses introduced by a word like “If.” I’m a proofreader for court reporters. One of my clients consistently puts a comma within this type of dependent clause, but I don’t think the comma should be there. The comma I’m asking about is the comma after the word “valve” in the following example: “If you had to replace a leaking valve, and it had external insulation, how long would that work take?” To me, the part about external insulation is restrictive and should not be preceded by a comma. I’d like to find some authoritative confirmation of that. Thank you.

A. We agree that the first comma in your example is best omitted, not because the clause following and is restrictive (and doesn’t normally work that way) but because that comma competes with the one that is normally required after an if clause. So:

If you had to replace a leaking valve, how long would that work take? (The comma is required; see CMOS 6.26.)

If you had to replace a leaking valve and it had external insulation, how long would that work take? (A comma after “valve” would be strictly correct but is usually best omitted.)

To make a long story short, omitting the comma after valve from your example makes it clearer that the introductory If applies to both clauses (i.e., “you had . . .” and “it had . . .”). For a similar take, see this previous Q&A (which refers to the 17th ed. but still applies).

To understand how all this works, it might also help to consider your example sentence with that instead of and (though in your situation any changes to wording presumably wouldn’t be an option):

If you had to replace a leaking valve that had external insulation, how long would that work take?

In that case, that introduces a restrictive relative clause (see CMOS 6.29). And though it doesn’t change the meaning of your example, the word that can help ensure a restrictive reading of the “insulation” clause.

We’ll try to clarify these options in a future edition of CMOS.

Q. Do you recommend capitalizing offices such as the Office of the Attorney General and the Office of the City Clerk? These are formal names for administrative bodies, but they are shared names. In other words, every city has an Office of the City Clerk.

A. You can usually use initial caps for the name of a specific office but lowercase when referring to such an office as a generic term or concept. For example, we’d advise lowercase for your last sentence, where the use is clearly generic: “Every city has an office of the city clerk.”

The main reason for lowercase in that last sentence is that “Office of the City Clerk” may not reflect the wording used by each such office. Here are a few more examples:

Illinois’s Office of the Attorney General was established in 1818.

The Office of the City Clerk for Chicago can be found at City Hall.

Portugal’s Ministry of Finance (Ministério das Finanças) is headquartered in Lisbon.

but

In many countries, the treasury department is known as the ministry of finance.

Most states have an office of the attorney general.

The city clerk’s office is on the first floor.

See also CMOS 8.63–71.

Q. How would you cite the publication date of a republished journal article? For example, a piece that was published in 1967 but can only (easily) be found as a republication in the fiftieth anniversary of the journal it was originally published in. Do you cite the newer date, the original date, or both? I’m procrastinating on writing a final paper by scrolling the Q&A, but I’m not having any luck finding the answer!

A. Normally, you should cite the version you consulted or the best version if you consulted more than one. But Chicago-style source citations can easily accommodate extra details, so if you think it’s important to acknowledge more than one version, you can do that.

To take an example that’s similar to what you’re describing, you could cite the following article as it appeared in a fiftieth-anniversary special issue of the Journal of Literary Semantics and then add info about the original publication (in this case, after the DOI):

Dry, Helen Aristar. “The Movement of Narrative Time.” In “50th Anniversary Issue,” edited by Marina Lambrou, Journal of Literary Semantics 51, no. s1 (2022): 19–53. https://doi.org/10.1515/jls-2021-2043. Facsimile of the article as originally published in Journal of Literary Semantics 12, no. 2 (1983), with the same page numbers.

The wording related to the extra info will depend on the source. In a bibliography or reference list entry, this info follows the period at the end of the main citation; in a note, it can follow a semicolon or start a new sentence (or, if the information is brief, it can go in parentheses).

In an author-date reference list, you could cite both publication dates:

Dry, Helen Aristar. (1983) 2022. “The Movement of Narrative Time.” In “50th Anniversary Issue,” edited by Marina Lambrou, Journal of Literary Semantics 51 (s1): 19–53. https://doi.org/10.1515/jls-2021-2043. Facsimile of the article as originally published in Journal of Literary Semantics 12, no. 2 (1983), with the same page numbers.

In the text, the author-date citation would be “(Dry [1983] 2022).” For more details and examples, see CMOS 14.16 (on reprint editions) and 14.77 (on journal special issues).

Q. Both CMOS 10.27 and Merriam-Webster style “post-traumatic stress disorder” that way, with a hyphen, but it occurs to me that since PTSD takes hold in the wake of traumatic stress, the prefix post- applies to “traumatic stress.” So shouldn’t the spelled-out term be styled “post–traumatic stress disorder,” with an en dash?

A. You’re right: It would make sense to spell post-traumatic stress disorder with an en dash. In fact, we could have used your editorial wisdom back in 2003, when we added posttraumatic to the fifteenth edition of CMOS to show that the prefix post- forms one-word compounds even if the result is two consecutive t’s.

That decision would have reflected the spelling posttraumatic in a list of 125 compounds formed with post- under the entry for that prefix in the then-current tenth edition of Merriam-Webster’s Collegiate Dictionary.

Still, we must have failed to register the hyphen in Merriam-Webster’s separate entry for post-traumatic stress disorder, which followed entries for posttranscriptional, posttransfusion, and posttranslational—all without hyphens. It’s clear in hindsight that, as you suggest, post-traumatic was hyphenated in that term (and not in the list of words formed with post-) for a reason, one that had nothing to do with consecutive t’s. (The latest iteration of Merriam-Webster, as of November 4, 2025, now hyphenates post-traumatic as a standalone term also.)

But the dictionary doesn’t do en dashes,* and in post-traumatic stress disorder, no one else seems to either (not even Wikipedia, which tends to feature a lot of en dashes compared with other publications). Instead, we’d advise saving your en dashes for more established uses (see CMOS 6.82–90—especially 6.86, on en dashes with compound adjectives).

__________

* The hyphenated main entries in printed editions of Merriam-Webster do feature en dashes rather than hyphens, but that’s presumably for the sake of legibility in very small type and therefore doesn’t count (see also this related Q&A).

Q. What’s the best way to format a euphemized profanity like “the f-word”? Should the initial letter of the profanity (in this case “f”) be italicized? Is a hyphen appropriate? Thank you!

A. Italics, maybe; hyphen, yes. We’d start with the entry for the term in Merriam-Webster, where it’s “the f-word” or “the F-word”; the two forms are equal variants, but we usually go with the first-listed term.

Then, considering that dictionaries don’t do italics in their entry words, you could apply Chicago style for letters used as letters and italicize the f—as in f-word. But most editors would probably leave that f alone.

See, for example, The F-Word, edited by Jesse Sheidlower (4th ed., Oxford University Press, 2024), which has “the F-word,” capital F, no hyphen—except on the cover, where the title is in all caps and the actual f-word is spelled out but obscured (except for the F) by what looks like a black Sharpie. On the title page, it’s “The F-Word” (capital F and W), which is in keeping with Chicago’s latest advice on capitalizing hyphenated terms in titles (see CMOS 8.162).

For more advice on handling expletives, see CMOS 6.99. For the grammatical term expletive, see 5.246.

Q. “Neither I/me nor my dog responded.” Should it be “me” or “I”?

A. The correct form of pronoun in sentences like yours can usually be figured out by considering the pronoun separately from the noun that it’s paired with. In your sentence, that process would look like this:

I/me didn’t respond.

plus

My dog didn’t respond.

equals

Neither I/me nor my dog responded.

If your goal is to use Standard English, the correct form would be “I didn’t respond,” which answers your question: “Neither I nor my dog responded.” See also CMOS 5.44; for more on Standard English, see 5.4.

Q. Good morning! I have a question about CMOS 8.183 (18th ed.). It says that a very long poetic work is usually italicized but that shorter poems are set in roman and enclosed in quotation marks. What is the cutoff point that distinguishes a longish short poem from a truly long poem? At what point do you stop putting poems in quotation marks and start putting them in italics? Thank you for your wonderful Q&A section!

Q. Could you clarify which texts—specifically sacred and ancient ones—should be written in roman type rather than italics? Is there a cutoff date for something to be considered “ancient”? Does it matter if an ancient text is sacred, literary, or philosophical? Is there some authority that rules whether a text is considered sacred? For example, some consider the Analects as sacred, but that title is widely italicized. Please advise.

A. Some of the advice in CMOS is there mainly to acknowledge how the rest of the world does things rather than to set a rule. Paragraphs 8.183 and (for the second question) 8.104 are in this category. Together, they amount to something like this: Note: Some titles may be treated differently from what we advise elsewhere. Here are a few examples.

This advice is supposed to warn people that one doesn’t generally refer, for example, to The Bible or Dante’s “Inferno” or Shakespeare’s Sonnets—stylings that might seem right based on the advice elsewhere in CMOS. Instead, that would be the Bible and Dante’s Inferno and Shakespeare’s Sonnets—not because we say so but because that’s the convention. And there’s no magic cutoff point. If the Analects of Confucius are normally italicized in your readings, then use italics.

The same thing goes for poems. Is it “The Rime of the Ancient Mariner” or The Rime of the Ancient Mariner? That one could go either way, but if you’re an editor, you know what to do: Choose the form that makes the most sense to you and be consistent about it.

Q. When writing out a citation on paper, should you underline instead of italicize?

A. Though you could try to approximate the look of italics with your handwriting, the convention for handwritten text is the same as it was in the days of the typewriter: underlining means italics. (The usual way to add underlining on a typewriter was to backspace over the letters to be underlined and then repeatedly press the underscore key, or _, the same key that’s typically paired with a hyphen on English-language QWERTY keyboards designed for use with desktops and laptops.)

Q. Hi! One of my authors wants to quote content from an app. It’s like a devotional or thought-of-the-day app, and the content is not available elsewhere online. What’s the best way to cite the material? Thank you!

A. No formal citation is needed. Instead, just make sure that the name of the app and the entity responsible for it (whether that’s a person or a company) have been clearly and accurately identified in the text as the source for the quotation.

Formal citations (as in a note or in a bibliography or reference list) can usually be limited to sources that readers could consult for themselves if they wanted to. Things that others wouldn’t be able to track down online or anywhere else (not even through a library) can usually be described in the text rather than cited in a footnote or the like.

Q. How would you format a credit line for a screenshot? Do you credit the person who took the screenshot or the website where the screenshot is from? For example, if a writer takes a screenshot of a notice on a website, do you cite the website, the writer, or both?

A. Unlike a photographer, the person who creates a screenshot doesn’t usually get credit. But if someone has added annotations or the like that aren’t part of the captured image, that fact should be made clear. For example, the caption and credit for this image—

—might read as follows:

The updated hyphenation guide includes linked cross-references to other terms and sections in the table. Reproduced by permission from The Chicago Manual of Style Online, 18th ed. (University of Chicago Press, 2024), 7.96; arrows added by the author.

See also CMOS 3.21–29 (for captions) and 3.30–38 (for credits).