The Oxford Comma and the Internet (2013) | Hacker News Reader

The Oxford Comma and the Internet (2013)

blog.anguscroll.com

23 pointsbookofjoe55 comments

> Throw in the Oxford comma and George has become a policeman: “Through the window she saw George, a policeman, and several onlookers”.

The problem here might also be that of artificially limiting the solution to commas. We have alternative or better punctuation available to demarcate the interrupting thought. One could emphasize the relationship better either with an em dash or with parentheses:

“Through the window she saw George — a policeman — and several onlookers.”

“Through the window she saw George (a policeman) and several onlookers.”

In text either of those could better express the different emphasis which verbally would be conveyed by a shift in the speaker's tone.

Since we’re being pedantic, canonically you aren’t supposed to separate an em dash with spaces. I assume that’s because typographically it already sets the parenthetical somewhat farther apart than a regular space would.

“Through the window she saw George—a policeman—and several onlookers.”

> you aren’t supposed to

In the world of grammar, these words are always a shot at your own toes. I'm missing a few myself and I learned: always specify a domain or a source.

The AP Style Manual, the bible of grammer for journalism (at least in the US), says to put spaces around the dash. However,

https://news.ycombinator.com/item?id=39436513

No, there should never be spaces around Em-dashes, while for En-dashes they are situational---spaced if being used in lieu of Em-dashes, and unspaced if used as ranges (e.g. 1938--1945) or comparisons (e.g. Blue--Red Dichotomy)

What basis do you have for saying that, and for refuting the AP Style Manual? Are you claiming it says otherwise?

You might be unaware that there are three types of dashes. For grammatical connections, you choose only one of the following and stick with it

- Em dash: unspaced

- En dash: spaced

For relations, it is always unspaced En dash

Do you see what I'm writing? Does it seem I'm unaware? There are even more such marks.

Still no source for your claim. Why do you think these things are true, and that always flawed claim, universally true?

The problem isn't the Oxford comma. The problem is lazy--or possibly rapid, if one is generous--writing and editing (if any). If the use or absence of a comma causes ambiguity, it's probably best to rewrite the sentence.

I believe it was Orwell who wrote a friend: "I am sorry this letter is so long. I didn't have time to make it shorter."

Blaise Pascal, several centuries earlier.

Well, if we are going to be authoritative then we might like to ... well:

"I believe it was Orwell who wrote [to] a friend: "

... soz, you are probably left pondian and have a habit of writing your friends. I like to get all transitive and write to mine.

> you are probably left pondian and have a habit of writing your friends. I like to get all transitive and write to mine.

"Writing your friends" is transitive; "writing to them" is intransitive.

> "Writing your friends" is transitive; "writing to them" is intransitive.

Does the infinitive [edit: ^H^H^H^H^H^H^H^H^H] made it intransitive? Also, 'writing' is a noun (a gerund), not a verb.

You packed a lot of confusion in there.

1. There is no infinitive in either phrase, though it is true that "writing" is a non-finite form. As you note, it is a gerund.

2. While you are correct to say that a gerund is a noun, you're wrong to say that it isn't a verb. You can clearly see that it retains the complements of the verb. A gerund is a verb cast into the syntactic form of a noun just as a participle is a verb cast into the syntactic form of an adjective.

3. A transitive verb is one that takes a direct object. An intransitive verb is any other verb.

4. "Them" is a direct object of "write" in "writing them", but an indirect object, marked by the preposition to, in "writing to them".

5. Just to be really clear, it is only possible to supply a direct object to the gerund "writing" because it is a verb that (in British usage) takes a direct object. There is no such concept as the object of a noun.

> 1. There is no infinitive in either phrase

you're right. how embarassing.

> 2. While you are correct to say that a gerund is a noun, you're wrong to say that it isn't a verb. You can clearly see that it retains the complements of the verb. A gerund is a verb cast into the syntactic form of a noun just as a participle is a verb cast into the syntactic form of an adjective.

Conceded also. This is going terribly. Have I been hustled? Are you a ringer?

> 3. A transitive verb is one that takes a direct object. An intransitive verb is any other verb.

> 4. "Them" is a direct object of "write" in "writing them", but an indirect object, marked by the preposition to, in "writing to them".

Sure, though I didn't say otherwise. I admit I wouldn't have said #3 part B so absolutely; I wonder if there are edge cases.

(Was the parent written by GPT?)

> I admit I wouldn't have said #3 part B so absolutely; I wonder if there are edge cases.

Sort of. There isn't a semantic difference between direct and indirect objects (there are some tendencies). Latin verbs with a prefix very commonly take indirect objects where you would expect a direct one. It's pretty normal (for us) to just think of that situation as the verb taking its object in a weird case; I have no idea how the Romans thought about it.†

As a matter of technical terminology, those are just the definitions of transitive and intransitive verbs. How you apply the concept of "direct object" may vary from language to language. In English a direct object is one that is an otherwise unmarked object of a verb (assuming we're happy with our ability to identify those) and an indirect object is one marked by a preposition. In Latin a direct object is one that appears in the accusative case, and an indirect object is one in the dative case. In Mandarin Chinese direct objects are unmarked and most typically appear after the verb, while indirect objects are marked with a preposition and appear before the verb... but if you want to rearrange the elements of a sentence, there is a preposition of this kind that serves no other purpose than to allow a direct object to be positioned before the verb.

† This is a hazier situation than occurs with deponent verbs, where the evidence is very strong that the verbs "really are" active in Latin despite requiring grammatically passive forms.

Thanks. If you like, share how you know grammar and syntax so well (I was joking about the GPT and of course nobody is required to disclose anything). I know a good bit; I am unused to being outgunned; and I'd like to learn more.

I have some undergraduate training in linguistics, though I wouldn't attribute much grammar knowledge to that. It does come in when you see me complaining that there's no semantic distinction between direct and indirect objects.

I've always been interested in language, so I am likely to pick things up and remember them.

You get fun facts about Latin and Mandarin because those are the languages I have studied. If you want to have examples available of how different languages work and how different categories might or might not be realized in different languages, there's not really a substitute for learning different languages.†

Studying Latin in particular is likely to familiarize you with a lot of English grammatical terminology, much of which was developed for the purpose of discussing Latin grammar. Latin's grammatical structures can be very explicit compared to English or Chinese (where the grammatical structure is of course still present, but it's more difficult to point at something concrete in the sentence and say "_this_ element of the sentence reflects _this_ grammatical feature!")

† Another example of what you might consider edge cases in the space of transitive/intransitive verbs or direct/indirect objects: in Latin, dono is a "ditransitive" verb, which is to say that it may take two direct objects, defined appropriately for Latin as objects which appear in the accusative case. This is pretty unusual. The two direct objects of dono (source of the English verb donate) are a gift and a recipient.

Latin has a much more common giving-related verb, do, which means "give". It is obviously possible to supply a gift and a recipient to do, but for this verb the gift is a direct object and the recipient is an indirect object. (In fact, this verb is the source of the dative case's name!)

So far these are just fun facts about Latin. But if you draw analogies to English, you find something interesting: once we define English direct objects as unmarked and indirect objects as marked by a preposition, we find that most English verbs accepting an indirect object may be regularly transformed into ditransitive verbs. "He bought a dress for his wife" and "he bought his wife a dress" are exactly equivalent statements. Here, some additional structure is provided by the ironclad English requirement that the first object in "ditransitive form" must correspond to the indirect object in "transitive form", and it's common to just call the first of the ditransitive objects an indirect object.

This might lead us to think that "ditransitive" doesn't really make sense as a separate class of verbs in English, but that's not quite true either; there are certain verbs, like bet, which can accept multiple direct objects none of which may appear as indirect objects instead. ("Bet [you] [four dollars] they won't.")

I believe that either object of an English verb in ditransitive form may be promoted to the subject of a passive form of the verb. That would be good evidence that it is really a direct object, except that in English the prepositionally-marked indirect objects of verbs may also be promoted to the subject of a passive form of the verb ("The baby wants to be sung to").

Thanks. I know about Latin from knowing a bit of about the development of English and about ideas of grammar and rhetoric developing from Latin.

What you are relating is the next step for me, and I'd love to find a good (and authoritative) book on it.

Too late to edit: The GPT comment was a joke; sorry that wasn't clear.

"I believe it was Orwell who wrote a friend [a letter]:"

It's perfectly correct and common to say "wrote a friend".

> correct and common

Those are not the same, my common friend.

In this case, however, it is both.

Most analyses focus on ambiguity/inconsistency of ,/and with other ways to read the same sentence, but I've found it interesting to focus merely on the grammar of serial lists alone:

  b and c // binary operator

  a and b and c // associativity

  a, b and c // abbreviation of the first 'and'

  a, b, and c // thus, the ", and" is arguably redundant

Starting with 'and' as a binary conjunction, the above have a sort of notational consistency in that the Oxford comma would be redundant.

  a, b, c // comma-separated list

  a, b, and c // the additional 'and' functioning as a kind of terminator/sentinel symbol, it is intentionally redundant.

But starting with ',' to denote a proper sequence, the 'and' becomes redundant.

It's not that redundancy is bad (redundancy can improve robustness against errors). But the above shows that even ignoring what a/b/c are children or cats or partial clauses, there are some interesting language questions to wonder about. Now I'm imagining flipping the lists around:

  a, and b, c, d, ...
  a and b, c, d, ...

I find this curious..

From OP: “She lives with her two children, a cat and a dog.”

The clause defines a list of type children, size of two, relationship: "her children". The correct punctuation character following a definition is usually : followed by object instantiation. Ideally we would flesh out the full structure to include "lives with" too.

lol etc. However we can avoid ambiguity with:

"She lives with her two children and a cat and a dog."

No need for any funkiness or whatevs. Keep it simple kids! the: "... and a <thing>" idiom can be repeated as much as you like without fiddling with commas.

But what if there is a goldfish too? Someday, you'll need a comma-delimited list.

The only rule is to be clear and consistent.

The fact that, at the time of writing, all of the comments on this post are about the Oxford comma rather than the actual thesis of the article goes a long way to proving its point.

Even though the Oxford comma has fallen out of fashion, there are a few cases where you really need it. My favorite example is the old joke:

There are two hard problems in computer science: naming things, cache invalidation, and off by one errors.

Leave out the comma and it makes a mess of things:

There are two hard problems in computer science: naming things, cache invalidation and off by one errors.

Don't take my word for it. Read both versions out loud, pausing for each comma, and not pausing where the comma was removed.

There are two hard problems in computer science: cache invalidation, naming things and off by one errors.

When reading a written work aloud: is it typical that the decision to pause before the last item is determined by the Oxford comma (or lack thereof)? Personally, I pause there regardless, for the same duration that I pause between items throughout the rest of the series.

> When reading a written work aloud: is it typical that the decision to pause before the last item is determined by the Oxford comma (or lack thereof)?

People are sometimes careless when they write and when they read but a comma does indicate a short pause. So yes, strictly speaking, the “decision to pause” (which is not really a decision) is determined by the presence of a comma.

Often the presence of a pause changes the meaning of the list; there are several examples in this thread. In these cases, adding or removing a comma does alter the meaning of the sentence and should be avoided.

You can't always know whether you've reached the last item in the list or whether it has just turned out that you're halfway through reading the last item, which is a compound clause. Unless, that is, serial commas are reliably present.

With Oxford comma: "We invited the strippers, JFK, and Stalin." [three distinct items in the list]

Without Oxford comma: "We invited the strippers, JFK and Stalin."[two named items in the list]

Reading with or without the pause is natural either way. And massively changes the meaning of the sentence.

Without Oxford comma: "We invited JFK, the stripper and Stalin." [three distinct items in the list]

With Oxford comma: "We invited JFK, the stripper, and Stalin."[two named items in the list with an appositive affirming that we're talking about JFK the stripper and not the former president]

Yes, changing the order and the number of items (plural strippers in original to singular stripper in yours) changes the meaning. That is unsurprising.

I cannot see a way that those sentences do not both mean the same (nonsensical) thing involving strippers, JFK and Stalin.

I think that your analysis in [] is wrong. Your second example says two named items but it still has three - strippers, JFK and Stalin.

> I cannot see a way that those sentences do not both mean the same (nonsensical) thing involving strippers, JFK and Stalin.

This specific example is nonsensical to demonstrate the possible confusion, but this sort of confusion is really common.

> I think that your analysis in [] is wrong. Your second example says two named items but it still has three - strippers, JFK and Stalin.

No, in the second example, JFK and Stalin are two instances of strippers. So there are three nouns, but only two people.

In the sentence with ann Oxford comma, there are a group of strippers, plus two men.

In the sentence without, there are two strippers, the first named JFK, the second named Stalin.

Here’s a drawing to help… http://helveticka.com/blog/wp-content/uploads/2014/08/oxford...

I went to college for journalism and had to memorize the AP Style Guide cover-to-cover. We got marked down for any violation of its rules in our writing, including the Oxford Comma. Repeated violations were usually mocked publicly.

My lifelong avoidance of it verges on the Pavlovian. I don't think I could start using it if I tried - just the thought makes me shudder.

> -

I see you placed spaces around your em dash, which is the AP way. Which I'm using to segue to this completely off-topic question, because you actually know these things: I see you did not use a proper em dash (I don't judge). Does the AP Style Manual see that as an em dash, different than an en dash and from a hyphen, or is it just a dash? I don't think the Style Manual ever uses the term em dash and doesn't necessarily differentiate. What would your journalism professor say?

I know, I know - who cares? I do!

Good question, but I honestly have no idea. Relooking at the sentence, I think a semicolon would probably be the most grammatical. There must be an online version of the Guide to search. Or actually, maybe ChatGPT would know.

Thanks. I searched an older edition of the Guide and my (lack of) results are the basis of my question; I think the current online version requires a subscription.

Discussed at the time:

The Oxford Comma and The Internet - https://news.ycombinator.com/item?id=5247718 - Feb 2013 (43 comments)

It's funny that even in 2013 people were already complaining that articles sounded like AI.

Worth your time, and not - I don't think - unrelated: https://www.ftrain.com/wwic

The Oxford comma is a must, and here's why:

"The U.S. President, a racist, and a misogynist." (Barack Obama, David Duke, Andrew Tate)

"The U.S. President, a racist and a misogynist." (Donald Trump)

I'd write that using a colon instead of the first comma, thus:

"The U.S. President: a racist and a misogynist." (Donald Trump)

That would be a misuse of the colon: The Chicago Manual (sec 6.67) advises against using it "in what would otherwise constitute a grammatical sentence", and only after an indepedent clause (though an assumed verb is allowed).

Dang, could we change the title to "The Oxford Comma, and the Internet (2013)" please.

Why? That wouldn't be an Oxford comma, it would be a bogus comma. The title is fine as it is.

He was joking, I believe.

"The Oxford, Comma, and the Internet (2013)"

There you go. Now that's an Oxford Comma!

See, now you just sound like Christopher Walken.

"I'd like to thank my parents, Ayn Rand and God."

"I'd like to thank my mother, Ayn Rand, and God."

Reply on news.ycombinator.com