Wikipedia talk:Manual of Style

Logical quotation

I noticed that Slim Virgin has been improving the section on logical quotation; nice work! I also noticed that there appears to be an error or omission contained therein. I'm specifically referring to this statement:

"On the English Wikipedia, place all punctuation marks inside the quotation marks if they are part of the quoted material and outside if they are not, regardless of the variety of English in which the article is written ... It does not require placing final periods and commas inside or outside the quotation marks all the time, but maintains their original positions in (or absence from) the quoted material.

This is not 100% accurate to LQ as described by Fowler, because a reproduction of what is technically a complete sentence does not necessarily justify inclusion of the full stop inside the quote marks, even when the full stop is present in the source material. E.g., if the quoted fragment does not represent the author's complete thought as presented in the source material. Fowler gives this sentence as an example:

Do not follow a multitude to do evil.

If we quoted a portion of that material that was also a complete sentence, such as We need not "follow a multitude to do evil", the full stop would not, according to Fowler, belong inside, because the quoted portion does not reproduce the author's complete thought. I'm not sure if I've explained this all that well, but hopefully others will understand what I mean. Rationalobserver (talk) 17:39, 19 September 2014 (UTC)

I also think it might be helpful to re-order the text a bit to make it clearer earlier that the final period of the quoted sentence is retained only when the end of the quoted sentence and the end of the matrix sentence coincide ("[w]here a quoted sentence occurs before the end of the containing sentence, a full-stop inside the quotation marks should be omitted"). I suppose that rule also applies when the only remaining part of the matrix sentence is a bracket, as in my last sentence. Similarly we should clarify (if correct) that the first word of the quoted sentence is capitalized only if the beginning of the quoted sentence coincides with the beginning of the matrix sentence; I couldn't find this anywhere, and there does seem to be some confusion about it. --Boson (talk) 21:17, 19 September 2014 (UTC)

The capitalization issue is discussed in Wikipedia:Manual of Style#Typographic conformity. Rationalobserver (talk) 21:23, 19 September 2014 (UTC)

Thanks! I thought I'd seen it somewhere. It still seems odd that capitalization of the first word and terminating a sentence with a period are dealt with so far apart. --Boson (talk) 23:57, 19 September 2014 (UTC)

This is one of the most contentious parts of the MOS, with multiple RFCs and pages of debate, from as early as last month. Adding advice about non-logical punctuation is confusing, since that's not what we use. The wording has been heavily debated too, and I think reverting to pre-debate versions is just likely to create the exact same debates all over again. __ E L A Q U E A T E 21:46, 19 September 2014 (UTC)

For now I think we should revert to the wording that was settled from the last few debates. I don't think SlimVirgin had a clear consensus to revert to wording that is guaranteed to cause major drama, as it is the exact wording that caused major drama in the past.. If people want to suggest changes to this section, suggest the wording here and get consensus, otherwise we'll have a lot of angry editors who invested days of their lives in the last few debates. __ E L A Q U E A T E 21:54, 19 September 2014 (UTC)

This section has an inline note, that I hope was just accidentally not noticed by the editor making the change: EDITORS PLEASE NOTE: Changes to this section may escalate into heated dispute. Please consider raising any proposed changes for discussion and consensus-building on the talk page before editing. __ E L A Q U E A T E 21:59, 19 September 2014 (UTC)

• We ought to retain a description of aesthetic punctuation, given that it's what most Wikipedians use (and probably most of the publishing world). It's otherwise not clear what the choice is. We also need a description of LQ. Was there consensus to remove it, and if so, can someone point me to the discussion? SlimVirgin ^(talk) 22:12, 19 September 2014 (UTC)

If we say what to use on Wikipedia, then it's clear what we use. I know there's a choice in the greater world. That's true of every other section in the MoS. We don't list everything we don't use, and in this case, we have a large number of examples in that section. Any attempt at describing LQ in greater detail to everyone's satisfaction have led to acrimonious debate. I can't dig up every discussion right now, but if you look in the archive under "Logical quotation" you'll see how heated and protracted the discussions were. __ E L A Q U E A T E 22:18, 19 September 2014 (UTC)

I think this was the last discussion to attempt consensus: [1]. I don't think anyone wants to do this from scratch by reverting to a version all of these editors squabbled over.__ E L A Q U E A T E 22:27, 19 September 2014 (UTC)

Someone removed the long-standing warning that LQ does not involve placing all periods and commas outside, which is what most Wikipedians think it means. I can't find consensus to remove that, so if it exists, please show me. I have added:

This practice does not require placing final periods and commas inside or outside quotation marks all the time. (In contrast, when using aesthetic punctuation (common in the United States, and much of British fiction and British journalism), periods and commas are always placed inside the closing quotation marks, whether or not they are part of the original quoted material.

From "in contrast," sourced to R. M. Ritter, New Hart's Rules: The Handbook of Style for Writers and Editors, Oxford University Press, 2005, pp. 155–156. SlimVirgin ^(talk) 22:39, 19 September 2014 (UTC)

Someone removed it after months of discussion. If you want it returned please discuss first. All it takes is one look at the archives to see that making broad undiscussed changes to this section is like starting a forest fire of wasted editor energy. __ E L A Q U E A T E 22:42, 19 September 2014 (UTC)

This instant reverting is really unacceptable, especially when it's a long-standing sentence that's being restored. The way that section is currently written is very unclear. Please point to the discussion where it was agreed to remove that sentence. Otherwise let me restore it. SlimVirgin ^(talk) 22:44, 19 September 2014 (UTC)

I think the fact you ignored the inline note to seek consensus first was regrettable first. Since that was changed with discussion on the talk page, it's now a bold addition to change it to the pre-discussed version. I think you should show you have a consensus for your change. This is a perennial hot issue, and big changes should be discussed first. I pointed you to this earlier discussion and you should also consider talking to FormerIP about the addition. What the other involved editors ended up with doesn't look my proposed text, so I'm not attached to it. But I do know that what you're attempting to add was seen as inadequate by multiple editors. I think you're making it in good faith, but past discussions haven't shown a consensus for that text.__ E L A Q U E A T E 22:56, 19 September 2014 (UTC)

It was in there a long time (for as long as I recall), and was removed on 1 June by Darkfrog. It needs to be restored, because it's an issue that most Wikipedians get wrong (which is why recommending LQ is a bad idea, but if we're going to recommend it, we must at least make people aware of that point). SlimVirgin ^(talk) 23:43, 19 September 2014 (UTC)

Yes, it was there a long time, and people constantly fought about it. There was a huge discussion after that, and the text was not re-included after it. At this point, I think that if we are putting Coca-cola on the menu, then we don't need a couple paragraphs explaining exactly what Pepsi is in great detail and how we don't have it. If you can think of a new way to explain it that will meet some kind of consensus, more power to you, but defining it as "part of the quoted material" was also fundamentally confusing to people. (Maybe I'll just come back here in a week and see if this newly opened can of worms was successfully dealt with, or if it's just a lengthy re-hash of all the arguments given a couple of months ago.)__ E L A Q U E A T E 00:05, 20 September 2014 (UTC)

Elaqueate, I've been watching this page for years, and I don't recall anyone fighting about that particular sentence (This practice does not require placing final periods and commas inside or outside quotation marks all the time). We do need to warn editors that LQ does not mean sticking periods and commas outside. Can you point me to the discussion where there was consensus to remove it? SlimVirgin ^(talk) 16:21, 20 September 2014 (UTC)

I've pointed you to the vigorous talk page discussion regarding the total wording of the section twice previously. The current section had its examples expanded to address that periods and commas shouldn't be placed outside in all situations. The very first example has the period on the inside, and currently illustrates what you say needs illustrating.__ E L A Q U E A T E 16:59, 20 September 2014 (UTC)

• I think that the line, This practice does not require placing final periods and commas inside or outside quotation marks all the time should be restored, because the key feature of LQ is that there is no definitive answer for every situation, or as Fowler says, "all signs of punctuation used with words in quotation marks must be placed according to the sense."(original emphasis). Rationalobserver (talk) 22:47, 19 September 2014 (UTC)

• That's a good way to put it. We ought to base this section on sources, rather than editors making up examples and descriptions of their own. SlimVirgin ^(talk) 23:45, 19 September 2014 (UTC)

You know... if policy/guidance really is descriptive and not proscriptive, then I would suggest the entire MOS could be shortened to: Just write as comes naturally to you. Someone will follow up and correct what ever errors you make (even if they are not errors). Then someone else will follow up and correct the correction... after which everyone will get into a long debate about what is and is not correct on the MOS talk page. Meanwhile you can go back to writing articles. That would accurately guide editors on what actual practice is. Blueboar (talk) 23:59, 19 September 2014 (UTC)

Too logical__ E L A Q U E A T E 00:08, 20 September 2014 (UTC)

• (edit conflict) If LQ is required of all articles, then this whole section is proscriptive. The guideline is also missing Fowler's suggestion to replace a full stop with a comma and include that comma inside the marks when quoting a fragment that ended with a full stop mid-sentence. As written now, this section is sorely lacking in its attempt to explain what LQ is according to Fowler. It sounds like some editors made this up. Rationalobserver (talk) 00:05, 20 September 2014 (UTC)

I think the LQ recommendation ought to go. Most publishers (and most Wikipedians) use inside punctuation. LQ is difficult to use correctly, and difficult to explain. But at the very least, the long-standing This practice does not require placing final periods and commas inside or outside quotation marks all the time should be restored, as should a description of inside punctuation for contrast.

I agree with Rationalobserver that the current version does not explain it well. Perhaps we should revert to before the May 2014 edits. SlimVirgin ^(talk) 00:14, 20 September 2014 (UTC)

This was the section on 28 May before the changes began. It seems clearer to me. SlimVirgin ^(talk) 00:16, 20 September 2014 (UTC)

Okay, that's two of you. There was a massive RFC on this last year. Are there any new arguments that weren't brought up then? __ E L A Q U E A T E 00:22, 20 September 2014 (UTC)

That RfC wasn't worded neutrally (it made AQ sound as confusing as LQ, and mistakenly implied that Brits always use LQ), and it should have been hosted in a neutral venue. The MoS ought to be descriptive on this point, not prescriptive. We ought to describe both systems (using sources, rather than making up our own descriptions and examples), and request only internal consistency. SlimVirgin ^(talk) 00:29, 20 September 2014 (UTC)

Well, I don't think that a name like J. R. R. Tolkien should have that many spaces after the initials, but it's Wikipedia's default style by consensus, and I would not demand that we include a paragraph about how my preferred way is popular and sensible. It's our style convention, even though it's not one I prefer myself. It sounds like you have a problem with the style and past consensus more than the wording.__ E L A Q U E A T E 00:43, 20 September 2014 (UTC)

Place all punctuation marks inside the quotation marks if they are part of the quoted material and outside if they are not. As has been pointed out above, this doesn't always hold. As it is often described, LQ is about placement according to sense (i.e. *not* source).

This practice does not require placing final periods and commas inside or outside quotation marks all the time. I'm not sure that guidance of this type is needed. It is something that should only be added if there is clear consensus for it. In the event of such a consensus, I would say that wording is unclear and, as it is, not correct. In my experience, LQ does indeed place final stops and commas inside or outside quotations marks all the time. Never above, below or in the same place, but always inside or outside.

On the question of whether we should cite sources, we generally do not in the MoS, although I don't see why it should never be done. However, in the current case, there is consensus to use LQ on en.wp, nothing more. Which means we need to ensure that we are not taking an over-prescriptive approach, so we need to be sure that anything we cite is uncontroversial.

I don't think we should revert to May. Even if the current version can be improved upon, that particular action would definitely not be an improvement. Formerip (talk) 00:25, 20 September 2014 (UTC)

Elaqueate, it's both. I think LQ shouldn't be recommended on WP because it's fussy and difficult to get right. It's a measure of how difficult it is that MoS editors have not been able to describe it clearly.

Most publishers don't use it. Most editors don't use it either, and of the few that do, most don't get it right. It's particularly difficult to use correctly on Wikipedia, because to do so you need access to the source, but our articles have multiple authors, many of whom don't have that access.

Given that we have STYLEVAR, CITEVAR and ENGVAR, it's puzzling that the MoS takes a different approach with this one, difficult, style issue – and now goes so far as to remove even a description of the more widely used, and much easier, alternative. SlimVirgin ^(talk) 14:42, 20 September 2014 (UTC)

Looking at the WP:MOS, we don't list all of the alternatives, just because there are alternatives. There are different standards in the world for where to add a possessive "s", for how to capitalize titles, for how to add punctuation to abbreviations, etc. This isn't an article describing all of the world's options when it comes to these issues. (We do have an actual article for that, where all styles are documented.) There are many popular styles we don't mention, because there's no consensus to recommend using them. I don't agree with every convention we have in the MoS, but it's a document to show where consensus has fallen on certain broad style issues in the past. For the less common examples where the advice isn't helpful, there's common sense. I don't want to get into a drawn out debate over why there's a consensus of one over the other or why you prefer one. You're welcome to your opinion, some people share it, most others on Wikipedia seem to have disagreed. __ E L A Q U E A T E 15:32, 20 September 2014 (UTC)

(ec) I was amused to see such a collection of statements with which I disagree (I am referring to SlimVirgin's last post here):

• That a practice cannot easily be codified to the last jot and tittle is no argument to abandon it. In this instance, when what is "right" is confusing, the detail probably does not matter that much. We may as well abandon English.
• "Most ... don't use it"; "more widely used": this has no real relevance and sounds overstated anyway – which style to use is a choice; there are real benefits in a coherent choice of style; which style dominates in other contexts is not important.
• xxxVAR: These are all symptoms of a lack of uniformity, but at root they are a rule to keep the status quo rather than warring. They are not at root motivating an introduction of style variation, which is what you seem to be motivating. ENGVAR is a particularly thorny one, not to be emulated.

—Quondum 15:38, 20 September 2014 (UTC)

@SlimVirgin: your comments above reinforce my concern that you are attempting to change the consensus view rather than clarify it. If you have such strong feelings against LQ, then you probably shouldn't be making these edits. Peter coxhead (talk) 17:15, 20 September 2014 (UTC)

• Comment The situation is somewhat more complicated than some of the discussion above implies. The English Wikipedia uses the same quotation marks for different purposes, including actual quotation, scare-quotes, and indicating a mention of a word rather than a use. For the last two, the advice is very simple: never include the punctuation within the quote marks. For quote marks used to indicate actual quotation, it is indeed hard to give clear advice, as we discovered in previous RfCs and discussions. However, TQ doesn't solve the problem entirely since it only applies to commas and full stops/periods. You still have to decide where to place colons and semicolons, for example, although these cases are admittedly less frequent. Peter coxhead (talk) 17:10, 20 September 2014 (UTC)

• Comment I think an issue that is complicating this process involves the varied interpretations of multiple editors. I suggest that we start by identifying which source we should use as a primary guide. I assume that source is Fowler, so why not re-write the guideline as a direct paraphrase of what Fowler actually suggests, and limit this section in perpetuity to only that which can be directly sourced to Fowler? In my Burchfield edition from 2004, Fowler outlines what we call logical quotation with two overview sections and eight sub-sections. Maybe we should attempt to structure this guideline similarly. I.e., as long as this section is subject to the opinions of editors we will not achieve a lasting consensus. However, if we all agree to only add that which is consistent with Fowler, we will arrive at a lasting consensus that is both accurate to external style guides and relatively easy to follow. Rationalobserver (talk) 18:10, 20 September 2014 (UTC)

I'm not sure about the degree to which there actually are varied interpretations of multiple editors here, RO. AFAICT, no-one actually disagrees with what is in the current version (unless I've missed it). It's more about whether we should be recommending LQ in the first place and/or particular introductory remarks would be useful. But what I would say is that we should not defer to any particular usage guide. If there are differences between different authorities, then what we should do is advise both or neither. Plus, from a copyright point-of-view, we can't just paraphrase Fowler's. Formerip (talk) 22:03, 20 September 2014 (UTC)

As I said above when opening this thread, the guideline as written is not 100% accurate to Fowler.

[LQ] does not require placing final periods and commas inside or outside the quotation marks all the time, but maintains their original positions in (or absence from) the quoted material.

This is not 100% accurate to LQ as described by Fowler, because a reproduction of what is technically a complete sentence does not necessarily justify inclusion of the full stop inside the quote marks, even when the full stop is present in the source material. E.g., if the quoted fragment does not represent the author's complete thought as presented in the source material. Fowler gives this sentence as an example:

Do not follow a multitude to do evil.

If we quoted a portion of that material that was also a complete sentence, such as We need not "follow a multitude to do evil", the full stop would not, according to Fowler, belong inside, because the quoted portion does not reproduce the author's complete thought. If we aren't basing this guideline on Fowler, then which style guide are we using? Have we invented a system that we call LQ from consensus of editor opinion? Rationalobserver (talk) 15:47, 21 September 2014 (UTC)

Umm, that's not the guideline as written; please note that you seem to be quoting the removed sentence as if it was in the guideline now. Many editors agreed with you that the description that contained "maintains their original positions" had problems of sense. That's why we don't have the wording now, and why it was removed in June. __ E L A Q U E A T E 16:06, 21 September 2014 (UTC)

My bad. I hadn't looked back the guideline since SlimVirgin's recent edits were reverted. Rationalobserver (talk) 16:47, 21 September 2014 (UTC)

Technically, it does not matter whether whether WP follows any external style guide, nor whether it chooses to apply the name LQ at slight variance to other style guides, and indeed it does not matter whether the style can be sourced. The purpose of a MoS is to set a style, and this inherently will be done via a consensus between the editors.

That said, using another style guide in refinements to the MoS can be helpful. —Quondum 16:18, 21 September 2014 (UTC)

I hear you, but that's not really my point, which has more to do with attempting to gain consensus amongst numerous editors without any frame of reference. I would bet that 90% of our MoS is taken directly from external style guides anyway, even if they aren't mentioned in-line by name. The way I see it, our style guide should be consistent with external style guides whenever it's not Wikipedia specific. This years-long dispute will never end as long as you and others hold the opinion that we can make this up as we go, and that we can create an LQ system that is not consistent with any one particular style guide. Having said that, I guess local consensus does trump all external logic and reasoning, and I think I'll just disengage here, as this is a mess that nobody wants to correct in any lasting way. Rationalobserver (talk) 16:47, 21 September 2014 (UTC)

One last point that I'll reiterate is that the guideline is currently missing Fowler's suggestion to replace a full stop with a comma and include that comma inside the marks when quoting – mid-sentence – a fragment that originally ended with a full stop. E.g., if the source material reads: Some dog breeds are obedient, and others are gentle but not obedient., and we quote the last portion: According to John, some dog breeds are "not obedient," but others are., because the quoted fragment ends with a full stop, that full stop should be replaced with an inside comma, per Fowler. Rationalobserver (talk) 17:05, 21 September 2014 (UTC)

Understood, and I don't disagree. But what we have is probably going to remain a poor reflection of a comprehensive style because it does not make sense to devote as much space to it as a comprehensive style guide does.

The example that you give here is unfortunate, since one would never include the period terminating the sentence with a sentence fragment. Hence, it would not be there to be replaced by a comma. Besides, the MoS already does in a sense include your final point: it says: "Where a quoted sentence is followed by a clause identifying the speaker, a comma should be used in place of a full-stop [...]", though I've now reworded this to apply to other clauses. —Quondum 19:54, 21 September 2014 (UTC)

I think it looks better now, and to clarify, all I meant was that because the source material being quoted included a full stop – e.g., "not obedient.", that full stop is replaced with a comma that is included, whereas the full stop, as you pointed out, would have been omitted because the quoted portion is not a complete sentence. However, if the quoted portion was "John is not obedient.", the same rule would apply when quoting this complete sentence mid-sentence. I.e., the full stop should be swapped out with a comma that is included inside the quote marks even though the original text did not have a comma in that position. Rationalobserver (talk) 20:04, 21 September 2014 (UTC)

Yes, I apologize for being obtuse; I should have phrased my reply to acknowledge what was obviously your intent. —Quondum 21:35, 21 September 2014 (UTC)

Rationalobsever, please could you give a citation for this thing about commas mid-sentence? As far as I can see, Fowler's recommends on p 647 that closing points should generally be omitted if the quote occurs before the end of sentence and not replaced by a comma (this is in section iii), but that a comma is used to represent the stop where what follows is "such words as he said" (section iv). This is the same advice as given by the Oxford Guide to English usage. So, I'm very unsure that the change made to the page was appropriate. Formerip (talk) 22:05, 21 September 2014 (UTC)

FormerIP, on page 647, in IV, a semi-colon is turned into a comma, and included inside the quote marks. Fowler says that "any punctuation at the point where it is broken off, a comma is placed within the quotation marks to represent this." Rationalobserver (talk) 00:49, 22 September 2014 (UTC)

You've quoted half a sentence, though. The full sentence is

When a quotation is broken off and resumed after such words as he said, if it would naturally have had any punctuation at the point where it is broken off, a comma is placed within the quotation marks to represent this.

This only seems to apply in the specific case of a clause identifying the speaker, as per the current guidance, not to any case where a quotation occurs mid-sentence. Formerip (talk) 16:38, 22 September 2014 (UTC)

How so? It seems to apply to whenever the quotation is interrupted, and has little or nothing to do with identification of the speaker. If you want to interpret the example too narrowly, you risk confining it to the specific phrase "he said". The simplest alternative is to interpret it as any interrupting clause. —Quondum 19:21, 22 September 2014 (UTC)

That's a different suggestion to what has been made above. I'm not sure, though, that it's really possible to have a clause interrupting a quotation other than one identifying the speaker particularly in an encyclopaedic style. Can you think of an example? Formerip (talk) 21:21, 22 September 2014 (UTC)

How about, "It has been said, 'we need not follow a multitude to do evil,' but Wikipedia editors do not necessarily agree." The speaker has not been identified, but the quoted fragment ends with a full stop that ought to be replaced with a comma. Rationalobserver (talk) 21:56, 22 September 2014 (UTC)

There's no interrupting clause in that example, so even if you take a wider interpretation of "words such as he said" than I think is warranted, the example is not supported by Fowler's (because the quote does not resume). Maybe you could have an interrupting clause as an editorial note, but that should really be placed as a footnote or in square brackets. So I think we may be talking about a situation that will never arise. Formerip (talk) 22:03, 22 September 2014 (UTC)

How about, According to Fowler, "it cannot be done," so Wikipedia accordingly recommends that we accept his advice and "give up the task". Notice that I did not include the full stop inside because, although a complete sentence, it does not represent the author's complete thought, which is another aspect of LQ that is currently not represented in the guideline. Rationalobserver (talk) 22:18, 22 September 2014 (UTC)

This example is also not supported by Fowler's because it doesn't contain a single quote which is broken up and resumes. It just contains two separate quotes. Formerip (talk) 23:19, 22 September 2014 (UTC)

How is this syntactically distinct from: "It cannot be done," he said; "we must give up the task.", which is the example that Fowler gives? Rationalobserver (talk) 23:28, 22 September 2014 (UTC)

Because you've attributed the first part to Fowler's and the second part to Wikipedia, so what you've ended up with two separate quotations. Also, because your sentence does not contain any "words such as he said". Formerip (talk) 00:00, 23 September 2014 (UTC)

Above you said, "I'm not sure, though, that it's really possible to have a clause interrupting a quotation other than one identifying the speaker particularly in an encyclopaedic style."

Source material: "It cannot be done; we must give up the task."

Possible quotation: Nonetheless, it was decided that if "it cannot be done," then "we must give up the task." The need to identify the speaker can be established by the preceding material, and in the proper context the above example is perfectly encyclopaedic. Rationalobserver (talk) 00:11, 23 September 2014 (UTC)

This example provides a good example of why the rule you are supposing is mistaken, whichever version of it we are talking about. Most people would recommend separating off the relative clause in your sentence with commas, which can be understood more easily if we convert to reported speech:

Nontheless, it was decided that, if it could not be done, then the task should be given up.

This is just common-or-garden punctuation, nothing to do with LQ. But if we also apply your LQ rule, we end up with this:

Nonetheless, it was decided that, if "it cannot be done,", then "we must give up the task."

Formerip (talk) 17:13, 23 September 2014 (UTC)

FormerIP, if your point is that we ought not include any language that refers to unlikely scenarios, will you please indicate which, if any, of the current examples are likely constructions on Wikipedia. Except maybe from an article about a movie or book that details a plot summary, every current example is incredibly unlikely to occur in encyclopaedic writing. Rationalobserver (talk) 00:24, 23 September 2014 (UTC)

I don't think it matters very much one way or the other if we describe a rule about something that is unlikely to occur, so long as following the rule will not produce an error.

Here's a challenge: can you find examples of a comma coming immediately before a quotation mark, where what follows is something other than a "he said" type clause, on the websites of The Guardian, The Times, The Telegraph or The Independent? Formerip (talk) 17:34, 23 September 2014 (UTC)

Here's one that took me about 10 minutes to find: "As Basil Fawlty famously exclaimed, as he struggled to start his car: 'If you don't go, there's little point in having you,' and Nico Rosberg might have sympathised with those sentiments on Sunday when his team-mate Lewis Hamilton roared off to win the Singapore Grand Prix, leaving the German stuck on the grid." Rationalobserver (talk) 18:30, 23 September 2014 (UTC)

My copyedit that you reverted was just trying to make it read better. But there's a bigger issue in the immediately preceding edits, such as this one in which the advice for whether the comma goes inside or out seems to be explicitly contradictory now, whereas before it was only slightly so. We need to decide which to recommend, or clarify the conditions that would make it go one way or the other. Dicklyon (talk) 22:57, 21 September 2014 (UTC)

Sorry, that was just me not paying attention to my own revert. I've done it properly now. Formerip (talk) 23:05, 21 September 2014 (UTC)

I've been trying to formulate a short description of the rules behind LQ, which has resulted in me feeling that the bit on the comma that is already there is an intrusion of an AQ rule into the LQ framework. Whatever the case may be, the piece on the comma substitution seems a little inconsistent as it stands. It would be nice to clarify the position on this. —Quondum 23:35, 21 September 2014 (UTC)

I can't follow who is arguing what here, but I want to support Rationalobserver's point about sourcing this to Fowler (and citing it), not making up examples and descriptions of our own, and not leaving out key points. I also can't see a reason to leave out a brief description of the more-common alternative style (commas and periods always inside). SlimVirgin ^(talk) 16:29, 22 September 2014 (UTC)

FTR, Fowler addresses the "alternative style" in Quotation marks, section 3, which he calls the American English style. Rationalobserver (talk) 16:42, 22 September 2014 (UTC)

FWIW, my Cambridge Guide to English Usage calls this a North American practice,(2004, p.454) and my edition of New Hart's Rules calls it a US practice.(2005, p.155) Rationalobserver (talk) 19:42, 22 September 2014 (UTC)

Most style guides refer to these as British and American, or North American, style. The obvious solution is to make it an ENGVAR choice. There has been resistance to doing that, because American style is also used in the UK (in British fiction, in particular).

But that just makes it all the more puzzling that we force the less-popular and more-difficult style on the entire English Wikipedia, including in North American articles, where readers are less likely to have encountered it before. SlimVirgin ^(talk) 20:25, 22 September 2014 (UTC)

It's interesting to look back at the archives and see how many editors deny this obvious fact. It's almost as though they know it's an EngVar issue, but they take the approach that it is not so that nobody can say that the BrE system is being forced on AmE articles, but as far as I can tell, it is. I agree that punctuation within quotations should be dealt with in the same way that we deal with the other EngVar issues, but I would be very surprised if a consensus would form around that, though I'm really not too certain why it isn't an option. Implementing LQ can be tedious for many AmE editors, whereas the so-called American Style could not be more simple. The previous RfC's suggest that this is – for no apparent reason – Wikipedia's EngVar exception, whereby AmE editors are not afforded the opportunity to reject LQ in favor of the American style, even if local consensus agrees, which seems like a very strange situation indeed. Rationalobserver (talk) 20:46, 22 September 2014 (UTC)

My own preference is that it be a STYLEVAR issue, i.e. the first major contributor chooses. But failing that, ENGVAR would solve the years-long controversy about it, and would make the MoS consistent with what most editors do anyway. SlimVirgin ^(talk) 20:56, 22 September 2014 (UTC)

"we force ... more-difficult style" that is an opinion not a fact (I think LQ is simpler and less confusing) and I think you ought to qualify such statements by indicating that it is you opinion and not a fact. You are in favour of wording in the MOS that forces reference tags after punctuation, is it not hypocritical of you to support one universal rule that you like and then complain about another that you do not? -- PBS (talk) 21:03, 22 September 2014 (UTC)

PBS, one of my objections to LQ is that we're forcing a minority preference onto the majority. Your ref preference is also a minority one. My aesthetic objection to LQ and your ref tags is that they leave dangling periods and commas, "like this", and this. If you like the danglers, at least you're consistent! SlimVirgin ^(talk) 22:24, 22 September 2014 (UTC)

For those who understand LQ's subtle machinations, it is not really more difficult, but it does require that the person checking for compliance with LQ has access to the source material, which is not always a convenient thing for a collaborative project like Wikipedia. Rationalobserver (talk) 21:21, 22 September 2014 (UTC)

It seems to me that you don't really have to do that. You can just always put the period outside the quote. That always complies with at least the idea of LQ — even if the period was part of the original quote, so was the part without the period. Similarly for commas. Question marks and exclamation points, if part of the original, need to go inside, but for those there's a clear difference in meaning.

Admittedly this becomes problematic when you're quoting multiple sentences inline. I think (I certainly hope) that that's a rare case. When quoting multiple sentences, editors should consider a pull quote instead. --Trovatore (talk) 21:29, 22 September 2014 (UTC)

"You can just always put the period outside the quote." That is decidedly not LQ; that's the North American style that is expressly forbidden on Wikipedia. Rationalobserver (talk) 21:32, 22 September 2014 (UTC)

Possibly I'm not following you — I thought the "North American" style was "always inside" rather than "always outside"??? If I had to give what I was talking about a name distinct from LQ, it would be something like "programmers' style". --Trovatore (talk) 21:48, 22 September 2014 (UTC)

Yes, in NA style it is "always inside", so that wasn't really accurate. What I meant was that to always do one thing or the other is not at all LQ, which explicitly suggests to place "according to sense", not to place based on a rule-of-thumb. Rationalobserver (talk) 21:54, 22 September 2014 (UTC)

It's not a "rule of thumb". It follows the logical structure. The period always has the meaning of closing the sentence at the same logical level at which it appears. You just might omit to close off the quoted sentence, but that's not as bad as omitting to close off the "main" sentence (the one doing the quoting).

To do perfect logical quotation, what you should do, when quoting a complete sentence, is allow a period, followed by a quote mark, followed by a period. The "inside" quote mark closes the quoted sentence; the "outside" one closes the main sentence. This seems to be disallowed for purely aesthetic reasons. So which period to omit? If you omit the "outside" one, my programmer brain says that the main sentence was never closed off at all, and it bugs me. Omitting the "inside" one bugs me much less. --Trovatore (talk) 22:17, 22 September 2014 (UTC)

RE: To do perfect logical quotation, what you should do, when quoting a complete sentence, is allow a period, followed by a quote mark, followed by a period." Sorry, but, according to Fowler, this is also wrong; he addresses this point in 2(v), and he recommends that "the point should be set inside the quotation marks and the point closing the main sentence omitted." Rationalobserver (talk) 22:28, 22 September 2014 (UTC)

I'm not talking about Fowler. I'm talking about logic. --Trovatore (talk) 22:30, 22 September 2014 (UTC)

(ec) That isn't LQ, Trovatore (and logical quotation has nothing to do with logic). The problem is that Wikipedians use their own punctuation style now, because they've had LQ forced on them, but few know how to use it. So our articles are internally inconsistent. This is bad for new editors, who need to be able to look at an article and discern consistent styles (punctuation, spelling, citations) so they can copy them. SlimVirgin ^(talk) 22:33, 22 September 2014 (UTC)

Ideal would be to use programmers' style consistently, and completely short-circuit the nationalistic aspects of the discussion. --Trovatore (talk) 22:36, 22 September 2014 (UTC)

Arb break

(ec) Right, Trovatore, and Fowler agrees that one could logically come to that conclusion, but as a matter of style, it would be absolutely ghoulish to have redundant full stops, so, despite its apparent logic, the suggestion is slightly absurd. To take it one step further, if a two-sentence quote ends with a complete sentence, then why wouldn't you do the same thing that you would do if you quoted only one complete sentence? Why would John said, "that dog is obedient." be correct, but John said, "That dog is obedient. He also said that the dog is gentle." be in correct? I don't see why the preceding material should help determine the placement of the final full stop, so that one sentence might have it included but two or more would not. Rationalobserver (talk) 22:41, 22 September 2014 (UTC)

Why would you quote two sentences inline? Use a pull quote. Granted, not always — rigid rules like that rarely work well. But for the most part we shouldn't have long quotations inline.

I think there's a key issue here that's not often discussed: Most of the disputes over quotation styles involve using quote marks to, you know, actually quote something. But that's not the main use model in context! In an encyclopedia, we should not be quoting stuff very much.

The main good uses for quotes, I think, are in order (1) use–mention distinction, (2) scare quotes (including the case of indicating that you're duplicating the nomenclature of a particular author which may be idiosyncratic), and only finally (3) actual quotations.

Now, for the first two uses, putting terminal punctuation inside the quotation mark just makes no sense at all, as far as I can see. --Trovatore (talk) 23:09, 22 September 2014 (UTC)

Italics are more appropriate for "mention distinction", and LQ does not apply to scare quotes, i.e., you would never under any circumstances include a full stop in them. Rationalobserver (talk) 23:13, 22 September 2014 (UTC)

Italics don't always work well. They work OK for words-as-words, provided you aren't using them elsewhere for emphasis. But use–mention is much more than words-as-words. --Trovatore (talk) 23:31, 22 September 2014 (UTC)

That may well be true, but, as with scare quotes, I cannot foresee an example where you could correctly include a full stop inside the quote marks while using them as |mention distinction; can you? Rationalobserver (talk) 23:38, 22 September 2014 (UTC)

Sure. In so-called "American style", how about this? There is considerable dispute over the correct use of the adjective "American." (Note by the way that I've given you an example where "American style" would require moving the comma inside scare quotes.) --Trovatore (talk) 23:42, 22 September 2014 (UTC)

In LQ, the full stop would not be included because the quoted portion is not a complete sentence, so I'm unsure what relevance this example has to this discussion, except to illustrate that American style might accept this. Per the Wikipedia MoS, the so-called American style is not an option. Rationalobserver (talk) 00:21, 23 September 2014 (UTC)

Oh right, for LQ there's no problem at all. I'm adducing it as a reason to avoid adopting AQ. --Trovatore (talk) 00:44, 23 September 2014 (UTC)

I think LQ is absolutely fraught with problems; I was merely pointing out that the current consensus is to disallow AQ and require LQ of all Wikipedia articles regardless of the variety of English in use therein. If it came up for a !vote, I would support allowing local consensus to adopt which ever they wanted, with some respect to EngVar if that's what editors of a particular article wanted. FWIW, the Chicago Manual of Style is 100% AQ, and as far as I can tell, that's the manual that is most often followed here, not that that has any particular bearing on consensus. Rationalobserver (talk) 00:55, 23 September 2014 (UTC)

Many of the problems would be obviated by modifying LQ to prefer to keep the period that ends the main sentence, rather than the one that ends the quoted sentence. Yes, I'm aware that it would probably be an innovation in terms of published style manuals, although it's the style that many have used on the Internet almost since the beginning. --Trovatore (talk) 01:00, 23 September 2014 (UTC)

Trovatore, I'd be interested to hear what your objection is to AQ. For example, what is wrong with this?

Discussing the view that "LQ is absolutely fraught with problems," Travatore recalled that placing periods outside is "the style that many have [always] used on the Internet."

It's easy to remember, it notes the comma pause, it uses a period to signal the end of the sentence, there's no dangling punctuation, you don't need access to the original source material, and new editors learning the style will know what to do with commas and periods in future based on this one example. SlimVirgin ^(talk) 01:14, 23 September 2014 (UTC)

Extra ping for Trovatore because I mistyped. SlimVirgin ^(talk) 01:15, 23 September 2014 (UTC)

Well, my objection is that it doesn't follow the logical structure. The quotes include punctuation that is not logically part of the material quoted (or mentioned — the "fraught" quote is a mention rather than a quote, given that no one is actually being quoted). And the main sentence omits punctuation that is logically a part of it; there is no period to close the sentence, because the last period that appears is part of the quote and not part of the main sentence. --Trovatore (talk) 01:22, 23 September 2014 (UTC) Oh — on further reading, the period at the end of the "Internet" quote actually does belong to it. Still, the period to close the main sentence is not there. To my mind, if you really can't stomach ."., it is worse to omit the period at the top level than the one inside the quote. --Trovatore (talk) 01:26, 23 September 2014 (UTC)
Wow — having trouble getting this quite right. Actually, you were directly quoting Rational, whereas you were not directly quoting me, so my previous "on further reading" intervention is more-or-less backwards. Still, with the obvious modifications, I hope my point is clear. --Trovatore (talk) 01:35, 23 September 2014 (UTC)

I'm not sure how you're using the word logic in this context (the fraught fragment is a quote from Rationalobserver, and the other fragment is a quote from you – just not the whole sentence).

These are just choices about how to present pauses visually. If you listen to someone speak, and you're transcribing, you have to make these choices. And if you're copying someone else's text, you can make your own choice about how to present the pauses (which is AQ), or you can use a mixture of their choice and your own, depending on the context (which is LQ). My question to you is: why not make the choice that is easier to follow, for you and those watching (e.g. new editors)? SlimVirgin ^(talk) 01:34, 23 September 2014 (UTC)

Come on, "logic" should be clear, shouldn't it? In the sense of Tarski, if you like. Punctuation to some extent does reflect pauses (sometimes I add commas that are not logically necessary, for that reason), but more importantly, it is a guide to the logical structure of the sentence. --Trovatore (talk) 01:38, 23 September 2014 (UTC)

I don't think we should be requiring British punctuation on articles written in non-British varieties of English, but as long as this rule is in place, we should get it right. If British punctuation doesn't really require original placement, then neither should the MoS. That being said, I don't think SlimV did anything wrong. The in-text warning advises editors to consider posting changes here first; it's not required.

Trovatore, the most logical way to write and punctuate is the way that will be understood and recognized as correct by one's readers. The human brain does not process visual information the same way that computers parse text. Example: Some fonts are easier to read with some light sources. We don't need colors to be inside the lines to recognize objects in paintings. The numerical ratio 1:1.66 usually looks uncannily appealing to most people.

Under actual use, American and British styles are roughly equal; neither of them causes non-hypothetical problems under real-world or Wikipedia conditions. They also have their particular ups and downs: American style is easier to learn, use and copy-edit (this last being an advantage on Wikipedia), but British style is more familiar to computer programmers, who make up a disproportionate number of Wikipedians (though not so much as in Wikipedia's early years). Darkfrog24 (talk) 07:45, 23 September 2014 (UTC)

OK, let's be a little careful here. It is important to distinguish between two senses of the word "logical". I am using it to mean "transparently reflective of the underlying logical structure"; you appear to be using it to mean "that which humans, using a rational process, conclude to be the most opportune course to take".

Now, I think the two senses happen to give the same answer here, but that is a different argument. When I use the term "logical" for a style of punctuation, I am referring to the first sense, not per se the second. --Trovatore (talk) 09:37, 23 September 2014 (UTC)

If you'll pardon the digression, where did you find that first definition. It's not quite what's in the OED. Is it a specialized definition, like the way "gender" differs in ordinary English and the social sciences? Darkfrog24 (talk) 14:43, 23 September 2014 (UTC)

Examples

Does anyone else think that we should construct new examples that are representative of the kind of encyclopedic writing that one might actually expect to find on Wikipedia? As they are currently written, I cannot imagine any of these constructions occurring on Wikipedia outside a plot summary. Rationalobserver (talk) 16:42, 23 September 2014 (UTC)

What did you have in mind? (If you're referring to the use of Finding Nemo characters, I don't think it's a problem.) Darkfrog24 (talk) 02:49, 25 September 2014 (UTC)

It's not the characters per se, it's the content. I cannot envision any of these actually occurring on Wikipedia outside a plot summary, which is okay for one or two, but at least some of the examples should represent things that Wikipedians would actually write in other types of articles. Also, I wonder if there are any copyvio issues involved with directly quoting material from the movie outside article space. We essentially have 10 quotes that we've made no attempt to justify based on fair-use. Maybe I'm wrong about this, but we could never include 10 unattributed quotes of material from a copyrighted work elsewhere on Wikipedia. Could we? Rationalobserver (talk) 15:53, 25 September 2014 (UTC)

We're probably okay for copyvio, but I see what you mean about focusing more on the type of quotation that would happen in an article. Things like song titles are more common than long dialogue-style quotations. Darkfrog24 (talk) 19:13, 26 September 2014 (UTC)

I would be happy to come up with a few alternates, but I get the feeling that FormerIP will revert any attempts to remove the existing examples. Rationalobserver (talk) 19:15, 26 September 2014 (UTC)

Whenever the issue of converting WP:LQ to an ENGVAR-based rule comes up, I usually return to these.

Bruce Springsteen, nicknamed "The Boss," wrote "American Skin." for American style and Eric Clapton, nicknamed "God", wrote "Cocaine". for British.

Is this more like what you had in mind? Darkfrog24 (talk) 16:16, 28 September 2014 (UTC)

Yes, that much closer to what I would expect to see on Wikipedia! Rationalobserver (talk) 18:03, 28 September 2014 (UTC)

Adjust that for accuracy; Eric Clapton did not write Cocaine (song). Better go with "Wonderful Tonight" if you want a Clapton solo composition. --Nat Gertler (talk) 18:12, 28 September 2014 (UTC)

Latin incipits

See discussion here: WT:AT#Italicization of Latin incipits --Francis Schonken (talk) 02:20, 26 September 2014 (UTC)

Please review changes at WP:ORDER

I changed WP:ORDER to be a text summary and list overview of Wikipedia_talk:Manual_of_Style/Layout. I would appreciate anyone commenting at Wikipedia_talk:Manual_of_Style/Layout#WP:ORDER_and_a_master_list_for_the_layout_order on the changes I made. Blue Rasberry (talk) 11:19, 26 September 2014 (UTC)

Someone asked for the diff - I am now sharing it here, but please put comments on the layout page because that is the natural place to archive discussion. Blue Rasberry (talk) 13:06, 26 September 2014 (UTC)

The Key to lists. No style guideline exists.

I have requested a discussion on whether the explanatory key to a list should be prominently displayed before, to the side or after a list, or is there a better, smarter, but still fully accessible to all users, alternative method of providing a key. As far as I can see, the key is not mentioned in any MOS guideline. Whilst the discussion started at a FLC review and continued at WT:CRIC, it is bigger than a single article or project, so should be discussed at Wikipedia_talk:Manual_of_Style/Lists#The_List_Key. The-Pope (talk) 01:54, 29 September 2014 (UTC)

NOTUSA in template parameters

We say here, "Do not use U.S.A. or USA, except in a quotation or as part of a proper name (Team USA), because these abbreviations are also used for United States Army and other names."

The rationale pertains to prose. Is there any reason to shorten or expand USA in template {{infobox person}} [2] and {{Persondata}} [3] birthplace and deathplace parameters? I doubt it.

--P64 (talk) 16:59, 29 September 2014 (UTC)

No means no. Radiopathy •talk• 23:17, 29 September 2014 (UTC)

Interesting discussion concerning ENGVAR

Wikipedia talk:WikiProject Video games. --John (talk) 21:58, 30 September 2014 (UTC)

Is close paraphrasing acceptable?

Opinions are needed on the following matter: Wikipedia:Village pump (policy)#Is close paraphrasing acceptable?. A WP:Permalink to that discussion is here. Rationalobserver (talk) 20:11, 1 October 2014 (UTC)

RfC - Animal breeds in lower case

Moved from Wikipedia talk:Manual of Style/Capitalization

The capitalization of animal breeds is not specifically called out in the Manual of Style. This has led to numerous disputes, and it seems to me the MOS should provide some guidance in this matter. There is no reason to leave it up to the individual editor. Obviously this would not apply to proper nouns within a breed name. Should the Wikipedia:Manual of Style#Animals, plants, and other organisms section, the Common names subsection of Wikipedia:Manual of Style/Capital letters#Animals, plants, and other organisms, the Wikipedia:Naming conventions (fauna) page be amended to add animal breeds to the list of items that are generally lowercase? Krychek (talk) 19:11, 1 October 2014 (UTC)

Support as nominator - Dictionaries and neutral style books tend not to capitalize breeds, while pet industry publications do. However, to cite the latter is an example of the specialist style fallacy. Krychek (talk) 19:16, 1 October 2014 (UTC) link refactored by SMcCandlish, after move of RfC.