Best practice for quote checking translations that use quote dashes

Question

I would like to gather suggestions from anyone who has experience checking translations that use quote dashes. I have a project in Spanish that has the following characteristics:

• First level quotes are
• double angle quotes «…» for OT quoted material and monologues.
• quote dashes — for dialogue
• Both use curly quotes for second “…” and third ‘…’ level quotes
• Quote dashes are never closed therefore the end of the quote is either only clear from the context or is simply ambiguous
• The same character u+2014 — is used for both quote dashes and for marking parenthetical material.

I certainty think I need to change quote dashes to use u+2015 ― but beyond that I am uncertain.

Any suggestions on how to proceed?

Answer 1

CrazyRocky:

Quote dashes are never closed therefore the end of the quote is either only clear from the context or is simply ambiguous

Yes, I’ve come across this issue with Russian and other Cyrillic-alphabet languages, too. Here’s an idea I had: we could request the creation of a USFM marker for < end-of-quote >.

In our context, one of the issues that was consuming my thoughts was: what if we use quote dashes and later decide for paired quotation marks? Manually changing them would be a lot of work, so how could we automatically change the quotation marks? An USFM marker would be an invisible pair to the dash, and would allow you to search-and-replace for paired quotation marks.

I’ll add a few other thoughts:

• It seems to me that quote dashes were quite common in English language novels decades ago, but that a shift was made and now they are completely out of fashion.
• In languages that use them, they usual context is dialogue, and people rarely speak more than a paragraph. Thus there is no ambiguity.
• In Scripture, we have long monologues, and these are often part of dialogues. Paragraphing the long utterances can give rise to ambiguity as to where the quote ends. This happens when the following text is not a quote (if it is a quote, the new quote dash shows you that the quote before has ended). In some examples I’ve looked at, the quote ends mid-paragraph! – for example the end-of-quote in Matthew 6 between vv27–28.
• Not to mention that there are often quotes within quotes … within quotes^⁎. I presume publications that use quote dashes always use quotation marks for these.
• Those of our translators that favour quote dashes say that they make dialogues look “more alive”. Personally, I think the same can be achieved by starting each speaker’s utterance in a new paragraph. Interestingly, novels usually do that, but printed Bibles often don’t.
• Another possible disadvantage of quote dashes is that the printed text spans more pages; when quotation marks are used, a whole dialogue can be in a single paragraph, which saves space (this, obviously, relates to the above issue).
• Quote dashes present a typesetting issue in justified (right & left justified) text: the justification is done by expanding (and, in professional software, also by contracting) the size of spaces. But when the width of a space after a quote dash is changed, the dialogue can look ragged, particularly if you have very wide spaces in a line (because the starts of the text of each utterance don’t line up vertically). The solution would seem to be to not allow the space after the dash to change width, but I’m not sure whether InDesign and PDF allow for that.^† (Not sure what this says about my personality and/or perfectionism, since I’ve never heard anyone else complain about this, but is sure bugs me when things don’t align vertically!)

^⁎ our translation of the Synoptic Gospels managed to use four levels of quotes in one instance! That was a problem because the PT automatic checks only support three levels (I wonder if it wouldn’t be much work for them to add a fourth).

^† It seems logical to me that spaces should have two formatting attributes: breaking/non-breaking and stretchable/non-stretchable. The legacy solution for the former was to have a separate codepoint for non-breaking space. Apart from the fact that this means you have two identical characters in your character set, this also has the disadvantage that you really need to do the same for all the other spaces that a full character set would have (½-em space, ¼-em space, etc.) I’d be interested to know whether programs like InDesign implement a better solution – and also whether the Unicode standard has anything to say on the issue.

(Talking of Unicode, I used one of the extra asterisk characters available in the standard while typing this posting – that seemed the only way to avoid my footnote being formatted as a bullet point (according to the markdown standard this forum uses). BTW, is there markdown for superscript? [edit: there is, and I’ve now used it above])

(If anyone’s interested, I built a custom US-English keyboard for Windows with extra characters like “ ” β • ⁎† ‡ .)

Answer 2

First, check the information found in the topic “How do I specify alternate quotation marks for a project?”

Checking quotations that include the em-dash (2014) or other quote dash is very difficult and no solution will get everything correct. The best I have come to is to get the quotation rule settings as accurate as possible and then begin looking for anomalies. The alternate quotation system only works in 7.6. Having em-dashes in locations other than the beginning of a quotation will cause problems - that may need to be denied.

I have on occasion created a duplicate project and then did some creative editing to remove known issues so that I could locate true problems with the text.

Answer 3

CrazyRocky:

I certainty think I need to change quote dashes to use u+2015 ― but beyond that I am uncertain.

That makes sense in terms of distinguishing the characters by usage, though in the case of these characters it may not get you what you want visually. I just checked out 8 fonts, and the results were interesting:

• Charis & Doulos have them identical;
• Arial & Times NR have the quote dash a fraction shorter; TNR also has it thicker;
• Cambria has the quote dash quite a bit shorter;
• but SIL’s latest Gentium & Andika have the quote dash longer.

(I’d attach my test file, but I can’t see a way to do that.)

My assumption was that most people would want them to look the same. Of course you can change the U+2015 back to U+2014 at publication time if it gives the visual effect you want.