Reason for this kind of double quotation marks

Lukas Horn

Hello,

I was wondering why the closing double quotation mark of courier looks wrong to me. I heard that type writer fonts based on the original letters of type writer machines are using " for opening and closing speech. Because it was a matter of using as less space as possible so they took just one letter for different uses (sometimes the same thing with zero and the letter O – it was the same on the keyboard). But the double quotation marks of the Courier of Adobe, Monotype and Microsoft are looking like this which is another thing (maybe it is a silly question but I don't know):

Kent Lew

This quote mark, which you view as a closing mark from the German perspective, is considered an opening quotation in English usage.

There is precedent for such “mirrored” quotes in several 20th century designs originating primarily in America. But this obviously leads to problems for German usage (and other related orthographies). So the practice is now best avoided as a default design.

This is also a problem in the original Verdana. When Font Bureau developed Verdana Pro for Ascender, the quoteleft and quotedblleft were fixed to address this shortcoming.

There are other examples.

Paul Miller

I always follow the positions and orientations in the Unicode Code charts where there is any doubt.

Florian Hardwig

There is precedent for such “mirrored” quotes in several 20th century designs originating primarily in America.

They are common in US sign painting and hence sometimes also referred to as sign painters’ quotes. Here are a few typographic examples for such mirrored quotes:

Uses tagged with “sign painters’ quotes” on Fonts In Use

Hrant Հրանդ Փափազեան Papazian

Mirrored quotes usually make more sense than calligraphic ones. But the biggest issue with quote marks is potential confusion with the apostrophe... which is one reason guillemets are generally superior.

John Savard

Kent Lew said:

There is precedent for such “mirrored” quotes in several 20th century designs originating primarily in America. But this obviously leads to problems for German usage (and other related orthographies).

I know that when I use the word "America", I mean the United States, but some people use it, following the practice of other languages, to mean the Americas. So as people will assume you are from the UK from your name, a potential for ambiguity exists.

Aside from that, to my mind, the original precedent for making quotes mirror-reflections of each other is an Italian design: Bodoni. There, quotes looking like a filled-in 6 and 9 from Roman typefaces are replaced by a reflected 9 and a 9.

I am puzzled as to what problems would be caused for German. A quote is in the wrong position to be confused with an accent mark. German does use apostrophes, but as the apostrophe and the closing quote are similar, how would changing the shape of the opening quote make any difference? (Oh, wait, I see my mistake. German quotes, as illustrated in the first post, use a closing quote pointed the opposite way from the apostrophe, so a change to mirrored quotes would indeed introduce that particular ambiguity.)

Hrant H. Papazian said:

But the biggest issue with quote marks is potential confusion with the apostrophe... which is one reason guillemets are generally superior.

That is a reasonable position to take. But at this point, ASCII is set in stone, as the beginning few characters of Unicode, and space on the typewriter keyboard is very limited, so I doubt there is much hope for change in the English-speaking world, at least.

John Hudson

I am puzzled as to what problems would be caused for German.

Modern German punctuation uses a raised left quote as a closing quote, and readers expect it to lean to the right.

André G. Isaak

John Savard said:

I am puzzled as to what problems would be caused for German. A quote is in the wrong position to be confused with an accent mark. German does use apostrophes, but as the apostrophe and the closing quote are similar, how would changing the shape of the opening quote make any difference? (Oh, wait, I see my mistake. German quotes, as illustrated in the first post, use a closing quote pointed the opposite way from the apostrophe, so a change to mirrored quotes would indeed introduce that particular ambiguity.)

Whether something creates ambiguity isn't really relevant. If something violates the conventions of a particular language, it is going to look wrong to speakers of that language even if they can easily figure out what is intended, just as a dollar sign constructed from a lower-case s and a slash would look wrong to Americans (just my s̸0.02).

Kent Lew

I know that when I use the word "America", I mean the United States, but some people use it, following the practice of other languages, to mean the Americas.

Fair enough. I will be more careful next time.

So as people will assume you are from the UK from your name, a potential for ambiguity exists.

I don’t know why anyone would make such an assumption about someone with a surname of Chinese origin. But if they are ever in doubt, they can check my profile.

André G. Isaak

Maybe he assumes anyone named 'Kent' must also be from Kent?

Nick Shinn

How about making the “German” left quote the default, and giving the “American” version a language tag?

<div>&nbsp; &nbsp;language ENG;</div><div>&nbsp; &nbsp; &nbsp; sub [quoteleft quotedblleft] by [quoteleft.alt quotedblleft.alt];
</div>

Nick Shinn

Here is an address, a riddle from before the era of postal codes:

Wood
John
Kent

John Savard

André G. Isaak said:

Maybe he assumes anyone named 'Kent' must also be from Kent?

Well, not only is Kent in Britain, but there's also that famous Grade fellow who makes movies.

Kent Lew

Nick —

I remember, several years ago, exploring various configurations of default with different OTL features and language tags when doing production for a Cyrus Highsmith face where he really wanted mirrored quotes for English default. And I recall sending versions to Indra Kupferschmid for testing to see if any of these approaches would be robust.

The problem with language tags is that they rely on an environment that tags text accordingly. Many do not, and certainly not in any automatic or reliable way. For instance, just because a user might have their system language set to English, doesn’t mean that text that they type in any given application will get tagged with an OpenType <ENG> language tag.

In the end, I wound up using a straightforward contextual substitution. Mirrored quotes are the default; and quoteleft/dblleft preceded by just about anything gets substituted by the conventional turned quote alternate.

To try to implement this as broadly and reliably as possible, I wound up registering it in {locl}, so that it is on by default and not turn-offable, but without any specific languagesystem designation, so that it gets registered to all languages and doesn’t rely upon the text being tagged in any particular way (since there are many other languages that use the „German“ quote configuration).

As long as an environment implements {locl} feature, it should work.

I’m not convinced that this is an ideal solution. But at the time, it seemed to satisfy Cyrus’s desire to have his preferred default for English, while satisfying Font Bureau’s desire to have it work for European clients without complaint.

Here was a test using the InDesign feature where language-tagged text gets dumb quotes converted on the fly to the preferred localized format.

John Savard

Interesting. I thought there were other languages that preferred the format shown for Slovenian... and I'm rather sure that Russian is wrong as shown there. Whatever quote format is preferred for Russian, it isn't the nesting of two different kinds of quote.

Oh, wait, they're doing that for all the languages that use guillemets. So the interior single quotes should just be ignored, I guess.

Nick Shinn

Here is the answer to yesterday’s riddle:

John Underwood

  Andover
    Kent

Hrant Հրանդ Փափազեան Papazian

John Savard said:
ASCII is set in stone, as the beginning few characters of Unicode, and space on the typewriter keyboard is very limited, so I doubt there is much hope for change in the English-speaking world, at least.

Although the overloading of quotes/apostrophe on one key (even more than the ASCII legacy) is a real problem, there is actually an apostrophe character in Unicode. We do need to greatly improve support for it however.

But most of all here:
— A type designer can decide to make rational (versus chirographic) quote marks. To me ideally where the opening ones are traditional (giving the reader their bearings) while the closing are vertically flipped (to avoid confusion with the apostrophe).
— A typographer can elect to use guillemets instead.

People who don't give readers enough credit for being able to adapt to either of the above are probably stuck in the conscious/display realm of typography.

BTW coincidentally the day after my post above I saw a publication (the Norwegian Air in-flight magazine) using both guillemets and mirrored quotes!

Kent Lew

Oh, wait, they're doing that for all the languages that use guillemets.

I neglected to explicitly mention that since I was testing my turned-quote substitution algorithm, for languages that use guillemets to mark primary quotation and hanging quotes for secondary, I set both to make sure the interior quotes worked adjacent to either opening or closing guillemet.

(It was not intended as a consistent or comprehensive documentation of international quotation styles.)

Denis Moyogo Jacquerye

Unicode had a character specifically for the mirrored double quote:

‟ ‎201F DOUBLE HIGH-REVERSED-9 QUOTATION MARK

= double reversed comma quotation mark
• has same semantic as 201C, but differs in appearance

Adam Jagosz

@Hrant H. Papazian The second example in the first picture looks like candidate for the “Blog” of “Unnecessary” Quotation Marks

Lukas Horn

Thanks for your opinions and all the nice information!

Vasil Stanev

Not exactly on-topic, but I feel I should post this here for anyone interested: 

Florian Hardwig

The problem with such maps is that they are simplistic. It’s correct that „…“ is the standard form of quotation marks in Germany – in office context and also in many newspapers. As soon as you start looking at printed books, you’ll find that the predominant preference there is »…«. I’m sure such nuances exist in other cultures, too.

Vasil Stanev

Florian Hardwig It would be super interesting if each culture showed these nuances.

Adam Jagosz

This kind of thing is what pisses me off about Unicode. Since Unicode is about characters and not glyphs, and the glyph at U+201C can be used for both opening and closing quote, ideally it would be split into two separate codepoints. Similarly, wouldn't it make sense to have four codepoints for guillemets?

But I bet if they could design this from scratch, they would assign one codepoint to a generic opening quote and one to the closing quote and then let vendors worry about providing localized and stylistic and contextual variants.

Khaled Hosny

In general, Unicode avoids duplicating identical characters unless they behave differently. Space and no break space are duplicated since the later does not allow line break. Colon and ratio are duplicated because they are spaced differently (in math at least). But the various quote marks behave identically (as far as Unicode is concerned), whether used as open or close marks.

Nick Shinn

If there were a separate language code for North American English, then we could make the <quoteleft> glyph (2018) look like an apostrophe, especially for display styles.

That would make sense, because we rarely use <quoteleft> as a quote mark, yet it appears everywhere bass-ackward, masquerading as a word-preceding apostrophe (e.g. rock ‘n’ roll, summer of ‘69, etc.), thanks to so-called “smartquote” apps.

Actually, I have done that for proprietary typefaces for North American companies (although with the strange sensation that John Hudson and Thomas Phinney were standing behind me going tut-tut-tut).

Adam Jagosz

Khaled Hosny said:

In general, Unicode avoids duplicating identical characters unless they behave differently.

Thanks for the clarification. That seams reasonable. I guess I forgot to add ;smirk; to the end of my last comment.

Henning von Vogelsang

Florian Hardwig said:

As soon as you start looking at printed books, you’ll find that the predominant preference there is »…«. I’m sure such nuances exist in other cultures, too.

Were these books printed in Switzerland? Because ever since Napoleon left, the Swiss adopted guillemet quotes from the French.

Florian Hardwig

No, I’m talking about German publishers. The Swiss indeed use guillemets, too, but typically the other way around («…»), like the French.

Nikola Kostic

Vasil Stanev said:

Not exactly on-topic, but I feel I should post this here for anyone interested: 

There's a mistake in this graph. In Serbia we use the same standard for quotes as Germans do.