Transcription of [German] double hyphen?

nina's picture

I'm currently working out the design of a German-language scholarly volume that quotes some texts from the 17th century. Now some of those quotations contain historically typical doubled hyphens (see Wikipedia entry on double hyphen).
In their transcriptions of quotes/citations/titles, the authors have opted to represent this by using an equal sign ("=") in place of the hyphen:

Die in einem der Texte enthaltene Anspielung auf den 1599 in Leipzig erschienenen Gemüths=Spiegel

I am doubting this is a good idea – mainly because the hyphen essentially is a hyphen, and its doubled form just a historical stylistic variant of the same base character; whereas the equal sign actually means something else. The "=" of course (and especially to typographic laymen, I guess) looks closer to the original than a single hyphen, but I wonder if it doesn't actually twist the original more.

Alternately, it looks like the double hyphen actually has a codepoint (U+2E17). It is however not present in the font that I'm looking at for this. I wonder if its usage here would be imperative.

Does anyone have experience or opinions on this? I will see the editors in a couple of days, but I'd love some opinions from the typographic side. Thanks.

Indra Kupferschmid's picture

The authors “transcribed” the blackletter into roman letterforms. So I think it’s okay to use a normal roman, single hyphen instead of the doubled. Maybe you can choose a typeface with a slightly angular one.

nina's picture

Thank you Indra. That sure seems a good angle.

"Maybe you can choose a typeface with a slightly angular one."
Hmm… why?

Queneau's picture

I think this is an interesting question. Also because these old texts would include both character and grammar that is different from the current one. The long s for instance is not being used anymore except for the ligature ß. Would you keep the long s or use the modern version. Is Historical/period accuracy the most important or the current relevance of the text (using the text but updating the grammar and the form). My opinion is that it is nice to take into account the fact that a text is historical, but one shouldn't take it too far. In my view the content is the most important thing, and the form can help to make the content more understandable/readable. Using old text from the same language is a bit the same as using foreign language texts: it needs translation to be understandable for most readers.

As for the short answer: I would use a single hyphen, as it does not structurally alter the content of the text, and minimizes the confusion a double hyphen might cause.

Indra Kupferschmid's picture

"Maybe you can choose a typeface with a slightly angular one."
Hmm… why?

Cause those are verly clearly hyphens and not dashes and it resembles the angular double hyphen as shown on your wikipedia example.

nina's picture

Ahok. Cheers!

Jeffrey – how close they want/need to stay to the original is exactly a question I'll have to discuss with them too. I think how one weighs this depends a lot on the «scholarliness» of the publication – in this case, being true to historical sources is crucial; but I'm beginning to suspect that orthotypography (if not orthography) can be «updated» (which would speak in favor of a normal hyphen). Anyway I'll report back :)

dan_reynolds's picture

Can you just add a double-hyphen to the font? I guess that depends on the font in question and its license, of course. I added a bunch of characters for the transcription of medieval Latin to a font recently, and the double hyphen was one of the glyphs the client wanted to have.

nina's picture

Yeah, I just checked and it should be OK (after written consent from the foundry). It's great to know that if the client wants one I could make one.
Interesting to hear that your client insisted on the double. Was this also on the grounds of staying true to the original source?

dan_reynolds's picture

I am not really up on this level of scholarship, I must admit. But I have been told that, in academic texts, old texts being discussed are now transcribed, not translated or converted. So in the case of medieval Latin or older German (in Fraktur, to boot…), at text will be transcribed exactly from the image into the same typeface as the article (or whatever typeface the author/editor/typesetter/etc) have chosen for transcription. This means that the font(s) for the article/book/etc must have every glyph that needs to be transcribed. I guess that (certain?) scholars view the double-hyphen as being unique. Certainly Latin abbreviations are not written out when you transcribe.

I hope to have images of the project I'm talking about by the end of the month. I'm sort of interested to see how it all worked.

Florian Hardwig's picture

Did the original font have ‘single hyphens’, alongside the ‘double hyphens’, used for different purposes? I guess not. A hyphen is a hyphen.
Would you add diamond-shaped dots to your font because the original featured those? If not, why? Because the diamond-shaped dot has no Unicode point? Hardly. A transcription is not a facsimile.

I must admit that I do like Dan’s suggestion of having a double hyphen, though.

dan_reynolds's picture

Well, I assume that transcribed texts are separated from the main body text in documents like this. So you would never confuse the two. So it would be ok for the hyphens in one section to be doubled, whereas they would otherwise be single.

Of course, the transcribed text will likely have lots of other weird glyph in it, too. And be in an archaic language.

nina's picture

" A transcription is not a facsimile.
I must admit that I do like Dan’s suggestion of having a double hyphen, though."

I must say I agree with both of these sentiments.
My client would be OK with a normal hyphen, but was very happy when I proposed to make a double hyphen, so I might just go for that. There's also the fun of making an extra glyph. :-)

BTW there is an original transcription (of a manuscript) in this book and it's very tricky to typeset – a text genre that definitely breaks «normal» conventions; for example all the characters that aren't actually in the manuscript but were added by the editor (to amend archaic abbreviations etc) are italicized, so you get lots of Roman/Ital breaks within words even. Kerning fun!

nina's picture

Dan, maybe you can share some pix of the project you mentioned when it's out? Always interesting to see «uncommon» glyphs in use.

Design question – do you people think it's OK to angle the double hyphen even when the standard hyphen is flat? (Feel free to yell at me if you think I should take this to Critique.)


To my eyes this looks quite fitting indeed (I'm glad I'm not using a Garalde) but I'm still so much of a learner.

eliason's picture

Hate to kill the fun, but this looks rather mannered and unfitting to me.

dan_reynolds's picture

Nina, of course I will share photos once I can! I am already shivering with anticipation to see the results myself. Anyway, here are two images similar to your text… one with the standard hyphen, and the other with the double hyphen.

Florian Hardwig's picture

For reference – here are a few non-blackletter double hyphens:

Store front lettering, Leipzig

Bold serif caps on memorial plaque, Braunschweig *

DIN street signs, Leipzig

*) In case you wonder about the very low position here, note that this once was en vogue for single hyphens, too – cf. this photo by Fritz Grögel.

nina's picture

Interesting to see those flat double hyphens. Hmm.
For some reason it looks like («normal» whiteletter) fonts that support U+2E17 tend to have it at an angle (see here), except when they have strange wiggly designs (I wonder where that came from).
Looking at Dan's examples, I can't help but feel the angle on the doubled one is really really nice, also when the single is flat. Maybe also because it's closer to the blackletter form, where this character is so much more common; and because it's less similar to an equal sign, so less like anything else.
Craig, I wonder, is it the angle you don't like – or is mine just a bit to, dunno, stiff or something? (I realize now it's probably a bit high.)

dan_reynolds's picture

>Looking at Dan's examples, I can't help but feel the angle on the doubled one is really really nice, also when the single is flat. Maybe also because it's closer to the blackletter form

Well, in this case, the client asked me for "a double hyphen, like those you see in blackletter faces." So, I can't take the credit.

kentlew's picture

Nina -- I think your angle is too steep.

Try taking an angle perpendicular to the italic slope, then flip horizontally. (I hope that makes sense.)

See what that looks like.

nina's picture

Ah! Thanks Kent – this does seem better.


I'm not sure, maybe it seems a wee bit too flat now? But it sure is more pleasing by being more moderate.

eliason's picture

Craig, I wonder, is it the angle you don't like – or is mine just a bit to, dunno, stiff or something?

No, my reservation was more along the lines of it looking like a wrong font, since I don't associate that glyph with that kind of font.* I would just expect a more conventional hyphen so this sort jumped out as looking kind of affected - analogous to a grotesque with a ct-ligature, perhaps.

But Florian opened my eyes to the variety of letters that can be accompanied by a double hyphen in German, so that softens my objection.And I do think that Kent's suggestion has made it more pleasing.

*What font is this, by the way? And how would you classify it if not a Garalde?

nina's picture

"What font is this, by the way? And how would you classify it if not a Garalde?"
Hm, maybe I'm not using the terms right? This is Kris Sowersby's Newzald, a Fleischmann/Rosart inspired design, so I'd rather call it transitional. Please correct me if I'm wrong. The reason why I said that BTW was that to my eye, the darkness and the relatively rigid «verticality» of the face is more reminiscent of blackletter than a soft oldstyle face would be, a quality/association the double hyphen nicely clicks with. But maybe that's just me (I'm one of these people who see blackletter everywhere), and now that I've seen it, I think Dan's sample shows this can work nicely in other styles too.

dan_reynolds's picture

I'm also one of these people who see blackletter everywhere…

Florian Hardwig's picture

Oh, my intention with those random samples was not to make a case for flat hyphens as a rule, but – as Craig states – to show that this form has been used with all kinds of styles (especially in lettering that is, not so much in type). Flat hyphens can be appropriate for static forms like DIN and bold Didones. I concur with you that an angled form looks better with a somewhat dynamic style like yours, especially in the italics.

David Waschbüsch's picture

Obviously I'm not anywhere near as qualified as my previous speakers but I recently did a angled double hyphen for my font. Despite it's origins I like setting a double hyphen when you have to seperate a compound word like "Fluss-Schifffahrt" exactly where the single hyphen is.

Angled because I won't have it confused with the equal sign & double because it is kind of doubled there.

eliason's picture

David, that's an interesting and persuasive case.

Nina, if Fleischmann is Transitional (on which many would agree), I think I would say it's on the Garalde side of Transitional, so I was wondering why you were thankful not to be dealing with a Garalde. Your explanation made sense. (I find it difficult to fit that whole strand of 17th/18th-c. "Dutch-English Oldstyle" into Vox categories.)

dezcom's picture

.

kentlew's picture

FWIW, I don’t consider Fleischmann or Rosart to be Transitional. They are Dutch Oldstyle, in my view. The sort of thing that leads to Caslon and Kis. I might be inclined to call this period Dutch Baroque, because of mixed stress and idiosyncratic terminals (and because I am musically minded).

I would say that, with Newzald, Kris has introduced a little Neoclassical quality to his interpretation of the underlying style, with more regularized upright stress and greater sense of symmetry and balance. In that sense, Newzald is more Transitional, perhaps, than its inspirations.

But we probably don’t need to go down the classification rabbit hole here.

Indra Kupferschmid's picture

I like your steeper angular one better. More distinct.

dan_reynolds's picture

Wasn't Kis active in Amsterdam a few decades before Fleischmann and Rosart? And doesn't Caslon's career overlap with Fleischmann and Rosart's careers?

nina's picture

Hm – could one of the Germans (or Germanicized ;) confirm if the term Barockantiqua translates/maps to Transitional Style – or are those different?
Because I would call this baroque rather than oldstyle, but now I'm not sure about the terminology. (Sorry to be asking such newbie stuff :-\ )

eliason's picture

Yes, I think "Barock=Antiqua" from DIN 16518 maps onto "Réales" from Vox/ATypI, which is the category that is most often rendered as "Transitional" in English.

The term "Transitional" may be what's complicating matters here. I'm actually currently writing a piece exactly on that subject.

On another note, my understanding is that "oldstyle" would usually encompass baroque types rather than be distinct from them, at least as the term was used originally to separate "modern" (didone) styles (and sometimes "transitional" styles meaning fonts like Baskerville's) from previous romans. But I guess you take it to mean Venetian/French pre-baroque styles only?

nina's picture

"Barock=Antiqua"
:=)

"'oldstyle' would usually encompass baroque types rather than be distinct from them"
Aah! Yes, I tried to use "oldstyle" as mapping to the German "Renaissance-Antiqua", which would include Garaldes and Venetians (or «humanes» & «garaldes» according to Vox/Atypi I guess), pre-Transitional. Is there no «exclusive» English term for these?

eliason's picture

I can't think of one.

kentlew's picture

Dan — I think you’re probably right. They are all very close in chronology. I always have to look it all up to sort them properly.

But I think of Kis as a development out of the Dutch tradition, because he was trained in Amsterdam (Voskens, if memory serves) but then went on to evolve independently, and his later work seems to share more in common with Fournier (in my eyes), making his types more of a stylistic bridge to neoclassical and modern (again: my personal take on it, not any historical truism).

Caslon is said to have been imitating Dutch models. If not Fleischmann, then precursors to Fleischmann. Maybe more Voskens or Van Dyck. Again, I think of Caslon as a development coming out of the Dutch style — influenced by, but then evolving into its own.

But I often create my own history. ;-)

dan_reynolds's picture

I definitely agree that Kis and Caslon both develop from Dutch Old Style traditions! ;-)

Bendy's picture

I love the way Nina's threads always generate so much food for thought, and how much more there is for me to learn about.

nina's picture

Same here! :-)

Nick Shinn's picture

The steepest (whiteletter) hyphen I have come across is in Polyphilus.
Not quite so extreme: Italian Oldstyle (Goudy).

mekka's picture

Take a look at Unicode’s nameslist.txt comments for U+2E17, where, among other things, you will find:

• hyphen in Fraktur text uses 002D or 2010, but with a '2E17' glyph in Fraktur fonts
→ (hyphen-minus - 002D)
→ (equals sign - 003D)
→ (hyphen - 2010)

David Waschbüsch's picture

Okay I'm sitting here in this meeting and this one woman across the board is wearing this t-shirt with an old Pepsi-Cola logo. Aaaaand guess what: There's a double hyphen in it. ;) Despite it isn't a blackletter logo type.

Florian Hardwig's picture

Excellent!

dezcom's picture

It may also be part of the Coke-Pepsi rivalry where they use an equals sign to indicate that Pepsi=Cola

eliason's picture



These are from The Inland Printer 1907-8.
Interesting that the double-hyphen accompanied tall fonts in these instances, since you mentioned "verticality," Nina.

nina's picture

Wow, nice finds. Was this common/frequent in English?

BTW, I just noticed that the usage that David (einfach ein) mentions above seems to be official practice of the Merriam-Websters, at least according to Wikipedia:
"In Merriam-Webster dictionaries if a word is divided at the end of the line, and the division-point happens to be a hyphen, it is replaced with a double hyphen to graphically indicate that the divided word is normally hyphenated, for example cross⸗
country"

Bendy's picture

Wow, I'm amazed. It's not something I've ever noticed before, but I guess now the double hyphen is on my radar, I'll see it everywhere.

eliason's picture

Was this common/frequent in English?

No, not common generally (I had to do some looking to come up with those examples). But now I wonder if it was a convention that attached to condensed fonts like those.

Your wiki link also mentions the Waldorf=Astoria.


Notice the double hyphen with the script and the single with the logo's blackletter!

Hannes Famira's picture

Glyph name 'doublehyphen'?

Hannes Famira's picture

Just in case I didn't manage to completely kill off this thread, what OT feature would make sense here? Any suggestions? Anyone? Bueller?

dan_reynolds's picture

Nina, here are the Malabar double hyphens as they are used in Gutenberg-Jahrbuch (for which I drew them). For transcription. That image lives at Flickr here, and I have a set of more photos from the book here.

riccard0's picture

Orthogonal: in Italy the (straight) double hyphen is (at least it was to me) taught in school as the proper hyphenation symbol to use in handwriting.
(I haven’t re-read the entire thread, so, sorry if it was already mentioned)

Syndicate content Syndicate content