The word hyphen/huphen is late Greek and is translated as "hyphen," so its use predates Gutenberg...

The hyphen, in both single stroke and double stroke forms, pre-exists moveable type and, like other aspects of Gutenberg's typography, is found in the manuscript tradition that he was emulating. Gutenberg used the double stroke hyphen, which seems to be most common in the 15th Century.

Remember also that the hyphen serves two functions: to indicate a word broken at the end of a line, and also to indicate compound words. The Hebrew equivalent, the maqaf was definitely in use by the end of the first millennium AD, as it is a standard feature of the Masoretic codices.

John’s right of course.

Double hyphens appear in MSS from the 14th century. Not to be confused with a similar looking double horizontal stroke that denotes an abbreviation. The single hyphen (or a hairline stroke) dates back to at least the 12th century. Can supply a scan if required.

@oldnick
>The word hyphen/huphen is late Greek

Good observation. The question to me, then, is whether the Greek term is a direct translation - a graphic arts term - or something borrowed from another sphere and applied to type as so often happens when you need to coin a term. (Where is my old friend Costa Stamos when I need him most? Gotta check Facebook.)

@john hudson
Thanks. I suspected it had roots.
I suppose it *is* fair to say - though a little tautological - that Gutenberg was the first to use the hyphen in print.
Yay? Nay?

Anybody else with info? Greatly appreciated. Gotta go Google Masoretic codices, now.

I have a few examples (French and Latin) in Codices illustres (Tashen) that date back to 1230. The hyphens were inserted anywhere in the word.

@John Boardley
>Can supply a scan if required.

If you would be so kind, I'd love to see it. Not sceptic, just inspired by such sort of things :) Thank you very much!

I'd love to see a scan of that, too!

> Gotta go Google Masoretic codices, now.

L codex:

re Greek origins, see OED 1st s.v. 'hyphen'; LSJ: derived from adverb ὑφέν = in one; [cup] 'written below two consecutive letters to show that they belong to the same word'

@david hamuel

Thanks. I was looking at the codices elsewhere but didn't really know what to look for.
My Hebrew is worse than rusty, but it looks like the "hyphen" is being used as a connector to re-inforce what's modifying what within certain types of phrases.
Wow, this is dredging up knowledge I forgot I had from my days as a student at an Orthodox temple. (My Dad had a lousy voice but he was a proficient cantor - he would pinch hit sometimes on Saturday mornings when one of the more observant (and therefore more appropriate) guys were out sick. He was an unlikely chazan (cantor) - a very secular, worldly guy - who had received a huge amount of training in his youth and had the whole repertoire down cold. Never got rusty, either. Weird - kind of like a photographic memory.)

So, can it be not a typographic but a musical cue? Saying, "sing as one phrase in one breath"? It's certainly a connector - a "hyphen" - but it's purpose makes more sense to me that way.
Waddaya think? Anybody know for sure?

@crych
Thanks, as well. That settles that.

rich

> So, can it be not a typographic but a musical cue?

musical...? definitely not at the beginning. The Tiberian system was developed in several phases; first — the accents above the words ("copied" the Babylonian tradition); second — the accents under the letters; third — the vowels.
We don't know all the rules of the maqaf. The most common rule: word preceding the maqaf changes in two ways: the vowel changes as well as its accent/ta'am.

This thread inspired me to do some Google searching which ended up piquing my curiosity about the older Babylonian vocalization and Palestinian vocalization used for Hebrew instead of the current Tiberian vocalization.

I found that one program which extended TEX to handle Hebrew allowed the use of alternate vocalizations: Makor.

But sadly, it was built to use Omega, which ended up being abandoned. There is current software for Hebrew typesetting, but apparently nothing currently available is fully the equal of Makor. Since Mem is based on Lambda, and Lambda is a successor to Omega, I suppose a port might be remotely possible...

John

The files in makor2 are rather messy. There are missing utilities, missing fonts (omsehe.tfm is even missing from texlive2009 even if omsehe.pfb is there), some fontinst files needed to be corrected, the ezra font was coming out badly... and I eventually had more clean up to do than I expected; the result is here: (makor2.tgz (3976 KB).

Just untar in kpsewhich -var-value TEXMFHOME, and apply texhash.

The test file with archaic vowel points is .../texmf/tex/makor2/doc/archaic.tex. With texlive 2009 on my mac, all I need to do to get the pdf is to execute:

aleph archaic
dvipdfm archaic

The same works with esther.tex, talmud.tex, vocal.tex, siloam.tex, siloamalt.tex, ohalph.tex, jabberwock.tex, german.tex, trpnv.tex. I found no no way to process latex files.

Michel

For LaTeX files, just use lamed instead of aleph.

@david hamuel

We don't know all the rules of the maqaf. The most common rule: word preceding the maqaf changes in two ways: the vowel changes as well as its accent/ta'am.
Makes sense.

As far as it doing double-duty as a rudimentary form of musical notation, (does anybody remember Mitch Miller and "follow the bouncing ball"?) one thing seems certain: if not chanted, these texts were read aloud because the general population was largely illiterate.

It just makes a lot of sense to me - I'm still speculating, though, of course - that the end of line hyphen was important as a visual cue to the reader saying, "hey, there's more of the word (or phrase) coming up on the next line."

The other thing I was curious about was to what extent the value of paper played a part in keeping the line lengths the same. In other words, did justification have roots in using paper, or vellum, or parchment, or whatever, to it's full value.
I assume it was very costly to produce.

Rich

Rich, Joshua R. Jacobson, in his standard word Chanting the Hebrew Bible, has this to say about maqaf: ‘A verse that contains too many accented syllables has a “choppy” rhythm. One solution for this problem was to join two or more short words together with makkef, a sign resembling a hyphen. The resulting compound word has only one accented syllable.’ He goes on to give some examples of using maqaf to avoid accents on consecutive syllables and of proclitic and enclitic words.

So yes, maqaf does have a role in the melodic interpretation of chanted text, but perhaps not in the way you imagined. When we talk about musical notation, most often we have something like that of the evolved western European tradition, a system indicating particular pitch and duration, in Jewish liturgical chant the accent marks — te‘amim, sometimes called cantillation marks — indicate melodic tropes of relative pitch and, to further complicate matters, the actual melodic trope associated with a mark will vary according to the day (e.g. regular shabbat vs high holy day) and also across different traditions. So, for example, following the same text with the same accentuation, an Ashkenazic cantor on a high holy day will be singing different melodies from a Sephardic cantor at a typical shabbat service. I would call the accent marks musical mnemonics rather than musical notation. The cantor has to learn the melodic tropes and how to apply them to text, he doesn't ‘sight read’ the accented text in the way that a musician can follow modern western musical notation, knowing the pitch and duration from the notes on the page.

The Byzantine and Latin chant traditions also use relative rather than fixed pitch, and historically one can trace notation from something similar to Hebrew accentuation, i.e. marks directly applied to the text indicating melodic tropes, to separate notation on a staff that is the precursor of the modern system.

The closest parallel to the Jewish model of melodic interpretation of marks in the Christian tradition may be prose chant, as used for instance to sing the prose readings during Mass or Divine Office. In this, punctuation marks are interpreted melodically: typically : and . will each indicate a short melodic cadence to be applied to the preceding syllable(s), and there will be a longer final cadence on the last syllables of the passage, which is otherwise sung recto tono. As with Jewish chant, the priest or lector has to know the melody of the cadences and how to apply them to the text, rather than following a musical notation.

>texts were read aloud because the general population was largely illiterate.

This isn't correct. Among Jews, there was universal education for boys by the second century, so the vast majority of Jewish men were literate. There are a bunch of other reasons for reading aloud, including ritual reasons, the expense of manuscripts, the displacement of Hebrew as a spoken language, etc.

Rich, in the classical period all reading was aloud. Silent reading is a relatively late development.

@jh

The cantor has to learn the melodic tropes and how to apply them to text, he doesn't ‘sight read’ the accented text in the way that a musician can follow modern western musical notation, knowing the pitch and duration from the notes on the page.

Yeah, I didn't mean "notes" or "pitch" I meant phrasing and accentuation.
Exactly.
Which group of words get sung in one breath, that kind of thing.
There are very precise equivalents in musical notation for that, too, of course.

@william berkson

Actually, in my head I *was* excluding Jews when I wrote that. Literacy was always presented to me as a religious requirement - growing out of the story of Joseph in Egypt.
However, literacy in what language? Hebrew was already dead by 600 BC. My understanding is that, as far as ancient texts, scholars were always needed. And support for them by the larger community was considered, likewise, a religious obligation.

English prof Richard Lanham - author of Revising Prose, a good book - believes that electronic text is, in many respects, a return to the oral style. I believe he's correct. The Official Style that emanates from government buracracies, corporate boardrooms, and (sorry Bill, but it's largely true) academia is stupid enough in print but onscreen, happily, it just comes off as ridiculous. Click.
(But I'm getting far afield, thanks for the feedback all. I have an article I'm being pushed to finish but I'm making a note to double back and edit that Wikipedia entry. It's misleading, really.)

Any more input, hey, we can keep going... but I gotta get ready for Typecon. Flying tomorrow.

Great stuff. Thanks again, Rich

>Hebrew was already dead by 600 BC.

That is not now the accepted view among scholars. As the wikipedia article on Hebrew explains, that old hypothesis was refuted by the Dead Sea scrolls. The general view now is that Hebrew as a living language lasted into the second century CE.

Richard,

> that the end of line hyphen was important as a visual cue to the reader saying,
> "hey, there's more of the word (or phrase) coming up on the next line."

What about makaf between two words that are written in the middle of the line/pagraph?

If you read the Talmud you'll find guidelines that are very similar to modern day typography, typesetting etc etc. For example, it is mentioned about the spacing, kerning, line length.For example, line length would not exceed three times the word למשפחותיכם (to your families) because with longer lines the reader might lose the lines; even Rashi in his commentary wrote about that issue (line length) !

Again, things were developed in stages. Though we don't have any document that states these developments there are several cues that show that musical function was developed later on. Early Babylonian MSS have ta'amim above words that don't bear the accented syllable. That said, their main goal was the division of the verse. In a later stage the ta'amim appear above the accented words.
The Tiberian system improved itself by adding more ta'amim, for example meshrtim (servants) which indicates that their functions are more musical. Probably at the beginning the reading was emphasizing the intonation, recitation (diklum) and the distance between intonation and melody/musical was not too far apart. And probably along the way came the maqaf (which has a grammar & musical role).
Actually the maqaf is not that important. Because if you have a word without ta'am it is basically supposed to be read as if there is a maqaf.

Given the evidence of the New Testament, and, for that matter, the use of Aramaic (often called "Chaldaean" in that context) even in parts of the Old Testament, I'd be inclined to suspect that the evidence of the Dead Sea Scrolls is enough to support a conclusion that Hebrew continued to exist as a literary language for a significant period of time - after all, knowledge of the Hebrew language was treated as a universal religious obligation by the Jews, the bar mitzvah continuing to the present, and having existed long before the revival of Hebrew by Ben-Yehuda - but not enough to conclude that it was a spoken language used in daily life right up until 200 A.D..

But I'll have to admit that we don't really know, and the Wikipedia article does note that the issue is still very controversial among scholars - it hasn't really reached a settled consensus.

John (Quadribloc) evidence is not just religious texts. It is letters from Bar Kochba, who led the rebellion in 135 CE and are written in Mishnaic Hebrew. The Encyclopedia Judaica article "Hebrew" takes this as pretty conclusive evidence of spoken Hebrew being alive at this time, and refuting the early view of Geiger that Mishnaic Hebrew was an artificial construction based on Aramaic and Biblical Hebrew.

There is also a story in the Talmud of Yehudah haNasi (c. 200 CE), in the Galilee, asking his maid about the meaning of a Hebrew word—indicating that its use was dying out, but some people had still grown up with it, probably in Judea.

That some people spoke Hebrew in the second century is I think generally accepted. The "hot debate" according to the Wikipedia article is what exactly was the mix of Hebrew, Aramaic, and Greek, and where exactly these were spoken.

Concerning the use of Hebrew as a spoken language, here is a citation from Yosef Ofer, The History and Authority of the Alleppo Codex (pdf 182K):

Yet another source provides surprising information about the use of Hebrew by the people of Tiberias. In the tenth century a Hebrew grammarian of the city wrote a work that has been preserved in part in the Cairo Genizah. He relates that he studied the accent of the people of Tiberias, especially their way of pronouncing the letter resh when reading the Bible and when speaking Hebrew. To that end he would spend long hours in the city streets and squares, listening to the speech of ordinary people, in Hebrew and in Aramaic (Allony 1970, 98-101). This description, which is written in Arabic, indicates that in the tenth century Hebrew was actually spoken in Tiberias and was not only a literary and liturgical language.

The reference for Alloni is Leshonenu, 1970, 75-105, 187-209. I can't copy paste from the pdf (and with hard to find latin diacritics I won't type it), so here it is in Hebrew (copy pasted):
אלוני, נחמיה עלי בן יהודה הנזיר וחיבורו "יסודות הלשון העברית" לשוננו לד (תשל) 75-105, 187-209 1970

Thanks, Michel, that's fascinating. It seems to strengthen the idea that the current system of nikkud, which is said to have been developed in Tiberias, in fact has roots in a continuous tradition of Hebrew speakers. I found this interesting as well: http://en.wikipedia.org/wiki/Tiberian_Hebrew

DH>If you read the Talmud you'll find guidelines that are very similar to modern day typography, typesetting etc etc. For example, it is mentioned about the spacing, kerning, line length.For example, line length would not exceed three times the word למשפחותיכם (to your families) because with longer lines the reader might lose the lines; even Rashi in his commentary wrote about that issue (line length) !

Those guys covered all the bases, didn't they? Hah!
(BTW - I know I'm not the first to observe that the Talmud was, in a way, a form of one-dimensional hypertext, too.)
That's really interesting.

Interesting, too, that Hebrew seems to have survived, at least in some form, within certain communities as a language of daily life.

Interesting thread. Please update the Wikipedia entry, it's embarrassing for scholars of scholarship.

@Propagandum

You're right. I'm going to log into Wikipedia and propose a change. (Something new for me - but I've been meaning to create a page or two about some other things for a long time anyway.)
I'll cite this thread.

BTW - TO ALL CONCERNED

I recently checked on the progress of Mathias Nater's Hyphenator project for the first time since I wrote about it both here and in an article on AListApart called The Look That Says Book and I'm happy to report that it's matured beautifully since then.
Very very smart work.
(I really must convince Garrick Van Buren of Kernest/Konstellations to incorporate it in an upcoming issue of Konstellations. Konstellations deliberately avoids scripted solutions, generally, but for this perhaps an exception should be made.)

It's still going to be a long while before native Hypenation & Justification will be ubiquitously found in browsers. So in the meantime, it's nice to have a first-rate tool like Hyphenator that does the job and does it well.
Kudos to Mathias for sticking with it and working as hard as he obviously has.

Richard Fink
Blog: Readable Web
Font Director: Kernest/Konstellations