Unattached Tironian notation

[henryso]

There should be a way to render Tironian letters, most especially x and e, unattached to any other neume in nabc. The x (expectare), which has rhythmic significance and therefore no other way of notation, is more important than e (equaliter) which has melodic significance and can be expressed in the square notes. Additionally, it should be possible to render the x or e midway between the neumes above adjacent syllables.

[henryso]

I'm willing to try implementing this because I need this feature, but it involves changing the parser. I don't want to disrupt the work on the Laon font. @jakubjelinek, should I wait?

[jakubjelinek]

x and e aren't Tironian letters, at least from what I know Tironian letters are used just in Laon, x and e are called significant letters. What syntax would you want to use for the unattached significant letters? Is it meant just for [cs:] etc. only, or do you want to use it somewhere in the middle of neumes? If it is e.g. lsX only, the question is how to disambiguate between say vilst2 and vi lst 2... Plus it would be nice to be able to specify vertical position (hN) for the significant letters. For letters in between two neume groups, where would you specify it (after an nabc group and let it be split into a separate box by gregorio, or before the next one, again, how to specify pitch). Generally when I'm typesetting some significant letter that appears in the MSS in between, I put it after the previous note or before the next note, usually they aren't that appart from each other.

BTW, for Tironian notes, I'm planning to use lt instead of ls, so e.g. unltdx3 would be uncinus with Tironian devexum in its upper right corner.

[henryso]

I've definitely read books where the letters in St. Gall have been referred as "Tironian notes." I don't know if that's an incorrect generalization, or an obsolete phrase, or whatever, but it was definitely in more than one book, so it's at least arguably a name that can be used.

As for an isolated x, take a look at SG 390 (Harker Antiphonary), folio 40, "O Rex Gentium" at the bottom (url is http://www.e-codices.unifr.ch/en/csg/0390/040), between hominem and quem. Looking at the manuscript, the letter is clearly by itself, centered between the neumes above the vowels on either side. My suggestion for syntax is "ni" (meaning nihil) for the neume, so nilsx2. There are also examples of e sitting between neumes in the O Antiphons (as well as e's biased to one side or the other as well).

If we're talking wish list, it would be nice sometimes to nudge the letters slightly in some direction from their general 1-4/6-9 position, as the algorithm occasionally causes them to touch the neume.

[henryso]

Note: here's a book that talks about Tironian notes in at St. Gall:

https://archive.org/stream/MN5060ucmf_0/MN5060ucmf_0_djvu.txt

Edit: I have no idea if they are referring to the musical manuscripts or some other manuscripts from the abbey.

[jakubjelinek]

For O Rex Gentium, I just sticked it before the next neume as in http://www.introitus.cz/files/adventusdominicis.pdf pg. 37 (especially because there is a line break there anyway, if there is syntax for in between, would it go at the end of previous line or at the beginning of next line? Generally, for the placement of significant letters, I've been thinking about adding the OpenType accent points for each of the positions to each neume and similarly have such points in the significant letters. But it would probably be lots of work. Having some syntax for minor position adjustments might be nice.

For the terminology, I've been mostly reading about Tironian signs in http://store.saintmeinrad.edu/scholarshop/our-authors/fr.-columba-kelly,-osb/an-introduction-to-the-interpretation-of-gregorian-chant-vol.-1-foundations/c-25/c-91/p-384 pg. 180-188, that book uses the term added letters for the non-tironian signs.

For Laon, I plan to add to the 1-4 and 6-9 positions also position 5, inside the neume, especially augete appears inside very often (though I expect one will always need a glyph with that positioned letter, it would be hard to position it inside of the neume in TeX without knowing details on the shape).

There is another thing in SG that nabc isn't able to express, sometimes a wide significant letter (e.g. lsew, lscw and lstw) appears above multiple neumes.

[henryso]

If it's possible to have the letter unattached so that it can be spaced away manually with (e.g.) //, then the urgency for automatically placing the letter in between is considerably lower. For my needs, an unattached letter would be sufficient.

[henryso]

About the terminology, I'm not picky and was just trying to explain why I used the term, perhaps (probably?) incorrectly. I can change the title of the issue, if you like.

[jakubjelinek]

I think ni is reasonable, the question is if we add a character for it, or special case it in the parser; and also what height it is going to have. E.g. if it has height of 0 and width of 0, then I think nilse1, nilse4 and nilse7 might look identically, similarly nilse3, nilse6 and nilse9. It would need to be documented in the nabc documentation.

[henryso]

I think you would be the best to opine on whether it should be a new character. There is precedent for having spaces in a font, so I think it's really what makes the most sense to you.