Missing some Dominican neumes

[olivierberten]

See Salve Regina, for instance...

[olivierberten]

The first one is actually an epiphonus (or pes deminutus, or podatus deminutus) as represented in Solesmes editions (1908 and 1961 Graduale Romanum, Liber Usualis, etc.)

[celide]

Please add them. :-)

[eroux]

Can someone propose a gabc notation for them?

[celide]

I would say: vg or any other vNOTE

Tomasz Grabowski OP
prezes fundacji

kom. +48 694 480 613
tel. 12 430 19 34
e-mail: t.grabowski@liturgia.dominikanie.pl

Fundacja Dominikański Ośrodek Liturgiczny
ul. Dominikańska 3/11, 31-043 Kraków
www.liturgia.dominikanie.pl

NIP: PL 6762419775, KRS 0000354036
47 1140 2004 0000 3802 7522 9738

Wiadomość napisana przez Elie Roux notifications@github.com w dniu 12 sie 2014, o godz. 20:46:

Can someone propose a gabc notation for them?

—
Reply to this email directly or view it on GitHub.

[rpspringuel]

The problem with that notation is that 'gvg' becomes ambiguous. Would that be the same as 'gv!g' or 'g!vg'? I'd prefer something that doesn't introduce that ambiguity if possible. That would, in my mind, mean keeping the note shape modifier after the note name as is currently done for everything except the initio debilis. Perhaps then using a 'u' for a vigra that appears on the left side of a note ('u' and 'v' are similar letters in shape and alphabet position so the similarity in their function is logical).

It also seems like something to indicate that the vigra goes up (instead of down) is needed to reproduce all the indicated notes. Maybe then '^' and 'n' (as these symbols are upside down versions of 'v' and 'u')?

The highlighted notes would then be (in order):
fng~ (the first two)
gnh~
c^n
dvu

[celide]

Great idea!
u n after note
v ^ after note

Agree.

Tomasz Grabowski OP
prezes fundacji

kom. +48 694 480 613
tel. 12 430 19 34
e-mail: t.grabowski@liturgia.dominikanie.pl

Fundacja Dominikański Ośrodek Liturgiczny
ul. Dominikańska 3/11, 31-043 Kraków
www.liturgia.dominikanie.pl

NIP: PL 6762419775, KRS 0000354036
47 1140 2004 0000 3802 7522 9738

Wiadomość napisana przez Br. Samuel Springuel notifications@github.com w dniu 12 sie 2014, o godz. 21:41:

The problem with that notation is that 'gvg' becomes ambiguous. Would that be the same as 'gv!g' or 'g!vg'? I'd prefer something that doesn't introduce that ambiguity if possible. That would, in my mind, mean keeping the note shape modifier after the note name as is currently done for everything except the initio debilis. Perhaps then using a 'u' for a vigra that appears on the left side of a note ('u' and 'v' are similar letters in shape and alphabet position so the similarity in their function is logical).

It also seems like something to indicate that the vigra goes up (instead of down) is needed to reproduce all the indicated notes. Maybe then '^' and 'n' (as these symbols are upside down versions of 'v' and 'u')?

The highlighted notes would then be (in order):
fng~ (the first two)
gnh~
c^n
dvu

—
Reply to this email directly or view it on GitHub.

[grzegorzrolek]

Notation for virga with stem on the left already exists, see Élie’s message on virga reversa. Moreover, note that the last two neumes highlighted involve a liquescence (they’re probably some kind of epiphonus and cephalicus, respectively) and this liquescence should be taken into account when choosing the notation.

Note also, that the first neume, as Olivier said above, is a regular epiphonus in the Solesmes editions mentioned, so there’s nothing to address at the gabc level in this particular case.

[olivierberten]

I would suggest dV> for the last one

[grzegorzrolek]

That could work assuming there are no cases of different ‘length’ of the liquescence, otherwise an entirely new approach would be needed. Does anybody have more examples of those neumes? Is this edition available online somewhere?

[olivierberten]

http://musicasacra.com/miscellany/dominican-liturgy/

http://gregobase.selapa.net/chant.php?id=4577

[grzegorzrolek]

Here’s what I’ve found from the books Olivier linked to. Harrison in his How to Sing Plain Chant, p. 31, discusses the usual liquescent neumes, after which he adds (with an illustration of what looks exactly like the apparent virga aucta in Salve Regina above):

If the liquescent [part of a neume] be absent and the note be finished with a small line only, then only one note is to be sung, but the double vowel sound (e.g. autem, ejus), or the several consonants (sanctus, virgo) must be as distinctly enunciated as if the note were there. Hence, this one note will necessarily be slightly prolonged.

Similar statement (or so I assume from a rough translation) can be found in Regulae Cantus of Cormier’s 1913 Processonarium, p. 340, with Salve Regina itself as an example:

Si tamen epiphonus et cephalicus punctulo caruerint, ut in his figuris, labiis quidem pronuntiatur syllaba liquescens, sed in melodia, uti nota una simplex figura talis habetur. Vide Antiphonam Salve Regina ad verba osténde et Virgo.

So it looks like they’re all simply puncta aucta or, for that matter, virgae auctae. This virga aucta is actually pretty common in these editions, and the simplest rendition of a punctum auctum with no special stems whatsoever is not uncommon either.

Pretty much every edition makes the ‘small line’ a single, constant length throughout. Rare exceptions are the neumes in Officium Hebdomadae Sanctae from both Fernandez, 1965 (p. 17, l. 2), and Paredes, 1925 (p. 38, l. 1). It’s only Cormier’s Processonarium that differentiates these lengths on a regular basis. It also appears that Cormier’s is the only edition that adjusts the small line’s length when it touches the staff line, the way it’s usually done with the stem of virga.

There are few cases, in both Cormier’s Processonarium (p. 46, l. 4) and Gillet's 1933 Antiphonarium (p. 750, l. 2) of a normal epiphonus having an elongated entry stem.

But the really tricky one is the neume over osténde in most renderings of Salve Regina, including the one pictured above. Its shape is counterintuitive for a punctum auctum ascendens, whether stemmed or not. Cormier is the only one that makes it into a ‘proper’ ascending form (p. 90, l. 4), exactly as pictured in his Regulae Cantus cited above. For some reason, though, all other editions have it in this awkward form, one exception being Gillet’s Antiphonarium with an ‘upside down’ punctum (p. 134, l. 1). There are at least several other examples of this awkward neume with stems of different length, all along cases of the supposedly proper form, so its peculiarity should probably be addressed in one way or another.

It all seems rather complex at first sight and I myself am not sure yet how to approach every aspect of it. It would be easier if someone with any background knowledge into these editions would share some insight, that’s for sure.

[olivierberten]

Edit: moved to #18

[eroux]

Olivier, IIRC, the neumes you show are what was presented to me as torculus resupinus flexus, I know them for a long time, but decided not to draw them directly in the font. The reason is that 5 notes makes 4 ambitus, and thus 5^4=625 neumes to draw (which was not really possible at the time, but would be now). Also, I thought the first ambitus was always one, so if you type the corresponding gabc, Gregorio is supposed to draw a punctum then a porrectus flexus with no bar at the beginning, so the output should be correct in this case... But you're showing neumes that don't fit in this case. Can you open another issue for them?

[olivierberten]

Done: #18

[henryso]

I'll admit to being confused by this discussion. Which shapes, exactly are missing? We have the epiphohus, so is it just the "u"-shaped note over "ten" and the inverted "u"-shaped noted over "Vir" that are missing?

[eroux]

Maybe there should be an alternative epiphonus shape that looks more like this one, but I think the new thing is the weird neume above "ten" (we don't see the text because of the red circle)... At first I thought it was like an alternative porrectus deminutus but without the first and last note... but can these bars above the note get higher than this or are they always at the same height?

[henryso]

What I'd like to understand is how much variety there is in the new shape. In addition to your question about height, can there be a line only on one side? A line going up and a line going down? I don't have any Dominican liturgical books, so I cannot see for myself. Are there any ones available on the Internet that I can look through?

[eroux]

There are somes PDF sources in this comment and on gregobase (in the by source section).

[henryso]

From what I've been reading, it seems that the only liquescents Dominican chant uses are the epiphonus, the cephalicus, the epiphonus-line neume without the final note, and the cephalicus-like neume without the final note. Should the latter two figures without the final notes just be alternate glyphs for the virga (reversa) auctus ascendens -- which would needed to be added -- and virga (reversa) auctus descendens -- which already exists? If this is the case, the notation would be gV> and gV<. If the preference is to add these neumes as regular figures (as opposed to alternates), then how about g&> and g&< for the epiphonus-line and cephalicus-like figures, respectively?

[eroux]

But doesn't the bar before the absent note vary in lenght?

[henryso]

As far as I can tell, that line height and whether to vary it within a score is a stylistic decision. Does anyone know for sure?

[henryso]

This really does feel like the (more ancient) Dominican version of the (more modern) Solesmes augmented and diminished liquescents, as the figure can be tacked onto a multi-note neume or can stand by itself, with or without a leading stem, and since Dominican chant doesn't have liquescents besides the deminutus, we already have a language for this in gabc (except that we currently lack augmented punctum and virga glyphs).

This paragraph (from #1 (comment)) ...

Pretty much every edition makes the ‘small line’ a single, constant length throughout. Rare exceptions are the neumes in Officium Hebdomadae Sanctae from both Fernandez, 1965 (p. 17, l. 2), and Paredes, 1925 (p. 38, l. 1). It’s only Cormier’s Processonarium that differentiates these lengths on a regular basis. It also appears that Cormier’s is the only edition that adjusts the small line’s length when it touches the staff line, the way it’s usually done with the stem of virga.

... is what leads me to believe that the line height before the absent note is a stylistic (i.e., a house style) decision. Do we want to be more Cormier-like, or like "pretty much every edition", or like the other exceptional cases?

My only concern is that we would have to draw as many glyphs as have augmented and diminished liquescents, and that's a considerable number of glyphs. This is true whether or not we decide to make in an alternate or independent glyph.

[eroux]

That's very interesting, thanks a lot for your research! I'm no gregorianist, so : when you say augmented and diminished liquescents, do you mean what is called auctus descendens and auctus ascendens in the code? If so, I think a new font could make sense here... Or maybe the alternate glyph mechanism could be extended to alternate fonts? Also, if the same syntax is used i gabc, this means that one cannot simply typeset normal auctus ascendens neumes and dominican auctus ascendens neumes... Well, I don't think it's a strong limitation, but it's just something we should be aware of...

[henryso]

I actually misspoke. I should have said (and I really did mean) ascending and descending augmented liquescents, and by that, I do mean auctus ascendens and descendens.

The alternative mechanism already supports alternate fonts and wildcards, giving you everything you need to implement this using that system (except for the ascending punctum and virga, which would need to be added).

I agree about the limitation, but that only extends to one score. You can switch it out between scores.

[eroux]

Well, it might be good to have @grzegorzrolek and @celide 's opinions on this, but otherwise this seems fine to me...

[grzegorzrolek]

I don’t have the time at the moment to try and find any actual examples, but as far as the stem lengths and the various 'note-head’ shapes are concerned, my intuition is that the editors of the Dominican editions simply followed what’s in the original manuscripts more closely than the more ‘mainstream’ Solesmes — and those manuscripts could obviously vary with different kinds of notational minutia. It would explain the apparent struggle of the editors over representing the unusual neumes with the pieces of type that were available. See, for example, the ‘rotated’ virga for the neume over osténde in Salve Regina, or similarly misused punctum in the same place in Gillet’s Antiphonarium. For what it’s worth, and to add to the argument further, in those few academic titles on the plainchant that I’ve had the opportunity to read, I can’t recall any author ever discussing the length of various liquescent shapes, at least not in any quantitative terms.

Now whether those stemmed shapes indicate some distinct set of liquescent neumes, aside from what you call puncta or virgae auctae, and the usual epiphonus and cephalicus — again, I can’t recall any author making such a distinction. I’m basically with Henry in that these are simply aucta neumes.

For those interested, in the notation of the early polyphony of the 13th century, which used the square notation of the contemporary plainchant manuscripts, there’s indeed a distinctive feature known as plica, which looks exactly like some of the neumes in question. It’s considered to have more of a rhythmic interpretation, though, and no resemblance to the liquescence as known from plainchant. I’ve once read David Hiley arguing in his article ‘The Plica and Liquescence’ that the plica as notated in early polyphony indeed carried at least some of the performance details of the liquescents, but this still doesn’t change anything on the plainchant side. In a somewhat different note, Timothy J. McGee in ‘“Ornamental” Neumes and Early Notation’, an equally worthy article, surprisingly makes use of the term ‘plica’ for all the single-note liquescents.

[henryso]

(a technical question unrelated to the immediately previous comment:)

It's about the same amount of work to generate a complete font (with all glyphs) as it is to create an extension font (with only the changed glyphs). Either could be used with the substitution mechanism, but the complete font could be enabled with a single statement (which, granted, could be true of the extension font as well, with a supplied macro), at the expense of duplicate glyphs and a larger download/installed size. Anyone have a preference?

[henryso]

How does this look? (score adapted from http://gregobase.selapa.net/chant.php?id=4577)

[henryso]

First cut work-in-progress implementation is at #542, pending some more testing and/or comments. Source for the Salve Regina above is at https://gist.github.com/henryso/b2f85e223a2f4b9fc946.

[henryso]

If there are issues with the implementation, please open a new issue since this one has become rather loaded.

[eroux]

@olivierberten what's the reference of the image you posted in the intial description?

[henryso]

http://gregobase.selapa.net/chant.php?id=4577

Completorium O.P. (Suarez), Dominican, 1949, p. 119

[eroux]

Thanks a lot!