Application of musica ficta

[carlos]

A couple of days ago I finally published my first edition on CPDL, a very simple Benedictus in canon by Brumel. The book I took it from had no indications of ficta whatsoever, and as the resulting sonority seemed good to me, I didn't bother about this issue. But today I coincidentally found a midi of this work in another site, and its editor did apply ficta in it (B flat was used instead of B natural). It sounds particularly strange to my ears, but as I'm a newbie when it comes to music theory, I'd like to hear the opinions of more learned members on the subject.

The wikipedia article on musica ficta says:

One common (but not exclusive) use of ficta was to avoid harsh harmonic or melodic intervals (for example the tritone, the 'diabolus in musica'). An example might be the use of a B-flat instead of a B-natural to avoid a tritone against an F in another part.

That's clearly not the case here, but I heard once that specific rules might apply for ascending/descending lines, not sure if it's valid in this situation. My question is: should I add an indication of optional ficta to the score, or would it be unnecessary/plain wrong?

[CHGiffen]

Thorny subject, Carlos!

I see a few B-flat ficta possibilities: (1) the descending B's in T1.m4 & T2.m5; (2) descending B's in T1.m11 & T2.m12 - this second possibility being more critical in that (with B-natural) they resolve an imperfect minor third to a perfect fifth, which by the attraction principle would ask us to alter the minor third to a major third by raising the B-natural to a B-flat).

By the same principle, it would appear that some F's should be raised to F-sharps: (1) T1.m5 (attraction principle, cadence) & T2.m6 (if only to preserve the exactness of the canon); (2) most (if not all) the F's from T1.m26 & T2.m27 to the end (attaction principle & cadence or leading tone). Rather dubious (to me) would be raising the F's to F-sharps in T1.m19 & T2.m20 (leading tone, and imperfect/perfect resolution via attraction principle in T1).

Here I've only considered ficta that would be applied to both voices of the canon, preserving its exactness. If this exactness is relaxed, then someone taking these ideas to extremes might argue for a C-sharp in T2.m4 (but not in T1.m3 which would result in a tritone) - but this is best avoided since there is no cadence to deal with. It's probably better to reject these outright, since then one is beginning seriously to tamper with the modality of the piece. Adding ficta here and there where the result stretches the (supposed) rules or simply because it sounds "better" to us (more than four centuries after the fact) is not such a good idea.

I've stressed and agonized several times over issues of musica ficta. Compare, for example, the different treatments of (essentially the same underlying music) Gombert's Je prens congie and prima pars of Lugebat David Absolon (especially from m.103 to the end). Or Crecquillon's Dont vient cela with and without ficta.

What would I do with the Brumel Benedictus? ... probably insert the B-flats of instance (2) m.10/11, as well as the F-sharps of instance (1) m.5/6 and some those in instance (2) m.26/27, m.28/(29?), m.(30?)/31, m.32/33. And, as always, musica ficta inserted by modern editors (especially) me are suggested possibilities ... it's up to the performer to decide whether to accept them.

By the way, the article/dissertation by Vincent Arlettaz is quite good - well researched and even-handed. There is an online summary at his website.

Chuck

[carlos]

Wow Chuck, that's a lot of information, thank you! I have to experiment with each variation you suggest, just to see how they sound. The resolution of major (instead of minor) thirds to fifths makes sense to me, considering the strict rules by which they used to abide in those times.

The difference in treatment between Gombert's madrigal and its sacred contrafactum was quite interesting. It's a lot of notes in the chords, hehe, I'm still trying to capture the harmonic rules you used in each case. By the way, the midi of Je prens is not opening, it seems the file is not where it should be.

Thanks also for the link to Arlettaz' article, I will read it with care.

[CHGiffen]

Thanks for your kind words, Carlos. In truth, I edited Je prens congie before Lugebat David Absalon, and was, by then, really beginning to absorb the music (and modalities) of the somewhat earlier Flemish masters, Obrecht and Ockeghem, as well as experience with Crecquillon and Josquin. The E-flat ficta in the Je prens congie are partly modal and also partly represent the top of the melodic line ... and the effect is to alternate the root G-minor tonality with a C-minor subdominant tonality in a sort of "plain" way. Not introducing these E-flats in Lugebat David Absalon creates a much different effect, which I now feel is more in keeping with the sense of Gombert's original intent.

The MIDI file you mentioned was originally on the Williams.edu server ... maybe it got lost? I think I've got another copy and will upload it. I wonder if there are other "lost" files from the williams.edu server (after transfer to our own server).

Chuck

[pml]

I agree with Chuck - this is a very thorny subject!

The application of accidentals in Brumel's work is often incredibly sparing. For example, in the Missa Et ecce terræ motus, which occupies 120 pages (60 folios) of music, there are only 23 accidentals in total: nine B flats and fourteen B sharps (i.e. B naturals). All of the sharps appear to be later additions, since they are not in the same hand as the principal scribe - whereas all of the flats were added explicitly (to confirm a vertical triadic sonority of what we'd call B flat major these days). The B sharps are there to indicate places where the B should not be flattened - though this occasionally causes other issues such as false relations when a preceding F should be raised to prevent a melodic tritone, while another part has a more or less fixed F.

In your setting of the Benedictus, fuga ex una, I'd be inclined to broach the following as optional "cautionaries" above the stave:
1st part.bar 5.note 3 and 2.6.3 = #
1.14.6 / 2.15.6 = # (and yes, this one would cause a momentary dissonance of the tritone in the latter case!)
1.28.3 / 2.29.3 = #
1.30.4 / 2.31.4 = #
1.32.4 / 2.33.4 = #

Possibly also 1.2.3 / 2.3.3 and 1.3.1 / 2.4.1 sharpened as well.

I'd be less inclined to go with the flats, as in the places suggested by Chuck they are rather fleeting; in the passage from bar 7 onward, if you add one flat somewhere, you inevitably have to begin adding them all over the place and as Chuck mentioned, then you're beginning to tamper with the modality of the piece...

I notice in a number of places (bars 24-27) you've tied minim to crotchet - these are almost all likely to be have been dotted minima/semi-brevis in the source (depending on whether note values have been halved or not).

As for the Gombert, there's also an 8-part Credo which has some material in common with both Je prens congie and Lugebat, and which would probably propound another treatment of ficta, depending on the construction of cadences and the way voices are handled.

Regards, Philip

[CHGiffen]

Philip's take on this issue varies to some extent from my own:

His T1.m14 & T2.m15 F-sharps seem out of place to me - they may sound "better" to him (and maybe even to me were I in a more modern frame of mind), but they do not conform to the principle of attraction in that they do not lead to a perfect cadence (at the unison, fifth, octave, etc.; a cadence at a major third is imperfect and the so-called "rules" do not apply, at least not until many years later.

Also, as I noted with question marks in my analysis, I am dubious about F-sharps for T2.m29 and T1.m30, because although the cadence is perfect, they do not expand the third to a fifth by contrary motion (there is a downward leap of a fifth in the other respective part).

Adding F-sharps to T1.m2 & T2.m3 might seem reasonable as they are part of an ascending scale, but then also to add F-sharps in T1.m3 and T2.m4 (why? to balance what was just done, even though the direction of motion is now downward?) is, to me at least, quite unwise for the simple reason that it destroys the modality of the piece from the outset (it is NOT in G-major!). Personally, I would refrain from tampering with the modality by adding these F-sharps. The only reasonable ones near the beginning would be the imperfect to perfect cadential F-sharps added in T1.m5 & T2.m6, which I happen to recommend.

On the other hand, while not defending the B-flats as being totally necessary, I do point out that B-flats in T1.m11 & T2.m2 completely fulfill the requirements of resolving an imperfect third to a perfect fifth at a cadence by contrary motion. The presence of these B-flats, if included, is not at all an invitation to add B-flats elsewhere.

Okay, there are many types of cookies (or biscuits) ... I happen to prefer Tim Tams (from Australia!) above most all others. Perhaps Philip prefers some sort of American biccie. At any rate, we do seem to disagree on what ficta to suggest (always suggest, not demand!). I continue to try to be conservative with these earliest works, and I definitely try not to impose late Renaissance (often Italian, at that) principles to such early Franco-Flemish music.

Chuck

[carlos]

Thanks Philip and Chuck for your thoughtful analyses and diverging opinions, I'm learning a lot from both. It's fascinating how the application of ficta can vary so wildly.

I'm planning to dedicate some time this weekend to create midis with the suggestions received so far, so that we may all have an idea of how they sound.

[pml]

To be honest, although I've marked rather a lot of ficta in practice I tend to leave more of the options out. Depending on how badly exposed they are to major key music, some modern singers will often put in raised leading notes anyway quite unwittingly; and any ficta really has to be tested with the ear to hear whether it adds anything to the music. For example I'm aware the descending Fs (1.3.1 / 2.4.1) should strictly not be raised, but the canon being only at a bars length leads to a rapid alternation of the raised and lowered leading note, and then the fugue would begin to sound more like a later composer such as Victoria rather than Brumel. So I qualified those as "possibly" - in practice I'd be very inclined to leave them all out. As for the end of the piece, I notice the other CPDL edition of this piece is a bit undecided, and has an explicit # at 28/29, natural at 30/31, explicit # at 32/33. I agree with Chuck that these sorts of additions should be clearly indicated as such: editorial suggestions, rather than prescriptions for performance.

And although I've tried Oreos, I have to say I am sadly parochial in my bickie preferences, and have been known to indulge in a Tim Tam Slam occasionally. ;-)

Cheers PML

[carlos]

I notice in a number of places (bars 24-27) you've tied minim to crotchet - these are almost all likely to be have been dotted minima/semi-brevis in the source (depending on whether note values have been halved or not).

Oh Philip, I was forgetting to comment about this; I have been totally faithful to my source (a 1885 book by J. B. Weckerlin), even to the ties. The only "editorial" change applied was by replacing the C-clefs with Tenor G-clefs. Weckerlin may well have done some kind of adaptation himself, when transcribing from his source, the book Musices practicae erotematum Libri II by Grégoire Faber.

[CHGiffen]

And although I've tried Oreos, I have to say I am sadly parochial in my bickie preferences, and have been known to indulge in a Tim Tam Slam occasionally.

Funny thing, I've just been going through a packet of dark choccy Tim Tams, with the occasionaly, but obbligatory, Tim Tam Slam!! I've another couple of packages of Tim Tams to hold me through the next week, then I'll have to track some more down.

Best,
Chuck

[Jaquick]

I'd only add 3 ficta here, all F#: vI, m3, 3rd note ; vI, m28, 3rd note; vII, m. 33 4th note -- all cadential.
It's a mixolydian piece, so Bb would be out of place, as "una nota super la" in the hard hexachord would be f natural anyway, and you'd only sharp it to be a cadential leading-tone to G.

Ficta usage can be a matter for much discussion. In rehearsal, the director's rule should be, "Any ficta discussion happens after rehearsal, over beer -- and the plaintiff buys the beer!" In general, I believe that less is more. You will need to sharp cadences and sometimes a final 3rd, and flatten tritones and "una nota super la", but not much beyond that. One suggestion I've heard from professional early music players is to read the piece without any ficta at all, and then add in only the ones that seem absolutely necessary.

I do however have a very strong opinion about how to notate ficta in Finale. There are several ways to do it. This is particularly important if you're posting Finale files instead of pdfs, because somebody might want to transpose your Finale file. First notate your ficta sign (flat or sharp) as if it were a normal note. This way your ficta will play back. Then use the Note Mover tool, select the accidental mover, check "Allow vertical positioning", and then drag that accidental to a position above the note. It's the right size. More importantly, if you transpose the piece up a step, the accidental will change if necessary. If you have Eb ficta and you transpose up a step, you will get F natural instead of Fb.

[CHGiffen]

I agree with everything about notating ficta in Finale, except I usually resize to 85%.

Chuck

[carlos]

Ficta usage can be a matter for much discussion. In rehearsal, the director's rule should be, "Any ficta discussion happens after rehearsal, over beer -- and the plaintiff buys the beer!"

Haha, I liked this rule!
Thank you for your suggestion on the usage of ficta, and for sharing what you learned about applying it.

Thanks also for explaining how to add ficta correctly in Finale, I was just trying to figure out how to do it in a way that the ficta could be played back.

[D-fished]

Hello all,

I'm not sure how much interest there will be in reviving this topic a year later, but here's my two cents' worth.

I've learned to avoid the term musica ficta for the application of accidentals to the score. The term had a very specific meaning in the Renaissance. It referred to hexachords that deviated from the standard gamut, that is, the Natural (c-a), the Soft (f-d, with b-flat) and the Hard (g-e, with b-natural). Anything that required solmization in another hexachord, say, b-flat to g, with e-flat, was called feigned or fictive music. Editorial tidying up of mi contra fa conflicts and leading tones at cadences don't necessarily involve fictive hexachords.

The first thing to remember is that the singers of the time did not use scores. Whatever decisions they made about accidental inflections had to be derived from what they saw in their parts and what they heard going on around them. They also didn't tend to rehearse extensively and work out the decisions ahead of time. They had to adjust on the fly, and there are certainly accounts of singers fumbling the solmization during Mass etc.

Another thing to remember is that they imbibed the system of reading in hexachords from the very beginning of their musical training. We think in terms of a C is a C is a C. They looked at their parts and saw this note is ut. Here, it must be read as fa. There, it must be sung as ut, but inflected up a semitone to make a major sixth with a cadencing tenor. If this C is sol, then the B below it must be fa, even if there's no flat sign provided. (Although it usually worked the other way around—the presence of a flat sign told you that you were in the soft hexachord, and therefore, C is sol.)

The two main reasons for accidental inflections are: avoiding mi contra fa and making sure cadential sixths were major. Melodically, avoiding mi contra fa involves observing where your part encompasses a tritone or diminished fifth (the two are different). A line that went f-g-a-b-a would typically call for the b to be sung as fa (b-flat in our parlance). Leaps from e to b-flat or from f to b, would similarly be softened. All this falls out from the rules, and if you know the rules, you don't need the flat signs in your part. The second category of mi contra fa gets a little hairier. The singers had to rely on their ears to tell them that another part had just sung, say, a b as fa, and seeing that they had a b coming up, they would sing it as fa as well. They had to be able to perfect imperfect fourths and fifths with other parts by ear as well.

Cadential sixths are a different matter. They aren't generally treated as mutations (changes of the current hexachord to a new one). They are also dependent on the mode. Because of the way dissonances are prepared in this music, a singer with a leading tone hears in advance that he (and we are typically talking about men and boys, here) has a 7-6 suspension over another part. A singer singing d-c-d against another part moving e-d would know that the c needs to be raised to make a major sixth before the octave. He would solmize it as re-ut-re, even though the 'ut' in this case is only a semitone below re. (Remember, the only semitone in this system is mi-fa; there are no ri's, fi's and si's.) To sing that c as mi would involve mutating to a hexachord where A is ut, and that would have dramatic consequences. Essentially, if the copyist or composer marks a sharp on any note other than b, that's what he's telling you—you are now in a fictive hexachord where d, or a, or e is ut.

In certain modes, it was necessary for the lower singer to inflect down a semitone and make a 'Phrygian' cadence. For example, up to about 1525, a tenor with a line moving c-b-a against which another singer sang a-g-a would have to sing the b as fa. The late 15th century theorists got quite hysterical about ignorant singers introducing this new-fangled g-sharp in cadences to a.

I've gone on long enough, but I'm happy to talk in greater detail with anyone by email. Just check my contributor page for the address.

One further point, though. Accidentals in Renaissance partbooks are warning signs telling you something unusual is happening. They usually mean, 'Watch out! All your musical training would leave you to believe that this note should be sung as mi. It's really fa.' (Or vice versa.)

Cheers,
David Millard