If you allow such a flexible lengthening of a syllable within a cadence, you have distorted the rhythm of 2s and 3s. This makes it awkward for the singers to enter together as they continue with the phrase or verse or begin the next verse.

You would need a director to coordinate the phrasing and entrances.

If the choir thinks and sings in accord with the rhythm of 2s and 3s, it is clear to everyone when it's time to cut off and enter.

In your article, you cite Dufay's Ave Regina caelorum as an example of misaccentuation. I agree on "Ave regina caelorum," but on "Salve radix sancta," there is another interpretation: it is a hemiola; within the 3/2 meter, there is a large 3/1, i.e., 3/2 sal-/ 3/1 ve / ra- dix / san-/ 3/2 cta. Final syllables on the first beat of a measure do not have to receive an accent and that is a commonplace in the period, so "ve" and "cta" though they are on the "downbeat" can be sung unaccented, with good result. There are several such passages further on in the piece; in performance, the accentuation comes out very well, once you realize that it is a hemiola. Concerning the initial phrase, I think that this is a matter of the composer's allowing the lilting triple rhythm to dominate the accentuation of the text.

WGS,

Well, to lengthen or not would depend on the position of the Latin accent in relation to the musical cadence, so would be predictable. That said, I don't have a problem with the presence of a schola director, ideally exercising a light, guiding influence on a group of singers used to singing chant together. I think that the best combination to encourage ensemble and movement.

Hang on--it looks like Fr. Kelly is saying that in this instance the word is sung differently than it would be spoken, and that the indications in the music effectively shift the accent to the second syllable?

So Fr. Kelly agrees with Jeff Ostrowski that musical considerations sometimes override normal speech patterns? :-)

Pages 144, 158, 172, 320, 321, 412, 418 & 526 of the Graduale Triplex all show notations of melodic units comparable to that on 'aperis' on page 343. Making a comparison between them and that on 'aperis', it appears that the section of melody at the very start of the melody for the second syllable of 'aperis' is not positioned at the very start of the melody for syllables containing comparable melodic units on the other pages.

While this, of course, isn't a conclusive full-blown study, a possibility which arises out of this is that the first two syllables of 'aperis' could be viewed as demonstrating an instance of a single melodic unit being divided over two syllables, a feature not unknown between chants using the same melodic material with different words. Stressing the first syllable of 'aperis' on p343 would certainly produce a matching melodic stress pattern to that found at the start of the word 'tuam' on p321.

With regard to lengthening the tonic accent, I find it significant that the notations for the opening and closing formulae of the mass psalm tones in Einsiedeln 121 include no lengthening indications which regularly shift from syllable to syllable to match shifting verbal stress . I find it even more significant that no such lengthenings to accommodate verbal stress are marked in the main body of the psalm tones either, and that the writer does not always even bother to write the whole psalm tone out.

It is not the case that lengthening indications do not appear in psalm tone notations in Einsiedeln 121. They occur with neumes of more than one note and with neumes on the final syllables of textual phrases, so the immobility of such lengthening indications in relation to verbal stress does not provide any support for the idea that psalm tones were sung using a system of durational flexibility which compensated for verbal stress.

I don't buy it.

gaude is a command to singular (implied) (second) person
specifying that person as the more specifically singular "you" (te) seems like an unlikely source for the plural form of the command.
(Just sayin')

Though the following is completely uninformed conjecture, it seems more likely for the etymology to have gone in reverse:
gaudete (and other commands), when used in the singular context, being mistaken for a contraction of gaude, te, thus birthing a singular form.

Of course that could be as wildly inaccurate as the other statement.
But let's not forget:
-in written Latin during the period when it was actually in use as a conversational language, there aren't accent marks
-the pronunciation of Ecclesial Latin is now governed by the Church and traditional practice, not by archeo-linguists trying to figure out how Latin was pronounced authentically

Adam Bartlett,

you wrote, "the fact that the indication "coniungatur" in the St. Gall manuscript tells the singer to conjoin "aperis" with "tu" which places the accent effectively on "-per" "

If you have proof that 'q' is short for 'coniungatur' here, I would like to see that. Notker writes, "Q. in significationibus notarum cur quaeratur? cum etiam in verbis ad nihil aliud scribatur nisi ut sequens V vim suam amittere queratur". Timothy McGee's translation of this reads, " 'Q' Query: what is the signification of this letter for notes, when even in words it is written only that the following 'u' may lose strength?"

However, the point that I would make that would be absolutely pertinent here is that I know of no proof that any of the significative letters are primarily intended to indicate verbal stress. Should there be proof of such, then more concrete information would be available to us about verbal stress in the latin used in these documents.

Dom Kelly's suggestion is anyhow irrelevant in relation to verbal stress. In speech, the same options of verbal stress would pertain to the word 'tu' regardless of whether the first or second syllables of 'aperis' are stressed because there is a third syllable. Obviously, it is impossible to separate the word "tu" from "aperis" in speech without the use of a gap of silence or an intervening word or an epenthetic vowel after the 's' of 'aperis'.

In practical terms, we are therefore left with the idea of a verbal stress indication being written on one musical sign (and notably not in the text) but intended for its effect on the verbal stress of a previous syllable struck many notes beforehand; in short, the indication is given too late in the notation. This would be like leaving a note saying, "the next syllable is stressed so the previous syllable should also be stressed and not the syllable prior to that".

A more practical interpretation of the probable use of "coniungete" (join together) in this instance, which would relate it to melody rather than to verbal stress, might be to warn the singer against ending the phrase at the last syllable of 'aperis' having arrived there after a substantial melisma. This would ensure that the phrasal end is made after 'tu'. This makes sense because the personal pronoun 'tu' is the subject of the sentence and is to be associated with the verb in phrasing, rather than with the object of the verb.

Ever since this post began I have become very self conscious about my tonic accents. it's just causing too much STRESS!
I think the importance of repose or a thetic shape must be made clear at the end of an incise even more so than the accent.
When chanting in English I do prefer to lengthen the accent in the termination of Tone ii when it is precisely pointed on an accented syllable. I really cherish real trochaic and dactylic English terminations which seem to be rare. Here comes one on the Christmas R. Psalm: "constancy."