The restoration of the dominant of mode III to ti was done by Solesmes in the Monastic Antiphonale in 1934, as an optional Tonus Antiquior in place of the Tonus Recentior (reciting on do); the ancient dominant has been completely restored by Solesmes in the new Antiphonale Monasticum and Antiphonale Romanum for the OF.

There is some debate about whether the recitation on ti can be used in the EF since it does not appear in the older Roman books: I have sung the Usus Antiquior (EF) with the Tonus Antiquior with the FSSP, which seems to say it is permitted.

If you are using the OF I would certainly use the New-Old Mode III as in the New Solesmes books; if using the EF I would use the Tonus Recentior. However, if you are singing with a group of people, the majority of which are already proficient in chanting, reciting on do could be too difficult a habit to break.

Just for the sake of completeness - the reciting tone of Mode III (and Mode VIII) was raised from ti to do during the days of polyphonic writing to avoid the diminished fifth (tritone) between the dominant (ti) and the second (fa), this could have been fixed by the sharpening of fa but that would distroy the Phrygian quality of Mode III, so it was decided to raise the reciting tone from ti to do. (This is the whole 'mi against fa is the devil in musica': the mi of the 'hard hexachord' [g(do)-a(re)-b(mi)] against the fa of the 'natural hexachord' [c(do)-d(re)-e(mi)-f(fa)].)

Within the Personal Ordinariates of the Anglican use, I think that Mode III psalm tones are always going to be preferred to be on the do clef. Thats the way too many people grew up hearing it. It is associated as being traditional for the end of Evensong/Compline.

Yet, there will always also be some people conflicted on whether Mode III should use more "C" or "B". I can never make up my mind and set it both ways, depending how the notes in the antiphon going with are or depending on whoever shows up wants to sing it...

Could someone post the ti version?

Yes, I noticed that Rorate article about ten minutes after I posted and promptly downloaded. I see the 'ti version now.

Thanks for that explanation, Salieri. I knew there must have been some justification for Solesmes editions returning to SI as the reciting note. I must say, though, that the discussion has left me more confused than ever as to which version I should personally use, and encourage the use of more broadly.

Modern editions seem to prefer SI, but I have only ever seen this in print. I have never ever heard the Nunc Dimittis sung this way, a point which was reinforced this morning by the music played as background music to drown out voices during a second-rite First Reconciliation, which seemed to be a random selection of chant (which began with an In Paradisum) but included the Nunc Dimittis, sung on DO.

Given that I only ever heard it sung on DO, perhaps it is too much of an effort to shift what seems to be established practice, and it would be easier just to follow the traditional usage, even if that wasn't the earliest practice.

The probable cause of the drift to DO is the inherent instability of the SI--even, today, amateur singers often get confused by semitones. Same goes for the drift to to the ti-flat in many Mode I pieces.