NOTE: this post is available in PDF at the end of the post, which may be easier to read for some readers.
The melodies of Latin office hymns have varied across time, up to the present day when the most “authentic” melodies were restored in the 1983 Liber Hymnarius. The last line of Veni, creator Spiritus, for example, continues to be reprinted the “old” way in congregational hymnals, misheeding the official version in the Liber Hymnarius:
This despite the document of the U.S. Catholic bishops Sing to the Lord, no. 80, which asks that everyone standardize to the Liber Hymnarius version of Gregorian hymns. Oh well.
I’ve had in mind this malleability of the chant melodies lately as I’ve been thinking about the use of the venerable Gregorian melodies with vernacular texts. Here the issue is more pressing, for it concerns the difference between Latin accentuation and English accentuation. According to a widely-held but certainly not unanimous theory, English accents are heavier than their Latin counterparts. (I argue this in the foreword to Chants of the Roman Missal, explaining the approach taken in the English chant of the 2011 English Missal: Study Edition, and won’t repeat it here.) Unlike English, Latin allows for “displacement” of melodic note groups onto unaccented syllables, oftentimes alongside just a single note on an accented syllable. The English accented syllable has more energy and has a sort of magnetic force pulling note groups to it. The melodic accents generally need to line up with the English textual accents. The note groups work better on accented syllables than unaccented ones in English.
Consider this noonday hymn from The Hymnal 1982, no. 19. Its text is in long meter (8 8 8 8 ), with accents on every other syllable: da DUM da DUM da DUM da DUM. The editors obviously thought that the Latin melody could be transferred to the English text with no adaptation, which is to say, with no attention to English accentuation:
The problem is most apparent in the first and third lines, where the melodic accent (which I’ve marked) is out of sync with the textual accent, giving us spi-RIT. But some would say that there is an infelicity in the second and fourth lines as well, where the E-D-D creates a dactyl emphasizing the first note and causing the last note, which is on a weighty accented syllable, to fade away.
We use the 1982 in St. John’s Abbey. I recall when we began singing this hymn at noon prayer some years ago that it was a bit “lumpy” and didn’t quite feel right. The choirs struggled with it. Through repetition we’ve gotten it down, and now it works, but it still doesn’t quite flow.
I’m considering adapting the melodies as follows in the big new hymnal I’m editing for the abbey:
I did not simply move the slurs over in the first and third line, with BA – F# on “SPI-rit” and “TO our.” There are two reasons for this, one having to do with modality and one having to do with rhythm.
- First, modality. The original Latin melody nicely emphasizes both the B and the A (and a Latin unaccented syllable, I’m arguing, can bear the heavy clivis A-F#). Shifting the slur obscures this, emphasizing only the B (and perhaps the F# a bit more than A).
- Second, rhythm. Shifting the slur sets up a waltz rhythm, with 2-1 | 2-1 resulting on “Holy Spirit.” Part of the beauty of the Latin melody, typically, is the creative interplay between melodic implied patterns of 2-1 and 1-2, and between groups of 2 and groups of 3.
Hence my proposal to add a note to the lone B, to give more melodic accent where it’s needed. There is a price, to be sure – there almost always is! – in that there are too many two-note patterns in the end of the line, from B to the end. But I think it works – but barely so – because there is rhythmic variety in that the line begins with a melodic group of 3 (on “holy”) followed by the sequence of all two-note groups.
For the third line, I shifted a slur on “forth” to create one three-note group, followed by a single note, so that the line would not have a monotony of all two-note groups. This preserves more of the “feel” of Latin, albeit with a solution not needed with the Latin text.
I’m less certain about the end of the second and fourth lines. Perhaps they could stand as they are, and perhaps I will revert to that. The adaptation fits the text slightly better, but I’m not sure if that’s worth the price of the resulting heavy feel. The Latin original, even with English, is more elegant in my view.
For the reader who is rebelling against or resisting the whole line thus far, let me note that all this is not some semiological heresy unique to Collegeville or that left-wing dissident Anthony Ruff. It is what has been done for several decades now in German-language chant books – which is to say, in places where chant scholarship historically has been especially advanced.
Here is an example – a Latin melody from Liber Hymnarius put next to its counterpart in the Antiphonale zum Stundengebet (I have the 1996 edition before me).
The German adaptation succeeds in most every respect: it fits the German text, it preserves the modally important notes, and its rhythm isn’t too bad. It could be argued, though, that the second and third lines have too many two-note groups in a row, and the elimination of the three-note group of the Latin (on “CHRIStum” and “..LUti”) make for a “goose-step” feel. What the Germans did at the end of the first and last line, btw, inspired my treatment of the end of the second and fourth line in my example from 1982 above.
Here’s another example from the German speakers – their rather heavily adapted melody for Ad cenam Agni. (It’s on p. 74 of the Liber Hymnarius if you wish to look it up.)
Here’s my dream: the U.S. bishops, or the bishops of ICEL, would take up this issue and standardize the melodic version of every hymn melody in the Liber Hymnarius for use with English hymn texts. We’ve seen so much work done translating Latin office hymns into English meter in recent years, and it sure would be nice to have standardized chant melodies that worked as well as possible with them.