Duruflé and Solesmes rhythm
  • 5744SHAW
    Posts: 11
    The following paragraph was posted by Incantu in June 2013. I welcome comments from anyone (including Incantu) who can elaborate on Duruflé's "entirely incorrect theories." I'm virtually illiterate in matters of chant and chant history and am dependent on others who are "fluent." I'm writing an essay on Duruflé's life as a chorister in Rouen and would like to have a better grasp of this issue.

    "We would not have, for example, the masterful choral works of Maurice Duruflé had he not based his compositional style on what has turned out to be entirely incorrect theories concerning the rhythm of Gregorian chant."

    Thanks to all!
  • MatthewRoth
    Posts: 3,437
    He’s saying that Dom Mocquereau’s theory was wrong, although as @Charles_Weaver reminds us regularly, Mocquereau was also establishing a grand, universal theory of rhythm as well.
    Thanked by 1CHGiffen
  • @5744SHAW

    I learned from Eaton's dissertation that Duruflé wrote the following in a set of program notes (1980) to the Requiem:

    Thus I have endeavored to reconcile as far as possible Gregorian rhythm, as it has been established by the Benedictines of Solesmes, with the requirements of the modern measure. The rigor of the latter, with its strong beats and weak beats recurring at regular intervals, is in fact difficult to reconcile with the variety and flexibility of the Gregorian line, which is nothing more than a series of successive impulses and relaxations.

    The strong beats had to lose their dominant character to take the same value of intensity as the weak beats, in such a way that the Gregorian rhythmic accent or the Latin tonic accent can be placed freely on any beat of our modern measure.


    Thus, a lot of the rhythmic suppleness that you hear both in the Requiem but also in the Messe cum jubilo and in the chant-based motets is a direct consequence of Mocquereau's theory. That theory was once the dominant way to sing chant in the Anglosphere and is still present in Church circles, but it is no longer considered the cutting edge of medieval performance practice, the practitioners of which generally favor semiology, mensuralism, or some other sui generis approach. Hence Incantu's comment.

    As Matthew says, I have maintained that Mocquereau is best considered as a turn-of-the-twentieth-century theorist of rhythm quite apart from the question of chant performance practice, and should be read as part of an ongoing discourse about: what exactly an accent is; the nature of musical meter; what is the nature of upbeats and downbeats; and how this interacts with text accentuation in Latin or in various other languages. Mocquereau in general is what you might call an upbeatnik, believing that all musical motion is a movement from an upbeat (arsis, élan) to a downbeat (thesis, repos). Other nineteenth-century writers who are somewhat along the line of Mocquereau are Momigny, Lussy, and Riemann. There were and are also plenty of downbeatniks (i.e., those who believe that music is best chunked in time units that begin on downbeats, and the upbeat is sort of an aftereffect). Among these we might include A.B. Marx (the good Marx!) and Heinrich Schenker.

    Duruflé and Messiaen were both heavily influenced by Mocquereau's theories in their compositions. I also find Mocquereau's ideas about rhythm/meter in general to be very sensible and musical. Plenty of others here and elsewhere disagree about their value! I have long been an incorrigible upbeatnik.
    Thanked by 3Liam CHGiffen davido
  • Liam
    Posts: 5,570
    Charles

    Perhaps - pure speculation here - for us Anglophones, our natural poetic and even speech patterns are so dominated by iambs and anapests that we have a bias towards being upbeatniks?
    Thanked by 1CHGiffen
  • 5744SHAW
    Posts: 11
    Charles,

    You've responded to my inquiry with grace and with compelling expertise. I appreciate it deeply.
    It's enormously informative--and brand new to me--what you have to say about Mocquereau and his influence upon Duruflé. I knew none of it previously and now have something worth further pondering. Thanks for that!
    I will re-read your message and may return to you with further questions.
    Jim Frazier
  • 5744SHAW
    Posts: 11
    Charles Weaver.

    jamesfrazier342@gmail.com
  • @Liam

    Absolutely, I think language has a lot to do with it. There is an excellent article out there by my grad-school advisor William Rothstein about "National Metric Types." His point in that article is that, on the whole, native German speakers are more downbeatnik and native French speakers are more upbeatnik. This difference becomes more apparent when you are looking at phrase rhythm (or hypermeter). Ask, for instance, which measures of Beethoven's Ode to Joy are strong and which are weak? The most interesting cases are when people cross national style boundaries (like Mozart's Italian operas). Haydn's emperor hymn was originally barred one way and then turned around, and of course his music has heavy Italian influence.

    These styles/language patterns also seem to change over time as well, and the language a composer speaks is not completely determinative. For instance, there is no more downbeat-heavy (thinking of the phrase level) composer I know of than Poulenc; he is practically the opposite of F. Couperin in that way.

    In his more recent book on Italian opera, Rothstein suggests that we should develop different ears for different language/meter styles: a wonderful idea. He offered a course on rhythm that was one of the most interesting classes I've ever taken. We spent a ton of time on Beethoven.

    I believe that Mocquereau's theory as laid out in Le nombre is one of the best and most coherent statements of an upbeat way of thinking.
  • MatthewRoth
    Posts: 3,437
    We spent a ton of time on Beethoven.


    in the spirit of a prospective conservatory student in 19th century Paris, perhaps… :)

    Yes, although ironically, Couperin lived at a time where certain forms of prepared speech (or so I'm told) were more like modern Italian to my ears a. There's a guy who makes videos in period dress with the rhetorical equivalent of HIP to recite the fables of Jean de La Fontaine; now, it leaves me a little queasy, but it's still interesting.
  • Mora
    Posts: 28
    @Charles_Weaver
    Regarding language - also remember that Durufle was writing for French Latin, not Italian Latin, which makes the “uptick” or “lightness” - even that more pronounced. If his Requiem is sung in French Latin, it is an entirely different experience than sung in Italian Latin (we tried it in a conducting lab in grad school - and it was amazing).
    @5744SHAW
    Although there were mensuralist theories at the time of Durufle’s compositional writing, Mocquereau’s theory was perhaps the best-known and most widely-practiced. Mocquereau was trying to develop a theory that would appeal to musicians with little experience with Gregorian chant, since it had all but disappeared until the Solesmes revival in the mid-19th to early 20th-centuries. It was widely successfully. However, Mocquereau was the best paleographer in his monastery and refuted his theory at the end of his life, knowing it was incompatible with the neumes he was studying.
    All that is to say, Durufle was a product of his time.
  • GerardH
    Posts: 646
    remember that Durufle was writing for French Latin, not Italian Latin

    @Mora citation needed

    Depends what you mean by 'French Latin' - if you're referring to Latin with French vowels like with historical performances of Charpentier's Te Deum, I think you might be mistaken. I'll find a source for that claim (or retract it) when I get the chance.
  • FSSPmusic
    Posts: 488
    Mocquereau was the best paleographer in his monastery and refuted his theory at the end of his life, knowing it was incompatible with the neumes he was studying.
    Citation needed on this too. I would be most eager to see it if true!
  • davido
    Posts: 1,186
    The umtramontane battles between the Curia and the French hierarchy were largely over before Duruflé was composing. Not sure how much Italian Latin pronunciation had permeated French choir culture, but I doubt church Latin was being intentionally taught with a French accent by the mid 20th century.

    Also, this is easily verifiable as we have recordings of choirs of this era.
  • kevinfkevinf
    Posts: 1,257
    However, I heard Madame Durufle singing some parts of the Requiem while she was explaining it. It was not Italian Latin. It was French Latin..you can hear the difference.

    Just saying....
    Thanked by 2Mora CHGiffen
  • kevinfkevinf
    Posts: 1,257
    And there is a 1948 recording of the Durufle Requiem on French radio. I will have to go back to listen. FYI; he revised it after this recording.
  • FSSPmusic
    Posts: 488
    Duruflé conducts his Ubi caritas (French choir):
    https://www.youtube.com/watch?v=JzGnfxsEF0g
    Sure sounds like Italianate church Latin to me.
    Thanked by 2Mora CHGiffen
  • MatthewRoth
    Posts: 3,437
    Mocquereau's own theory accounted for more accord with the manuscripts.

    French choirs have a very forward sound and the middle ground is between that and what Italians produce. I’m probably the one who asks for IPA symbols the most here and uses them a lot, but if you use Italian interpretations of that you get « Mario ! » lol

    Anyway, that is to say that there is less and less /y/ and /ʒ/: you won’t hear /ʒe.zy/ in modern performance for the name of the Lord, but it’s a decidedly French sound more often than not. And on that one: the voiced consonant is one that everyone seems to devoice correctly…would that people could do it consistently.
    Thanked by 1Mora
  • Mora
    Posts: 28
    I must’ve been mis-remembering something! I apologize! Dom Mocquereau never recanted his rhythmic theory. However, I found some other interesting pieces of information that point at least to the inconsistencies between his rhythmic theory and his research of the neumes. When asked by Pope Pius X in 1904 how long it would take for a performance practice of Gregorian chant to be produced, he replied, “Oh, about fifty years”. The Pope essentially told him he only had a couple years to figure it out. That is revealing because it shows that he knew from his neumatic studies that understanding all the neumes would take decades(1), but he was given an ultimatum to produce a theory that could be disseminated throughout the world. What ensued was his Solesmes method. Cardine, importantly, saw his work as a continuation of Mocquereau’s scholarly work, stressing that Mocquereau was concerned that someone would continue in the pursuit of a more critical edition of Gregorian chant(2). Cardine was never convinced of the theory(3), and I find it difficult to believe that Mocquereau could reconcile his theory of two’s and three’s with the manuscript evidence that pointed to the contrary. I know human beings are full of contradictions - myself having plenty - but these are very large contradictions. I guess I wonder if he ever had private reservations about the system he was promoting.
    I’m sorry again for the mis-information! Not intended!

    1. “The Role of Semiology”, Fr. Columba Kelly, published in Sacred Music (no volume number given), originally given at a conference on Gregorian Semiology between June 26-28, 1988 at California State University, pg. 6.
    2. Ibid, pg. 6
    3. Private interview with Professor Godehard Joppich, Frankfurt, March 2005, pg. 6.

    https://isidore.co/misc/Res pro Deo/Greg Chant/semiology_sacredmusic.pdf
    Thanked by 1FSSPmusic
  • FSSPmusic
    Posts: 488
    Yes, and there were aspects of Mocquereau's interpretation that weren't incorporated into the "official" Solesmes method, such as the quilisma as a portamento and repercussion of each repeated note. Cardine discarded the ictus placement theory but retained the theory of rhythmic nuances. I have no problem with binary and ternary groupings per se—4 is simply 2+2, 5 is either 2+3 or 3+2, and an isolated upbeat has a silent eighth rest before it forming a binary grouping—but a lot of the Solesmes ictus placement is demonstrably wrong. He managed to convince nearly everyone that the long clivis and pes quadratus were long-short figures rather than entirely long, based on a faulty application of the "golden rule" principle, and he erroneously treated the normal syllabic value as short and indivisible on that same basis.
  • MatthewRoth
    Posts: 3,437
    In practice Mocquereau is not wrongly applying the golden rule. It’s much more sensible at the end of lines or words (and trying to get people to not separate out syllables is a challenge no matter what). But Fontgombault has long modified certain examples of the clivis where only the first is explicitly marked long.

    But I can only refer people back to Dr Weaver’s post and let them see for themselves.

    And in Moc’s defense he does put the episema over both notes sometimes, plus it’s Pothier who gave him this tradition!
  • @Mora

    I think it is useful to separate Mocquereau's paleographical studies from his theoretical concerns. Also, it would be good to separate questions of melody and rhythm. I believe the fifty-years idea was primarily about melodic readings and the work of the Vatican commission, which was separate from his rhythmic theorizing. The end result of the fifty-years-versus-a-few-years controversy was not the Solesmes theory of rhythm (which anyway was almost fully developed by Paléographie Musicale 7 a few years before) but rather the fact that Mocquereau and the other Solesmes monks left the Vatican commission over the lack of deep and substantive debate about melodic readings, so that the Vatican Edition was based more or less on Pothier's early melodic readings.

    As for separating theory and paleography, there is, of course, no basis for 2s and 3s in the Gregorian manuscripts, and Mocquereau never would have suggested there was. He based that idea on basic musical psychology and physiology: humans tend to chunk series of undifferentiated pulses into small prime numbers; groups of two remind us of walking; groups of three remind us of breathing. This is quite separate from his paleographic study, which was quite sophisticated. On the other hand, Mocquereau would suggest that any string of pitches of more-or-less equal length is going to be grouped into 2s and 3s just by nature of how our brains work and regardless of how the neumes are drawn. I don't think that's such an unreasonable idea. I have run across many writings of Mocquereau where he sounds more semiological and less dogmatic about the ictus, but then he turns around and is completely dogmatic. The little monograph on Alleluia Ostende is a case in point of this. He gives a very nuanced and detailed study of each neume and a kind of proto-semiological reading, but he also gives a full Solesmization, which is also replicated in Suñol's textbook. I don't think these things have to be mutually exclusive or contradictory. I've attached examples of each.

    alleluia1.png
    1094 x 792 - 523K
    alleluia5.png
    752 x 914 - 319K
  • As an addendum, the whole controversy over rhythmic signs and the reception of the Vatican edition meant that the time was not ripe for semiological editions in the early twentieth century. I think this was why Mocquereau preserved these investigations for his more speculative rather than practical publications. Happily, living as we do more than a century later, it is possible to read Pothier, Mocquereau, Cardine, and Guilmard as part of a single tradition of performance with many varied branches.
  • FSSPmusic
    Posts: 488
    the Solesmes theory of rhythm (which anyway was almost fully developed by Paléographie Musicale 7 a few years before)
    Which is a problem of great consequence, as Laon 239, a nearly complete Graduale from perhaps 880 and a manuscript source of the greatest value, was not rediscovered until 1906—five years later—according to Mocquereau himself, and published in PM X, 1910. They didn't fully know what they were talking about yet.

    Regarding the golden rule, the preface to the Vatican edition states that, “there must be no pause at the end of any neum followed immediately by a new syllable of the same word; by no means must there be a lengthening of sound still less a silent beat, for this would break up and spoil the diction.” Elsewhere in the preface, pause means a lengthening, not a rest or breath. In contrast, Engelbert of Admont (ca. 1250–1331) uses the word pausa to refer not to a lengthening, but a rest. I think the thirteenth-century golden rule refers to a breath/rest, not an episema, long note, or mora vocis. Regardless, it tells us nothing about first-millennial performance practice. In the oldest manuscripts, long notes before syllabic breaks are more the norm than the exception; more on that momentarily.

    In the Al. Ostende nobis example above, note the different ictus placement at -lu- than in the official Solesmes edition, where there is an ictus on the first and third notes, and the liquescent note is disregarded. The authentic rhythm according to the ancient manuscripts is short-long-short-long. In my edition, I have placed an ictus mark on the lower note as a warning of syncopation. The fourth note is also ictic. Another possibility is that the lower note is a grace note and the second note is dotted. Either way, the neume has a value of three long beats, not two. Of the seven syllabic breaks in Mocquereau's quintuplex example, four of the preceding neumes conclude with a definitely long note, and two others are long by interpretation, leaving -no- as the sole example of a short note before a syllabic break. At least the Cardinians have more sense about this "golden rule" than the Mocquereauvians and generally respect more of the long values.
  • MatthewRoth
    Posts: 3,437
    But the Cardinian semiologists still prefer SG to Laon anyway…
  • FSSPmusic
    Posts: 488
    And?
  • MatthewRoth
    Posts: 3,437
    Well you probably have an idea of what I might say: it is perhaps not that earth-shattering in their view, and for whatever departures Cardine — and Saulnier — made, they did not jettison the traditions of the abbey, and indeed despite removing printed signs they sing as if they are there. Saint-Benoît du Lac added them I. by hand. Who knows what the Solesmes monks do in choir now, especially since the Liber Hymnarius is now without signs and also without Cardine’s preface.

    Even if, perhaps, the dogmatization comes out at the last minute in Mocquereau’s writings, if the Laon tradition really was that big of a deal, it’d come up in PM.

    I also do think that while it is possible that we have exhausted our sources, the idea that Mocquereau et al. had not fully understood something leads me to, well, get nervous; what will the next revolution be? I’ve posted before the article on the Deus in loco sancto introit, I’m pretty sure. The rhythmic and melodic questions are separate but certainly overlap, and for various reasons, I’m dubious of attempts to reform the melodies even if they are less, well, clunky than the attempts to reconcile mode 5 and tone 7. (I tell Matthias Bry every now and again that I’ll eventually buy his books, but we’re not changing Christmas or Tenebrae.)
  • FSSPmusic
    Posts: 488
    it is perhaps not that earth-shattering in their view
    But that's precisely the problem! It is a manuscript of great clarity, written perhaps four generations earlier than the oldest St. Gall sources for 60% of the Mass chants.
  • To circle back momentarily to the OP's point, it is Mocquereau's theory—specifically that chant (and indeed all music) has downbeats and upbeats but that the downbeat is not necessarily strong—that was so influential on Duruflé.

    I don't know how Mocquereau's paleographic/semiological views might have evolved. At any rate, his description of the Laon source in PM 10 seems pretty clear. But again, he was limited by what he could print in his chant editions, which still had to adhere more or less to the Vatican edition. The Solesmes theory as we have it (and as still transmitted by Laus in Ecclesia etc.) is designed for users of the Solesmes books, who don't have all the paleographic information to hand. It produces a good result (a subjective judgment, but one I still experience viscerally and frequently when chanting or listening to chant). We must indeed admit that as a historical reconstruction of ninth-century performance practice it is way too full of conjecture to pass muster. But inasmuch as that is not the goal of most chanters or most conductors of Duruflé, it can still be useful.
  • MatthewRoth
    Posts: 3,437
    Right, and Solesmes seems content with the Graduale Triplex. I don’t think that they feel inclined to move beyond the Vatican Edition as some would wish they would.

    I’m always ready to call out longwinded poetic nonsense from French scholars, but I don’t see why it’s unreasonable to hold that the Laon manuscripts are not entirely disruptive or inconsistent with what those more aligned with Cardine hold about chant. Perhaps the unspoken assumption is “earlier is not necessarily better in the end”. I don’t know. Maybe Dom Guilmard would have some more insights. Or Père Damien at Le Barroux who is also very open-minded IMHO (he is working with Matthias on the Nocturnale).
  • FSSPmusic
    Posts: 488
    If we add the relatively late St. Gall 376 to the mix, relying on St. Gall sources alone, we get definitely long notes at five of the seven syllabic breaks, circled in red:

    image

    image

    The notes at syllabic breaks circled in blue are of a neutral character open to some degree of interpretation, not marked definitely long or short, but when we compare the manuscripts to each other, the correct reading becomes clear. My position is not that the Laon Gradual is "disruptive or inconsistent with what those more aligned with Cardine hold about chant," but rather that it is a preeminent manuscript source of great antiquity that merits serious attention. It absolutely refutes the long-short pes and clivis misinterpretation of the Solesmes method, but I suppose that inconvenient fact alone may be disruptive and inconsistent enough! Acknowledging the length of both notes doesn't require any modification to the notes of the Vatican edition, just a longer episema. If the golden rule really means no long notes at syllabic breaks, it isn't much of a rule when it applies only 14% of the time.
    Ostende.png
    1094 x 792 - 374K
    Ostende376.jpg
    1182 x 308 - 85K
  • MatthewRoth
    Posts: 3,437
    But you have very particular ideas about what is true or not true about chant. I don’t really care about the long-short thing. I don’t, in the end, for pretty much every reason that Dr. Weaver has laid out, just as much as I don’t sing with historical ternary meter for polyphony as much as I’d like to do that.

    But I think that you’re in a bit of a tough spot if people (as early as Gajard) have more clearly acknowledged the problem and when those folks today already agree with you but don’t clearly agree about the need to really focus on Laon and what it brings.

    I don’t speak for Solesmes and the Atelier paléographique of course. But I would say that pretty much no one has a problem with the examples from the end of word syllables so that may be why folks like me sort of shrug our shoulders in the end.

    But anyway my point is simpler: I think good, reasonable scholars not singing from Laon is probably bad if what you say is true, but on the other, that’s the case, so we should ask why this is.
  • FSSPmusic
    Posts: 488
    If you "don’t really care about the long-short thing," then why make the claim that, "Mocquereau is not wrongly applying the golden rule"? Your position seems incongruous, as Mocquereau's misapplication of the golden rule presumes that it involves proscribes long notes before syllabic breaks rather than breaths or rests. As I have shown above, this has nothing to do specifically with Laon and can be demonstrated entirely from the St. Gall sources.
  • I don't think we are really in disagreement, nor do I think you're really refuting my point. I have suggested, based on what I have read of Mocquereau, that the reason the episema-over-clivis figure is drawn over the first note only in the Solesmes editions (with a few exceptions at ends of sections, like in the introit for Advent I) is because he felt constrained by the rules (specifically the so-called "golden rule") laid out in the preface to the Vatican edition, which he did not write. I do not think Mocquereau did this because he believed, as a matter of semiology, that the non-cursive clivis was long-short. The upshot of this is that singers working within the Solesmes/Fontgombault tradition should feel empowered to lengthen both notes of the clivis-with-episema; and indeed this is something that I have observed in the oral tradition.

    To the extent that there are singers who lengthen only the first note, this seems like a predictable consequence of the particularly acrimonious debate over what signs should be allowed in chant editions in the early twentieth century. I'm not sure there's really anyone defending such an interpretation on semiological grounds. Of course, people are free to follow the notation they are singing from, without delving too much into the history behind it. What lesson can we draw from this? Especially as the Church has moved away from a heavy-handed involvement in performance practice and towards a hands-off lack of implementation of the pro-chant language of Sacrosanctum concilium? Can you imagine an official of the DDW now telling one of us that we cannot extend an episema over an entire two-note neume in a published edition because it would undermine the interpretation of the Vatican Edition?

    In Duruflé, I'm trying to think of examples where such a single-note-lengthening happens. As far as I can recall, Duruflé often either ignores the episema or applies it to both notes. One spot where that is not true that comes to mind is the podatus in "In paradisum" at "suscipiant," where just the first note is lengthened. Maybe I even sing it that was as a result of listening to Duruflé!
    Thanked by 1FSSPmusic
  • FSSPmusic
    Posts: 488
    1. The thirteenth-century golden rule was a proscription against breath before a syllabic break, not against a long note there.
    2. The Preface to the Vatican Edition misinterprets the golden rule as pertaining to note length.
    3. The Solesmes editions do not respect the long note values of the various manuscripts at syllabic breaks, both single-note and longer neumes, presumably because of the golden rule, but see no. 5 below.
    4. This is a fundamental difference between the classic Solesmes method and Cardine's semiology.
    5. The Solesmes editions often write a long-short form of the clivis and pes in contexts other than immediately before syllabic breaks, even within a melisma far removed from the syllabic articulation, and the clivis sometimes also before quarter and half bar lines. Now here's my question: What Vatican edition rule applies in such contexts that would have constrained Mocquereau?

    Here is my initial claim above, which opened up this whole can of worms:
    [Mocquereau] managed to convince nearly everyone that the long clivis and pes quadratus were long-short figures rather than entirely long, based on a faulty application of the "golden rule" principle, and he erroneously treated the normal syllabic value as short and indivisible on that same basis.
    I stand by what I wrote. Are we really not in disagreement here?
  • FSSPmusic
    Posts: 488
    image
    In the third instance, the Solesmes edition contradicts the melismatic mora vocis of the Vatican edition.
    image
    Ostende2.jpg
    577 x 135 - 23K
    Ostende3.jpg
    2958 x 495 - 205K
  • MatthewRoth
    Posts: 3,437

    The thirteenth-century golden rule was a proscription against breath before a syllabic break, not against a long note there.


    Certainly that it is your interpretation. But even if it’s right: in practice long notes lead to breaths. So I see how Pothier got there, and if Mocquereau has to work within that, then I can also see why he wouldn’t push back.

    Notably, and this is why I’m just not that interested in the long-short question, in actual practice coming from Fontgombault no less you hear long-long at places where it will not introduce a breath in an inappropriate place.

    Whether they lengthen the second note mid-word is another matter. The In Paradisum is a place where this can actually work practically. Dom Gajard certainly hints at that the possibility. But context matters too.

    I suspect the tradition of the abbey, consistency, and context are the answer as with nearly everything. In other words lengthening both notes when only one is written long at a bar would not faze Mocquereau and that he felt that it was redundant. Again, take the Nocturnale chants. That’s a perfectly valid approach but honestly I’ve heard a lot of what amounts to yapping from people who don’t like rhythmic signs at all; I would say that they should be happy that Mocquereau did not intend (as best he could and with some exceptions) to completely determine performance by means of musical notation and that you had to know the bigger picture or otherwise judge for yourself. Maybe this is why Saulnier went the other way and removed the signs (to our detriment given the abbey’s change in official interpretation in the new Liber Hymnarius).