Low B-flat

I'm working on a project of performance editions of Dominican chant using medieval manuscripts of Dominican chant sources to correct the versions in the mid-20th century Dominican Graduale. One issue I've come up with for a couple of chants (e.g. Offertorium 'Meditabor' and communio 'Domine dominus noster') is that the mid-20th century edition adds a flat on certain low B notes that is not present in the manuscript sources. I notice that in parallel chants in the Solesmes editions there is no low B-flat given. I understand that the low B-flat was not included in some medieval theoretical systems, but is there any consensus presently about whether a low B would have been sung flatted or not?

You need to be a member of Musicologie Médiévale to add comments!

Join Musicologie Médiévale

Email me when people reply –


  • Hello, Fra Innocent -

    Re. the music-theoretical aspect of the question -- 

    As I'm sure you know, the Guidonian Gamut, which was the standard medieval scale system from about the mid-11th century, especially for plainchant (as opposed to polyphony), lacked the Bb (properly B-fa) in the lowest octave.

    But treatises of the south German school (later 11th-12th c.) do include it, or argue in favor of its inclusion. See, for example, Wilhelm of Hirsau, Musica, cap.xx, cap. xxxvi & cap. xxxviii. The note in question is there referred to as "grave synemmenon" and "inferius synemmenon." (Whyb-flat was called synemmenon in the first place is explained in The Cambridge History of Music Theory, ed. Thomas Christensen, Chap. 11, pp. 320 & 340-41,)

    I don't know whether this information is of any help to you in your project, but I thought it might perhaps be of interest in any case.

  • Thank you all for your helpful comments. The Dominican sources and the methodology of the project are outlined here: https://cantusop.wordpress.com/2016/08/22/graduale-op-introduction/ 

    The two Dominican manuscript sources I'm primarily consulting (Rome, Santa Sabina XIV L1 and London, British Library Ms. Add. 23935) both present the offertory 'Meditabor' with the F-clef rather than transposed C-clef. I've attached two different ways of notating the flat -- one by putting the flat in the line and the other in the "ficta" modes. I'd welcome feedback on these two approaches.

    I appreciate all of your insights!


    Introducing the Graduale O.P. project
    The Graduale O.P. project is aimed at producing a new edition of the Graduale drawing on Dominican chant sources and arranging them for use in the co…
  • I can only second the points made so clearly by Franz Karl Prassl. His point about transposition in the 2nd mode graduals is especially à-propos.  The Haec dies family, for example, is notated in the Graduale Romanum (and in many MSS) with final on a and a signature of B-flat, thereby enabling the singing of both low B-flat and the E-flat immediately above the final. There are references to the low B-flat in other theoretical sources, as well, the most extreme, perhaps, being the first treatise of the Berkeley manuscript.  Via the coniuncta it can rationalize chromatic inflections of virtually every degree of the Gamut.

  • Thank you to Franz Prassl for his answer. As I told, the b flat is perfectly right in the quoted examples . In any case, Guido suggests a transposition, and this confirms the b-flat !

  • Cher Père,

    Pour commencer, je serais heureux de savoir sur quelles sources dominicaines vous travaillez...

    J'ai l'habitude de construire des tableaux synoptiques, et cette méthode permet, dans beaucoup de cas, de connaître  l'avis de la tradition médiévale (la construction de tableaux, grâce aux procédés informatiques, s'avère extrêmement rapide. Pour information, il ne m'a pas fallu fallu plus de trois heures de travail pour construire un synoptique de 70 témoins, de zone germanique, pour la COM Scapulis suis).

    Le cas du SI bémol au grave s'avère plus complexe. Cependant, une observation incessante de la tradition permet de confirmer un fait très connu ici, à savoir que l'absence de signe ne signifie pas l'absence de pratique.

    Bien que ce ne soit pas le sujet, il me semble que la tradition dominicaine est plutôt cohérente sur le témoignage du SI bémol aigu (bien que quelques divergences puissent s'observer, par exemple au sommet mélodique de l'ANT Nos qui vivimus). Ces fluctuations mineures s'observent parfois aussi chez les chartreux...

  • 1. A low B WAS SUNG but very often not written. According to general theory they used very often a transposition to the higher 5th. Finalis D = Finalis A. Transposition was the miraculous tool to hide forbidden notes (forbidden only in theory) One may find it in the graduals of mode 2 e.g.

    2. Some manuscript ignore theoretical prescriptions and write a low B flat too, like A-Gu 8o7 (Gradual of Klosterneuburg) ands the Klosterneuburg antiphoners.

    3. Comparing sources helps to decide whether to use B flat or not. Very often a SI bemolle may be a SI naturale.

    4. In modern edition there is no reason to avoid the low B flat. Why should we avoid it? We are not the disciples of Guido.

    5. Such questions should not be decided by theoretical considerations, but with an analysis of the real practice. 

    For such questions consult: Graduale Novum, Regensburg - Rom 2011

  • Yes, Charles, you are right, this is the exception that confirms the rule. I won't reopen the question of the strange scale of the Musica enchiriadis, but we must agree, I think, that it is an exception, to say the least.

  • The tone-system of the Musica enchiriadis includes a low B-flat in the tetrachord of the Graves, G A B-flat C in modern notation.

  • If the the problem is essentially theoretical (as it is), I think that the b flat, in these two pieces, could be accepted in a practical edition without problems, right?

  • The low B-flat, unless I am mistaken, was never included in the medieval theoretical system.

    The usual way to explain this is by considering the hexachords: the first hexachord was built on Γ (Gamma) and went up Γ A B C D E, with B♮. In order to obtain B♭, it would have been necessary to add a hexachord from F, which was not done before about the 14th century.

    A tetrachordal conception of the musical system would have produced the same. The theoretical system would then begin on A, with the first tetrachord A B♮ C D. There was no "synemmenon" (=additional) tetrachord below G a b♭ c an octave higher. This appears to be the doctrine in all treatises, starting with Hucbald, dealing with the musical system.

    That chants may have been transposed to avoid this (essentially theoretical) problem, as Charles M. Atkinson rightly suggests, raises additional questions. The transposition, indeed, may exist in the notation exclusively; the question, in other words, would then be whether the chant would have been sung in its entirety at one single pitch standard, or the "transposed" part, if transposed say a fifth higher, would have been sung a fifth lower than notated.

    Such questions certainly will remain unsanswered, as it is unlikely that we will ever discover medieval sound recordings. But your question, "is there any consensus presently about whether a low B would have been sung flatted or not?" raises another one, to know whether such a consensus could have existed in the Middle Ages. I cannot imagine it. The theoretical descriptions of musica ficta seem to indicate changing usages or, even more, variable usages - that is, usages that may have varied depending on the mood of the moment.

    If the Offertorium Meditabor that you mention is the one reproduced in the Graduale Romanum (1961 edition, p. 114 or 301), you will note that it is written there transposed on A: this obviously is in order to avoid the problem of the B♭ on dilexi, now written as F. This notation of course has no incidence on the pitch level at which the piece could be sung. Domine Dominus noster, p. 117, is written without transposition and with B♮, even although the passage, on universa, is quite similar. In both cases, the melody goes F D C B(♮?♭?) = c a G F[♯?♮?] - the hesitation may arise in the case of B, much less in that of F. The transposed version avoids the tritone, the untransposed one does not.

    It would be wrong, I think, to presuppose that a consensus existed in the Middle Ages as to how this was to be sung: the choice may have been done "by fantasy" (per lasciviam), and neither version can be considered more "authentic" than the other. Modern performers, of course, will need a clear answer; but I think that the sensible one would be to tell them that they may choose, as in the Middle Ages...

    [On these questions, see also my paper "Jacques de Liège et la pratique de la transposition partielle", Revue belge de musicologie 47 (1993), pp. 43-48, reproduced in Musica. Liber amicorum Nicolas Meeùs, Presses universitaires de Paris-Sorbonne, 2014, pp. 171-77, where I comment other cases, mainly Alleluia iudicabunt and Terribilis est.]

This reply was deleted.


and your logo here...

 We need other partners !


Soutenir et adhérer à l'Association Musicologie Médiévale !

Support and join The Musicologie Médiévale Association!