music, so this distributional layout came to be called “choirbook style.” The Bamberg Codex is one of the
earliest examples of it.
We do not know exactly when or where the virga and punctum were first used to represent the long
and the breve. It happens in practical sources, like Ba, before any surviving theoretical source discusses
it. (In fact, hints of such a distinction can be found in chant manuscripts as far back as the tenth century.)
The earliest known theorist to prescribe the practice was Magister Lambertus, probably a University of
Paris instructor like Garlandia but of a later generation, in a treatise of ca. 1270. His description of
mensural shapes and their relationship closely accords with the notation found in Ba, which probably
means that Ba, though possibly copied in Germany, contains a Parisian repertoire. (The piece in Fig. 7-3
is obviously Parisian, of course, since it is just a texted version of a Notre Dame clausula.)
The fullest discussion of early mensural notation is found in a famous treatise called Ars cantus
mensurabilis (“The art of measured song”) by a German writer, Franco of Cologne, whose name has
become attached to the notation he so definitively described. The principles of “Franconian” notation,
first formulated by ca. 1280, though much supplemented and modified over the years, basically held good
for the next two to three centuries.
Despite the mensural breakthrough, and not to take away from it, Franco’s rhythmic notation did not
absolutely transcend or replace the contextual aspects of “modal” notation. It represented a compromise
of sorts between the intrinsic and the contextual (which is why there had to be all that supplementing and
modifying over the years). A virga unambiguously represented a long rather than a breve, but that long
could either be a perfect (three-tempora) long or an imperfect (two-tempora) long, depending on the
context. In Fig. 7-3, the longs that alternate trochaically with breves in the motetus (texted duplum) and
triplum parts are imperfect, while the longs that congregate spondaically in the tenor are perfect. A
punctum unambiguously represented a breve and not a long, but that breve could be a normal one-tempus
breve (brevis recta) or a two-tempora “altered” breve (brevis altera) as originally devised for the
dactylic or “third mode” meter at Notre Dame, depending on the context. The contexts, which can be
complicated, are spelled out to a degree in the accompanying table (Fig. 7-4), which outlines the basic
principles of Franconian notation.
One of the cleverest Franconian innovations had to do with ligatures, where some apparently new
graphemes were introduced. The new shapes, however, were based very systematically on the old. As
observed in the previous chapter, the usual Gregorian binariae—the pes (ascending) and the clivis
(descending)—happened to assume the rhythm BL in the first and second rhythmic modes as specified by
Garlandia. Under the Franconian rules this rhythmic assignment was made intrinsic to the shapes
irrespective of context. Then the fun began.
When written in their familiar pes and clivis forms, binariae were “proper” and “perfect.” The
former word applied to the appearance of the first note, the latter to the appearance of the second. If the
first note in the ligature departed from its normal shape, whether by adding a tail to the pes or taking it
away from the clivis , then the first note received the opposite meaning and the ligature became LL. If
the second note in the ligature departed from its normal shape, whether by reversing the termination of the
pes or making the square termination of the clivis oblique , then the second note received the
opposite meaning and the ligature became BB. If both notes were affected—whether or —then the
whole ligature received its opposite meaning and became LB. That covered all possible two-note
combinations. Additional notes were considered interpolations and were always read as breves.
EX. 7-2 Principles of Franconian notation