Around 1600 a remarkable piece of music surfaced in a publication entitled Delle imperfettioni della moderna musica by Giovanni Artusi. It is a piece that, without a doubt, is still one of the lingering, unsolved riddles of music literature. It confounded the musicians of it's day, both performers and theorists, and debates over the work have lasted well into the twentieth century. The piece was a two-voice composition by Adriano Willaert. Artusi had placed the title "Quidnam ebrietas" on the duo. Unlike any other work of its day, this composition apparently ended with the interval of a seventh between the two parts. No text was printed with the music, but instead, Artusi includes a letter from Giovanni Spataro to Pietro Aron, dated September 9, 1524, which discusses this unique piece. Following the letter, Artusi tries to explain the solution given it by the theorists of his day. He goes on to share his personal viewpoint as well. In each case, the work is examined from the standpoint of mathematical acoustics only.
From all indications, the letter published by Artusi was just one in
a series of letters between the two famous sixteenth-century theorists,
Spataro and Aron. Unfortunately, most of the letters written by Aron
to Spataro have been lost. The letters of Spataro, for the most part,
have been preserved and one in particular gives some insight into the background
of the duo. It is dated May 23, 1524 and reads:
"I think three years have passed or perhaps four since I was told
by... a private singer of Pope Leo, that a duo which ended with the
interval of a seventh, was sent by Messer Adriano, a very famous
musician attached to the court of the Duke of Ferrara, to his Holiness
Pope Leo. He told me, furthermore, that the singers of his Holiness
found it too difficult to sing; it was then performed on viols but not
particularly well. In the meantime, the thought of this composition
intrigued me, and having always cherished a desire to learn as much
as possible, I kindly requested one of my Bolognese friends who at
present lives in Ferrara to intercede, and as the result received a
copy of the duo, specially written for me from Messer Adriano himself,
which was sung, played and carefully examined by our Bolognese musicians.
It was praised as a very fine and learned work. Since there
are many among us who are very critical and entertain doubt in this
matter, I am sending your excellency a copy of the duo in order that
you and other Venetian musicians may have an opportunity -- perhaps,
the last, as this composition is very rarely seen in our times --
to examine and to resolve it, and in order that I may learn what
you candidly think of it." (Levitan, p. 174-175)
If we accept the letter of Spataro as correct, then we must assume that Willaert composed this piece before 1520, at a young age, since he was born around 1490. This gives evidence of the progressive ideas of this master even as a young composer. When attempted by the musicians of Pope Leo, it evidently proved to be too odd, too daring, and too puzzling for them to solve its riddle and perform it. It, according to the letter, was also studied and analyzed by the musicians of both Bologna, with Spataro, and Venice, along with Aron. Without question, the appearance of this duo created quite a sensation in its time, and probably contributed to Willaert's appointment at St. Mark's in 1527.
The duo has been classified as a madrigal by many authorities. Its style is characteristic of the madrigal, employing contrapuntal treatment with frequent points of imitation. As the most artistic and expressive form of secular music of that time, the madrigal was the experimental testing ground for every emotion of musical thought, which made it an ideal medium for Willaert's enigmatic duo.
Since the text weighs so heavily in this concept of musical expression,
consideration must be given to Willaert's choice of text. Although
Artusi published no text for the duo, he did allude to a text based on
Horace's fifth epistle. In one of Spataro's letters to Aron he refers
to the duo having the text "Quid non ebrietas designat..." in the tenor
voice. From this indication, we must assume that Artusi's publication
is correct in connecting Willaert's duo with the fifth epistle of Horace,
which contains the same poetic text. Coming to this conclusion raises
two other issues. First, if the text is based on Horace's fifth epistle
which begins, "Quid non ebrietas designat...", then Artusi must be in error
in publishing the duo with the title "Quidnam ebrietas." Secondly,
since the text is Latin, the piece should not be classified as a madrigal
and suggests instead that it be regarded as a motet, secular and humanistically
inspired, which was a common form of the day.
One other piece of evidence concerning the text comes in the form of
an alto part discovered later that effectively fits with the soprano (cantus)
and tenor published by Artusi. This will be discussed in more detail
later, but at this point, suffice it to say that this alto part has Horace's
text included with it, which would help solidify the assumption made by
Artusi and mentioned by Spataro.
In Horace's fifth epistle, the poet invites his friend Torquatus to
cease for a day the struggle for wealth, dismiss his legal clients, and
join him... for dinner in the country: flowers, friends, good conversation,
and above all good wine. He then praises man's state of mind under
the influence of wine. "What a miracle cannot the wine-cup work.
It unlocks secrets, bids hopes to be fulfilled, thrusts the coward into
the field, the load from anxious hearts, teaches new arts." (Lowinsky,
p. 31) How appropriate the text becomes in light of Willaert's composition.
The later discovery of the alto part raises questions of its own.
Was the duo conceived in the form published by Artusi or was it originally
written for four voices? No bass part has yet been uncovered to give
this piece its full complement of voices, but correspondence between Spataro
and Aron indicates that Willaert originally composed the piece for just
the two voices then later enlarged it to a quartet. Spataro, for
purposes of theoretical study, requested only the original two parts be
sent to him from Willaert. It was through Spataro's thirst for knowledge
and his request for a copy that the original duo survives.
What is it about this duo that is so intriguing? How could a piece
of two-voice music be such a mystery and inspire ongoing debate for hundreds
of years? In our quest for insight, let us focus our attention on
the music itself. Looking at the unaltered version of Figure 1, the music
looks relatively straightforward and not extremely difficult. There
are a few accidentals provided by the composer, but nothing that appears
dramatically chromatic. On closer inspection of the contrapuntal
relationships, nothing seems out of the ordinary until we reach measure
21. At that point, the tenor voice seems to go wildly astray.
It suddenly appears to be moving in intervals of a ninth and then tritones
and sevenths. From here to the end of the piece, there are intervallic
relationships that are completely contradictory to the common practice
of the day. How could this be?
The obvious solution would be musica ficta , the common practice of
that day to alter notes chromatically in order to avoid undesirable intervals
between the voices and in the melodic flow, and to create sonorities more
beautiful in character. The musicians in Willaert's time would have
understood and known where to alter the pitches without the composer having
to write it into the parts.
Figure 2 is a harmonically altered version of the duo with musica ficta added to enable us to see how the tenor line changes to avoid inappropriate musical collisions. The alterations look very typical in the first few measures of music. In measures 11 and 12, the E-flat is added to avoid the interval of the tritone with the soprano B-flat. In measure 13, the A-flat is added to avoid the melodic leap of a tritone from the E-flat. In measure 15, the composer has given us a D-flat, so in measure 17 the D-flat is retained to maintain that harmonic character. This creates a need to change the A to A-flat in measure 18. We can see that from this point on, the chromatic alterations begin a type of domino effect -- a spiraling down of altered pitches in order to maintain the intervallic and melodic consonance.
What we end up with in the tenor line is a section of eighteen measures where every note is double-flatted except for the F, which has simply become F-flat. Has Willaert lost his musical map? Is there any logical direction or structure to this seemingly bizarre trek? If we look at the harmonic progression outlined in the tenor, we will discover something extremely unique for this period in musical history.
Willaert begins the piece on G then quickly progresses to C and F. In measure 10, we see a strong cadence in G minor, which, as we know, is the relative minor of B-flat major. Willaert's progression of tonality should be quite obvious by now -- he is taking the tenor line through the circle of fifths. From measure 11 he continues his journey by passing through E-flat to A-flat then D-flat. This systematic cycle takes the tenor line completely around to its final cadence on E-double flat. This in itself is an interesting notion, but it becomes even more intriguing because the soprano line never really modulates. By our system today, there would be absolutely no conflict whatsoever with having the tenor on E-double flat and the soprano on D. Enharmonically they are the same pitch... or are they?
Before we deal with the issue of enharmonics, another topic needs to be touched upon briefly. Modulation of this sort was nonexistent in the early sixteenth century. Composers did not allow their music to stray far from its initial tonal/modal center. When a piece employed an accidental, it basically shifted the hexachord -- a type of temporary modulation. A flat added to a note indicated that that pitch was now functioning as fa . A natural or sharp given to a pitch indicated the location of mi . A typical musical work of this era rarely strayed more than one or two accidentals away from its original center.
Willaert took this principle of the shifted hexachord and extended it to its limit. Instead of a typical, temporary shift of center, the accidentals placed in the tenor of our duo never allow the hexachords to return to their point of origin, but continue to move them farther away with each added alteration, until they complete the circle. Willaert's shifted hexachords break with the accepted practice and explore the realm of modulation, which was a bold and daring innovation.
It is hard for us to realize, with today's musical practice, just how controversial this modulation concept was in Willaert's time. There were accepted laws of acoustics that would not allow music to be performed in multiple tonal centers. An accepted concept of enharmonic pitches did not exist. As alluded to before, an E-double flat was not the same pitch as a D.
For around two thousand years, the discoveries of the Greek mathematician Pythagoras, had dictated the way in which music was tuned. Every pitch relationship was based on a mathematical formula. These formulas manifested themselves through a system of dividing the octave. Today we take for granted that the octave is equally divided into six whole tones and each whole tone is divided into two equal semitones. At the time Willaert's duo was composed, this was not the case.
The practitioners of Pythagorean tuning did not divide the whole tone into two equal semitones. It was divided into two unequal semitones, one greater and one lesser. The greater semitone was referred to as a major semitone and the lesser, a minor semitone. The octave was divided into five whole tones, not six, plus two minor semitones. In this system of tuning, if a pitch was flatted, it would be lowered by a major semitone, leaving the minor semitone as the distance to the pitch immediately below it. If a pitch was sharped, it was raised by a major semitone, leaving the minor semitone as the distance to the note immediately above it. Therefore, F-sharp is not the same pitch as G-flat. F-sharp would lie closer to G while G-flat would lie closer to F. This system was the main reason for limited tonal/modal centers, especially if keyboards were incorporated. The keyboard instruments of the day were tuned using this Pythagorean system. Choices had to be made concerning how to tune, for lack of a better phrase, the black notes. The pitch between F and G was generally considered an F-sharp, not a G-flat. The pitch between A and B was tuned as a B-flat, not an A-sharp.
This predetermination of accidentals, did not allow for what Willaert was proposing with his duo. As we can easily see, this would present significant problems for his concept of modulation. Let us take the last pitch of the duo as an example. The E-double flat would be located two semitones lower than E, but with Pythagorean tuning, these would be major/greater semitones (more than half the distance to the lower whole tone) and therefore combining two would put the pitch slightly lower than a D, making the distance from the tenor E-double flat to the soprano D more than an octave. Needless to say, if this is the case for the last pitch, it would certainly be the same for each of the other unusually altered tones.
Why would Willaert compose the duo in this manner, knowing the common practice of his day? Many believe he was advocating equal temperament. They associated his ideas with those of Aristoxenus whose system of tuning divided the octave into six equal whole tones and in turn, each whole tone was divided into two equal semitones. This would allow every one of the twelve pitches in the octave to be used equally and a tone or mode to be centered around any one of them. It would also make the concept of enharmonics feasible as well as the practice of modulation practical. If this system of tuning had been established, the duo would have worked perfectly well.
Before any conclusions are drawn, one matter needs to be attended to. Since it was established earlier that an alto part had been discovered, Quid non ebrietas seems to call for a bass part to be reconstructed and allow us to see and/or hear the piece in a fully-rendered form. Edward Lowinsky has accomplished that task for us. With the addition of the alto part, and a reconstructed bass line, we see Lowinsky's version of Willaert's duo, now fully expanded, in Figure 3.
What can be said of such a piece? It combines a light, humorous text with such advanced theoretical composition. It's a piece so bold that, inspite of its mere 40 measures, was destined to intrigue the best theoretical minds, to stimulate the finest creative forces of the century, and to point the way to the style of the future by ushering in equal temperament and modulation in the circle of fifths. (Lowinsky, p. 31)
It is a coincidence of symbolic significance that in the year 1519 -- in which Ferdinand Magellan undertook the first circumnavigation of the globe -- Adriano Willaert composed a piece which, for the first time in history, circumnavigated the whole tonal space through the complete circle of fifths. It was the same spirit of adventure, the same desire to open new and unexplored spaces that lured the sailors across the sea and beckoned the musicians to their discoveries of remote and distant keys and new harmonic conquests. (Lowinsky, p. 22)
Adrian Willaert's Chromatic "Duo" Re-examined
by Edward E. Lowinsky