article 7:   Cavazzoni and the Birth of the Harpsichord Toccata

Additional notes to
NAXOS 8.572998: Marco Antonio Cavazzoni, Complete Works

Except for dances, Italian keyboard music of the 16th century - a time of extremely important developments in the history of the genre and for instrumental music in general - is consigned in the musicological literature and in editions, over and over again, exclusively to the organ. Prima facie this ought raise eyebrows, but in the light of massive evidence to the importance of the harpsichord in the period, not only in Italy but in the rest of Europe, this automatism is nothing short of astounding. Harpsichordists, with their lopsided fixation on the 18th-century literature (Bach, Scarlatti, Rameau, and Couperin le Grand ad infinitum) are partly to blame for the loss of some of their richest repertoire.

Let a few selected title pages begin to illustrate the organ fallacy:

  • The first collection of keyboard music printed in Italy, Andrea Antico's Frottole Intabulate (1517, a collection of frottolas arranged for keyboard - per sonar organi ) shows the editor seated at a harpsichord.

  • Musica Nova (1540), the landmark collection of ricercars for keyboard or ensemble, which survives in a single bass part book, can be reconstructed only by using the Lyons reprint called Musique de Joye, which was published as a means of learning to play "les epinettes" (a term encompassing the harpsichord and its smaller cousins) and melody instruments. The organ is not mentioned.

  • Antonio Valente's Intavolatura de Cimbalo ("Harpsichord Tablature") (1576) contains (besides dances) fantasias, ricercars and a Salve Regina. Again, there is no mention of the organ.

For music from outside of Italy, similar prejudices in favor of the organ can consistently be observed. One striking example is Bernhard Schmidt the Elder’s Zwey Bücher Einer Neuen Kunstlichen Tabulatur auff Orgel und Instrument (1577). “Instrument” at this time translates as “harpsichord”. Das Erbe Deutscher Musik published the collection in 1997 as an “Orgeltabulatur”, and the editor does not mention the Instrument once in his extensive prefatory material.

Many non-musical sources testify to the harpsichord’s importance and popularity in 15th- and 16th-century Italy. Letters and other documents in official archives reveal that the line of musical patrons which I would call the most distinguished in all of musical history were harpsichordists: Ercole I, Duke of Ferrara, who established that magical city as the center of the musical world around 1470, owned several harpsichords housed in a special music room in the castle, and sent one to his famous son, Cardinal Ippolito. Duke Ercole was consoled on his deathbed by Vincenzo da Modena’s harpsichord playing. Ippolito and his sister Isabella d’Este (who was called the “First Lady of the World”) had harpsichord lessons from Girolamo da Sestola; and Isabella’s daughter Eleonora was the patroness of the harpsichordist Marco Antonio Cavazzoni after she became Duchess of Urbino.

Also belonging to the first century of the harpsichord’s existence are the following three highly prominent performers:

- Isacco Argiropulo (Argyropoulos) was given a contract as harpsichordist in 1472 by the Duke of Milan, Galeazzo Maria Sforza, as was already revealed in an article from 1881. This brutal tyrant, who quickly built his private chapel into one of the finest in Europe, names Isacco as “nostro Cortesano et Sonatore de gravacymbolo”. The organ is not mentioned in the document; nevertheless, in all literature and lexica he is consistently referred to only as an organist and organ builder. Isacco was later appointed cubicularius secretus (private chamberlain) of Pope Sixtus IV, possibly the first mention of keyboard player in such a capacity at any court (Pirotta).

- The Spaniard Laurenzo of Cordova, documented from 1476 at the brilliant Aragonese court of King Ferrante I of Naples, and given an introduction by him to Lorenzo de’ Medici in 1484, is praised for his interpuncta facilitas (able precision) on the Gravecordium by the papal secretary Paolo Cortese (or Cortesi) in his famous treatise De Cardinalatu (1510). This term has been incorrectly translated (Pirrotta, Atlas) as “clavichord”, but clavicordio was the Spanish name for the harpsichord, and the prefix “grave-” is only associated with this instrument.

- The earliest-born known composer of keyboard ricercars, Jacobo (or Giacomo) Fogliano da Modena, whose memorial plaque can still be seen in the cathedral where he spent most of his long life, is already described when he was still a teenager in 1483 as a master of the harpsichord.

Here are a few later sources of various kinds:

  • A correspondence of 1524 regarding Adrian Willaert's controversial chromatic part-song "Quid non ebrietas" refers to the famous theorist Pietro Aron playing the piece on his harpsichord (which, by the way, puts the instrument at the center of furious arguments about tuning and temperament).

  • Ortensio Lando's book of lists (1552) mentions two of the most important keyboard composers of the era as harpsichordists only: Giulio Segni, the initiator of Musica Nova, is "most excellent on quilled instruments", and the great Jacques Brunel of Ferrara is "miraculous on quilled instruments and expert in chromatic music" (cf. the previous remark).

  • Segni, one of the greatest masters of the early ricercar, is mentioned again in Cosimo Bartoli's Raggionimenti Academici (1567), which includes a list of active keyboard players in which the harpsichord dominates. Segni is said to play beautifully on the organ, but to be "worth far more" on quilled instruments. Bartoli tells two stories of important political conferences, one of them involving Pope Clement VII, where Segni's quiet background music on the harpsichord caused the pressing subjects at hand to be dropped, and those involved to wander over to the instrument and listen, spellbound. The harpsichord is specified in the source, but an organ, with its accompanying inconvenience and indiscretion of a bellows-blower, would be in any case inconceivable in such situations.

  • Bartoli also has one of his speakers tell of hearing "il Moschino" (the nickname of the Florentine Baccio Moschini) play alone on the harpsichord for an hour "for his own pleasure and study", causing his listener to be cleansed of all bitterness and anger. He played the whole while "in contrabattuta", in which style he is said to be unrivaled; I take this to mean what was later called "senza battuta" or "con discrezione", in which case we may have here the earliest reference to free preluding on the harpsichord.

  • The Venetian envoy to the Papal court reports as follows on a 65-course dinner given by Cardinal Francesco Cornaro, the dedicatee of Marco Antonio Cavazzoni's epochal publication of 1523 which is the focus of our recording, on the occasion of Isabella d'Este's visit to Rome: "At the end of the meal we rose from the table...deafened by the continual concert, carried on both within and without the hall and proceeding from every instrument that Rome could produce - fifes, harpsichords and four-stringed lutes in addition to the voices of hired singers."

  • A letter from Antonfranceso Doni to the sculptor Giovanni Angelo refers to Claudio Veggio, composer of a few brilliant ricercars preserved at Castell'Arquato, as a harpsichordist. (Both Doni and Veggio were from Piancenza, members of the local Academy of Gardeners dedicated to the garden-fertility god Priapus, which was disbanded by the authorities for bad behavior. Veggio makes a rare personal appearance in Antonfrancesco Doni's Dialogo della Musica (1554) as a feisty young fellow trying to interrupt the loose talk and get on with music-making; he is also jealous of the attention which the multi-talented Girolamo Parabosco (Segni's succesor as first organist at San Marco) is getting.

  • Vincenzo Galilei (the astronomer's father), in a critique of contrapuntal music found in his famous Dialogo (1581), criticizes composers for using extremely long notes, which render their performance on lute and harpsichord, "those noble instruments", impossible unless "the experienced performer" re-strikes them.

These are some of the results of a few winter evenings’ trolling of the internet; I feel certain they could easily be multiplied by a more diligent researcher than I.


Why is this bias in favor of the organ - which has slowly become as irritating to me as all those Urtext editions of Bach for “Piano” or Klavier - so powerful? I think it is mostly because of the massive presence of these formidable and dangerously powerful instruments in churches all over the world, under which the fragile remains of 16th-century harpsichords, largely mute and untouchable in museums, are simply crushed. Then there is the Roman Catholic Church itself, housing so many magnificent organs, with its tremendous influence on society at large, then and now. Nothing could be published in 16th-century Italy without her blessing. Even such a raucous gathering as that described in Doni’s Dialogo had to begin with the singing of motets to the Virgin. The harpsichord was a private instrument; for personal delectation, gatherings of friends or “Academies”, small concerts and entertainment during potentates’ meals, and for purposes of learning the keyboardist’s art at home.

There is also considerable confusion regarding the term “organo” itself, perhaps best illustrated by Antico’s cover mentioned above. The full title includes “per sonare organi”, which has to include the harpsichord, given the picture of Antico playing one on the cover. The literal German translation of the original Greco-Latin word “organon/organum”, Instrument, is an extreme example of this conflation in that it quite simply means “harpsichord”. The most influential theorist of the period, Zarlino, says the term organo covers “any man-made instrument”, the original usage that goes back to the ancient Greeks. The invention of the instrument we now call the organ of course goes back to Alexandria in the third century BC, and the term slowly shifted in later languages to apply, at least in music, to that complicated apparatus, without ever quite losing the original senses, which included “tool” and “sensory organ”. When the clavichord and harpsichord were invented in the 14th century, the fact of their having keys like their predecessor nudged them into the same general category.

Particular confusion has been caused for centuries by the Psalmist’s exhortation to praise God in organis, which simply means “with instruments”, and has nothing whatsoever to do with the modern organ, which had yet to be invented when the original Hebrew was composed (although St. Jerome, who wrote the Latin Vulgate translation quoted here certainly knew the organ).

The term “intavolatura d’organo”, meaning a keyboard score, is similarly ambiguous. The emphasis should be on “intavolatura”, the making of a “table” of many notes, not on “organo”; it just happened that organists were already looking for ways to notate their music before the younger instruments came along and took it over. It is telling in this respect that some of the earliest mentions of the clavichord, in its earliest form called the chekker, say that it is “played like” or “resembles the organ”. There are numerous examples of harpsichord music in sources described as being notated in “intavolatura d’organo”. The old, somewhat oxymoronic Spanish term canto de organo - “organ (or instrumental?) song” - is a similar case; it means “mensural music”, as opposed to canto llano - plainchant. Here again, a term was taken from what was conveniently nearby, although in this case proximity concerned the location of choir and organ, not a similarity of mechanism.

This association of organ and harpsichord is borne out by the extraordinary prevalence of the combined instruments called claviorgana in the 16th century, which culminated in the monstrosity containing a harpsichord and three separate spinets built in the 17th century for Palazzo Verospi Vitelleschi in Rome. Inventories from several countries cite dozens of them. A notable example is famous the portrait of Paul Hofhaimer in the “Triumph of Emperor Maximilian”, where he is shown sitting at a positive organ. But on the camel-drawn mythical float, a claviorganum (with “Baldachinorgel” and upright harpsichord elements) and a clavicytherium (upright harpsichord) in their protective cases are also clearly visible. In 1585 at Palazzo Pitti in Florence, Claudio Merulo played a particularly elaborate one, with a register that initiated a mock naval battle on its top, to the astonishment of four Japanese boys who had been sent by the Jesuits to Europe to illustrate their success in training the people of that otherwise stubborn nation.


Italian keyboard music of the 16th century centers on the dance, on arrangements of vocal music, but especially on the ricercar. The form began its life, as far as we know, as an improvisatory prelude on the lute, but the term came to mean almost anything (dances and variations excepted) played on solo instruments which involved no text. Eventually it developed into the most complex form of contrapuntal music.

This lost art of polyphony - the simultaneous interweaving of more than one melodic line - was once the glory and pinnacle of Western music. As it was practiced in Flanders and Italy during the early 16th century, it had the straightforward, muscular beauty of ancient Greek, and is now equally as dead to a world drowning in ephemera. Its Mannerist late flowering in the 17th and 18th centuries, the fugue, still echoes faintly, thanks largely to the harpsichord works of J. S. Bach (and his three-part ricercar in the Musical Offering returns the circle to improvisation). But the foundation upon which that last great master of polyphony built his edifices - the Italian ricercare for instrumental ensemble or keyboard - has been consigned, if anywhere, to university classrooms, where one imagines reluctant students of music history yawning as they consult their social media. Musica Nova represented the decisive shift in the form from something reflecting the word’s literal meaning (“to search out”), which evolved into the toccata, to the sober counterpoint that challenged composers to exercise their craft to the highest degree.

There are a very few examples of a type of ricercar which fits neither the category “improvised prelude”, nor that of “strict counterpoint”. A distinguished Canadian musicologist, Warren Kirkendale, has called the former “Aristotelian” and the latter “Ciceronian”, citing the Stagirite’s descriptions of extemporaneous exordia, and contrasting them with Cicero’s recommendations for careful preparation, as propagated by Cavazzoni’s friend and patron, Cardinal Bembo. This discursive, in-between type might, by the same token, be called “Epicurian”, since I think they were composed for gatherings of friends, which in the philosophy of Epicurus represent the highest form of pleasure. The prime examples of this third type are the two which appear in Marco Antonio Cavazzoni’s print of 1523, which forms the centerpiece of the recording for which this essay provides some background, and which since they were first discussed (Jeppensen, 1943) and published have been called “organ music”. There is, however, strong internal evidence that they represent Cavazzoni’s work as a harpsichordist, which is strongly documented:

  • We know of his youthful service at the court of Urbino because of letters from Duchess Eleonora in 1512 (the year when Cavazzoni's native Bologna was conquered by her husband) which call him "mio musico"; this implies work as a chamber musician, which could involve small organs, but more likely stringed instruments. Some years ago I saw, in the Museo Nazionale located in the palace where he played, a drawing of a small group of people around a harpsichordist marked "15th-16th century". I made a mental note at the time and thought no more about it. In spite of my best efforts to date, and those of curators at the museum, I have not been able to trace what may be a picture of Cavazzoni performing in a gathering similar to those described in Castigione's famous Cortegiano, set in Urbino. The frottolist Marco Cara and a few other musicians are mentioned there, but the brief period of overlap between Cavazzoni's and Castigione's residence in Urbino (the author left for Rome in 1513), the fact that the book was already begun in 1508, and the older man's close association with the peripatetic duke, have robbed us of a description of Marco Antonio d'Urbino role in intimate gatherings at the most refined court in the world.

  • A Venetian ambassador to the Papal court writes that Leo X kept Cavazzoni dal gravicembalo "very close to him".

  • The dedication of a 1523 edition of Petrarch refers to Cavazzoni's service with the Pope, and calls him the world's greatest master of the harpsichord.

  • The salary records of the papal Curia refer to him only as harpsichordist.

  • A papal singer reports back to Ferrara that Cavazzoni performed feats on the harpsichord at a concert for the Pope which were "miraculous, not to say which might resurrect the dead."

  • Pietro Aretino, besides addressing two of his famous letters to Cavazzoni, mentions him alongside Giulio da Modena in his comedy Il Marescalco as players of the "cimbalis bene sonantibus", a Biblical reference which puns on "cembalo".


The book of 1523, entitled Recerchari Motetti Canzoni Libro Primo, contains no reference to any specific instrument whatsoever; there is only a standard mention in the printing privilege, granted in the name of Pope Adrian VI, to the praise of God in organis, as explained above. It is divided into two parts of four pieces each, for a total of eight, then considered a perfect number. The first part is sacred, the second, secular; domains which are usually assigned to the organ and harpsichord respectively. That division can be easily affirmed for part two, heavily-ornamented arrangements of four chansons, which, while certainly playable on a chamber organ, rely to an extraordinary degree on long written-out trills. This points rather clearly to primacy of the harpsichord. So does the fact that (judging by their titles) the lost originals, efforts by Cavazzoni in the fashionable French-language form, are eminently of this world, not the next.

The first part, consisting of two ricercars each followed by an arrangement of a motet in the same key, requires considerably more comment. I hope to show that they, too, are conceived for the stringed instrument.

There can be no objection to playing Cavazzoni’s two motet arrangements on the harpsichord; all the harpsichord collections mentioned at the beginning of this essay contain sacred music, as do several dance sources. In France, three collections published by Attaingnant in 1531 containing ONLY music for the liturgy are presented for “organ, harpsichord or clavichord”. Private devotion, general keyboard study, and even performance in religious houses that had no organ are some of the reasons which can be given for this practice. Restricting such music to the organ on pious grounds cannot be justified.

Having liberated Cavazzoni’s part one from the organ monopoly, we must next sever it from the organ and from the church altogether. The main reasons for this are found within the two ricercars themselves. They are threefold.

First, range: The first ricercar uses the range G-e’’’, the second goes even farther, to F-f’’’. This is beyond the range of Italian organs of the time. In the 15th century this was fairly standardized at F-f’’. Larger “doppio” organs extended this down an octave, to cover pull-down pedal range. By Cavazzoni’s time extensions up to a’’, and down to C or CC were becoming common. It was the harpsichord and spinet which went up as far as f’’’. It is sometimes asked what was played up that high; these are the only pieces I know of that answer that question, with one exception, a set of mass versets by “Jaches” in Castell’Arquato, which goes up to d’’’. This piece looks texturally like harpsichord music, and I would suggest it is just such a work for such small chapels with no organ as I mentioned a couple of paragraphs back. Of course, it could also have been intended for a small organ with a higher range; such did exist, but not in sufficient numbers to justify an expensive publication such as Cavazzoni’s, assuming he ever had the organ in mind for it.

I am aware that organists like to claim an historical right to transpose an octave higher or lower at will if their instrument allows it. Both these ricercars could be played an octave lower on, say, the large organ at San Petronio in Bologna, but that only leads to a confused growling in the nether regions of the pipework, even if the principal registers are not drawn. Such an expedient is unnecessary in any case when there is such a clear historical alternative: harpsichords and spinets with C-f’’’ (or F to f’’’, which fits Cavazzoni to perfection) ranges. There is no need whatsoever to change the pitches notated in the source.

Secondly, the very texture of the two ricercars points clearly to the harpsichord. All music of the period which is definitely for the organ respects the flowing, connected lines which the instrument’s sustained sound requires. By contrast, the melodies in our two ricercars leap around wildly, and often break off suddenly, even at leading notes.The number of voices is also subject to constant and often sudden changes, as is the general Affekt. These pieces are floods of sudden inspirations; they manipulate a small number of motifs and modulate violently in a way never before seen in music history, fairly careening from one end of the emotional spectrum to the other and back again. All this is typical for the chameleon-like, thrust-and-parry character of the harpsichord, and is utterly foreign to the early organ literature.

In addition, the frequent parallel fifths and octaves in left-hand chords, which resemble the octave and quint registers of an organ, are also characteristic of the harpsichord’s wonderful dance literature, and are very rarely found in organ music, simply because they are redundant there. And to me, at least, the dissonant parallel thirds in the left hand against the repeated chords in the right at the beginning of the Recercare primo sound simply awful on the organ, whereas the transparency of the harpsichord renders them delightful. This is not Gothic music, as this instance, as well as some of the many errors in the print (if taken as they stand) might cause one to suspect. (Initial repeated chords like these are found in organ music as far away as Poland, and are not, as might be thought, indicative of the quickly diminishing sound of the harpsichord.)

Thirdly, both pieces are inconceivable in the liturgy because of their extreme length. All sources instruct organists to be brief; these two ricercars compare in duration with the massive ricercars of Jacques Buus. The Roman Catholic church was at this time under great pressure from the Reformation to simplify its ceremonials, and came within a hair of eliminating all church music except Gregorian chant a few year later at the Council of Trent.

The only clearly liturgical pieces which can match the 1523 ricercars in length come more than a century later, in Frescobaldi’s Fiori Musicali, which contain very long ricercars. But he supplies many cadences where, as he tells us, the organist can break off when necessary; and anyway, I think he was just showing off his latest masterpieces in print.

Cavazzoni’s pieces are not for the church; they are distant mirrors of the long improvisations on chamber organs and harpsichords which made Paul Hofhaimer, keyboardist to the Holy Roman Emperor Maximilian I, famous all over Europe. This might seem far-fetched, but it must be remembered that the Venetian Dionisio Memmo, another widely-travelled virtuoso, studied with Hofhaimer, and worshipped him as a demi-god. I think it can be assumed that Cavazzoni, as a young man in Bologna, heard and knew Memmo, and possibly even studied with him. Hofhaimer, who constantly sought out the companionship of humanists to supplement his poor education, was working at the time of his death in 1537 on a setting of Latin poetry which is purest Renaissance. His four surviving organ pieces are Gothic, and must be much earlier. I am reminded by this contrast of two adjacent bishops’ tombstones by Tilman Riemenschneider in the cathedral of Würzburg, where I live; one late Gothic, the other early Renaissance. And let me say again that the much younger Cavazzoni was emphatically on the later side of that divide.

So much for the internal evidence; but there is one more objection to performance of Cavazzoni and other ricercarists on organs. The way the ranks of the Italian church organ are built up weighs them heavily to the treble. There are no mixtures to balance the bass (although smaller registers break back when their pipes become too small). Large organs even sometimes have additional ranks of principal pipes in the discant. This shows, to my mind, that their main function was the harmonization of choral melodies (falso bordone). Recordings, as well as my own experiences as performer and listener, confirm that lower contrapuntal voices come through poorly. This, however, is a general problem with all big organs, as noted by Arnold Schlick in his Spiegel as early as 1511, which the Germans tried to solve by adding independent pedal divisions. It is no problem at all on the harpsichord, which is why I, for one, will always prefer it for polyphony. In addition to this difficulty, there are many passages in Cavazzoni’s ricercars where all the voices cluster in the lower range. These are reduced to mud on any organ.

The word “ricercar” first appears in a church context in 1531, in the report of a competition for the organist’s post in Treviso. It refers there to improvised music (preludes, postludes, responses, elevations) with no cantus firmus. The three preludes in the three Attaingnant collections mentioned previously (which rather surprisingly contain some organ music as advanced as anything from Italy) provide a sensible example of what a good organist might have provided: highly expressive, but stable, clear, mostly three-part textures.

Cavazzoni himself provides us with one ricercar which can be seen as suitable for the church, his only work not preserved in print, but rather in the priceless manuscripts at Castell’Arquato. It is of a suitable length; it stays within the organ’s range; the lines, while sometimes virtuosic, do not leap around; and the fugal sections are short and clear. It leads directly to the toccatas of the later Venetian school, and the contrast to the two pieces under discussion is striking.

Marco Antonio Cavazzoni, close friend of Cardinal Bembo, Pietro Aretino and Titian, was very much part of the artistic and literary life of Venice during its most famous flowering. Music often lags behind the other arts, because of its cumbersome, arcane burden of theory and the difficulty of achieving its practice at the highest level. In these two ricercars we see a brief flash of congruence, perhaps even of a leading edge of some kind. They are so far in advance of their musical neighbors that one must seek the source of their inspiration in the improvising mind of a genius who watched Raphael and Michelangelo at work in the Vatican, and who could have observed Titian while he painted one of his portraits of Eleonora née Gonzaga, Cavazzoni’s early patroness, in whose service the composer earned the sobriquet “d’Urbino”. Who knows, he might have even helped to inspire them by his playing on the sober little instrument from 1516 shown on our cover, formerly the property of Leo X, which happens to be the oldest harpsichord to have survived the havoc of the intervening centuries. This project celebrates the 500th anniversary of its existence.

In Greek mythology, the satyr Marsyas challenged Apollo to a contest: his panpipe vs. the god’s lyre. The defeat of the impudent creature is invoked in Renaissance literary discussions of the relative merits of wind and stringed instruments, in which strings obviously come out on top. I am not suggesting that organists who play Cavazzoni and other early ricercars should be flayed alive, as poor Marsyas was; but a little less presumption would be in order.

Glen Wilson

- back -