Faà di Bruno, Giovanni Matteo [Horatio, Orazio] 83

2. Medieval and Renaissance vocal music.

(i) 14th and 15th centuries.

The words ‘fuga’, ‘chace’ and ‘caccia’ all denote the chase or hunt, and in the 14th century all acquired the same musical meaning, namely a piece of music consisting either entirely or principally of two or more voices in canon. Canonic technique was, along with Stimmtausch, the earliest type of imitative counterpoint in Western music, and it may therefore fairly be said that the word ‘fugue’ has been associated with imitative techniques since their first formulation. By the 15th century, ‘chace’ and ‘caccia’ had largely fallen from use and ‘fugue’ became the term of choice for any piece in which all voices participated in the canonic performance of a single melodic line (see, for example, the fugues by Oswald von Wolkenstein).

As early as the mid-15th century, however, ‘fugue’ had begun to take on new meanings. As canonic technique came increasingly to be incorporated into pieces that also included non-canonic voices, musicians often continued to apply the word, though not to the piece as a whole; they reserved it instead for the canonic voices. Also at this time, musicians began to use ‘fugue’ to designate the compositional technique itself. Johannes Tinctoris, in his dictionary of musical terms written about 1472, defined it as ‘the sameness [idemtitas] of the voice parts in a composition. The notes and rests of the voice parts are identical in [rhythmic] value, name [i.e. hexachord syllable], shape and sometimes even location on the staff’. Here fugue is not a piece of music or group of voices governed by canonic technique; it is the technique itself, the quality of having made the voice parts identical. Perhaps the best-known use of the word in this sense is Josquin des Prez’s Missa ad fugam, or ‘mass by means of fugue’.

(ii) 16th century.

During the 15th century, as composers gradually abandoned the compositional process of writing one voice at a time above a cantus firmus in favour of the simultaneous composition of all voices a few bars at a time, the point of imitation began to replace canonic writing as the pre-eminent technique of imitative counterpoint. To write a motet or mass movement using this technique, the composer first created for each phrase of text a musical phrase that fitted its Latin declamation well. The piece then proceeded as a series of imitative sections, each devoted to its own textual/musical phrase which was treated in a manner similar to (if much freer than) the fugal exposition described in §1 above. These imitative sections, usually referred to today as ‘points of imitation’ but called ‘fugues’ by musicians of the time, might also alternate with occasional homophonic sections for contrast. A classic early example of a piece composed in this manner is Josquin’s four-voice Ave Maria … virgo serena. Musicians never adopted the word ‘fugue’ as a genre designation for such pieces, however. The first composers to import point-of-imitation technique into instrumental music – beginning with the Musica nova (1540²²) of Julio Segni, Willaert and others – chose instead the designation ‘ricercar’, a word meaning ‘to seek out’ or ‘to search’. Throughout the 16th century, only the strictly canonic piece might bear the genre designation ‘fugue’.

North of the Alps, composers of the post-Josquin generation, most notably Clemens non Papa and Nicolas Gombert, made point-of-imitation technique the cornerstone of their style. As described by the German theorist Gallus Dressler, an enthusiast for the music of Clemens, in his manuscript treatise Praecepta musicae poeticae dated 1563, a ‘fugue’ (i.e. point of imitation) that appeared at the beginning of a piece needed to be constructed in such a way that the mode of the composition was made clear at the outset. This meant that the melodic motion of the voices should emphasize the important modal notes of final, dominant, mediant and psalm tone tenor(s), and that the imitation should show a certain clarity of structure. As a result the opening ‘fugue’ of a 16th-century motet resembles in many respects the exposition of an 18th-century fugue: its voices are most likely to enter on final (tonic) and dominant, its theme is also likely to feature those notes prominently, and each voice is likely to wait until the previous one has completed its thematic statement before entering.

Dressler borrowed the tripartite division of beginning (exordium), middle (medium) and end (finis) from classical rhetoric to describe a composition’s overall structure, and indicated that ‘fugues’ appearing in the body of the piece (i.e. the medium) could be handled much more freely than the opening one (which Dressler defined as the exordium). This freedom extended to the allowing of thematic entrances on notes other than final and dominant, the greater altering of the theme from statement to statement, the incorporation into the theme of notes ‘outside the mode’, and considerable use of stretto. For the finis, however, the composition should close with a strong reaffirmation of the mode. Here again certain parallels can be drawn, this time between the structure of a motet and that of an 18th-century fugue: both begin and end ‘in the key’ (or mode) and with greater regularity but may wander (i.e. touch on other notes or keys) and behave more freely in the body of the composition. Most of the motets of Clemens and Gombert fit Dressler’s model well.

South of the Alps, meanwhile, Clemens’s and Gombert’s Netherlandish contemporary Adrian Willaert and his Italian pupil Gioseffo Zarlino took a different attitude towards imitative counterpoint and attempted instead to preserve Tinctoris’s 15th-century definition of fugue as exact imitation. In his Istitutioni harmoniche (1558) Zarlino subdivided imitative counterpoint into four categories, for which he coined the expressions fuga legata, fuga sciolta, imitatione legata and imitatione sciolta. Whereas musicians in the north had already begun to label all imitative counterpoint ‘fugue’, Zarlino insisted that the word be reserved exclusively for instances in which the following voice (which he called consequente) reproduced exactly all the intervals and rhythmic values of the preceding voice (guida). Imitatione should refer to instances when the consequente did not reproduce the guidaexactly. The adjectives legata and sciolta then distinguished further between, respectively, passages in which the consequentecontinued to imitate the guida from beginning to end of the piece, and those in which the imitation was broken off at some point. Zarlino also allowed for both fuga and imitatione in contrary motion, for which he offered the modifier per arsin et thesin (a Greek expression literally designating upbeats and downbeats).

Zarlino pointed out that fuga was possible only when the consequentewas a perfect interval from the guida. This requirement was a direct descendant of Tinctoris’s insistence that the solmization syllables of the two voices be identical, which restricted the imitation to the three hexachords. However, a perfect interval does not guarantee exactness of imitation. For instance, any imitative counterpoint written at the 4th or 5th can be melodically exact only so long as the voices remain within the bounds of their respective hexachords. Thus, if Zarlino’s distinction between exact and inexact replication is to be strictly maintained, it must be conceded that his technique of imitation can take place at any interval, perfect or imperfect.

Scholars are not in full agreement about whether Zarlino meant to allow for the technique of imitatione at a perfect interval, but in any case the theorist himself seems to have recognized a conflict between categories of imitative counterpoint based on the degree of exactness of replication and those based on imitation at perfect as opposed to imperfect intervals. To address this problem, Zarlino created a final category, which he called ‘part fugue and part imitation’ but admitted was often called fugue. The example he offered involves two voices that imitate each other canonically at the 5th. At only three places does the second voice answer inexactly, and in each case the inexactness consists of an F answered by a B natural rather than the B that would be required for precise replication.

Willaert’s music reflects Zarlino’s thinking closely in its penchant for introducing imitative counterpoint horizontally into a composition voice by voice. The point of imitation almost never serves as the building-block for Willaert’s mature works as it does for the works of Clemens and Gombert, whereas canon is frequently encountered (see, for example, Verbum supernum and Praeter rerum seriem from Musica nova, 1559). Nevertheless, despite Willaert’s prestige and the lasting influence of Zarlino’s writings, the wave of the future was not canon but the point of imitation, which allowed a much more flexible treatment of the words and thus fitted better with humanistic ideals about text expression and clarity of declamation. In the end, therefore, Zarlino’s role in the history of fugue is a peculiar one. On the one hand, the traditional idea of fugue as precision of imitation and its manifestation in canonic writing quickly faded. Yet, at the same time, the categories of imitative counterpoint which he invented lived on, though with very different meanings from those intended by their creator.

As the 16th century drew to its close, the rise of humanism, with its emphasis on clarity of text, caused many musicians, especially in Italy, to question altogether the suitability of fugal techniques in vocal music, since their use virtually ensured that different words were being sung simultaneously by the various parts. By the 1580s, when Vincenzo Galilei began to call for the abandonment of fugal writing, it was apparent that the most important and innovatory genre of music was the Italian madrigal, in which fugue played no significant role. Even such conservative composers as Palestrina and Lassus began, partly under the influence of Counter-Reformation concerns for textual clarity, to show much greater caution in introducing points of imitation into their motets, in contrast to its almost ubiquitous presence in the works of Clemens and Gombert. Introduce it they did, however. In fact, Lassus continued to be cited by German theorists writing about fugue throughout the first half of the 17th century, and in the 1660s Christoph Bernhard chose Palestrina’s offertories to illustrate his chapters on fugue. The first major composers of vocal music in the new Baroque style, however, all but abandoned fugal techniques for their seconda pratica music. (A piece such as Monteverdi’s Piagne e sospira, from the fourth book of madrigals, is a rare exception.) Fugue found no place in the new genres of opera, monody or cantata, nor, surprisingly, did it play a role in the early development of the oratorio. Even such a retrospective collection as Schütz’s Geistliche Chor-Music (1648) is of little significance in the history of imitative counterpoint. It was not until towards the end of the 17th century that fugue once again re-entered the sphere of vocal music in a significant way. By that time, a great deal had changed.

Fugue