updated 13 November 1995
EOL: Magrini music

EOL: Magrini

4. Music analysis of ballads

Four types of two-voice songs found in Bettinelli sisters repertoire
Solo melodies derived from two-voice repertoire
Examples from Nigra
Summary of musical analysis

In Canti popolari del Piemonte, Nigra says nothing about performance practice, but he includes musical transcriptions showing a single line. This seems to imply that those ballads were monodic songs. Since this agrees with monodic assumptions about ballad traditions in Europe and USA, the monodic nature of Italian ballads has been accepted without further discussion. But Nigra generally worked with single informants because of the methodology adopted, dictation of ballad texts. Nigra was interested in the ballad as poetry. His interest in music and performance practice was minimal, as it was for many other folklorists working in the nineteenth century, for example, Giuseppe Ferraro (1845-1907).

Nigra's notes to the texts offer clues about ballad performance practice. He transcribed most of the texts from solo recitations by a woman, such as a peasant, a servant or even a door-keeper. But texts transcribed from singing were often performed by groups of women, generally "peasants," but also "grape-harvesters" (Nigra 1974:48, 148) and in one case by a mixed group (ibid.:403). This is the first evidence that ballads were performed by groups in the nineteenth century.

Besides the rich collection of ballad texts, Nigra's Canti popolari del Piemonte includes twelve musical transcriptions. Notated from sung performances rather than recitations of lyrics, many are group performances, according to Nigra's notes (page numbers in parenthesis refer to Nigra 1974):

  1. "Donna lombarda," sung by Domenica Bracco (34)
  2. "Fior di tomba," sung by a group of peasant women (159)
  3. "Cattivo custode," sung by a group of grape-harvester women (349)
  4. "La pesca dell'anello," sung by a group of peasant women (417)
  5. "Il tamburino," sung by Domenica Bracco (450)
  6. "Convegno notturno," from a collaborator, Luigi Franchelli, (no information about the performance) (460)
  7. "La bevanda sonnifera," sung by a group of peasant women (471)
  8. "L'uccellino del bosco," sung by a group of peasant women (528)
  9. "La bionda di Voghera" (no information about the performance) (551)
  10. "Girometta" (no information about the performance) (556)
  11. "Il pellegrino di Roma" (no information about the performance) (569)
  12. "Carolina di Savoia," a historical song sung by a peasant woman (645).
Although apparently monodic, half of these transcribed melodies, all clearly tonal, end on the third degree of a major scale (numbers 2, 3, 4, 6, 7, 12). Melodic lines ending on the third degree of a major scale have been widely documented in the twentieth century, together with ballads performed by two voices in parallel thirds (Magrini 1990a, 1990b, in press). These two-voice songs have different configurations, which can be compared with examples from the rich repertoire of the Bettinelli sisters (Mantovani 1979).

Four Types of Two-Voice Songs found in Bettinelli Sisters Repertoire

Type 1

Score, audio

Type 2

Score

Type 3

Score, audio

Type 4

Score

Solo melodies derived from two-voice repertoire

Besides songs performed in two-voice style (7), one finds in Italy, in the literature and in recordings, many solo melodies which seem to derive from songs conceived for a two-voice performance. Emanuela Lagnier shows that in Valle d'Aosta, a region situated between France and Piedmont, where Nigra collected his songs, "most of the melodies belong to the tonal system and their tonal range is not extended. As regards almost all melodies sung in Italian and in Piedmontese dialect (and part of the melodies sung in French), the singer may be accompanied by a second performer, who sings a third above the line. This kind of performance is typical of the polyphonic folk music in Northern Italy" (Lagnier 1984:26)( translation, T. Magrini). Lomax also notes that Piedmont ballads "are invariably sung in chorus" (Lomax 1955-56:126).

A well-known ballad from Lagnier's collection demonstrates a solo melody derived an older two-voice performance practice. This song may be classified as belonging to Type 3. The melody begins with a characteristic jump from the fifth to the first degree and ends on the fifth degree in the first section. When the melody starts again it presents the third degree where one expects the tonic and ends on the third degree. After the first section of the song, the melody clearly reflects an older two-voice practice, but the lower part (the "tenor") no longer appears in the repetitions.
Example: "Cattivo custode"

Morelli and Starec report that in Trentino and Friuli, two regions of Northern Italy, all ballads are sung either in thirds or in solo style, retaining the voice that the performer is used to singing. In his commentary on Wassermann's collection, Starec says, "The melodies ending on the third degree, that are conceived as upper lines of a polyphonic performance, are twice the number of the melodies ending on the tonic. It is well known that the singers, even in solo performances, choose the upper or the lower part on the basis of which part they like best and of the role they a used to performing in polyphonic performances" (Wassermann 1991:313) (translation, T. Magrini).

In an example from Wasserman's collection, (ibid.:320), a old woman sings the melody taking the upper part of a song transcribed in G Major (originally in D Major). If originally a two-voice song, it presumably belongs to Type 1, with no solo section. It begins and ends on the third degree, has a limited range, and, characteristic of upper voices, the seventh degree descends instead of ascending to the tonic.
Example: "La prova"

Thus, melodies intended for two-voice performance have acquired a separate existence, to be sung solo or in unison, sometimes by performers used to singing in two-voice style. A melody from the Emilia region of Northern Italy exemplifies this type. In D Major, it ends on the third degree, clearly belonging to Type 2. The second voice would enter just after the triadic incipit stating the tonality of D Major. When the incipit is repeated in measure 2, it is modified because of the virtual presence of the lower part (8).
Example: "Ratto al ballo"

Teresa Viarengo is one of the most important singers of monodic ballads in Northern Italy (Leydi 1977, 1990). She performed the few cases of modal ballads in Italy and adopted a particular style of singing ballads, the "enunciativo" style, (Magrini 1990a) consistent with the "moderate, restrained and impersonal" style typical of North European balladry (Porter 1988:76). Viarengo's repertoire also includes melodies related to two-voice singing, such as the "Donna lombarda," which seems to belong to Type 3 and takes the form ABB.

In the A section, the melody begins by outlining the tonic chord of D Major; then the melody moves to the region of the fifth degree and ends on it. The B section narrows the range, avoids the tonic, and concludes on the third degree, F-sharp. The B section appears to be the upper voice of the two-voice second section of the song. In addition, in the first measure G-sharp is performed as a kind of lower neighboring tone, which is allowed in this solo section. In the second section, G is generally natural, necessarily so for the two-voice rendition. But the pitch of this tone is sometimes uncertain in the repetitions, probably because of the practice of solo singing and the lack of the accompaniment of the "second" voice.
Example: "Donna lombarda"

Musical Examples from Nigra Collection

"La pesca dell'anello" seems to belong to Type 4. It is divided into two parts, AB. Part A was presumably solo or in unison. Part B begins repeating the same melody and certainly seems to call for parallel thirds in the last three measures of B. The voice moves to the upper part and ends on the third degree in Part B.

"La bionda di Voghera" begins in C Major, then repeats the same melody a third above, slightly varied. This seems to belong to Type 3. It seems to represent a soloist who sings the first phrase alone, then moves to the upper part doubling the melody a third above.

"Convegno notturno" seems to belong to Type 1. Apparently the upper voice of a two-voice song in B-flat Major, it begins and ends on the third degree, following throughout in parallel thirds above a virtual main melody:

Summary of musical analysis

These examples strongly imply a practice of two-voice singing, perhaps widespread since at least the second half of the nineteenth century. The survival of a number of upper voices sung in solo style may suggest other considerations. Singers are used to a fixed role in two-voice performances, singing the top voice or the bottom voice. When a singer accustomed to two-voice singing performs a solo, she performs as though she were singing in two-voice performances. This results in some melodies ending regularly on the tonic (lower voices). Other melodies begin on the tonic, then move a third above and end on the third degree, indicating a singer accustomed to perform the upper part.

The high number of apparently upper voices documented as solo melodies may have resulted from singers with good voices and a wide register regularly performing the upper part, having the capability to do so. In many cases, singers on the upper voice act as leaders in the performance, for example, the Natalina Bettinelli, who leads her sisters (Mantovani 1979:41). These women are generally good singers, know many songs, remember long texts, and is therefore likely that are transmitters of a fair number of songs in the "upper voice" version.

When the practice of ballad singing in two-voice style became less frequent, many melodies survived either in the "upper voice" version or in the "lower voice" one. Then, because they sang without accompaniment, the performers may have sometimes worked out these melodies in new ways, inserting little variations, typically mixing fragments of the upper and lower parts (Sassu 1983:161), inconsistent with the old style of two-voice singing. Nevertheless, when singing "upper voice," they retain the typical ending on the third degree.

Upper lines of two-voice songs acquired an existence of their own. Performers sing these lines as if they were monodic. They do not need another singer to perform. They sing these lines before or after other melodies ending on the tonic without making any distinction between the two types of melody. They comfortably end a melody on the third degree of the scale, suggesting that this degree has acquired the status of a substitute finalis, a sort of confinalis, to borrow a term from the theory of Gregorian chant (Powers 1980:394-96), a note on which it is possible to end a transposed melody, in this case, the melodic section transposed a third above. Unlike Gregorian chant, this tone does not possess the quality of "tonic." Even if these melodies are often sung in solo style, their connection with the two-voice performance practice is easily recognizable, in their melodic configuration too, and is further confirmed by field research, as demonstrated here.

All balladic singing in the past did not use two-voice singing. Cases of clearly monodic melodies, sometimes endowed with modal features, have been documented (Leydi 1973, 1977; Magrini 1990a), so that we can assume two-voice and monodic practice existed together (Sassu 1983:160). Yet the practice of two-voice singing was widespread in the nineteenth century and many ballads documented in the past and in the present were conceived for this performance practice. This contrasts with the common representation of ballad singing as a traditionally solitary activity of women, who liked to recall and perform old narrative songs mainly for their personal entertainment. On evidence provided by older literature and recordings made since the beginning of the 1950s, one can argue that in Italy ballad singing was often conceived in the past as a form of musical interaction.


Continue to "Comparing male and female ballad singing" page
Comment to author, editor, or public discussion
Return to Magrini article table of contents