|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
CHAPTER THREE
One of the strongest tendencies that emerges
from a survey of the 884 different rhyme schemes
devised by the troubadour poets is the preference for
patterns that articulate the first four lines of the
stanza, regardless of its length, as a unit. This is
seen clearly in the popularity of the abab and abba
patterns, which are found to begin almost half of all
the rhyme schemes in Frank's Répertoire
métrique; it can also be seen in the other
patterns employed. Therefore, one way of understanding
the ABAB musical form is as a reflection of this
pervasive articulation of the song's first four lines
as a unit. Indeed, the most obvious correlation
between musical form and rhyme scheme is found in this
category, with 40 songs, exactly half of the group,
matching their ABAB musical form with an abab rhyme
scheme. This represents almost two‑thirds of all
the songs transmitted with music that use the abab
rhyme scheme. The abba scheme was used with 28 of the
ABAB forms; this can be compared with a total of 96
songs in which this rhyme scheme was used.
With the form of the first section of the song
fixed, it is the remainder, from the fifth phrase
onwards, that allows for formal variation and
different sets of options. The ABAB songs therefore
fall naturally into a handful of sub‑groups
according to whether or not there is phrase repetition
in the second half, and the kind of repetition used.
These sub‑groups will be considered in turn, but
first it may be appropriate to look at the ABAB
section more closely, since it is the defining feature
of the form. Given the clear individual integrity of
this four‑line unit from the metric point of
view, one is led to ask whether the A and B musical
phrases tend to be distinguished musically in any
special way from other phrases. To investigate this in
a manner as objectively quantifiable as possible, the
interval between final and initial pitches of
consecutive phrases was taken as the main indicator of
the strength of articulation at phrase junctures,
supplemented in some cases by register. In other
words, a small interval between the end of one phrase
and the beginning of the next is understood to provide
a smooth juncture between the phrases, and thus a
relatively weak articulation, while a large interval,
such as a fifth, is taken to imply a strong
articulation. However, there may be situations where a
pronounced shift in register occurs in a following
phrase, which is not manifested in the interval
between the two phrases, because it may not take place
until slightly later in the phrase; in such cases the
shift of register is still considered a sufficient
indicator of a relatively strong articulation.
All 80 songs from this category were examined
for the intervals between the final of the A phrase
and the initial of the B phrase; between the final of
the B phrase and the initial of the second A phrase;
and between the final of the second B phrase and the
initial of the C phrase.
In addition, the finals of the A and B phrases
were compared with the song final, on the assumption
that the use of the latter in an earlier phrase could
be a further indicator of a strong articulation. It
was found that the smallest intervals occurred between
the end of the A phrase and the beginning of the B
phrase, while the largest intervals were found between
the end of the B phrase and the beginning of the
following A phrase. In about 85 percent of the cases,
the interval between the A final and the B initial was
either a unison or second; the third was used in ten
songs, while the fourth, the largest interval found in
this position, appeared only three times. This
suggests a smooth transition between the A and B
phrases as a general feature, and the linking of the
two phrases as a pair, separated by a weak
articulation.
Conversely, the widest intervals were found
between the final of the B phrase and the beginning of
the A phrase. Unison and second combined make up 37
percent of the total, the remainder being composed of
intervals from the third to the octave. The interval
of the fifth is the most common of all at this
position, occuring 30 percent of the time. This strong
articulation at the juncture between each AB phrase
pair further underlines the role of the two phrases as
a single unit. At the juncture between the second B
phrase and the C phrase, one finds three options used
with equal frequency. In a third of the songs, an
interval of a fourth or fifth or larger, and/or a
shift in register indicates a strong articulation
between the two main sections of the song. An equal
number of songs have a smooth transition between these
phrases, as indicated by the narrow interval of the
unison or second, and the absence of other forms of
marked differentiation. The remaining third of the
songs have what could be termed a more or less
neutral articulation produced by the interval of the
third.
Some commentators have gone so far as to
interpret the relationship between A and B phrases as
similar to the antecedent/consequent pair of later
styles, or as the genesis for the ouvert/clos
formations in the centuries immediately succeeding the
troubadour and trouvère period. One may indeed find
some analogies to these relationships in the ABAB
phrases, but it would be going much too far to view
the two phrases as indissolubly linked. For one thing,
tonal relationships in this repertoire do not have the
clear definition and fixed meaning that gradually
marked the development of a functional tonality, and
therefore, in many cases, an ouvert/clos sequence could just as easily be understood as clos/ouvert
or in other, less specific terms. Furthermore, any
individual A or B phrase in a given song still has as
much relative closure as any other phrase, with the
typical initial and cadential formations common to
other phrases; the proof of this, if proof were
needed, lies in the songs where either the A phrase or
B phrase is repeated in the latter half of the song,
independently of its partner.
It is important to understand that the
antecedent/consequent type of effect that seems to be
naturally associated with the A and B phrases may be
attributed in large measure to their immediate
repetition as a pair, rather than to any intrinsic
features of the phrases themselves. To be convinced of
this, one may try the experiment of taking any two
consecutive phrases from a song without any phrase
repetition, and repeating them as in the ABAB form; in
most cases, the same effect will be produced. With
these cautions in mind, we may compare the finals of
the A and B phrases in all the songs from this
category as a partial measure of the types of
relationships to be found between the two phrases,
given the important role of cadences in both ouvert/clos and antecedent/consequent formations. In 20 percent of
the cases, both phrases end with the same final pitch.
In almost 30 percent, the final of the B phrase is
either a second higher or lower than that of the A
phrase, with the former situation slightly more common
than the latter. When the interval between finals is a
third, fourth or fifth, the overwhelming tendency is
for the final of the B phrase to be lower rather than
higher. The lower third occurs as final in the B
phrase in 17 percent of the songs, while the upper
third is used in only 3 percent; the lower fourth
occurs in 12 percent, while the upper fourth is found
in only one song; the upper fifth is used in 13
percent, while the lower fifth is found in only a
single song. The only other interval found in this
location is the lower octave, which appears in the B
phrase of three songs.
In approximately 40 percent of the ABAB songs,
there is some alteration in the second pedes, which may affect either the A or B phrase, or both. For
roughly 12 percent of these cases, the change takes
the form of a minor variation or elaboration affecting
any part of the phrase. The majority of the other
cases can be divided into two distinct sub‑groups
of equal size, according to the manner in which the
cadence of the second B phrase is modified. In one sub‑group,
repesenting about a quarter of these songs with a
change in the second pedes,
the cadence of the second B phrase is altered to lead
more smoothly into the following phrase, in effect
eschewing the articulation of this juncture in favour
of what can be termed a melodic enjambment. Examples
include P‑C 70,12, P‑C 30,16, P‑C
167,30, P‑C 406,23, P‑C 366,6, P‑C
366,12, and P‑C 248,18. As an illustration, the
first five lines of Gaucelm Faidit's "Jamais nulh
tems" (P‑C 167,30) are shown in Example 20,
in all three of its surviving versions. The
interesting point to note in this example is the way
in which the same ungapped link is made between the
end of line 4 and the beginning of line 5 in all
versions, in spite of the difference in pitches
between the version of R and the other two. Example
20. Gaucelm Faidit, P‑C 167,30, ll. 1‑5.
Example
22. Pons d'Ortafas, P‑C 379,2, ll. 1-5.
In
"Nuls hom no sap d'amic" (P‑C 456,26),
by Uc de Saint Circ, the ouvert/clos
relation is effected between the two pedes
with the shift of the cadence of the second B phrase
down a fifth. Then the third B phrase, immediately
following, has a further cadential alteration, with
another shift up a fifth in a variant of the first B
phrase. The unique situation represented in this song
is shown in Example 23 below. Example
23. Uc de Saint Circ, P‑C 457,26, ll. 1‑5.
Another
song deserves mention here for the unique role the
caesura plays in the modified second pedes. In
"Ben dei chantar" (P‑C 366,3), by
Peirol, consisting entirely of decasyllables, the
caesura after the fourth syllable is underlined by
cadential formations in every line in this position.
In the second A phrase, the music for syllables 1 to 4
is replaced with new music, while the rest of the line
repeats that of the first A phrase, except for the
minor change of pitch at the fifth syllable. Then, the
music for the first four syllables of the first A
phrase returns as the initium
for the last phrase of the song, again with minor
variations. The song is reproduced below in Example
24.
Taking the latter case first, we may compare
the formal graphs for the two versions of Peirol's
"Coras que mi fes doler" as given below:
As the graph shows, the version of R diverges from
the ABAB form by virtue of its third phrase, which is
different enough from its first phrase to be
considered new, even though it repeats the cadence of
the first phrase. It would have been a simple matter
for the scribe to have made the third phrase closer
to, or exactly the same as, the first phrase, and the
main scribe of this manuscript has often been accused
of regularizing the forms of songs in just this way. That he did not
do so here may perhaps be attributed to a recognition
of another form of symmetry in the version he was
transmitting, a feature which is not present in the version
of G.[15]
This is the repetition of the pair of phrases making up lines
3 and 4 as lines 6 and 7, both of which are then
recalled in line 8.
The difference in form between the two versions
of P‑C 167,53, by Gaucelm Faidit, are a result
of the marked melodic divergence found in almost every
phrase of the song. The melodies are not completely
different; line 5, for example, is practically
identical in both versions, and lines 4, 6, and 9 are
similar enough to be recognizable as having a common
source. In all the other lines, however, the
divergences, several of them radical, outweigh the
points of similarity. Following is the graph for the
two versions of this song:
Example
25. P‑C 406,2 and P‑C 406,13, ll. 1 and 3.
It now remains to consider the kinds of formal
design found in the second half of the songs with an
ABAB form, that is, from the fifth line onward. The
songs fall into two main groups of equal size, those
without phrase repetition in the latter part of the
song, and those that repeat one or more phrases. The
latter group can be further subdivided into five
distinct subgroups, according to the kinds of
repetition displayed. In one design, the B phrase only
is repeated as the last line of the song;[16]
in another the B phrase only is repeated, but not as
the last line.[17]
Another group of songs repeats both A and B phrases in the
second half, as a pair;[18]
in the fourth group, only the A phrase is repeated.[19]
What distinguishes the final group is the repetition of one or
more phrases from the second half of the song, rather than the A
or B phrase.[20]
There are a few songs that belong to more than one of these
groups, and these will be pointed out as each group is
examined in turn.
In ten songs, including the famous
"Lanquan li jorn son lonc en may" (P‑C
262,2), by Jaufre Rudel, the B phrase is repeated as
the last line of the song. Typically, these songs also
feature an upward shift in register at the fifth line,
heightening the contrast between sections. With the
return of the B phrase at the end, an effect of
balance and closure is produced which our contemporary
ear finds appealing, for it recognizes the familiar
structure of the ABA forms of the common practice
period. To illustrate this design, "De la jensor
c'om vei' al meu semblan" (P‑C 47,5), by
Berenguier de Palazol, is given below in Example 26.
In all the other songs of this type, the entire B
phrase is repeated without change at the end, but in
this one the initium
of the phrase is modified, and its first four
syllables recall the descending passage at syllables 4
to 7 in phrase A. One may also note the matching of
cadence and rhyme with the repeat of the b rhyme in
the last line. Example
26. Berenguier de Palazol, P‑C 47,5.
In nine songs, the B phrase is repeated at some
position before the last phrase. If it is repeated
immediately after the second B phrase, as the fifth
line of the song, it can be understood as an extension
of the ABAB section. This is the case in four of the
present songs (P‑C 457,26, P‑C 456,40, P‑C
406,47, and P‑C 248,82), and since all of them
also bear a modified second pedes,
they were discussed together in the section treating
this feature, to which the reader is referred. In one
song, the varied repeat of the B phrase at every other
line creates a refrain‑like structure that
extends the ABAB form into the second half of the
song. This occurs in Bernart de Ventadorn's "Pois
preyatz me, senhor" (P‑C 70,36), in the
version of G, which can be graphed as follows:
Example
27. Raimon
de Miraval, P-C 406,20.
Example
28. Arnaut
de Mareuil, P-C 30,16.
Example
29. Raimon de Miraval, P-C 406,24.
P‑C 293,30
A B
A B C C
D P‑C 366,19
A B
A B C C* D
E P‑C 406,14
A B
A B C C
D E
F P‑C
366,26
A B A
B' C C*
C C*'* P‑C 10,25
A B
A B*
C D E
C*
(R) P‑C 248,66
A B
A B C D
E D
F P‑C 406,21
A B
A B C D
C D P‑C 248,65
A B
A B C D
*C E
D' D'* P‑C 248,7
A B
A B C D
*D* E F
G H G* *E*
I P‑C 248,12
A B
A B C
D *D*
E C'
F
Example
30. Raimon de Miraval, P-C 406,21.
«
Chapter IIIb
Contents
Chapter IIId »
[13]
They are: P‑C 406,21, P‑C 366,26, P‑C
248,66, P‑C 70,36, and P‑C 457,40.
[14]
The songs are: P‑C 323,15, P‑C 70,1, P‑C
70,6, P‑C 70,12, P‑C 70,36, P‑C
70,41, P‑C 167,30, P‑C 167,37, P‑C
167,43, P‑C 167,53, P‑C 10,25, P‑C
406,2, P‑C 406,13, P‑C 366,9, P‑C
366,12, P‑C 372,3, and P‑C 375,27.
[15]
Note that I am not using the term symmetry in its
precise and literal meaning of mirror symmetry, an
example of which would be the rhyme pattern abba.
Rather, I extend the term to include various types
of parallelism, balance and regularity in the
matching of groups of phrases.
[16]
See P‑C 262,2, P‑C 47,5, P‑C
70,12, P‑C 46,2, P‑C 335,67, P‑C
406,13, P‑C 248,21, P‑C 248,30, P‑C
248,45, and P‑C 248,56.
[17]
See P‑C 262,5, P‑C 70,4, P‑C
70,36, P‑C 457,26, P‑C 457,40, P‑C
406,20, P‑C 406,47, P‑C 248,8, and P‑C
248,82.
[18]
See P‑C 379,2, P‑C 30,16, P‑C
167,30, P‑C 248,5, P‑C 248,18, P‑C
248,29, and P‑C 248,78.
[19]
See P‑C 70,6, P‑C 406,20, P‑C
273,1, P‑C 406,8, P‑C 406,24, and P‑C
248,57.
[20]
See P‑C 457,40, P‑C 70,6, P‑C
70,36, P‑C 273,1, P‑C 293,30, P‑C
10,25, P‑C 406,14, P‑C 406,21, P‑C
366,19, P‑C 366,26, P‑C 248,7, P‑C
248,12, P‑C 248,65, and P‑C 248,66.
[21]
It is the version of ms. R that is referred to in
connection with P‑C 167,30; the other two
versions differ significantly from that of R, and
do not repeat the A or B phrase.
|