Simon Furey wrote:

> Don,
> Your exposition is very helpful. Nevertheless, I still have a difficulty:
> are you implying that we cannot determine the mode of a melody without
> adding harmony? If so, that's an important principle that I was unaware of.
> My consideration of the Beatles' tunes was done by taking them away from
> their musical settings and dealing with them in isolation, on the basis that
> there are many quite different ways of arranging a tune. I quite accept that
> the musical settings that the Beatles provided may not have been in the
> least modal. But doesn't the same thing happen with folk songs when sung by
> pop groups or arranged (say) for orchestra?  If you do treat the Beatles
> tunes I mention extracted from their harmonic settings I don't see why you
> can't consider them as modal. 

Well, you've put your finger on one or two of the quandaries.  A tune in
isolation can in fact be interpreted in a number of different ways. 
When accompanied, there are many options for chords; the Maritime
pianist will use a lot more chords and transitional structures when
accompanying a fiddle tune than the New England French-Canadian pianist
or Appalachian guitarist.  Times change, tastes change, and broad
distribution of recorded music has influenced local cultural styles.  

But if you choose to evaluate a tune in isolation, it has to be
performed in isolation - and what is it you are performing?

1. If the Beatles wrote the tune in conventional major, and performed
and presented it as such, with full accompaniment, and you decide to
reset as an unaccompanied modal piece (I'm not sure how you do that
without changing the notes), are you performing their song or your own
song derived from theirs?

2. Melodies do have relationships among their notes; these relationships
shift around depending on the structure of the melody, but when you've
developed the ear for it, you can pick out the tonic, which is inherent
in the structure of the melody and the scale.  Without a tonic, music
would sound like some of Phillip Glass's less successful efforts
(forgive me - I saw "Akhnaten" Friday night!); to some extent the very
essence of melody requires that it exist in relation to an "anchor". 
That anchor is not necessarily the starting or ending note - although
it's more often the ending note than the start - but you have to find it
before you can characterize either the scale or the mode.

3. Each mode - including authentic and plagal variants as separate modes
- has a distinctive sound.  There exist several medieval sung mnemonics
which simultaneously illustrate the structure of melodies for each mode
while describing the type of music best fitted to it (i.e. the emotional
effect they perceived as conveyed by music in this mode; e.g. our
current characterization of minor keys as sad).  So a mode is not simply
a matter of scale; it's also a melodic structure which exploits the
chosen scale.

4. We have lost the ear for modes.  A collector of sea songs around the
turn of the century (I really have to chase this reference down) found
that he could only get modal songs from the older sailors, who observed
that the younger sailors didn't have the ear for them; their versions
had moved into the major and minor which characterized the music all
around them, fitted to instrumental accompaniment.  Anyone who sings
rounds and is attentive will note that the contemporary oral tradition
turns modal to minor, and minor to major; a round written as modal is
likely to be adjusted into major in three passes, because that's what
people hear constantly, and are most familiar with.

5. All music has context; it depends on a listener and a cultural
environment (and sometimes a language) as well as a performer, even if
the listener and performer are the same. Analytically, there is a
quandary inherent in any attempt to evaluate a tune or piece of music in
isolation, or "objectively" - I put the term in quotes, because the
evaluator is irretrievably stuck with his/her distinct musical
background, which influences how melodies are perceived.  But without
such a background, the "melody" might as well be a random string of
notes - which we all agree is not, by our standards, "music".

Where does this all leave us in evaluating traditional music?  Well,
Bronson concluded several interesting things in his analysis of
traditional tunes; one was that "pure" modes and/or gapped scales were
relatively uncommon - i.e. that singers freely added the sixth or
flatted the seventh if they wanted the effect at that point in the song.
 My own feeling (bearing in mind I'm not a scholar in this area) is that
the traditional ballad is the descendant of plainsong, where the melody
was constructed (according to certain rules) as a function of the text
being sung, and its meaning.  Singers clearly didn't feel constrained to
fit their texts or tunes to what has become conventional meters and line
lengths - so why should they have felt constrained by the then
contemporary preoccupation with the regular rhythms, scales and chordal
progressions which were derived from contraints relating to instrumental
and orchestral performance - or even to academic definitions of modality?

I also theorize that traditional modal singing does not use the
equal-tempered scale - and I don't just mean the use of unique notes
which fall on quarter-tones in certain traditions.  The equal-tempered
scale is for instruments; you don't sing it naturally, even in
multi-voice a capella arrangements.  And in unaccompanied singing, you
don't even necessarily sing conventional intervals; a "repeated" note
may be flatter coming up to it, and sharper going down, or vice versa
depending on what the singer is trying to do.  All of this disappears
the minute the music is transcribed, and can't be replaced if you
accompany the music with chorded instruments, even if you can bring
yourself to ignore all the music you've been inundated with over your
lifetime, and the lessons of all choral leaders and most voice teachers.
 I'm sitting here listening to a tape of a friend singing "Sweet
William's Ghost"; if I pay attention, I can perceive that few of the
transitional notes are "in tune" by ET standards, but if I simply listen
to the song, it sounds just right, and presents the text exceedingly
well.  For instance, when she sings up to and back down from the fourth
leaving it flat, the effect is sadness, not out-of-tuneness; part of the
reason it doesn't sound out of tune is because she flatted the second
and then sang a more-or-less "true" interval from there, then sharped
the third a little (shorted the interval) on the way back down, hit the
tonic low and slowly brought it back up to pitch.  To analyze this sort
of singing, one requires a tape recorder and one or both of an
exceptional ear or a computer.  You don't find it in the collectors' records.

Lastly, the major and minor scales are modes too.

So when we take a contemporary song, written and performed in major
mode, and try to force it into another mode because of some arbitrary
relation between its notes, we're stretching a *lot* of different
conceptual and perceptual structures.

This sort of thing happens in fiddle tunes, but they usually end up with
differrent names.  It's easy enough to flat the third to move a piece
into minor from major, for instance, but the fiddler quickly discovers
this implies other changes, and if he follows his instrument and ear,
the changes cascade from one to the other until the tune settles in its
new configuration - usually still recognizable compared to its brother,
but clearly distinct.

One of the groups I sing in does an 1850 "set piece" called "The Jolly
Soldier", which is a version of "The Rambling Soldier" with textual
modifications to make it a paean to George Washington.  Listening to Roy
Harris sing "The Rambling Soldier" with his simple concertina
accompaniment, it's clearly modal with a tonic of G (although I seem to
recall he shifts a note to make it fit the instrumental scale).  But the
arranger who added two harmony voices presumably found the same thing I
found - you can't do a conventional vocal arrangement in G (remember
Bronson).  His arrangement of the same tune in the same scale is in C
Major (since it's hexatonic, it fits the definitions of two modes
depending on what you decide is the tonic), and now instead of ending on
the tonic, it ends on the fifth even though it doesn't want to.  I
strongly dislike the piece; it feels out of balance to me, fighting
itself internally.  This, to me, is the effect when you take a melody
from the mode it "wants" to be in, and force it into another.

You can arbitrarily change the tonic, moving the song into another mode
(e.g. "Yesterday" as Phrygian), but the tune won't cooperate; it will
surely need some hammering about before it will wear its new mantle
lightly, if it ever does.

> "Yesterday", for example, has no accidentals
> in the melody line. - which brings me back to my original question.

The Beatles played it in F; if you transpose it to C - which is where it
ends on E - it needs an F# and G# in the second phrase, on "my" and
"trou-" of "troubles", and again on "I'm _not_ _half_ the man..." and
then on "love _was_ _such_ an easy...".  They *are* accidentals, because
coming back down they play the naturals; presumably they're there
because the B's wanted to use the E chord.  If you play it without them,
you're already changing the melody to fit your new scheme....

-Don Duncan


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%