Stephen Heersink asks what's wrong with the following:

>The Harvard Dictionary of Music define melody as "a coherent succession
>of pitches," where pitches are understood to be a "stretch of sound whose
>frequency is clear and stable enough to be heard as not noise; succession
>means that several pitches occur; coherent means that the succession of
>pitches in accepted as belonging together."

Surely it's obvious that "belonging together" is hopelessly vague -- and
allows much too much to count as a melody, including whole major concert
works.  ALL the notes in Beethoven's succession of pitches known as the
Emperor Concerto (for example) are "accepted as belonging together".  I
surely wouldn't want to change a one.  Does that make the entire concerto
a single melody?

Of course not.  Long ago in this thread, which I initiated, and which has
gotten way off base onto the question of defining harmony, I began by
saying that I started from the idea that the sequence of notes (or
"succession of pitches") formed a single musical "thought" -- and asked if
anyone could shed light on what should count as a single musical thought.

Others agreed with this, but observed that (i) there were musical
units shorter than a whole thought -- such as a "phrase" -- and some that
were longer as well; (ii) that the analogy to language and grammar is
illuminating -- a sentence is a unit, yet there can be smaller, and larger
units.  Etc.  Some were happy to stop at that point -- it's a unit if you
think it is etc.  -- others, like me, who think nothing is just "whatever
you think it is", hoped for more.  "To the store" is NEVER a sentence, and
a three page sentence ain't really a sentence either.  As for the key next
question, what does it take to produce the sense of a whole musical thought
or unit, I have some ideas, but they involve a broader theory of what art
in general (and music in particular) communicates, and I wanted rather to
hear what the state of thinking was on this, rather than to sketch out my
own view.

Some suggested that part of what makes a melody the melody it is, is
something about its rhythm: change the rhythm enough and you change the
melody.  That seems clearly right, and a useful insight.

Some insisted that there was a harmonic element to melody -- the notes
must stand temporally in the sorts of relationships they need to stand to
at the same moment in simple harmonic chords.  This seemed to me a really
interesting idea, worth development.  Some directions for development were
suggested, but they turned into polemics and a fair bit of obscure language
(my friend), and the discussion got so involved in what harmony was and how
pervasive it was that the connection with melody got entirely lost.
Somewhere in the process my friend pooh pooh-ed those of us who prefer
sustained melodic content in our music as preferring "simple little
melodies" or some such, and the discussion really got off.

On that issue I must interject that what I think is at issue is rather that
some late romantic and much post-romantic music begins to express a sense
of life (i.e.  an emotional metaphysics) rarely expressed in music before,
one that is darkly malevolent and pained.  and that some of us reject that
sense of life, indeed violently (notwithstanding that others think that
their increased currency is proof of their validity).  Although I can
appreciate the technical accomplishments of such works, and their success
at conveying their view of the world, I philosophically reject that family
of views of the world, and demand rich and lush melody embedded in complex
but affirmingly resolving structures in music I personally like.

But, having said that, and restored, at least in my soul a balance on
that issue, I'd like to return us to the main issue -- what melody is.
Is there more to say about what makes an auditory unit, a single melody?
Going back to the language analogy, if we struggle to say what makes "a
complete thought", we think in terms of the facts being expressed, or
opined.  Something is said about a subject.  Feed in separate subjects
and separate things said about them and you start saying it's really two
or more sentences, or should be, even if there's only one period.  Even a
philosophical essay has sentences, as does this letter, and it too has a
subject about which something is said, etc.  If we knew more about WHAT IS
SAID in music, I'm suggesting, we might be better able to say when we HAVE
SAID a single unit.  Any thoughts about WHAT MUSIC SAYS, then? That seems
the next question.

Allan Gotthelf