sounds. Indeed it is obvious that whenever we feel the expression of
a song to be due to its quickness or slowness of movement to smoothness
of flow, loudness of utterance, and so on we are, in fact, interpreting
the muscular actions which produce sound, in the same way in which we
interpret muscular action generally. But this leaves unexplained the
more subtle and more specific effect which we call the _musical_
expression of the song the delight given by its melody, or even by the
separate sounds which make up the melody. This is an effect indefinable
in language one which, so far as I am aware, no one has been able to
analyse, and which the ingenious speculation of Mr. Herbert Spencer as
to the origin of music leaves quite unexplained. For it is certain that
the _melodic_ effect of a series of sounds does not depend in the least
on their loudness or softness, or on their _absolute_ pitch. A tune is
always the same tune, whether it is sung loudly or softly, by a child
or a man; whether it is played on a flute or on a trombone. The purely
musical effect of any sound depends on its place in what is technically
called a scale; the same sound producing absolutely different effects
on the ear, according as it is heard in connection with one or another
series of sounds.
It is on this _relative_ association of the sounds that all the
essentially characteristic effects which are summed up in the phrase
musical expression, depend. But why certain associations of sounds
have such-and-such effects, is a problem which yet remains to be
solved. These effects must indeed, in some way or other, be connected
with the well-known arithmetical relations between the rates of
vibration of the sounds which form a musical scale. And it is
possible but this is merely a suggestion that the greater or less
mechanical facility with which the vibrating apparatus of the human
larynx passes from one state of vibration to another, may have been a
primary cause of the greater or less pleasure produced by various
sequences of sounds.
But leaving aside these complex questions and confining ourselves to
the simpler sounds, we can, at least, see some reasons for the
association of certain kinds of sounds with certain states of mind. A
scream, for instance, uttered by a young animal, or by one of the
members of a community, as a call for assistance, will naturally be
loud, prolonged, and high, so as to penetrate to a distance. For
Helmholtz has shown[407] that, owing to the shape of the internal
61