Phonozoic Text Archive, Document 102
Helmholtz's Vowel Theory and the Phonograph
Nature, Vol. XVII, March 14, 1878, p. 384.
THE following experiments with the phonograph are of
interest as throwing light on the nature of vowel sounds:--
Let a set of vowel sounds as A, E, I, O, U
(pronounced in Italian fashion) be spoken to the phonograph in any pitch, and
with the barrel of the instrument turned at a definite rate. Then let the
phonograph be made to speak them, first at the same rate, and then at a much
higher or lower speed. The pitch is, of course, altered, but the vowel sounds
retain their quality when the barrel of the phonograph is turned at very
different rates. We have made this experiment at speeds varying from about
three to one, and we can detect no alteration in the quality of the sounds.
According to Helmholtz, the characteristic
quality of each vowel is given by the prominence of a constituent note or notes,
of definite or approximately definite absolute pitch, in the sounds uttered.
Now obviously, the absolute pitches of the constituents of the vowel-sounds in
the above experiment were all altered in the same proportion, so that the
absolute pitch of the prominent notes varied greatly; but yet the vowel quality
was unchanged. This experiment, therefore, seems to give results in
contradiction of Helmholtz’s theory as we understand it.
At the same time we have found, in the course
of experiments, of which a full account will shortly be communicated to the
Royal Society of Edinburgh, that if a scale be sung to the phonograph with one
vowel sound, such as O, the wave-form of the marks on the tinfoil does not
remain unchanged at all pitches. We have not yet had time to analyse the curves
so obtained into their harmonic constituents. Such an analysis will show
whether the changes we have observed in the wave-form as the pitch rises, are
due to a change in the relation of the amplitudes of the constituents present,
or only to a variation of phase.
FLEEMING JENKIN
J. A. EWING
Edinburgh, March 11