it long; the other, by the _smart percussion_--of the voice, as
necessarily _makes it short_"--_Ib._, p. 57. Now it is all a mistake,
however common, to suppose that our accent, consisting as it does, in
stress, enforcement, or "percussion of voice," can ever _shorten_ the
syllable on which it is laid; because what increases the quantum of a vocal
sound, cannot diminish its length; and a syllable accented will always be
found _longer_ as well as _louder_, than any unaccented one immediately
before or after it. Though weak sounds may possibly be protracted, and
shorter ones be exploded loudly, it is not the custom of our speech, so to
deal with the sounds of syllables.

OBS. 11.--Sheridan admitted that some syllables are naturally and
necessarily short, but denied that any are naturally and necessarily long.
In this, since syllabic length and shortness are relative to each other,
and to the cause of each, he was, perhaps, hardly consistent. He might have
done better, to have denied both, or neither. Bating his new division of
accent to subject it sometimes to short quantity, he recognized very fully
the dependence of quantity, long or short, whether in syllables or only in
vowels, upon the presence or absence of accent or emphasis. In this he
differed considerably from most of the grammarians of his day; and many
since have continued to uphold other views. He says, "It is an _infallible
rule_ in our tongue that no vowel ever has a long sound in an unaccented
syllable."--_Lectures on Elocution_, p. 60. Again: "In treating of the
simple elements or letters, I have shown that some, both vowels and
consonants, are _naturally short_; that is, whose sounds _cannot possibly_
be prolonged; and these are the [short or shut] sounds of ~e, ~i, and ~u,
of vocal sounds; and three pure mutes, k, p, t, of the consonant; as in the
words _beck, lip, cut_. I have shown also, that the sounds of all the other
vowels, and of the consonant semivowels, may be prolonged to what degree we
please; but at the same time it is to be observed, that all these may also
be reduced to a short quantity, and are capable of being uttered in as
short a space of time as those which are naturally short. So that they who
speak of syllables as absolutely in their own nature long, _the common cant
of prosodians_, speak of a nonentity: for though, as I have shown above,
there are syllables absolutely short, which cannot possibly be prolonged by
any effort of the sp