On Quantitative Metre 339into the category of longs by the force of the consonant weigh-
tage, but these conditions are not naturally present in English
verse.
There is therefore no good reason, or at least no essential
reason, for the admission of a rule allowing or obliging a throw-
back of influence from a following word upon its predecessor.
In accentual or stress metre no such rule prevails, — one never
thinks of this element in arranging one’s line; there is nothing
that compels its adoption in quantitative verse. If these initial
consonants created an obstacle to the pace of the voice sufficient
to make it linger or pause, then such an effect would be justified,
— the closing short syllable of the preceding word would or
might be lengthened: but, normally, the obstacle is so slight that
it is not felt and the voice takes it in its stride and passes on with-
out any slackening or with only a slight slackening of its pace.
The distinctness of each word from another does not, indeed,
create any gap or pause, but it is strong enough to preserve for
it its independence, its separate self-value in the total rhythm of
the line, the word-group or the clause. This does not destroy
the value of consonant weight in the sound system; it is evident
that a crowding or sparseness of consonants will make a great
difference to the total rhythm, it will produce a greater or less
heaviness or lightness; but that is a rhythmic effect quite distinct
from any imperative influence on the metre. A trochee does not
become a spondee, a dactyl does not become a cretic because its
final syllable is followed by a consonant or even by a group of
consonants. There is, then, no sense in dragging in the classical
rule where its admission is quite contrary to the natural instinct
and practice of the language.
If these considerations are accepted as valid, the way lies
open for the construction of true quantitative metre; a sound
and realistic theory of it becomes possible. Four rules or sets of
rules can be formulated which will sum up the whole base of the
theory: —
(1) All stressed syllables are metrically long, as are also all
long-vowel syllables even without stress.
All short-vowel syllables are metrically short, unless they