Let us observe, in passing, that when a vowel sound changes through transposition of the ac cent, the Hebrew grammarians—instead of trust ing that the voice will of itself modify the vowel when the accent is shifted —generally think it necessary to depict the vowel differently : which is one 'principal cause of the complicated changes of the vowel points.
A second concomitant of the accent is less marked in English than in Italian or Greek ; namely—a musical elevation of the voice. On a piano or violin we of course separate entirely the stress given to a note (which is called forte and staccato) from its elevation (which may be A, or c, or F) ; yet in speech it is natural to execute in a higher tone, or as we improperly term it, in a higher key, a syllable on which we desire to lay stress : possibly because sharp sounds are more distinctly heard than flat ones. Practically, therefore, accent embraces a slide of the voice into a higher note, as well as an emphasis on the vowel ; and in Greek and Latin it would appear that this slide upwards was the most marked peculiarity of accent, and was that which gained it the names accentzts. Even at the present day, if we listen to the speech of a Greek or Italian, we shall observe a marked ele vation in the slides of the voice, giving the appear ance of great vivacity, even where no peculiar sentiment is intended. Thus, if a Greek be requested to pronounce the words crocDta (wisdom), rapapoki) (parable), his voice will rise on the f and 7) in a manner never heard from an Englishman. In ancient Greek, however, yet greater nicety existed ; for the voice had three kinds of accent, or slides, which the grammarians called flat, sharp, and circumflex ; as in TiS, It is at the same time to be remarked, that this flat accent was solely oratorical ; for when a word was read in a vocabu lary, or named in isolation, or indeed at the end of a sentence, it never took the flat accent, even on the last syllable ; except, it would seem, the word cis, a certain one. In the middle of a sentence, however, the simple accent (for we are not speaking of the circumflex) on a penultima or antepenultima was always sharp, and on a last syllable was flat. Possibly a stricter attention to the speech of the best educated modem Greeks, or, on the contrary, to that of their peasants in isolated districts, might detect a similar peculiarity : but it is generally believed that it has been lost, and some uncertainty therefore naturally rests on the true pronunciation.
On the whole, it is most probable that the flat ac cent was a stress of the voice uttered in a lower note, much as the second accent in gnintifather; that the sharp accent was that which prevails in modem Greek, and has been above described ; and that the circumflex combined an upward and a downward slide on the same vowel. The last was naturally incapable of being executed, unless the vowel was long; but the other two accents could exist equally well on a short vowel.
In English elocution various slides are to be heard, more complicated than the Greek circumflex ; but with us they are wholly oratorical, never vocabular. Moreover, they are peculiar to vehement or vivacious oratory ; being abundant in familiar or comic speech, and admissible also in high pathetic or indignant declamation : but they are almost entirely excluded from tranquil and serious utter ance.
Secondary Accent. —On the same word, when it consists of many syllables, a double accent is frequently heard, certainly in English, and probably in most languages ; but in our own tongue one of the two is generally feebler than the other, and may be called secondary. If we agree to denote this by the flat accent C) of the Greeks, we may indicate as follows our double accent : considerAtion, disobedience, impretending ; secondary, accessory, peTempthrily.
We have purposely selected as the three last ex amples cases in which the secondary accent falls on a very short or obscure vowel, such as can never sustain the primary accent.
In some cases two syllables intervene between the accents, and it may then be difficult to say which accent is the principal. In aristocrat, equalize, anti dale, the first syllable has a stronger accent than the last ; but in aristocratic, equalization, tintediltivian, they seem to be as equal as possible, though the latter catches the ear more. In aristocracy, the former is beyond a doubt secondary ; but here the two are separated by only one syllable. Predeter mination has three accents, of which the middle most is secondary.