his reader accept certain assumptions and agree to certain definitions throughout a given discussion; you cannot logically follow his argument unless you accept his meaning of the terms number, straight line, or parallelopipedon Metrical discussions to be at all fruitful require a similar agreement on the meaning of terms like stress, accent, or foot, throughout the same argument.

CHAPTER II

METER STRESS-ACCENT

Most readers will agree that the first obvious difference between verse and prose is that verse is divided into certain units called lines, and that these lines must be "metrical." It is the definition of "metrical" that causes disagreement. Let us try to form a definition which may be one basis for a study and classification of verse.

Suppose we grant that the following indicates a possible metrical reading of the opening lines of Henry IV, Part I:

So sháken ás we are so wán with cáre

Find we a tíme for frighted peace to pánt

And breathe short-winded áccents of néw bróils

To bé comménc'd in stránds afár remote.

No more shall trénching wár channel her fields . . .

The first and fourth lines might be explained as metrical because every other syllable receives emphasis, but this will not explain why the second, third and fifth, when read as indicated, are also metrical. Furthermore, the following lines from Shelley and Tennyson, though they seem quite different in the distribution of emphasis from the first quoted above, occur in contexts of the same kind of verse:

When night makes a weird soúnd of its own stillness.
(Shelley: Alastor.)

The little innocent soúl flítted away.

(Tennyson: Enoch Arden.)

One listening to the indicated metrical reading of the passage from Henry IV can detect a huddling of the syllables "Find we a," "accents of," and "channel her" and also a

314

slight lengthening of "new" and "war." The reader seems to jump from one point of emphasis to the next at something approaching equal intervals of time, and to let all syllables between take care of themselves. The lines of verse when read metrically are divided into equal time parts which correspond to the bars of a phrase in music. That is, both verse and music primarily depend upon rhythm. To make the parallel between verse and music. clearer we may represent by musical notation one possible reading of the first line quoted above. The bars, as in written music, precede the emphatic syllables.

1

Possible readings of the less regular lines might be these:

We may carry the parallel further by reading the lines to the rhythm of a metronome, the ticks of which occur at exactly equal intervals of time. The reading will sound stiff and expressionless, but it will still satisfy us as "metrical." And we must remember that playing a musical instrument in the unmodified tempo of the metronome would have as awkwardly stiff an effect as our experiment with verse. Expression in good reading or in good playing may

"Rhythm is the harmonious repetition of certain fixed sound relations, time being the basis, just as in dancing or music. . . ." (Gummere: Poetics, p. 136.)

Of course some readers may prefer:

necessitate frequent slight departures from an exact equality of time divisions, but the departures must not be so considerable as to destroy the feeling that rhythm is present. The ear is a very imperfect instrument, and our sense of rhythm is satisfied if the phenomena marking the rhythm occur at only sensibly equal intervals of time. In fact, the sophisticated ear finds greater pleasure in these slight variations, just as the tempo of a virtuoso subtly playing with individuality about a norm of exact rhythm pleases us much more than the inevitable rhythmic accuracy of the pianola. "In a fine oriental rug the hand-made, slightly irregular ornament and intentionally varied symmetry are more interesting and more beautiful than the dead mechanical precision of a machine-woven pattern; but a geometrically perfect pattern may be said to lie at the basis of the Persian weaver's design. So in verse there is an exact pattern underneath, to which the reader approximates, now more closely, now less, as the phonetic character of the words or the requirements of sense and expression permit or demand." 4

A possible way, then, of explaining the basis of verse is to say that a line is metrical when it is divided into sensibly equal time parts. This definition assumes a unit called a line, which does not exist in prose. There is, after all,

The margin of inaccuracy in the perception of the different senses is one of the common subjects for experiment in psychology. "It is a familiar condition that two stimuli must differ at least by a minimal amount, in order that we may become aware of their differences. . . . For lifted weights it is, we shall see, about one-thirtieth. For pressure on the finger-tip it has been found to be about one-twentieth, for brightness of light about one-hundredth, and for intensities of noise, about one-third; two sounds of different loudness can just be distinguished as different, provided that the intensity of one is greater by about one-third than that of the other." C. S. Myers, A Text-book of Experimental Psychology, 1909, pp. 255-6.

4 T. D. Goodell: Chapters on Greek Metric, p. 82. The chapter on "Rhythm and Language" is an admirable discussion of the subject of rhythm in general.

much to be said for the popular distinction that each line in verse begins with a capital but in prose does not. The sentence you are reading now, for instance, may be divided easily into verse of equal line lengths, each with its five stresses, thus:

The sentence you are reáding nów, for instance
May bé divíded easily into vérse

Of équal líne lengths, each with its five stresses.

But you were not conscious of any such divisions when you read the sentence first. Read as prose it has a very different sound from when read as verse. For one thing, each group of words which forms a line when read as verse, your prose reading probably divided into a different number of time parts, so that the groups could not be recognized as three units of the same metrical form. A more detailed discussion of the differences between prose and verse will be taken up later."

6

Further support for the definition at which we have arrived -a line is metrical when divided into sensibly equal time parts may be found in the exhaustive researches of M. Verrier and other psychologists, who have shown by actual measurement that the time parts of verse are much more nearly equal than those of singing. Then, too, many of the great makers of poetry have shown by the chanting manner, often monotonous, of their own recitations that they felt a musical rhythm in verse as its fundamental quality. There is an agreement of evidence that this was characteristic of Tennyson's and Poe's reading; Scott composed verses on horseback, and Wordsworth marked his rhythm by beating the Cumberland hills with his cane. The principle is likewise borne out by Professor Gummere's

'See Chapter 4.

6 Principes de la Métrique Anglaise, III, 241. M. Verrier finds this true of even the most irregular verse compared with the most regular songs sung to a single syllable.