in the following chapters cannot in fact be properly read without the use of metrical pause, and its importance will become more evident as the more complicated lines are considered.

TECHNIQUE

The technique of the metrical pause offers little difficulty. It is concerned with the position of the pause in the line or couplet, the number of syllables omitted, and to a less degree with the proportion of paused lines. The regular position of the pause in the line is that produced by omission of the fourth stressed syllable. There is, however, a very limited use at other positions, as in the line already quoted:

Son Davie, son Davie.

Whitman's venture in formal verse offers a good example of this type of omission:

O Captain! my Captain! our fearful trip is done,

The ship has weather'd every rack, the prize we sought is won.

Such omission must occur in conjunction with a logical pause, since otherwise the reading would be confused. Its usage is limited and comparatively unimportant.

The regular type of metrical pause is that which occurs at the division between the two half-lines (i.e. by omission of the fourth stressed syllable). In poems in which metrical pause is used the lines may be classed in five main groups according to the nature of the pause.1

The examples are all from Stevenson's Ticonderoga, Part I. Anacrusis and trisyllabic substitution are not regarded except where they influence the nature of the pause.

(0)|—u(0)|— u(u)|—^/^^|- ~(0)|— ~(0) |

Of these types of line the first might stand in regular ballad meter and does not necessarily represent a pause. If the line above, for instance, should be printed in the unbroken form, it might be classed as regular:

And cried to him for vengeance on the man that laid him low. -

If any one prefers to classify this as a variation of caesura, the present writer will not quarrel with him. (The study of metrics is the observation and description of phenomena rather than their nomenclature and classification into water-tight compartments.). There is, however, very little difference between the structure of Type I and of Type II, which is the characteristic case of an omitted syllable. The distinct arrangement of phrasing also gives the impression that a missing stressed syllable must be compensated by a pause. The notation used above adopts this view.

Type II represents the simplest form of metrical pause, and the most common. This form of omission is the one used most frequently by Lockhart, Macaulay, and most of the earlier poets. The types III, IV and V represent longer pauses with the omission of one or more unstressed as well as the stressed syllable. Type V reaches what is apparently the maximum omission in simple verse-a stressed syllable both preceded and followed by an unstressed. In verse without musical accompaniment the omission of two consecutive stressed syllables would be so extreme that some other explanation would probably be more satisfactory.1.

The five types of paused line fall evidently into two divisions according to ending. Types I, II and III have a feminine ending at the pause, while Types IV and V resemble the regular line in having a stressed syllable at the end of the first half-line. All of the types with feminine ending are subject to the variation which we have noticed as characteristic of such lines in song verse, that is, the last two syllables may both be logically important so that a Type II line (for example) approaches the structure:

1 The situation in dipodic verse is different. Here the statement must be changed to "two consecutive syllables of primary stress." (See p/a)

In such a line the first half-line can only with difficulty be said to omit a stressed syllable. In most poems such lines are in negligible proportion, but in a certain type of verse they become so frequent as to be a real factor in the verse structure.

The proportions of regular lines and of the various types of paused lines vary so greatly with different poems that little can be gained from percentage computations. Usage shades all the way from verse in which the regular line predominates to that in which it is exceptional. A percentage comparison of some poems by various writers may be offered although little metrical significance can be attached to it.

These percentages are not generally of much value, since in most cases they do not represent any consistent metrical practice. The groupings established by proportion of trisyllabic feet determine a particular verse style maintained throughout the whole poem, but no such practice can be determined in the use of metrical pause. Ticonderoga, for example, opens and closes with a very high percentage of Type IV lines, while almost all the regular lines are at the end of Part II and the beginning of Part III. There is not, however, even in this poem any hesitancy in combining different types of line in the same couplet. We may have, for instance, five different types in six consecutive lines (i.e., II, I, IV, III, I, and regular):

"My blood is on the heather,

My bones are on the hill;
There is joy in the home of ravens
That the young shall eat their fill.
My blood is poured in the dust,
My soul is spilled in the air;
And the man who has undone me
Sleeps in my brother's care."

"I'm wae for your death, my brother,
But if all of my house were dead,

I couldnae withdraw the plighted hand,
Nor break the word once said."

A line of such extended pause as Type IV may even be coupled with a regular line:

It fell in the dusk of the night

When unco things betide,

That he was aware of a captain man

Drew near to the waterside.

1 Penitence of Don Roderick, Lockhart; Last Buccaneer, Kingsley; Ticonderoga, Stevenson; Song of Diego Valdez, Kipling; the others are all Macaulay's.

The conclusion must be that the meter remains always fundamentally the same; syllables may bé omitted, but it is always felt that their place is filled by the pause.

There is, however, a distinct tendency in the grouping of various types of lines-the second line of each couplet is more likely to be regular than is the first. We may compare the proportions in this respect for the poems listed above:

This tendency is in fact more marked than the percentages show. In The Last Buccaneer the refrain line (Type III) keeps the proportion of regular lines low in couplets. In Ticonderoga the concentration of usa linen regula has artificially the effect; as a matter of fact in only one couplet of ne same this poem does a regular line stand in the second half of a couplet preceded requter by a paused line in the first half.

It is in The Lays of Ancient Rome that this arrangement of paused and regular lines may be best observed. Macaulay's own statement would lead us to believe that every first half-line omitted the final syllable.1 While this is not absolutely correct, it nevertheless holds almost regularly for the three of the Lays printed in the broken line. This arrangement of paused and regular lines seems to have rhetorical value. The period is kept in suspense through the paused line, and then descends to a conclusion in the regular second line of the couplet. This tendency is noticeable also in units larger than the couplet, when the poet is working up to a distinct climax. A good example of this is the offer of Horatius to defend the bridge; this speech begins with a regular line, runs through the unusual succession of nine paused lines and concludes with a line in which syllabic regularity is reinforced by internal rhyme. (Similar passages are the two concluding stanzas of Horatius, the speech of the consul nominating Aulus, and the reply of the Twin Brothers to Aulus.)

While it is the method of the present chapter not to consider doubtful cases, there are certain poems which the evidence renders scarcely doubtful. These poems can best be discussed in the present connection. The metrical form in question is that used principally by William Morris in Sigurd and other poems. A few lines from the opening of Sigurd will give a specimen

of the style:

There was a dwelling of Kings ere the world was waxen old;
Dukes were the door-wards there, and the roofs were thatched with gold;
Earls were the wrights that wrought it, and silver nailed its doors;
Earls' wives were its weaving-women, queens' daughters strewed its floors,

1 See Trevelyan's Life, Chapter IX. Macaulay counts the Latin -ius, -ia, and -ium endings as only one syllable.

And the masters of its song-craft were the mightiest men that cast
The sails of the storm of battle adown the bickering blast.
There dwelt men merry-hearted, and in hope exceeding great
Met the good days and the evil as they went the way of fate:
There the gods were unforgotten, yea whiles they walked with men,
Though e'en in that world's beginning rose a murmur now and again.

Various readers might give different decisions as to the metrical norm of these lines. A frequent opinion would probably be that they represent a six foot line with free variation of unstressed syllables. Many lines do in fact make good six foot lines (e.g. lines 1 and 6). On the other hand this hypothesis goes entirely to pieces in view of the considerable number of lines which cannot by any twist of the tongue be compressed into six feet:

Though e'en in that world's beginning rose a murmur now and again.

In such lines it is necessary to run into one foot four syllables (two of them logically important), to say nothing of a caesural pause. No simple English foot can stand under such a load.

There are two other possible hypotheses for the meter of Sigurd: (1) it represents two three-foot lines printed as one with free variation both within the feet and at endings and beginnings of lines; (2) it represents a full ballad line varying freely within feet and regularly employing a metrical pause between the half-lines. According to the former hypothesis the meter should be printed:

There was a dwelling of Kings

Ere the world was waxen old;
Dukes were the door-wards there,

And the roofs were thatched with gold.

This interpretation will accommodate in a loose way the various line types of Sigurd, but little support for it can be drawn from analogical forms or from the evidence of history. In addition to this the meter thus printed does not rhyme at the end of every line as would be expected. And finally we should not lose sight of the fact that the meter is not so printed by the poet himself; if Morris printed as he did, he evidently had a reason for considering the whole line as his unit.

While it is not possible to prove finally that the verse of Sigurd is a form of ballad meter, evidence converging from different lines tends to make this interpretation certain beyond any reasonable doubt. The principal argument in the way of a certain proof is the rather serious one that there is no single line of this verse which is necessarily ballad meter and cannot be anything else. On the other hand every line of the poem conforms to one of the five types of ballad meter with metrical pause.1 In the lines quoted we have for example a succession of types: IV, IV, II, II, I, II, I, I, II, I. One result of this conformation to types is to produce the unsymmetrical half-lines characteristic of paused ballad meter. An even stronger factor in the evidence is that there is a large proportion of Type I lines. As we have seen, these lines frequently represent merely a variation

The proportion of types varies with different sections of the poem. For the first hundred lines the percentages are: I, 34; II. 45; III, 1; IV, 19; V, 1. For the battle in the hall of Atli: I, 34; II, 63; IV, 2.