Part 2 — CHAPTER X THE SEPTENARY, THE ALEXANDRINE, AND THE THREE-FOOT LINE

§ 135. The Septenary is a favourite Middle English metre, going back to a Mediaeval Latin model. It cannot, however, be definitely determined whether this is to be found in the (accentual) catalectic iambic tetrameter, an example of which is preserved, among other instances, in the Planctus Bonaventurae (1221–74) printed by Mone in his Latin Hymns of the Middle Ages, which begins as follows:

or possibly in another Latin metre which was a far greater favourite with the Anglo-Norman Latin poets. This is the (accentual) brachycatalectic trochaic tetrameter, which frequently occurs, among other instances, in the poems ascribed to Walter Map, e.g. in the still popular verses:

The result of an attempt to adopt this metre in Middle English might, on account of the preference of the language for iambic rhythm, very naturally be to transform it into the iambic catalectic tetrameter by the frequent addition of an unaccented opening syllable at the beginning of each half-line. Probably the latter verse-form was the model, as may be seen from Leigh Hunt’s Modern English translation of the Latin drinking-song just quoted.[145]

Moreover, many mediaeval Latin verses also have a wavering rhythm resulting in a form at times characterized by level stress, e.g.

§ 136. These verses correspond pretty exactly, in their metrical structure, to the opening lines of the Moral Ode, which, as far as is known, is the earliest Middle English poem in septenary lines, and dates from the twelfth century:

The other common licences of even-beat metre which affect the rhythm of the line, the metrical value of syllables, and the word-accent, also occur in the Moral Ode. Suppression of the anacrusis is very often met with; it occurs, for instance, in the first hemistich, in lines 1 and 4 above; in the second hemistich, ér ic hít iwíste l. 17, in both, þó þet hábbeð wél idón | éfter híre míhte, l. 175; so that a pure iambic couplet seldom occurs, although the iambic rhythm is, on the whole, predominant. The omission of unaccented syllables in the middle of the line is also often found (although many verses of this kind probably require emendation), as Ne léve nó mán to múchel 24; also in the second hemistich, as and wól éche dede 88. Transpositions of the accent are quite usual at the beginning of the first as well as of the second hemistich: Elde me ís bestólen ón 17; síððen ic spéke cúðe 9. Level stress is also not absent: For bétere is án elmésse bifóre 28. We often meet with elision, apocope, syncope, slurring of syllables, and the use of a disyllabic thesis both at the beginning of the line and in other positions: þo þet wél ne dóeþ þe wíle he múȝe 19; nís hit búte gámen and glíe 188. A noteworthy indication of want of skill in the handling of the Septenary in this first attempt is the frequent occurrence of a superfluous syllable at the close of the first hemistich, which should only admit of an acatalectic ending, e.g.: Hé scal cúme on úuele stéde | búte him Gód beo mílde 26; Eíðer to lútel ánd to múchel | scal þúnchen éft hem báthe 62, &c. The end of the second hemistich, on the other hand, in accordance with the structure of the metre, is in this poem always catalectic.

§ 137. The irregularity of the structure of the Septenary rhyming line of the Moral Ode stands in marked contrast with the regularity of the rhymeless Septenary verse of the Ormulum. The first hemistich here is always acatalectic, the second catalectic, and the whole line has never more nor less than fifteen syllables.

Hence the only metrical licences that occur here are elision, syncope, and apocope of the unaccented e of some inflexional endings, and the very frequent admission of level stress in disyllabic and polysyllabic words, which are to be found in all places in the line:

In all such cases, in the versification of Orm, whose practice is to count the syllables, there can only be a question of level stress, not of inversion of accent. Ennglisshe at the beginning of the second hemistich of the above line, 322, is no more an example of inversion of rhythm than in the hemistich Icc háfe wénnd inntill Ennglíssh l. 13

§ 138. After the Moral Ode and the Ormulum the Septenary often occurs in combination with other metres, especially the Alexandrine, of which we shall speak later on.

In some works of the thirteenth and fourteenth centuries the Septenary was, however, employed in a fairly unmixed form, as, for instance, in the Lives of Saints, ed. Furnivall, 1862, the Fragment of Popular Science, ed. Wright in Popular Treatises on Science, London, 1841, and several others.

The most important deviation from the Septenary of Orm and of the Moral Ode is the frequent occurrence of long lines with a masculine instead of the usual feminine ending. Both forms are to be found in the opening lines of the Fragment of Popular Science:

It may fairly be assumed that the structure of the Alexandrine (which, according to French models, might have either a masculine or a feminine ending) may have greatly furthered the intrusion of monosyllabic feet into the Septenary verse, although the gradual decay of the final inflexions may likewise have contributed to this end. For the rest, all the rhythmic licences of the Septenary occurring in the Moral Ode are also to be met with here; as, for instance, the suppression of the anacrusis in the first hemistich of l. 4 of the passage quoted, and in the second of l. 2, and the omission of the unaccented syllable in the second hemistich of the fourth line, the inversion of accent and disyllabic thesis in the first hemistich of the third line, and other licences, such as the anapaestic beginning of the line, &c., in other places in these poems (cf. Metrik, i, p. 246)

§ 139. In lyrical poems of this time and in later popular ballad poetry the Septenary is employed in another manner, namely, in four-lined stanzas of four- and three-foot verse, rhyming crosswise, each of which must be looked on as consisting of pairs of Septenaries with middle rhyme inserted (interlaced rhyme), as is clearly shown by the Latin models of these metrical forms quoted above (p. 192). Latin and English lines are thus found connected, so as to form a stanza, in a poem of the fifteenth century:

Political Poems, ii. 249.

In many lyrical poems of the older period some stanzas rhyme in long lines, others rhyme in short lines, which shows the gradual genesis of the short-lined metre, rhyming throughout. Thus, in the poem in Wright’s Spec. of Lyr. P., p. 90, the opening verses of the first stanza rhyme in long lines:

whereas those of the second rhyme in short lines:

Instances of this kind are frequent; but the four lines of the single stanzas are never completely rhymed throughout as short-lines, as, for instance, is the case in the opening parts or ‘frontes’ of the stanzas of the poems in Wright’s Spec. of Lyr. P., pp. 27 and 83, the lines of which are far more regularly constructed. The rhymes are in these compositions still generally disyllabic.

The metrical structure of the old ballads The Battle of Otterborn and Chevy Chase is similar to that of the poem just quoted. In those ballads some original long lines are provided with middle rhyme, others not, so that the stanzas partly rhyme according to the formula a b c b, partly according to the formula a b a b. The versification is, moreover, very uneven, and the endings are, as a rule, if not without exception, masculine:

The ballads of the end of the Middle English period are generally composed in far more regular lines or stanzas. The feminine endings of the Septenary are, however, as a rule replaced by masculine endings, whether the lines rhyme crosswise or only in the three-foot verses. Cf. the ballad, The Lady’s Fall (Ritson, ii. 110), which, however, was probably composed as late as the Modern English period:

§ 140. In Modern English the Septenary has been extensively used, both in long and in short rhyming lines. One special variety of it, consisting of stanzas of four lines, alternately of eight and six syllables (always with masculine ending), is designated in hymn-books by the name of Common Metre.

In the long-lined form this metre occurs at the beginning of this period in poems of some length, as, for instance, in William Warner’s Albion’s England, and in Chapman’s translation of the Iliad. Here, too, the ending of the line is almost without exception masculine, and the rhythm, on the whole, pretty regular, although this regularity, especially in Chapman, is, in accordance with the contemporary practice, only attained by alternate full pronunciation and slurring of the same syllables (Romanic -ion, -ious, &c., and Germanic -ed, &c.) and by inversion of accent. The caesura is always masculine at the end of the first hemistich, but masculine or feminine minor caesuras are often met with after the second or in the third foot, sometimes also after the first or in the second:

Occásioned thús: | Chrýses the príest || cáme to the fléet to búy. i. 11.

To plágue the ármy, | ánd to déath || by tróops the sóldiers wént.ib. 10.

Secondary caesuras also occur, though less frequently, in other places in the line, particularly in the second hemistich:

But íf thou wílt be sáfe begóne. || This sáid, | the séa-beat shóre. ib. 32.

All mén in óne aróse and sáid: || Atrídes, | nów I sée. ib. 54.

These last examples suffice to show the rich variety of the caesura, which may be referred perhaps to the influence of blank verse, in the management of which Chapman displays great skill, and to the frequent use which he makes of the enjambement. Rhyme-breaking also sometimes occurs in his verse. Occasionally three consecutive lines rhyme together, as in W. Warner, whose versification is otherwise extremely regular, similar to that of lyrical poetry. In this branch of poetry the Septenary, with the simple rhyme-order a b c b and especially with the more artistic form a b a b, has continued to be very popular from the time of Wyatt down to the present day. The three-foot line has naturally in most instances a masculine ending, but lines also occasionally occur with feminine rhyme. In many poems the feminine rhyme is, moreover, regularly employed in this metre; as, for instance, in Burns’s To John Taylor (p. 158):

In ballad poetry, on the other hand, the Septenary metre tends to assume a somewhat freer construction, similar to, though not so capricious as that in the old ballads edited by Percy. A well-known example is offered by Coleridge’s Rime of the Ancient Mariner:

Two unaccented opening syllables and two unaccented syllables in the middle of the line are, in particular, often met with.

§ 141. The Septenary in combination with other metres. After its occurrence in the Moral Ode and the Ormulum the Septenary, as we have seen, appears at first very seldom by itself, but generally in connexion with other metres, especially the old long line in its freer development, the four-foot metre (though more rarely), and, particularly, the Alexandrine.

The Middle English Alexandrine was constructed on the model of the Old French Alexandrine—except for the use of Teutonic licences in even-beat rhythm—and it thus possessed four different types, which the following examples from On god Ureison of ure Lefdi[146] may serve to illustrate. We give the corresponding Old French metrical types from the Roman d’Alixandre (Bartsch, Chrestomathie de l’ancien français, p. 175).

a. Masculine caesura with masculine line-ending:

En icele forest, | dont voz m’oëz conter. 24.

Nim nu ȝéme to mé, | so me bést a béo ðe béo. 129.

b. Feminine (epic) caesura with masculine line-ending:

nesune male choze | ne puet laianz entrer.25.

vor þín is þé wurchípe, | ȝif ich wrécche wel iþéo.130.

c. Masculine caesura with feminine line-ending:

Moult fut biaus li vregiers | et gente la praële.1.

Þine blísse ne méi | nówiht únderstónden.31.

d. Feminine (epic) caesura with feminine line-ending:

Moult souëf i flairoient | radise et canele.2.

Vor ál is gódes ríche | an únder þíne hónden.32.

Alexandrines of this sort, particularly of the last type, are found in a group of poems of the close of the twelfth, or beginning of the thirteenth century, intermingled with Septenaries, and also, though more seldom, combined with four-beat alliterative rhyming long lines and with four-foot verses. Such poems are On god Ureison of ure Lefdi (quoted above), A lutel soth sermon (Old English Miscellany, ed. R. Morris, pp. 186 ff.), and A Bestiary (ib. pp. 1–25).

The following lines from A lutel soth sermon may serve to illustrate this mixture:

Here we have Septenaries (ll. 1, 4, 7) and Alexandrines (ll. 2, 3, 5, 6, 8) intermixed in ll. 1–8, eight-foot long lines resolved by means of sectional rhyme into four-foot lines in ll. 9–12, and four-beat rhyming alliterative long lines of the freer type in ll. 13–16. The easy intermixture of metres may be explained by the fact that in all these different long-lined metrical forms four principal stresses are prominent amid the rest, as we have indicated by accents (´).

§ 142. In the Bestiary this mixture of metrical forms has assumed still greater proportions, inasmuch as alongside of the long-lined rhyming Septenaries and alliterative long lines there are found also Layamon’s short-lined rhyming verses and Septenary lines resolved into short verses by middle rhyme.

The following passages may more closely illustrate the metrical construction of this poem; in the first place, ll. 384–97:

Here we have unmistakable long lines of the freer type.

In other passages the alliterative long lines pass into Septenaries, as, for instance, ll. 273–98:

In a third instance (ll. 628–35) Septenary and four-foot lines run into one another:

In many passages in the poem one or other of these different types of verse occurs unmixed with others. Thus we have short couplets in the section 444–5; in ll. 1–39 alliterative rhymeless verse, occasionally of marked archaic construction, concluding with a hemistich (39) which rhymes with the preceding hemistich so as to form a transition to the following section (ll. 40–52), which again consists of four-foot and Septenary verses. These are followed by a section (ll. 53–87) in which four-foot and three-foot lines (that is to say, Alexandrines) rhyming in couplets are blended; and this is succeeded by a further section (ll. 88–119) mostly consisting of Septenaries resolved by the rhyme into short lines. (Cf. Metrik, i, §§ 79–84.)

Hence we may say that the poet, in accordance with his Latin model (likewise composed in various metres), has purposely made use of these different metrical forms, and that the assertion made by Trautmann and others,[147] that the Septenary of the Ormulum and the Moral Ode, which is contemporary with Layamon, represents the final result of the development of Layamon’s verse (the freer alliterative long line), must be erroneous.

§ 143. In On god Ureison of ure Lefdi, on the other hand, the alliterative long lines play only an insignificant part, a part which is confined to an occasional use of a two-beat rhythm in the hemistichs and the frequent introduction of alliteration. Septenaries and Alexandrines here interchange ad libitum.

The following short passage (ll. 23–34) will suffice to illustrate these combinations of metres:

Lines 26 and 34, perhaps also 25 and 30, are Septenaries, l. 28 is the only line of the poem which contains two beats in both hemistichs (hemistichs of this sort are further found in the first hemistich of ll. 3, 12, 44, 72, 77, and in the second of ll. 30, 45, 46, 52, and 70); the remaining lines of this passage are most naturally scanned as Alexandrines.

§ 144. Now, this unsystematic combination of Alexandrines and Septenaries is a metre which was especially in vogue in the Middle English period. In this metrical form two religious poems, The Passion of our Lord and The Woman of Samaria (Morris, Old English Miscellany), were composed so early as the beginning of the thirteenth century. From the first we quote ll. 21–4:

Many lines of these poems may be scanned in both ways; in the third line of the preceding extract, for instance, we may either take the second syllable of the word ofsprung, in the manner of the usual even-beat rhythm, to form a thesis (in this case hypermetrical, yielding an epic caesura), or we may regard it as forming, according to ancient Germanic usage, a fourth arsis of the hemistich, which would then belong to a Septenary. At any rate, this scansion would, in this case, be quite admissible, as indeed the other licences of even-beat rhythm all occur here.

It is in this metre that the South English Legends of Saints (Ms. Harleian 2277) and other poems in the same MS., as the Fragment on Popular Science (fourteenth century), are written. The same holds good for Robert of Gloucester’s Rhyming Chronicle (cf. Metrik, i, §§ 113, 114). Mätzner (in his Altengl. Sprachproben, p. 155), and Ten Brink (Literaturgeschichte, i, pp. 334, 345) concur in this opinion, while Trautmann (in Anglia, v, Anz., pp. 123–5), on a theory of metrical accentuation which we hold to be untenable, pronounces the verses to be Septenaries.

The following passage (Mätzner, Altengl. Sprachproben, i, p. 155) may serve to illustrate the versification of Robert of Gloucester:

§ 145. At the end of the thirteenth century the Septenary and Alexandrine were, however, relegated to a subordinate position by the new fashionable five-foot iambic verse. But we soon meet them again in popular works of another kind, viz. in the Miracle Plays, especially in some plays of the Towneley Collection, like the Conspiratio et Capcio (p. 182), and actually employed quite in the arbitrary sequence hitherto observed, Alexandrine sometimes rhyming with Alexandrine, Septenary with Septenary, but, more frequently, Alexandrine with Septenary. A passage from the Towneley Mysteries may make this clear:

This metre is also employed in many Moral Plays with a similar liberty in the succession of the two metrical forms.

But we may often observe in these works, as, for instance, in Redford’s Marriage of Wit and Science (Dodsley, ii, p. 325 sq.), that Alexandrines and Septenaries are used interchangeably, though not according to any fixed plan, so that sometimes the Septenary and sometimes the Alexandrine precedes in the couplet, as, for instance, in the last four lines of the following passage (Dodsley, ii, p. 386):

§ 146. In other passages in this drama, e.g. in the speech of Wit, p. 359, this combination (Alexandrine with Septenary following) occurs in a sequence of some length. It existed, however, before Redford’s time, as a favourite form of stave, in lyrical as well as in narrative poetry, and was well known to the first Tudor English prosodists under the name of The Poulter’s Measure.[148]

The opening lines of Surrey’s Complaint of a dying Lover (p. 24) present an example of its cadence:

Brooke’s narrative poem Romeus and Juliet, utilized by Shakespeare for his drama of the same name, is in this metre. Probably the strict iambic cadence and the fixed position of the caesura caused this metre to appear especially adapted for cultured poetry, at a time when rising and falling rhythms were first sharply distinguished. It was, however, not long popular, though isolated examples are found in modern poets, as, for instance, Cowper and Watts. Thackeray uses it for comic poems, for which it appears especially suitable, sometimes using the two kinds of verse promiscuously, as Dean Swift had done before him, and sometimes employing the Alexandrine and Septenary in regular alternation.

§ 147. The Alexandrine runs more smoothly than the Septenary. The Middle English Alexandrine is a six-foot iambic line with a caesura after the third foot. This caesura, like the end of the line, may be either masculine or feminine.

This metre was probably employed for the first time in Robert Mannyng’s translation of Peter Langtoft’s rhythmical Chronicle, partly composed in French Alexandrines. The four metrical types of the model mentioned above (p. 198) naturally also make their appearance here.

a. Méssengérs he sent | þórghout Ínglónd

b. Untó the Ínglis kýnges | þat hád it ín þer hónd.

p. 2, ll. 3–4.

c. Áfter Éthelbért | com Élfríth his bróther,

d. Þát was Égbrihtes sónne, | and ȝít þer wás an óþer;

p. 21, ll. 7–8.

The Germanic licences incidental to even-beat rhythm are strikingly perceptible throughout.

In the first line we have to note in both hemistichs suppression of the anacrusis, in the second either the omission of an unaccented syllable or lengthening of a word (Ing(e)lond). The second line has a regular structure: in the third the suppression of the anacrusis is to be noted and the absence of an unaccented syllable in the second hemistich. The last line has the regular number of syllables, but double inversion of accent in the first hemistich. A disyllabic thesis at the beginning or in the middle of the line also frequently occurs.

To purvéie þám a skúlking, | on the Énglish éft to ríde;

p. 3, l. 8.

Bot soiórned þám a whíle | in rést a Bángóre;

p. 3, l. 16.

In Wéstsex was þán a kýng, | his náme wás Sir Íne;

p. 2, l. 1.

There is less freedom of structure in the Alexandrine as used in the lyrical poems of this period, in which, however, the verse is generally resolved by middle rhyme into short lines, as may be seen from the examples in § 150.

§ 148. The structure of the Alexandrine is, on the other hand, extremely irregular in the late Middle English Mysteries and the Early English Moral Plays, where, so far as we have observed, it is not employed in any piece as the exclusive metre, but mostly occurs either as the first member of the above-mentioned Poulter’s Measure, and occasionally in uninterrupted sequence in speeches of considerable length. We cannot therefore always say with certainty whether we have in many passages of Jacob and Esau (Dodsley’s Old Plays, ed. Hazlitt, vol. ii, pp. 185 ff.) to deal with four-beat lines or with unpolished Alexandrines (cf. Act II, Sc. i). In other pieces, on the other hand, the Alexandrine, where it appears in passages of some length, is pretty regularly constructed, as, for instance, in Redford’s Marriage of Wit and Science (Dodsley, ii, pp. 325 ff.), e.g. in Act II. Sc. ii (pp. 340–1):

The caesura and close of the line are in this passage, which comprises eighteen lines, monosyllabic throughout.

§ 149. In Modern English the Alexandrine is also found in a long-lined rhyming form, as, for instance, in the sixteenth century in certain poems by Sidney, but notably in Drayton’s Polyolbion.

The Modern English Alexandrine is particularly distinguished from the Middle English variety by the fact that the four types of the Middle English Alexandrine are reduced to one, the caesura being regularly masculine and the close of the line nearly always so; further by the very scanty employment of the Teutonic rhythmical licences; cf. the opening lines of the Polyolbion (Poets, iii. pp. 239 ff.):

Minor caesuras seldom occur, and generally in the second hemistich, as, e.g., minor lyric caesuras after the first foot:

Wise génius, | bý thy hélp || that só I máy descrý.

240 a;

or masculine caesura after the second foot:

Ye sácred bárds | that to || your hárps’ melódious stríngs.

ib.

Enjambement is only sporadically met with; breaking of the rhyme still more seldom.

Less significance is to be attached to the fact that Brysket, in a poem on Sidney’s death, entitled The Mourning Muse of Thestylis (printed with Spenser’s works, Globe edition, p. 563), makes Alexandrines rhyme together, not in couplets, but in an arbitrary order; further, that Surrey and Blennerhasset occasionally composed in similarly constructed rhymeless Alexandrines (cf. Metrik, ii, p. 83).

Of greater importance is the structure of the Alexandrine when used as the concluding line of the Spenserian stanza and of its imitations.

It is here noteworthy that the lyric caesura, unusual in Middle English, often occurs in Spenser after the first hemistich:

That súch a cúrsed créature || líves so lóng a spáce.

F. Q. I. i. 31;

as well as in connexion with minor caesuras:

Upón his fóe, | a Drágon, || hórriblé and stéarne. ib. I. i. 3.

The closing line of the Spenserian stanza is similarly handled by other poets, such as Thomson, Scott, Wordsworth, while poets like Pope, Byron, Shelley, and others admit only masculine caesuras after the third foot. By itself the Alexandrine has not often been employed in Modern English.

Connected in couplets it occurs in the nineteenth century in Wordsworth’s verse, e.g. in The Pet Lamb (ii. 149), and is in this use as well as in the Spenserian stanza treated by this poet with greater freedom than by others, two opening and medial disyllabic theses as well as suppression of anacrusis, being frequently admitted, while on the other hand the caesura and close of the verse are always monosyllabic.

§ 150. The three-foot line has its origin theoretically, and as a rule also actually, in a halving of the Alexandrine, and this is effected less frequently by the use of leonine than by cross rhyme.

Two Alexandrine long lines are, for instance, frequently resolved in this metrical type into four three-foot short lines with crossed rhymes, as, e.g., in Robert Mannyng’s Chronicle, from p. 69 of Hearne’s edition onwards.

From our previous description of the four types of the Middle English Alexandrine, determined by the caesura and the close of the verse, it is clear that the short verses resulting from them may rhyme either with masculine or feminine endings, as, e.g., on p. 78, ll. 1, 2:

In accordance with the general character of the metre the verses in this Chronicle are, even when rhyming as short lines, printed as long lines, especially as this order of rhymes is not consistently observed in all places in which they occur.

In lyrical poetry this metre is naturally chiefly found arranged in short lines, as in the following examples:

With another order of rhymes these verses are also met with in tail-rhyme stanzas of different kinds, as, for instance, in Wright’s Spec. of L. P., p. 41:

As a rule, the verses in such lyrical compositions intended to be sung are more regularly constructed than in those of narrative poetry, where the usual Germanic metrical licences occur more frequently.

In Modern English the three-foot verse has remained a favourite, chiefly in lyrical poetry, and occurs there as well with monosyllabic as with disyllabic rhymes, which may either follow one another or be crossed, e.g.:

We seldom find three-foot verses with disyllabic rhymes throughout. There is, on the other hand, in lyrical poetry a predilection for stanzas in which disyllabic rhymes alternate with monosyllabic, as, for instance, in Sheffield, On the Loss of an only Son:

Rhythmical licences, such as suppression of the anacrusis, seldom occur in such short lines. The species of licence that is most frequent appears to be inversion of accent.