The Heroic Age

A Journal of Early Medieval Northwestern Europe

12 (May 2009)  |   Issue Editors: Larry Swain

Founded 1998   |   ISSN 1526-1867

Issue Navigation

Issue Homepage

Office for St. Cuthbert

| Metrical Foregrounding |

King Alfred's Scholarly Writings

Forum—Digitizing Numismatics

Forum—Regna et Regnum

Forum—Saints and Sacred Space

Electronic Medievalia

Continental Business


Philological Inquiries


Heavy Hypermetrical Foregrounding in the Old Saxon Heliand and Genesis Poems

Douglas Simms
Southern Illinois University Edwardsville

© 2009 by Douglas Simms. All rights reserved. This edition copyright © 2009 by The Heroic Age. All rights reserved.

Abstract:  The Old Saxon Heliand and Vatican Genesis and the Old English Genesis B contain a number of metrically aberrant verses known generally as heavy hypermetrical verses. This paper argues that these verses serve to highlight portions of the text for the audience.

1.1 Introduction

§1.  In comparison to the relatively wealthy poetic remains of Old English, the continental Germanic alliterative traditions have left us few examples.  Most significant among these is the Old Saxon Heliand of the early ninth century which, with 5,983 lines, is the longest single poem in an early Germanic alliterative verse (Cathey 2002, 20–22).  Beyond that, however, the only significant texts are the Vatican Genesis, at approximately 334 lines, and a partial Anglo-Saxon translation of the same Genesis poem, Genesis B.  Though some aspects of Old Saxon meter may indicate a "tradition in decline" (Russom 1998, 170; see also Lehmann 1956, 105–113), the Heliand poet and the poet of Vatican Genesis did not lack verbal, and as will be argued here, metrical artistry.  Of interest to us here is the use of extraordinarily long metrical constructions, with which the poets were able to catch the audiences' ears and redirect their attention to passages of particular import.  As the Heliand poet has affectively used the variation and placement of alliteration (Jeep 2002), fitt structure (Murphy 1992, 221–230; Haferland 2002, 243–45), as well as markers in other linguistic components (Rauch 2002), it is natural that poets who bestowed this much attention to rhetorical devices should likewise vary the metrical structure of the work.

§2.  Little work has been done in studies of Old Saxon meter to explain a handful of irregularities, known as 'heavy hypermetric' verses, a term to be defined below.1   Furthermore, it has escaped notice that a number of them appear at significant moments within the poem and have the effect of highlighting the passages metrically against the surrounding verse.  These anomalies can be characterized as foregrounding devices, namely poetic devices employed to bring attention to a certain verse within a given passage.

§3.  Whether one chooses to view Old Saxon verse as sub-par in comparison with what we know of Old English meter, or whether one seeks to argue that Old Saxon meter is merely organized differently than Old English meter (Suzuki 2004; 2001), one cannot help but see that the poets of these works were conscious enough of metrical variation to employ it as a stylistic device at key moments.  Although it is tempting to claim that a particular metrical variant possesses a certain function within poetry, one must exercise caution in doing so.  In this study I will first establish particular, formal, metrical criteria for the anomaly of interest.  After presentation and scansion of the verses adhering to the criteria within their contexts, these metrical variants will be sorted according to possible functional tendencies.  The strongest tendency we shall encounter is their placement at transitions between narration and direct speech.  This most apparent tendency will thus present sufficient reason to support the possibility that the poets of these works employed the metrical variants elsewhere to achieve particular effects, namely to emphasize certain verses and passages to the listening audience.

1.2 Formal Criteria

§4.  As with Old English and Old Norse alliterative verse, the Heliand and Genesis poems are composed of lines containing two verses, or half-lines.  The two verses are bound to one another by means of alliteration, located on the first stressed syllable of the second verse and on one or both of the two stressed syllables of the first verse.  The majority of verses contain four metrical positions.  Each position is filled either with a syllable carrying primary or secondary stress, or with one or more unstressed syllables.2   Verses with four, and on some occasions, five positions (Cable 1974, 80–82) are termed 'normal'.

§5.  There are, however, longer verses approximately half as large again as a normal verse.  These larger verses, whose structure has yet to be fully explained, are termed 'hypermetric' verses.  Below is an example of a typical hypermetric verse scanned with six clearly identifiable metrical positions in Heliand 1314a:3

mildi mahtig selƀo

To provide a short synopsis of the structure of Old Saxon hypermetrics, one may say that the six metrical positions may contain two types of filler: 'strong' and 'weak.'4   Strong positions (S) may be filled by:

Strong positions as a rule cannot hold unstressed syllables (x).  Strong positions are permitted to be adjacent within a verse.  However, once three strong positions are adjacent, one will be subordinated.

Weak positions (w) containing the dips of the alliterative verse, may:

Despite their ability to contain multiple syllables, weak positions are prohibited from being adjacent to one another, unless they fill the first two positions, in which case only one syllable is permitted in the first position (which will be represented here as W), with or without anacrusis.  The prohibition against adjacent weak positions is justified because the ability to contain multiple syllables could make the two indistinguishable from one weak position.5  Weak positions are, nonetheless, regulated.  The further to the right a weak position might be, for example, the fewer syllables it is permitted to hold.  In Old Saxon the right-most position may hold a maximum of two syllables, the antepenultimate may hold up to four, whereas the position second from the left may contain up to 13, visible in Heliand 605b.6

§6.  Pope, in his classification of Old English hypermetric verses, referred to those which began with an alliterating lift as 'strong' hypermetrics, whereas verses with a series of unstressed syllables prior to the first lift are 'weak' hypermetrics (Pope 1966, 126–27, 134–35).7  Below are examples of a 'strong' hypermetric and a 'weak' hypermetric with alliterating staves underlined.

§7.  Heliand 1314b shown above is a typical 'strong' hypermetric:

mildi mahtig selƀo

whereas Heliand 2213b is a typical 'weak' hypermetric:

quâðun that uualdand selƀo

§8.  In order to get an idea of the usage of Old Saxon hypermetrics, we might look to the hypermetric verses found in the Old English tradition.  The larger sample and corpus sizes might better serve as a point of departure for a sense of the usage of hypermetric verses in general, which we might then apply to Old Saxon verse as well.  Timmer points out that Old English hypermetric verses are used to mark the beginnings and ends of passages and of poems, and to mark a general 'elevated' style (1952, 229 et passim).  The insertion of hypermetric verse into normal verse often takes the form of gradual, bell-shaped increases in verse length, or in clusters (Bartlett 1935, 63–64).8  In a diachronic sense, the 'elevated' style of hypermetrics might be, as suggested in Suzuki (1991), a throwback to general Indo-European metrical practices of using shorter verses for lyric and longer verse for epic and gnomic verse.  Synchronically, however, one might also say that the stylistic usage of hypermetric lines is less dependent on tradition than it is on their usage.  The disruption of the default, unmarked meter with an alternate, or marked, variant creates a change which can then be taken advantage of rhetorically.  Yet, it is perhaps no mere coincidence that in Genesis A, as Fulk notes, "God tends to speak in hypermetrics" (2001, 151).

1.3 Heavy Hypermetric Verses

§9.  The metrical anomaly of particular interest to us is found in thirteen possible instances within the Heliand (verses 1144a, 3062a, 3990a, 5690a, 5916a, 5920a, and 5975a),9 the Vatican Genesis (verses 71a and 235a), and Old English Genesis B (verses 260a, 356a, 403a, and 507a).  Whereas the vast majority of hypermetric verses contain six metrical positions, these anomalous verses contain eight, though they adhere to the same criteria set forth here for six-positioned hypermetrics.  Following Bliss's terminology (1958, 95–96), these eight-positioned verses are known as "heavy" or "double" hypermetric verses.10  The term 'heavy hypermetric' will be used consistently in this paper.  Vatican Genesis 71a exemplifies the octopositional verse:

libbian an thesun landæ lango huila

Because heavy hypermetrics are identical to regular hypermetrics, differing only in the number of positions, there are weak heavy hypermetrics in addition to strong ones. An esample is the last heavy hypermetric in the Heliand 5975a sôhta imu that hôha himilo rîki, where the introductory syllables preceding the first alliterating lift exceed the three-syllable limit for anacrusis.11

§10.  Problematic for the heavy hypermetric is that as octopositional verses they are identical in shape to two normal verses of four metrical positions each, accompanied by either a hypermetric or normal off-verse.12  Though we must begin by establishing metrical criteria for the selection of verses, the goal here is not the argument for one model of Old Saxon verse over another, but rather an attempt to approach an explanation for the distribution of these anomalies.  As we shall see in the examination of these verses, the heavy hypermetric verses are employed as rhetorical markers which emphasize the beginning or end of a speech within the poem, as well as a general means of foregrounding for effect.  For ease of reading, the relevant verses will be scanned only with the stress values of the syllables indicated, with positions separated by vertical bars.

2.1.1 Genesis B

§11.  Genesis B, that part of the Junius codex identified by Eduard Sievers solely by the meter as a translation of an Old Saxon original (Doane 1991, ix), contains perhaps three passages able to be scanned as heavy hypermetric verses.  I have chosen to begin the discussion of heavy hypermetrics in Old Saxon with these, technically speaking, Old English verses because they border the two languages.  They are clearly written in Old English, yet at the same time we know that they are a translation.  Unfortunately, one is not able to simply translate these words back into Old Saxon, because the translator/scribe was not shy about making metrical and phraseological alterations, and in some cases made mistakes in reading the Old Saxon original (Doane 1991, 56).  It is thus problematic to assume that the metrical patterns depicted here are truly representative of their Old Saxon originals, because the surviving text of the Vatican Genesis cannot provide parallel text.  Nonetheless, these two verses belong to both poetic traditions, to the Old English because of its acceptance through translation, and to the Old Saxon because of the extant original Vatican Genesis.

2.1.2 Genesis B 356a

§12.  Satan is the subject of the first heavy hypermetric in Genesis B, verse 356a, cited in the context of ll. 353b–357, as he begins his description of Hell:13

Weoll him on innan
hyge ymb his heortan,         hat wæs him utan
wraðlic wite.         He þa worde cwæð:
'Is þæs ænga styde ungelic swiðe         þam oðrum þe we ær cuðon,
hean on heofonrice,         þe me min hearra onlag,'

[His mind welled up within, all about his heart, the loathsome punishment was hot to him on the outside.  He spoke then with a word: 'This troublesome place is very dissimilar compared to that other which we knew before, the high one in the heavenly kingdom, which my Lord granted me,']

If we treat 356 as a single heavy hypermetric line, we are faced with a switch from normal to a heavy hypermetric (as versified below into eight positions), which marks and emphasizes the beginning of Satan's speech:

x    x |  S | x |  sx |  S | x x |  S |x
Is þæs ænge styde  ungelic swiðe

This passage marks Satan's incarceration within the abyss.  In addition to marking the shift of voice from the narrator's to Satan's, the poet has managed to use the meter to underscore the fall into the abyss within the first verse of the speech with the wry, drawn-out comment that "this place is very different."  The syntax, likewise, does not miss out on adding to the foregrounding, because the normal prose order of ungelic swiðe is the exact opposite, i.e. swiðe ungelic, reversed for the sake of alliteration.14

2.1.3 Genesis B 403a

§13.  One can observe similar behavior in the other heavy hypermetric verse in Genesis B, 403a, located in a lengthy hypermetric cluster beginning at 388b and continuing to 408b.  Below are lines 402b–405:

Ne magon we þæt on aldre gewinnan,
þæt we mihtiges godes mod onwæcen.         Uton oðwendan hit nu monnum bearnum,
þæt heofonrice, nu we hit habban ne moton,         gedon þæt hie his hyldo forlæten,
þæt hie þæt onwendon þæt he mid his worde bebead. Þonne weorð he him wrað on mode.

[We are unable to fight in life such that we weaken the mind of mighty God. Let us turn it away from the sons of men, that heavenly kingdom, we cannot have it now, (let us) work such that they might leave his favor that they overturn that which he commanded with his word.  He will then become angry with them in his mind.]

Left unemended by Krapp and treated as a triplet by Doane, verse 403a is best scanned, in my opinion, as a heavy hypermetric on-verse:

  x    x | S | x x |    sx  |  S  | x |  S | x
þæt we mihtiges godes mod onwæcen

With its greater metrical weight and complexity, the poet foregrounds this verse against the already lengthy hypermetric verses in the cluster and uses it to highlight Satan's inability to struggle in any meaningful way against the might of God.  Realizing his inefficacy against the divine power, Satan exhorts his fellow fallen angels to concentrate their efforts against humanity, a comparatively easier target, in order to pit them against God.  In terms of function, this passage demonstrates the use of the heavy hypermetric verse to highlight points within hypermetric clusters.  If six-positioned hypermetric clusters are employed for foregrounding effects against normal verses, a verse would need more than six positions to stand out against the already expanded lines.

2.1.4 Genesis B 507a

§14.  Adam and Satan are the participants in the portion of Genesis B where the final heavy hypermetric is found.  In his effort to tempt Adam to eat of the forbidden fruit, Satan cloaked as the serpent approaches Adam with ligenum 'lies'.  Somewhat reminiscent of the tendency of hypermetric verses pointed out by Bartlett (1935, 63–64) to form bell-shaped curves, the serpent's temptation of lies, ll. 496b–521, is punctuated roughly in the middle (11 lines from the opening of the speech and 14 lines from the end) by the heavy hypermetric in verse 507a, ll. 504b–508:

'Nu þu willan hæfst,
hyldo geworhte         heofoncyninges,
to þance geþenod         þinum hearan,
hæfst þe wið drihten dyrne geworhtne.         Ic hyrde hine þine dæd ond word
lofian on his leohte         and ymb þin lif sprecan…'

['Now you have a choice, obtained the favor of the King of Heaven, served your Lord to satisfaction, have made yourself dear to the Lord. I heard him praise your deeds and words in his light and speak about your life…']

Located within a distinct portion of the temptation we find a weak heavy hypermetric:

  x  |    x  x  |  S | x |  S | x    |  S  |  x
hæfst þe wið drihten dyrne geworhtne

One should note that the three introductory syllables in this verse can very well be seen as an anacrusis to the verse proper.  Here the fact that we have a translation here is worth considering, as the Old Saxon equivalent to West Saxon hæfst would be the disyllabic haƀes, adding an additional syllable beyond the trisyllabic limit of anacrusis.

§15.  Although I have, for the sake of convenience, chosen to translate 507a using the modern English present perfect, it in no way reflects the same Old English verbal category.  One may see the presential nature of the verb in the opening of the passage.  Within the temptation of falsehoods stands a bit of truth.  Up until this point, indeed up to the point where Eve convinces him to eat of the fruit, Adam has obeyed God and does stand within his favor.  Immediately following the speck of truth within the misleading verse 507a, the serpent begins again at verse 507b to deceive Adam.  The truthful passage, culminating in the heavy hypermetric, forces the audience through progressive referentiality to compare the obedient Adam with the fallen Adam who is to come.

2.2 Vatican Genesis

§16.  Turning now to a copy of Genesis B's forerunner, the Vatican Genesis, we find two verses whose metrical composition marks them as heavy hypermetrics: 71a and 235a.  The first among these is found in the retelling of Cain's punishment for having slain Abel.  As the audience might expect, Cain remarks that as he walks his path of exile whatsoever he encounters will be in a position to slay him (ll. 67b–69a).  God replies, his direct speech beginning in typical Hakenstil in 70b, that it will not be so:

thuo sprak im eft selƀo angegin
Heƀanes uualdand:         'Hier scalt thu noh nu,' quad he,
'libbian an thesun landæ lango huila.         Tho thu sus aleđit sis
mid firinum bifangan,         thoh uuillik thi frithu settean,
togean sulic tekean         so thu an treuuua maht
uuesan an thesero uuerolde,         thoh thu is uuirdic ni sis:'

[Then spoke back to him in return Heaven's Ruler Himself: 'Here you must still,' he said, 'live on this land for a long time.  Though you might have become hated so, wrapped up with sins, nonetheless I wish to establish a surety for you, draw (?) such a mark, so that you can, in peace, remain in this world, though you might not be worthy of it:']

Cain's curse to walk the earth until the end is projected out from verses of normal metrical length, in addition to marking the transition from indirect to direct speech in the first on-verse of the quote.  The importance of this transition cannot be understated.  Instances where the introduction of direct speech correlates to a switch to a heavy hypermetric provide us with the clearest examples of a rhetorical usage of metrical variants.

§17.  Though Doane scans Vatican Genesis 228a as a verse which might fall under the scope of heavy hypermetrical verses,17 the only other heavy hypermetric to be found in this work closes the series of three speeches in which Abraham beseeches God to spare the city of Sodom, two lines prior to the end of the quote, ll.233b–38:

'…huuat uuilis thu is thana, fro min, duoan
ef thu thar tehani         treuhafte maht
fiðan under themo folca ferahtera manno,         uuilthu im thanna hiro ferh fargeƀan
that sia umbi sodomaland         sittian muotin,
buan an them burugium         so thu im abolgan ni sis?'

['…what will you do, my Lord, then, if ten faithful ones you there might find among that people, pious men, will you restore their lives to them then, such that they could dwell around the land of Sodom?']

Verse 235a is metrically simple and possesses proper alliteration, having the equivalent shape of two Type-A verses, making it easier to identify as a heavy hypermetric:

S |x  x    x      x  x |  S |x | sx |  x x |  S |  x
fiðan under themo folca ferahtera manno

The affective prominence of the meter here is evident, if we view the verse as an example of Behaghel's Gesetz der wachsenden Glieder (1932, 6) as well as Olrik's episch[es] gesetz der dreizahl (1909, 3–5), in that the longest and/or most significant member of a series is positioned last.  One can see Abraham's entreaties to God concerning Sodom build from l. 203:

Ef thu thar fiđis fiftig         ferahtaro manno

[If you find there fifty pious men]

to l. 214:

Ef thu thar thritig maht         thegno fiđan

[If you can find there thirty thanes]

culminating in the third, l.234–35:

ef thu thar tehani         treuhafte maht
fiðan under themo folca ferahtera manno,         uuilthu im thanna hiro ferh fargeƀan

[if ten faithful ones you there might find among that people, pious men, will you restore their lives to them then]

In the third instance the ef-clause spans two lines, whereas the first two contain the clause within the line.  As the third time is the charm, so to speak, it is appropriate that the verse coincides with a modulation in the meter.  Even though one might expect the modulation to pair with the numeral tehani, the heavy hypermetric draws attention to the echoed ferahtaro and fiđan as well as to the question following in the off-verse.

3.1 The Heliand

§18.  According to the criteria set forth above, the Heliand is home to seven heavy hypermetric verses.19  Of the two main manuscripts of the Heliand (the Cottonian and Munich MSS), only two verses are attested in both (1444a and 3062a) due to the more fragmentary condition of the Munich MS.  No heavy hypermetrics are present in the fragments from Prague, Straubing, the Vatican, or from the newly-discovered fragment from Leipzig (see Schmid 2006).

3.2 Heliand 1144a

§19.  First among the Heliand's heavy hypermetric verses is 1144a, cited within its context of ll.1141–45:

'nu is it all gefullot sô,
sô hîr alde man         êr huuanna sprâcun,
gehêtun eu te helpu         heƀenrîki:
nû is it giu ginâhid thurh thes neriandan craft:         thes môtun gi neotan forð
sô huue sô gerno uuili         gode theonogean,'

[It is now thus entirely fulfilled, just as old men here long ago at one time spoke, they promised to you as a help the heavenly kingdom: Now it has drawn nigh to you through the strength of the Savior: this ye can enjoy henceforth, whosoever wants eagerly to serve God,]

This exhortation to begin proselytizing and the annunciation of the approach of the heavenly kingdom, introduced first with normal verses, is punctuated quite abruptly with the heavy hypermetric construction, its stress on the initial 'now,' connecting alliteratively with genâhid 'drawn near' and neriandan 'of the Savior:'

  S|x    x    x  x| S |x    x      x    |S | s |  x  |    S20
nû is  it giu ginâhid thurh thes neriandan craft

The particle , which participates in the alliteration, is at the forefront of this verse and highlights the immediacy of Christ's mission on earth to the audience, whom the poet is, in part, attempting to convert with the poem (Cathey 2002, 12–15; Murphy 1992, vii–viii).  In addition to the topicalization of the adverb, the unusual heavy hypermetric punctuates the verse and assists in contrasting this statement with the parallel construction in 1141b. Verses 1141b and 1144a stand in variation to one another with 1144a expressing the means by which the prophecy of old has been fulfilled in the present.

3.3 Heliand 3062a

§20.  The next passage, cited by Sievers as an example of a heavy hypermetric verse in his Altgermanische Metrik (1893, 164), is verse 3062a, which precedes the bestowal of the keys of Heaven to Peter, ll. 3061b–3064:

Thô sprac imu eft is hero angegin:
'sâlig bist thu Sîmon', quað he, 'sunu Ionases;         ni mahtes thu that selƀo gehuggean,
gimarcon an thînun môdgithâhtiun,         ne it ni mahte thi mannes tunge
uuorden geuuîsien,         ac dede it thi uualdand selƀo,

[His Lord then spoke back to him in return: 'Blessed art thou Simon,' he said, 'son of Jonas you are unable to think that yourself, mark it in your innermost thoughts, nor can the tongue of man guide you with words, rather the Ruler did it to you Himself,]

The three alliterating substantivals of 3062a provide clear identification of the first three lifts in this heavy hypermetric:

S| x    x      x | S | x |  sx |  S| s| x
sâlig bist thu Sîmon sunu Ionases21

§21.  One should not be surprised to see the Heliand poet highlighting the role of Simon Peter.  Because the Carolingian emperors had obtained their primacy, to a certain extent, through papal coronation, an emphasis on the sanctity of Peter would serve to support the rule of Christian Franks over the Saxons.  We find here an example of Murphy's note that:

This is an era when popes consistently referred to themselves as the vicars of Peter and were seen in iconography to be on the throne of Peter.  Of course there is no Vatican hierarchy in the Heliand, but "Peter" is everywhere and, as we will see, is described both in the scene of the walking on the water and in the swordplay at the arrest of Jesus as being especially close to Christ. (Murphy 1989, 4)

In the case of 3620a, the Heliand-poet chose to mark Peter's special place among the apostles with metrical texturing to project Christ's blessing of Peter out to the audience (cf. Cathey 2002, 203–4).

3.4 Heliand 3990a

§22.  Verse 3990a, located within a paraphrase of John 11:8, makes use of a heavy hypermetric verse to highlight the danger facing Christ as he goes to raise Lazarus.  In this instance, the heavy hypermetric construction adds to the sense of the apostles' anxiety which, in turn, serves to present Christ as a man fulfilling his duty despite danger to his person, a sentiment that must have greatly pleased an audience accustomed to heroic poetry, ll. 3988–3992:

'…Ni that nu furn ni uuas
that sia thik thinero uuordo         uuîtnon hogdun,
uueldun thi mid stênon starcan auuerpan?:         nu thu eft undar thia strîdigun thioda
fundos te faranne,         thar ist fîondo ginuog,
erlos oƀarmuoda?'

['Now, was it not long ago that they intended to kill you on account of your words (that) they wanted to throw you down with strong stones?: now you set out under that contentious people to travel back, there where there are enough enemies, arrogant earls?']

Under some circumstances one might wish to read the initial uueldun 'they wanted' as an unstressed anacrusis to the verse proper.  However, the prohibition against more than three syllables in anacrusis combined with the alliterative pattern of the on-verse, where the required alliteration on st- (stênon ~ starcan) is couched between another, secondary alliterative pair in w- (uueldun ~ auuerpan), indicates that the first syllable of uueldun could be filling position I as an initial weak, WwSwSwSw:

    x | x    x    x |    S |x |  S |  x  x | S |  x 
uueldun thi mid stênon starcan auuerpan

The apostles' concern for Christ's well-being, underscored here by the heavy hypermetric, sets the scene for Thomas, much like Wiglaf in Beowulf ll. 2631–2660, to proclaim to the retinue the duty of a thane to his lord in a time of need (Heliand ll.3992b–4002a).  A departure from the typical metrical scheme in 3990a is followed immediately by the Heliand poet's departure from Tatian in the embellishment of a biblical scene with characteristics of traditional Germanic heroic poetry (Haferland 2002, 246; Murphy 1989, 99–100).

3.5 Heliand 5690a

§23.  The next verse to be examined is 5690a, found in the crucifixion scene, ll. 5689–5692a:

Than uuas sido Iudeono
that sia thia haftun man thuru thena hêlagan dag         hangon ni lietin
lengerun huîla         than im that lîf scrîði
thiu seola besunki:

[The custom of the Jews was then that they did not permit captive men to be hanged during the holy day for a longer time than that life might pass from them, (that) the soul might have sunk away:]

Although most hypermetric verses occur in clusters, this verse is similar to 1144a in that it appears alone among several normal verses.  This is, however, more common in the Heliand than in Old English poetry (Hofmann 1991, I, 153).  The effect served here could be to highlight the contrast between the sinful captive men, thia haftun man, and the Sabbath day, thena hêlagan dag:

  x    x    x | S | x |    S  |  x x    x x | S | x x |  S
that sia thia haftun man thuru thena hêlagan dag

3.6 Heliand 5916a and 5920a

§24.  Two verses, 5916a and 5920a, are to be found in a relatively lengthy hypermetric cluster describing Mary Magdalene's vigil at the empty tomb and her encounter with the risen Christ, ll. 5915–17a:

Maria uuas that Magdalena:         uuas iro muodgithâht,
seƀo mid sorogon sêro giblandan:         ne uuissa huarod siu sôkian scolda
thena hêrron, thar iro uuârun at thia helpa gilanga.

[That was Mary Magdalene: her innermost thoughts, her mind, were sorely mixed with sorrows: she knew not whither she should search for the Lord there, where (He would be) truly ready to help her]

Although Sievers (1878) marks his disapproval with the structure of the line by placing parentheses around sêro 'sorely, very', we can scan the manuscript reading as an eight-positioned hypermetric verse:

  sx  |  x  |  sx | x |  S| x  x |  S | x
seƀo mid sorogon  sêro  giblandan

§25.  A mere four lines after the previous passage, Christ is reintroduced to Mary Magdalene, in what is perhaps the most striking use of a heavy hypermetric verse in the Heliand.  It should come as no surprise that Christ's return after the Resurrection is thrown into relief against the surrounding verses with a heavy hypermetric, ll. 5919b–5921a:

Thuo gisah siu thena mahtigan thar standan,
Christe, thuoh siu ina cûðlîco ankennian ni mohti,         êr than hie ina cûðian uuelda,
seggian that hie it selƀo uuâri.

[She then saw the Mighty One standing there, Christ, though she could not clearly recognize him, before he wished to announce himself, to say that it was He Himself.]

The structure of this verse, indicated by alliteration, is a strong heavy hypermetric:

      S | x    x      x x x | S | x x x |  S |x x  x |  S | x
Christe, thuoh siu ina cûðlîco ankennian ni mohti

In addition to the metrical foregrounding, the verse is marked by a threefold figura etymologica, cûðlico ~ ankennian ~ cûðian, within the line as a whole.  The etymological figure raises the tension and irony of the scene by repeating the root for 'know' while simultaneously expressing Mary Magdalene's inability to recognize Christ.

3.7 Heliand 5975a

§26.  The final heavy hypermetric verse to appear in the Heliand as it survives is 5975a, relating Christ's ascension to Heaven after the Resurrection.  Having led the apostles to Bethany, Christ blessed them all and (ll.5974b–5978):

Giuuêt imo up thanan
sôhta imo that hôha himilo rîki         endi thena is hêlagon stôl:
sitit imo thar         an thea suîðron half godes,
alomahtiges fader,         endi thanan all gisihit
uualdandeo Crist,         sô huat sô thius uuerold behaƀet.

[He departed thence up above, sought for Himself the high kingdom of the heavens and that holy throne of His: He Himself sits there at the right hand of God, of the Father almighty, and the Ruling Christ from there sees all whatsoever this world contains.]

Verse 5975a stands as a lone heavy hypermetric amidst normal verses.  As with 3990a and 5690a, this verse is a weak heavy hypermetric:

  x | x x  x    x |  S |x | sx | x |S|x
sôhta imo that hôha himilo rîki

Though Christ's departure is overtly stated in the preceding off-verse, again in typical Hakenstil, a variation on the same theme is presented in the first on-verse of the Ascension, overshadowing the surrounding normal verses.  To be sure, there is no way to know whether this is the absolute final occurrence of this metrical anomaly in the Heliand, as the final portion of the text is lost to us.  It seems, nonetheless, uncanny that this last heavy hypermetric should co-occur with Christ's Ascension to Heaven.  In a small, though by no means insignificant way, the poet altered the passive ferebatur in caelum (Tatian Diatessaron 244, 2) in order to present an active Christ, beginning the verse sôhta imo (bringing with it a verse-internal rhyme sôhta ~ hôha) and presenting a Christ rising to heaven more reminiscent of the phraseology of the Apostles' and Nicene Creeds than of Luke 24:51.

4 Summary

§27.  Although the amount of material to work with is too sparse for meaningful statistical analysis, the distribution of these metrically expanded lines correlate in the Heliand and Genesis poems with at least one distinct function.  In four instances, heavy hypermetric verses occur within close proximity of the beginning or end of direct speech, three being found in the first verse of direct speech or immediately adjacent to it.22   In this fashion, heavy hypermetric verses serve as demarcators of direct speech, indicating to the audience that a shift in voice has occurred, as is the case with the beginning of Satan's speech in Genesis B, at the beginning of Christ's blessing of Simon Peter in the Heliand, and in the Vatican Genesis at God's placing of the mark upon Cain's forehead.

§28.  Although the association with direct speech and shifting voices is the less common tendency observable in heavy hypermetric verses, it is the most demonstrable.  This clearer usage of heavy hypermetric verses justifies, in part, the positing of a second usage, that as a pure foregrounding device meant to highlight a 'significant' passage.  Without the connection to marking direct speech we could encounter a circular situation, in which one might claim that the metrical variant indicates a marked passage, and then read the verse in such a manner so as to justify its special status.  Because we have already established a possible relationship between the eight-positioned verse variant and rhetorically significant placement of those verses, we are in a better position to view other, similar metrical constructions as being marked in some respect.  The remaining examples of heavy hypermetric verses in Old Saxon poetry may be construed as offering metrical emphasis to verses containing significant content.23   One may note that the two functions are at the core identical and not mutually exclusive, because Heliand 3062a may fit both categories equally well.

§29.  In the case of the Heliand, we find one instance corresponding to a statement that Heaven is nigh and brought closer by Christ (1144a), one corresponding to Christ's blessing of Simon Peter (3062a), one corresponding to Christ's return to Jerusalem, leading up to his healing Lazarus (3990a), one corresponding to the contrast of executing captive men and the proscriptions of the holy Sabbath (5690a).  Two occurrences of heavy hypermetric verses in the Heliand correspond to Christ's Resurrection.  The first highlights Mary Magdalene's sorrow as she comes upon the Holy Sepulcher.  Christ's disappearance, however, is paralleled by His reintroduction to Mary Magdalene, and the first reintroduction of Christ into the Heliand after the Resurrection.  In verse 5975a, the last heavy hypermetric to occur in the poem as a whole, we find Christ's ascension to Heaven.  Though they are few in number, these metrical anomalies have the semblance of something other than mistakes or metrical hiccoughs, because they surface at points within the text where the addition of emphasis to the passage is arguably present.  Treating these verses as instances where the poet has departed from the expected metrical pattern in an organized manner enables us to add a sense of regularity to the meter and give a better understanding of the rhetorical and affective uses toward which the poets of these Old Saxon works employed metrical variants.


1.   Bliss (1971) addressed the identical phenomenon in Old English verse, treating them as tripartite metrical structures analogous to the Old Norse ljóðahá ttr, a view adopted by Doane and applied to Old Saxon meter (1991).  The comparison, however, is problematic.  Even if one were to ignore the difficulties of comparing the first two verses of the ljóðahá ttr with one W. Germanic long-line, the characteristics of the ljóðaháttr lie in the third Vollzeile, which contains two independently alliterating syllables, and tends to end in one heavy, stressed syllable, or its resolved equivalent (Árnason 1991, 52–56).  Moreover, as will become apparent from the verses presented here, a triplet interpretation of these passages would place the third portion, as the point of departure from the metrical norm, in a position of prominence, when, if there is any foregrounding, it seems confined to the initial portion of these passages.  Russom's word-foot model views these verses as aberrant according to his Universal Overlap Constraint, which states "Avoid feet that resemble verses and verses that resemble feet" (Russom 1998, 219).  Because these verses contain feet resembling verses they run counter to the constraint.  This violation of the UOC is, however, not a problem for the aims of this paper, rather it would seem to support the view that these verses are highly marked. Of the recent works on Old Saxon meter, Suzuki (2004) and Russom (1998) have not dealt with the heavy hypermetric. Hofmann (1991) discusses the phenomenon, though his formal criteria suffer from the same difficulties as Bliss' upon which his work is based.  [Back]

2.   There is, of course, no room here to present a full model of early Germanic alliterative verse, but a synopsis of Old English verse can be found in Fulk (2001, 129–158) and characteristics of Old Saxon verse in Russom (1998, in particular 136–170) and a full account of Old Saxon meter in Hofmann (1991) and, more recently, Suzuki (2004).  [Back]

3.   One should consult Sievers (1893), Bliss (1958), Pope (1966), Russom (1987, 1998), Hofmann (1991), and Suzuki (2004) for varying views of the hypermetric verse.  The metrical system presented here is a hypothetical model, which requires further inquiry (particularly into the nature of anacrusis and double-drops) to be fully accurate, but suffices for the needs of this paper.  [Back]

4.   This model of hypermetrical verses bears some similarity to the model of the Old Norse dróttkvætt posited by Kristján Árnason (1991), in that I treat the underlying structure of the hypermetric verse as the rhythmical arrangement of six strong and weak positions.  Though Kristján Árnason's treatment of the dróttkvætt has been superceded by others, e.g. Gade (1995), the analogy is helpful. Furthermore it seems possible that the dróttkvætt verse is the N. Germanic cogener of the W. Germanic hypermetric, the two being separate developments of a once common poetic tradition (Simms 2003).  [Back]

5.   One might conjecture that verses which have an initial augmented weak syllable (W) are permitted to do so, because the syllable gains relative prominence after the verse-initial boundary and before a syllable with relatively weaker stress.  [Back]

6.   saga ûs undar huilicumu he sî thesaro cunneo afôdit  [Back]

7.   These verses are the hypermetric analogues to the normal type A3, or "light" type-A, e.g. Heliand 34a that sea fan Cristes, x | x x | S | x, where the verse might be represented in terms of position strength as WwSw.  The filler for position I in weak hypermetrics is, in the main, constrained to conjunctions, finite verbs, and pronouns (including deictic, interrogative, and reflexive pronouns), prepositions not adjacent to nouns or adjectives, and some adverbial particles, e.g. .  Because there would be an obvious clash of terminology to speak of 'light heavy hypermetrics' they will be referred to as 'weak heavy hypermetrics.'  Pope's dichotomy of hypermetrics fails, unfortunately, to account for verses like Heliand 621a that he scoldi on Bethleem giboran uuerðan, in which five unstressed syllables precede the first alliterating lift, though the verse contains three lifts (-boran is a resolved sequence).  According to the system described here, Heliand 621a would be depicted at an abstract level as wSwSSw, thus actually strong, because it possesses three strong positions, and no augmented-weak initial.  [Back]

8.   See also Hieatt (1980) for an analysis of the envelope-patterning function of hypermetric lines in the Old English Judith.  [Back]

9.   Because they fall outside of the criteria established here, and are, most likely, regular hypermetrics, verses 604a, 621a, 1687a, 1730a, 3344a, 4374a are excluded from the study, though scanned by Hofmann (1991) as heavy hypermetrics.  [Back]

10.   Hofmann calls them vierkolige Schwellverse 'hypermetric verses with four cola' (1991, I, 169–170).  [Back]

11.   Depending on one's view of anacrusis in Old Saxon, one might treat this verse as a hypermetric with five syllables in anacrusis.  However, it seems more prudent to accept Russom's constraint on Old Saxon anacrusis to three syllables (1998, 147–58).  In line with the criteria established here, one might treat 1994b as a regular weak hypermetric with the underlying structure of WwSwSw.  [Back]

12.   One could differentiate between a pair of normal verses and one heavy hypermetric in certain circumstances.  For example, a weak heavy hypermetric should be substantially different from a pair of normal verses, the first of which possesses a Type B rhythm.  The second dip of a Type B verse in Old Saxon, in general, does not contain more than two syllables, and those instances of three or more syllables in the dip are exceptional (Russom 1998, 139–144).  Suzuki in his treatment of hypermetric verses in the Heliand notes that hypermetrical verses beginning with a Type B rhythm do not adhere to the usual constraints on the number of syllables in the second dip (2004, 300–301).  This might suggest that the constraint on the second dip of the Type B verse is a constraint on the penultimate metrical position, rather than on Type B rhythms.  Thus, the treatment of a hypermetric verse as being equivalent to two overlapping verses does not capture the structure of all hypermetric verses.  If the heavy hypermetric is a metrical unit, rather than a pair of normal verses, then one should not expect any such constraint on the number of syllables in the second dip in verses beginning with a Type B rhythm.  Unfortunately no extant verses provide any positive data on the absence of the constraint.  In the end, though, even if the underlying structure of these passages is in actuality two normal verses paired with an off-verse, as Doane (1991) treats them, the effect is the same.  [Back]

13.   Due to the at first ambiguous structure of heavy hypermetric verses, Krapp emended this passage in the ASPR as follows, ll. 353b–358:

Weoll him on innan
hyge ymb his heortan,         hat wæs him utan
wraðlic wite.         He þa worde cwæð:
Is þæs ænga styde         ungelic swiðe
þam oðrum [ham]         þe we ær cuðon,
hean on heofonrice,         þe me min hearra onlag,

One should note that Krapp attempts to make sense of a metrically difficult passage by inserting the word hām 'home' after ōðrum and treats the heavy hypermetric verse in question as two verses within one line.  [Back]

14.   Lehmann notes the preference for non-poetic syntax in the poem, mentioning the "Heliand poet's use of colloquial rather than traditional patterns" (1956, 12).  [Back]

15.   Again, Doane treats line 507 as a triplet:

hæfst þe wið drihten         dyrne geworhtne.
Ic hyrde hine þine dæd ond word

16.   Doane's arrangement of Vatican Genesis 71:

libbian an thesun landæ         lango huila.
Tho thu sus aleđit sis

17.   Although Vatican Genesis 228 is not a heavy hypermetric according to the metrical criteria set forth here, Doane's scansion is worth examination, because he treats the verse as a triplet:

uuslea uuider thi         mid minum uuordum.
ik uuet that ik thas uuirðig ni bium

Doane scans the first portion of this verse, uueslea uuider thi, as a type 2E1b, i.e. / x | x x /, despite the double alliteration and the lack of identifiable secondary stress on a preposition, as well as violating the prohibition against the placement of the head-stave on a final stress.  Hofmann scans the same verse as a normal type A verse, by his system type avIIv2, i.e.  / x x x x x x x x | / x, (1991, II, 222), with stresses on the first syllables of uueslea and uuordum.  Treating the verse as such provides a more acceptable scansion and is compatible with the non-hypermetrical environment.  [Back]

18.   Doane's arrangement of Vatican Genesis 235:

fiðan under themo folca         ferahtera manno,
uuilthu im thanna hiro ferh fargeƀan

19.   We will be examining Heliand 1144a, 3062a, 3990a, 5690a, 5916a, 5920a, and 5975a.  All Heliand passages are cited from Behaghel-Taeger (1996).  All translations are my own.  [Back]

20.   Hofmann does not place any stress on nu; however, given the emphasis, the alliteration, and the availability of a long, stressed variant , it seems likelier that it would be stressed, provided that it behaves like alliterating particles in Old English (Cable 1991, 22).  See also Hutcheson (1992, 136).  Robinson (1993) thoroughly discusses OE and its participation in stress in verse-initial position.  One might note from one of Robinson's examples (1993, 116) that there is an almost formulaic similarity between it (Seasons for Fasting l. 152) and Heliand 1144a: Nū wæs æt nēhstan þæt ūs nergend Crīst.  [Back]

21.   The quotative quað he 'he said' is extrametrical (Russom 1998, 138–39).  The second syllable of Ionases carries tertiary stress because it acts like the second half of a compound, in analogy with native Germanic names, e.g. Beowulf vs. Beowulfes, where -wulf- only takes stress when medial.  [Back]

22.   Genesis B 356a, Vatican Genesis 71a, 235a, and Heliand 3062a.  Heliand 1144a, in a sense, could also belong among this group, though it begins a separate passage within direct speech initiated three lines prior.  [Back]

23.   Genesis B 403a, 507a, Heliand 1144a, 3990a, 5690a, 5916a, 5920a, and 5975a.  [Back]

Works Cited

Árnason, Kristján. 1991. The rhythms of dróttkvætt and other Old Icelandic metres. Reykjavík: Institute of Linguistics, University of Iceland.  [Back]

Bartlett, Adeline C. 1935. The larger rhetorical patterns in Anglo-Saxon Poetry. Morningside Heights, N.Y.: Columbia University Press.  [Back]

Behaghel, Otto, ed. 1996. Heliand und Genesis. 10th ed. Rev. by Burkhard Taeger. Tübingen: Max Niemeyer Verlag.  [Back]

———. 1932. Deutsche Syntax: Eine geschichtliche Darstellung. Band IV: Wortstellung. Periodenbau. Heidelberg: Carl Winter Universitätsbuchhandlung.  [Back]

Bliss, Alan J. 1971.  Single half-lines in Old English poetry. Notes & Queries 18: 442–49.  [Back]

———. 1958. The metre of Beowulf. Oxford: Basil Blackwell.  [Back]

Cable, Thomas. 1991. The English alliterative tradition. Philadelphia: University of Pennsylvania Press.  [Back]

———. 1974. The meter and melody of Beowulf. Urbana: University of Illinois Press.  [Back]

Cathey, James E., ed. 2002. Hêliand: text and commentary. Morgantown: West Virginia University Press.  [Back]

Doane, Alger N., ed. 1991. The Saxon Genesis: an edition of the West Saxon Genesis B and the Old Saxon Vatican Genesis. Madison: University of Wisconsin Press.  [Back]

Fulk, Robert D., ed. 2001. Eight Old English poems: edited with commentary and glossary by John C. Pope. 3rd ed. New York: W.W. Norton & Compary.  [Back]

Gade, Kari E. 1995. The structure of Old Norse dróttkvætt poetry. Ithaca: Cornell University Press.  [Back]

Haferland, Harald. 2002. Der Haß der Feinde: Germanische Heldendichtung und die Erzählkonzeption des Heliand. Euphorion: Zeitschrift für Literaturgeschichte 95: 237–256.  [Back]

Hieatt, Constance B. 1980. Judith and the literary function of Old English hypermetric lines. Studia Neophilologica 52:251–257.  [Back]

Hofmann, Dietrich. 1991. Die Versstrukturen der altsächsischen Stabreimgedichte Heliand und Genesis. 2 vols. Heidelberg: Winter.  [Back]

Hutcheson, Bellendon R. 1992. Kuhn's Law, finite verb stress, and the critics. Studia Neophilologica 64:129–139.  [Back]

Jeep, John. 2002. The rhetorical significance of the alliterative tradition in the Heliand. In New insights in Germanic linguistics III. Edited by Irmengard Rauch and Gerald F. Carr. New York: Peter Lang.  [Back]

Krapp, George P. and Elliot V.K. Dobbie, eds. 1931–1953. The Anglo-Saxon poetic records. 6 vols. New York: Columbia University Press.  [Back]

Lehmann, Winfrid P. 1956. The development of Germanic verse form. Austin: Univeristy of Texas Press.  [Back]

Murphy, G. Ronald, S.J., trans. 1992. The Heliand: The Saxon Gospel. New York: Oxford University Press.  [Back]

———. 1989. The Saxon savior: The Germanic transformation of the Gospel in the ninth-century Heliand. New York: Oxford University Press.  [Back]

Olrik, Axel. 1909. Epische Gesetze der Volksdichtung. Zeitschrift für deutsches Altertum 51:1–12.  [Back]

Pope, John C. 1966. The rhythm of Beowulf. 2nd ed. New Haven: Yale University Press.  [Back]

Rauch, Irmengard. 2002. Historical pragmatics: Pervasive evidence from Old Saxon. In New insights in Germanic linguistics III. Edited by Irmengard Rauch and Gerald F. Carr. New York: Peter Lang Publishing, Inc. Pp. 211–219.  [Back]

Robinson, Fred C. 1993. The accentuation of in Cædmon's Hymn. In Heroic poetry in the Anglo-Saxon period: Studies in honor of Jess B. Bessinger, Jr. Studies in Medieval Culture XXXII. Edited by Helen Damico and John Leyerle. Kalamazoo, Michigan: Medieval Institute Publications. Pp. 115–120.  [Back]

Russom, Geoffrey. 1998. Beowulf and Old Germanic metre. Cambridge: Cambridge University Press.  [Back]

———. 1987. Old English meter and linguistic theory. Cambridge: Cambridge University Press.  [Back]

Schmid, Hans U. 2006. Ein neues 'Heliand'-Fragment aus der Universitätsbibliothek Leipzig. Zeitschrift für deutsches Alterthum und deutsche Literatur 135: 309–323.  [Back]

Sievers, Eduard. 1893. Altgermanische Metrik. Halle: Max Niemeyer.  [Back]

———, ed. 1878. Heliand. Halle: Buchhandlung des Waisenhauses.  [Back]

Simms, Douglas P.A. 2003. Reconstructing an oral tradition: Problems in the comparative metrical analysis of Old English, Old Saxon and Old Norse alliterative verse. Ph.D. diss. University of Texas at Austin.  [Back]

Suzuki, Seiichi. 2004. The metre of Old Saxon poetry: The remaking of an alliterative tradition. New York: D.S. Brewer.  [Back]

———.  2001. The metrical organization of the Heliand: Gradation and harmonization. International Journal of Germanic Linguistics and Semiotic Analysis 6:11–39.  [Back]

———. 1991. The Germanic hypermetric line in Indo-European perspective. In Perspectives on Indo-European language, culture and religion. Edited by Roger Pearson. 2 vols. Mclean Virginia: Institute for the Study of Man. Pp. 480–501.  [Back]

Tatian. 1892. Tatian: Lateinisch und altdeutsch mit ausführlichem Glossar. 2nd rev. ed. Edited by Eduard Sievers. Paderborn: Ferdinand Schöningh.  [Back]

Timmer, B.J. 1952. Expanded lines in Old English poetry. Neophilologus 35:226–30.  [Back]