Its Origin and Context
Hamðismál reports Hamðir’s last words and describes his death in a struggle against superior forces in these words:
“Vel höfom við vegit, stöndom á val Gotna,
ofan, eggmóðom, sem ernir á qvisti;
góðs höfom tírar fengið, þott scylim nú eða í gær deyia,
qveld lifir maðr ecci eptir qvið norna.”
Þar fell Sörli at salar gafli,
en Hamðir hné at húsbaki. (sts. 30–31)
[“We have fought well, we stand on Goth corpses, weary from the sword-edge like eagles on a branch; we have won great glory if we die now or yesterday, after the norns have given their verdict, no man outlasts the evening.” Then Sörli fell at the end of the hall, and Hamðir sank behind the house.]
Guðrúnarhvöt depicts Guðrún’s reaction when her sons have ridden off to battle after predicting that they will die there:
Guðrún grátandi, Giúca dóttir,
gecc hon tregliga á tái sitia,
oc at telia, táruchlýra,
móðug spiöll á margan veg (st. 9)
[Weeping Guðrún, the daughter of Gjúki, went sorrowfully to sit on the threshold and to recount, with tears on her face, grievous stories, many times.]
Both poems belong to the heroic cycle of the Edda, but the two passages make a different impression, and in the scholarly literature, the whole poems have been considered essentially different. In both quotations, the depicted character experiences something painful, but the depiction of this is very different. In the first case, the speaker exhibits a stoic emotional control; he does not complain at all at his violent death. In the second case, however, the pain is openly expressed through weeping and lament. In the first instance, the central character is an active fighter, in the second a passive mourner.
What is the reason for this difference? For a long time, Hamðismál and Guðrúnarhvöt were seen by scholars as simply different poems: they were different in character, and differences such as those mentioned were completely natural. But in the early twentieth century, their differences were regarded as fundamental not only between these two poems but between entire groups of Eddic poems. The difference between them was explained by differences in age, mentality, and the essential character of the poem. Guðrúnarhvöt came to be classified as belonging to a particular group of Eddic poems, the “elegies,” which were considered to be different as a group from the other poems in the heroic cycle. This view has since been revised and questioned, but it still has supporters and remains, albeit often in less rigid form, the standard picture presented in reference works.
It was Andreas Heusler who in a number of works from the early twentieth century laid the foundation for this view. He presented a general picture of the Edda poems, a “Gesamtbild” (1941, 188), where form, motifs, mentality, and time of composition were brought together. Among the heroic poems, he distinguishes a group of five, which he claims to be Norse representatives of pan-Germanic heroic poetry with roots in the era of the Great Migration: Brot af Sigurðarkviðo, Hlöðskviða, Völundarkviða, Hamðismál, and Atlakviða. He describes them as “doppelseitige Ereignislieder” (double-aspect poems of action), by which he means that they include both narrative passages and dialogue and they revolve around action. He dates them to the early Viking Age (153–154), that is, from the ninth or tenth century. He also dates another group of heroic poems to the Viking Age, “einseitige Ereignislieder” (single-aspect poems of action), which are purely dialogic, yet action-oriented: Reginsmál, Fáfnismál, and Helgakviða Hundingsbana II (178). Finally, he distinguishes a group that he describes as an “isländische Nachblüte der Heldendichtung” (late Icelandic flowering of heroic poetry), and this group is dated much later: the Christian High Middle Ages (eleventh and twelfth centuries), and they stand out more for “Ausmalen und Begründen der Seelenvorgänge” (causes and elaboration of psychic processes) than for depiction of actions. The group consists of the long narrative poems Sigurðarkviða in skamma and Atlamál, but primarily the so-called elegies. This group consists mainly of five “weibliche Rückblickslieder” (retrospective women’s songs), or “Frauenklagen” (women’s complaints): Guðrúnarkviða I, Guðrúnarkviða II, Guðrúnarhvöt, Oddrúnargrátr, and Helreið Brynhildar (181–188). Based on the formal patterns of these poems, he argues, a number of male retrospective poems were composed, which are found outside the Codex Regius (such as Örvar-Oddr’s death song and Hrókr’s wooing song; Heusler 1941, 185–186). But it is the group of female poems that Heusler is mainly interested in, and these are the poems that later scholarship has discussed as Eddic “elegies.”
Heusler’s primary argument is that these “elegies” are fundamentally different from what he describes as the “old” heroic poems. He speaks of a deep gulf (“tiefe Kluft”) of half a millennium between the groups (188) and generally ascribes to them different structures and literary character. He argues that they also express different mentalities: the spirit of the “elegies” is marked by an era of peace (187) in a Christian culture (Heusler 1906, 252).
Heusler neither argues much for his general model nor for its constituent parts. It is the system itself, the fact that everything—form, content, mentality, and time of composition—is linked together, that should support his view. His general model is formally oriented, and this applies not least to his main contrasts, the five supposedly ancient poems versus the five “elegies”: on the one hand, primarily narrative action-filled poems, and on the other hand, contemplative retrospective utterances with a lot of direct speech. But the reason that these formal features should be the basic criterion for a division of the Eddic poems is not made clear. The vague reference to the supposedly peaceful spirit of the “elegies” is in fact Heusler’s only real dating argument, in this case to the friðaröld (the peace age) of Iceland, that is: in the twelfth century (1941, 187).
Before Heusler, leading Edda scholars had identified neither a separate group of “elegies” nor the relative dating implied by Heusler’s general picture. Some of Heusler’s “elegies” are usually dated early and some of his “old” poems late; when the Eddic poems were categorized, poems from the opposite categories of the Heusler system were grouped together.1 These scholars also saw distinctive category features, but they were different from those of Heusler. They saw no fundamental gap (“Kluft”) between different groups of the heroic poems, neither in terms of time of composition nor in terms of character (cf. Sävborg 2004, 65–67).
But Heusler’s Gesamtbild became generally accepted, and since then the idea of a group of “elegies” with both markedly different character and a younger age has been the standard view. Subsequent criticism has mostly been concerned with details and can usually be described as a critique inside the system.2 There has been a tendency to move the dating of the “elegies” even further forward. While Heusler believed in a dating to the friðaröld, the twelfth century, later scholars, such as Ulrike Sprenger, date them all to the thirteenth century (i.e., just a little older than the manuscript) (e.g., Sprenger 1983, 193).
Heusler’s idea of the origin of the elegy group was, however, elaborated with other, more specific hypotheses. For Heusler, the Eddic elegy was a specifically Icelandic creation, and the condition for its emergence he explained vaguely by historical circumstances in Iceland. But in 1938–1939, Wolfgang Mohr published two articles in which he presented a new thesis about the origin of the allegedly younger Eddic poems (Mohr 1938, 1939). He builds on Heusler’s system of categories and dating, but according to Mohr, the “elegy layer” (“Elegienschicht” 1939, 150) is not an Icelandic phenomenon. Instead, he argued that these poems are imitations of Danish and Low German rhymed poems. The model is what he calls “das novellistische Spielmannslied” (the novelistic minstrel-song). No examples in this genre have been preserved: it is totally reconstructed by Mohr on the basis of Danish medieval ballads, which he claims were created out of this genre. Ballad is thus a sister genre of the “Eddic elegies.” He supports his theory by pointing out similarities between the “Eddic elegies” and Danish ballads (e.g., 1938, 280; 1939, 149). These alleged similarities consist of single words and motifs of all kinds, not just traditional “elegy” features like lament and retrospective focus. Mohr also devotes a long and detailed section to words and motifs related to needlework (1938, 236–237). Mohr’s theory became widely accepted in German and English scholarship. His view of the “Eddic elegies” is presented in, for example, Jan de Vries’ Altnordische Literaturgeschichte (1964–1967), Joseph Harris’s overview in Strayer’s Dictionary of the Middle Ages (1984, 390), and Heinrich Beck’s overview in Kindlers neues Literatur Lexikon (1992, 667). It should be noted that Mohr’s theory conflicts with the results of ballad scholarship. The Nordic ballad genre is much younger than Mohr thought, even younger than the Codex Regius Edda manuscript; nor does it have the German background that Mohr assumed, while the existence of the genre he claims to be the basis of both the ballads and the “Eddic elegies” is entirely rejected by ballad scholars.3
Ulrike Sprenger devoted several works to the “elegies” in the 1980s and 1990s, including the first complete monograph on them, Die altnordische heroische Elegie, from 1992. She presents a new general picture of the group, its distinctive character, and its origins. She accepts Heusler’s categorization and relative dating, but she magnifies the contrast he claimed between the “elegies” and other heroic poems even further. The character portrayal of the “elegies” differs fundamentally, she argues, from that of the “old” poems, especially in terms of emotional and moral interest that is lacking in the earlier ones (e.g., Sprenger 1988, 249, 268). Her conclusion is that the “elegies” emerged in the thirteenth century under the direct influence of religious literature. Their distinctive literary character is explained neither by a historical situation (Heusler) nor by impact from another poetic genre (Mohr). According to Sprenger, it originated under the influence of the ideals, spirit, and language found in, for example, homilies, saints’ lives, hymns to the Virgin, and the Bible.
But at the same time, a number of scholars published works that in different ways challenged the traditional image of the “elegies.” Joseph Harris attempted on the basis of Mohr’s theory and dating to broaden the perspective and argued for an Old Germanic background behind the “Eddic elegies.” He is interested less in the existing poems themselves than in the “formal elegy” he detects as a part of them. A number of key elements in the “Eddic elegies” are traced in West Germanic poetry, especially in Hildebrandslied and the Old English elegies (Harris 1988). A different perspective was chosen by Gísli Sigurðsson. He adopts modern theories of oral poetry and argues that the Eddic poems should be classified according to their audience. For him the “elegies” are not necessarily manifestations of a different age or a different literary fashion from other Eddic poems, but rather their distinctive character is explained by the fact that they were performed by women for a female audience (1990, 252). My own monograph of 1997, Sorg och elegi i Eddans hjältediktning, further developed some of Joseph Harris’s and Gísli Sigurðsson’s ideas, but it also argued that the sharp distinction between “elegies” and other Eddic poems was unjustified. In an analysis of Guðrúnarkviða I from 2005, Lars Lönnroth argued that oral-formulaic theory makes the distinction between young and old poems meaningless and that the “elegies” and their distinctive character combine an Old Germanic legacy with late foreign influences.
A response to several of these critical voices was made in 2005 when Vésteinn Ólason presented a defense of Heusler’s basic division of a group of Viking-Age “action-oriented poems” and a relatively uniform group of poems dated to about 1200, Heusler’s “Nachblüte” (late flowering) (Vésteinn Ólason 2005, 190). As their main characteristic, he distinguishes an “interest in emotions and inner life,” which he considered to be influenced by Continental courtly poetry (186–187).
Different scholars have focused on different matters when they discuss the characteristics of the “Eddic elegies” and their literary origin. Heusler focuses on form, the retrospective outlook, and the static, almost lyrical character of the “elegies.” Mohr collects an abundance of scattered elements, words, and motifs of different kinds. For Sprenger, a supposed moral character is a key feature, while for Vésteinn Ólason, emotional analysis is the distinguishing feature. Harris’s theory focuses on form once again but also on a number of fixed content-related elements. Gísli argues from the gender of the characters and the motifs linked to the different gender roles.
The focus may be different, but there are still common features. The idea that the “Eddic elegies” are something fundamentally different from other Eddic heroic poems is shared by all these scholars. All but Gísli claim a late dating of the “elegies.”
There are actually several key issues to consider:
It is clear that both Heusler and later scholars who have devoted themselves to the “Eddic elegies” have perceived them as young in relation to the rest of the Eddic poems. When the origin of the group is explained, the late dating is sometimes a precondition. For Mohr, they are by definition “younger” Eddic poems (“jüngere Eddalieder” in the titles of his articles), and this governs his comparisons (Mohr 1938, 1939). Sprenger also presumes a high medieval dating for the “elegies” when she searches for a possible model (e.g., Sprenger 1988, 283). It is therefore necessary to remember that the usual datings of the heroic poems of the Edda are not based on external, such as linguistic, criteria, but on preconceived assumptions about which motifs are young and old, respectively, which mentality is young or old, etc. (see Sävborg 1997, 52–57; 2004, 80–81). This means that dating is an uncertain basis for establishing the origins of the “Eddic elegy.”
*
Despite the differences in focus among scholars, it is clear that they all regard grief and lamentation as a characteristic feature of the “elegies.” Even those who focus on formal characteristics describe the poems as lamentation or grief poems (“Frauenklagen,” Heusler 1941, 184), and this also holds true for later scholars such as Sprenger (“Klage” 1988, 270) and Gísli Sigurðsson (“tregrófskvæði” [chain of grief poems] 1986, 127). Grief and open expressions of emotions are very prominent in Guðrúnarkviða I and II, Guðrúnarhvöt, and Oddrúnargrátr, more so than in most other Eddic poems. Investigation of the “Eddic elegy” as a phenomenon might therefore focus on grief.
Hitherto, scholarship on the “Eddic elegies” has assumed that their distinctive features are fundamentally alien to Norse tradition, violent heroic poetry, and mentality of a feud society. Thus, it has become necessary to look for parallels in later foreign traditions to explain these features as produced by external influences. Mohr, Sprenger, and Vésteinn Ólason propose totally different models, but they agree regarding the characteristics of the “elegies” as something alien to the Eddic-Norse-Germanic context. They base their conclusions on comparative arguments, but they all restrict their comparisons to the late period when they, following Heusler, assume the poems were composed. However, one can also consider the whole issue from another angle, choosing a general and broader comparative perspective and placing the “Eddic elegies” in a larger generic, literary, and social context.
Heroic poetry is an ancient worldwide genre with recurrent features. Western literature traditionally begins with a heroic poem, the Iliad (ed. Mazon 1955–6). Like the heroic poetry of the Edda, the Iliad revolves around struggles and violence in a relatively “primitive” warrior society. Killing and death are the focus of the narrative, in the Iliad as in the Edda, and this is characteristic of heroic poetry as a phenomenon. But not only violent acts are depicted in the Iliad. After the acts of violence, grief over the dead is often mentioned. This is most notable in song 22 and 24, after Hector’s death, and the entire epic ends with the lament of the women at his funeral. Grief and lamentation are fundamental elements of the depiction of struggle and killing in the Iliad. European medieval heroic poetry is similar. The Old French Chanson de Roland depicts a war adventure, but here, too, grief is a prominent motif. After the long battle scenes, long scenes of grief follow; exceptional warrior ability has a parallel in exceptional lamentation (lines 2414–2932, etc.). Middle High German heroic poetry also revolves around heroic deeds and acts of violence: grief over the dead is a central element (e.g., Nibelungenlied (ed. Bartsch 1961) lines 1002–1103, 2377–2379; Rabenschlacht lines 1107–1109, 1055–1058; Diu Klage (ed. Bartsch 1875)). Grief and lament are elements that seem to belong to heroic poetry as a genre, regardless of time and culture. The violence that the plot revolves around is almost regularly followed by grief over the dead.
West Germanic heroic poetry is closest to Eddic poetry in genre and tradition: they can be seen as sister traditions. Since Old English heroic poetry is written down in the tenth and eleventh centuries, in this context it represents an ‘early’ period, the period to which the supposedly old Eddic poems exhibiting a harsh, non-emotional and non-elegiac character are dated. Like other heroic poems, Beowulf depicts the deeds of heroes, mostly fighting against monsters. But together with the violence, death is also present, and together with death, there is mourning over the dead. The poem begins and concludes with sorrowful funeral scenes. In the final scene, the hero himself is buried after a heroic death, and the grief and lamentation at his funeral are described in detail (lines 3032–3178). Lamentation over the dead resounds through the poem for Grendel’s many victims. But of particular interest is the Finnsburh episode, a retelling of a heroic lay within Beowulf. It has a typical heroic plot, the treacherous killing of a relative, a temporary reconciliation, and subsequent revenge on the oath-breakers. But the depiction does not focus on the acts of violence but on the grief to which they lead (lines 1075, 1117–1118, 1148–1149, etc.). Deor (ed. Krapp 1936) is another poem cataloguing heroic motifs. It consists of a monologue by a sorrowful poet who lists cases of grief and misery from heroic legend to bring consolation in his unhappiness. In this poem, too, lament appears as a natural consequence of the depiction of the main characters and events of ancient Germanic heroic poetry.
The prominent position of grief in heroic poetry may seem surprising. Depiction of such a relatively static and “weak” emotion as grief may seem to contradict heroic poetry’s focus on killings and violent acts. But it seems to be an essential connection. Nor does it seem to be a pure literary phenomenon; rather it is deeply rooted in the reality of violent societies. Margaret Alexiou has noted, “the dirge is always strongest where the law of vendetta flourishes” (22). The remark is made in a study of role of ritual lament in connection with blood revenge in Archaic Greece, but she stresses that this applies also in contemporary Sicily and the Greek Mani peninsula. There is a strong link between poetry of grief and a society, where feud and blood revenge are central elements.
Skaldic poetry is the sister of Eddic poetry in Norse tradition. It is metrically related to Eddic poetry, composed and transmitted in the same milieu. Skaldic poetry is an important witness to Old Norse poetic tradition, not least because it is more datable and its chronology can be reconstructed. The dominant form is the praise poem, in which princes are praised for their deeds, in both Viking-Age examples and high medieval poems several centuries later. Grief is not a dominant theme in this type of poetry, but it is frequently found there. The grief of the skald and the people over the dead prince is an important element of the praise in, for example, the poems Hallfreðr composes on Óláfr Tryggvason (Erfidrápa 5; Skj BI: 151), Sigvatr on Óláfr Helgi (Erfidrápa 9, 19, 25; Skj BI: 239–245, lv. 22, 24, 26; Skj BI: 251–252), Erlingr Skjálgsson (Str 8; Skj BI: 230), and Oddr kíkinaskáld on Magnús inn góði (Str 2 and lv.; Skj BI: 327–328). The last two cases are entirely laments for the dead king. The skald both describes his own grief and chronicles the weeping of the king’s men. Skaldic poetry also has instances of personal grief over dead relatives (cf. Sigvatr [lv. 22; Skj BI: 251] and Egill Skallagrímsson [lv. 10, 13; Skj BI: 44–45]). But most important is Egill Skallagrímsson’s Sonatorrek, whose main subject is the skald’s own grief over the death of his sons: a pure lament poem of 25 stanzas in which Egill describes the physical effects of grief, expresses several different emotions, and relates the events that caused his grief (Skj BI: 34–37).
Eddic heroic poetry is thus part of an international genre in which grief and lament over the dead are essential elements. It is composed in a feud society, that is, in a type of society that generally promotes lament poetry. It belongs in the context of Germanic poetry in which grief over the dead is frequently depicted and is prominent in individual poems. In the Norse poetic context, grief over the dead is a well-known feature. Seen from this perspective, it would be strange if the Eddic heroic poetry did not contain grief and lamentation. As it does. Grief is a well-established motif in most of the heroic poems of the Edda, and it is a prominent feature of many: in four of the “elegies,” Guðrúnarkviða I, Guðrúnarkviða II, Guðrúnarhvöt, and Oddrúnargrátr, and also in Sigurðarkviða in skamma, Helgakviða Hundingsbana II, Hamðismál, and Brot af Sigurðarkviðo. Grief appears as a natural element, seen in the cultural and social contexts to which the poems belong, and as a natural part the traditional poetic heritage. But that is not how scholars have usually viewed grief in the heroic poems of the Edda. The “Eddic elegies” as a group have been separated out as a secondary phenomenon, as non-genuine heroic poems, primarily because of their focus on grief. Lament has been regarded as a phenomenon alien to the indigenous legacy of heroic tradition. Characteristic is Ulrike Sprenger’s statement: “Generell gesehen ist auch eine Klage im Rahmen eines alten Heldenliedes mit seinem springenden Stil kaum denkbar” [In general within the frame of an old heroic poem with its energetic style, lament is scarcely imaginable] (Sprenger 1992, 53–54). The change comes in a later age, marked by foreign influences: Christian ideals and Christian literature: “Die alte heroische Haltung, die Gefühlsäußerungen ausschloß, wurde nun aufgegeben, und es wurde ein anderes Heldenideal vertreten, wo man frei seine Gefühle äußern, klagen dürfte” [The old heroic attitude, which excluded the expression of feelings, was now abandoned, and another heroic ideal appeared, in which one might freely express one’s feelings, might lament] (1988, 259). Klaus von See takes the same basic view as a criterion for his re-dating of Hamðismál. According to him, the absence of “die starre, gefühlsharte Haltung des Heldenpersonals in der älteren Dichtung” [the unyielding, unemotional behaviour of the heroic characters in the older poetry] indicates that the poem is young (von See 1981, 258). Richard Leicher writes about the hero of the old Eddic poetry: “die Kraft ihres Schmerzes setzt sich in Tat um, nicht in Klage” [the strength of their pain is converted into action, not lament] (Leicher 1927, 34). Rose Zeller expresses a similar view: “Ein Verweilen bei der Klage würde von dem alten Dichter als unwichtig und vielleicht sogar als würdelos empfunden worden sein” [Lingering over lamentation was felt by the old poet as unimportant and perhaps even unworthy] (Zeller 1939, 53). This view has often constituted the main criterion for dating the “Eddic elegies.” Gustav Neckel, for example, dates the Guðrún lays with the following argument: “Und nicht bloß stofflich, auch in geist und stimmung geben sie sich als mittelalterliche dichtungen zu erkennen. Weiche gefühle, liebe und trauer, und seelisch verfeinte motive spielen in ihnen eine rolle, die in heidnischer zeit unerhört wäre” [Not only in terms of content, but also in spirit and tone they are recognizably medieval poems. Gentle feelings, love and sorrow, and psychically refined motifs play a role in them which would have been unheard of in pagan times] (Neckel 1908, 234).
This view is generally taken for granted by scholars and is therefore rarely argued for directly. When comparative arguments are used, they are usually put forward within Heusler’s model, motivated by contrast with those Eddic poems that Heusler identified as old and are thus perceived as a norm for what is old in Norse tradition (e.g., Sprenger 1988, 283). The argument is not entirely logical, since “elegiac” features in many cases are also found in the poems that Heusler and other scholars perceive as old. Brot af Sigurðarkviðu and Völundarkviða contain lament scenes and depict weeping women (Brot sts. 14, 15; Vkv sts. 29, 31). In Hamðismál, a poem that Heusler and his followers believed to be very old, numerous stanzas depict grief or express lamentation (e.g., sts. 1, 5, 7–9). This should reasonably indicate that the difference between “elegies” and other Eddic heroic poems is not so clear-cut, yet many scholars have not wanted to see it that way, explaining such stanzas as “elegiac” interpolations from a later time (e.g., Gering 1931, 428; Dronke 1969, 182; Sprenger 1983, 185). However, these ad hoc arguments are hardly scientific in their method. They presuppose what should be proved.
But viewed from a truly comparative perspective with literature of a known age and background, the case looks different. This perspective suggests that grief and lamentation over the dead are traditional native motifs in Eddic heroic poetry, and are not alien to indigenous mentality, poetic tradition, or heroic poetry, and thus there is no reason to regard the phenomenon of “Eddic elegies” as originating under foreign influences.
*
I have discussed the occurrence of grief as a motif in heroic poetry; now I turn to its depiction. The “Eddic elegy” is a complex literary form. To explain the background and literary context of these poems, it is necessary to look in detail at their treatment of grief. A number of recurrent features in the presentation of grief in the “Eddic elegies” appear to be typical of these poems as a group.
Weeping is the most common expression of grief in the “Eddic elegies,” occurring in the majority of the heroic poems, even in those that are perceived as old. It is, however, common in poetry from all periods and cultures, in high medieval courtly poetry and religious literature, as well as in West Germanic poems like Beowulf or in Viking-Age poetry by skalds like Oddr kíkinaskáld. Of greater interest is the gesture of striking the hands together, höndum slá, mentioned in connection with Guðrún’s grief after Sigurðr’s death in three Eddic poems, Guðrúnarkviða I st. 1, Guðrúnarkviða II st. 11, and Sigurðarkviða in skamma st. 25. The gesture has in modern times been associated with positive emotions (like applauding), but in the three Eddic cases, it is clearly an expression of grief. The gesture is not found in high medieval continental literature, where grief is rather expressed by wringing one’s hands. However, in a work from 1935, Gustav Neckel showed that the hand-striking gesture occurs several times in the ninth-century Old Saxon Hêliand (“handun slôg” 2183, etc.; Neckel 1935, 64). Both wording and metrical position are the same as in the two Guðrún lays, thus it must be a common poetic formula. The gesture is not found in the Latin prose model of Hêliand but is an addition by the Germanic poet. Quoting Neckel, we can talk about a unique “altgermanische Geste des Schmerzes” (old Germanic grief gesture) (the title of Neckel’s article). There is no doubt that this detail of the depiction of grief in these “elegiac” Eddic poems reflects an indigenous tradition originating in common Germanic tradition, not in late foreign influence.
Guðrúnarkviða I opens with this description:
Ár var, þatz Guðrún gorðiz at deyia,
Er hon sat sorgfull yfir Sigurði. (st. 1)
[It was long ago that Guðrún intended to die, when she sat sorrowful over Sigurðr.]
The verb construction sitja yfir recurs in two other Eddic poems depicting Guðrún at the body of the dead Sigurðr: Guðrúnarkviða II (“þá er sárla satc yfir Sigurði” [as I sat grieving over Sigurðr] st. 12) and Hamðismál (“saztu yfir dauðom / glýia þú né gáðír” [you sat over the dead man / gave no thought to happiness] st. 7). The first two poems are “elegies.” Hamðismál has, since Heusler, been regarded as old in contrast to the “elegies,” but here the poems show similarities in depicting grief. This seems to be a formula linked to description of a mournful vigil over a dead person and thus has attracted the attention of Ulrike Sprenger as her main argument for the origin of the “elegies” in Christian-religious literature. In support of this thesis, she refers to several instances of sitja yfir from Icelandic prose, all with religious affiliations (Sprenger 1988, 260). Only one of these instances, however, really describes sitting over a dead person, the late version of Marte saga ok Marie Magdalene from the fourteenth century, and this appears to be the only relevant example. The link between the formula “sit over a dead person” and religious-Christian literature is not strong. Sprenger attempts to trace the formula to the Vulgate, where Mary Magdalene is outside Jesus’s empty tomb: “Stabat ad monumentum plorans” [stood at the sepulchre weeping] (John 20:11; Sprenger 1983, 190). But this passage does not refer to sitting or to a vigil over a dead body at all. However, the formula is frequently found in connection with mournful vigils over dead people in entirely different literary contexts. It occurs in Icelandic saga prose, in Heimskringla’s description of Haraldr hárfagri’s grief at the body of the dead Snæfríðr (Heimskringla 126). Ute Schwab has also shown that the formula is well established in lament scenes in German heroic poetry (382): “manec heidnischez wîp, / die sâzen ob den tôten und qualten sêre ir lîp” [many heathen women / they sat over the dead and greatly tormented their bodies] in Ortnit (ed. Ame-lung and Jänicke 1968, st. 427). Further cases are found in Rabenschlacht (sts. 977, 1129) and Dietrichs Flucht (ed. Martin 1967, l. 9903). These appearances in heroic poetry support the argument about the formula’s occurrences in Eddic poetry. And more importantly, Joseph Harris has shown (Harris 1988, 87) that the formula is also found in an Old English heroic poem, Beowulf, in the same context of vigil over a dead hero as in the Edda: “Wiglaf siteð / ofer Biowulfe, byre Wihstanes” [Wiglaf, Wih-stan’s son, sits over Beowulf] (ed. Wrenn 1958, lines 2906–2907). Again there is no doubt that we are dealing with an indigenous element in the grief depiction of the “Eddic elegies” ultimately inherited from common Germanic heroic tradition.
The starting point and key topic in the first “elegy” in the Poetic Edda, Guðrúnarkviða I, is the heroine’s remarkable inability to express grief, caused by grief itself (see Hill, Chapter 5, this volume). The poem exhaustively and repeatedly describes how Guðrún cannot weep (sts. 1, 2, 5, 11) or perform the traditional grief gesture of striking her hands together (st. 1), nor express her grief in words (st. 23). It is clear that this inability is a consequence of her great grief (e.g., st. 2). The poem depicts attempts to help Guðrún to cry: the first two fail, while the final attempt is successful. Then she weeps, laments her grief, celebrates the dead Sigurðr, and verbally attacks those responsible for his death. By giving open expression to her grief, she regains her strength. There is a marked parallel to this in Norse poetry. In Egill Skallagrímsson’s Sonatorrek, the difficulty of expressing grief is the poem’s starting point. “Mjök erum tregt / tungu at hrœra” [It is very grievous to move the tongue] (Skj BI: 34) are the introductory words of the poem, and the first two stanzas revolve entirely around this motif. Here, too, it is made clear that grief causes this difficulty (“ekki veldr” [does not cause]). Later in the poem, there are further parallels to Guðrúnarkviða I. Sonatorrek, too, depicts how the difficulty of expressing grief is overcome. Gradually Egill expresses his grief more openly, and, like Guðrún, he later attacks those responsible for the grief (sts. 7, 8). At the end, the grief is still there, but the paralysis it caused is gone. It is difficult to determine whether there is a direct influence between the two poems or whether both connect to a common tradition. But it is clear that no poem is closer to the “Eddic elegy” Guðrúnarkviða I in structure and main theme than this indigenous skaldic poem.4
Little imagery can be clearly linked to grief depiction in the “Eddic elegies.” But there is one example in Guðrúnarhvöt, in Guðrún’s words before she ascends the pyre after having listed the sorrows of her life:
Megi brenna brióst bölvafult eldr,
… um hiarta þiðni sorgir! (st. 20)
[May fire burn up the breast so full of wrongs, … may sorrows melt about the heart.]
The frozen breast is a metaphorical description of grief or sorrow (“sorgir”). Metaphors of cold are frequent in depictions of grief, as in Völundarkviða (st. 31) in connection with Niðuðr’s grief over his sons’ death. The late medieval Skáldhelgarímur refers to “rygdar iokul” [the ice of grief] (Rímnasafn 105). Such metaphors are well known in Old English poetry. In The Wanderer, the word “wintercearig” [winter-sorrowful] is used (l. 24), and in Deor it is said of Welund: “hæfde him to gesiþþe sorge ond longaþ, / wintercealde wræce” [sorrow and longing he had as companions / winter cold grief] (4–5). In addition, realistic descriptions of cold, frost, and hail are often symbolically connected with personal grief (cf. The Wanderer 2–4, 48–50; The Seafarer 8–11). Against this context, Sprenger’s attempt to link the metaphors of coldness in the “Eddic elegies” with religious literature seems implausible. She notes cases where words for ice and frost are connected to shame and envy (Sprenger 1985, 171–172). The argument is unconvincing, since in Guðrúnarhvöt it is explicitly said that it is Guðrún’s “sorgir” that should melt; there is no connection with the moral defects of Sprenger’s religious parallels. A Norse-Germanic context gives a better explanation for a phenomenon in the “Eddic elegies” than later foreign influences.
The depiction of grief in Eddic poetry is not hyperbolic as European high medieval depictions often are. The Eddic poems lack the descriptions of violent collective weeping and fainting in Chanson de Roland (ed. Atkinson 1924), the long monologues of lament in Rabenschlacht, and the grotesque self-injuring behavior of Dietrichs Flucht (see examples in Sävborg 1997, 263). In fact, this is a marked difference between the “Eddic elegy” and the poetry usually identified as its contemporary. The “Eddic elegies” emphasize grief through quantity instead. A recurring technique is to intensify current grief by connecting it with other sorrows, often in a kind of catalogue of grief. When Guðrún in Guðrúnarhvöt bursts into tears after her sons’ departure to their death, she relates in a monologue all the major sorrows in her life. The starting point is the death of her daughter and her sons, but this grief is intensified in the context of other sorrows. She herself calls it a “tregróf” [a chain of sorrows], and she wishes that it could give comfort to others (st. 21). At the beginning of Guðrúnarkviða I, two women, Gjaflaug and Herborg, recount their unhappy lives, listing all the sorrows that have befallen them. Their aim is to release Guðrún from her emotional paralysis, but the two monologues also work to emphasize Guðrún’s grief: when later in the poem she begins to lament, this becomes the climax of a series of lament monologues (brilliantly described in Sprenger 1992, 19). The technique of intensifying grief by listing further sorrows is also a characteristic feature in Egill Skallagrímsson’s Sonatorrek. The starting point is the skald’s grief over his son Böðvarr’s death, but Egill also mentions the death of his parents, his brother, and his second son. The effect is to suggest that Egill’s whole life is full of sorrows, using the same tregróf technique as Guðrúnarhvöt and Guðrúnarkviða I. There are also some parallels to this phenomenon outside Norse poetry. Old English Deor is basically a monologue in the form of a grief catalogue, where the starting point is the speaker’s misery, which he parallels with sorrows from heroic legend. The aim here is less the intensification of grief than consolation, an aim that, however, is also present in Guðrúnarkviða I and Guðrúnarhvöt. No continental parallels to this phenomenon have been found. Mohr does note a parallel in a Danish ballad, Hustrus og Moders Klage (DgF 286; Mohr 1938, 247), in which two women outbid each other regarding sorrows in life. But since we now know that the ballad genre is much younger than Eddic poetry, this ballad does not solve the problem of the origin of “Eddic elegy” (see e.g., Jonsson 1991, 164–166). The tregróf in these Eddic poems seems to have an indigenous background, and it would be reasonable to see the ballad as a part of that Nordic tradition, too.
Drawing conclusions from the terminology of grief is problematic, since the lexis (sorg, harmr, gráta, etc.) is indigenous and present in both older skaldic poetry and in younger works. Yet the terminology has played a role in the discussion of origins. Sprenger shows that the word sárr used for emotional pain (in Ghv, Gðr I, Gðr II, and Sigsk) is never found in the older skaldic poetry but is common in religious literature (1992, 227–231). But as Joseph Harris has shown, the word is commonly used for emotional pain in both Old Saxon and Old English poetry, including the heroic poem Beowulf (1988, 86). Thus, there is an indigenous context for this typical “elegy” word, too. More important is the word tárughlýra (with tears on the face), used of the weeping Guðrún in Guðrúnarhvöt st. 9 (see the second of the quotes that open this chapter). It is a hapax legomenon in Norse literature, but a cognate does occur in the Old English Genesis: tearighleor (ed. Krapp 1931, l. 2276), preserved in the tenth-century Junius manuscript. It is unlikely that the poet of Guðrúnarhvöt was influenced directly by the Old English poem. We have obviously to do with a specific poetic word for depiction of open grief, a legacy in the oral tradition from a common Germanic concept.
*
The tregróf sections in Guðrúnarhvöt and Guðrúnarkviða I are retrospective in outlook, one of the aspects that was the basis of Heusler’s distinction of the “elegy” as a group. Heusler argued that this aspect connected these two poems with Oddrúnargrátr, Helreið Brynhildar, and Guðrúnarkviða II, and also with seven poems with male speakers from the fornaldarsögur and Saxo’s Gesta Danorum. However, the retrospective mode is a diverse phenomenon in these poems, in both function and form.
In Guðrúnarhvöt and Guðrúnarkviða I, the retrospection is integral to, and thus dependent on, the depiction of grief. It can hardly be described as a category-founding feature. Neither of these poems is solely retrospective. The first is half retrospective, and in Guðrúnarkviða I, the retrospective passages occupy a remarkably small space. Oddrúnargrátr contains a long retrospection that lists the great sorrows of the speaker’s life, the unhappy love affair with Gunnarr, and his death, so in this case, too, a clear relationship exists between retrospection and grief. Helreið Brynhildar, however, is an overview of the valkyrie’s whole life, in which grief is not prominent. Guðrúnarkviða II is a first-person narrative poem that covers a long time. Large parts consist of dialogue, and grief depiction occurs alongside much other narrative material. The retrospective mode is thus of a different kind than in Guðrúnarhvöt and Guðrúnarkviða I. The male-voiced poems mentioned by Heusler are retrospective monologues, pure catalogues of warrior deeds, such as Örvar-Oddr’s death song, Hrókr’s wooing song, and two Latin poems spoken by Starkaðr in Saxo’s Gesta Danorum. Víkarsbálkr and Hildibrandr’s death song have a more tragic tone, but none of them openly expresses grief. Neither in form nor function do these poems have any similarities with Guðrúnarhvöt and Guðrúnarkviða I. In Hjálmarr’s death song, the dying hero establishes an elegiac contrast between his death and his former happy life at the Swedish royal court with the king’s daughter. Although he expresses grief, no literary similarities are found between this depiction and the ones in the Guðrún lays.
In form and function, the retrospections summarized above are completely different phenomena in different Eddic poems; thus, Heusler’s categorization rests on shaky ground. The focus on retrospection is just one manifestation of Heusler’s markedly form-oriented view of Eddic poetry, its chronology, and its Old Germanic context. He argued that Germanic heroic poetry developed from a “doppelseitiges Ereignislied” (double-aspect action poem), with a larger amount of narrative verse and a smaller amount of direct speech, to a “situation-oriented” type of poetry, with monologue and dialogue as the dominant elements. Yet the West Germanic heroic poems preserved hardly support this idea. Hildebrandslied contains 69% direct speech and the Waldere fragments contain 94% (Heusler 1902, 195), and both poems are in their extant form distinctly situation-oriented. Heusler’s model presupposes that Eddic poems like Atlakviða represent the most original form and thus presupposes what should be proved (cf. Sävborg 2004, 61–64).
The retrospective lament poems preserved in Old English have also been called “elegies,” and in recent years a possible connection with the Eddic “elegies” has been suggested. In several studies, Joseph Harris has claimed a generic link between the two traditions, a thesis that would strengthen the theory of an Old Germanic context behind the “Eddic elegies” and their distinctive character (Harris 1988). But even if a vague similarity exists between Deor and a few of the Eddic “elegies” regarding the tregróf technique, the differences are more apparent. The female-voiced elegies Wulf and Eadwacer and Wife’s Lament describe a longing for a man from whom the woman is separated, and the happy past is contrasted with the sad present. The monologues are lyrical, detached from all narrative context. This is markedly different from the Eddic “elegies.” There the narrative poems occur in a larger context of narrative heroic poetry, and the laments concern specific events in heroic legend: grief over dead men. This is a kind of grief, as we have seen, which is characteristic of heroic poetry in general, and it is rather in this context the “Eddic elegies” and their grief should be placed. The other Old English elegies, like The Seafarer, The Wanderer, Ruin, Resignation, and Rhyming Poem, differ even more from the Eddic poems. The epic context is absent here, too, and their purely lyrical character is even more pronounced. The grief is produced by longing within exile, but the poems also express sorrow over the world’s vanity and transitoriness (The Seafarer). This is philosophically reflective poetry, not heroic engagement with death and grief for the fallen. As Germanic alliterative poems, they of course have an ultimate common origin, but it is difficult to see a closer relationship between Old English and Eddic “elegies” on the basis of “elegiac” peculiarities.
*
Guðrúnarhvöt does not begin with Guðrún’s lament or with any depiction of grief. It begins with a whetting scene. Svanhildr is dead, and Guðrún incites her sons with harsh words to avenge their sister’s death:
er harðhuguð hvatti at vígi
grimmom orðom Guðrún sono. (st. 1)
[The fierce-spirited Guðrún whetted for the fight, with grim words, her sons.]
Guðrún continues by implying that her sons are cowards, and she contrasts this to her brothers’ courage and fierceness. First the sons are reluctant, but then they capitulate. Laughing, Guðrún chooses helmets and mail-coats for her sons. Only after their departure does she burst into tears.
When in Guðrúnarkviða II Guðrún receives the message about Sigurðr’s death, she responds with a vengeful verbal attack against the killer, Högni: “þitt scyli hiarta hrafnar slíta” [may the ravens tear out your heart] (st. 9). Guðrúnarkviða I is dominated by Guðrún’s grief. When she is finally able to express it, she laments openly, but in stanza 21, she also curses the murderers and predicts the killing of Gunnarr. Such features have puzzled many of the scholars of the Heuslerian school. Vengefulness and fierce heroism have been linked to the “old” Eddic poems contrasted to the “Eddic elegies.” So, for example, Franz Rolf Schröder argued that the whole whetting scene in Guðrúnarhvöt was an interpolation from an old poem, arguing that a whetting did not belong in a young, grief-focused poem (Schröder 434). Sprenger claims similarly that stanza 21 of Guðrúnarkviða I, where Guðrún curses Gunnarr, is an interpolation of an old “doppelseitiges Ereignislied” [double-aspect action poem] arguing that such a stanza would be strange in a poem devoted to Guðrún’s grief (Sprenger 1992, 6).
These arguments clearly illustrate the Heuslerian view about the special character of the “Eddic elegies” and the distinction between them and the other, apparently older Eddic poems. Heusler himself connected the spirit of the “elegies” with Iceland’s friðaröld (“age of peace”; Heusler 1941, 187), and he claimed that “young” and “old” Eddic poems evaluated the events of heroic legend fundamentally differently: “Der vorkirchliche Geist b e j a h t diese Welt der Untaten” [the pre-Christian spirit valued positively this world of wicked deeds] (1906, 252). Jónas Kristjánsson sees in the “young” Eddic poems “the influence of Christianity and its message of mercy and love” (Jonas Kristjánsson 1988, 64). Sprenger generally supposes a connection between the emotional orientation of the “elegies” and their supposed rejection of the old heroism (e.g., Sprenger 1985, 967).
But Guðrún’s words and acts in these three “elegies” are not pacificistic. Moreover, her curse of Gunnarr in Guðrúnarkviða I occurs in the stanza immediately following her lamentation and hard-won tears, and her desire for revenge in Guðrúnarkviða II is articulated in the same stanza in which she mentions her grief (her harmar). The vengeance-whetting in Guðrúnarhvöt is introduced by a stanza relating Guðrún’s grief, and after the whetting, her detailed lament monologue follows. Grief and vengefulness do not seem to be opposite phenomena in the “Eddic elegies.” This also applies to other Eddic poems. In Helgakviða Hjörvarðssonar, Sváva reacts, in the same stanza, with both lamentation and a vengeful outburst when she receives news of the death of Helgi (st. 38). In Hamðismál, the introductory stanza 2 explains that Guðrún whetted her sons to revenge, but when she does speak, she laments her lack of joy and her sorrows (sts. 5, etc.). The last example is discussed by Carol Clover, who claims it shows that Guðrún’s lament was an implicit incitement to revenge; her sons also interpret her words thus (Clover 1986, 161). She concludes, “the apparent confusion in Guðrúnarhvöt and Hamðismál between incitement and lament is not accidental, the result of a redactorial botch, but organic” (162).
Clover’s thesis has good support in Germanic material. Whetting and lament, vengeance and grief, are not contradictory phenomena; rather they are complementary in Eddic heroic poetry. The link appears to be almost generic. Revenge and grief are also closely related in Old French and Middle High German heroic poetry. After a description of violent grief reactions in Chanson de Roland (e.g., 2414–2422), this request follows: “Car chevalchiez, vengiez ceste dolor!” [To horse! avenge this sorrow] (2428). A mighty battle follows. Lamentation over the dead is a means of incitement to vengeance. As previously mentioned, the dirge is strongest in cultures where the law of blood-vengeance prevails. In the Iliad, Achilles’s immoderate grief goads him to his greatest feat, the slaying of Hector. Scenes of lament and heroic deeds follow one another in Beowulf. From this perspective, there is nothing strange about the presence of whetting and the vengeful outbursts in the “elegies” or about the elements of grief in other Eddic poems. In this regard, there is no difference between them. Neither the “elegies” nor the other Eddic poems are peaceful works but rather martial, heroic poems.
*
The Finnsburh episode in Beowulf is of great importance for the understanding of Germanic heroic poetry. It is, of course, only an episode within the larger epic, but what is described is a performance of a short poem about an episode from heroic legend. The poem performed in the Danish hall is thus, with regard to its length and scope, more comparable to the Eddic heroic poetry than is Beowulf as a whole with its epic length and myriad events. Although the Finnsburh episode is a summary of a lay, perhaps part of a larger epic cycle, and not the poem itself, it probably gives information about the theme and focus of this poem.
The summary of the poem about the aftermath of the battle at Finnsburh begins with these words:
Ne huru Hildeburh herian þorfte
Eotena treowe; unsynnum wearð
beloren leofum æt þam lind-plegan
bearnum ond broðrum; hie on gebyrd hruron
gare wunde; þæt wæs geomuru ides.
Nalles holinga Hoces dohtor
meotodsceaft bemearn. (v. 1071–1077)
[Nor indeed did Hildeburh need to praise the faith of the Jutes; guiltlessly she was bereft of dear ones, brothers and sons at the battle; they fell in succession, wounded with spears; that was a mournful lady. Not without cause Hoc’s daughter bewailed fate.]
After this section on mourning, there is a brief explanation of the background to the fight that gave rise to the grief. Grief continues to be a prominent theme. Hildeburh laments over her slain son and brother (1117–1118), and two men’s grief over the events is also portrayed (1149). Throughout the episode, grief for slain family members occupies considerably more space than the battles that caused their death. The central acts of violence in the poem are alluded to through the grief they cause the main female character.
Not only is it remarkable that grief should dominate the representation of a story about killing and revenge, it is also noteworthy that the focus of this whole violent episode from heroic legend is on a woman. For it is Hildeburh who is the central character in the depiction of the events in the episode, it is with her that the summarized poem begins, and she is the one who dominates the rest. She does not dominate through direct actions but as a grieving person, a mourner.
This problematizes the views of several Edda scholars. For those who accept the Heuslerian model, the focus on static scenes of mourning is a late and foreign phenomenon contrary to the focus on action in the “old” heroic poems (typical comments are e.g., Turville-Petre 1972, 132–133; Holtsmark 1957, 153). For many, the focus on women in Eddic poems is a late influence (e.g., de Vries 1964, 302). But as is shown by the Finnsburh episode in Beowulf, these phenomena are neither recent nor foreign in Germanic heroic poetry. The tendency we see in the Finnsburh episode is the same as we see in the “Eddic elegies” Guðrúnarkviða I, Guðrúnarkviða II, and Guðrúnarhvöt, where the central acts of violence, the killing of Sigurðr and Svanhildr, are represented through the grief they cause the central female character, Guðrún.
In the Finnsburh episode, the violence causes misery for several of the poem’s characters, and in one case, this is also mentioned (1149). But the principal mourner is a woman. Women do not appear often in Beowulf, but when they do, their role is typically, as in Hildeburh’s case, that of mourner. At the end of Beowulf, a woman appears at the mourning ceremony after Beowulf’s death. The section also describes collective male grief in general terms, but elaborated, individual, and verbalized grief is indicated only by this anonymous woman, whose sole function in the poem is to mourn (3150–3155). The tendency is the same in the three Guðrún “elegies” mentioned. The killings in the poems must have distressed many people, and some men also seem sad, but it is the woman, Guðrún, who is the main mourner.
The similarities between these “Eddic elegies” and the Finnsburh episode are remarkable. A woman is placed at the center of the narrative, although the episode in heroic legend revolves around acts of violence between men. But these acts of violence are primarily portrayed through the woman’s grief, and thus the narrative is focalized through the woman. This focus is one of the most characteristic features of “Eddic elegies” as a group, and it is one of the features that has been regarded as most significantly different from the “old” Eddic poetry. But the Finnsburh episode shows that there is no reason to see this as a foreign feature. This, too, belongs to the context of indigenous Germanic heroic poetry.
The focus precisely on a woman in a story depicted primarily through grief is significant for the understanding of “Eddic elegy” as a possibly distinct phenomenon. This feature really does distinguish the “elegies” from other Eddic heroic poems. In two articles, Gísli Sigurðsson has presented a view of the Eddic poems where he focuses on the main characters’ gender as crucial to understanding the poems as a whole. Gísli’s starting point is modern theories of oral poetry, and his aim is to classify Eddic poetry by its audience during oral transmission. The same episodes from heroic legend would have been mediated in totally different forms at the performances for female and male audiences, respectively. Following Parry and Lord, he believes that Eddic poetry as oral poetry was re-created at every performance, so it is not meaningful to split the poems into old and recent works. He accepts the Heuslerian division into two basic groups of Eddic heroic poems, but for him, they are not old versus young but rather “male-orientated poems” versus “female-orientated” (Gísli Sigurðsson 1990, 252). He believes the two groups to be fundamentally different.
It is not necessary to accept everything in Gísli’s theory. His strict Parry-Lord view on the non-memorized character of oral poetry has not received strong support by other researchers. Nor do the poems in the two designated groups seem as profoundly different as Gísli, like Heusler, argues. The theory that the differences between the poems might have derived from different audiences has no support in the sources; the only medieval description of an actual oral performance of an “Eddic elegy,” Norna-Gests þáttr (ed. Guðni Jónsson 1.326), describes a male audience. But the idea that the poems Heusler identified as “elegies” really do have something in common—their focus on women—appears to be basically correct. Here Gísli Sigurðsson seems to have come close to a very important conclusion.
Guðrúnarhvöt and Hamðismál depict the same episode from heroic legend, the events following Svanhildr’s death. Moreover, in their opening sections, they depict the same scene: Guðrún’s whetting. In both poems, grief occurs in this scene. Both poems show Guðrún lamenting openly (Hm 4–5; Ghv 11, 16–17, 20); she is associated with weeping in both (Hm 8–10; Ghv 9); both emphasize her grief in the present as well as the past (Hm 5, 8, 10; Ghv 1, 5, 8), and both problematize her emotions over her murder of her own children (Hm 8; Ghv 5). But Guðrún is the only character who laments openly, and this is also common to both poems. Her sons, the men in the two poems, have lost their sister and are aware of their imminent death, but they neither lament nor show open grief in either poem. On the contrary, their belligerence is marked in both poems (Hm 11, 16; Ghv 6) as well as their hard, stoic attitude toward death (Hm 10–11; Ghv 6–8). There is no difference between the two poems in the general reaction to tragic events or in the expression of grief; rather the difference is found within the poems, between men and women. The woman is “elegiac,” a mourner, in the “old” Hamðismál, and the men are terse and stoic in a supposedly “old” way in the “elegy” Guðrúnarhvöt. Scholars of the Heuslerian school generally believe that the characters in heroic legend were re-shaped, “umgestaltet,” in the “elegies” in contrast to the “old” Eddic poems. Sprenger has repeatedly referred to an “Umgestaltung heroischer Figuren” [reshaping of heroic figures] (the title of Sprenger 1985a) and has argued that the “old” poetry portrays these characters as avengers who never mourn, while the “elegies” depict them as grieving and lamenting (Sprenger 1988, 245–247, 259; 1992, 224). She identifies representatives of the two supposedly contrasting types of poems. Typical of the heroic figure of the “old” poems is Hamðir (Sprenger 1988, 246) while the typical figure of the so-called young “elegies” is Guðrún in Guðrúnarhvöt (247, 258, etc.). The same view is found among several other scholars (e.g., Heusler 1934, 12; von See 1981, 258; Dahlstedt 1962, 31, 34, 39; Andersson 1988, 293). But the shared scene in Guðrúnarhvöt and Hamðismál contradicts this view. The Guðrún character has not been “umgestaltet” [reshaped], and neither has Hamðir nor Sörli. That Hamðismál is experienced as emotionally harsher and more heroic and Guðrúnarhvöt as more emotional is because, after the departure of Hamðir and Sörli, Hamðismál chooses to follow the men while Guðrúnarhvöt stays with the woman. In these continuations, men and woman maintain the same roles as in the first scene, but the one-sided focus on the men in Hamðismál’s continuation and on the woman in Guðrúnarhvöt’s continuation colors our reading of the poem. Thus, one appears to be harsh-heroic and the other emotional-elegiac, a difference resulting from the focus on men and a woman, respectively. Hamðismál shows how a woman who is prominent in a supposedly old Eddic poem is associated with grief and lament there too, exactly as in the “elegies.” In the same way, Guðrúnarhvöt is an “elegy” that is not generally more focused on the characters’ emotions.
Gender differentiation is also evident in Brot af Sigurðarkviðu, another of Heusler’s “old” Eddic poems. Sigurðr’s death affects both men and women in the poems, but the only one who laments openly and even weeps is a woman, Brynhildr (sts. 14, 15). Certainly there are cases where women in the Edda are active and personally commit acts of violence, but they are exceptions. These events occur in heroic legend and are mentioned in both “elegies” and “old” poems (Guðrún’s slaying of Atli and her sons are mentioned in Atlakviða, Atlamál, Guðrúnarkviða II, Guðrúnarhvöt, Sigurðarkviða in skamma, and Hamðismál; i.e. two “old” poems and four “young” ones). Certainly there is also evidence of female emotional harshness in the “old” group (as we would expect from Heusler’s and Sprenger’s arguments); there is, however, only one real case, Atlakviða, and it is a rare exception. The usual and primary role of the woman is that of mourner, in all Eddic heroic poems, and in both Eddic and Old English heroic poetry.
That brings us close to a possible solution to the problem of the distinctive character of the “Eddic elegies.” They are not fundamentally different from the other Eddic poems. There is no reason to regard them as belonging to a different tradition or to ascribe to them a different origin or literary context. They do have a partially distinctive character within the group of Eddic poetry by virtue of their greater focus on grief and emotions, but this does not in itself seem to be a feature that constitutes a subgenre, as it is secondary to their focus on women. That is what distinguishes the “Eddic elegies,” and to some extent it makes them a separate group within the category of Eddic heroic poetry. “Eddic elegies” depict the violent events of heroic legend by focusing on the women who were affected by them. This takes place in an indigenous Norse-Germanic heroic tradition.
*
The twentieth century gave us a new understanding of oral poetry. The new knowledge problematizes the traditional philological view of the Eddic poems as composed in a fixed form at a certain moment and then memorized until the time they were written down. Hardly any scholars would claim today that we have access to or can reconstruct an original version of an Eddic poem. This is important for assessing the age and date of the Eddic poems and what can be dated. Gísli Sigurðsson has argued from Parry’s and Lord’s Oral Formulaic theory that there are no “young” or “old” Eddic poems. Although not everyone has accepted that the Parry-Lord theory is fully applicable to Norse poetry, it has influenced Old Norse scholarship in general. Most scholars accept that the oral Eddic poems have a fundamentally more unstable character than later written poetry. Eddic poems that originated as oral compositions during the Viking Age must necessarily have changed significantly during the centuries of transmission through large numbers of singers.
Lars Lönnroth has argued for such an origin of the perhaps most characteristic “Eddic elegy” of all, Guðrúnarkviða I. He accepts the arguments presented by Joseph Harris and by me for elegy as an essentially old phenomenon in Norse poetry. Formulas such as “sit over a dead person” and “strike the hands together” bear witness, he argues, to an old Germanic tradition of grief poetry. But he thinks that through the centuries, the old Eddic poem incorporated new features, from later times and foreign literary traditions. He accepts some of Sprenger’s parallels in religious literature and also suggests borrowings from courtly poetry (Lönnroth 2001, 116–119). For him, Guðrúnarkviða I is not young or old: “It is neither or both” (Lönnroth 125).
Generally, it is a reasonable approach to regard Eddic poems, in their character of oral poems, as so open to re-shaping that much must have been added since the poem was first composed. Thus, the evidential value of individual elements as a means for determining the age and origin is significantly reduced. However, it is not impossible that the Eddic poems had a more stable shape than Lönnroth, following Parry and Lord, believes, even if they were open to re-shaping during transmission. This makes it tempting to try to re-date the “Eddic elegies,” since so many of the arguments put forward for their late date were proved problematic or unsustainable. But the uncertainties are so great that such an attempt does not seem necessary. The fact that the poems are based primarily on indigenous tradition can, of course, entail that they themselves are old, but the genre they belong to could also be relatively stable and productive over a long period. Skaldic court poetry is an example of that. This type of poetry survived relatively unchanged for four hundred years, and similarly the individual “Eddic elegies” might have been composed late but within a conservative tradition (cf. Sävborg 1997, 452–453). Yet the uncertainty of dating the “elegies” applies equally to Eddic poetry in general, and thus also to the poems traditionally regarded as old. Perhaps it is most reasonable to conclude that the heroic poems of the Edda, both “elegies” and others, have a long history, but that they were composed over a long period and underwent several changes during oral transmission.
The issue of dating is not as crucial as Heusler and his followers contended; in this I think Gísli Sigurðsson is right. The important thing is that the fundamental characteristics of the “Eddic elegies” are part of an indigenous tradition and do not differ essentially from other Eddic poems. They are not a foreign phenomenon in Norse literature or Germanic heroic poetry.
1. E.g., Grimm 1829, 4, 12, 365–355; Ettmüller 1847, 74; Jessen 1871, 40, 50, 60; Mogk 1904, 639, 643, 646, 656; Edzardi 1878, 337; Vigfusson and York Powell 1883, 168, 306, 348; Finnur Jónsson 1920, 67.
2. One case is when Klaus von See, contra Heusler, attempts to move Hamðismál from the group of “old” poems to the youngest layer (de Vries 1981, 258, etc.). But his argument for this is entirely Heuslerian: he claims that Hamðismál has a peaceful and emotionally open character and thus differs from the tone that, according to Heusler, characterizes the old Eddic poems.
3. Cf. e.g., Jonsson 1991a, 155, 164–166, etc. For critical analyses of Mohr, see Sävborg 1997, 440–444; 2000, 60–61; Vésteinn Ólason 2005, 178–183.
4. The similarities in the depiction of emotion in Sonatorrek and the “Eddic elegies” are generally remarkable. One of the most famous and distinctive features of the “Eddic elegies” is their sensibility, “Innerlichkeit” in Heusler’s words (1941, 183). The emotional complexity they depict and their “long speeches highlighting the psychological reactions of the characters” have been identified as traits that can only be explained by influence from high medieval continental literature (Vésteinn Ólason 2005, 187). But all of this is characteristic also of Egill’s indigenous Viking-Age poem.
Amelung, Arthur, and Oskar Jänicke, eds. Ortnit. Deutsches Heldenbuch 3: Ortnit und die Wolfdietriche. Zürich: Weidmann, 1968.
Atkinson Jenkins, T., ed. La Chanson de Roland: Oxford Version. Rev. ed. Boston: D. C. Heath, 1924.
Bartsch, Karl, ed. Diu Klage. Leipzig: F. A. Brockhaus, 1875.
——, ed. Das Nibelungenlied. 16th ed. Helmut de Boor. Wiesbaden: Brockhaus, 1961.
Bjarni Aðalbjarnarson, ed. Heimskringla. I. Íslenzk fornrit 26. Reykjavík: Hið íslenzka fornritafélag, 1941.
Guðni Jónsson, ed. Norna-Gests þáttr. Fornaldar sögur Norðurlanda. 4 vols. Reykjavík: Íslendingasagnaútgáfan, 1954. 1: 305–335.
Grundtvig, Svend, et al., eds. Danmarks gamle Folkeviser. 13 vols. Copenhagen: Samfundet til den danske literatures fremme, 1853–1976. (DgF)
Finnur Jónsson, ed. Rímnasafn: Samling af de ældste islandske rimer. I. Copenhagen: S. L. Møller, 1905–1912.
——, ed. Den norsk-islandske skjaldedigtning. B: Rettet tekst I–II. Copenhagen: Gyldendal, 1912–1915. (Skj)
Krapp, George Philip, ed. Genesis. The Junius Manuscript. The Anglo-Saxon Poetic Records 1. New York: Columbia UP, 1931.
——, and Elliott van Kirk Dobbie, eds. The Exeter Book. The Anglo-Saxon Poetic Records 3. New York: Columbia UP, 1936.
Martin, Ernst, ed. Dietrichs Flucht. Deutsches Heldenbuch 2: Alpharts Tod, Dietrichs Flucht, Rabenschlacht. 2nd ed. Zürich: Weidmann, 1967.
Mazon, Paul, ed. Homère, Iliade. Paris: Belles Lettres, 1955–1956.
Neckel, Gustav, ed. Edda: Die Lieder des Codex Regius nebst verwandten Denkmälern. 4th ed. rev. Hans Kuhn. Heidelberg: Carl Winter, 1962.
Wrenn, C. L., ed. Beowulf. With the Finnesburg Fragment. Rev. ed. London: Harrap, 1958.
Alexiou, Margaret. The Ritual Lament in Greek Tradition. Cambridge: Cambridge UP, 1974.
Andersson, Theodore M. “Sigurðarkviða in skamma.” Dictionary of the Middle Ages. Ed. Joseph R. Strayer. New York: Scribner, 1988. 11: 293.
Beck, Heinrich. “Guðrúnarkviða I-III.” Kindlers neues Literatur Lexikon. 18. München: Kindler, 1992. 677–679.
Clover, Carol J. “Hildigunnr’s lament.” Structure and Meaning in Old Norse Literature: New Approaches to Textual Analysis and Literary Criticism. Ed. John Lindow et al. Islandica 45. Ithaca, NY: Cornell UP, 1986.
Dahlstedt, Karl-Hampus. “Gudruns sorg: Stilstudier över ett eddamotiv.” Scripta Islandica 13. Uppsala: Almqvist & Wiksell, 1962. 25–47.
de Vries, Jan. Altnordische Literaturgeschichte. 2 vols. Grundriss der germanischen Philologie 15. Berlin: de Gruyter, 1964.
Dronke, Ursula. The Poetic Edda: Heroic Poems. I. Oxford: Clarendon, 1969.
Edzardi, A. Kleine Beiträge zur Geschichte und Erklärung der Eddalieder. Germania XI/XXIII. Vienna: Gerhold, 1878.
Ettmüller, Ludwig. Handbuch der deutschen Literaturgeschichte von den ältesten bis auf die neuesten Zeiten, mit Einschluß der angelsächsischen, altskandinavischen und mittelniederländischen Schriftwerke. Leipzig: Verlagsbureau, 1847.
Finnur Jónsson. Den oldnorske og oldislandske literaturs historie. I. 2nd ed. Copenhagen: Gad, 1920.
Gering, Hugo. Kommentar: Heldenlieder. Die Lieder der Edda 3. Halle: Buchhandlung des Waisenhauses, 1931.
Gísli Sigurðsson. “Ástir og útsaumur: Umhverfiog kvenleg einkenni hetjukvæða Eddu.” Skírnir: Tímarit hins íslenska bókmenntafélags 160 (1986): 126–152.
——. “On the Classification of Eddic Heroic Poetry in View of the Oral Theory.” The Seventh International Saga Conference: Poetry in the Scandinavian Middle Ages, Spoleto, 4–10 September 1988. Spoleto: Presso la sede del Centro studi, 1990.
Grimm, Wilhelm. Die deutsche Heldensage. Göttingen: Dietrichsche buchhand-lung, 1829.
Harris, Joseph. “Eddic Poetry.” Dictionary of the Middle Ages. Ed. Joseph R. Strayer. New York: Scribner, 1984. 4: 385–392.
——. “Hadubrand’s Lament: On the Origin and Age of Elegy in Germanic.” Heldensage und Heldendichtung im Germanischen. Ed. Heinrich Beck. Ergänzungzbände zum Reallexikon der Germanischen Altertumskunde 2. Berlin: de Gruyter, 1988.
Heusler, Andreas. “Der dialog in der altgermanischen erzählenden dichtung.” Zeitschrift für deutsches Altertum und deutsche Literatur 46 (1902): 189–284.
——. “Heimat und Alter der eddischen Gedichte.” Archiv für das Studium der neueren Sprachen und Literaturen 116 (1906): 249–281.
——. [Preface to:] Edda. 1: Heldendichtung. Tr. Felix Genzmer. Jena: E. Died-erichs, 1934.
——. Die altgermanische Dichtung. 2nd ed. Potsdam: Akademische Verlagsgesellschaft Athenaion, 1941.
Holtsmark, Anne. [Notes to:] Paasche, Fredrik, Norges og Islands litteratur inntil utgangen av Middelalderen. Norsk litteratur historie 1. 2nd ed. Oslo: Aschehoug, 1957.
Jessen, Edwin. “Über die Eddalieder: Heimat, alter, charakter.” Zeitschrift für deutsche Philologie 3 (1871): 1–84, 251–252, 494.
Jonas Kristjánsson. Eddas and Sagas: Iceland’s Medieval Literature. Tr. Peter Foote. Reykjavík: Hið íslenska bókmenntafélag, 1988.
Jonsson, Bengt R. “Bråvalla och Lena: Kring balladen SMB 56.” Sumlen: Årsbok för vis- och folkmusikforskning 1989. Stockholm: Samfundet för visforskning, 1991. 49–166.
Leicher, Richard. Die Totenklage in der deutschen Epik von der ältesten Zeit bis zur Nibelungenklage. Germanistische Abhandlungen 58. Breslau: M. and H. Marcus, 1927.
Lönnroth, Lars. “Heroine in Grief: The Old Norse Development of a Germanic Theme.” Inclinate Aurem. Oral Perspectives on Early European Verbal Culture. A Symposium. Ed. Jan Helldén et al. Odense: Odense UP, 2001. 111–127.
Mogk, Eugen. Geschichte der norwegisch-isländischen Literatur. 2nd ed. Strassburg: Trübner, 1904.
Mohr, Wolfgang. “Entstehungsgeschichte und Heimat der jüngeren Eddalieder südgermanischen Stoffes.” Zeitschrift für deutsches Altertum und deutsche Literatur 75 (1938): 217–280.
——. “Wortschatz und Motive der jüngeren Eddalieder mit südgermanischem Stoff.” Zeitschrift für deutsches Altertum und deutsche Literatur 76 (1939): 149–217.
Neckel, Gustav. Beiträge zur Eddaforschung mit Exkursen zur Heldensage. Dortmund: Ruhfus, 1908.
——. “Über eine altgermanische Geste des Schmerzes.” Archiv für das Studium der neueren Sprachen 167 (1935): 63–64.
Sävborg, Daniel. Sorg och elegi i Eddans hjältediktning. Acta Universitatis Stock-holmiensis, Stockholm Studies in History of Literature 36. Stockholm: Almqvist & Wiksell, 1997.
——. “Filologi och litteraturvetenskap inom eddaforskningen: Konflikter, kombinationer och uppslag.” Den fornnordiska texten i filologisk och litteraturvetenskaplig belysning. Ed. Kristinn Jóhannesson, Karl G. Johansson, and Lars Lönnroth. Göteborg: Litteraturvetenskapliga institutionen, 2000. 51–63.
——. “Om eddadikternas ursprung och ålder: Gamla och nya tankar.” Arkiv för nordisk filologi 119 (2004): 55–104.
Schröder, Franz Rolf. “Die Eingangsszene von Guðrúnarhvöt und Hamðismál.” Beiträge zur Geschichte der deutschen Sprache und Literatur 98 (1976): 430–436.
Sprenger, Ulrike. “Heroische Elegie und geistliche Literatur.” Akten der fünften Arbeitstagung der Skandinavisten des deutschen Sprachgebiets: 16–22 August 1981 in Kungälv. Ed. Heiko Uecker. St. Augustin: Dr. Bernd Kretschmar, 1983.
——. “Die Umgestaltung heroischer Figuren unter dem Einfluß geistlicher Literatur.” The Sixth International Saga Conference 28/7–2/8 1985: Workshop Papers. København. 1985.
——. “Zum Ursprung der altnordischen Heroischen Elegie.” Heldensage und Heldendichtung im Germanischen. Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 2. Berlin: de Gruyter, 1988.
——. Die altnordische heroische Elegie. Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 6. Berlin: de Gruyter, 1992.
Turville-Petre, Gabriel. Nine Norse Studies. London: Viking Society, 1972.
Vésteinn Ólason. “Heusler and the Dating of Eddic Poetry—With Special Reference to ‘isländische Nachblüte der Heldendichtung’.” Germanentum im Fin de siècle. Wissenschaftsgeschichtliche Studien zum Werk Andreas Heuslers. Ed. Jürg Glauser and Julia Zernack. Basel: Schwabe, 2005. 165–193.
Vigfusson, Gudbrand, and F. York Powell. Corpvs Poeticvm Boreale: The Poetry of the Old Northern Tongue from the Earliest Times to the Thirteenth Century. 1: Eddic Poetry. Oxford: Clarendon, 1883.
von See, Klaus. Edda, Saga, Skaldendichtung: Aufsätze zur skandinavischen Literatur des Mittelalters. Heidelberg: Winter, 1981.
Zeller, Rose. Die Gudrunlieder der Edda. Tübinger germanistische Arbeiten 26. Stuttgart, Berlin: W. Kohlhammer, 1939.