The Elder Edda

A Book of Viking Lore

Translated with Introduction and Notes by
ANDY ORCHARD

PENGUIN BOOKS

PENGUIN CLASSICS

Published by the Penguin Group

Penguin Books Ltd, 80 Strand, London WC2R 0RL, England

Penguin Group (USA) Inc., 375 Hudson Street, New York, New York 10014, USA

Penguin Group (Canada), 90 Eglinton Avenue East, Suite 700, Toronto, Ontario, Canada M4P 2Y3 (a division of Pearson Penguin Canada Inc.)

Penguin Ireland, 25 St Stephen’s Green, Dublin 2, Ireland
(a division of Penguin Books Ltd)

Penguin Group (Australia), 250 Camberwell Road, Camberwell, Victoria 3124, Australia
(a division of Pearson Australia Group Pty Ltd)

Penguin Books India Pvt Ltd, 11 Community Centre, Panchsheel Park, New Delhi – 110 017, India

Penguin Group (NZ), 67 Apollo Drive, Rosedale, Auckland 0632, New Zealand
(a division of Pearson New Zealand Ltd)

Penguin Books (South Africa) (Pty) Ltd, 24 Sturdee Avenue, Rosebank, Johannesburg 2196, South Africa

Penguin Books Ltd, Registered Offices: 80 Strand, London WC2R 0RL, England

www.penguin.com

This translation first published in Penguin Classics 2011

The moral right of the translator has been asserted

Except in the United States of America, this book is sold subject to the condition that it shall not, by way of trade or otherwise, be lent, re-sold, hired out, or otherwise circulated without the publisher’s prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser

ISBN: 978-0-14-194347-3

Acknowledgements

Chronology

Introduction

Acknowledgements

My first debt, appropriately enough for a book that deals so much in myth and speculation, is to a man I never met, Roger Lancelyn Green, whose Myths of the Norsemen I read at the age of eleven and immediately resolved both to become a medievalist and to visit Iceland; the latter I have been doing for nearly thirty years now, and the former I still aspire to. As a shy but somehow surly and hitherto largely self-taught undergraduate, I was I now realize my own worst nightmare to teach, and yet it was my great luck (and certainly not theirs) to sit at the feet of the likes of Ursula Dronke, Michael Lapidge, Ray Page, Sverrir Tómasson and Maureen Thompson, where I learned much and very likely should have learnt very much more. But I did also (and do!) pay close attention to the superb work of contemporaries at Oxford and Cambridge much more gifted in many areas of Old Norse–Icelandic studies than I, including Paul Bibire, Vicky Cribb, Matthew Driscoll, Mike Fox, Carolyne Larrington, Guðrún Nordal, Richard North, Peter Robinson and Clive Tolley. I am grateful in various and different ways to them all. In addition, I have learnt much from those I sometime tried to teach, and it has been wondrous to share a classroom or a supervision or just the odd pint or three with the likes of Chris Abram, Shami Ghosh, Jonny Grove, Guðrún Edda Þórhannessdóttir, Alaric Hall, Roberta Hamilton, Paul Langeslag, Emily Lethbridge, Marteinn Helgi Sigurðsson, Ralph O’Connor, Helle Falcher Petersen, Emily Thornbury and Al Vining; Shami and Marteinn and Jonny I should single out for specific and invaluable help. I would also like to apologize to the successive generations of students in Cambridge and Toronto that I subjected to earlier versions of these translations, as I tried to make the words sound less clumsy than they sometimes seemed.

There are, again appropriately, giants in this field, even beyond those named already, and I would urge anyone with a serious interest in the material to consult anything and everything written by several scholars it has been (largely through the good offices of the above named, though serendipity played its customary part) my privilege and pleasure to meet and even to discuss these and other related texts: especially Paul Acker, Fred Biggs, Bob Bjork, Anthony Faulkes, Chris Fell, Alison Finlay, Peter Foote, Roberta Frank, Tom Hall, Joe Harris, Tom Hill, Judy Jesch, David and Ian McDougall, John McKinnell, Rory McTurk, Heather O’Donoghue, Russell Poole, Judy Quinn, Margaret Clunies Ross, Tom Shippey, Andrew Wawn, David Wilson and Charlie Wright; they won’t need me to tell them that the mistakes and infelicities are my own. I am deeply grateful to everyone at Penguin for their patience and professionalism, and especially to Lindeth Vasey for her eagle-eyed editing, superb suggestions and extraordinary expertise. And then there are the folks at Mullins, who mainly made me welcome and left me alone, even if they occasionally wondered what I was up to. I give my warmest thanks to all.

Translation, especially of poetry, is a fraught business, a masking of another’s words, and an overlaying of often intrusive and unnecessary interpretation. Only the outcome, if it provokes those who know and piques the interest of those who don’t, can be judged successful, and if this effort to transmit some of the wonder I feel every time I read these texts seems to some simply a tissue of errors and an opportunity squandered, then I trust others will step up and fill the gap. Even as I can now see from the scholarly perspective how the simplifications, mystifications and misunderstandings of a well-educated fan like that man I never met can muddy the waters but still inspire, I do hope that these pages will provoke others to go beyond what I could. After all, these poems have been read and reread and remembered for centuries, and for reason: they can speak to us still.

For Steve, best of brothers and first of friends

Chronology

117 Death of Tacitus, who wrote Germania (Germany and Her People)

376 Death of Ermanaric the Goth (who appears here as Jörmunrekk)

453 Death of Attila the Hun (as Atli Budlason)

526 Death of Theoderic the Ostrogoth (as Thjódrek)

584 Death of Chilperic the Neustrian (as Hjálprek)

787–95/6 Paul the Deacon writes the Historia Langobardorum (History of the Lombards)

793 Viking raid on the monastery at Lindisfarne

c. 850 Beginnings of Norse settlement in Anglo-Saxon England; Bragi Boddason composes Ragnarsdrápa (‘Poem for Ragnar’)

c. 870 Beginnings of Norse settlement in Iceland (traditional date is 874)

871 Accession of Alfred the Great to the throne of Wessex

c. 885 Harold Fairhair becomes king of all Norway

c. 885–c. 920 Þjóðólfr of Hvinir composes Haustlöng (‘Autumnlong’)

899 Death of Alfred the Great

c. 900 Þorbjörn hornklofi composes Hrafnsmál (‘The raven’s lay’) for Harald Fairhair

930 Foundation of the national parliament (Althing) in Iceland

954 Death of Eirík Bloodaxe; composition of Eiríksmál (‘Poem for Eirík’)

961 Death of Hákon the Good fighting the sons of Eirík Bloodaxe; composition of Hákonarmál (‘Poem for Hákon’) by Eyvíndr Finnsson skáldaspillir

c. 980 Egill Skallagrímsson composes Sonatorrek (‘Painful loss of sons’)

982 Beginnings of Norse settlement in Greenland

c. 985 Úlfr Uggason composes Húsdrápa (‘House-poem’)

986 Beginnings of Norse settlement in North America

995 Ólafr Tryggvason becomes king of Norway

999–1000 Christianity accepted in Iceland at Althing

c. 1000 The Beowulf manuscript is written

1010 Burning of Njál

1014 Battle of Clontarf

1030 St Ólafr killed at the Battle of Stiklastadir

1056 First bishop at Skálholt, South Iceland Sæmundr Sigfússon inn fróði (‘the wise’) is born

c. 1065 Arnórr jarlaskáld composes Þorfinnsdrápa (‘Thorfinn’s poem’)

1066 Battle of Stamford Bridge; death of Harald Harðráði Battle of Hastings; death of Harold Godwinsson

1067 Ari Þorgilsson inn fróði is born

c. 1085 Death of Adam of Bremen, author of Gesta Hamaburgensis ecclesiae pontificum (History of the Bishops of the Church at Hamburg)

1106 First bishop at Hólar, north Iceland

c. 1130 Ari compiles Íslendingabók (Book of Icelanders)

1130–36 Harald Gilli (Harald IV), king of Norway

1133 First monastery is founded in Iceland, at Thingeyrar Sæmundr inn fróði dies

1136 Death of King Harald Gilli of Norway

c. 1175–1250/1 Þiðreks saga (The saga of Thidrek) written

1178/9 Snorri Sturluson is born

1185–1216 Saxo Grammaticus composes Gesta Danorum (History of the Danes)

c. 1210–59 Óláfr Þórðarson, the grammarian, active

1215–18 Snorri is Lawspeaker

1218/19 Death of Gunnlaugr Leifsson, author of Merlínusspá (‘The prophecy of Merlin’)

c. 1220 Snorri composes the Prose Edda

1222–31 Snorri is Lawspeaker for a second time

1241 Snorri is killed at his home Reykholt

c. 1250 Laxdæla saga (The saga of the people of Laxdale) is written

1259 Death of Óláfr Þórðarson, author of the Third Grammatical Treatise

c. 1260–70 Völsung Saga is written

1262 Icelandic Commonwealth ends; Iceland is ceded to the king of Norway

c. 1270 Codex Regius is written

c. 1275 Gísla saga (The saga of Gísli) is written

c. 1280 Njáls saga (Saga of Njáll) is written

1294 Haukr Erlendsson is Lawspeaker

c. 1300 AM 748 Ia 4to written

1302–10 Hauksbók (AM 544 4to) is written

1334 Death of Haukr Erlendsson, the main scribe of Hauksbók (Haukr’s Book)

c. 1350 Codex Wormianus of Snorri’s Prose Edda (AM 242 fol.) is written

1387–94 Flateyjarbók (The Book of Flatey [Flat-island]) is written

1643 Bishop Brynjólfur Sveinsson of Skálholt acquires the Codex Regius

1662 Codex Regius is given to the Danish king

1665 First edition of Völuspá and Hávamál by Peder Hansen Resen, with Latin translations

1787 Arnamagnaean Institute begins to publish editions of the Codex Regius poems with Latin translations

1797 Amos Cottle publishes the first English translation of the Codex Regius poems

1944 Iceland regains independence

1971 Codex Regius is returned by Denmark to Iceland

Introduction

Iceland owns one of the literary wonders of the medieval world. Little to look at, there survives in Reykjavík today a plain manuscript of only 45 brown and heavily written leaves, each measuring around 19 × 13 cm, more or less the same size as the pages you are holding now, bound as a book. It is called variously the Codex Regius (‘Royal manuscript’), the Elder Edda, the Poetic Edda and Sæmundar Edda (‘Sæmund’s Edda’). The traditional and apocryphal association of the manuscript with the Icelander Sæmundr Sigfússon (1056–1133), also known as Sæmundr inn fróði (‘the wise’), a priest who is said to have studied in Paris, again marks a clear desire to authenticate its contents as not only old, but worthy of sustained and serious study, sanctioned by the most celebrated and learned of early Christian Icelanders. Most of these titles likewise distinguish the Codex Regius from a composite treatise on similarly mythological and heroic themes by another learned and politically active Icelander, Snorri Sturluson (1178/9–1241), that also survives in several copies (predictably enough as the Younger Edda, the Prose Edda and Snorra Edda (‘Snorri’s Edda’)). Although the term ‘edda’ appears first to have been coined with regard to Snorri’s compilation, given the extent to which he seems to have employed many of the same poems and even prose texts found in the Codex Regius, albeit in sometimes different forms, it seems appropriate to designate the collection presented here by the more traditional name Elder Edda, even if the manuscript of the Codex Regius itself was clearly written roughly a generation after Snorri’s death. The use of Elder Edda has the further advantage of reminding readers that the book contains authentic echoes of an age already aged when the words were written down.

The manuscript, which now carries the shelf mark of the Royal Library in Copenhagen where it once was held (as GkS 2635 4to), has over time suffered loss, misreading and botched repair, much like its texts, largely in verse, but also with connecting prose passages, and originally seems to have had 106 pages, rather than the 90 that now survive. Loss of probably eight leaves after folio 32 (the so-called ‘great lacuna’) has left a gap in the narrative of the life of the mighty hero Sigurd the Völsung that can partly be filled by the surviving mid-thirteenth-century Völsunga saga, which, as we shall see, itself bears witness to many of the same poems preserved in the Codex Regius. It seems feasible that the missing leaves offered, among other things, the further wisdom of a valkyrie (one of the supernatural warrior-women who picked the best of the dead warriors for the service of the god Odin) given to her chosen hero, as well as an account of a heroic wooing of a matchless man and a marvellous maid; it is possible that apart from the ending of one damaged poem and the beginning of another, the great lacuna contained the longest single account of Sigurd’s exploits. As such, the self-contained booklet may not so much have been simply lost as removed for private perusal by an over-zealous reader. A naturally occurring hole in the stretched vellum of folio 41, which affects nine lines, is carefully written around by the scribe, and a stitched repair on folio 28 covers most of the width of the page; in such ways one can begin to guess at how much the texts were valued that the modest volume contains.

Each page is densely written, in an often abbreviated fashion that has sometimes compounded the problems of interpretation, but which again may suggest the poems were so well known that readers could reconstruct them without the full wording, and each page contains between thirty-one and thirty-eight lines of text, with the number of the lines increasing and the size of the writing decreasing as the single scribe comes to the end of his stint. As it is, he (and it was almost certainly a man writing) seems to have overcompensated somewhat for the need to finish before the vellum ran out: the last third of the final page is blank. Yet what that unknown scribe wrote in this plain manuscript comprises the bulk of what we now know of viking lore, and how that developed over several centuries. If some of its contents have been traced back to the ninth century, the manuscript seems only to have survived because in 1662 an Icelandic bishop with a keen interest in his country’s past gave it as a gift to a Danish king, then overlord of Iceland. This together with the joyous reception and outpouring of national pride when the Codex Regius came home to Iceland under naval escort from Copenhagen in 1971 say much for the symbolic power of the extraordinary texts that this battered and basic book contains; we can only wish we had more.

The Codex Regius was written around 1270, and has been associated with similar productions from the Benedictine monastery of Thingeyrar, in North Iceland, founded in 1133, a wealthy estate rich in salmon and seals. Indeed, according to recent investigations, seals may have supplied the raw material for at least some medieval Icelandic manuscripts just as sheep, goats and calves did further south. On the first page of the Codex Regius there is the distinctive mark of Bishop Brynjólfur Sveinsson of Skálholt in South Iceland (1605–75): a pair of intertwined ‘L’s to signify Lupus Loricatus (‘Armoured Wolf’), a Latinization of his Norse name. Brynjólfur, whose role as collector of manuscripts for the Danish King Frederick III (1609–70) is well documented, added the date 1643 (presumably the date he acquired it), and called the book ‘Edda’, a name that has since stuck, though its meaning is still debated. ‘Edda’ seems to signify ‘Great-grandma’ in the poem Rígsthula, which is composed in a similar style to many of the texts in the Codex Regius and is found in a single manuscript, the Codex Wormianus (AM 242 fol.), written about 1350, that largely contains the composite prose and poetry text on Norse mythological and legendary lore (also called ‘Edda’) composed by Snorri c. 1220 and now known as the Prose Edda. An attractive alternative suggestion is that ‘edda’ is formed after the model of kredda, which derives from the Latin credo (‘I believe’; also used for the Christian Creed), and would then suggest that ‘edda’ derives from the Latin verb edo (‘I produce’, ‘I publish’), which can have the specific sense ‘I compose (poetry)’. If that argument is accepted, an Edda would be a kind of compilation of traditional poetry, which Snorri’s work certainly is; the application of the term to the Codex Regius would then be an extension, in effect an acknowledgement of the distinct overlap between its poems and those in Snorri’s collection.

The twenty-nine poems (and various prose pieces) to which the Codex Regius is the main or only witness are, then, whether complete, composite or otherwise compromised, compiled from various places and periods, the works of different poets of divergent powers and discrepant beliefs, but they all deal with a common stock of inherited stories that together speak of gods and heroes from an ancient era. They are projections back into the pagan past, yet evidently preserved over the years and in many manuscripts by the skill of Christian scribes long after Iceland’s conversion to Christianity in 999–1000. Some of these pagan gods survive into English in the names of the days of the week (in so far as Tuesday, Wednesday, Thursday and Friday can still be claimed to commemorate Týr the war-god, Odin the psychopomp and god of frenzy, Thor the thunder-god and Freyja the goddess of love and sex), while some of the heroes likewise live on in legend in culture both high and low: opera-goers will readily recognize Siegfried, the direct descendant of the hero Sigurd, and others will easily identify the infamous cruelty of Attila (or Atli) the Hun, who here has his just deserts, unwittingly eating the flesh of his sons before being stabbed and burnt alive in his own hall. These and other poems like them, scattered in other medieval parchments and, after the medieval period, on paper that a combination of chance, careful antiquarianism and often careless copying has allowed to survive, comprise what is known as ‘eddic’ or ‘eddaic’ verse, a term used to distinguish these anonymous texts on mythological and heroic themes from the skaldic poetry, generally on contemporary topics composed in a different metre than those used in the eddic poems (see below) by mainly named and known poets (also known as skalds).

When the eddic texts in this volume were first collected we do not know, though some groups of poems seem to have circulated together in earlier compilations. But as the last of a long line, the single scribe of the Codex Regius was evidently eager to save every scrap of such lore – whether he understood it well or not – to judge by the tidy wee writing and multiple abbreviations he employed; indeed, there are so many of the latter that when the same set of words and phrases recur, as they do often in orally transmitted texts, the scribe simply gave the first letter of each word. In other examples, such as in the brief prose text entitled Frá dauda Sinfjötla (‘About Sinfjötli’s death’), the scribe has carefully left gaps for a character’s name to be filled in when it became available. He did well: we should realize that for every letter that has lasted, and whose sense we can still guess, much more has been lost in other texts, and much of what remains was changed by chance or circumstance; so that in reconstructing whole stanzas and verses from these tiny traces, we run the risk of merely repeating what we know from other sources, but it is simply all we have. Despite these drawbacks, the Codex Regius is a crucial witness to how viking lore developed, as well as to its preservation and alteration long after the viking era was over: it is an eclectic, incomplete and composite collection of often augmented and interpolated poems from various dates and provenances, which individually and in groups have been butchered, badly transmitted and recast elsewhere not only by Snorri in his Prose Edda, but also, for example, by the anonymous author of the Völsunga saga (The saga of the Völsungs). Certainly, these poems were preserved in part for their agedness and authenticity, but also surely for their intrinsic worth as verse, and for the pure pleasure that they still bring. While it would be quite wrong to regard the Codex Regius manuscript as a definitive collection, it clearly represents the last link in a conscious chain of compilation, with some of the poems having been copied in groups from pre-existing gatherings, and with separately-titled prose passages both within and between poems summarizing, introducing or filling gaps. Often the prose, particularly in the heroic section of the manuscript, explains the fate (and usually the death) of major figures, and occasionally provides different perspectives, so helping to connect a narrative from poetry that would otherwise offer an even more disjointed story than it currently conveys.

The first eleven poems in the Codex Regius are all concerned broadly with mythological themes, and seem to have been arranged in clusters: the opening four texts, Völuspá (‘The prophecy of the seeress’), Hávamál (‘The lay of the High One’), Vafthrúdnismál (‘The lay of Vafthrúdnir’) and Grímnismál (‘The lay of Grímnir’), all deal with various aspects of the god Odin as a searcher after and a source of wisdom. Much mythological lore is imparted, largely through a series of conversations and wisdom-contests between Odin and creatures of various kinds, specifically a dead seeress (Völuspá), a wise giant (Vafthrúdnismál) and a young prince (Grímnismál). The clearly composite nature of Hávamál precludes any obvious sense of audience, although the sections commonly known as Loddfáfnismál (‘Loddfáfnir’s lay’), stanzas 111–37, and Rúnatals þattr (‘The episode of the tally of runes’), 138–63, are each based on wisdom-dialogues featuring Odin. The fifth poem, För Skírnis (‘Skírnir’s journey’), again concerns a dialogue between inhabitants of different worlds: Skírnir, the servant of the god Frey, and the giantess Gerd. Five of the next six poems feature the giant-slaying thunder-god Thor, who, along with Odin and Frey, seems to have been the most widely worshipped of the Norse gods. In the first poem of this group, Hárbardsljód (‘Grey-beard’s poem’), Thor and Odin engage in an abusive versified dialogue or flyting, that invites comparison between them, while in the next three Thor is paired with the mischievous and disruptive Loki, both as a travelling companion on quests (in Hymiskvida (‘The song of Hymir’) for a giant cauldron (although earlier commentators and translators have assumed that Thor’s companion was actually Týr: see its headnote); in Thrymskvida (‘The song of Thrym’) for his lost hammer, Mjöllnir); and as antagonist (in Lokasenna (‘Loki’s home-truths’)). In Lokasenna, after Loki has abused many of the gods in an extended formal flyting, Thor returns from a giant-slaying adventure in the East (as he does in Hárbardsljód and Hymiskvida), and terminates the abuse with threats of violence. None of the gods emerges well from the poetic accounts, where Frey dithers, Thor quivers, and even Odin in multi-faceted majesty cuts such a figure that to call him two-faced is to limit the options. ‘[H]ow can his truth be trusted?’ the text says (Hávamál 110), where we can suspect that the poet has already put his faith elsewhere.

The final two poems in the mythological section of the Codex Regius, Völundarkvida (‘The song of Völund’) and Alvíssmál (‘The lay of All-wise’), are often considered to be out of sequence, since while Alvíssmál is a wisdom-dialogue between the rather unintellectual god Thor and All-wise (Alvíss), described as a ‘pale-nosed’ dwarf, Völundarkvida depicts the grim vengeance of a ‘prince’ or ‘leader’ of elves. In terms of the kind of distinction between beings, as well as of the morality represented by the gods, it is notable that male gods sleep freely with giant women (Odin often; Thor, Frey and Njörd on occasion), but goddesses are generally spared such attentions from the giants, with which indeed they are threatened. So, while various giants might covet the sex-goddess Freyja, who is one of the fertility gods, the Vanir, with a record of sexual adventuring, there is no record of her accommodating them. As ever, the giant-born Loki is an exception: he claims in Lokasenna to have cunningly conquered almost every goddess that we know, but he also has had sex with a giant stallion, and his all-embracing appetites are widely condemned in Lokasenna itself.

In the mythological world of the Codex Regius, women are largely scheming and suspect, when they are not simply victims or the objects of unwanted sexual attention. Indeed, even giant-women can be threatened with giant lust, as Gerd is as the unwilling object of Frey’s passion in För Skírnis. In the course of threatening Gerd he condemns her to ergi (here translated as ‘cock-craving’), an insatiable yearning to be sexually penetrated that is beyond the pale. When used, as in För Skírnis 36, of a female, it connotes a wanton lasciviousness that is greatly frowned upon (see further headnote to Lokasenna). By contrast, the heroic poems of the Codex Regius quite regularly present the female perspective, and often with great sympathy and a deep acknowledgement of feminine authority; one can hardly argue that heroines such as Gudrún, Grímhild or Brynhild lack authority, even as they endure deep suffering. Such a contrast underlines a chief concern of the heroic poems of the Codex Regius: the tensions between the family ties of blood and marriage. In fact, the compiler foreshadows the subject: all three of the final mythological poems, Thrymskvida, Völundarkvida and Alvíssmál (not to mention För Skírnis), concern forced, failed or thwarted marriages, a theme that is to become a leitmotif throughout the tragic intertwined tale of the mighty families of the Völsungs, the Gjúkungs and the Niflungs (linked through the doomed marriages of Sigurd the Völsung to Gudrún, the daughter of King Gjúki, and of Gudrún in turn to Atli and Jónakr), which forms the background for the heroic poems of the Codex Regius. Other repeated motifs link the mythological and heroic poems: so, for example, just as many of the first eleven poems concern wisdom-dialogues and predictions between different kinds of creatures, with Odin playing a principal role, so too Sigurd, Odin’s favoured hero among a great family of favourites, is involved with a number of dialogues of wisdom or prophecy with a wide range of beings, including his maternal uncle Grípir (Grípisspá (‘Grípir’s prophecy’)), the disguised god Odin (Reginsmál (‘Regin’s lay’) 16–25), the dying dragon Fáfnir (Fáfnismál (‘Fáfnir’s lay’) 1–22), the wise giant Regin (Fáfnismál 23–27 and Reginsmál 13–14), several talking nuthatches (Fáfnismál 32–44), and the newly woken valkyrie Sigrdrífa (Sigrdrífumál (‘Sigrdrífa’s lay’)).

Nonetheless, a particularly large capital marks the first line of Helgakvida Hundingsbana in fyrri (‘The earlier song of Helgi, the slayer of Hunding’) in the Codex Regius, and so highlights the sensitivity of the compiler to the differences between the mythological and heroic material. Indeed, many of the same themes of love, loyalty and longing for knowledge are shot through both sections. The poems document several phases in the family saga of the doomed Völsung family, beginning with Sigmund, and his three sons by three different liaisons: Sinfjötli, whom he conceived with his own sister, Signý; Helgi, born of Borghild; Sigurd, born of Hjördís. The first three heroic poems in the Codex Regius (Helgakvida Hundingsbana I, Helgakvida Hjörvardssonar (‘The song of Helgi Hjörvardsson’) and Helgakvida Hundingsbana önnur (‘The second song of Helgi, the slayer of Hunding’), together known as the Helgi-lays), conclude with the death of Helgi. An intervening prose passage, Frá dauda Sinfjötla (‘About Sinfjötli’s death’), marks the death of Sinfjötli, as well as the birth of Sigurd, Sigmund’s most famous son, and recounts the death of Sigmund himself.

Grípisspá, the poem that immediately follows, offers a complete prophecy of Sigurd’s life made by his maternal uncle, Grípir: his revenge for his father, Sigmund, and the slaying of the dragon, Fáfnir, and his brother Regin, so acquiring the doomed treasure of the Völsungs; then Sigurd will journey to the court of King Gjúki, encounter an unnamed valkyrie, who will teach him runic wisdom, and travel to the court of King Heimir, where he will meet the lovely Brynhild, daughter of Budli, with whom he will fall madly in love. But the meeting marks a turning-point in Sigurd’s fortunes, since on his return to King Gjúki’s court he will be tricked by Grímhild, Gjúki’s queen, and made to forget Brynhild, to woo her instead for Grímhild’s son, Gunnar (by taking on his appearance), and to marry Grímhild’s daughter, Gudrún, himself. In binding himself by ties of marriage and blood-brotherhood to Gjúki’s family, Sigurd effectively seals his own fate: when Brynhild discovers how she has been tricked and betrayed, she asserts her seniority as Sigurd’s first love, and drives Gudrún and Gunnar, with the help of their brothers Högni and Gutthorm, to contrive Sigurd’s death. Sigurd and Grípir part with the much chastened young hero accepting his fate.

The events described in Grípisspá are duly played out in the succeeding poems, beginning with Reginsmál, which narrates Sigur further education at the hands of the mysterious Regin, described by the prose as ‘handier than any man, and a dwarf in height: he was wise, stern and skilled in magic’. Reginsmál begins by telling Regin’s own family history, and how they acquired a cursed treasure-hoard, which his brother, Fáfnir, now transformed into the shape of a dragon, was guarding. Rather than comply immediately with Regin’s request that he help him kill Fáfnir, Sigurd instead heads off to avenge his own father, and apparently meets up en route with the god Odin in disguise, who offers him still more useful advice. When Sigurd returns successful from the battlefield, he accedes to Regin’s request, and Fáfnismál, which immediately follows and indeed is seamlessly connected with Reginsmál in the Codex Regius, describes Sigurd in conversation with the dying dragon, in what seems yet another form of wisdom-dialogue. Once Fáfnir is dead, Sigurd and Regin engage in a discussion of bravery and cowardice that appears to end with Sigurd spouting two pieces of gnomic wisdom on the value of the former (stanzas 30–31), even as he roasts the dragon’s heart for Regin to eat, illustrated alongside other scenes from the same part of the Sigurd legend on the Ramsund rune-stone (usually dated around 1000) in Södermanland in south-eastern Sweden. The hero tests the spitted heart to see if it is done, burns his thumb on the dragon’s roasted heart-blood, and when he sucks it, finds that he can miraculously understand the language of birds. Several nuthatches (the precise number is debated) perched nearby urge him to kill Regin, which he duly does; they then direct him to a hall surrounded by flames, where a valkyrie disobedient to the wishes of the god Odin slumbers. The account of Sigurd’s encounter with Sigrdrífa, the name given to the valkyrie in both Fáfnismál and Sigrdrífumál, the next poem in the Codex Regius, is told in full in the latter text, where once she has been awakened, the valkyrie offers the hero more runic lore and sound advice very close in spirit to some sections of Hávamál. Unfortunately, Sigrdrífumál breaks off after the valkyrie has given Sigurd only eleven pieces of advice (compare the twenty-one pieces in the Loddfáfnismál section of Hávamál), just before the missing pages.

As far as the ‘great lacuna’ is concerned, a rough calculation suggests that the Codex Regius scribe averages around 100–125 lines of verse per page (written as elsewhere in the manuscript in 31–8 lines of text: the poetry was written continuously as though it were prose to save space). Given differences of metre and levels of abbreviation, this produces on average about 15 stanzas per page, so suggesting that the eight lost folios may have contained roughly 200–250 stanzas, allowing for occasional prose inserts. Given that, if the two poems affected, Sigrdrífumál and Brot af Sigurdarkvidu (‘A fragment of the song of Sigurd’), are excluded, the average length in the Codex Regius is about 50 stanzas, the loss could amount to as many as four complete poems. Later paper manuscripts of Sigrdrífumál help to supply the missing material, with the further clue of four stanzas in Völsunga saga 27–9. The very fact that the Codex Regius also contains a poem explicitly called Sigurdarkvida in skamma (‘The short song of Sigurd’), which at 71 stanzas long is significantly above average length, may encourage us to speculate that a significant portion of the ‘great lacuna’ was made up with the missing ending and beginning (and no poems were lost). When J. R. R. Tolkien sought to repair the loss, he composed two new poems, which he called Völsungakviða en nýja eða Sigurdarkviða en mesta (‘The new song of the Völsungs, or The longest song of Sigurd’) and Guðrúnarkviða en nýja eða Dráp Niflunga (‘The new song of Guðrún, or The killing of the Niflungs’), using the existing poems, and bridging the gap from a combination of his own imagination and the sometimes contradictory narratives of Völsunga saga and Snorri’s Prose Edda. In so doing, Tolkien was effectively aligning himself with the long line of poets who, across many centuries, added or elaborated further links to an existing chain of song. Albeit unwittingly offering him a fine opportunity to demonstrate both his command of the material and his own poetic imagination, whoever was responsible for the missing folios certainly ripped the heart out of the Codex Regius.

Again, there is a change of tone and style in the prose passage (Frá dauda Sigurdar (‘About Sigurd’s death’)) that immediately follows Brot af Sigurdarkvidu. We are told that there are different versions of the killing of Sigurd, and even disagreements about whether he was killed inside or outside; that he was killed while in bed with Gudrún may represent a later tradition that recalls the similar death of Atli, Gudrún’s second husband – a death-scene for Atli (as Attila) at the hands of a Gothic or Germanic bride is attested early (by for example Priscus in the fifth century, and Jordanes and Procopius in the sixth), and on to which the character of Gudrún herself seems to have been grafted. Like so many of the bridging passages in prose, here too we learn of the fate of key characters, so tying up the loose ends and allowing the broader poetic narrative to continue. Certainly, the poems that follow all switch the focus from Sigurd, safely dead, to his young Gjúkung widow, Gudrún, whose three marriages (Sigurd, Atli, Jónakr) echo the three liaisons of Sigmund the Völsung (Signý, Borghild, Hjördís) with which the heroic portion of the Codex Regius began.

Clearly, the focus of the next five poems is squarely on the perspective of the hapless women in Sigurd’s life. Starting with the poignant account of his grieving widow in Gudrúnarkvida in fyrsta (‘The first song of Gudrún’) and the flashback sequence of Sigurdarkvida in skamma, the whole sorry sequence is blamed on fate, the female norns and the feminine wiles of the wounded Brynhild, who in stanzas 53–64 continues the tradition of prophecy, outlining all the hideous events ahead, and then in 65–70 makes requests for her own funeral, concluding in the final stanza with words that seem to echo those of the seeress in Völuspá. Brynhild’s rewritten role (now identified with Sigrdrífa) as a valkyrie and prophetess is more sympathetically presented in Helreid Byrnhildar (‘Brynhild’s Hel-ride’), where her tragic and tainted love for Sigurd is laid bare, before the focus shifts back to Gudrún in Gudrúnarkvida in forna (‘The ancient song of Gudrún’), and forward to her marriage to Atli, Brynhild’s brother, and his own foreboding dreams (stanzas 38–43), that pre-empt (physically, if not necessarily chronologically) the dismal dreams of Atlamál in grœnlenzku (‘The Greelandic lay of Atli’). The next poems in the Codex Regius can best be regarded as bridging passages or later additions to the final two vengeful phases of Gudrún’s life, recounting as they do her alleged adultery (in Gudrúnarkvida in thridja (‘The third song of Gudrún’)) and the role of Oddrún, (in Oddrúnargrátr (‘Oddrún’s lament’)), sister of Atli and lover of Gunnar (who therefore slept with both Brynhild and her sister), in helping Borgný, daughter of King Heidrek, to give birth. Oddrún’s sorry tale, encompassing as it does Gunnar’s death in a snake-pit (stanzas 25–33), looks forward to the two poems that immediately follow, namely Atlakvida (‘Atli’s song’) and Atlamál. The former is brisk and brutal, the latter languid and leisurely; though they may have been composed several centuries apart, they both testify to the fascination of a lurid and inherited tale that they treat in quite different ways.

So Gudrún’s grief is renewed, as it were, for the third and final time in the closing pair of poems in the manuscript. Here Gudrún, whose actions hitherto have been those of wife or sister, becomes in Gudrúnarhvöt (‘Gudrún’s inciting’) and Hamdismál (‘Hamdir’s lay’) Gudrún the vengeful mother, sacrificing one set of children, those she had by Jónakr (Hamdir and Sörli), to avenge another child, by Sigurd (Svanhild). Given that the previous two poems (Atlakvida and Atlamál) document in detail the ways in which Gudrún is prepared to sacrifice her sons for her own concerns, such a redemption is limited, but does at least serve to emphasize the point that Sigurd remains her ‘real’ husband: she will happily sacrifice the sons of her next two husbands to reassert her loyalty both to her own birth-family and to the one she shared with Sigurd. Likewise, Hamdismál looks back to the texts dealing with the death of Sigurd, by echoing the pattern of two close brothers who act out of heroic pride (there Gunnar and Högni; Hamdir and Sörli here), and a third who is somehow seen as inferior or at least disposable (Gutthorm and Erp, respectively). Certainly, these final poems in the Codex Regius, focusing as they do on the plight of the later generations of the children of Gjúki and their heirs (the Gjúkungs), have a force and allure of their own, and appear to have had a particular following; two thirteenth-century sagas from the West of Iceland, Gísla saga (The saga of Gísli) and Laxdæla saga (The saga of the people of Laxdale) both contain traces of poems about the Gjúkungs, even if there seems little else to connect them. In such ways did these texts continue to exert their poetic power.

We may reasonably assume, as we likely should, that the Codex Regius was compiled by a thirteenth-century Icelandic cleric who – because he was aware of his countrymen’s traditional role as collective keeper of Norse lore, because he could appreciate the intrinsic value of texts he could not always fathom or understand, and because he could see that the knowledge of such ancient and traditional texts was dying out – chose to preserve texts that did not always agree with his own beliefs. Nonetheless, that cleric, his patron or whoever made it clear that such things should be saved, evidently had sufficient respect for the material that he assembled the texts in a specific pattern and order, demonstrating to Christian readers the extent to which the ancient world of gods and men was one, and that it ended in death and destruction. There are no great ethical lessons to be learnt from the poems of the Codex Regius, except a stubborn insistence, exemplified by the god Odin and the heroes Helgi, Sigurd, Gunnar and Hamdir, as well as the heroines Gudrún and Brynhild, to fight on in the face of unfriendly fate. Even when they know they are doomed, they enjoy whatever pleasures they may partake of on the way, happy in the sense of being individually gifted and specifically subject to individual doom.

Beyond the Codex Regius, its poems are variously witnessed mostly in fragments and often in strikingly different forms in those manuscripts. The next most important manuscript witnesses to eddic verse are two fragments of six leaves, AM 748 Ia 4to, written around 1300. The first two leaves begin partway through Hárbardsljód (towards the end of stanza 19), then give the full text of Baldrs draumar (‘Baldr’s dreams’; not in the Codex Regius), and the first twenty-seven stanzas of För Skírnis, here called Skírnismál (‘The lay of Skírnir’); the other four leaves witness the final twenty-three stanzas of Vafthrúdnismál, all of Grímnismál and Hymiskvida, and the beginning of the prose introduction to Völundarkvida. We have no way of knowing how much has been lost from this manuscript, but even the two scraps show less conscious organization than the Codex Regius, mixing up primarily Odinic material with that pertaining to other gods, and with Völundarkvida again apparently appended to mythological lore.

A further important witness to eddic poetry is the Hauksbók manuscript (AM 544 4to), written between 1302 and 1310 by Haukr Erlendsson (who became Icelandic Lawspeaker in 1294 and died in 1334), and containing on folios 20r–22r a variant version of Völuspá that was added by another hand after Haukr’s death, likely around the middle of the fourteenth century. Some twenty-eight stanzas of the same poem are found in various manuscripts of the part of Snorri Sturluson’s Prose Edda known as Gylfaginning (‘The beguiling of Gylfi’), as well as single stanzas of Hávamál, Vafthrúdnismál, Grímnismál, Lokasenna, Alvíssmál and Skírnismál. Other verses from the Codex Regius are also attested in various manuscripts of Snorri’s Skáldskaparmál (‘Treatise on the art of poetry’), to which other eddic poems not found in the Codex Regius are also occasionally attached, notably Rígsthula (‘Ríg’s list’) and Grottasöngr (‘Grotti’s chanting’). The late thirteenth-century Völsunga saga bears witness to eleven of the heroic poems of the Codex Regius, found in a single manuscript that was also at one point in the Royal Collection of Copenhagen (Ny Kgl. Sml. 1824 b 4to), written around 1400, and seems to be using Helgakvida Hundingsbana in fyrsta, Gripisspá, Reginsmál, Sigrdrífumál (called ‘The song of Brynhild’ in Völsunga saga), Brot af Sigurdarkvidu, Sigurdarkvida in skamma, Gudrúnarkvida in forna, Atlakvida, Atlamál in grœnlenzku, Gudrúnarhvöt and Hamdismál, as well as the prose passage known as Frá dauda Sinfjötla. From the citations in the saga, it seems clear that its author had access to a manuscript containing the poems in a form very close to but not identical with the Codex Regius. There are also a few citations from Reginsmál and Helreid Brynhildar in the Nornagests páttr (‘Story of Nornagest’) found uniquely in Flateyjarbók (GkS 1005 fol.), a huge manuscript of 225 leaves, written 1387–94, and which likewise uniquely contains Hyndluljód, including the section known as Völuspá in skamma (‘The shorter Völuspá’). Many of the texts discussed here, sometimes in variant forms, were copied into paper manuscripts in the post-medieval period, again attesting to continuing interest in them. Likewise, certain other texts which can broadly be called eddic are either embedded within other texts, or, like Svipdagsmál consisting of Grógaldr and Fjölsvinnsmál, preserved in several paper manuscripts, of which the earliest is from around 1650. In this volume, I include as an appendix four further poems which share so many affinities with those in the Codex Regius that they have regularly been included in editions and translations of eddic verse.

The first published edition of the first two poems in the Codex Regius (Völuspá and Hávamál), together with Latin translations and a version of Snorri’s Edda, was edited by Peder Hansen Introduction à l’histoire du Danemarch où l’on traite de la religion, des moeurs, des lois, et des usages des anciens DanoisMonuments de la mythologie et de la poèsie des Celtes, et particulièrement des anciens ScandinavesReliques of Ancient English PoetryNorthern AntiquitiesFive Pieces of Runic Poetry

The Elder Edda

Contents

Acknowledgements

Chronology

Introduction