Royal Ideology in Early Scandinavia: A Theory Versus the Texts

Margaret Clunies Ross

JEGP, Journal of English and Germanic Philology, Volume 113, Number
1, January 2014, pp. 18-33 (Article)

Published by University of Illinois Press

For additional information about this article

Access provided by Uniwersytet Warszawski (5 Mar 2014 07:06 GMT)

http://muse.jhu.edu/journals/egp/summary/v113/113.1.ross.html

Journal of English and Germanic Philology—January

© 2014 by the Board of Trustees of the University of Illinois

Royal Ideology in Early Scandinavia:

A Theory Versus the Texts

Margaret Clunies Ross, The University of Sydney

INTRODUCTION: THE THEORY

In 1989 the Norwegian historian of religion, Gro Steinsland, successfully
defended her doctoral dissertation at the University of Oslo, and in 1991
published it as a book, with the title Det hellige bryllup og norrøn kongeideologi
(The Sacred Marriage and Norse Royal Ideology)

1

. When they were first

published, Steinsland’s arguments received considerable largely positive
attention,

2

and she has continued to promote them up to the present

time in numerous publications and lectures, the latest presentation being
her plenary lecture to the Fifteenth International Saga Conference in
Aarhus in August 2012.

3

In fact, with the passage of time, a number of

her arguments have grown bolder and less nuanced. For the most part,
scholars appear to have accepted the validity of her theories, and there
have been few who have gainsaid them or offered detailed critiques of
them.

4

More than twenty years after its first publication, the present article

1. The book was published in Oslo by Solum Forlag, with the subtitle En analyse av hierogami-

myten i Skírnismál, Ynglingatal, Háleygjatal og Hyndluljóð (An analysis of the hieros gamos myth

in Skírnismál, Ynglingatal, Háleygjatal and Hyndluljóð).

2. See, for example, the review of Det hellige bryllup in Alvíssmál, 3 (1994), 111–13,

by Carolyne Larrington, which is largely positive, though it does express some cautious

reservations about some aspects of Steinsland’s argument.

3. The lecture, delivered on August 6, 2012, was entitled “Memory, Myth and Rulership.” A

selection of Steinsland’s publications on the topic include “Die mythologische Grundlage für

die nordische Königsideologie,” in Germanische Religionsgeschichte. Quellen und Quellenprobleme,

ed. Heinrich Beck et al. (Berlin: de Gruyter, 1992), pp. 736–51; Den hellige kongen: om religion og

herskermakt fra vikingetid til middelalder (Oslo: Pax, 2000); “Origin Myths and Rulership. From

the Viking Age Ruler to the Ruler of Medieval Historiography: Continuity, Transformations

and Innovations,” in Ideology and power in the Viking and Middle Ages: Scandinavia, Iceland,

Ireland, Orkney and the Faeroes, ed. Gro Steinsland et al. (Leiden, Neth.: Brill, 2011), pp.

15–67; Mytene som skapte Norge: myter og makt fra vikingetid til middelalder (Oslo: Pax, 2012).

4. One of the most thoughtful critiques, which adumbrates a number of the issues addressed

in greater detail in the present article, is Anders Hultgård’s review of Det hellige bryllup in Maal

og minne (1994), 75–80. More recently there has been muted criticism of Steinsland’s theory

as too all-embracing and as ignoring different constructions of the relationship between

humans and gods in medieval Scandinavian traditions; cf. Olof Sundqvist, “‘Religious Ruler

Ideology’ in Pre-Christian Scandinavia. A Contextual Approach,” in More than Mythology.

Narratives, Ritual Practices and Regional Distribution in Pre-Christian Scandinavian Religions, ed.

seeks to examine the textual bases of Steinsland’s hypothesis in some
detail, grounding itself upon some of the new resources made available
by major contemporary research projects that have as their objectives the
better understanding of Old Norse poetry, which is the foundation upon
which her arguments are based. Chief among these new resources are the
extensive set of commentaries to the poems of the Poetic Edda now almost
completed by a research team at the University of Frankfurt and the new
edition of skaldic poetry being published in hard copy by Brepols from
2007 and online by the editors.

5

Available resources for the study of pre-Christian modes of thought
among the early Nordic peoples are largely confined to early vernacular
texts, usually in poetic form, together with material objects like standing
stones and archaeological finds of various kinds, which are considerable
and growing in number all the time, thanks to the vigor of contemporary
archaeology. Both material objects and texts require interpretation and
understanding. The early texts are methodologically problematic, because
for the most part they were recorded in writing long after they were allegedly
composed, and by people who were Christian. Although certain properties
of early Norse poetry, like metre, rhyme, and alliteration, may offer some
guarantee of authenticity, its evidentiary value as witness to earlier cultural
ways of thinking and behaving must be carefully assessed. The context in
which this poetry has been preserved or used is also something that requires
careful evaluation. As is well known, much skaldic poetry is recorded in
prose texts of a later period, whether in historical sources in which the
poetry is quoted as support for the prose author’s narrative, or in sagas with
mixed historical and entertainment objectives. Very frequently, what were
probably original long poems were chopped up by later historians in order
to provide quotable quotes for their narratives. Further, the major part of
our extant textual material was written down in Iceland by Icelanders and

Catharina Raudvere and Jens Peter Schjødt (Lund: Nordic Academic Press, 2012), pp. 225–

61, especially p. 241 and p. 257, n. 60. In my own Prolonged Echoes. Old Norse Myths in Medieval

Northern Society, vol. 1: The Myths (Odense: Odense Univ. Press, 1994), pp. 127–43, especially

p. 127, n. 28 and pp. 135–36, n. 35, I offered a critique of Steinsland’s interpretation of the

poem Skírnismál and the myth of the god Freyr’s wooing of the giantess Gerðr and provided

an alternative analysis that I still stand by and will refer to in the present article, but at that

time I did not engage with her arguments concerning the skaldic poems Ynglingatal and

Háleygjatal, which will be the main focus of the present article. Space precludes a discussion

of Hyndluljóð here, but Steinsland’s use of this poem is secondary to her argument apropos

Skírnismál and the two skaldic poems.

5. Kommentar zu den Liedern der Edda, ed. Klaus von See et al. (Heidelberg: Winter, 1997-);

Skaldic Poetry of the Scandinavian Middle Ages, ed. Margaret Clunies Ross, Kari Ellen Gade,

Guðrún Nordal, Edith Marold, Diana Whaley, and Tarrin Wills, 9 vols. (Turnhout: Brepols

(2007–); see also the skaldic website at www.abdn.ac.uk/skaldic/db.php. This edition is

referred to hereafter as SkP.

Royal Ideology in Early Scandinavia

19

not in Norway, Sweden, or Denmark, where some of the events represented
took place and the poets who composed a good many of the verses came
from. There is good reason, then, to be very circumspect in the use of poetic
resources for the interpretation of modes of thought of the preliterate
period. This makes it all the more important that a theory, such as Gro
Steinsland’s, based on a holistic analysis of early Nordic poems as mythical
structures, is also able to stand up to detailed textual analysis in such a way
as to provide sufficient internal evidence for the theory itself.
The textual material upon which Steinsland’s arguments are based
includes two poems in eddic metres, Skírnismál (The speech of Skírnir) and
Hyndluljóð (The song of Hyndla)

6

and two genealogical poems in the skaldic

metre kviðuháttr, namely Ynglingatal (Enumeration of the Ynglingar) and
Háleygjatal (Enumeration of the Háleygir).

7

While the two eddic poems are

anonymous and difficult to date, the two skaldic poems are attributed to the
late ninth- or early tenth-century skald Þjóðólfr ór Hvini (Ynglingatal) and
the late tenth-century skald Eyvindr skáldaspillir the Plagiarist (?) Finnsson
(Háleygjatal), respectively. The former comprises twenty-seven stanzas,
recorded in manuscripts of the historical compilation Heimskringla (Globe
of the World, ca. 1230), usually attributed to the Icelander Snorri Sturluson
(1179–1241). The latter is now extant as thirteen stanzas in total, recorded
in a number of manuscripts dating from the fourteenth century and later.
Ynglingatal enumerates legendary Swedish and Norwegian rulers from a
certain Fjolnir to Óláfr Geirstaðaálfr (Elf of Geirstaðir) and concludes
with praise of a late ninth-century Norwegian ruler named as Rognvaldr
heiðumhár (High with Honours), about whom little is known. The poem
mentions the often bizarre manner in which these rulers met their deaths
and sometimes includes information on their burial places. Háleygjatal,
which is probably modelled on Ynglingatal, enumerates the ancestors of
Hákon jarl Sigurðarson of Hlaðir (ca. 970–ca. 995), tracing them back to
the god Óðinn and his consort, the giantess Skaði. It also gives an account
of the ways in which the rulers died. Háleygjatal is probably incomplete in

6. The texts of both eddic poems are cited from the standard edition, Edda, Die Lieder

des Codex Regius nebst verwandten Denkmälern, ed. Gustav Neckel, revised Hans Kuhn, 5th ed.

(Heidelberg: Winter, 1983), pp. 69–77 (Skírnismál) and 288–96 (Hyndluljóð). Skírnismál is

recorded in two medieval manuscript compilations, the Codex Regius of the Poetic Edda (GKS

2365 4° of ca. 1270) and (up to stanza 27) AM 748 1 a 4° of ca. 1300–1325. One stanza (42)

is also found in some manuscripts of Snorri Sturluson’s Edda, dating from the fourteenth

century onwards. Hyndluljóð occurs only in the compilation Flateyjarbók, GKS 1005 fol, of

ca. 1387–95. For detailed commentary on these two poems, see Kommentar, ed. Klaus von

See et al. II, 45–151, and III, 667–836.

7. For recent editions and commentary on Ynglingatal and Háleygjatal, see Skaldic Poetry

of the Scandinavian Middle Ages, vol. 1: Poetry from the Kings’ Sagas. 1: From Mythical Times to c.

1035, ed. Diana Whaley, 2 Parts. (Turnhout: Brepols, 2012), Part 1, pp. 3–60 (Ynglingatal,

ed. Edith Marold et al.), and 195–213 (Háleygjatal, ed. Russell Poole).

20

Clunies Ross

the texts we know, and Ynglingatal may be lacking its conclusion and possibly
its beginning.
In assessing the value of the two skaldic poems as evidence of Steinsland’s
or any other theory, it is important to remember that they are editorial
constructs, however thorough and discerning their editors may be. This
is particularly true of Háleygjatal, whose stanzas are scattered through
the prose texts of several manuscripts. Both poems are placed within
prose contexts by medieval historians, and the interpretations offered by
those writers, mostly of the thirteenth century and later, have unavoidably
influenced the understanding of modern scholars, particularly where the
poetic texts are at their most obscure or laconic, as they often are. Thus,
in using such sources, it is methodologically necessary to differentiate what
can be inferred from the poetic sources themselves and what the later
prose sources say about them. It is precisely this failure to differentiate
between different kinds of medieval sources that, in my opinion,

8

vitiates

Steinsland’s interpretation of the myth of Freyr and Gerðr, and the same
can be said of her use of the early skaldic poems Ynglingatal and Háleygjatal
in conjunction with the prose narratives in which they have been embedded
by later medieval historians.
What is new in Steinsland’s approach to her subject is that she replaces
an older theory of sacral kingship among the early Scandinavians,

9

primarily involving Norwegian and Swedish rulers, with one that sees
the essence of the ideology of royal power among them as residing in a
hieros gamos (sacred marriage) between a god and a giantess. This sort of
union was abnormal, according to her understanding of the conventions
of Norse mythology, because it brought together beings from opposing
worlds, namely gods and giantesses. Hence, in Steinsland’s view, any
offspring from such a union would represent a new type of being, one
that combined the qualities of a male deity with those of a giantess, the
former associated with power, authority and control, the latter with the
fruitfulness of the land. This mediating combination of qualities gave
early Scandinavia its ruling dynasties, “and the holy marriage between
the god and the giantess may be interpreted as a symbol of the king’s
intimate relationship with his land.”

10

8. Cf. Clunies Ross, Prolonged Echoes, p. 141.

9. The literature on this subject is extensive, and the concept of “sacral kingship” is itself

problematic, as has recently been discussed by Sundqvist, “‘Religious Ruler Ideology.’” For

a useful review of older theories of sacral kingship, see R. W. McTurk, “Sacral Kingship in

Ancient Scandinavia: A Review of Some Recent Writings,” Saga-Book of the Viking Society, 19

(1974–77), 139–69. Hultgård (review of Det hellige bryllup, p. 79) casts doubt on whether the

concept of a hieros gamos, found in Near Eastern myth, is really applicable to early Scandinavia.

10. Steinsland, Det hellige bryllup, p. 349 (English summary).

Royal Ideology in Early Scandinavia

21

THE PRIMARY EDDIC TEXT: SKÍRNISMÁL

Skírnismál is the foundation of Steinsland’s theory, just as it was the
foundation of Magnus Olsen’s theory, published in 1909,

11

that Skírnismál

was the literary representation of a prehistoric Norse myth of the
sacred marriage between a deity of fruitfulness and light, Freyr, and
the representative of the fertility of crop and field in the person of the
giantess Gerðr. In Steinsland’s view, the marriage between Freyr and the
initially reluctant Gerðr, arranged through the god’s emissary Skírnir,
is a prototype of the hieros gamos that produced the ideal Nordic ruler.
She argues that there are several details of the myth that need to be
interpreted in the light of this sacred marriage; the fact that Freyr directs
operations from a high-seat (mentioned only in the prose introduction
to the poem in the Codex Regius and by Snorri Sturluson in his prose
narrative of the myth)

12

and that Skírnir offers Gerðr, as an inducement

to the marriage, golden apples and a gold ring, the latter a possession
of the god Óðinn, as well as a staff. All these Steinsland sees as the
insignia of royalty, although all can equally be understood as among
the mythic possessions of the gods as a group with rather different
symbolic meanings, and, in two cases (high-seat and ring), objects usually
associated with Óðinn rather than Freyr.
There are two serious problems with Steinsland’s theory as applied to
Skírnismál. The first is that there is absolutely no evidence internal to the
text of the poem that links it to the ideology of royal power in human
society. The apples (stanza 20/1 epli ellifo) are most closely comparable to
the apples described as ellilyf ása (the old-age cure of the gods) that the
goddess Iðunn provides to the gods to keep them young,

13

rather than to

the globes symbolizing royal authority in medieval Europe that Steinsland
links to her theory (Det hellige bryllup, pp. 146–48). Granted, and this is a
point I have argued myself,

14

the corpus of Norse myth as a whole can be

seen to engage with most of the major concerns of early Nordic society,
which we can see expressed through other kinds of texts and through
what we know of Nordic habitation patterns, social organization, and
religion. But this particular text cannot be made to yield up such a precise

11. Magnus Olsen, “Fra gammelnorsk myte og kultus,” Maal og minne (1909), 17–36.

12. Snorri gives a prose account of the myth in the Gylfaginning (Delusion of Gylfi) section

of his Edda (ca. 1225) and quotes the last stanza of Skírnismál at the end of his narrative. For

the Icelandic text, see Snorri Sturluson Edda. Prologue and Gylfaginning, ed. Anthony Faulkes,

2d ed. (University College London: Viking Society for Northern Research, 2005), pp. 30–31.

13. According to Þjóðólfr ór Hvini, Haustlöng (Autumn Long) stanza 9/3; Snorri Sturluson,

Edda, Skáldskaparmál, ed. Anthony Faulkes, 2 vols. (University College London: Viking Society

for Northern Research, 1998), I, 30.

14. Clunies Ross, Prolonged Echoes, pp. 85–102.

22

Clunies Ross

interpretation as Steinsland seeks to place upon it without the substantial
use of extraneous evidence that is of questionable relevance to an early
Norse poem. The second problem, and it is a fundamental one, is that
Skírnismál itself does not mention any offspring, royal or otherwise, of the
union between Freyr and Gerðr, nor does Snorri Sturluson in his prose
account of this myth in his Edda. There is only one place that does so,
and that is the prose text of Heimskringla.
While it is certainly sometimes legitimate to support one’s understanding
of early poetic texts with reference to more explicit later ones, it is worrying
in the case of Skírnismál that the interpretative framework for understanding
this poem as an expression of the Nordic ideology of kingship as based on
the union of a god with a giantess comes from outside the poem itself in
later prose sources. The very same issue concerns the two skaldic poems
Ynglingatal and Háleygjatal that Steinsland also invokes in support of her
theory: the meaning that her theory requires of them depends on the
same thirteenth-century sources, namely Heimskringla, its Prologue and
first, legendary section, Ynglinga saga (Saga of the Ynglingar), together
with the Prologue to the Edda, also the work of Snorri Sturluson.

THE TWO SKALDIC TEXTS:

YNGLINGATAL 1 AND HÁLEYGJATAL 2

Before we investigate how Snorri contextualizes the relevant stanzas
of these two early skaldic poems, it is important to present the texts
themselves, and I do so in the new editions of Edith Marold and her
assistants (for Ynglingatal) and Russell Poole (for Háleygjatal).

15

In each

case there is only one stanza crucial to Steinsland’s argument, the first
stanza of Ynglingatal and what is usually regarded as the second of
Háleygjatal. The first stanza of Ynglingatal runs as follows, with a prose
word order and English translation below:

Varð framgengt,

þars Fróði bjó,

feigðarorð,

es at Fjolni kom.

Ok sikling

svigðis geira

vágr vindlauss

of viða skyldi.

Feigðarorð, es kom at Fjolni, varð framgengt, þars Fróði bjó.

Ok vindlauss vágr geira svigðis skyldi of viða sikling.

15. In SkP I, ed. Whaley, Part 1, pp. 9–10 (Ynglingatal 1) and 199–200 (Háleygjatal 2).

Royal Ideology in Early Scandinavia

23

(The word of doom that fell upon Fjolnir was fulfilled where Fróði lived. And

the windless wave of the spears of the bull [horns > beer] was to destroy

the prince.)

According to the text of Heimskringla,

16

its only medieval context of

preservation, this stanza tells of the death of the Swedish king Fjolnir, son of
Yngvi-Freyr, at the court of the Danish king Fróði at Hleiðra (Lejre). After
becoming drunk and then drowsy, Fjolnir falls into a barrel of ale and is
drowned. In Chapter 10 of Ynglinga saga, which precedes the citation of
this stanza, the prose text accounts for the parentage of Fjolnir, which is
not mentioned in Ynglingatal, as follows:

Gerðr Gymisdóttir hét kona hans [Freys]. Sonr þeira hét Fjolnir. Freyr hét

Yngvi oðru nafni.

17

(Gerðr daughter of Gymir was the name of his [Freyr’s] wife. Their son was

called Fjolnir. Another name for Freyr was Yngvi).

The information about Gerðr’s and Freyr’s marriage plus her giant-
father’s name probably comes from Skírnismál, though in Ynglinga saga the
characters are humans, not gods or giants, but the statement that Fjolnir
was their son is only found in this passage. It cannot be inferred from
Ynglingatal, where no information about his parentage is given and he is
referred to as a siklingr (prince), which, it has been argued, indicates that
he is a mortal, not a god.

18

At the very least, this way of referrring to Fjolnir

does not advertise him as the first of a new breed of being, supremely suited
to the exercise of royal power, as Steinsland’s theory requires.
Unlike some of the other stanzas of Háleygjatal, the second one is only
found in manuscripts of Heimskringla, and it is not absolutely certain that
lines 9–12 belong to this stanza. Some editors and commentators have
treated them as separate.

19

Þann skjaldblœtr

skattfœri gat

Ása niðr

við járnviðju,

16. For the Icelandic prose text and the stanza in context, see Snorri Sturluson Heimskringla

I, ed. Bjarni Aðalbjarnarson, Íslenzk fornrit, 26 (Reykjavík: Hið íslenzka fornritafélag, 1941),

pp. 25–26.

17. Snorri Sturluson, Heimskringla I, ed. Bjarni Aðalbjarnarson, p. 24.

18. See Siegfried Beyschlag, Konungasögur: Untersuchungen zur Königssaga bis Snorri. Die älteren

Übersichtswerke samt Ynglingasaga (Copenhagen: Munksgaard, 1950), p. 37; and Walter Baetke,

Yngvi und die Ynglinger. Eine quellenkritische Untersuchung über das nordische “Sakralkönigtum,”

Sitzungsberichte der Sächsischen Akademie der Wissenschaften Leipzig, Phil.-Hist. Kl. 109/3

(Berlin: Akademie-Verlag, 1964), pp. 85–88.

19. On this point and many other issues regarding this stanza, see Poole’s extensive notes

in SkP I, Part 1, pp. 199–200. It is cited in Bjarni Aðalbjarnarson’s edition of Heimskringla I

on pp. 21–22 (Ynglinga saga, ch. 8).

24

Clunies Ross

þás þau mær

í manheimum

skatna vinr

ok Skaði byggðu,

sævar beins,

ok sunu marga

ondurdís

við Óðni gat.

Skjaldblœtr niðr Ása gat þann skattfœri við járnviðju, þás þau mær,

vinr skatna ok Skaði, byggðu í sævar beins manheimum ok ondurdís

gat marga sunu við Óðni.

(The shield-worshipped kinsman of the Æsir <gods> [= Óðinn] begat that

tribute-bringer [jarl = Sæmingr (?)] with the female from Járnviðr, when

those renowned ones, the friend of warriors [= Óðinn] and Skaði <giantess>,

lived in the lands of the maiden of the bone of the sea [(lit. “maiden-lands

of the bone of the sea”) rock > giantess > = Jotunheimar ‘Giant-lands’], and

the ski-goddess [= Skaði] bore many sons with Óðinn.)

The prose context of citation of this stanza in Heimskringla reminds its
audience that it is about what happened when the giantess Skaði rejected
her marriage with the Vanir god Njorðr, which had been part of the gods’
compensation package to her for their killing of her father Þjazi, and later
married Óðinn (giptisk síðan Óðni).

20

The prose text then claims: “Áttu þau

marga sonu. Einn þeira hét Sæmingr. Um hann orti Eyvindr skáldaspillir
þetta” (They had many sons. One of them was called Sæmingr. Eyvindr
the Plagiarist composed this about him), and this stanza is then cited.
The stanza is followed in the prose text by the comment that Hákon jarl
Sigurðarson reckoned his lineage back to Sæmingr. We may note here, for
further discussion later, that Sæmingr is not actually named in the stanza,
and that the information from the prose source that Skaði married Óðinn
after Njorðr is not found anywhere else in Old Norse texts.

THE MEDIEVAL PROSE CONTEXT

OF THE TWO SKALDIC STANZAS

The prose contexts in which the relevant stanzas from Háleygjatal and
Ynglingatal have been preserved have to be evaluated carefully. They cannot
be taken at face value, and indeed they are inconsistent, if not contradictory,
as the following analysis will show. These sources include the Prologue to

20. The gods’ compensation to Skaði for her father’s killing is narrated by Snorri Sturluson

in the Skáldskaparmál section of his Edda (Snorri Sturluson Edda. Skáldskaparmál, ed. Faulkes, I,

2), while the incompatible marriage partners Njorðr and Skaði are the subject of a passage

in Gylfaginning (Snorri Sturluson Edda. Prologue and Gylfaginning, ed. Faulkes, pp. 23–24).

Royal Ideology in Early Scandinavia

25

Heimskringla, in which the author, probably Snorri, gives an account of
both his oral, vernacular poetic sources and the works of historians such
as the Icelander Ari Þorgilsson (1068–1148), whose writings he mentions
there. Another important source is the first part of Heimskringla, Ynglinga
saga, where the author follows and expands upon the text of Ynglingatal,
giving an account of the rulers of Sweden and Norway from legendary
times up to the period immediately before the ninth-century Norwegian
king Hálfdan svarti (the Black). A third source of relevance here is the
Prologue to Snorri’s Edda.
Although these works are written in the vernacular and are based on an
early thirteenth-century Icelander’s knowledge of early Nordic myth and
the poetry in which it was often expressed, they are also part of learned,
historiographical traditions that manifested themselves in many medieval
European textual cultures. These have been widely studied, and their bases
are well understood.

21

They aimed to situate the pre-Christian societies of

early medieval Europe within the mainstream of Christian and classical
traditions by creating genealogies for the rulers of those societies that saw
them descend from Biblical figures or from euhemerized versions of the
pre-Christian gods. It is probable that they drew support, at least to some
extent, from indigenous ideas that the early rulers of these pre-Christian
societies did indeed descend from the gods, and both Ynglingatal and
Háleygjatal support this concept, assuming them to be old poems,

22

but the

prose texts that make use of such genealogical poetry were undoubtedly
the product of medieval learning, blending the two traditions, as the
Prologue to Heimskringla indicates.
The latter part of the Prologue to Snorri’s Edda and the early chapters
of Ynglinga saga present a historically oriented narrative of the migration
of the euhemerized Nordic deities from Asia to Scandinavia and tell
how they established themselves as rulers over Sweden (Svíþjóð). The
narrative does not present Óðinn, Njorðr, and Freyr as gods, but rather
as humans who persuaded the native inhabitants of the region to give
them the power to control their society by imposing taxes, carrying
out sacrifices to the gods, and so forth. The narrative is concerned
to establish the euhemerized Óðinn, Njorðr, and Freyr as successive

21. See, among others, Andreas Heusler, Die gelehrte Urgeschichte im altisländischen Schrifttum,

Abhandlungen der Preußischen Akademie der Wissenschaften, Phil.-Hist. Kl. 1908, 3 (Berlin:

Reimer, 1908); and Anthony Faulkes, “Descent from the Gods,” Mediaeval Scandinavia, 11

(1978–79), 92–125.

22. It has been argued by Claus Krag, Ynglingatal og Ynglingesaga: En studie i historiske kilder,

Studia humaniora, 2 (Oslo: Universitetsforlaget, 1991) and in other writings that Ynglingatal

is a work of the Christian age and is probably to be dated to the twelfth century. This theory

has not received a great deal of support.

26

Clunies Ross

rulers of Sweden, in the process incorporating a number of themes
and motifs from the mythological narratives we know best from Snorri’s
Edda, though many of them are attested in skaldic and eddic poetry too.
These mythological motifs are subordinated to the main historicizing
line of the narrative.
The Prologue to Heimskringla includes a very clear statement of Snorri’s
understanding of the purpose and contents of both Ynglingatal and
Háleygjatal:

Þjóðólfr inn fróði ór Hvini var skáld Haralds konungs ins hárfagra. Hann orti

kvæði um Rognvald konung heiðumhæra, þat er kallat Ynglingatal. Rogn-

valdr var sonr Óláfs Geirstaðaálfs, bróður Hálfdanar svarta. Í því kvæði eru

nefndir þrír tigir langfeðga hans ok sagt frá dauða hvers þeira ok legstað.

Fjolnir er sá nefndr, er var sonr Yngvifreys, þess er Svíar hafa blótat lengi

síðan. Af hans nafn eru Ynglingar kallaðir. Eyvindr skáldaspillir talði ok lang-

feðga Hákonar jarls ins ríka í kvæði því, er Háleygjatal heitir, er ort var um

Hákon. Sæmingr er þar nefndr sonr Yngvifreys. Sagt er þar ok frá dauða

hvers þeira ok haugstað.

(Þjóðólfr the Wise from Hvinir was a poet of king Haraldr the Fair-haired.

He composed a poem about king Rognvaldr [High with Honours], which

is called Ynglingatal. Rognvaldr was the son of Óláfr Geirstaðaálfr, brother

of Hálfdan the Black. In that poem thirty of his ancestors are named and it

is told about the deaths of each of them and their burial places. That one is

named Fjolnir, who was the son of Yngvifreyr, the one to whom the Swedes

sacrificed for a long time afterwards. The Ynglingar are named after him.

Eyvindr skáldaspillir also enumerated the ancestry of Jarl Hákon the powerful

in the poem which is called Háleygjatal, which was composed about Hákon.

Sæmingr is there named as the son of Yngvifreyr. It is also told there about

the deaths and burial places of each of them.)

Here it is made clear that, according to Ynglingatal, the Yngling dynasty,
which originated in Sweden but came to rule in Norway, descended from
Fjolnir, the son of Yngvifreyr, an euhmerized avatar of the Norse god Freyr,
while, according to this account of Háleygjatal, the dynasty of the jarls of
Hlaðir, based in the Trøndelag region of northern Norway, claimed descent
from a certain Sæmingr, also said to be a son of Yngvifreyr. No mention is
made in either case of the mothers of these two ancestral beings.
As we have seen, Háleygjatal 2 is extant only in manuscripts of Heimskringla.
It is cited in Chapter 8 of Ynglinga saga in the context of an extended
account of the rule of the euhemerized Óðinn in Sweden after the Æsir’s
migration there from Asia:

Um alla Svíþjóð guldu menn Óðni skatt, penning fyrir nef hvert, en hann

skyldi verja land þeira fyrir ófriði ok blóta þeim til árs. Njorðr fekk konu

þeirar, er Skaði hét. Hon vildi ekki við hann samfarar ok giptisk síðan Óðni.

Áttu þau marga sonu. Einn þeira hét Sæmingr. Um hann orti Eyvindr skál-

daspillir þetta:

Royal Ideology in Early Scandinavia

27

[Háleygjatal st. 2]

Til Sæmings talði Hákon jarl inn ríki langfeðga kyn sitt. Þessa Svíþjóð kol-

luðu þeir Mannheima, en ina miklu Svíþjóð kolluðu þeir Goðheima. Ór

Goðheimum sogðu þeir morg tíðendi.

23

(Throughout all Sweden people paid tax to Óðinn, a penny per head, and he

was obliged to defend their country from hostility and sacrifice for them for

prosperity. Njorðr married the woman called Skaði. She did not want to have

intercourse with him and afterwards married Óðinn. They had many sons. One

of them was called Sæmingr. Eyvindr skáldaspillir composed this about him:

[Háleygjatal st. 2]

Jarl Hákon the powerful traced the genealogy of his family to Sæmingr.

They called this Sweden Mannheimar, but they called Sweden the Great

Goðheimar. They told many stories out of Goðheimar.)

The quotation from Háleygjatal 2 is the only citation from this poem in
Ynglinga saga, and it precedes the enumeration of the Yngling kings that
begins in Chapter 11, with the support of the poem Ynglingatal. There
appear to be two reasons for the citation of this stanza here. The first is
that it allows Snorri to account for the existence of the jarls of Hlaðir
as a dynasty different from that of the Ynglingar, who were to be his
main subject in Ynglinga saga, and he does this, backed up by the stanza,
by assigning their genesis to a union of Óðinn with the euhemerized
giantess Skaði. In terms of the main narrative of Ynglinga saga, and indeed
of Heimskringla, this was a side step from Snorri’s account of the chief
inheritance of power among the euhemerized Æsir, which descended
from Óðinn to Njorðr and then, through Freyr, to the Ynglingar in the
person of Fjolnir. The second reason for quoting Háleygjatal 2 was likely
its reference to man(n)heimar, which Snorri interpreted etymologically
as a place, “the world of humans” and an alternative name for Sweden,
which fitted very well with his general euhemeristic frame, though it was
probably not what Eyvindr had in mind.
If we compare these two contradictory accounts of Sæmingr’s paternity,
it looks rather as though, in the case of Ynglinga saga Chapter 8, the
passage from Háleygjatal 2 gets Snorri out of a dilemma; he may well
have known a tradition assigning Sæmingr’s paternity to Freyr, but, in his
euhemerized account of Nordic dynastic inheritance, he needed to keep
Njorðr free for rulership and paternity in the line that led through Freyr
directly to the Ynglingar, so Skaði, who was otherwise well known in myth
for her marriage with Njorðr,

24

could be shunted off into the line that led

23. Snorri Sturuson Heimskringla I, ed. Bjarni Aðalbjarnarson, pp. 21–22.

24. Not only was this marriage well known to Snorri himself and written into the Edda, but

in Skáldskaparmál he cites a fragment of poetry that alludes to it by the skald Þórðr Særeksson

(ca. 960–1020 [?]) Fragment 3/2, 6

III

snotr goðbrúðr nama una Vani (the wise god-bride did

not begin to love the Vanr). Cf. the description of Skaði as scír brúðr guða (the bright bride

of gods) in the eddic poem Grímnismál 11/5.

28

Clunies Ross

to the jarls of Hlaðir, promoters of the old religion, who were not going
to be the main subject of Ynglinga saga nor of Heimskringla.
Still another of Snorri’s euhemerist narratives, this time in the Prologue
to the Edda, tells how Óðinn, the leader of the Æsir from Troy, appointed
his sons as rulers over Sweden and Norway, again with reference to
Háleygjatal:

Eptir þat fór hann norðr þar til er sjár tók við honum, sá er þeir hugðu at

lægi um oll lond, ok setti þar son sinn til þess ríkis er nú heitir Nóregr. Sá

er Sæmingr kallaðr, ok telja þar Nóregskonungar sínar ættir til hans ok svá

jarlar ok aðrir ríkismenn, svá sem segir í Háleygjatali. En Óðinn hafði með

sér þann son sinn er Yngvi er nefndr, er konungr var í Svíþjóðu, ok eru frá

honum komnar þær ættir er Ynglingar eru kallaðir.

25

(After that he travelled north until the sea, which they [the Æsir from Troy]

thought lay around all lands, stopped his advance, and there he established

his son to rule over that kingdom which is now called Norway. He is called

Sæmingr, and the kings of Norway reckon their genealogies back to him, as

well as jarls and other powerful men, as it says in Háleygjatal. And Óðinn had

with him that son of his named Yngvi, who was king in Sweden, and those

lineages that are called Ynglingar are descended from him.)

In this passage Snorri glosses over the fact that Háleygjatal refers only to the
jarls of Hlaðir, and asserts that the poem gives an account of the descent
of all the kings of Norway as well as jarls and other powerful men, which
is of course a fudge, necessitated because he is here saving the Ynglingar
for kingship of Sweden alone, even though Ynglingatal indicates that a
branch of this dynasty also ruled in Norway and Snorri himself connects
them with the rulers of Norway in Heimskringla. This third passage is com-
pletely silent about the identities of the mothers of these sons of Óðinn;
it is concerned only with descent in the male line and in that respect is
similar to most other medieval Norse historiography, like Ari Þorgilsson’s
Íslendingabók (Book of the Icelanders) and the Latin Historia Norvegiae,
which mention royal and noble dynasties in early Scandinavia and are
also probably dependent on the two skaldic poems discussed here. In the
context of Gro Steinsland’s theory, though, the maternal line of Fjolnir
and Sæmingr, if these were the progenitors of the Ynglingar and the jarls
of Hlaðir, is as important as the male line, if not more so.

WEIGHING UP THE EVIDENCE

The upshot of the foregoing comparison of three prose sources bearing
on Ynglingatal 1 and Háleygjatal 2 is that it is very difficult to place great
trust in medieval historians’ euhemerist uses of Old Norse mythology, even

25. Snorri Sturluson Edda. Prologue and Gylfaginning, ed. Faulkes, p. 6.

Royal Ideology in Early Scandinavia

29

when the writer is as well-informed an individual as Snorri Sturluson was.
Yet it would also be unwise to dismiss out of hand the general likelihood
of a royal ideology of descent from the gods in early Scandinavia, though
it is doubtful whether that was as precisely articulated as Steinsland
has argued. In particular the notion of a hieros gamos between a male
god and a giantess is very weakly attested and depends completely (in
the case of Fjolnir) or largely (in the case of Sæmingr) on otherwise
unattested statements in Ynglinga saga. However, even though there is
no other evidence of the Skaði-Óðinn liaison in Old Norse literature,
we should not necessarily dismiss what stanza 2 of Háleygjatal says about
it. It is possible that Eyvindr knew a myth about Óðinn’s relationship
with Skaði and that it had currency in the circles of Hákon jarl and his
family, who, with their pagan sympathies in an age increasingly turning
toward Christianity, would have been an appreciative audience of a poem
celebrating a connection, however vague, between their dynasty and a
god–giantess liaison. Another poem for Hákon jarl, Einarr Skálaglamm’s
Vellekla (Lack of Gold), also celebrates Hákon’s descent from Óðinn with
the kenning Yggs niðr (descendant of Yggr <= Óðinn>) (Vellekla 19/8),
though here only the ruler’s patriline is mentioned. Given this dynasty’s
original northern Norwegian base, too, in Hálogaland, it would seem
appropriate for the liaison to have taken place in Jötunheimar, as the giants
of Old Norse myth are frequently imagined to have lived in northerly and
rocky environments. And we know from other evidence, such as Orkneyinga
saga and related texts, that the jarls of Møre, ancestors of the earls of
Orkney, also claimed descent from a giant, Fornjótr, though in their case
in the male rather than the female line.
What neither Ynglingatal stanza 1 nor Háleygjatal stanza 2 can do
independently is to support Steinsland’s theory of a general Nordic
ideology of royal power based on the ruler’s dynastic connection with an
ancestral figure who was the product of a hieros gamos between a male god,
whether Óðinn or Freyr, and a giantess, whether Skaði or Gerðr, which
produced a new class of exceptional being, fit to rule as a king or jarl.
The evidence of Háleygjatal 2 points rather to Óðinn’s relationship with
Skaði as one of many informal alliances that this god had with giantesses,
and which produced sons: Þórr with Jorð, Váli with Rindr, Viðarr with
Gríðr, and possibly others whose mothers are unrecorded.

26

If Eyvindr had

wished to celebrate a hieros gamos giving power and authority to the rule of
the jarls of Hlaðir, he would surely have chosen a more impressive kenning
to refer to the product of that marriage. Instead he refers to the unnamed
offspring of Óðinn and Skaði as þann skattfœri (that tribute-bringer) (2/1,

26. A list of Óðinn’s many sons is the subject of a þula (Þul Ása 1).

30

Clunies Ross

2). A glance at the skaldic database’s current list of kennings for jarl shows
that þann skattfœri conforms to a pattern that modestly defers to the higher
authority of rulers who are kings. It does not assert the kind of power that
the descendant of a hieros gamos might expect to exert.

27

And the scion

of Óðinn and Skaði is not even named or referred to again, at least in
that part of Háleygjatal that has survived. Indeed, as we have seen, even
Snorri seems to have been confused about Sæmingr’s parentage. Sæmingr
is a figure who is never named in skaldic poetry except in a single, and
probably late þula that lists a number of Óðinn’s sons (Þul Ása 1). The
literary presentation of Fjolnir in Ynglingatal invites similar skepticism.
Although he is named in the text, nothing is made of his parentage, nor
is he referred to in a kenning as the son of a god or a special new type of
ruler, as one might expect. Instead, as we have seen, he is simply called
siklingr (prince), a designation one would expect of a human ruler.

28

There are in fact several kennings in both Ynglingatal and Háleygjatal
that support the idea that the early kings of Sweden and Norway were
considered to be god-descended, but these do not favor a god–giantess
liaison. Rather they suggest a more generalized sense of divine descent.
Two such kennings stand out: áttungr Freys (descendant of Freyr)
(Ynglingatal 16/7 and Háleygjatal 7/7) and áttungr Týs (descendant of Týr)
(Ynglingatal 14/3 and Háleygjatal 10/7). This kenning pattern is unusual,
the only similar one appearing in Sigvatr Þórðarson’s Vestrfararvísur
2/6, 7, where the kenning áttungr Gorms (descendant of Gormr) is used
to refer to the Danish king Knútr. In each case, however, though the
kennings are the same, the referents are quite different, which, together
with the fact that no female partner is named, casts considerable doubt on
the hieros gamos theory. In Ynglingatal, áttungr Freys refers to the Swedish
king Aðils, while in Háleygjatal 7, it refers to Hákon jarl Grjótgarðsson.
In Ynglingatal 14, áttungr Týs refers to the Swedish king Egill, while in
Háleygjatal 10, it refers to Sigurðr jarl Hákonarson, father of Hákon jarl.
It looks very much as though this type of kenning may have been in use
as a formulaic honorific ca. 900 when Þjóðólfr composed Ynglingatal
and was then borrowed by Eyvindr skáldaspillir about a century later
to give a genealogical boost to the pedigree of the jarls of Hlaðir. What
is highly unlikely, on the basis of this evidence, is that there ever was a
hieros gamos myth involving a god and a giantess underlying the kenning
pattern exemplified here.

27. There are in fact relatively few kennings of this type. Cf. orðheppinn spjalli jöfra (the

speech-blessed confidant of princes) (Glúmr Geirason, Gráfeldardrápa 11/7–8) and nefi jarla

(the kinsman of jarls) (Sigvatr Þórðarson, Austrfararvísur 18/4).

28. Fjolnir is elsewhere one of the god Óðinn’s names (Grímnismál 47/5; Reginsmál 18/7)

and occurs as such in several skaldic kennings.

Royal Ideology in Early Scandinavia

31

A similar conclusion may be drawn from a study of skaldic kennings for
king and/or ruler. If Steinsland’s theory were correct, one would think
that, in early poetry at least, divine descent from the prototypical god
and giantess might be a productive element in such expressions. There
are too many to discuss here in detail, but the dominant patterns are:
“lord of X,” where X is the name of a Norwegian or Swedish region and/
or its people, like “the lord of the Mœrir” or “the ruler of the Svíar” or
periphrases like “the ruler of men.”

29

Except for those discussed above,

none of the kennings accumulated under these headings claim descent
from any god whatsoever. This is unlikely to be the result of Christian
bias, given the considerable amount of skaldic verse surviving from the
late ninth through to the early eleventh century.

CONCLUSION

This article has been concerned with using the tools of textual analysis
and source criticism to disable a theory of the origins of royal power in
early Scandinavia, particularly in Norway and Sweden, promoted by the
Norwegian historian of religion, Gro Steinsland. Although the theory has
been in play on the conference circuit and in the writings of scholars in
the field of Norse mythology and the history of religion in Scandinavia for
over twenty years, its fundamental premises have never been thoroughly
discussed. In undertaking this largely philological and literary analysis, I
have left out of account some other fundamental questions that can equally
well be raised in connection with Steinsland’s ideas, questions in particular
that relate to the appropriateness of the concepts of sacral kingship and
hieros gamos to ruler ideology among the Pre-Christian Scandinavians. I
leave these to historians of religion who are better qualified than I am
to tackle them.
The question of the male bias of our sources is also something that I have
not discussed directly. It is quite clear that, although female figures play
an important part in Old Norse myth generally, most of the sources that
relate to royal or aristocratic genealogy, whether traditional or learned,
show an almost exclusive interest in the male progenitors of dynasties, or
at least record descent in the male rather than the female line. It could be
argued that this bias makes it difficult to prove the validity or otherwise

29. See Rudolf Meissner, Die Kenningar der Skalden. Ein Beitrag zur skaldischen Poetik,

Rheinische Beiträge und Hülfsbücher zur germanischen Philologie und Volkskunde, 1

(Bonn: Schroeder, 1921), pp. 351–63, for a list of kennings for king or ruler. The relatively

small number of kennings that use a determinant referring to an ancestral figure are listed

on p. 361.

32

Clunies Ross

of Steinsland’s theory, and this is true to an extent. Unfortunately, an
argument from silence in the sources, unless it can be made good from
other relevant witnesses, cannot be sustained.
With regard to the sources themselves, I have been concerned with
demonstrating that a methodology that uses one kind of text to make
good deficiencies in another (from the perspective of a particular
theory) is academically unacceptable, just as it is not legitimate to use
later, euhemerist interpretations of early myths to provide a bridge, as
Steinsland does,

30

between a poem (Skírnismál) that makes no mention

of human affairs and two early skaldic poems (Ynglingatal and Háleygjatal)
that require the commentary of a thirteenth-century historian, Snorri
Sturluson, to make them yield the myths of Nordic royal origin that are
being sought.

31

30. E.g., Steinsland, “Die mythologische Grundlage für die nordische Königsideologie,”

pp. 741–43; Steinsland, “Origin Myths and Rulership,” pp. 22–23, 57–58.

31. Mytene som skapte Norge, p. 87: essensielt i den norrøne opphavsmyten er at den prototypiske

herskeren avles på tvers av gude- og jotunverdenen (The essential thing in the Nordic myth of

origin is that the prototypical ruler is begotten across the divine and giant world).

Royal Ideology in Early Scandinavia

33