-
Beginnings and Endings in the Icelandic Family SagasAuthor(s):
Kathryn HumeSource: The Modern Language Review, Vol. 68, No. 3
(Jul., 1973), pp. 593-606Published by: Modern Humanities Research
AssociationStable URL: http://www.jstor.org/stable/3724996
.Accessed: 16/01/2011 19:25
Your use of the JSTOR archive indicates your acceptance of
JSTOR's Terms and Conditions of Use, available at
.http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's
Terms and Conditions of Use provides, in part, that unlessyou have
obtained prior permission, you may not download an entire issue of
a journal or multiple copies of articles, and youmay use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this
work. Publisher contact information may be obtained at
.http://www.jstor.org/action/showPublisher?publisherCode=mhra.
.
Each copy of any part of a JSTOR transmission must contain the
same copyright notice that appears on the screen or printedpage of
such transmission.
JSTOR is a not-for-profit service that helps scholars,
researchers, and students discover, use, and build upon a wide
range ofcontent in a trusted digital archive. We use information
technology and tools to increase productivity and facilitate new
formsof scholarship. For more information about JSTOR, please
contact [email protected].
Modern Humanities Research Association is collaborating with
JSTOR to digitize, preserve and extend accessto The Modern Language
Review.
http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=mhrahttp://www.jstor.org/stable/3724996?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=mhra
-
BEGINNINGS AND ENDINGS IN THE ICELANDIC FAMILY SAGAS
The difficulties inherent in starting and ending a narrative
increase with the degree of mimetic accuracy sought, since
absolutely first causes or utterly final effects are not found in
real life. The family sagas derive their matter from Icelandic
history (however much the true course of events is reshaped or
mangled), and hence impose on their creators special technical
problems. Neither those problems nor the Icelanders' solution,
their patterns of beginning and ending, have received attention.
Yet those introductions and conclusions are not only illuminatingly
consonant with the impulses behind saga creation, they also deserve
respect as a highly novel method of spanning the gap between art -
necessarily finite - and the continuum of human experience.
Recent criticism has done much to increase our sensitivity to
some areas of saga aesthetics like ethos and plot structure,1 but
disparagement of the sagas' extremities remains the standard
attitude. Scholarly discussions of individual sagas often ignore
such material; many translators omit it without bothering even to
acknowledge the cuts; and reference works like Einarsson's History
apologetically deprecate 'the weakness for genealogy and personal
history'.2 The fundamental problem is well put by Theodore M.
Andersson, who recognizes some functions of the long introductions,
but observes that they 'apparently give information for
information's sake and are not integral in the sense that they
contribute something vital to the later story. They could be
dropped without depriving the reader of any hints about things to
come' (p. 9). He is, of course, perfectly correct. What the modern
reader finds wanting is organic unity, logical and causal bonds
uniting every portion of a work from first word to last. We are
conditioned to expect this aesthetic by most of western literature,
and the sagas do not conform to our expectations. Such a unified
action can be found in most of them, as Andersson's brilliant plot
scheme demonstrates, but the conflict pattern he traces is not the
whole of these works. Almost all of the sagas contain material
which we instinctively feel to be 'pre- beginning' and
'post-ending'; not integral in the sense of not contributing to the
conflict story, and yet apparently considered necessary according
to saga aesthetic, and hence integral to the form, if not to that
portion we call the plot. The nature and function of this material
are the concerns of the present study.
That we may approach the family saga beginnings and endings
afresh, with renewed sense of their uniqueness, we might glance
first at the sagas with forms
1 Some interesting work has been done on saga narrative mode by
Robert Scholes and Robert Kellogg in The Nature of Narrative (New
York, 1966). Problems relating to ethos in the sagas have been
dealt with illuminatingly by M. C. van den Toorn in Ethics and
Moral in Icelandic Saga Literature (Assen, 1955); Lars Lonnroth in
'The Noble Heathen: A Theme in the Sagas', Scandinavian Studies, 4I
(1969), I-29; and Theodore M. Andersson, 'The Displacement of the
Heroic Ideal in the Family Sagas', Speculum, 45 (1970), 575-93. Our
understanding of plot structure has been greatly advanced by
Andersson's The Icelandic Family Saga (Cambridge, Massachusetts,
I967). Citations of Andersson's ideas are from the book, not the
article.
2 Stefan Einarsson, A History of Icelandic Literature (New York,
I957), p. 134, italics added. W. P. Ker, whose comments are still
influential, remarks that 'the local history, the pedigrees of
notable families, are felt as a hindrance, in a greater or less
degree, by all readers of the Sagas; as a pre- liminary obstacle to
clear comprehension'. See his Epic and Romance: Essays on Medieval
Literature (1897; reprinted New York, I957), p. I85.
38
-
Beginnings and Endings in the Icelandic Family Sagas
more in line with European tradition, known variously as sagas
of kings, bishops, saints, and knights, as well as lying sagas and
sagas of the legendary past.' Examples of each, though not then
differentiated as they are now with separate generic names, were
available for imitation throughout the period in which the family
sagas were written, and the family sagas do indeed show their
influence: one notes the many informed allusions to Norwegian
history, or the obvious knowledge of the heroes described in the
legendary sagas; the Spesar pdittr (an episode in Grettis saga) and
the romantic impulse behind several plots seem to derive from
European chivalric romance; dreams and marvels often prove
familiarity with hagiography.2 Almost all of the works belonging to
these other types of sagas conform to one of two structural
patterns, vita and romance. Since these two forms were imported,
the European patterns most likely to have influenced the family
saga writers can be taken into account at the same time.
The vita pattern is, as its name implies, biographical, and
derives its organiza- tional unity from the subject's life span.
But modifications of the pure form, especially at outset and
conclusion, had become conventional: a saint's birth was often
accompanied by pre-natal portents of sanctity, his corpse was
almost invariably the occasion of further marvels to confirm that
sanctity, and it became customary to list the miracles he had
performed as a kind of finale to the vita. The accretion of such
details, inorganic to the subject's life (the plot) but relevant to
its religious significance (an interest extrinsic to aesthetic
concerns), is common among hagiographies, and also among such
derivative forms as the bishops' and kings' sagas. Thus the birth
of Sverrir is heralded by a portentous dream, as are the nativities
of many saints. And while the sagas of canonized bishops and kings
might end with a list of miracles, those of the less blessed
substituted alterna- tives to round out the literary formula: a
'secular miracle' (6Oafr Tryggvason's hound killing itself at the
news of his death); a list of public works (Hdkonar saga gamla); a
list of famous men who died during the subject's episcopate (Pals
saga byskups); a cluster of verses (both Hdkonar saga goda and
Olafs saga Tryggvasonar in Heimskringla). This tendency to sprout
list-like codas beyond the hero's death probably influenced saga
aesthetics. In addition to this overlap, there are two other
features shared with the family sagas which should be noted: the
practice of dating stories with Norwegian regnal allusions, and the
fascination with genealogy. So strong are the alignment in terms of
Norway and the interest in pedigree that they intrude even into the
early, imitative, latinate vitae like Porlaks saga byskups and Jons
saga helga (older version), and these features are extensively
developed in such pieces as Gu6mundar saga Arasonar.
Knights' sagas and a substantial number of lying and legendary
sagas exhibit the structure variously known as the romance or
folktale pattern. This form was
1 That these are all known as sagas is, of course, no indication
of generic kinship. As Lars Lonnroth has argued in European Sources
of Icelandic Saga-Writing (Stockholm, I965), p. 6, saga is not a
technical name for a separate genre, but only the Icelandic
equivalent to historia or narratio, terms to designate narrative as
opposed to hortatory or instructional works.
2 The essential Christianity of many dreams and marvels is only
beginning to be recognized. See Dag Str6mbick, 'Some Remarks on
Learned and Novelistic Elements in the Icelandic Sagas', in Nordica
et Anglica: Studies in Honor of Stefdn Einarsson, edited by Allan
H. Orrick (The Hague, 1968), pp. I40-7, and Paul Schach, 'Symbolic
Dreams of Future Renown in Old Icelandic Literature', Mosaic, 4
(I970-I), 5I-73?
594
-
KATHRYN HUME 595
immensely popular throughout the Middle Ages in Europe, and is
notable for its tidy, organic form. Typically, a single hero
undertakes a quest or series of tests, and ends by succeeding,
marrying, and assuming power.' At the outset, we are told his
homeland, parentage, and rank, but nothing more unless it is
relevant to his adventures. At the end, the author may remark that
he had heirs, and may even refer to his eventual death, but the
terminal mood in such a work is festively celebratory, for the hero
has successfully developed from untried youth to seasoned warrior
and leader.2 This pattern seems to have exerted less influence than
that of the vita over the family sagas. However, lurking among the
conventional knightly and legendary works are a few pieces which
deviate from this romance formula, and these strongly resemble the
family sagas.
Among the knights' sagas and among European romances generally,
Tristrams saga stands out as unusual; we are told far more about
his parents' ill-fated love than is necessary to explain Tristram's
sad name and natural gentility.3 The tragic finale is also
exceptional. Anecdotes about prior generations, the unseasonable
death of the hero, and (if the hero is of northern extraction)
lengthy genealogies, are common among the legendary sagas. We even
find disregard for the neat folk- tale formula of one hero;
sometimes the emphasis so shifts between two or more men that we
are left to wonder just whom the story is meant to be about.
Vilsunga saga, for instance, is arguably about Sigur6r, yet he
occupies the stage for fewer than half the chapters, and the
detailed, anecdotal account of his heroic ancestors
1 The fundamental economy and clarity of romance beginnings and
conclusions are demonstrated by Valdimars saga:
Filipus hefer kongr heitet. hann red fyrir Saxlandj. hann atte
vid sinnj drottnjngu tuau baurn. son hans het Ualldjmar en Marmoria
dotter. Valldjmar var stor ok sterkr ok vann ok aungum likr at
j1rottum eigi at eins um Saxland helldr fannzt eigi hans likj j
nordrhallfunj. hann kunnj allar tungur at tala ok suo listir at
eingi uar honum jafn. There was a king named Filipus and he ruled
Saxland. He had by his wife two children. His son was named
Valdimar, his daughter Marmoria. Valdimar was large and strong and
handsome and had no match in accomplishment, not only in Saxland
but indeed none his like was to be found in all Europe. He knew all
arts and could speak all languages so that no one was his equal.
siglir herra Ualldjmar nu heim til Saxlandz ok hans drottnjng.
Filipus kongvr tok sott ok anndadizt. sitja 1au herra Ualldjmar ok
Florida nu med sinum heidri ok attu maurg baurn. en la lyktz her
letta afintyr. hafj sa laukk er las ok sa er skrifadj ok heir er
til hlyddu en hinir skamm er ohliod gerdu. Sir Valdimar and his
wife sail home now to Saxland. King Filipus took sick and died. Sir
Valdimar and Florida rule now with all honor and have many
children. Thus ends their adventure. Thanks be to the one who read
aloud and the one who copied, and to those who listened, but shame
on those who made noise.
This example can stand for many; though neat, it is not
unusually so. The text is that edited by Agnete Loth in Late
Medieval Icelandic Romances I, Editiones Arnamagnazane, series B,
Vol. 20 (Copenhagen, 1962), pp. 53-78.
2 For different analyses of this archetypal pattern, see
Northrop Frye, Anatomy of Criticism: Four Essays (Princeton, New
Jersey, 1957), pp. I86 ff., and Vladimir Propp, Morphology of the
Folktale, translated by Laurence Scott, second edition (original
Russian edition I928; translation, Austin, Texas, 1968). Some
structural differences between this kind of romance and hagiography
are discussed in my 'Structure and Perspective: Romance and
Hagiographic Features in the Amicus and Amelius Story', JEGP, 69
(1970), 89-I07.
3 One wonders to what extent family saga aesthetics may have
been influenced by the peculiar structure of this a-typical
romance, with its long inorganic treatment of the previous
generation. It was the first European romance to be translated into
Norse (I226), and Paul V. Rubow has in fact argued that it was the
foundation of 'Norse prose fiction', that no family sagas were
written prior to its appearance. See 'The Icelandic Sagas: The
Icelandic Family-Novel', in Two Essays (Copenhagen, 1949), pp.
30-64 (p. 50)-
-
596 Beginnings and Endings in the Icelandic Family Sagas
and his descendants keeps him from being a focal hero of the
sort found in most romances. Such diffuseness of subject affects
Egils saga (to some extent), Njdls saga, Eyrbyggja, and Laxdaela.
Indeed Laxdoela resembles Vilsunga saga precisely in its most
problematical features: a family is traced through several
generations; the memorable heroic figure (Kjartan O(lafsson) is
killed tragically young and cannot be called focal; and both Guorun
Gjukad6ttir and Guruin Osvifrsd6ttir are present throughout much of
their respective sagas, yet neither seems central enough to be
called the heroine. But if these sagas seem oddly constructed by
modern standards, what about Gautreks saga? That starts with a
comic experience of King Gauti's. Though Gautrek is born as a
result, the anecdote is related for our delectation, not to impart
needed information about the hero. To perplex us further, much of
the saga is not about Gautrek at all, but rather Gjafa-Refr, a
character straight from folktale. Even if we use the alternative
title Gjafa-Refs saga, our usual critical constructs are unable to
account for this distribution of emphasis. Introductory episodes as
inorganic as Gauti's occur in Droplaugarsona saga, Fostbra?ra saga,
and Vatnsdcla, to name but three. Vita-form sagas offer a few
parallels to the family sagas, and most types display interest in
genealogy and alignment with Norway, but only among these
exceptional knightly and legendary tales do we find equiva- lent
structures. Regrettably, the structures are no more easily
understood here than in the family sagas themselves.
If we turn now to the family sagas, and keep these rival forms
in mind, we find many differences. No family saga has completely
organic structuring. A few have very spare introductions,
consisting of a list of dramatispersonae followed immediately by
the story - Porsteins saga stangarhdggs, Hdvardar saga Isfirdings,
and Valla Ljots saga - and a few have economical finales, but no
saga I can find possesses both. Though few attain the length of the
introduction of Laxdola, most at least start with pedigrees, and
concerning these, two features should be noted. First, over three-
quarters of the sagas trace at least one genealogy back to the
settlement generation or before, thus connecting Icelandic and
Norwegian history directly or implicitly. Second, the genealogies
traced are not necessarily those of the sagas' central characters;
the fame of forebears outweighs logical consideration of relevance.
A clear demonstration of the force of ancestral fame is seen in
Fostbraodra saga: after the curious Grettir episode, the heroes
Dorgeirr and Dorm6or are identified by parentage but nothing more.
A relative of Dorgeirr's is mentioned however, apparently because
the author wished to trace his pedigree; the man himself plays no
real part in the story, but he is descended from SigurSr
Fafnisbani.
Concern with fame may even affect the layout of some sagas: both
Gunnlaugs saga and JVjals saga start with lesser characters. Had
the former been a European romance, it would have begun either with
'Gunnlaugr het ma6r...' or with a statement on Illugi inn svarti,
giving his name, rank, and the fact that he had a son named
Gunnlaugr. Instead, we hear about Dorsteinn Egilsson. His dream,
which gives this saga a kind of unity beyond the merely
biographical, is one justification for his preceding the others.
Another however is the fact that he is important in his own right
(despite Gunnlaugr's slighting valuation) and blessed with an
extra- ordinarily famous father, Egill Skalla-Grimsson. The
antecedents of Gunnarr and Njall are obscure, but those of Hoskuldr
Dala-Kollsson, one of the first men to be shown in action in Njdls
saga, are so illustrious that it would be difficult to dream up
better, for he is descended both from Siguror ormr-f-auga (and
hence from
-
KATHRYN HUME 597
Ragnarr lo6br6k and SigurOr Fafnisbani) and from Bj6rn Buna,
progenitor of the single most distinguished settlement family.
Perhaps even the disturbingly in- organic Grettir episode in
F6stbradra saga can best be explained as an extreme example of this
concern with fame.l Vermundr appears little in the saga, and his
wife less, yet Porbj6rg in digra had had dealings with one of
Iceland's most famous heroes, so the episode is recounted because
it sheds lustre upon everyone connected to Dorbj6rg, including the
characters in this saga. That Icelanders enjoyed hearing about the
famous men from their past scarcely needs repeating, and my doing
so at such length may seem naive. Interest in past heroes is a
well-documented charac- teristic of Icelandic culture, and, indeed,
of early Germanic peoples generally. But, I would argue, critics
are mistaken when they dismiss the 'digressions' on famous heroes
and the illustrious genealogies of very minor characters merely as
unshapely manifestations of this cultural phenomenon. The logic of
good story- telling is too egregiously violated by many of the
aberrations for this explanation to be satisfying. Anyone
attempting to illuminate saga construction must try to find a
better.
When we consider the inorganic matter other than genealogies in
the opening passages, we find two basic types: one is the colourful
anecdote recounted for its own charm or interest; the other is the
historical survey like that in Laxdala, usually from settlement to
the saga generation. Since neither pattern appears to be more
relevant or organic than the other, it is likely that both were
once considered equally attractive. The human-interest anecdotes
still entertain, but intervening centuries have made the family
history boring or even repellent. Nevertheless, we cannot hope to
understand the principles on which sagas were constructed unless we
recognize the kind of effect the history could have had, and
attempt to deter- mine the reasons for its appeal.
As a way of double-checking the opening pattern principles I
have postulated, I would like to call attention to two sagas which
share the unusual property of
possessing alternative beginnings. When a writer wishes to see a
piece changed, we can logically hope to be able to deduce a good
deal from the contrasting versions, and that is the case with these
two. Droplaugarsona saga and Vdpnfirdinga saga both exist in a
standard form, but each inspired a later writer to try his hand at
a different introit, the results being the Brandkrossa dattr and
Porsteins saga hvita. Vdpnfirdinga saga has a very spare preface,
and disposes of preliminary genealogies and family history in about
ten sentences. Porsteins saga hvita takes the story hinted at in
those sentences and expands it to short saga length. Evidently the
later author felt it in keeping with saga aesthetics thus to
elaborate, even though the story he tells has no bearing on that of
VdpnfirJinga saga; and his assumption is justified by the practices
of the authors of Grettis saga and Egils saga. The story of
Dorsteinn inn
fagri is attractive, and because it concerns Brodd-Helgi's
ancestor Dorsteinn inn hviti, it may be worked in at the beginning
of Brodd-Helgi's saga.
The original opening of Droplaugarsona saga is a tale about the
great-grandfather of Droplaug's sons, Ketill lrymr, and about
Arnei6r, the unusual servant girl he married. It is entirely
irrelevant to the main feud; its inclusion can be justified
only
1 This episode is not found in all manuscripts and has been
called an interpolation, but even if it is, critics have still to
explain why the interpolator did not feel that his contribution
violated saga aesthetics.
-
598 Beginnings and Endings in the Icelandic Family Sagas
if the rules one is judging by state that colourful family
anecdotes may be related for their intrinsic interest or fame. This
saga has, in fact, a typically perplexing inorganic commencement.
The Brandkrossa pdttr however begins with a helpful statement of
authorial intention: Dar hefjum vdr upp Helganna sggu, er Ketill
Drymr er, bvi at ver vitum hann kynsaelstan verit hafa leira manna,
er i bessari sQgu er fra sagt. Eru fra honum komnir Si6umenn ok sva
Krossvikingar ok sva beir Droplaugarsynir. Dat viljum vdr ok segja,
hversu Helgi Asbjar- narson er kominn af landnamsmQnnum, er
g9fgastr ma6r er i lessari sggu at vitra manna vir6ingu. (xi,
i83)1
The author mentions tracing a family to the settlement
generation, and also descendants, some of whom may have been among
his audience. More important however, he acknowledges that the
standard commencement is orthodox in con-
centrating on the family of Helgi Droplaugarson because it is
the most
distinguished; ancestral fame does seem to be the primary
criterion for deciding whose family shall be traced. But the author
wishes to deviate from what he con- siders standard practice and
tell also of the ancestors of Helgi Asbjarnarson because that Helgi
is the man in the saga he respects most. Then follows an anecdote
about Helgi Asbjarnarson's antecedents, which is no more organic
than that about Arnei6r and Ketill. The importance of ancestral
fame and the settlement
generation, and the type of inorganic episode considered proper
are all thus indicated by these two unusual sagas.
When we turn from beginnings to endings, we find no equivalent
to the romance
pattern finish. This may seem strange, considering that form's
popularity elsewhere, until we remember that the joyful romantic
connotations attaching to marriage and assumption of power do not
fit the normal saga subject matter. Saga writers knew all too well
that marriage did not mean 'he lived happily ever after'. At the
end of a feud, with its revenges and counter-revenges, the most one
could realistically hope for was weary equilibrium, not jubilant
festivity. The triangular love situation in Dorsteins saga hvita is
good romance material: the hero wins his love after treachery and
suffering, and makes honourable settlement with Dorsteinn inn
hviti, father to Porgils, his enemy Einarr's closest ally. With
Einarr and Porgils dead and this settlement made, we would expect
the hero to be able to sit back and live happily in peace. And he
does, but only until Dorgils's son becomes old enough to seek
vengeance. Marriage figures in the finale of this saga, and also
in Hwensa-Pdris saga, Bandamanna saga, and Njals saga, but more as
social event than as hieros gamos, symbolic expression of cosmic
harmony.
The vita form of ending finds some equivalents if only because
some sagas con- clude with the hero's death. Hallfredar saga ends
with a 'miracle'; his beloved lord 61Oafr Tryggvason gives
directions in a dream to the man who can seek out and bury
Hallfrebr's body. The termination of Egils saga is adorned with a
secularized corpse miracle: when Egill's bones are translated, his
skull is seen to be extraordinarily
We begin the saga of the two Helgis with Ketill prymr because we
consider him to have been the most blessed with famous kin of those
men mentioned in the saga. From him are descended the men of Si6a
and Krossavik, and also Droplaug's sons. But we also wish to tell
how Helgi Asbjarnarson is descended from the settlers, because he
is the most worthy man in this story, according to the opinion of
wise men.
All quotations from the family sagas are from the fslenzk
Fornrit editions, identified by volume number and page.
-
KATHRYN HUME
thick, and proves so indestructible that even after years in the
ground, a vigorously wielded axe cannot split it. Eyrbyggja and
Viga-Glums saga end essentially at their protagonists' deaths, but
remarks follow on others involved in the feuds, and descendants. In
Egils saga, after the miracle, we learn much about his later kin;
that they perpetuate the dark/fair genetic contrast, that many are
skalds, that Helga in fagra is of his line and exerted fatal
attraction on two skalds, Gunnlaugr and Hrafn. We saw the list of
miracles adapted to secular uses in the kings' and bishops' sagas.
In the family sagas, the most common equivalent is a listing or
discussion of descendants, a device we might have anticipated from
the concerns displayed in the opening passages. Though not organic,
such an ending is compre- hensible.
When the hero dies in the course of the feud, the need for
revenge takes us, of necessity, beyond the natural biographical
end. Once beyond that logical terminus, the author has to find some
arbitrary method of finishing the story. He may end when the feud
is finally settled (Bandamanna saga, Njals saga). But frequently
authors traced the avenger to his death: in Droplaugarsona saga,
Grimr Droplaugarson avenges his brother Helgi, but we then follow
him to Scotland, where he dies of sorcery and an infected wound,
after a fight unconnected to the main story. He does leave
descendants though, and they are named (supposedly) right down to
the writer of the tale. In Grettis saga, one of the most extreme
examples of this solution, the author follows the avenger Porsteinn
as he weathers a dangerous love affair in Constantinople, marries
the woman, takes her back north, lives to old age; then the two
leave their property to their grown children and retire to Rome to
become anchorites that they may atone for the trick they played on
God with their Tristram- and-Isolt oath. The final words are on
Grettir, but the Spesar bdttr, as this afterpiece is sometimes
called, does seem untidily trailing. Perhaps one might classify as
a variant of the avenger-ending the finale which just follows one
other person close to the hero up to that person's death, even if
he or she is not an avenger. Helga in fagra in Gunnlaugs saga and
Gu6run Osvifrsd6ttir are such figures.
In almost all the family saga closures, descendants are
mentioned, and again the claims of fame outweigh relevance to the
main characters. Thus Porsteins saga stangarhiggs ends not with
Porsteinn and his family but with his more famous overlord Bjarni.
We are told that Iorsteinn followed Bjarni until his own death, but
then the narrator regales us with details about Bjarni, his
pilgrimage to Rome, and his illustrious descendants. In
Vdpnfirdinga saga, the later kin of a rather ineffectual would-be
avenger are listed because they number famous bishops among them,
not because the man himself deserved any respect or attention. In
Laxdoela, we learn, albeit sketchily, about two or three
generations after Gu6run Osvifrsd6ttir, much as we follow several
prior generations during that saga's commencement. Descend- ants
may have seemed so much a part of the saga aesthetic to some
writers that they invented them to fit literary need: the existence
of Ari, Glsli's brother, is apparently considered 'extremely
doubtful; nothing is known otherwise of any settlement by him or of
any men who trace their descent from him'.1 Yet the closing words
assure us that he had many descendants. But after reading these
1 See the note by Peter Foote to George Johnston's translation
The Saga of Gisli (London, 1963), p. 85.
599
-
6oo Beginnings and Endings in the Icelandic Family Sagas
endings, we rarely leave the works thinking, 'yes, it had to end
here. This is the only logical place'.
We have now seen the nature of opening and closing material.
Whether expressed in brief or at length, the saga writers' concerns
at those two junctures were essen- tially the same: distinction of
kin and placement of the conflict within the span of Icelandic
history - first to the past with references to famous Icelanders,
indica- tions of time elapsed since settlement, and references to
Norwegian kings; then to the future, with discussion of famous
friends, relatives, and descendants of the saga's characters. In
their various ways, these concerns are all social, and all seem
unnecessary additions to something which strikes us as already
complete. Some prologues actually hinder our grasping the plot, and
many epilogues leave us feeling stranded, cheated, deprived of the
pleasing sensation that the story is over.1 A reader's sense of
finality is governed by formal expectations, and the sagas neither
arouse nor gratify our expectations because we are conditioned to
respond to literature controlled by other conventions. For a start,
they possess a different type of plot, one which does not fit our
pigeonholes of romance, comedy, or tragedy. Gunnlaugr's troubles
stem from flaws of character, but we do not think of the saga as
the tragedy of Gunnlaugr. Nor do we feel Njdls saga to be Njall's
tragedy, even if we feel his death to be a 'tragic' loss to his
country. Nor is Porsteins saga hvita a comedy or romance, though it
resembles them superficially. The conven- tional terms do not apply
because conflict is the usual subject of the sagas, as Andersson
has shown; our engagement with the action is such that we wonder
how the feud will end, not whether a certain character will
survive. Occasionally we can guess: Brodd-Helgi's personality, and
the outlawed state of Grettir and Gisli lead us to anticipate their
demise; but then Egill's temper is such that we might well expect
him to be wiped out, yet he lives to cantankerous old age. From the
begin- ning of any saga, we can expect the ultimate return of
social equilibrium, but whether any one character will be left
alive or not does not become clear from the story's form.
The sagas' concern with social conflict rather than an
individual means that we may find no focal figure, and their
laudable avoidance of idealization often leaves those figures who
do emerge so far from ideal that we are disconcerted; we expect a
hero, defined by both morals and placement, in tales of blood and
violence. And because equilibrium at the end is social rather than
personal, we sometimes feel at a loss to accept as final the state
in which some characters are left: in comedy or tragedy, the hero
either obtains happiness, or is killed off and thus freed from
misery. But saga personages may realistically live unhappily for
many years until natural and unclimactic death. Helga in fagra,
Kormakr, Samr (in Hrafnkels saga), and to a lesser extent Gu6ruin
(svlfrsd6ttir all live on in a state of unresolved tension. Egill's
life, as he advances in years, is far from happy. The sagas do not
provide us with the emotional simplicities we are used to.
Given the sagas' historical and social concerns, we can see why
they might require different beginnings and endings from the
romance-form works
1 Barbara Herrnstein Smith comments on this sense of finality,
saying 'the occurrence of the terminal event is a confirmation of
expectations that have been established by the structure of the
sequence, and is usually distinctly gratifying. The sense of stable
conclusiveness, finality, or "clinch" which we experience at that
point is.. . closure'. See her Poetic Closure: A Study of How Poems
End (Chicago, I968), p. 2.
-
KATHRYN HUME 6oi
current elsewhere at the time sagas were committed to writing.
But treating the
sagas as history or biography does not offer a really
satisfactory explanation either. Andersson indeed has suggested
biography as an analogue: In most cases, however, this introductory
matter has no proleptic function and seems rather to spring from
the author's historical or antiquarian interest. It is a kind of
scholarly preface. Just as the modern biography inevitably begins
with a sketch of what the author has been able to unearth about his
hero's forebears, so the saga author sketches in the family history
and gives whatever information he can about the genealogy, prowess,
and especially the colonization of his hero's ancestors. (pp.
8-9)
This seems good as far as it goes, but the introductory material
in Laxdela or Egils saga or Eyrbyggja is so extensive that no
modern biographer's preface is likely to resemble it: such material
seems more important to saga writers than to us. Nor does this
analogy justify the family trees of very minor characters,
especially when those of the main characters are truncated.
Furthermore, the sagas very clearly are neither biography nor
history as we know these forms. The regularity of Anders- son's own
conflict scheme bespeaks literary shaping, and other research
confirms that characters were freely altered for aesthetic
reasons.l
I would like to suggest that the opening and closing patterns
can best be under- stood as a method of satisfying one of the
cravings which brought the sagas forth, but in a different mode
from that in which the conflict-plots operate. The modern reader's
apperception of the sagas is an aesthetic response only; we cannot
relate to the history they concern. The continuity between plot and
extremities however is not in the aesthetic plane at all (and hence
not organic to the story) but in affective function. The
commencement and ending are part of the historical
impulse behind saga writing, and they have a definite purpose
mediative between
saga and audience on the historical level, but do not contribute
directly to the aesthetic effect of the main story except in
isolated instances.
Consider the historical forces at work during the thirteenth and
fourteenth
centuries, one of the ugliest periods in Icelandic history.
European pressures, both secular and ecclesiastical, so exacerbated
local power struggles that Norwegian governance seemed to offer
more likelihood of peace than the crumbling and destructive
remnants of the native institutions. The Church, by granting easy
absolutions for treacheries men would have hesitated to commit when
constrained
by the old notions of drengskapr, weakened the indigenous
customs. Political contests were now fought with armies rather than
words at the alpingi, and though most of the quarrels were
ostensibly over matters of regional concern, Norwegian fomenta-
tion and claims to sovereignty were usually present. Snorri
Sturluson, one of the most active politicians in the land, was
enjoined by the king in I220 to make Iceland submit to Norway. That
pressure may have influenced his writing, for
during the following decade he penned the stirring speech
delivered by Einarr
Dveraeingr in (lafs saga helga which refuses Olafr any foothold
in Iceland; friend-
ship and dealings as between equal nations he could have, but no
dominion, no
1 Artistic shaping of saga personalities is discussed by Einar
01. Sveinsson (in 'The Icelandic Family Sagas and the Period in
which their Authors Lived', Acta Philologica Scandinavica, I2 (I
937-8), 7I-90, especially pp. 82-3) and Lonnroth (in his 'Noble
Heathen' article (see page 593, note I), p. 28). The best known
example of the creative side to saga writing is Hrafnkels saga;
once considered the most historically accurate of the sagas, it was
shown to be almost pure fiction by E. V. Gordon (Medium /Evum, 8
(I939), 1-32) and Sigurour Nordal (Hrafnkatla, Studia Islandica, 7
(Reykjavik, 1940) ).
-
Beginnings and Endings in the Icelandic Family Sagas
matter how minor.1 We should remember too that Snorri may have
written Egils saga, a saga whose main concern, after biography of a
skald, was bitter indictment of royal tyranny. The period after
I264 was superficially quieter than what had gone before (though
discontent swelled to rebellion in I300), but remained emotionally
unsettled. The remaining shreds of the once-strong social fabric
disintegrated as reciprocity between chieftains and farmers died;
because chief- tains now received their power from the king rather
than from their followers, they could mistreat their subordinates
with impunity. Taxes and loss of independence in various fields
gradually strangled Icelandic commonwealth culture, and eventually
the writing of family sagas.2
The sagas are the product of this age, and, I suggest, partly
owe their existence to the historical conditions. Time and vellum
were indeed necessary to their production, but those commodities
were not new acquisitions.3 What brought sagas forth was a
combination of contributing forces: the accessibility of writing
materials; the presence of literate men with secular tastes,
whether in or out of the Church; discontent with the contemporary
age because it seemed less heroically admirable than the past;
desire to keep alive memory of that past - Snorri's Prose Edda is
another document from the saga-writing period which was composed to
preserve knowledge of the superior practices of the past; and,
finally, fears and discontents related to Norwegian rule. In
addition to all of the sagas' literary merits, whether measured in
terms of style, characterization, thematic richness, plot
structuring, psychology, or story-telling, the sagas are an
affirmation of national heroic identity in a time of pressure and
crisis. Men wanted to be reminded that theirs was a country
deserving respect, capable of governing itself. Anxieties about the
future and nostalgic longing for those times which had been free of
Norwegian and ecclesiastical influence seem to me important in the
sagas. I would argue that the sagas were not received only as good
stories, though of course they are that too. In their historical
capacities, they satisfied psychological cravings extrinsic to
aesthetic appreciation of a well-shaped plot, and the opening and
closing patterns were designed to help satisfy those same
desires.
Consider the tendency to go back to the settlement generation
for a starting point. Though some settlers came to Iceland with the
current king's good will- Ingi- mundr Porsteinsson in Vatnsdela is
one - many came as the result of what they considered royal
tyranny. Any saga mentioning the descendants by blood or marriage
of Kveld-Ulfr or Bj6rn Buna automatically reminds its audience of
Norwegian tyranny; sagas of this sort include Eyrbyggja, Gunnlaugs
saga, Kjalnesinga saga, Laxdcla, Njdls saga, and Viga-Glums saga.
Sometimes a saga merely starts 'Iat var a dogum Haralds konungs ins
harfagra . . .' and we are expected to know what he was famed for.
Kormdks saga, Hrafnkels saga, and Grettis saga use Haraldr in this
fashion. Occasionally the sagas are very explicit. The younger
version of
1 The speech occurs in Chapter I25 of Snorri's 6ldfs saga helga
in Heimskringla. There are similar instances in some family sagas
where contemporary rather than saga-age events seem to have been
uppermost in the writers' minds - the satire on the goJar in
Bandamanna saga is one such.
2 For information on the political problems of this period, I am
indebted to Sigurbur Nordal's fslenzk Menning I (Reykjavik, 1942);
to Einar 61. Sveinsson's The Age of the Sturlungs, translated by
J6hann S. Hannesson (Ithaca, New York, 1953); and to R. George
Thomas's 'The Sturlung Age as an Age of Saga Writing', Germanic
Review, 25 (I950), 50-66.
3 SigurBur Nordal, 'Time and Vellum', his MHRA presidential
address, published in the MHRA Annual Bulletin for 1952, pp.
15-26.
602
-
KATHRYN HUME 603
Gisla saga starts almost belligerently: 'bat er upphaf a sggu
pessi, at Haraldr inn harfagri re6 fyrir N6regi. Hann var fyrstr
einvaldskonungr um allan Noreg, - af hans ett hafa allir si6an
h9f6ingjar verit i N6regi, - ok ur6u margir landflotta fyrir honum
til ymissa landa, ok varu hat meir st6rmenni, fyrir Pvi at Peir
vildu eigi yfirgang hans hola, en h6ttusk eigi hafa styrk til
m6tstQ6u vi6 hann' (vI, 3).1 The manner of most future Icelanders'
leavetaking was actually considered insult- ing to the royal power,
as Haraldr tartly indicates in Vatnsdoela: 'en pat hygg ek, at
Dangat munir pu' koma, ok er pat uggligt, hvart Pu' ferr i lofi
minu e6a leynisk Pui, sem nu tekr mjQk at ti6kask' (vIII, 34).2 Not
only is Norwegian royalty implicitly twitted at the commencement of
most sagas, Norwegians in general come out distinctly inferior to
Icelanders whenever an Icelander travels to visit the king's court.
As skalds and warriors, the Icelandic travellers are always
portrayed as superior.3 Icelanders of the saga-writing period could
be proud of their Norwegian ancestors, and retained a close
interest in that country, but the saga writers did not let their
audience forget why men had left the country. The reason was still
a living one.
The ancestor lists going back to settlement then are not just
antiquarian (as Andersson and others label them), with the
connotations that word has of dry, bloodless, academic interest.
Rather, most of them carry an implicit comment on relations with
Norway, historical and contemporary. But that is not their only
function. We who rarely know our forebears before great-grandfather
fail to realize how real past kin can be to a man, or how great a
role kinship played in Icelandic history. Nordal performs a
valuable service when, working from the Landndmabdk comment 'Fra
Birni er naer allt stormenni komit a fslandi', he traces Bjorn
Buna's descendants down to the time the commonwealth was
established, and shows that an amazing percentage of the magnates
then were of Bjorn's blood.4 According to the law of geometric
progression, the four centuries between A.D. 900 and A.D. I300
would have provided Bjorn with something like eight thousand
descendants living at the close of the thirteenth century, even if
no member of the family ever had more than two children who
succeeded in having children of their own. That statistic suggests
that any member of a Sturlung-age audience would be
1 This story begins with Haraldr inn harfagri ruling over
Norway. He was the first monarch to govern the whole country - from
his line come all later princes of the realm - and many men became
emigrants because of him, going to various other lands, and most
were important men who would not tolerate his tyranny but did not
have the strength to oppose him.
A similar comment appears at the beginning of Fareyinga saga,
which is technically not a family saga because its action is not
laid in Iceland, but which resembles the Icelanders' sagas in
structure and concerns:
Maor er nefndr Grimr kamban; hann bygg6i fyrstr Faereyjar a
dQgum Haralds hins harfagra. Da fly6u fyrir hans ofriki fjQl6i
manna, settuz sumir i Faereyjum ok bygg6u par, en sumir leituou til
annarra ey6ilanda. There was a man named Grimr kamban; he was the
first settler in the Faeroes in the days of Haraldr inn harfagri. A
great many men fled because of his tyranny; some came to the
Faeroes and settled there, but some sought out other empty
lands.
The edition cited is that of Finnur Jonsson (Copenhagen, 1927),
p. i. 2 This I believe, that you will go there [Iceland]; the
question is whether you will go with my
leave, or sneak away as is now the fashion. 3 A. Margaret Arent
discusses this manifestation of wishful thinking in the
introduction to her
translation of Laxd&ela (Seattle, Washington, I964), p.
xxvii. 4 Islenzk Menning I, pp. I 1-20.
-
604 Beginnings and Endings in the Icelandic Family Sagas
related to at least one of the principal settlers, and we might
expect them to be aware of this. Indeed, elaborate kinship
reckoning is still common in Iceland today, especially in country
districts.1 A thirteenth-century listener would have known at the
very least whether, according to family tradition, he was related
to any of the heroes of the saga age. The leisurely study of Ketill
flatnefr's descendants at the beginning of Laxdela is more readily
understandable when we remember the direct personal interest the
audience would have felt in the people described; a good many
listeners may actually have counted Ketill among their kinsmen.
Local and national conditions were so unpleasantly changed, and
so destructive of cultural identity that men looked to their
nation's past for reassurance. The saga plot satisfied this need at
one level by telling of the great men of the heroic past whose
excellence in various fields of endeavour made them a heritage to
be proud of, and one which in some sense reflected glory on their
present descendants. The opening and closing passages served to
anchor the tale firmly in historical reality and offer members of
the audience various channels through which to relate the events to
their personal history. The prelude might well supply the name of
some settler from whom they could trace their own lineage, thus
making the main characters collateral relatives, as it were, and
the plot and its heroics a family experience. The coda, by
mentioning famous friends, relatives, and descendants of the saga's
characters, provided more possible family ties. Thus both plot and
extremities share the affective function of helping the listeners
relate present to past.
If most commencements and endings work on the historical plane
alone, what do they do to the aesthetic? Non-Icelandic readers,
even sensitive ones like W. P. Ker, have always felt that at the
very least that element is thrown off balance, and at worst
seriously mutilated. Even the exceptional Egils saga whose
historical introduc- tion can be said to contribute aesthetically
with its plot foreshadowings and genetic characterization, seems to
modern readers ill-proportioned. Organic unity is still flouted. A
work may be laid out to follow a mechanical pattern however, and
for an audience familiar with that form's conventions, expectations
will be duly aroused and gratified. References to descendants in
the sagas, for instance, surely function as a conventional
indicator of closure. Moreover, symmetry may play a part in our
sense of beginning and end; genealogical and historical references
at both ends of a saga may give the properly conditioned audience a
sense of rounding off and completeness. Writing of this kind makes
active demands on the listeners, who have to know the conventions
for them to work, but given that knowledge, the conventions should
be effective. Frank Kermode, in The Sense of an Ending, defines
literary form (and the saga, albeit oral and traditional by
background, may be called a literary form) as 'humanly needed
order', and works of art as 'things ... designed, in collaboration
more or less close between producer and consumer, to accommodate,
confirm, and extend that order'.2 That statement is particularly
illuminating when applied to saga form and subject matter, for in
both the colla-
1 My thanks to Vilhjalmur Bjarnar and Pardee Lowe of Cornell
University for this information about modern Iceland; thanks also
to them and to Thomas D. Hill of Cornell for other helpful comments
and criticisms.
2 Frank Kermode, The Sense of an Ending: Studies in the Theory
of Fiction (London, I966; paperback edition, New York, 1968), p.
123.
-
KATHRYN HUME 605
boration between author and audience to produce the desired
ordering of ex- perience - relating past to present - is
particularly close.
I started this article by stating that saga beginnings and
endings are a novel answer to a technical problem which faces any
writer, but especially one using historical material, the
difficulty being that 'relations [meaning the interconnected- ness
of events and people] stop nowhere, and the exquisite problem of
the artist is eternally to draw, by a geometry of his own, the
circle in which they shall happily appear to do so'.1 Human
experience is a continuum, but the artist, even one writing
fiction, must artificially separate his material from all that is
continuous with it or would be were the story true. He has to give
tidy endings to experiences which have no endings at all in real
life. Looked at in this fashion, there is something dreadfully
arbitrary about lovers walking off through the sunset to live
happily ever after. In human affairs, marriage is not a terminus at
all; treating it as such is merely a literary technique, one
involving considerable (if pleasing) falsification of experience.
Organic works assume that there can be a terminal effect, and hence
by definition are cut off from the flow of life. Though we know
that the organic method can succeed, we should also be able to see
that the solution cannot be called ideal. There ought to be other
ways of relating literature to the continuum, and it is this, I
believe, that most of the family sagas do.
Saga narrative technique suggests that writers worked by looking
at the span of Icelandic history from settlement to their own
times, and composed by running an eye along the whole length,
usually along one blood line. A writer documented pre-plot material
genetically rather than logically, pausing occasionally to mention
some interesting incident which befell the hero's grandfather;
sometimes he would dilate upon the incident if it were unusually
colourful or particularly satisfying politically. Or if the hero's
family were obscure, minor characters might be able to provide a
colourful historical background for the story among their more
famous ancestors. Then, reaching the main action chronologically,
the writer deals with it, trying (in so far as that is possible) to
trace the action to the return of social equili- brium. Then he
looks at other people, either contemporaries of the hero or descen-
dants; in either case he is widening his perspective beyond one
focal figure and one point in history, and thus works outward from
the plot back into the historical continuum which, by common
knowledge, comes down genetically to his own audience.
The particular nature of the total saga structure can perhaps be
made clearer with a couple of similes. Icelandic history can be
viewed as the warp on a loom, the individual threads being families
seen as unbroken lines from beginning to present. A saga writer may
select a thread and pull it slightly outward to inspect it at some
point (the conflict portion), but the thread remains fastened at
either end to the frame, whose one end is the settlement and whose
other is, by implication, the writer's audience. Or picture
IUtgar6a-Loki's 'cat'. o6rr strains to lift it, and we see it with
its muscles knotted and back arched, but its feet (except one)
remain firmly planted to the floor; indeed, though the phenomenon
is invisible on the physical plane, the feet really go long past
the floor and encircle the world. The saga writer, like Iorr,
raises one section of artistically shaped history for our
delectation - the conflict plot. The opening and closing passages
however ensure
1 Kermode, p. 176.
-
606 Beginnings and Endings in the Icelandic Family Sagas
the connexion of that piece of history to the continuum - they
are the cat's legs, and on the physical plane, they lead us back to
whichever ancestor mentioned is earliest, and forward to whichever
descendant mentioned is closest to the audience. And invisibly,
beyond what is actually mentioned, they lead back to Norway and the
tyranny which sent men to Iceland, and forward to each member of
the audience. As modern readers, we react negatively to saga
beginnings and endings because we are neither conditioned to
respond to the conventions of this form nor aware of personal
genetic connexion to the actors. Seen in light of this analysis
though, perhaps future readers will recognize the original function
opening and closing material performed, and value its intrinsic
significance as a highly original attempt to relate art to the
continuum of human experience.
ITHACA, NEW YORK KATHRYN HUME
Article
Contentsp.[593]p.594p.595p.596p.597p.598p.599p.600p.601p.602p.603p.604p.605p.606
Issue Table of ContentsThe Modern Language Review, Vol. 68, No.
3 (Jul., 1973), pp. i-viii+481-720Front Matter [pp.i-viii]Juvenal
and Restoration Modes of Translation [pp.481-493]"King Lear" in the
Eighteenth Century [pp.494-506]Some Wordsworthian Transparencies
[pp.507-520]The Lament of Edward II [pp.521-529]Machiavellism in
Etienne Pasquier's "Pourparler du Prince" [pp.530-544]"La Revue des
Deux Mondes" in Transition: From the Death of Naturalism to the
Early Debate on Literary Cosmopolitanism [pp.545-550]Proust's Novel
in a Novel: "Un Amour de Swann" [pp.551-558]Cervantes's "El
Licenciado Vidriera": Meaning and Structure [pp.559-568]Salomon
Gessner's Idylls as Prose Poems [pp.569-576]From the "Neue
Gedichte" to the "Duineser Elegien": Rilke's Chandos Crisis
[pp.577-592]Beginnings and Endings in the Icelandic Family Sagas
[pp.593-606]The Rag with Ambition: The Problem of Self-Will in
Dostoevsky's "Bednyye Lyudi" and "Dvoynik"
[pp.607-618]Reviewsuntitled [pp.619-621]untitled
[pp.621-622]untitled [pp.622-623]untitled [pp.623-625]untitled
[pp.626-628]untitled [pp.628-629]untitled [pp.629-630]untitled
[pp.630-634]untitled [pp.634-636]untitled [pp.636-637]untitled
[pp.637-638]untitled [pp.638-639]untitled [pp.639-641]untitled
[pp.641-643]untitled [pp.643-644]untitled [pp.645-646]untitled
[pp.646-647]untitled [pp.647-649]untitled [pp.649-650]untitled
[pp.650-651]untitled [pp.651-652]untitled [p.653]untitled
[pp.654-655]untitled [pp.655-656]untitled [pp.656-657]untitled
[pp.657-659]untitled [p.659]untitled [pp.659-661]untitled
[pp.661-662]untitled [pp.662-663]untitled [pp.663-664]untitled
[pp.664-665]untitled [pp.665-666]untitled [pp.666-668]untitled
[pp.668-670]untitled [pp.670-671]untitled [pp.671-672]untitled
[pp.672-673]untitled [pp.673-675]untitled [pp.675-676]untitled
[pp.676-677]untitled [pp.677-678]untitled [pp.679-680]untitled
[pp.680-682]untitled [pp.682-683]untitled [pp.683-685]untitled
[pp.685-687]untitled [pp.687-688]untitled [pp.688-689]untitled
[pp.689-690]untitled [pp.690-691]untitled [pp.691-692]untitled
[pp.692-693]untitled [pp.693-695]untitled [pp.695-696]untitled
[pp.696-697]untitled [pp.697-698]untitled [pp.698-699]untitled
[pp.699-700]untitled [pp.700-701]untitled [p.701]untitled
[p.702]untitled [pp.702-703]untitled [p.703]untitled
[pp.704-705]untitled [pp.706-707]untitled [pp.707-711]untitled
[p.711]untitled [pp.711-712]untitled [pp.712-713]untitled
[p.713]untitled [pp.714-715]untitled [pp.715-717]untitled
[pp.717-720]
Back Matter