Folktale Morphology and the Structure of "Beowulf:" A Counterproposal

Author(s): Bruce A. Rosenberg

Source: Journal of the Folklore Institute, Vol. 11, No. 3 (Mar., 1975), pp. 199-209
Published by: Indiana University Press
Stable URL: http://www.jstor.org/stable/3813934
Accessed: 12-02-2017 22:18 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

Indiana University Press is collaborating with JSTOR to digitize, preserve and extend access to Journal
of the Folklore Institute

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
BRUCE A. ROSENBERG

Folktale Morphology and the Structure of Beowulf:

A Counterproposal

Another methodological episode in the century-long romance of medie

ists and folklorists - often a tempestuous and passionate affair - w
added in an article appearing in Speculum, "Folktale Morphology
the Structure of Beowulf". In it, Daniel R. Barnes analyzed our old
English epic according to the morphological system, now commonp
to folklorists, of Vladimir Propp. The result has been to reinforce seve
opinions about Beowulf, notably that it is much like a folktale, and
"delineate more precisely those areas in which the poet as conscious art
... is free to create".' Propp thought that folktale form was governed b
laws; the teller was free, however, to choose the names and attributes
his characters, the linguistic means of developing his tale, and the sele
tion and modus operandi of functions.2 If Beowulf is a folktale, then w
Propp observed of the Russian Mdrchen would be true of the epic
Finally, the most striking result of Barnes's examination is his identif
tion of Unferth as the morphological Donor whose alleged treacher
"really irrelevant to his function in the poem".3 This new approac
fraught with possibilities, but also - as is always the case with new
exciting ideas - with dangers.
The first important folktale study of Beowulf was made by Friedric
Panzer, who employed a method that seems to many medievalists so
what paradoxical: the folktale, by definition an oral genre, can only ex

1 Daniel R. Barnes, "Folktale Morphology and the Structure of Beowulf," Spec

45 (1970): 432.
2 Vladimir Propp, Morphology of the Folktale, ed. Louis A. Wagner (Austin, T
1968), pp. 112-113.
3 Barnes, p. 433.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
200 BRUCE A. ROSENBERG

in manuscript form from the early Middle Ages. Therefore Panzer stu
the modern oral reflexes of "The Bear's Son" on the assumption
their form had remained stable for over ten centuries. The validit
this assumption has made the comparative method in folktale stu
possible. The very existence of the Aarne-Thompson's Types of the Fol
tale is based upon the existence of the folktale, independent of its tell
as a discrete, coherent, relatively consistent entity, however abstract t
tale types may seem.4 Not only Thompson but Propp also has insis
upon the folktale's stability,5 for without it morphology would hardly
possible.
In effect, Panzer abstracted from more than two hundred versions of
Type 301, "The Three Stolen Princesses", a structurally consistent sub-
type, "The Bear's Son", in which the objective is not three maidens, but
the monster's defeat. He then assumed that such oral tales must have been
in circulation before Beowulf was written down. If there was any weak
aspect to his methodology it was here, because Beowulf is very old even
in terms of most Marchen; it is possible that our epic came first and that
the folktales are reflexes of it. Scholars have, however, considered it
likely that this particular Mdrchen, at least, had historical precedence,
and that the structural deviants shown in the Beowulf story should be
considered variations from its folktale original.
If some version of "The Bear's Son" - or even several versions which the
scop may have heard during his life - is the origin of our epic, then Un-
ferth's role in the story is an interesting one. Specifically, there is no such
taunter or tester or court gadfly in the reconstructed tale type. It is hard
to believe, therefore, that such a character was ever part of Type 301,
since he would have had to be deleted from all the collected versions.
It seems much more likely that he never was a part of the original
folktale, but rather an addition by the scop when the Mdrchen became an
epic. The Thompson paradigm is explicit on this point: no Unferth
appears:

I. The Hero is of supernatural origin and strength: (a) son of a bear who has
stolen his mother; (b) of a dwarf or robber from whom the boy rescues himself
and his mother; (c) the son of a man and a she-bear or cow; or (e) engendered

4 Antti Aarne and Stith Thompson, The Types of the Folktale, Folklore Fellows
Communications, no. 184 (Helsinki, 1964).
5 Propp, pp. 23-24.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
 FOLKTALE MORPHOLOGY AND THE STRUCTURE OF Beowulf 201

by the eating of fruit, (f) by the wind, or (g) from a burning piece of wood. (h)
He grows supernaturally strong and is unruly.

II. The Descent. (a) With two extraordinary companions (b) he comes to a
house in the woods, or (b1) a bridge; the monster who owns it punishes the
companions but is defeated by the hero, (c) who is let down through a well
into a lower world. - Alternative beginning of the tale: (d) The third prince,
where his elder brothers have failed, (e) overcomes at night the monster who
steals from the king's apple-tree, and (f) follows him through a hole into the
lower world.6

Nor do we find much assistance when we examine medieval literary

analogues. The best known of such stories is in the Grettis saga, the life
of the most famous of Icelandic outlaws, Grettir the Strong. In the
Sandhaugar episode Grettir wrestles with a troll whom he meets in a
deserted house in the woods, works off her arm and follows her as she
disappears behind a waterfall. A priest, Steinn, has been enlisted to hold
a rope for the hero while he descends into the lower world. In the depths
Grettir defeats a giant whom he finds sitting beside a fire, but when he
shakes the rope as a signal to Steinn to raise him, there is no response.
The priest has assumed Grettir dead and has returned to the house.
Grettir has to climb up the rope himself and, once on the surface, seeks
out Steinn to rebuke him.
The Grettis saga, like the modern folktale analogues, does not have an
Unferth figure, nor does the fourteenth-century Saga of Rolf Kraki;
in fact, as Chambers has pointed out, as we search further back into
Scandinavian antiquity - closer to the time of the composition of Beowulf
- the figure of Unferth or of his counterparts disappears.7 The importance
of this fact is two-fold: it suggests that the enfance material, that narrative
which relates the adventures of the hero while a youth, which is present
in most of the folktales and present in the literary analogues, has been
displaced in Beowulf to the "Unferth digression".8 But far more important
for the methodology under question, it suggests that for whatever pur-
pose, Unferth and his flyting with Beowulf was the insertion of a literate
poet, the writer of Beowulf, and apparently not that of any of his folktale
sources.

6 Aarne and Thompson, pp. 90-91.

7 See R. W. Chambers, Beowulf. An Introduction to the Study of the Poem (Cam
Mass., 1959), pp. 57-61.
8 See Bruce A. Rosenberg, "The Necessity of Unferth," Journal of the F
Institute 6 (1970): 50-60.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
202 BRUCE A. ROSENBERG

Barnes argues that Unferth is the Donor of the morphology; if so, we

have the curious phenomenon of a literate scop adding an oral folkt
element to his manuscript which is missing from his folktale sour
Such an emendation is possible; it is even possible that Unferth was in t
oral versions, but that modern analogues delete him. But the latter chan
is not very likely.
And yet there is an interesting point to be made here still. For if our
assumptions are true - that an "oral" element was added by a litera
teller of tales - then we have a great insight into the influence, howeve
unconscious, of the oral tradition and its power to shape the literature
the learned poet. To speculate upon other possibilities, for an instant, w
might well reexamine the Middle English romance, Sir Orfeo, in this ligh
All the known versions of the Orpheus story end with the lovers back-
sliding into the underworld; the romance ends with their happy return
Orfeo's kingdom. It is just possible that the oral tradition, with i
convention of ending folktales - and romances - on a happy note
reconciliation (whether in death as with Tristan, or in respective co
vents as with a few others), has also influenced the author of Sir Or
so that he gave it the appropriate ending - appropriate that is to or
tradition. A recent article in the Chaucer Review on Sir Gowther9 makes
a similar point: the conflated source tales have been joined in such a way
that the finished product stands up under morphological analysis, as
though it were a single, coherent story which, of course, in the finished,
extant version, it has become. Again, oral shaping has made of literary
material a literary result that bears many of the structural traits of oral
composition. If we can establish our understanding of this phenomenon
more firmly we will know substantially more about the intricate and subtle
relationship between oral tradition and literature.
The Aarne-Thompson indexes have their benefits and their shortcom-
ings when applied to literary criticism. The newest generation of folklorists
has not been happy with them, however, and for the same reasons that
some medievalists still find them useful. There is nothing in the indexes
that enables the student to define the relationship of one motif to another,
and consequently nothing with which to define form. Morphology is
particularly concerned with the relationship of structural parts to other
parts, and was designed to define structure. Nothing in the motif and
9 Shirley Marchalonis, "Sir Gowther: The Process of a Romance," Chaucer Review 6
(1971): 14-29.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
 FOLKTALE MORPHOLOGY AND THE STRUCTURE OF Beowulf 203

tale type indexes permits a distinction between the folktale and other
traditional genres, such as the anecdote or the fable. But then the morpho-
logical system, with its emphasis upon deep structure, is nearly useless
in tracing and relating sources. Motif analysis, so detailed and so precise
in its description of event, is ideal for the source-hunting medievalist.
But we must grant that each method has its strengths; morphology is
still new in the United States, particularly among literary critics, and
holds its own promise.
Essentially it is a kind of deep structure of traditional narrative, a
system of analysis that allows one to construct a formula for folktales
which will correspond to an almost infinite number of surface structures
- the varied forms which the style and characters and events give to the
individuality of the tale. The elements of composition, in morphological
terms, are not words or motifs, but "functions", the acts of the tale's
dramatis personae, which are defined from the point of view of their
narrative consequences within the tale. There are no better illustrations
of this concept than Propp's own:

1. A tsar gives an eagle to a hero. The eagle carries the hero away to another
kingdom.
2. An old man gives Sucenko a horse. The horse carries Sucenko away to
another kingdom.
3. A sorcerer gives Ivan a little boat. The boat takes Ivan to another king-
dom.
4. A princess gives Ivan a ring. Young men appearing from out of the ring
carry Ivan away to another kingdom.10

The constants as well as the variables in these illustrations are clear;

although the surface structure form of a particular agent may vary (eagle,
horse, boat, young men), what remains constant is its function, its pur-
pose. Propp found, surprisingly, that an examination of over one hundred
Russian folktales revealed a surprising morphological consistency, lead-
ing him to conclude that there is really only one structure for the folktale
(allowing for a few transformations) and only one source.ll
The repercussions of the Morphology of the Folktale since 1958, when
it first became available in English, have been profound and widespread.
Hundreds of articles, monographs, and books have discussed, modified,
and applied the system to a wide range of traditional genres. Narratives

10 Propp, pp. 19-20.

1 Propp, pp. 23-24, 106.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
204 BRUCE A. ROSENBERG

were not treated exclusively: Dundes expanded the concern of m

logy to the riddle, and others have used it to investigate the b
proverbs, myths, even "minor genres of obscene folklore".12 I th
fair to say that though most of these applications spring from P
initial idea, they have been rather flexibly adapted when applied to
genres as the riddle or the proverb. It is important in the present e
remember that Propp's purpose was to provide a means to defin
precision those folktales which Aarne had classified between nu
300 and 749 in his index,13 and to find an equally precise met
distinguishing the folktale - as genre - from other oral genres.
Barnes' essay analyzes Beowulf with Propp's morphology and finds
all of the functional elements are present and, equally important, tha
are present in the correct order. Although he disclaims any intentio
trying to prove that Beowulf is a folktale,l4 he nevertheless analyze
poem as though it were a Mirchen. Of particular interest to Bar
the role of Unferth in this scheme,15 a major character who is one
most enigmatic in the poem. Unferth's morphological role, we ar
is that of the Donor, which Propp has defined as follows:

Now a new character enters the tale: this personage might be termed the
or more precisely, the provider. Usually he is encountered accidentally -
forest, along the roadway, etc. ... It is from him that the hero (both the
hero and the victim hero) obtains some agent (usually magical) which p
the eventual liquidation of misfortune. But before receipt of the magic
takes place, the hero is subjected to a number of quite diverse actions
however, all lead to the result that a magical agent comes into his h

Barnes' interpolation of Unferth is that, again in the morpho

sense, the taunter of Beowulf "fulfills the functions of a test admini
in behalf of Hrothgar and the Danes by Unferth".17 Unferth's
12 Dan Ben-Amos, "The Situation Structure of the Non-Humorous English
Midwest Folklore 13 (1963): 163-176; George Herzog, Jabo Proverbs from
(London, 1963), pp. 4-5; Claude Levi-Strauss, "The Structural Study of
Journal of American Folklore 68 (1955): 428-444; Alan Dundes and Robert G
"Some Minor Genres of Obscence Folklore," Journal of American Folklore 75
221-226.
13 Propp, p. 19; the "index" for Propp was Antti Aarne, Verzeichnis der Mirchentypen,
Folklore Fellows Communications, no. 3 (Helsinki, 1911).
14 Barnes, p. 432.
15 Ibid., pp. 421-424, 433.
16 Propp, p. 39.
16 Propp, p. 39.
17 Barnes, pp. 422-423.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
 FOLKTALE MORPHOLOGY AND THE STRUCTURE OF Beowulf 205

Donor explains why Hrothgar does not chide him for insulting his guest,
and then accepts Beowulf's offer of aid - subsequent to the revelation of
the Breca episode. From this it follows that the proffering of the sword
Hrunting should not be viewed with suspicion, as Rosier has suggested,18
but is the inevitable concomitant to Beowulf's having passed the test.
Barnes has done what the literary critic should do and what many folk-
loristic approaches leave undone: he has not been content to leave un-
touched the implications of structural analysis, and has gone on to
criticize the poem on the basis of that analysis. But he falls victim, I
believe, to several assumptions that have plagued folklorists and their
diverse approaches to Beowulf for many years, namely, that whatever
its origins, whatever its sources, whatever its structure, the poem that has
come down to us on Cotton Vitellius A XV is not a folktale. And while
some precursor of it may have been in oral tradition for centuries, the
poem as we have it was written down on vellum.
The clearest and most cogent warning about applying the morpho-
logical system of analysis to material other than folktales comes from
Propp himself:

The job [of decomposing tales into their constituent units] is complicated, how-
ever, by the fact that uncorrupted tale construction is peculiar only to the
peasantry - to a peasantry, moreover, little touched by civilization. All kinds
of foreign influences alter and sometimes even corrupt a tale. Complications
begin as soon as we leave the boundary of the absolutely authentic tale.19

Therefore, we should be suspicious when we are told that "I have applied
Propp's theory to Beowulf by testing the poem against the structural model
presented in Morphology of the Folktale". And we should proceed with
extreme caution when the author asserts that one of his premises is
"that examination of the poem in terms of a synthetic structural theory
which attempts to describe the principles common to all folktales -
independent of content similarities - offers, if not a less hypothetical means
than the comparative method, at least a radically new approach to the
question of the 'folk-tale element' in Beowulf".20
Barnes argues that Hrothgar's genealogy "in the prologue to the poem"
is Propp's "Initial Situation" with which folktales begin, and which

18 James L. Rosier, "Design for Treachery: The Unferth Intrigue," Publications of

the Modern Language Association 77 (1962): 1-7.
19 Propp, p. 100.
20 Barnes, p. 418.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
206 BRUCE A. ROSENBERG

commonly introduces the villain's victim and his family. Further

we are told, "the members of a family are often enumerated in a m
similar to the Beowulf- poet's introduction of Hrothgar's lineag
The assumption behind this comparison must be that the folkta
literary works which derive from folktales - let me grant that Beow
one of them - develop in much the same ways. Such an assumpt
seems to me, flies in the face of literary traditions, of which the gen
of heroes and kings is one. Even if such elements are not in the folk
an epic poet might be likely to add them. Again, we have the possibi
that a literate man has added an "oral" element to his manuscrip
we know very well that folktales and epics (or romances or lais)
develop in the same ways: the Thompson indexes are a tribute t
stability of the folktale; the frequent redactions of popular stor
the Middle Ages, the inclination to amplificatio, and the individu
occasionally eccentric forms they assume, is eloquent testimony
quite different development of manuscript traditions. La Chans
Roland developed orally, but it is not a Mirchen nor did it develop q
like one.

Finally, there is the matter of syllogistic logic. Barnes applies the

morphological system to Beowulf and finds that it does in fact account for
all of the characters. This leads him to support the old view - which I
also hold - that, as Klaeber says, the folktale is "the germ pure and simple
of the Beowulfian legend". The unstated premise in these pages22 may
fairly be restated as a categorical syllogism:

1. Morphology describes all folktales.

2. Beowulf is describable by morphology.
3. Therefore, Beowulf is a folktale.

The flaw is that of the undistributed middle term; reversing the order
of the proposition reveals that the assumption is invalid: folktales are not
the only works that morphology describes, a fact which Propp himself
noted: "On the other hand the very same structure is exhibited, for
example, by certain novels of chivalry. This is a very likely realm which
itself may be traced back to the tale".23 And, as I have argued elsewhere,24

21 Ibid., p. 419.
22 Ibid., p. 432-434.
23 Propp, p. 100.
24 Bruce A. Rosenberg, "Morphology of the Middle English Romance," Journal of
Popular Culture 1 (1967): 63-77.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
 FOLKTALE MORPHOLOGY AND THE STRUCTURE OF Beowulf 207

the Middle English metrical romances also bear the structure of folktales.
The methodology we have been examining, then, is open to some
question. Although Barnes cites Propp's admonition about applying
morphological techniques to other than "authentic tales", though he
admits that Beowulf bears "obvious traces" of literary artistry, yet he
examines the poem as though it were a folktale. Literary history is
ignored, and I believe to the detriment of the author's argument: because
Unferth does not appear in the analogues he becomes an even more
interesting creation of the Beowulf scop. And we should view with a
certain reserve any criticism based upon the application to literature of
morphological analyses of folktales.
Specifically, I hope that it will not be too tactless to repeat an earlier
observation that while Unferth may have been the creation of the poet
as a means to relate the enfance of the hero, his dramatic function, his
purpose, is also to contrast with the hero. Obviously he fulfills several
dramatic purposes; but one of them is certainly as a device on the part
of the artist to realize the hero. Again, I should like to recall a personal
conversation with Francis Lee Utley, who noted that Thersites is also
an objectionable gadfly in the Iliad, as is Humbaba in Gilgamesh, Dinadan
in Malory, Falstaff in parts one and two of Henry IV, perhaps even
Pandarus in the Troilus. And I added - as I reconstruct that conversa-
tion - Sir Kay in the romances of Chretien. The ubiquity of this artistic
tactic, often in instances where literary convention cannot be operative,
suggests that it is a natural way for the artist to make his hero great.
Could this not be the case with Beowulf- whether one accepts the "dis-
placed enfance" theory or not - and is it not possible that the figure of
Unferth, like the genealogy, is a literary tradition of early medieval
poetry?
I am hesitant when Barnes says that "Unferth's supposed 'treachery'
is really irrelevant to his function in the poem",25 because Unferth as a
traitor would be crucial to our understanding of his role in the poem - as
literary critics, though not as ethnologists. I am hesitant when told that
"such an interpretation does ... provide a satisfactory explanation for
Hrothgar's failure to rebuke Unferth".26 It is true, such an "interpreta-
tion" does explain Hrothgar's silence, but again, this is not a matter of
literary criticism but of folktale sources. Furthermore, we might well
25 Barnes, p. 433.
26 Ibid., p. 423.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
208 BRUCE A. ROSENBERG

invoke the "Epic Laws" of Axel Olrik here, particularly the "
Two to a Scene" (das Gesetz der scenischen Zweiheit),27 which o
that only two characters ever speak in a scene ("tableau"), and
others are present they remain silent. Olrik does not supply th
- perhaps it is not important or even too obvious, but a three-w
versation, I believe, is too much for the oral teller to handle, and p
too much for his aural audience to follow. Barnes does provid
explanation for Hrothgar's silence; I prefer others.
Finally, "decomposing" the poem into three "Moves" does n
anything to our knowledge or understanding of the structure that
obvious upon a first reading. It even sounds a little like the ninete
century Liedertheorie.28 Such a decomposition and analysis d
confirm the view of those critics who find "(contra Tolkien)" in th
an "integrally whole, three-part structure".29 This implies that Be
three Moves (or parts) have been stitched together, when it does n
likely that the epic developed from oral tradition in quite th
Integral wholeness is not demonstrated, any more than in a poo
folktale which conflates two or more tale types.
And yet, despite all that has been said above, I feel that morp
can be a valuable method for criticizing text literature.30 The prob
that, as with any theory, it has to be implemented properly. In th
a morphological analysis it is often to complement the conclusions
from literary history: certainly this is true when the work in
has come down to us in manuscript form; most certainly this is tr
Beowulf, no matter how much like a folktale it may be. Propp
defers to the historian on matters of origin, the connection between
day life and religion, and the like.31 And we must be careful about

27 "The Epic Laws of Folk Narrative", in The Study of Folklore, ed. Alan
(Englewood Cliffs, N.J., 1965), pp. 131-141.
28 See Arthur Brodeur, "Beowulf: One Poem or Three?" in Medieval Litera
Folklore Studies, ed. Bruce A. Rosenberg and Jerome Mandel (Englewood
N.J., 1970), pp. 3-26.
29 Barnes, p. 433.
30 See William O. Hendricks, "Folklore and the Structural Analysis of Li
Texts," Language and Style 3 (1970): 83-121, and my essay on "The Aesth
Traditional Narrative," in Directions in Literary Criticism, ed. Stanley Wein
Philip Young (University Park: The Pennsylvania State University Press, 1973),
which is based on a paper that I read at the American Folklore Society me
Los Angeles in 1970.
31 Propp, pp. 106-107.

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms
 FOLKTALE MORPHOLOGY AND THE STRUCTURE OF Beowulf 209

ing that the conclusions of morphology could not have been possible
without morphology; this assertion of exclusivity of approach does no
good service to the real advantages which Propp's system genuinely
offers the critic.
Morphology of narrative is still a new idea in the United States. As
I said near the beginning of this essay, it has already been fruitful, it is
more promising still, and it is exciting. But it is still new; as we experi-
ment with it we shall make errors, no doubt, but that is the inevitable
side-effect of innovation. Morphology is a method that we will have to
deal with increasingly in many areas of our intellectual lives in the
immediate future, and Propp is a name that will become even more
familiar. But only if his system is treated properly.

The Pennsylvania State University

University Park, Pennsylvania

This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:18:43 UTC
All use subject to http://about.jstor.org/terms