A Recent Theory of Ballad-Making
by Louise Pound
Source: PMLA, Vol. 44, No. 2 (Jun., 1929), pp. 622-630
XXV- A RECENT THEORY OF BALLAD-MAKING
PROFESSOR GORDON HALL GEROULD'S article entitled "The Making of Ballads"[1] is an attractive essay, written in the fluent and polished manner that we are accustomed to expect from this scholar. It has charm of style, and its positions, taken as
a whole, may be termed accepted positions. Because of its literary
quality, because it brings together in one paper what has hitherto
been stressed in scattered places, and because of its appreciation
of the poetical quality of those English and Scottish ballads sought
out by the notable collectors of the earlier nineteenth century and
made available in the volumes of Professor Child, the paper has
real value for the student. That "The Making of Ballads" is a
research article, the product of painstaking investigation, Professor
Gerould would not, I think, himself maintain. He is a
literary theorist in the realm of traditional song, rather than an
experienced field worker or a practical folk-lorist. He brings
forward little that has novelty for the special scholar. This circumstance
would call for no particular comment except for the
fact that the paper has been announced as new and subversiveas
something independent of old theories. It has been referred to
by several scholars as "The Gerould Theory of Ballad Origins."
The author himself leads us to expect something revolutionary
when he asks us todismiss
from our minds, for the time being, our preconceived and wellbuttressed
theories as to the narrative lyrics we call ballads; forget, if
we can, our arguments;a nd .... look at certain .... indisputablep henomena
of the ballad. Oddly enough, though they are perfectly well
known, they have been much neglected. Very rarely has their existence
been noticed in writings on the ballads, while never, I believe, has their
true significance been fully recognized.
In view of the claims made for it, it seems in place to examine
the article carefully, to ask what is its content and what are its
conclusions. In the first place, what are those overlooked characteristics
on which Professor Gerould's argument is to be based?
The author remarks in his opening pages that ballads have
a profusion of widely different versions, and that they still cir-
Mod. Phil. XXI, 15 ff.
622
Louise Pound
culate in unlettered communities. He believes that the fact of
their variation is familiar to all but that the nature of their variation
has been passed over in silence. Yet this is the phenomenon,
he finds, that throws the clearest light on ballad making. Ballads
do degenerate. Oral tradition fails to represent the original without
change. There are many versions.
What may legitimately surprise us, however, .... is the large number
of ballads of which more than one excellent version have been brought
to light. I beg you who are ballad lovers to consider this phenomenon
carefully..... The point is that there is the widest discrepancy among
what we may call "good" versions of popular ballads; a fact that has never
been emphasized, even though you and I have always known it.
Professor Gerould quotes for illustration two stanzas from
"The Wife of Usher's Well," the first from Scott, whose texts are
always poetical, and the second from Kinloch, also a collector
who specialized in good or pleasing texts. He thinks it surprising
that the second stanza is independently good, not a distorted
reflection of the first. His next step is to ask us to take the several
versions of such a ballad as "The Wife of Usher's Well" and try
to reconstruct from them a composite original. It will be found
that the variants cannot be satisfactorily fitted together. All the
pieces cannot be used. A composite cannot be made that will
embody everything good without wrecking the narrative structure.
Does this generalization seem novel to Professor Gerould?
Probably not. Folk-lorists have long known that it is true in all
times and places for any song or bit of folk-lore that had good
elements in it in the first place or was handed on by those from
whose background of lore it could gain improving incrustations.
Anything in folk-tradition takes multiple forms, songs, ballads,
carols, dance songs, tales, proverbs-lore of all kinds. Whether
they add good or debasing elements depends upon who preserves
them and where and at what time they are preserved. That orally
transmitted songs assume the color of their surroundings, domesticate
themselves in their new environment and accommodate
themselves to the background, regional and individual, of their
singers has been demonstrated many times for many people and
many places.2
2 See works like Jamaican Song and Story, edited by Walter Jekyl, 1907, Old
World songs preserveda mong the Southwesternc owboys,w hite songsa mongt he
Negroes,e tc. An exampleo f an originally" good"p iece bequeathingg oode lements
to its progeny is "0 Bury Me not in the Deep Deep Sea," from which come the
many attractive texts of "0 Bury Me not on the Lone Prairie."
623
A Recent Theory of Ballad-Making
We come next to a pivotal paragraph.
Why should these things be? If, on the one hand, a ballad text is nothing
but an orally preserved copy of a narrative poem made by some anonymous
bard of uncertain date, how can there be in existence several more
or less mutually exclusive versions, all of them with merits of their own?
Something must be wrong with the theory, for by misquotation merely,
fine poetry, it may safely be said, has never been achieved. No: variant
texts of differingle ngths,i n whicht he sames toryi s told with irreconcilable
divergences of incident and phrase, yet finely told, can scarcely be the
flotsam of a poetic wreck. Some better explanation must be found.
Professor Child may have had the truth in mind, thinks Professor
Gerould, but neither Professor Gummere nor Professor
Kittredge nor Professor Frank Sidgwick has given the proper
solution. None of these scholars considered sufficiently, he thinks,
the phenomena of textual differences. There may be mutually
inclusive versions, all having merit as lyrical narratives, but they
cannot be put together without scrapping the virtues of the
several variants. An "original" cannot be reconstructed from
them. Surely this is a safe generalization. There are few or no
scholars that would hold of a song in popular tradition, the history
of which has been lost, that an authentic original could be reconstructed
from its multiple texts. For my own part I am inclined
to question how far any of the various scholars mentioned would
find the premises or the conclusions of "The Making of Ballads"
unfamiliar.
The point is that .... they [the variations] are inexplicable by anything
that requires us to believe in the ballad as a fixed entity and to
view the variants as mere corruptions. All versions that have been
collectedf rom folk-singersh ave equal authority,t hougho ne may be very
noble and one very base. The ballad does not exist .... except in its
variants.
This is well said but not for the first time. The underlying
thought may seem new to the author of the article but it does
not to others. It has been assumed by most practical collectors
of folk-song for many years. Surely Professor Child did not believe
that his texts could be pieced together into one authentic original
text, of which the variants he gathered were mere corruptions.
He may have printed the best or the oldest texts first, but he
prints many texts when he has them, and on an equal footing.
But whatever Professor Child did or did not believe, Professor
624
Louise Pound
Gerould would have less confidence in the novelty of his positions
had he read John Meier's work, printed as far back as 1906.3
The core of his doctrine is:
Als Volkspoesie werden wir daher diejenige Poesie bezeichnen dtirfen,
die im Munde des Volkes-Volk im weitesten Sinn genommen-lebt, bei
der aberd as Volkn ichts von individuellenAnrechtewn eisso derempfindet,
und der gegeniiber es, jeder einzelne im einzelnen Falle, eine unbedingt
autoritare und herrschende Stellung einnimmt.
He might well have read also the articles of Phillips Barry4 in
Modern Language Notes and in the Journal of American Folk-Lore.
In one article Mr. Barry defines a ballad as a "theme" treated in
many ways in many texts. Elsewhere he defines a ballad as of
"individual creation" plus "communal re-creation." Certainly he
does not think of it as a fixed entity. He gives no one text priority
over another unless he is comparing chronologically later texts
with an original still in existence. For my own part I have never
held at any time that orally transmitted texts could be pieced
together into an authentic original, nor have I thought of a folkballad
as a fixed entity. In my American Ballads and Songs (1922)
I wrote: "Traditional pieces, handed on orally from mouth to
mouth, are in a state of flux. They have no standard form but
are continually changing..... Criteria of origin for genuine
folk-song have no dependability." Professor Gerould has gone a
long way around to arrive at something that most scholars who
are not arm-chair theorists but practical collectors would have
conceded without discussion.
This explains what, I think, cannot be accounted for in any other
way; the amazing variety in ballad texts.
This variety is not amazing to folk-lorists, but is taken for
granted. Nothing else is to be expected when there is preservation
in popular tradition. The expression "communal re-creation" as
over against the old doctrine of "communal creation" has been
employed to account for and to describe it, as by Phillips Barry.
I have used it in the past to describe the multiplication of texts in
3 Kunstlied und Volkslied in Deutschland, Halle, 1906, pp. 12-26, especially
p. 14. This work is a reprint, according to its preface, of articles that appeared
in 1898.
4"An American Homiletic Ballad," Modern Language Notes, 1913; "The
Origin of Folk Melodies," Journal of American Folk-Lore, 1910; "The Transmission
of Folk-Song," ibid., 1914; "William Carter," ibid., 1912, etc.
625
A Recent Theory -of Ballad-Making
folk-transmission. Later I discarded it,5 for the reason that the
epithet "communal" has no real validity. The oral re-creation of
texts is by a succession of individual hands, not by a community.
One singer in a community makes one set of changes, another
makes another set. The same singer does not always sing a song
in the same way or with the same words. There is no community
text but many shifting texts in the mouths of many singers. And
such refashioning resulting in a variety of texts is not distinctive
of ballads, as Professor Gerould seems to imply. It is to be expected
of anything that enters into oral tradition. Folk-lorists
everywhere have recognized the variety and the "equal authenticity"
of such variants.
Let us take the next step, which follows from this quite logically, and
let us take it quite without regard to our theories as to ballad origins.
If the balladb e consideredn ot as a singlet ext, whichh as sufferedv arious
alterations good and bad, but as a group of versions, collected and uncollected,
which have circulated in oral tradition, it becomes clear that
any ultimateo r originalt ext is not only undiscoverablbe ut comparatively
unimportant. In whatever way the ballad originated, that is, it would
be submitted to the same processes of remaking, once it came into popular
favor. Provided it were in the suitable rhythm, a poem of sophisticated
origin might well, it seems to me, have a long history as a ballad, alongside
another poem that had sprung crude and simple from the excitement
of a rural festival. Both narratives would pass under the same set of
influences,w ouldb e dominatedb y the samem usicala nd poetic traditions.
"Grant this"-there is nothing new in conceding it-"and the
old quarrel between communalists and individualists seems absurd.
Why dispute about the origin of ballads if it is what happens to
them in their diffusion that really matters?"6 Have we not to
do with an instance of non sequitur here? Surely it is of value to
inquire how songs taken up in popular tradition originated, so
long as false ideas of their composition are upheld and repeated,
and so long as the ideal of scholarship remains the quest for truth.
It is not very long ago that Professor Gerould, terming himself
5 "The Term: 'Communal,"' PMLA, XXXIX (1924), 440-454.
6 Compare my "To most lovers of traditional verse, the source of a song seems
a negligible matter. The problem of its origin is of little interest except to the
specialist. The fact of popular transmission and the circumstance that generations
of singers have contributed to its modification, curtailment, or expansion, lend it
attraction." American Ballads and Songs, 1922, p. xxiii.
626
Louise Pound
a communalist and a critic of the individualist position, felt that
the question of origins did matter and pronounced those "fatuous"
who did not hold as he did. Both questions have importance for
the scholar: first, how folk-songs originate (they originate not
in the one "communal" way once assumed for "pure folk-song"
but in many ways); and, second, what happens to them after
they have started on their course in popular tradition. In a following
paragraph Professor Gerould sums up his conclusions.
I fail to see how it is possible to escape the conclusion that in certain
regions, long before the beginning of popular education, there developed
a tradition of poetic utterance that enhanced the powers common to
most illiteratef olk and made an extraordinaryn umbero f personsc apable
of putting into noble form such tales as they chose to sing ..... For a
few happy centuries, it appears, the men and women of the countryside
lived under such conditions that they could not only preserve in good
form but actually improve the stories they sang to traditional melodies.
.... This is no mystical doctrine. There was a tradition of good music
and good poetry by which the unlettered peasant was so affected that he
did not mar but rather make the ballads that he knew.
A few statements here probably need qualification or modification.
For one thing "in certain centuries" and "long before the
beginning of popular education" are too vague. The author
probably means the seventeenth, eighteenth, and nineteenth
centuries, from which most of the Child texts were recovered.
Next, the "powers common to most illiterate folk" are probably
mythical. Even primitive peoples have their professional bards.
Alexander Keith7 states matters more accurately when he writes
that folk-songs are usually recovered from a few people with
especially good memories. All collectors know that the illiterate
have not especial powers. The collector must go to selected people
for his best texts. Sometimes it is an unlettered person that has
the excellent memory, and sometimes it is a lettered person. In
general, during the period when the English and Scottish ballads
had the greatest vitality, some singers may have improved the
texts that they knew, while others may have marred them, much
as folk-singers do at present. James Rankin's garrulous versions
of Buchan's ballads may, for all we know, actually have better
represented the general popular tendencies in transmitting ballads
7 Introduction of Gavin Greig's Last Leaves of Traditional Ballads and Ballad
Airs. (Publicationos f the Universityo f Aberdeen,1 925).
627
A Recent Theory of Ballad-Making
than did the versions of special persons on whom thegreat collectors
of the early nineteenth century relied. Professor Gerould would
do well to work through the volumes of the Child collection in
order to examine what proportion of them can be proved to
have come from illiterates. The Child ballads are mainly from
manuscripts, from cultivated persons like Mrs. Brown of Falkland,
a professor's wife, from Sir Walter Scott, who retouched what he
transmitted into poetry, and from the great collectors, who also
specialized in the most lyrical texts that they could lay their
hands on. Texts from manuscripts are obviously not directly
from the illiterate, and the earlier Child texts come necessarily
from manuscripts. The later texts are mostly selected texts, the
best available, coming often from exceptional persons. The nineteenth
century collected and preserved what had special appeal
for it, first establishing definite criteria of selection. The twentieth
century collects and preserves with a minimum of selection the
bad as well as the good, and structureless songs and fragments
as well as narrative songs or ballads.
The last pages of "The Making of Ballads" are devoted to a
contrast between Appalachian versions of the English and Scottish
ballads and the Child versions of earlier date. The degeneration
of the American versions is clear. It might be added that there
is degeneration also, though in less degree, in Greig's Aberdeenshire
versions, from the twentieth century, of the texts of ballads
preserved by the great collectors of the nineteenth century. Professor
Gerould's final positions are: first, that the ballad as a
poem has submitted to processes of moulding under the influence
of a definite tradition of music and verse-making, and no sharp
division need be drawn among ballads thus formed; and, second,
that the day of the best balladry has past. The new ballads cannot
equal the old because the tradition of song making has decayed.
In the view of the present writer both positions are valid but
not new. The first should be enlarged (though we limit our consideration
to English ballads) in order to recognize that there
were a number of different traditions moulding ballads of different
types, not only within the Child ballads but for ballads that
Child's criteria did not let him take into account at all. That no
sharp divisions as to origins need be made among ballads is a
conception familiar since John Meier. The second generalization,
that the day of the best balladry has passed, would be contested
by none. It is what Professor Kittredge meant when he said that
628
Louise Pound
"ballad-making is a closed account." It is a closed account for
ballads of the Child type. Just as our present stage songs are
inferior in poetical quality to the Elizabethan stage songs and our
play-party songs of modern origin inferior to those handed down
in tradition, so our present popular ballads lack, most of them,
the old fine lyrical qualities. Fifteenth and sixteenth century
popular song on the text side, not only ballads but pure lyrics,
had a special manner that gives it high place. It is trite to point
out that Scotch song from the fifteenth century onward had
distinct superiority over song of the corresponding types in
Southern England. The special attractiveness of Elizabethan song
and of Scotch folk-poetry has been emphasized in too many
histories of English literature and by too many class-room teachers
to need reiteration. In view, then, of the high quality of fifteenth
and sixteenth century lyrics when compared to nineteenth and
twentieth century popular song, why should any one be surprised
at the large number of excellent texts coming out of Scotland and
finding preservation in nineteenth-century ballad collections?
Would the student of folk-song expect anything else, when looking
over the Child volumes, or expect later traditions in popular song
to have the same appeal?
I did not feel that I was remarking anything especially new
when I wrote in 19148:
To the present writer it seems a mistake to make style standard-giving
in a collection of folk-song. There are many who seem to hold as standardgiving
the style prevailing in the fifteenth and sixteenth centuries: all
songs conforming to these in tone and diction are "genuine"; all others
spurious .... When we contrast the older and the newer in folk-song
it becomes obvious that the superiority for persistence in the popular
mouth belongs to the former; nor is this to be wondered at. The older
singer composed for the ear; otherwise his work was vain. The newer
writesf or the eye, both wordsa nd music;insteado f professionaml usicians
we now have printing. Skill in creating memorable songs is more likely
to characterize the first type than the second. Much in modern song is
unsingablea nd unrememberablen; o one can expect it to make a deep
impression on the popular mind. In the fifteenth and sixteenth centuries
poets, whatever their class, were likely to be singers too. If we approach
popular song from the side of musical history, it is clear enough that contributions
to folk-song should be especially rich at a time when the
8 Folk-Song of Nebraska and the Central West: a Syllabus. Publications of the
Nebraska Academy of Sciences. 1915.
629
A Recent Theory of Ballad-Making
connection between composition and delivery was very close. In the
sixteenth century, song was as nearly universalized as it is likely to be
for a long time to come. Some musical proficiency was demanded of
nearly everybody, whether belonging to the upper classes or the lower.
Acknowledgmentt hat the period of the English renaissanceh ad the
more memorable style in folk-song is not the same thing however as
acknowledgmentt hat only such folk-songsa s conformt o this style are
"genuine." The making of popular ballads is not a "closed account,"
though the making of ballads or songs in the older and more memorable
style may be.
Once more, Professor Gerould's paper on "The Making of
Ballads" is an excellent essay and it deserves to be read attentively
by ballad students. It presents matters that it was well to bring
together. But I think it regrettable that he did not take into
account the fact that most of the ideas he advances were held by
his predecessors. Few, I think, among the leading ballad scholars
of the present day would have failed to concede his leading
positions before his article was written.
LOUISE POUND
University of Nebraska