Full text of "European Balladry"

See other formats

OXFORD UM VII RSI I V RK

AMKN HOI'S! , F.<‘. 4
London Edinburgh 0,l,i-<;;nw N» w
Toronto Melbourne O.ipetov, a I‘<
Calrntbt M.nir.o
H UM I’H K F V MIM-i ' R

K 1'ROPIiAN

balladry

li'. ;• * ■■

^ ^ . i \ \ i i. i \ i \\ i -.'i » j

air:: u: ■ m

Poesie ist tiefes Sehmer/en,

Und es kommt das echte Lied
Einzig aus dem Menschenherzem
Das ein tiefes Leid dmvhgluht.
Doch die hdehsten Pocsiecu
Schweigen, wie der hnchste Seime
Nur wic Geistersehatten ziehen
Stumm sic durehs gehrodme Hem

Most of all the Spanish ballad
Haunts me oft, and tarries Iona.

Of the noble Count Arnaldos
And the sailor's mystic song.

Like the long waves on a sea-beai h
Where the sand as silver shines,
Witli a soft, monotonous cadence,
Flow its unrhymed Ivric lines;

Till my soul is full of longing
For the secret of the sea,

And the heart of the grc.it ocean
Sends a thrilling pulse through me.

iwi>® m mm mnmn

PREFACE

A CRITIC has been known to complain that he could not
imagine for what sort of reader a given book was devised.
Lest this difficulty should again arise, it may be well to say that
this book counts on one reader: its author. It is an attempt to set
down answers to a number of small questions which have been
nagging at the base of his brain for several years: what are ballads ?
who made them, and for whom ? what purpose do they serve ? why
do they die ? and so on. Once the answers are fair and feateously
written down, no one will have motives more keen than those of
the author to see whether they stand their ground when the en¬
thusiasm of invention has died; no one—because past experience
promises that these questions, insufficiently answered, will raise
their heads and nag, nag again.

The author would have been contented to accept any complete
solution. At first indeed, it seemed sufficient to repeat the words
of the masters whenever the course of professional duty brought
up the topic of ballad poetry. (The field happened to be Spanish
ballads, where the masters are insuperable.) Curiosity leading the
writer to follow up some references to the ballads of his own
tongue, he realized that there were other masters holding very
different language. There are, perhaps, no two ideas in common
between Sr. Menendez Pidal and Andrew Lang on this subject.
So new questions raised their heads: are all ballads really the same
thing? are the different explanations compatible within some
larger answer ? The author nibbled at the cheese in some articles on
points of detail, and then it seemed possible to cut a way through
to a solution (since a conviction of ballad oneness was steadily
gaining strength) by comparing the evidence of some of the greatest
national balladries on a few fundamental problems: this was the
purpose of an article in Medium Aevum, vol. i, a comparative study
of Spanish, Danish, and Yugoslav balladries. What effect that
article had on its readers, if any, is concealed from the present
writer: on himself it had the disconcerting effect of reve alin g the
insufficiency of his data. To answer the very simplest inquiries
about the nature of ballads, it seemed necessary to traverse all the
ground of dispute. All the ballads of all the nations or communi¬
ties in Europe have some evidence to contribute. This was the

vi PREFACE

point at which a prudent man, mindful that only specialists can
be truly scholarly, would have desisted from the enterprise. It
would have stopped, but for the nagging of the little questions
themselves, which would not be at peace without an answer.

At this point the author-^a-reader, who is a personage at
least as distinct as Launcelot Gobbo’s conscience, and so able to
enter into a dialogue with the writer, came in with his fallacious
assurances.

The task is not herculean. The ballad is, so far as you are con¬
cerned, a definitely European and medieval phenomenon; and it is
possible to acquire a reading knowledge of the languages, especially
as ballad vocabulary and style are intrinsically simple. You do not need
to read all ballads. So long as you handle really representative collec¬
tions, it will be safe to let hundreds stand for thousands. Keep your
questions perfectly simple; look for the answers with your own eyes,
not forgetting to consult all available authorities. You should not be
seriously in error, if you describe what you have actually seen for your¬
self ; and the truth, if not precisely what you make it, must be of that
kind.. 5

It is to be feared that the author-gmz-reader was optimistically
misinformed about the difficulty of the undertaking. He did not
know how many languages there are in Europe, and that the least
accessible are often the most abundant in ballads. It did not occur
to him that the songs, in addition to being old in speech, were
often maddeningly dialectal, so that the dictionaries were always
liable to fail his needs. It was not possible to envisage such a case
as the Erlangen manuscript of Yugoslav songs which, in addition
to being Hercegovinian (for whatever dialectal difficulties that
might imply), have been copied out by a German with a copious
supply of German mispronunciations; nor could one imagine how
hard it is to read Greek ballads, in the vulgar language, by the aid
of dictionaries all anxious to demonstrate the likeness of modern
to ancient Greek. Let it be admitted that the rule of trusting only
one’s own eyes has been broken in the case of Finnish and
Esthonian ballads, lest this inquiry should be spun out to infinity.
For Hungarian poetry it has been necessary to rely too heavily on
report, since the material that came to hand was generally not of
the right type. Into Gaelic balladry, if the term be rightly applied,
the writer has gone only far enough to be aware how slippery that
ground may be. There are still gaps, but nothing, I hope, which

PREFACE vii

may invalidate the inquiry. The authority of half a dozen funda¬
mental stores of ballad poetry should be enough to guarantee the
soundness of the general plan.

Temptation was offered to carry the inquiry outside Europe.
Perhaps some one else may do so; but for the present it seemed
that Europe is a sufficiently homogeneous mass, and that the
movement of ballads within Europe could be related to place and
time, so that the whole study could have the solidity of history.
European ballads are such and such; they resemble each other
broadly so; they rise at approximately similar times, under
approximately similar circumstances; there has been a traffic in
certain motifs, so marked that one can establish the trade routes;
the motifs are often associated with certain facts which give us a
chronology of sorts. Poems rather like ballads have been composed
and sung outside the Ural frontier: the Confucian odes, perhaps
old Arabic raiding poems, Armenian and Caucasian folk-verse,
and the ballads of Rajputana, of which my friend Dr. Kalidas
Nag has made me aware. All these things might have come into
the picture had I been concerned to discuss balladry in the
abstract; but my problem was more concrete and historical, and
in any case there was the necessity of bringing the work to some
end, especially when it was composed on the lip of the erupting
volcano of Europe.

So this book is an attempt to write down answers to questions
which have perplexed its author; but he does not suppose himself
so singular that no one should be found to share his wish for greater
assurance. Indeed, he has had evidence to the contrary. In the
course of pondering and puzzling, portions of this book have
bubbled over in the presence of friends of quite a number of
different nationalities, and they have seldom been treated with
indifference. Members of small nations with rich balladries have
expressed delight at the recognition of their genuine worth;
others have been attracted by the width of the theme, or have had v
an interest in some special aspect. A critic, who believes ballads
on the whole to be very small beer, exempts Sir Patrick Spens ;
and one does not have to ask many questions before finding that
almost every sensitive person fondles in his heart some favourite
ballad: Tam Lin, La Pernette, Count Arnaldos, The Maiden of
Kosovo, Unter der Linde, Holger Danske, Constantine and Arete,
Dobrynja’s Return, or Sadko’s Voyage. There are those who like

viii PREFACE

the music or the dancing, or the racy language of peasants, or the
memories of times gone by, or stark realism or wild magic, but
there is something in ballads which stirs every one. Nearly all i
have talked to have found some one of the plain little questions
reallv interesting, and have raised the writer’s hopes by encourage-
ment But most of all, I owe thanks to the Oxford Medieval
Society who received the pith and gist of this argument with
kindlv encouragement at a meeting in October 1937-

Mr. John Goss, in his pleasant Ballads of Britain , has com¬
plained of "the professors and minor poets who have made this
subject their own’. Though a member of one of these depressed
classes, and not having risen so high as the second, the author
hopes to have added a little to the pleasure found, in reading or
hearing ballads by mentioning their wider associations. Lady
Isabel and the elfin Knight is fine in itself, but it calls up such rich
associations when one thinks of Rico Franco and Frere Renaud
and Ulinger and the original Dutch Halemjn, all so strangely
begotten, as it would seem, by the old story of Judith and Holo-
femes. One may be impatient with the Suffolk Miracle , and
wonder that so strong a plot could be so tamely told; it is surely
a gain for us to read it in the Greek Constantine and Arete in the
fullness of its supernatural horror. There may be some who find
these associations no addition to their enjoyment. An English
critic has written that, to describe ballads, the English ones
will suffice; a Spanish one tells us that no nation has anything
comparable to the ‘Romancero’; and a German writer claims for
the Wolkslieder 5 that they are entirely German. For those who
must have their pleasures in secret I can do nothing. For my part
I am fascinated by the spectacle of all Europe creating song and
enjoyment, without ambition or rivalry, and readily imparting
the best creations; and it is with amazed respect that one hears the
heroic singing of peoples so often shattered, destitute, and des¬
perate. Ballad comparisons give ground not merely for pleasure
of a gayer or graver cast; they become important when we notice
how, in land after land, ballads have educated national conscious¬
ness and literary taste. There are countries whose very existence

depends on ballad traditions, and many a great master in many
a land—Cervantes, Shakespeare, Goethe, Oehlenschlager, Cole¬
ridge, Heine, Rimski-Korsakov, and others—is what he is partly
because of ballads. Were the racial and literary importance of

PREFACE

ballads throughout all Europe generally realized, Englishmen
would surely take more than a fleeting and occasional interest in
their own very rich heritage.

In writing this book I have contracted many and deep debts
and had much encouragement from friends. Doubtless, from one
pomt of view I ought to have contracted still deeper obligations,
by handing over the sections of the work to experts. Almost every¬
where in this wide European field, my position cannot but be that
o an amateur, running the risk of heavy censure from the experts.
But I saw no way of determining the proportions of the study
save by passing all through one mind; and, apart from the diffi¬
culty of finding authorities on many of the remoter fields among
English scholars, it seemed to be of the essence of the inquiry
t at the questions should be kept simple and always asked in the
same sense. How difficult that is, only one who has tried it can
know; but any alert reader is aware that symposia seldom keep
to one path and one manner. In any of its separate chapters this
study does not claim to compete with expert authority; but it is
hoped that the specialists will be guided when they step off their
own ground, and that they may rephrase some of their problems
(and perchance conclusions) in the light of a more general experi¬
ence. rhe labours of the great collectors and critics are the first
grounds for my gratitude. Sr. Ramon Menendez Pidal’s books and
articles I think I have read nearly all of them—have been a model
and stimulus to me as to all other hispanists; our indebtedness is
not less deep if we have sometimes been tempted to cut our teeth
by disputing one or other of his conclusions. I owe a great deal
to my fellow members of the Medieval Society, and to colleagues
at the Taylor Institution: Professors H. G. Fiedler, G. Rudler
A. Ewert, J. Boyd, C. Foligno, S. Konovalov (on many occasions)'
Dr. J. Bostock, Miss Olga Bickley, and to Professor R. M.
Dawkins and Mr. N. Coghill, my companions at Exeter College.
Professor Norman, of London, has given me some useful pointers
and I am very much in the debt of Mrs. Chadwick, not only for
her share m the monumental Growth of Literature, with its
admirable chapters on Russian and Yugoslav poetry, but also on
more private occasions. Sir William Craigie has allowed me to
consult his Icelandic and Rumanian books. Dr. L. F. Powell, of
the Taylorian Library, has indefatigably warned me of books I
ought to know. I am grateful to comrades of the P.E.N. Club

s PREFACE

also, and notably to Professor Semper for advice as to Esthonian
songs, Mr. J. Karklins for the gift of a splendid collection of
Latvian ‘dainas’, and Professor Vladeta Popovic for much en¬
couragement and the gift of his copy of the Erlangen poems,
without which my outlook on his country’s poetry would have
been sectional. Professor lorga Iordan, of Iasi, kindly gave some
suggestions about Rumanian. Professor Myles Dillon advised me
to read certain things on Irish traditional songs, though I have not
ventured to incorporate Gaelic evidence into this work. I am
grateful also to Dr. Kalidas Nag, of Calcutta, for suggestions con¬
cerning Indian ballads, which I have not followed up. ‘Forse altri
cantera con miglior plettro.’ In naming these names, and that of
my dear friend and constant encouragement Professor Morley, I
desire to exonerate them one and all from any responsibility for
the errors I may have committed. That responsibility must be
mine.

W. J. E.

CONTENTS

Book L BALLADS IN GENERAL

I. People and Poets ....

II. What is a Ballad? ....

III. Performance .....

IV. Tunes ......

V. Kinds and Dates ....

VI. How Ballads Spread ....

VII. The Descent of Ballads

VIII. The Ascent of Ballads

Book II. BALLADS IN PARTICULAR

I. Romance Ballads

1. France, Provence, North Italy, and Brittany

2. Central and Southern Italy and Sicily

3. Spain, Spanish Jewry, Portugal, Ibero-America

4. Catalonia .....

II. Nordic Ballads

1. General Considerations .

2. Scandinavia ....

3. England, Scotland, America .

4* Germany and the Low Countries

5. Czechoslovakia, Hungary

6. Lusatia, Poland ....

7. Lithuania, Latvia ....

8. Esthonia, Finland ....

III. Balkan Ballads

1. Asia Minor, Greece ....

2. Yugoslavia, Bulgaria

3. Rumania .....

IV. Russian Ballads

1. Great Russia

2. Ukrainia .....

NOTES . . .

INDEX .

109

132

148

152

192

*95

208

228

242

272

277

284

293

302

3 20

345

354

375

381

393

BOOK I

BALLADS IN GENERAL

PEOPLE AND POETS

I N a vivid passage the Russian collector Rybnikov has spoken of
the moment when he first encountered a community in which
ballad-poetry was still alive. He had for a long time sought this
meeting, but had been foiled by the shyness of the singers, since
ballad people are slow to expose their intimate treasures to the ears
of strangers. At last his wanderings took the collector to a hut by
the shores of Lake Onega. There were peasants in the hut, and
conversation and fatigue made Rybnikov drowsy, so that he began
to doze by the fire.

Being warmed by the fire, I gradually fell asleep. I was awakened
by strange sounds. Up to now I had heard many songs and religious
poems, but such singing as this I had not heard. Vivacious, fantastic
and gay, now it grew quicker, now it slowed down, and recalled by its
tune something very long ago, forgotten by our generation. For a long
time I was unwilling to awaken, and listened to every word of the song—
so happy was I to remain totally overpowered by this new sensation. 1

These were the first sounds of the popular ballads which grew,
under his hand, into one of the richest of the Russian collections.
The people who cherish and enjoy the ‘byliny’ have since been
more fully described. 2 They are free peasants living in the rude
and invigorating tracts of Russia lying to the north of the great
forest belt. The landscape is rough, mighty, stately, and of
fascinating primitiveness. The distances between villages often
amount to hundreds of versts, and the intervening spaces are
diversified by swamps, lakes, rivers, and rocks. In autumn and
spring the peasants are often separated from each other for weeks.
A rigorous climate compels men to work strenuously in the few
months at their disposal; fishing and hunting cease when the long,
white winter drives them to seek the warmth of their wooden houses.

1 -Apud H. M. and N. K. Chadwick, The Growth of Literature, Cambridge,
1936, ii, pp. 239-40.

2 F. Trautmann, Die Volksdichtung der Grossrussen, Heidelberg, 1935, i

pp. 17 ff.

2 PEOPLE AND POETS

The inhabitants have a natural sense of form and a ready in J“^ 10 ”’
and thev are thrown on their own resources for entertammen.

Storv-teningandsingingarethedelightofthepeasants.fewofwhom

(none of the good singers) know how to read or write. ^ools are

few; the nearest rail-head is sometimes hundreds of miles distant,

and a vast forest zone sunders these regions from the rest of Russia.
This belt, and the rigours of their life, have saved them from e
serfdom which has cursed the easier lands to the south, preserving
their independence and initiative. The singers are men of the same
class as the listeners, but have acquired fame over wide areas for
their gifts of memory or voice. Frequently they have been special-
ized to their profession by some natural calamity, such as blind¬
ness, which has lessened their usefulness in their proper trades.

Across the width of Europe and the Atlantic from these Russians
it is possible to encounter another ballad folk, of our own blood. 1
The mountains of Virginia and North Carolina lie, indeed, at no
great distance from flourishing centres of American civilization,
but are scarcely less isolated than the peasants we have described.
No commercial prospects exist to entice traffic by rail and road.

The roads are few, and rough; when they peter out at some mission
school, the traveller must take to horse. A primitive cultivation
and soil erosion make the winning of livelihood a hard task ; after
a severe season, famine cannot be staved off. The log-cabins lie
at distances apart, but relative prosperity confers on certain per¬
sons the English title of ‘squire’. There are few social services;
none maintained by the districts themselves. In such a community
old customs survive, together with traditional turns of speech and
the words and airs of old ballads. The ballads are sung unaccom¬
panied, or to a ‘dulcimer’, and there are persons in the community
who have a superior repute for their repertoire. Transmission is
from mouth to mouth. Sometimes the words are jotted down as
an aid to memory, and note-books containing such jottings are
‘ballet’ books. To ‘have the ballet’ of a ballad is to have a written
text, but these texts are not for circulation, still less for printing.
The ballads, or ‘love songs’ as they are called, are prized but not
praised. Any new jingle from the towns causes more admiration
for a while, and a certain disrepute attaches itself to ‘love songs’ in

1 Described by Dorothy Scarborough, A Song Catcher in Southern Mountains ,
New York, 2937, An older account is that given in C. J. Sharp, English Folk-
Songs from ike Southern Appalachians, Oxford, 1933. There are several others.

PEOPLE AND POETS 3

the face of hymns and moral poems. Patience is needed to win the
confidence of those who know ballads, to overcome their coyness.
Once they have been persuaded to sing, they bring forth things old
and new. Not only are many of Child’s ballads still alive in the
American mountains; but the use of old scales proves that the tunes i
are equally ancient. The historical sense is vague, though there'
are some pieces recognized as ‘way back yonder songs’. By way
of compensation, the sense of immediacy is such that traditional
ballads are given a new setting as events within living memory. In
Scotland the details of The Douglas Tragedy have been fitted to the
Yarrow at Douglas Craig; but Miss Scarborough encountered in
Virginia a ninety-year-old gentleman who claimed to have been a
witness. ‘The Seven Sleepers (he told her) was a true song. It
happened way back yonder in Mutton Hollow. I was there myself.
Somebody got killed over the girl. I was there soon after it
happened. Another man was after the girl and one man shot him.’
Frequently the reciter’s memory fails, and mention is made of some
relative who had known the song much better. At other times, one
will exclaim, ‘Oh, if only I were driving the cows home I could
sing it at once!’ so closely are the words of the song associated with
some feature of the singer’s usual vocation.

Crete and Asia Minor have proved fruitful sources of Greek
ballads, thanks to their remoteness; in Spain the Asturian valleys
have given asylum to archaic versions. The remoteness of the four
East Baltic states is linguistic rather than geographical. Shut in to
the use of their own resources, they have each developed copious
and versatile balladries. In the larger states of western Europe
social groups may approximate to the conditions of a ballad people,
since their interests may be centred on themselves as a well-defined
community. Sir William Craigie has told me that at St. Andrew’s
University in his day it was customary for students to gather out¬
side the mathematical class-room before a lecture, to sing for half
an hour. Among the songs one might encounter genuine, if late
plebeian, ballads, such as Clementine or Riding down from Bangor.
The gathering of these and other songs into Students' Song Books
seems to have had the untoward effect of converting students’
songs into concert pieces, having burst open the oral tradition
which was their firmest guarantee. In the United States new
ballads have arisen out of the conditions of cowboy life and among
the negroes. The great conscript armies of Europe create self-

PEOPLE AND POETS

centred communities, depending for entertainment on
and ears. The German armies have been particularly rich y
and ballads. In those of Austria the same events °r reourre
situations are recorded in German, Czech, Hungarian, ’

and Croat songs; and there is a soldiers’ section m the ballad col¬
lections of Poland, Esthonia, Latvia, and Lithuania, due to service
in the Russian army. The ballads express the peasant s dislike
of conscription and" the exile it causes, the march to a rontier,
some desperate defence, the soldier’s dying testament associated
now with one battle, now the other, the sad or joyful return
home. These professional songs are the direct descendants o
those sung by reiters and landsknechts in the sixteenth and seven¬
teenth centuries, and by Swiss pikemen in the fourteenth and
fifteenth.

The Faeroe Islands may be classed as a remote community, but
their flourishing cult of the ballad in our day has this additional
feature: it is common to the entire population. 1 New songs are
required for the festival of the patron saint and other important
dates, and they are multiplied by the convention that the same
words should not be used twice in a year. The demand for new
mutter is satisfied by importing ballads from Denmark and Norway,
either in the original or in translation, and by excavating new
ballads from Icelandic sagas. As late as the year 1830 a bonder
named Djurhuus composed his Long Serpent (Ormurin langi) from
materials in the Saga of St. Olaf and it was not committed to
print until 1884. Curiously enough, the equally remote and self-
sufficing community of the Icelanders has shown indifference to
the ballads, being wholly devoted to the cult of an elaborate
rhetorical form: the < rfmur\ 2 The poetry of the Gael has also
developed under a strict discipline in bardic schools. These were,
says one authority, 3

the university system of the nation—granting degrees, or what corre¬
sponded to such, and bestowing privileges on both professors and
students simply because they were professors and students.

He goes on to remark that ‘one searches Europe in vain for the

1 See the classical description in V. U. Hammershaimb, Fcemsk Anthology
Copenhagen, 1891.

2 See Sir William Craigie’s Taylorian lecture, The Art of Poetry in Iceland,
Oxford, 1937, and Skotlands Rimur, Oxford, 1908.

1 Daniel Corkery, The Hidden Ireland, Dublin, 1925, p. 63.

PEOPLE AND POETS 5

equivalent of our bardic school system’. It lasted in Ireland until
the disasters of the seventeenth century, and was even then pro-
onged by the Courts of Poetry which met in taverns surrepti¬
tiously to maintain the prestige of poetic art. In Scotland there
were bardic schools in the eighteenth century. This highly
academic atmosphere did not prevent poems entering into oral
radition, so as to be recovered in our days on the lips of ‘illiterate’
peasants in Kerry and Galway. It is not impossible for a traditional
culture to prefer elaborate to naive art; but such instances lie away
trom the beaten track of this book. J

Of another ballad people we have had detailed descriptions in
recent years: the Montenegrin Serbs, who treasure and still com¬
pose ballads. The minor particulars are not of general application,
and may be omitted here, as also the possible relation between this
hvS ! nd n that gaVe rise t0 the Primitive (and largely

t? n 3 } epi A S , ° f Germany ‘ In the hi § h P laces of Montenegro
and the Dmanc Alps a heroic form of poetry and society has per¬
sisted in vigour to our day. The social unit is small; it is the tribe
or family state. Wealth, commerce, and elaborate hierarchies
require an ampler space for their growth, and in fact have come to
possess the Serbian river valleys. But in the mountains, the
criterion of aristocracy is personal prowess; allegiance is given to
a leader who joins birth to valour, and it is perturbed by personal
euds. Raids on traditional enemies, vendettas, marriage under the
hgure of bride-stealing, and calamitous battles make up the stuff
not only of entertainment, but of instruction. Where all are poor’
the poverty of the poet is not exceptional; his status is acknow¬
ledged, and his improvisations are the acknowledged standards of
achievement m the group. Like a bard or a skald he has his allotted
place beside the tribal leader, and is a master of a definite technique
though it is not erudite like theirs. On the plains the heroic ballad
is no longer at home, and traditional songs are cultivated chiefly by
t e women in their dances; by women, among whom traditions
linger longer than among men.

The authority of the Montenegrin ballads encourages us to take
at^ their face value those social descriptions implied for countries

und l Iel K enlie o eS b “ den Serbokr °aten’, M. Braun

O. T h ' ’ Heldenlied, both m Beitrage zur Geschichte der Deutsche*

Sprache und Literatur lix, 1935; M. Murko, La Poesie epique en Yougoslavie au

’ I929; W - WtoSCh> Hddm ^ * SUdostZopa,

PEOPLE AND POETS

inhabitants of the Blue Mountains of Virginia, but* d J r

was more organized and more equal to their age.

ballads the people is no proletariat. The pro^ is ^

tioned; the people consists of those bold-handed, m

grouped round headers like Johnny Armstrong and WiHi ^

L aristocracy of their own sort of world. The Spanish fiontier
ballads, also, imply the people of southern f °^ es ® ’ J lonso
persons as Bishop Gonzalo ofJaen, Fajardo of Murcia Don A o
de Aguilar, the soldier Sayavedra, &c. It ^ ^wdent tha the

American ballads have suffered‘deminutio capitis . Danish balkds

now crooned above cradles or danced by peasants in ^ms were
formerly the delight of the squire’s garth The ^al grand
would lead the dance, and the king might be an onlooker, f
thought the queen safe in the palace. It is the society of the ideal
republic: the number of persons within range of a leader s voic .
It is true that ballads mention more august ranks, and have even
a fondness for kings and emperors. But they describe them as
magnified squires, all their acts and feelings being on a personal
scale. So we have those intimate, charming, and unreal scenes:
Queen Bengerd, waking up in the morning, and demanding her
morning gift, which is a general persecution of the Damshbonders;
Charlemagne, told of his son’s misdeeds, and agreeing to have him
punished at once; Niels Ebbeson protecting the Danish frontier
for purely private reasons; the king talking to Robin Hood as
easily as Robin to Friar Tuck. In the Spanish ballads of Monte-
sinos we are told that the hero rode to Paris from the Pyrenees
(whence the city was visible!) to avenge his father’s wrongs. He
rode into the royal courtyard and strode into the chamber. A
game of chess was in progress between the king and the traitor
Tomillas. Montesinos lounged by, as an onlooker. He saw
Tomillas was cheating, snatched up the board and beat out his
brains. The king was annoyed, but a few words of explanation put
things right. In the French epos of Aiol, on which these ballads
depend, it is not considered possible for unknown young men to
gate-crash on kings in this way. Aiol has to gain the king’s favour
by feudal service and prowess, and he has to collect a party which
will enable him to make headway against Macaire and all his
numerous and powerful lineage. Aiol, in fact, has national pro-

PEOPLE AND POETS 7

portions and is set in complicated feudal conditions; the Montesinos
ballads imply a more informal and restricted community.

A combination of these snapshots will provide a recognizable
portrait of the ballad people. Minor differences have already been
apparent, but there are a few common features which are constant.
The community supposed by ballad poetry is small, stable, and
self-sufficient. In the east it is composed of the Slavic ‘knjaz* and
his ‘druzina’ or body-guard, and the latter came to be fixed at a
conventional total of thirty in Serbian tradition. To build the
Bridge at Arta, according to a Greek ballad, there gathered

Five and forty master-wrights, apprentices full sixty.

A Spanish ballad begins

See from Jaen sally forth a good four hundred gentlemen.

They are attached to particular places: Branxholm, the Sherwood
Forest, Antequera, Kiev, Senj, &c. Despite all the activity dis¬
played in the ballads, they relate no great movements of peoples;
nothing like the Germanic Migrations, the settlement of Iceland,
or the Crusades. The numbers involved in some of these move¬
ments may not have been greater than those of ballad poetry, but
the scale of their actions corresponds to epos or saga, not to ballads.
Nationality is too big a concept for people so tribally organized.
One encounters instead simple antitheses which explain the recur¬
rence of the same sort of action in ballad after ballad. There are
the Scots and English of our Borders, naturally at feud; Moors
and Christians; Serbs and Turks; Holy Russian ‘bogatyrs’ and
pagans, indifferently drawn from Lithuania or Tatary. Doubtless
these conflicts have amounted in fact to formative influences in
national history, but the ballad poets see them only episodically, as
they concern individuals. The great ‘geste’ is broken into frag¬
ments. The career of Ruy Diaz de Bivar gives 205 different narra¬
tives ; the disaster to the Serbs at Kosovo is recounted as it affected
Milos Obilic, Prince Lazar, the nine Jugovici and their mother, a
nameless maiden, but there is no attempt at a conspectus on a
grand scale, such as occurs in the heroic epics. These small com¬
munities are self-centred and self-sufficient, attached to their own
soil by instinctive patriotism, and led by leaders who command
their personal devotion. They have a lively intelligence, as their
metaphors show. In the best periods there is restraint and decency
in the ballads, however tragic or amusing the pieces may be.

g PEOPLE AND POETS

Criminal themes and vulgar jesting are signs that a given ballad
tradition is on the wane, through the withdrawal of the best ele¬
ments of society. Singers and hearers are normally analphabetic.
‘Illiterate’ is no term to apply to the creators of so much exquisite
literature; but theirs is a literature for the ear, not for the eye. i he
spread of the habit of reading—the habit, rather than the mere
ability—is everywhere a principal cause of ballad decline. Gradu¬
ally more and more members of the community seek their pleasure
away from, the ballads, which fall to the exclusive possession of the
vulgar. In the best periods the ballad people is homogeneous.

The relationships maintained wdthin such societies are entirely
personal. ‘A man’s a man for a’ that 5 in ballad poetry, and he is
found to act upon the simplest and most universal motives. The
sphere of his activities is uniform, and even monotonous. Feuds
arise out of quarrels among tipplers, brides are stolen with the same
precautions, duels are between puny heroes and gigantic villains,
there are heroic massacres of a common pattern, love comes at first
sight and is irresistible, and a simple ethical code distinguishes
right and wrong without the sanction of religion. These repetitions
are not due, one feels, to poverty of imagination, but to a feeling
that everything has its due ritual and form. An event, whatever its
actual circumstances, readily pours itself into one of the established
moulds, using the appropriate form of words. Professor Popovic
has told me that ballads on the Great War already circulate in the
hills of Montenegro. The details are those of ballad tradition:
Tsar William writes a letter to Tsar Peter demanding submission
or tribute. Tsar Peter receives the letter, says not a word, springs
to his nimble feet, and mounts his horse, taking lance and pistols.
The rivals meet and exchange their blows, and the Serb is inevit¬
ably the victor. A pattern of this kind seems to run through these
efforts to express poetically a recent event which is known to have
taken quite another form. Vuk Stepan Karadzic has told us how
one of his informants was suddenly moved to compose a poem on
the scuffle for which he had to flee from his home. In place of
a mean affray with a Turkish gendarme, the rhapsode produced
a full-dress duel with a gigantic pagan adversary, first with the
lance, and then with pistols. Each thing has its proper rhythm.

This rhythm and these conditions of society may be truly de¬
scribed as medieval. Ballads do not begin anywhere in Europe
until the mass migrations have subsided and left people engaged

PEOPLE AND POETS 9

by small groups in occupying the soil. Actual dates, as we shall see
later, can be assigned for the beginning of some of the cardinal
collections. Medieval England, writes Mr. H. S. Bennett, 1

was almost exclusively rural England, and rural England congregated
m small groups of people, here fifty, here a hundred, and (much less
°ft en ) here several hundreds, living in rude houses which clustered
together in some places, while in others they straggled endlessly down
the village street. Agriculture and its allied occupations engaged the
energies of everyone as they strove to win a living from the soil. These
small groups, on closer examination, are often found to be sharing
certain rights and privileges and to be discharging certain duties, under
the protection and control of a lord. There were many thousands of
these little groups throughout England, and they formed the manorial
society which existed for some hundreds of years after the Conquest.

The bonders of the Danish songs were similarly grouped round the
garth of a lord. The borderers, in Scotland or Spain, and the
meinees of Serbian princelings were like this, save that the occupa¬
tion of those worthy to be mentioned in poetry was war, not agri¬
culture. Such conditions did not commence everywhere in the
same age, and the dates of ballad beginnings vary correspondingly.
One has to allow also for the fact that the ballad is an art form,
which does not arise spontaneously, but more often has some
literary antecedents.

The end of medieval society comes with the practice of reading.
This was due in the West to humanism and the art of printing.
Books, suddenly much more plentiful and cheap, drew to them-
selves the middle and upper classes, abandoning the ballads to the
vulgar. Even then, however, ballad communities might survive,
though on a humbler level, thanks to difficulty of communication,
local backwardness, or some cause like professional segregation.
The conscripts of the great European armies, as we have seen,
prolong in some measure the medieval self-sufficient, homo¬
geneous, analphabetic ballad societies. In the Balkans and Russia
the end of medieval conditions has come much later. The lettered
public is relatively small. One notices chiefly the drying up of
patronage in the wealthier centres, where probably the ballads were
first developed. The songs are scattered to the circumference of
culture. Those of Kiev, for instance, having been transformed by

1 Life on the English Manor: a study of peasant life, 1130-1400, Cambridge
1937, p. 41*

461s

I0 PEOPLE AND POETS

Muscovy, are now to be encountered in the far north, sheltered
bv the deep forest belt. In such places composition still goes
on Professor Dawkins has shown me a curious instance in th
Cretan ballad of Daskaloianni, printed roughly in a smallpopubr
pamphlet. The poem is of more than a thousand lines, an g
with an elaborate invocation:

Dear God, now grant enlightenment, a heart like to a cauldron,
to me as 1 commemorate the schoolmaster Johannes.

Dear God, wit and ability grant as I make beginning

in ballad-wise to sing a song of that renowned teacher. .

Dear God, now grant me memory, within my head put wisdom,
here to compose and utter forth the sad affair at Sphakia.

So he continues, with an all too common tale of Turkish oppres¬
sion and wild revenge. The affair occurred in i 77 °- n *7
Anagnostis, son of the priest Joseph Skordylis, took it down from
the dictation of the illiterate poet, writing a little each day; he
wrote at Papura, beyond the Giverti, where he was a dealer m milk
and cheese.

And it was I that held the pen, ’twas I that held the paper,

but he dictated me the tale, and word by word I wrote it,

for he dictated me the tale of the schoolmaster Johannes,

and ever tears streamed from his eyes, as he the deed remembered,

till all his speech was broken off, his narrative was ended,

and all his inmost heart was poured in black and bitter groaning.

The poet, so deeply afflicted but (it must be confessed) not skilful
in conveying his grief, was the ‘barba’ Patzelios. Another poet in
the same pamphlet was George Pateros. 1

Such poets were not professionals with a high standard of train¬
ing. Any one could compose a ballad who knew how to express
events in the ballad manner. King James the First is among the
reputed authors, Fray Ambrosio Montesinos, a future bishop, and
Gil Vicente, a goldsmith of Lisbon. Many another name is known,
chiefly hacks. The scholars of the bardic schools who forged and
preserved the Ossianic poems and underwent a kind of university
training cannot be deemed to have written ‘ballads' within the
same sense of the term, nor can Icelandic ‘rimur’ or poems in the old
Norse Edda be fairly called ballads. Such work is too specialized.
The vocation for ballad poetry is more intuitive and casual. To
lead a Bulgarian round-dance the precentor must be distinguished

1 Kretikai Rimai, Athens, *Estia, 1888, edited by Emm. Bardide, Cretan.

PEOPLE AND POETS

Spun!

dav T 1 d b d ^ aSOnides have ^ny descendants to this
implies another kind of speciaSltion f,! ’ Pllgnm ’ wh ^h

=‘tr ,io r the smaI1 pa,riarcM enjoj;,

Dobrynja a, the court of 4, Golden Prince VMndr of Key

I* 16 ,.! 1 * 16 minstrel Paid no heed to their words
the little minstrel took no notice of their words’

then burst into speech the litde minstrel:

Ah thou Vladimir, royal Sun of Kiev!
where stands our place, reserved for the minstrel?’

1 hen answered Vladimir, royal Prince of Kiev
tour place, reserved for minstrelsy
is at the stove there, behind the stove.’

I he h tt i e minstrel was not squeamish about that place

but jumped upon the glazed oven. P ’

Then he played on his furious gusli
even on that glazed oven;

Dobrynja played in the manner of Kiev,
and played in addition in the manner of Tsararad
concerning the old and the young 8 ’

he played, naming their names.

All those at the tables fell a-wondering,
all to his playing gave closest hearing, ’
all those at the tables burst out saying:

PEOPLE AND POETS

‘But here is no bold minstrel,
it is a stout, worthy young warrior,
a mighty bogatyr of Holy Russia.

Then spake royal Vladimir of Kiev:

‘Ah thou bold minstrel!

Get you down from the oven, behind the oven,
we give thee three places, the three choice ones:
one place here beside me,
another place facing me,
a third place where thou wilt.’ 1

No doubt Dobrynja was more fortunate than most working min¬
strels, who probably did not enjoy the advantages of silken strings
and golden chords, but his advancement in the profession shows
that the minstrel’s place at the feast was known. ^ ^

Poets and poetry of this sort might justly be called ‘popular’.
The word does not involve the slur of vulgarity; plebeian ballads
belong to the decline of the art, but in the best epoch the whole
people sings. Nor need the word imply adherence to the mystical
doctrine of the people’s authorship, about which much ink has
been shed. 2 The audience, drawn without abstentions from the
whole community, conditions the minstrel’s performance as reciter
and creator. The matter belongs to them all, and any one who
knows a better version may produce it. After a song has been sung,
the audience falls into a discussion, and mention may be made of
variants. 3 There is no professional wall separating the performer
from his hearers; he is no more than ‘primus inter pares’. If he
composes a new song, he must meet the people’s expectations.
The course of the tale must have the prescribed order and formulas.
The art is greater than the artist, who must not show his hand if he
is to be believed. Karadzic says somewhere that Yugoslav minstrels
claim to hand on from tradition even songs they have composed
themselves about the events of yesterday; if the German and

1 Gil’ferding, Onezskija Byliny, St. Petersburg, 1873, pp. 43-4; cf. N. K.
Chadwick, Russian Heroic Poetry, Cambridge, 1932, pp. 87-8.

2 Most recently between Mr. John Goss and Sir J. C. Squire in the former’s
Ballads of Britain, London, 1937. Mr. Goss puts his case in such general terms
that there is room for both disputants to be right. The best treatment of this
subject is R. Men&idez Pidal’s lecture, Poesia popular y poesia traditional,
Oxford, 1922, to which I shall recur in the next chapter.

3 ‘She was firm in her notion as to the correct way of rendering songs, and
when mention was made of a version differing from hers, she would say hotly,
“That ain’t right. This is the way it goes.” ’ (Dorothy Scarborough, A Song
Catcher in Southern Mountains, New York, 1937, p. 61.)

PEOPLE AND POETS x 3

French songs of the trades are wont to declare the poet’s profession
and claim that the song is new, that also is a matter of formula,
since the claim to be new adheres still when the ballad has become
very old.

In this way ballads come to be the completest definition of the
community which enjoys them. One must allow that not all things
are considered suitable for verse; but that which finds expression,
finds it in the commonest manner. If we wish to know what sort
of mind is the Spanish one, it is of less advantage to consult Don
Qmjote, which is a work of exceptional genius, or the ‘comedias’,
which were Castilian and of a certain epoch, than the ‘romancero’.
The romances’ are at home not only in Castile, whence they
sprung up, but wherever Castilian is spoken: in Mexico or Chile,
or among the exiled Jews in Oran and the Balkans. They have
spread into all the dialects, and into the languages akin to Castilian.
So the same ballads and ballad-types are encountered in Portugal
and her islands, and in Brazil. In Catalonia ballads are Spanish,
save for an older French stratum. But when these songs travel
farther, it is by way of translation, and they appear as exotics.
Every Spaniard has seen his image in Don Quixote or Don Juan,
the one with his unbridled passion for justice, the other with his
unbridled will, both dynamic figures. The Cid of the ‘romances’
is a froward youth and an upright old man, typically Spanish in
either way, without the need to discount any of his qualities; the
Cid is neither mad, like Don Quixote, nor a hedonist, like Don Juan,
nor are Spaniards like that in general. Genius may offer pictures
which are more subtly true, more various or more brilliant; but
nothing more broadly acceptable than the portraits of ballad
poetry. The features are generalized and motives are broadly
human; the situations are those which occur at all times. The
heroic exaltation of Roland when he refuses to blow his horn, or
the cold fury of Hagen in the pit of serpents, are moments of
tension which cannot be for ever maintained; Milton’s puritanism
or Tasso’s synthesis of Christendom are attitudes that have passed.

Robin Hood s good humour and sense of fair play are, one
hopes, qualities for ever English; Niels Ebbeson’s self-reliance is
the manly Dane; Il’ja and Dobrynja are Russians of the Russians,
and Marko Kraljevic is everything a Serb would like to be. They
have their faults, which are also characteristic. Save for their
stature, ballad heroes are average leaders of their race.

PEOPLE AND POETS

„ . , +w ^ii-ds which (as we have seen) know so little

So it happen _ ^ t0 ’ crea t e and sustain the sense of nation-
ot nations tj, However crushed by the rapacity

hood in times lose heart s0 , 0 „g os he sang of

of pashas, the B g . One reads, without surprise,

S to life and limb. Though the Bulgarian and Serbian tongues
are almost one, the ballads of the latter perpetuate the racial dif-
ferencehi the term ‘black Bulgar’. The Greek klepht led a life of
wM exhLation in summer and bitter privation in winter; it was
sure to end on the gallows, probably after torture, yet his
Woudia’ let him believe that Olympos and Kissabos talked of his
exploits, and convinced him that ‘freedom is a glorious thing.
Thanks to the ballads, the razzias of the kleph s were joined
into one common demand for Greek freedom. Still more marked
was the influence of his ballads on the Serb. The best of them
sang of the defeat of Kosovo, redeemed from shame by patriotic
devotion and martyrdom; in Marko Kraljevic they had a Serb
whom no odds could daunt; and the haiduk poems displayed the
same wild valour on a more normal scale and in more recent times.
We can well believe that such memories were in their minds as
the Serbian army crossed the field of Kosovo in the war of 1912.

As soon as the soldiers felt under their feet the liberated plain, they
fell on their knees, whispering prayers and kissing the sacred soil. And
when they rose again, they instinctively marched over it softly, on tip-toe,
in order not to disturb the sleep of the heroic dead who, more than five
hundred years before, had there given their lives for Cross and Liberty. 1

At the same time, the sight of Marko Kraljevic astride his Dapple
broke a detachment away from the line to storm a hillock with an
irresistible rush, before the officers could restore the discipline

required by modern war.

These Balkan instances are striking, but are not unparalleled. In
the north, Holger Danske became a symbol of Danish freedom
through his defence of their land against the German Dietrich, and
Danish soldiers chanted his ballad as they manned the Daneverk
in 1864. The recovery of the Kalevala and Kalevipoeg , and of the
balladry underlying these poems, has restored the national spirit
of Finns and Esthonians.

1 J. Lavrin in Helen Rootham, Kossovo: Heroic Songs of the Serbs, Oxford,
1920, pp. 19-20.

PEOPLE AND POETS i S

I have the liveliest recollection (wrote Dr. Jakob Hurt) how on a fine
summer morning—-it was a holiday—I ran to Dr. W. Schultz, then
secretary of the Esthonian Learned Academy, in real enthusiasm, and
got myself a copy of the eagerly awaited heroic poem. With hasty im¬
patience I cut the pages, and with feverish excitement I ran over the
introduction and the individual songs. No other work in my youth so
gripped me, so electrified me, as these first cantos of the Kalevipoeg.
The sweet accent of my mother-tongue, beloved echoes of home, mar¬
vellous memories of golden childhood, the original thoughts and figures,
the popular form and art of composition—all these worked mightily on
me, weaving a magic halo around the newly resurrected national hero.
... I understood and fully comprehended the enthusiasm of Dr. G.
Schultz of St. Petersburg when he said: ‘Just think what an inspiring
influence it must be for a people to become aware and conscious of its
historical existence and grandeur! They would feel like that beggar
who was abruptly told, “You are a king’s son!” Is there any more in¬
contestable evidence of a people’s significance than the possession of its
own epos?’ 1

History and interest formerly united Lithuania with Poland;
ballads and language now keep them apart. The languages of
Lithuania and Latvia are closely akin, but the countries have had
different historical experiences; the sweet elegies of Lithuania
show a different temper of mind from that which forged the
Latvian epigrams.

Ballads have power to declare nationality and separate neigh¬
bours, but they have also power to unite those whom history has
put asunder. The old common feelings of Scotland, Norway, and
Denmark, revealed in the ballads, have been to some extent re¬
covered in modern times by Jamieson’s rendering of the ‘viser’, 2
and Grundtvig’s of the Scottish pieces. The scores of English
ballads alive in the American mountains, sung in the manner of
English folk-songs and occasionally in modes we have forgotten,
are firm testimony to the cousinship of the two nations. The
greatest services to English literature by the United States have
been F. J. Child’s ordering of the English and Scottish Ballads, and
the discovery by his pupils how lively these traditions of ours are
to-day in America.

* J. Hurt, VanaKannel (Alte Harfe), Tartu, 1886, i, pp. xi-xii
2 Selected ‘viser’. A more amply representative collection is R. C. A. Prior’s
Ancient Danish Ballads , 3 vols., London, i860.

WHAT IS A BALLAD?

T O the word "ballad 5 it will be necessary to attribute a special
meaning for the duration of this book, since none of those
commonly accepted precisely cover the subject. The material o
ballad concerts—‘a light, simple song of any kind: now specially;
a sentimental or romantic composition, each verse of which is sung
to the same melody—does not concern us here; still less have we
to do with "popular songs, often scurrilous or personal 5 or^with
"posies 5 . The etymological sense of the word is "dancing-song 5 , but
many such songs are in use which we should not call ballads, and
many, perhaps most, ballads were not composed to accompany a
dance. There remains a fifth definition in the Oxford Dictionary ,
supported by the excellent instance of Sir Patrick Spens: in this
acceptation a ballad is "a simple, spirited poem in short stanzas,
narrating some popular story 5 . The definition stands much nearer
to the intention of this book, but it is, on the whole, too narrow.
One might cavil at the word "spirited 5 since there are too many
base and dispirited ballads which we must accept, though without
dwelling upon them, but the demand for stanzaic structure is ful¬
filled only by English ballads and those of the countries within a
certain radius of our shores; three out of four principal types of
European ballad are not stanzaic at all. One might, of course, use
the English w r ord to describe the English thing, which is of
sufficient importance in itself, were it not for the fact that no more
general term is available. In Spain traditional narrative poems are
called "romances 5 and these cohere in the "romancero 5 . The collec¬
tive term is of great significance, since it is a witness to the fixity
of Spanish technique in this matter, to the mutual relations of the
various "romances 5 , and to their collective weight and influence.
It has spread to France, Germany, and Italy, and "romance 5 in
France and "romanza 5 in Italy have been distinguished from
"roman, romanzo 5 , "romance, novel 5 . But in our country these
words have no circulation beyond the limited company of English
hispanists, nor is naturalization feasible, since it would not be
possible to keep apart the two acceptations of the word "romance 5 .
There is no option but to employ the word "ballad 5 in the widest
sense as meaning any short traditional narrative poem sung,

WHAT IS A BALLAD? I7

with or without accompaniment or dance, in assemblies of the
people.

Narrative songs of this nature could be heard all over Europe
in the later years of the fifteenth century or the first half of the
sixteenth, and they are still enjoyed or newly created in central
south-western, and south-eastern Europe and in parts of America
where English, Scandinavian, and Spanish ballads still live. The
area covered by balladry is vast and the period since they first
appeared m the twelfth century is long, yet their unity as a literary
type is convincing. The same or similar subjects recur in them all
the same situations, the same generalized handling, the same habit
of repetition and stock phrases, the same rejection of claim to
authorship, the same instinctive response by the unlettered audi¬
ence to the often blind and illiterate singer, and even the same
reward for the entertainer—‘un vaso de bon vino’. Within the
type, however, there are differences which have a regional signifi-
cance and may be defined with reference to those balladries in
which a more or less uniform usage has been established. In
h ranee Doncieux distinguished between ‘complaintes’ and ‘chan¬
sons a danser’; the latter are verses with refrains, capable of being
anced; the former are not strophic, and their use of tragic material
goes some way to justify the somewhat unsatisfactory term ‘com-
plamte It is not, however, in the presence or absence of the
dance that one can seek a criterion of ballad differences. Many
German and English ballads are unsuitable for dancing either
because of their words or their tunes, though it is possible that
their form was originally that of the ‘carole’. In Serbia there is an
apparently firm distinction between the songs of men or warriors
(junacke pesme) and those of women (zenske pesme); the latter
only are danced, and the former are not so much songs as recita-
tives. One is tempted to use these terms to distinguish between
traditional narratives and traditional lyrics, especially as this dif-
rerentiation of function between men and women could be sup-
ported by the example of other countries, such as Portugal. But <
the women’s songs are not only by women or for women, nor are '
they uniquely lyrical or all suited to the dance. Lyrical qualities
enter into ballads in various proportions, and have been held bv ■
some writers, such as the American Gummere, to be essential to ?
the type. It is difficult to grant that all ballads are lyrical, since

many Spanish ‘romances’ are wholly narrative or dramatic,

l8 WHAT IS A BALLAD?

and the same is generally true of Serbian ‘junacke pesme and
Russian ‘byliny’. They are only lyrical in so far as any relatively
short poem excites a swift and simple response, different from the
complex reactions to epic or dramatic poetry. Traditional verse
mav be dramatic narrative, purely narrative, lyrically narrative,
narrative lyric, or purely lyrical. 1 The fifth possibility lies out¬
side the province of this book, but cannot be wholly ignored, since
the lyrical injection into balladry varies greatly m different coun¬
tries. There are some, such as Lithuania, which have no purely
narrative ballads, but a shading of lyric into narrative as subtle
as the spectrum; in other countries, such as Spain and Portugal,
the ballad is formally marked off from the lyric. Owing to their
stanzaic form it might be held that English ballads are lyrical
narratives, not pure narratives; want of such form does not prevent
some short Greek ‘tragoudia’ from being almost purely lyrical.
When themes migrate, that which is a story in one land is an ex¬
pression of mood in another; when technique migrates, the dances
which have made Portuguese women’s songs lyrical are probably
akin to those which have given cause for the narratives of Scandi¬
navia. One cannot set an absolute frontier between ballads and
folk-songs, though this book would become unmanageable were
it to attempt to cover all traditional poetry. We can, however,
observe the distinction of more or less, and keep our attention fixed
on the narrative end of the spectrum. Our attention may be
focused by those balladries (the Spanish, Danish, Serbian, and
Russian) which have the most marked individuality, the most uni¬
form usage, and the greatest aesthetic value.

In the Scandinavian North ballads are sung and danced, and
they are named by words equivalent to the Danish ‘viser’ (singular
‘vise’), but also to ‘kvaeder’ and ‘rimer’. The terms are, like most
of those historically applied to ballad poetry, vague and by no
means exclusive; but there arose in Denmark, and spread to
Sweden, Norway, the Faeroes, and Iceland, a clearly defined ‘vise’
style, which entitles us to use this name for the northern type of
balladry. The Danish ‘viser’ are narratives; the lyrical element is
strictly subordinate, and chiefly enters through the action por¬
trayed or the parallelism to which the strophic pattern gives rise.
The verses are arranged generally as quatrains, but sometimes as

See the discussion in L. K. Goetz, Volkslied und Volksleben der Kroaten und
Serben, Heidelberg, 1936, i, pp. 2-4.

WHAT IS A BALLAD? 19

couplets, and less frequently in other forms. The couplets assonate
together; the lines of quatrains alternatively. Assonance, however,
is too strict a term to apply to the elusive echoes of these verses, in
which exact equivalence of rhyme gives way to assonance and
assonance to echo, and the pattern of sound is like the ringing of
a bell, now clear, now barely perceptible, in a high wind. This
elusiveness one may take to be a consciously applied resource of
the ballad-monger’s simple art. The verses of the ‘viser 5 concern
the precentor; the chorus of dancers joins in with the refrain,
which may be some simple ejaculation (as £ Binnorie, O Binnorie 5 ),
or some vaguely lyrical evocation (as ‘when the norland flowers
spring bonny 5 ), or something relevant to the story itself (as ‘Thor
is taming his foals on the common 5 in the ballad of Thor of Jl$gciY(P );
it may be repeated within the quatrain, or some other refrain be
there used. The rhythmic pattern of these lines is very free; the
verses, however, are essentially four-accent lines, with consider-
able freedom in the use of unstressed syllables and the loss of one
accent in the second and fourth lines. Many Scottish ballads have
this form, though others have no refrain. There is also a remark-
able community of subjects in the oldest strata of English and
Scandinavian balladry which points to an original unity of impulse.
Despite the ancient and respectable testimony of Sir Aldingar , it
would be too bold to base this unity on the Danelaw and Canute’s
empire, which joined our island to Denmark and Norway in an
imposing thalassocracy. Though that empire vanished, there re¬
mained commercial and cultural links between the seafarers of the
North Sea. Scottish affairs interested Scandinavian poets as late
as the seventeenth century, and Swedish balladry sings of an
individual Scot, Malcolm Sinclair, who perished at Breslau in
June 1739. While the Scandinavian countries show a virtual
identity of ballad resources, the English and Scottish poems stand
also in special relationship with the songs of France and Germany.
Whether Denmark or the Danelaw witnessed the birth of this kind
of folk-poetry is debatable; it is certain that the most numerous
and worthy representatives of the style are Danish.

The same word (Weise) is found in Germany to denote rather
the melody than the text of German ballads. For them there is
only one word historically justified, ‘Lieder 5 . In more recent times
the word has been modified as ‘Volkslieder 5 , folk-songs, without
greatly increasing its precision. ‘Lied 5 may be applied to almost

WHAT IS A BALLAD?

any kind of poetry; there were the lost heroic poems of the
Barbarous Age (Heldenlieder), the fragments that survive (as the
Hildebrandslied '), the Eddalieder , the swollen medieval epics like
the Nibelungenlied , as well as the ballads themselves; and the use
of the same word in so many quite different senses makes German
accounts of epic and ballad origins confusing reading. To get rid
of the ambiguity the word 'Ballade’ was imported from English on
the wings of Bishop Percy’s fame. ‘Ballade’ stands for both ‘ballad’

and ‘ballade’, though the Villonesque ‘ballade’ has had little
interest for Germany. ‘Romanze’, from Spain, is another term
implying a narrative style, and still further definition may be
sought by bracketing the two words, ‘Balladen und Romanzen’.
By some such means it is possible to pick out the narrative ballads
of Germany, Holland, Switzerland, and Austria from the immense
mass of traditional lyrical poetry which also enjoys the title of
\olkslied’. The form taken by these narrative poems is stanzaic;
the refrain is characteristically, though not invariably, absent from
German ballads; end syllables are often echoed; the strophic forms
are more numerous than in the north; the rhythm is intrinsically
that of four-accent verse; and the rhymes are elusive. It would
appear that North Germany enjoyed the products of ballad-singers
before South Germany, and Saxo Grammaticus, who does not
mention any Danish ‘vise’, speaks of the presence and activity in
twelfth-century Denmark of Saxon singers. In respect of subject-
matter there is ready intercourse between Germany and Scandi¬
navia, but no such early and striking correspondences as to indicate
an original unity.

German ballad poetry has exerted a stimulating influence on
ht l k r° m n Ba I tlC ; and Finnic peoples, her neighbours. To the

tlT fl r H ech ° s 0vakia ’ a Slavonic spearhead embedded in

4 Artt r rr,L and end ° Sed by Saxons ’ B^arians, and
Austrians. Czech ballads are stanzaic, not addicted to the use of

The sTmrf/t 6906 ?^ 111 ^ immediate debt of Ge ™ originals.
£n ther n r 5 ° f Magyar ballads

he music Ttlv°7 J r e t 1 ’ “ Ae Gi Psy spirit infused into

hi 1 Vends ° f Lusatia cultiv ate ballads of German
theme and form, though in a Slavonic language and with it is said
Slavomc characteristics in music. In Poland we encounter ’

Stol’‘KeT;r da ^

’ h Athenians of Galicia and the Ukraine follow

WHAT IS A BALLAD? 21

the narrative tradition. In the Ukraine itself, the assonating,
measured poems of the western type encounter the unrhymed
‘dumi’ in free verse, which are typically Russian. The lyrical
preferences of the Poles are more marked among the Lithuanians,
who have no pure narratives. The German rhymes were, as we
have seen, elusive and frequently failed. In Poland these failures
were the more noticeable since the rhymes could be attributed to
accidents of declension and conjugation. The Lithuanian ‘dainos’
are formally unrhymed; but the accidental rhymes are numerous,
thanks to grammatical forms and to the use of liquid diminutives.'
The poems are stanzaic, receiving shape from the use of simple
parallel clauses. In Latvia the scale has been reduced; the verse
is very often octosyllabic, and the number of lines only a quatrain.
The effect is epigrammatic, like Greek distichs or Portuguese
quadras’. The notion of stanzaic form is much weakened by this
change; and in the poetry of their neighbours, the Esthonians, we
find rhymeless octosyllables, marked by parallelism so as to form
groups of verses, in many cases, but not actual stanzas. In the more
narrative pieces there is no stanzaic structure. Finnish verse is of
the same description, but insists on the use of alliteration as a
formal element. The word ‘runo’ is significant; both Ger man (as
I think) and certainly Swedish influences have been exerted on
Finnish folk-verse, alongside the indigenous turn for magical
incantations.

The Anglo-Scandinavian and German areas, with the minor
balladries, form one vast Nordic region, divided into two main
parts. The Romance area is similarly divided between the Franco-
Italian and the Hispanic parts, which commingle in the songs of;
Catalonia. The chief link between them is the fact that adventure
ballads in Spain are chiefly of foreign, and immediately of French,
origin; the same is true of ballads based on imaginative literature.
The Franco-Italian popular songs (chansons populaires, canti
popolari) fall largely outside the scope of this work, being no more
than lyrical snatches. Those which we should class as ballads arise
out of the narrative element in lyrical poetry, coagulating into
lyrical narratives. Some of these have refrains (chansons a danser),
but others are plain (complaintes). They are assonating; the older
stratum uses the tirade as its form; the younger ballads, some
dating back to the late fifteenth century in all probability, are in
quatrains. The difference between tirade and quatrain is diminished

22 WHAT IS A BALLAD?

by the circumstance that the lines of the tirade are usually
equipped with a medial pause, and the phrases of the quatrains
usually extend over two lines. The difference lies in the more per¬
ceptible stanzaic pattern of the latter. A peculiar feature of this
poetry is the rule that the two halves of a line must end in different
cadences; if the one be masculine, the other must be feminine. This
mark is found in Catalan ballads of the old French stratum, though
not perfectly. The effects may be perceived in some Castilian
ballads of French origin, though the rule is not known in
Spain.

The narrative songs of the Bretons and Provencals are of this
French type. So, too, are those of Lombardy and Piedmont, both
as to form and subject. These songs tend to be younger than
their French parallels, but are rougher and more narrative. Central
and southern Italy depend on popularized forms of the Tuscan
octave. The great mass of songs are either mere lyrical ejaculations,
like the Sicilian ‘ciuri’, or epigrammatic octaves. Narrative ballads
are few and derivative throughout the central Italian area, but they
are rather more numerous in Sicily, so much further removed from
the prestige of the literary dialect. It is a curious and significant
fact that France and Italy, the two lands of intense culture in the
Middle Ages, are weakly endowed with traditional narrative poetry;
and something of the sort seems to have applied to the folk of the
rich south of England. It is evidence of the incompatibility of
complex society and literate culture with ballad poetry. On the
other hand, French achievements in the ‘chanson populaire’ and
popularized forms of literature have a unique resonance in Europe,
and have much affected the ballads of Germany, Spain, and more
distant neighbours.

The Spanish ballads are rhymed, but not stanzaic. The unit of
composition is the octosyllable, which is not identical with the
French line of the same name. There is no pause at the end of the
octosyllable, but both the sense and assonance impose a pause after
sixteen syllables have been sung. The style is strictly narrative.
Style and subject-matter associate these ‘romances’ with the older
epic poems which flourished from the eleventh to the thirteenth
and fourteenth centuries; and a theory holds the field that the
romances’ originated in fragments of epics. This explanation is
not entirely satisfactory even for the epical ballads; as to ballads of
adventure or of real life it can tell us little that is helpful. What is

WHAT IS A BALLAD? 23

remarkable about the Spanish ballads is their uniformity of metre
and technique. They are homogeneous and cohere as a ‘roman-
cero> - The ‘romancero’, or ballad corpus, exerts a magnetic attrac-
tion by its own mass. Into it have been incorporated lyrical and other
materials, formerly expressed in a different form. The grouping
of ballads greatly increases their influence on thought and conduct.
It is through the ballads that the Spaniard obtains his knowledge
of national history and develops a consciousness of his racial
characteristics. The ‘romances’ have radiated from Castile into
Portugal, where old Spanish and new Portuguese ballads live to¬
gether ; and into Catalonia, where a later Castilian stratum overlays
the old Franco-Proven9al popular songs, often employing a mixed
language. On the other hand, Catalonia has introduced to Castile
many adventure ballads of French or foreign origin.

A third area is found in the Balkans. In the Balkans and in
Russia ballads are unrhymed and unstrophic; their musical back¬
ground is Byzantine, not the Gregorian chant of the West; they
have themes and devices in common. Assonance is found in
Rumania, though in other respects Rumanian songs belong to the
Balkan group; and it has penetrated into the Ukraine in com¬
paratively recent times. It is even to be encountered in Greek, in
ballads indebted to Venetian influence. The true Greek practice,
however, is not to employ rhyme or assonance; and this abstention
is common to the lands which Byzantium has civilized. Thus a
main line of division in European balladry separates the rhymeless
East from the rhyming West, Roman from Byzantine.

The priority of Greek ballads from Asia Minor over all others in
the Balkans is absolute; but the Greek collections are less rich or
admirable than the Serbian. The verse-form preferred by the
Greek ‘tragoudia’ is the iambic tetrameter catalectic; there is con¬
siderable use of the iambic trimeter and other more lyrical metres.
These pieces are scanned by their accents (or more precisely, by
the accents of their musical settings), and are thus different from
the quantitative measures of classical Greece. They cannot be due,
as one grammarian urged, to erecting as a norm those few verses
in ancient tragedies which happen to scan accentually; yet it is hard
to dissociate the accentual verses from the quantitative lines other¬
wise^ so similar to them. The transition between ancient and
medieval metres in Greece, as in Latin lands, is obscure. The
politic tetrameters, as they are called, are a continuous narrative

2 4 WHAT IS A BALLAD?

measure. They are very flexible, but the rush and gabble of

syllables detracts from their dignity. . , ■

" The finest metre used in popular poetry is the Serbian deca¬
syllabic; it is flexible, weighty, definite, and the vehicle of some
poignantly tragic poetry. Serbia lies on the frontier between
Roman and Byzantine influence. On the one side, Dalmatia is
Roman; on the other Bulgaria is Byzantine. The oldest recorded
poems are in the peculiar measure called ‘bugarstica’, with an
irregular number of syllables in the first half, and three trochees m
the "second. As ‘bugarstica’ implies Bulgarian provenience, and
as Bulgaria is now a ballad area dependent on the . Serbian, it
would seem that we have in this measure the corruption of some
Byzantine metre. The bulk of Yugoslav narrative pieces is ex¬
pressed in decasyllabics, with a fixed pause after the fourth syllable.
They are unrhymed, but in other respects too closely resemble the
old French epic metre to be considered indigenous to Serbia. The
French measure appears in Italian verse as a hendecasyllable with
pause after the fifth syllable; and so it would pass from Venice to
the dependencies of Venice on the Dalmatian coast. The first
Italian syllable is often unaccented. Omitting it, one obtains the
Serbian trochaic decasyllable with obligatory pause after the fourth.
A certain number of themes came from the West along with the
verse, but most of the subjects and all the music is Balkanic.
These narrative poems, when dealing with warlike subjects, are
called warriors’ or men’s songs (junacke pesme) and are chanted,
not sung. Women’s songs (zenske pesme) are amatory, and are sung
and danced in a wide variety of metres. Between the two types
lie certain amatory narratives in the heroic verse. The decasyllable
is found in Bulgaria, especially in ballads taken over from the
Serbians. Otherwise the octosyllable is more common, and there
are some interesting poems in free verse. In Rumania the line is
an octosyllable or less; the style and many important themes are
Balkanic, but the presence of assonance reveals the influence of
Germany and Hungary. A special feature of Rumanian balladry is
its pastoral background.

The fourth great ballad area is the Russian. The Russian
‘starkly’ or ‘byliny’ relate the events of a distant age: that of Kiev
under Vladimir or Novgorod in its glory. Other narrative or his¬
torical themes have been added, but this sense of antiquarianism
is a characteristic note. The lines are unrhymed, but they are also

WHAT IS A BALLAD? 25

unmeasured, save by a pause at the end and (generally) one about
the fourth syllable. The only scansion applied to them is that of
their music, which is still insufficiently known. Free in form, they
enjoy an amazing freedom of spirit. No ballads show such wild
variations of content. An individual ballad is a nucleus of character
and incident; but the rest may vary without restrictions. The
Ukrainian ‘duma’ is another ballad manner of this unrestricted
type.

The unsophisticated adventures of many ballad heroes connect
this manner with that of fables and fairy-tales. In Russia, between
the folk-tales (skazki) and the ballads (byliny), there stands a class
of prose-ballads (probyvalsciny), employing the measured formulas
of the ballads, but not the linear arrangement. Elsewhere, also,
ballads are seen to break down into prose or to be deliberately
transformed. This occurs particularly to historical ballads, accepted
as true accounts of events of national importance, and incorporated
in medieval chronicles. So there are ballads behind a number of
chapters of the Spanish Trastamaran chronicles as there are epics
su PP or ^ n § chapters of the earlier General Chronicles. The verse
sources betray their presence by runs of assonance in the prose and
by clauses which are almost metrical.

Returning to the definition offered for balladry, it has been laid
down that a ballad must be short. Short and long are relative
terms, and must be so understood. There are many ballads which
extend to only a few lines, it is true, but there are others which run
into hundreds. The longest of the Spanish ballads, Count Dirlos,
contains some 700 double octosyllables; Stepan Dushan y s Marriage
(Zenidba Dusanova) has almost the same number of decasyllables;
Russian ‘byliny’ are extremely variable, but among them we may
cite Kalinin’s version of Mihailo Potyk in about 900 lines. There
are also ballad-sequences, such as the long ballads of Robin Hood
in England and Marsh Stig in Denmark which cover much paper
and are longer than some epics. In general, the two eastern types
of ballad tend to be long; those of the west to be shorter. The
strophic form of northern balladry is an encouragement to diffuse¬
ness of a parallelistic sort, but tends to limit the length, since
stanzas become wearisome when continued too long, especially to
one tune only. Spanish ‘romances’ have been shortened by the
tendency to drama, which has led to the omission of as much

narrative as possible so as to leave the dialogue in the forefront of
4615

26 WHAT IS A BALLAD?

the poem; they have been shortened also by the cuts applied by

the early compilers, such as the anonymous collector of the

Cancionero de Amberes (about i 54 S)> cu * s ^ ue a PP aren ^y
pressure of entertainers at the reunions in which ballads were sung.
The length of the extant ‘romances' is therefore less than that of
their first form, which may normally have been that of short
epics. The old French Gormond et Isemhart has 661 lines, though
it is a true epic and the text is virtually complete, the Danish
Vise* of Tori af Havsgaard is actually longer than its original, the
Thrymskmda. None the less, it is characteristic of ballads to be
relatively short, since their range is from a few lines to several
hundreds; the epic range is from hundreds to thousands and tens
of thousands.

Ballad shortness is also qualitative, and that is what makes it
worth notice. When It is possible to compare them with epics or
novels, they are found to be either episodic or summary. The ballad
is concerned either with one striking moment of the whole tale, or
it conveys to an unlettered audience a general impression of a
work from written literature. Sometimes both tendencies work to¬
gether, as in ballads dealing with Tristan. In the Tristan story
there is one scene of paramount interest: Isolt dying of heart-break
stretched on the corpse of Tristan. To this everything else in the
long novel is either subordinate (episode of the love-drink) or
irrelevant (Tristan's numerous battles). In Spain and in Iceland
an episodic treatment is given to the theme. The ballad-poets take
into account only the last scene, though they work from different
versions of the event. In Spain the poet has squeezed the last three
chapters of Don Tristan de Leonis into a few intensely dramatic
lines; the Icelandic poet takes more space since he heightens the
pathos by repetitions. In the Faeroes it is still the death scene that
occupies the attention, but the author of Tistrams tattur introduces
the characters in a regular way and presents a complete novelette.
This summary treatment appears in the Danish Grimhild’s Ven¬
geance , the Spanish Marquis of Mantua and Montesinos as con¬
trasted with the French epics Chevalerie Ogier and Aiol % and the
Danish Holger Danske when compared with the Enfances Ogier .

These summaries are not only abbreviations of the originals but
void of circumstance. It is in this respect that they differ from the
summaries presented in such Eddie poems as the Gnpispd and
Atfamal, which are not as long as many ballads, and in this way

WHAT IS A BALLAD? 27

also the Thrymskvida is seen to be a more detailed narrative than
the longer ballad Tord af Havsgaard. Precisely why Tristan and
Isolt should be lovers, yet unable to marry, requires knowledge of
many attendant circumstances in the prose romances. One must
know about the love-drink, that Tristan wooed on behalf of Mark
and was honourably bound to deliver Isolt, that there was no
question of taking her away by force, and that there were quarrels
and reconciliations and moments when secrecy seemed no longer
possible. It is not so in Tistrams tdttur. The ballad poet has
selected the simplest of all motifs for this situation. Tristan’s
father and mother frown on his marriage with Isolt and send him
to France to marry the French king’s daughter. So, too, in the
Thrymskvida we are given all the circumstantial details attending
Thor and his associates as gods of Asgard or giants of Jotunheim,
but any Squire Western would serve as the hero of Tord af Havs¬
gaard. Ogier the Dane fought the giant Brunehaut to win the lovely
Gloriana for his friend the Saracen Karaheu. How Ogier came to
be in the Saracen camp and Karaheu in the Christian, why Ogier
should fight for Karaheu on this occasion and how they had
become acquainted, this is matter for hundreds of lines in the
Enfances Ogier-, but in Holger Danske the hero’s imprisonment is
taken for granted, and his friend is written off as a coward.
Similarly, comparing the Chevalerie Ogier with the ballads of the
Marquis of Mantua we find that the lynch-pin of the epic poem
has been removed by the ballad-monger. The general theme is
that a vassal, grievously wronged by his lord’s son, is at last
avenged. In the epic poem Charlemagne refuses satisfaction and
Ogier is forced into rebellion, involving a siege, a long pursuit
along the road to Rome, and years of imprisonment. In the ballad
Charlemagne agrees to deliver his son to the judgement of his
peers, and the action is brought to a swift conclusion, though the
style is diffuse and verbose.

avoidance of circumstantial detail, the broadest generaliza-
tion of motif, situation, and character, is a leading characteristic of
ballad poetry and essential to its traditional preservation. A ballad
king is the king, a ballad Saracen is a dog of a Moor, ballad lovers
are taken for granted and have all our sympathy, ballad situations
repeat themselves and are transferable, ballad motives are the
primary loves and hates, ballad language is formula, and ballad
style is precedent.

WHAT IS A BALLAD?

These are the elements of traditional art. 1 Composed in common
form, the ballad becomes at once common property, like a fairy¬
tale or legend. The author has no copyright, and the ballad only
exists by virtue of each successive performance when it is what the
performer makes it. It is not that ballads were, as the Romantics
insisted, the product of the community working as a creator.
Artistic creation under such conditions would be impossible; each
ballad has its author and its moment of birth. On both these points
we have information in respect of a number of ballads in different
countries. We learn, in Spain, that Garcilaso de la Vega performed
a remarkable feat of arms in the year 1455, on the nth July, and
that became the subject of a ballad; and that the conversation
between Juan II and Prince Yusuf of Granada, reported in
Abendmaty Abendmar, occurred on June 27th, 1431. In 1491 Queen
Isabel the Catholic ordered her confessor, Fray Ambrosio Monte-
sinos, to compose a ballad on the death of Prince Afonso of
Portugal; the ballad survives and is in the veritable traditional style.
The names of Alonso de Salaya, Pedro de Palma, Diego de Zamora,
and Juan de Leyva, preserved by flying leaves of the sixteenth
century, are as likely to refer to composers as to performers. In a
number of French and German pieces of a domestic and somewhat
vulgar type it was an established convention to indicate the real or
putative authors by some description. The formula is: ‘if you wish
to know who made this song, it was made by three maidens’ or
‘an apprentice’ or ‘a soldier’, &c. In 1812 F. L. Jahn visited
Ferdinand August and commissioned a satirical ballad on the ruin
of the Grande Armee. He gave a few phrases, such as ‘cuirassiers
in frocks’ and ‘ensigns without ensigns’ and the opening couplet:
‘With man and horse and wagon, so has the Lord them stricken.’
August composed six verses which became popular for their
vigorous scorn, and they were adapted in 1871 to Bourbaki’s
retreat into Switzerland. About 1830 Jens Christian Djurhuus, a
crofter from Kollefjord in the Faeroe Islands, borrowed a copy of
the Saga of St. Olaf to excavate from it new material for the dance.
In the method of composition his Ormurin langi and Longfellow’s
Building of the Long Serpent are identical; but the one is personal
poetry, the other is an excellent example of traditional balladry.

There is no personal right arrogated by such authors over their
work, which is the absolute property of each reciter, to shorten,

1 See R. Menfedez Pidal , Poesia popular y Poesia traditional, Oxford, 1922.

WHAT IS A BALLAD? 29

extend, mingle with others, and transmit. Once launched the
ballad is everybody’s possession. Personal or local details will be
pared away; situations, motives, and characters will be generalized.
It will only exist at each moment of performance, and it will never
be twice performed alike. The accuracy of the reciter’s memory
and his private interests will affect the recitation: he may make a
humble parade of his store of technical devices, or accentuate the
ornate, pious, or bellicose elements in his original. He may prefer
a new tune or forget an old one; if his new tune be borrowed from
another ballad (which is more likely than not), it will bring some
of its old words with it. The ballad is fluid in the performer’s
mouth; and yet its variations are not infinite. As remarkable as
the multiformity of ballads is their rigid conservatism. In the
sixteenth century the ballad on the Master of Santiago's Murder
( : 358 ) had an altered beginning which may have come into exis¬
tence within thirty years of the event; but last century the original
commencement was recovered from Asturian tradition. Similarly,
the second half of Count Arnaldos remained unknown until en¬
countered among African Spanish Jews quite recently. In the
Faeroe Islands there existed a Norse ballad, Ole Morske, without
one Faeroese expression, which had been quite lost in its country
of origin. Authentic Danish ‘viser’ have been discovered in the
Faeroes and in outlying parts of Denmark, such as Jutland. The
cycle of the Kiev heroes flourishes still on the shores of Lake
Onega, but is unknown at Kiev, where a new lyrical manner has >
completely dislodged the epic style since the sixteenth century. A
long memory is a pearl of price. If the singer forgets, there may be
those in the audience who can correct him, like children reciting
at a children’s party verses known to them all. Singers give and
accept challenges, and the one with the poorer repertoire or
versions is likely to learn from the other. Ballad-singers must
learn their art and work within the strict conventions of the tradi¬
tional style; they are professionals or semi-professionals. We have
many words which denote such persons: ‘minstrels’, ‘histriones’,
jongleurs, juglares’, ‘skomorohi’, ‘guslari’, ‘forsangere’, ‘recita-
doras’, which denote function, and also ‘kaleki’ (‘lame beggars’)
and ‘ciegos’ (‘blind beggars’), terms which show why the singer
has been forced to adopt the profession. In the last resort ballads,
like fairy-tales and incantations, are the property of grandmothers
and nurses, singing their babies to sleep or keeping children quiet

3 o WHAT IS A BALLAD?

by telling stories. But such a restriction of public is a mark of

extreme decline. .

Ballads are popular. The word has given rise to difficulties
because of the confused thinking which caused the Romantics to
Imagine a populace of poets, and because of the identification of
the people with the piebs. This confusion is of long-standing, for
the Marquis of Santillana, writing in 1445, speaks of ‘the lowest
order of versifiers, those who without any order, rule or count
make these “romances” and songs wherein persons of low, servile
condition delight 5 . No doubt they were delighted; but so, too, was
the great Isabel the Catholic, and Queen Sophia of Denmark, and
Tsar Ivan the Terrible. The generous heart of Sir Philip Sidney
was stirred by ballads ‘as with a trumpet 5 . The ballad people was
the whole people, organized under its natural leaders. They
prefer aristocracy to the proletariat, and often seem to be mainly
interested in the lesser nobility, the ‘KleinadeP of Germany. It
is the local chief, not the nation’s king, who counts for so much
in frontier ballads: the Percy and the Douglas, Kinmonth Willie,
Sayavedra, Bishop Gonzalo of Jaen, Fajardo, Niels Ebbeson,
Mark) Kraljevic, Ermak, Kolokotronis. In this respect also ballads
differ from traditional epics, for the traditional epic has a national
scope and its heroes are the leaders of the nation. Its inspiration
comes from large movements of the folk: the Barbarian irruptions,
the Viking raids, the crusades and pilgrimages, the defence of
Anatolia against the Turk. The ballads are concerned with small
settled communities, local heroes, raids, and excitements of no
more than episodic value in the nation’s history. Even a national
movement, such as the Conquest of Granada, is broken up by
ballad-poets into disconnected episodes of personal and local
interest. Both the epic and the ballad, however, were directed to
an unlettered public, not necessarily quite illiterate, but accustomed .
to get entertainment orally. They are addressed to those who have!
ears to hear; not to readers. When a ballad is written down and
printed it may find a new circle of readers, but it becomes stereo¬
typed and begins to shed the characteristics of traditional litera¬
ture. A schism appears in the public; readers are different from"
listeners, and the former have more, the latter less and less
prestige. Instead of addressing the people, the ballad-monger has
before him the piebs, and ballads become vulgar and insignificant.
It is to humanists that we owe the preservation of so many delight-

WHAT IS A BALLAD? 31

ful ballad texts; far more than were contained in the repertoire of
any traditional minstrel; but humanism, however pious, is a deadly
opponent of the genre. It takes ballads out of the line of oral trans¬
mission, into the textual; they are the less heard or performed, the
more they are read and discussed. The advance of the habit of
reading involves the shrinkage of oral entertainment. As they lose
the better sort of patronage, ballads fall off in art and vigour; they
are driven from the centres of mental life into outlying provinces;
their topics lose elevation. In place of tragedy there comes horror;
for energy, puffy pretentiousness; for humour, personal abuse and
levity; for heroes, brigands and malefactors; for events of some
consequence, crimes and casualties. Such vulgar ballads of the
decline are very different from those of the best period. Chaucer
summoned Wordings’ to listen to the rhyme of Sir Thopas:

Now listen, lor dings, to mine intent,
and I wol telle you verrayment
of mirth and of solas.

Chaucer was probably not serious when he penned these lines; but
a Swedish minstrel is rightly in earnest when he sang of his hero:
Him shall men praise
in courtly lays

amid squires and dames.

The ballad people—the whole people without distinction, lords
and commons alike—danced in Ribe:

There dance the knights in scarlet braid—

{Tread it so featly, noblemen!)
and there goes Chrissie, so fair a maid—

{for men honour young ladies in the dance).

The dance goes down through Ribe’s street,
the knights they dance both glad and fleet.

The dance goes down the Ribe’s stream,
the knights they dance in shoes that gleam.

Sir Riber-Wulf he danced the first—

{Tread it so featly, noblemen!)

King’s man was he in truth and trust—

{for men honour young ladies in the dance).

With the swirling polkas of the sixteenth century the old knightly
round dances of Denmark retired to their present-day fastnesses

3 2 WHAT IS A BALLAD?

_the barn and the hearth; but the old ballads keep, their air of

good breeding. We have the less reason to be surprised at their

discretion and simple elegance, their avoidance of the coarse and
clumsy, and the easy courtliness of many a phrase. There are
crudities which belong to the epoch; but, though they appear in
ballads, it is without exaggeration or prurient emphasis. Even m
those scenes which testify to a code of manners less sophisticated
than our own, the old ballads keep their poise and delicacy, like
that of the good knights of old at whom Ariosto smiled, not without

tears: #

0 gran bonta de s cavallieri antichi!

The klephts and haiduks of Balkan poetry are much diminished
from the pristine excellence of Digenis Akritas and Tsar Lazar,
Robin Hood is below the Douglas and the Percy; but they, too,
represent ideals and give flesh to communal aspirations after liberty
and generosity.

Ill

PERFORMANCE

T HE ballad, then, is a short traditional narrative poem sung,
with or without accompaniment or dance, in the assemblies
of the people. It is a complete entertainment, and is alive only when
so performed. It satisfied more requirements than does a printed
poeim The story told was enlivened by mime; it had a sprightly
or plaintive tune, and if there were a chorus the public could join
in, in many cases the ballad was made a pretext for dancing. The
various types of folk-song encountered in Rumania have been
named and classified in such a way as to give oral literature the
same range and variety as written literature: there are lyrical
f doine\ ‘hore^ ‘colinde’, and ‘bocete’, epic ‘balade’ and ‘plugusori’,
dramatic ‘oratie’ and hrozi’, novelistic ‘basme’, and miscellaneous
charms and formulas which may be classed as didactic. Similar
divisions might be established for the songs of the Balkans and the
Ukraine. Were the distinctions exact, they might be more generally
applied. In Spain, for instance, the 'romance’ is epical, with only
a lyrical tinge in certain ballads. There is even a formal dis¬
tinction between epic and lyric in oral poetry, since the use of the
romance metre implies narrative intent. In some of the finest of
these ballads the narrative lines are reduced to a minimum or quite
omitted. There remain only the speeches of the opposing heroes;
and with the aid of mime, these pieces would be one-man
dramatic entertainments. The closing ballad of the Infantes de Lara
cycle is a breathless dialogue between the villain and the avenger
in the crisis of the action; it is more poignantly dramatic than the
corresponding scene in Lope de Vega’s play El Bastardo Mudarra .
Such scenes may have satisfied the popular demand for drama, and
so answer a paradox: for the Spanish people are endowed with a
dramatic instinct which gave the immense harvest of their Golden
Age dramas, and yet seem to have neglected the theatre throughout
the Middle Ages. The Danish Viser’ are as vivid and abundant as
Danish medieval literature is meagre.

The earlier modern collectors of ballads were careful to recover
the texts, but tended to ignore the tunes. The tunes were omitted
altogether, or consigned to an appendix, and it was only with the
advent of Ludwig Erk that the same care was given to the music

461s

performance

* ,0 the words. His work was “ n,inued fn ^ m ^ t ®2“S s “ d f

has resulted in a collection which serves for model to the rest
Europe ' Germany still leads the world in the number and orderli¬
es o'her studies of tunes. In other countries practice varies
”ueid rab v and, as those bools which record tunes do no, often
records// the tunes of a given ballad, it is extremely difficult to
determine their interrelations. American collectors record ftrnes
and words of living ballads, and lovers of English ballads are
deeplv indebted to S Baring-Gould, Cecil Sharp and other enthu¬
siasts" We lack, however, such a compendium of words and tunes
together as would give us the advantages that Germany enjoys.

For their remissness in this respect the older collectors have been
severelv censured. It has been suggested that ‘perhaps the fairest
explanation is that literary scholars are tone-deaf, and as incapable
of being moved by the melodies of Lord Gregory or Geordie as, for
the most part, they are of appreciating that, from a literary point
of view, the most significant thing about The Wife of Usher s Well
or Edward is that the one will cause strong men to weep, and the
other make their hairs stand on end’. The names exempted from
this rule, however, are those of the scholars who have attained the
highest perfection in ballad studies; and there are, in any case,
some palliating pleas. To print texts only is a tradition dating back
to the ballad collections of the sixteenth century in Spain.. The
reason was not tone-deafness, since these texts served as aids to
the memory in concerts of chamber music. The publisher could
rely on his clients to know the traditional tunes, though they might
trip over the words; or, alternatively, they might prefer to sing the
words to one of the new polyphonic settings which contemporary
composers produced in abundance. On either supposition, those
who bought the books of words did so because of their fondness for
music. Further, though a ballad is not a ballad except when sung,
there is no indissoluble connection between the tune and the
words. There being no other scansion than the music, ballads were
composed to tunes already existing; 1 2 3 an unemotional style is

1 L. Erk und F. Bohme, Deutscher Liederhort , Leipzig, 1893-4.

2 Mr. J. Goss’s Ballads of Britain, London, 1937, is an excellent anthology,
giving a standardized ballad text and a choice of tunes. It is valuable on this
account. For comparative purposes, however, one requires all the tunes, and these
should be related, as in Erk und Bohme, each to its own proper variant of the ballad.

3 See the headings in R. Liliencron’s Historische Volkslieder der Deutschen ,
Leipzig, .1865-9', ■

PERFORMANCE 35

adopted by the singers, so that the adaptation of music to words is
not intimate; new tunes were fitted to the same words, and there
was much borrowing; and the words do have artistic qualities
which can be savoured separately. We are accustomed to do so
with the Renaissance lyrics and sonnets, though these were devised
for music. Ballads, indeed, are not complete when we associate
words and tunes, for that may be very much less than their full
performance. Singing Tam Lin in a drawing-room is something
remote from the real Tam Lin, which was probably danced in the
open air by the villagers on some specially appointed day.

The classic description of ballad dances is that of V. U. Hammer-
shaimb in his Fcemsk Anthologi. The Faeroese dances occur on
Sundays and on three special festivals of the year. The islanders
have few entertainments: chess, tricks with string, a ball game,
wrestling, and field sports. The opportunity to dance thus stands
supreme among their pastimes, and attracts the bonders from long
distances, especially for St. OlaPs feast, when they honour the
patron of the archipelago. Theirs is a simple ring dance, generally
in 6/8 time. The dancers hold hands, and if there are too many
for one ring, others are formed within, so far as space permits.
They also know a variety of ‘Sir Roger de Coverley’, though it is
considered rather fatiguing. The precentor sings a ballad and the
rhythm is stamped with the feet. The dancers pay close attention
to hi& words, which must come clearly, since the characteristics of
the narrative are brought out by the mime: hands are tightly
clasped in the turmoil of battle, and a jubilant leap expresses vic¬
tory. All the dancers join in the chorus at the end of each stanza,
but the stanza is sung only by one or two persons of special repute.
They must have good memories, since there is a prejudice against
using the same verses twice in a year, and also against the use of
too many different tunes on any one occasion. Hammershaimb
went on to specify the steps of the dance.

Very similar conditions apply in the Balkans. A. Dozon 1 quoted
from the brothers Miladinov the following account of a Bulgarian
‘horo’:

At Straga, on the days of less solemn festivals, a round is formed
separately in each quarter, but on the great feasts (as Easter, St. George’s
Day, &c.), all the girls collect in a garden outside the town and form an
immense ‘branle\ led by one of them who sings. Half the dancers

1 A. Dozon, Chansons populaires bulgaires inedites , Paris, 1875, pp. xiv-xv.

3 6 performance

accompany her and the other half repeat each line, and so on to the end
of the song. The leader (horovodka) then yields her place and function
to her next neighbour and takes up a position at the other end of the
round. If the dance lasts long enough, these changes go on until every
dancer has had her turn. Normally, however, the dance is led by the
girl with the best voice and best-furnished memory. . . . One or two
such are to be found in each village or township.

In later medieval times, writes Rodney Gallop, 1
the name given to the French Round dance was Branle , a name which
still survives not only in the country districts of France, but as far afield
as Ronmania where the ‘Braul* is a favourite and very lively ‘Hora’.^ In
his Orchesographte Thoinot Arbeau gives the steps of the Branle as
danced in the sixteenth century, together with the tune which will be
familiar to many from Peter Warlock’s fascinating ‘CaprioP. We should
describe these steps today as two ‘chassis’ to the left followed by one
to the right, and in this very form it still survives in the ballad dances of
the Faeroe Islands, to which the Chain dance came very probably by
way of England and Scandinavia. These ballad dances are done at the
Feast of St. John and at the National Festival (July 29th). They are
accompanied in the best traditional style by the singing of old ballads
and folk-songs. The chain, closed at first, is later opened by the leader,
and in its labyrinthine windings the dance resembles the end of the
Lancers’ at a rowdy house-party.

These steps are called ‘stigingarstev’ by Hammershaimb, who also
described the ‘trokingarstev’ (steps backwards during the verse,
forward during the refrain) and the ‘bandadansur’, performed by
men and women in two lines, with ribbons.

The Bulgarian ‘hero* and Rumanian ‘hora’ are the same as the
Serbian ‘kolo’ and Greek ‘chores’. The Greek word is the source
of all others, as it may also be of the French ‘carole’. In Greek
balladry we encounter the tableau of the maidens’ round dance
which is such a favourite in French and medieval Latin poetry. A
king, out rabbiting, comes suddenly on the scene:

The golden damsels danced before, the brown girls danced behind
them, and in the midmost of them all fair Zerbopoula tripped it,
and as she moved her sleevelets gleamed, her collar flashed as lightning.

Ludunt super gramina
virgines decore,
quarum nova carmina
dulci sonant ore.

1 Violet Alford and Rodney Gallop, The Traditional Dance, London, 1935.

PERFORMANCE 37

Beneath the castle of Beauclair,
they shortly raised large beams in air.

The damsels to the ‘carole’ go,
at jousting squires their prowess show,
and belted knights regard the fair.

The girls of Galicia and Portugal observed this ritual in the thir¬
teenth and fourteenth centuries. Their exquisite 'cossantes’ show
how they accompanied their mothers on local pilgrimages, and
while the older women prayed, they danced on the glebe, taking
care to tighten their bodices so that the watching gentlemen could
admire their rounded charms. These 'cossantes’ are purely lyrical.
In France, in the thirteenth century, men began to take part in the
women’s dances.

These dances raise the curious and intricate question of woman’s
contribution to poetry. In extreme decline ballads are to be found
on the lips of grandmothers and nurses, protected by the im¬
memorial sameness of house and cradle. The memories of men
are more capricious; they may even show aversion from folk-songs.
While ballads flourish, the more lyrical and domestic poems are the
special concern of the women. This may be acknowledged, as in
the opposition between 'zenske’ and 'junacke pesme’ in Serbia; or
it may be implicit in the texts. Greek ballads to be danced are
often feminine in sentiment or theme; the Lithuanian 'dainos’ are
open to both sexes, but with a feminine bias; the same bias charac¬
terizes not only the French 'chanson populaire’, but also, to a great
extent, the whole of French poetry. A distinction may arise, as
in Serbia, between lyrical pieces (whether ballads or not) with
developed melodies and associations with the dance, in which love
is the favourite theme, and chanted narratives of warlike acts,
never danced; the former being feminine, the latter masculine.
But the antithesis is not complete, since there is a considerable
body of non-heroic verse which might be classified either way.
Further, it appears that the women’s songs go back to a century
or two m which 'junacke pesme’ were unknown. 1 Similarly, in the
steps of the dance, where both sexes take part, the traditional round
is the affair of the women in particular, while the men execute the
leaps of 'virtuosi’. In France the poetry of the troubadours is
under feminine patronage; doubtless for many reasons. One
which must not be overlooked is that the 'chorese rusticarum

EX Subotic, Yugoslav Popular Ballads t Cambridge, 19325, p. 1421.

3 g PERFORMANCE

mulierum’, with their songs, had been a feature of French life
from, at least, the sixth century. 1 These considerations raise issues
too distant for the present book; namely, the indebtedness of all

poetry to women.

Ballad performances, in which there is no dancing, have been
described by Cecil Sharp: 2

The mountain singers sing in very much the same manner as English
folk-singers, in the same straightforward direct manner, without any
conscious attempt at expression, and with the even tone and clarity of
enunciation with which all folk-song collectors are familiar. ... So far
as I have been able to comprehend his mental attitude, I gather that,
when singing a ballad, for instance, he is merely relating a story in a
particularly effective way which he has learned from his elders, his
conscious attention being wholly concentrated on what he is singing and
not upon the effect which he himself is producing.

Mr. Neville Coghill has described to me his impression of ballads
heard in the Blue Mountains of Virginia. They were sung to him
without accompaniment, chorus, or dance. The singers were all
women, who adopted a harsh, clear, nasal intonation completely
devoid of expression. At the end of each piece the singer merely
said, 'That’s a pretty ballad’, and passed on to the next without
change of expression. (Similarly, Mila y Fontanals noticed that the
only comments were ‘pretty’ or ‘sad’; more lively epithets were
used only for up-to-date ditties from the towns.) The effect of this
impersonal singing was to leave the words of the ballad to do their
own work, and for that a clear enunciation is essential. The singer
plainly originated nothing in her repertoire. She might add or
omit, as her memory served, or change the order of narrative, but
the only modifications of which she could be capable were of a
mechanical kind. Her repertoire probably went back to the seven-

1 The texts (chiefly ecclesiastical protests) are collected in K. Voretzsch,
Einfuhrung in das Studium der cdtfranzosischen Literatur, Halle, 1925. See also
G. Paris, ‘Les origines de la po^sie lyrique en France’, Journal des Savants , 1892
(reprinted in his collected essays). The ‘cossantes’ in the old Cancioneiros are
the work of men, but are attributed to women. There seems small reason to
doubt that this poetry was autochthonous. C. M. Bowra, Greek Lyric Poetry ,
Oxford, 1936, connects Sappho’s genius with folk-songs. In Chinese verse one
notes the prominence of women in the Confucian Odes, which are traditional.
Later lyrics are chiefly by men, but such conventions as the tableau of the
Deserted Wife must originally have been feminine.

Olive Dame Campbell and Cecil J. Sharp, English Folk-songs from the
Southern Appalachians , New York and London, 1917, pp. i x and x.

PERFORMANCE 39

teenth century in some of its items; but the fact was probably
unknown to her, or if known, of no importance.

Though these songs were unaccompanied, Cecil Sharp had come
across one singer who used a guitar, and the use of the ‘dulcimer’—
a shallow wooden box with four sound holes, three or four strings,
plucked with the fingers—has been more than once described. In
the Balkans and the Ukraine there are a number of instruments
associated with ballad-singing, such as the one-stringed ‘gusle’ of
Montenegro, the three-stringed Bulgarian ‘gadulka’, and the
Ukrainian bandura’ with many strings, as also wind-instruments
like the bagpipe among the Greeks. The simpler the instrument,
the more heroic the style.

A good number of English and Scottish ballads were sung only,
but others must have been danced. Even when they were merely
s ung, the performance was probably more extensive than it is now in
America. A number of preliminaries are gone through by Balkan
singers before they are in the mood to commence the ballad, or
sure of the attention of the audience. The programme of a fifteenth-
century Spanish entertainment has been preserved (not that of a
ballad-singer), 1 and shows that the performer began with a pious
ejaculation, then recited saws to catch attention, then burst into a bit
of the Libro de buen Amor , then more saws, an appeal to the public for
money, &c. The opening sections of the Libro de buen Amor are them-
selves admirably adapted for recitation in the streets and squares,
and include even the appeal to the audience not to scrape their feet.’

Midway between dancing and not dancing is the curious ritual
described in the Kalevala. The performance was, and is still,
accompanied by an instrument, the ‘kantele’. Vainamoinen’s
technique was as follows (Kirby’s translation):

In his hands the harp then taking,
very near he felt his pleasure,
and the frame he turned to heaven,
on his knees the knob then propping,
all the strings he put in order,
fit to make melodious music.

When he had the strings adjusted,
then the instrument was ready;
underneath his hands he placed it,
and across his knees he laid it,

1 R. Men&dez Pidal, Poesia juglaresca y juglares, Madrid, 1924, pp. 462-7.

performance

with his ten nails did he play it,

and he let five active fingers

draw the tunes from out the harp-string ,

making most delightful music.

Then the aged Vanaimoinen _

answered in the words which follow.

‘Are there any who are youthful,
of the noblest of the people,
who will clasp their hands together,
hook their hands in one another,
and begin to speak unto us,. _
swaying back and forth in singing,

that the day may be more joyful

and the evening be more blessed.

, CU stom of discarding clothes as they sing,

5ome singers hav ra dition must have been to sing naked.

m d it appears thal.the tine tradiionm ^ ^ ^

1 sway ng motion presumably belongs to the round dance.

Thl dance^ is thus seen to be a secondary element in ballad
but one which has exerted a profound influence In
Scandinavian countries all ballads are danced; but it is something
of a ‘tour de force’ to have associated dancing with a style so
wholly narrative. The older Scottish ballads were probably danced,
like their Scandinavian congeners, but later Scottish ones andmost
English ballads were more probably just sung. , n r ^ C -
‘complainte’ resembles the older ‘chansons de toile.; if that be just,
then both classes of popular song were feminine in bias, and the
difference lay between those which accompanied the dance and
those for sedentary occupations. It is unlikely that many German
ballads were danced, and the only connexion would lie m the pos¬
sible origin of all lyrical forms as dances. In Poland, however, an
intense interest in dancing has accentuated the lyrical aspect of
balladry so as almost to efface the narrative. In Spam the dance
has given lyrics only, and also in Portugal; the ballad is sung or
recited. In the Balkans danced songs are feminine, either explicitly
or by implication, and have richer versification and melodies than
the heroic chants of the men. Russian ‘byliny’ and ‘dump are
designed to be chanted.

TUNES

T HE history and mutual relations of ballad tunes constitute a
subject as complex and interesting as the study of their texts.
It is a field for musical experts, and into it I must not intrude, for
lack of skill, save in so far as the study of the words drives me. For
the words imply tunes. It is true that words and tunes are not
inseparable, and that even when we have associated the two we are
still far from reproducing the atmosphere of the traditional ballad.
We are, however, further on the way towards sharing in the
performance as it really was, and we know that the ballad could
not exist traditionally without an accompanying tune, though not
necessarily the same tune. Hence the tunes are found to have
exercised a formative influence on the texts, which even strictly
textual studies cannot ignore. The ballad is scanned by its tune.
In Greece and Yugoslavia this scansion involves shifting the
normal accents of the words. A Serb reading his heroic poetry finds
enough rhythm to enjoy, and may even appreciate as reliefs the
cases in which the accent conflicts with the trochaic rhythm; but
if he will learn to sing, he will find there is much greater regularity
in the scansion, coupled with a different sort of relief from
monotony. In Russia, those who have heard ‘byliny’ sung recog¬
nize an elusive rhythm which cannot be discovered in the words;
the words may be slightly altered, by the addition or omission of
syllables, under the influence of the tune. The syllabic irregularity
of English and Danish ballads is controlled by the melodies. In
Czechoslovakia we encounter a number of pieces in distichs and
triplets combined; by consulting the music we see that the metre
is triplet, and that the distichs must be lengthened by repeating
one of the lines. There are French 'chansons populaires 5 which
might seem to be composed in tirades like the Spanish ‘romances 5 ,
until one notices that the tunes imply in the one case a highly
lyrical utterance, in the other a narrative.

We may also feel reasonably certain that tunes have commended
ballads from one people to another, when the words were unknown.
There has been borrowing of tunes between ballads, as there has
been borrowing of words. It is a likely assumption that the tunes
may have assisted the words to migrate. Changes of musical

4615 «

42 TUNES

fashion have, in time, led to changes of ballad form. In both Fiance
and Spain, and also to some extent in Germany, there seems to
have been a passage from melodies composed for distichs to
melodies for quatrains, which has duly led to a preference for
quatrain form in the texts.

The study of tunes encounters a number of difficulties which
have not been adequately surmounted. The first of these is the
lack of tunes, properly authenticated and organized, to study. 1
The earlier collectors tended to ignore £ wild snatches of song and
obsolete airs 5 , so that far too many have perished. They are now
recorded eagerly. The process is difficult if the collector have only
the usual rudiments of a musical education, since he may be misled
as to rhythm, time, and pitch in pieces which employ fluctuating
rhythms, unfamiliar times, and intervals no longer known to
western music. The harm done by ignorant enthusiasm has been
put at a maximum by J. Tiersot. 2 Collectors, he states,

under the influence of men of letters, who have understood the interest
of the subject long before the musicians were aware, have either
neglected the musical part (which was no doubt better than to mis¬
handle it), or have treated it with an incompetence which reveals the
utter inexperience of the authors.

This censure may be too severe, since the writer has not troubled
to descend into details. When those are given, it may appear that
there is a good deal that can be used even in an erroneous tran¬
scription. 3 The naive zeal of the collectors at least preserved them

1 Apart from the magnificent work of Erk and Bohme, we owe to Germany the
Deutsche Volkslieder mit ihren Melodien, Berlin-Leipzig, 1935 ff., still in process
of edition. John Meier, the editor, and his helpers have reconsidered the words
and melodies already known and added new ones. They have traced, like their
predecessors but more fully, the history of the texts, but they have, for the first
time on a large scale, discussed the history and interrelations of the tunes. The
forty-eight pieces published so far are all narrative ballads. H. Holler’s Das Lied
der Volker, Mainz—Leipzig—London, Schott, n.d., 3 vols., contains specimens
of all European folk-poetry arranged on an ‘ethnophonic’ plan. It is a valuable
book, and has many interesting notes, especially on the influence of given tunes
on the great composers. The standpoint is predominantly lyrical; what this
book calls ballads are sparsely represented, even for the great ballad countries.
This is only natural, since the interest of lyrical tunes is often much greater than
those simpler airs used for narrative purposes. The word ‘bearbeitet’ occurs
trequermy and is, for comparative criticism, somewhat unsettling.

J. Tiersot in Doncieux’s Romancero, Paris, 1904, p. xli

' See ’ for “stance, C. Obreschkoff, Das bulgarische Voikslied, Bem-Leipzig,
ttaiipt, 1937, p. 91, where he complains of two transcriptions by E. Biicken.
i he transcriptions give us the right notes and bars, but (according to Obresch-

TUNES 43

from the gravest musicianly error, which is to set up a standard
tune tor a ballad, incorporating variants and ignoring different
versions The heresy of the standard text vitiates both words and

Tiersot° f ^ ° therW1Se admirable Romancero of Doncieux and'

To pursue this theme of the difficulty of transcribing tunes
accurately, we must further notice that the tunes are never quite
constant, either throughout the individual ballad or at different
performances If the air has been fixed on paper, we still do not
know ns subtle application to the ballad. To meet this difficulty

stored Tndln?^ tT ^ ph ° n0gra P h ’ and the records are then
ffisufficlm, 1 ?? at l61SUre - Even this process a PPears to be

t t0 CaptUrC aH the nUanCCS 0f BaIkan sin ging,
v th difficult rhythms, irrational intervals, optional flourishes and

other complexities. C. Obreschkoff demands for this purpos^ the

“hahJ Z TJ Z K t * pr0 f is mi6ht be some cerai "‘y »to

bether the intervals of one-third and quarter tones are reallv
theear mterV3lS ° rS P oken - The y occur in passages so rapid that
tries*** coT

appears to he ,nT nt d ° eS n0t OCCUr ’ and the greatest difficulty

heptatonic scales. 6 reC ° gmZing of modes and of the pentatonic and

much S mo e m to 6 eXtremeI y difficuIt to describe ballad tunes; how
much more to compare them. I cannot do better than reneat

£* *■ p -** ^ ^ ss

Stsrs**** W* are

TheZa"”it ^ T Z 7 T SitesaM

alterations IZSgfoZ aSem £ ^ there are
W, hue ^ leloS' tZ

If -ou.d be fap „,

since rhythms are very variable without afWt°° Sf mt ® matI0na l comparisons,
collector who has doZ yTomm sUce n^h^^ CUrve ‘ ° f ^°th e ;

transcriptions were uninteresting. * r ’ 11 was c °mplained that his

between the closest variations of the best-preserved melodies one always

fiSTdifietenc^, however slight, of f f ” “iS.he^gStS
some also whose relationship can only be divined through their general
to 5h similar rhythms and melodic cadences falling a ways on the
same places, though the outline of the song differs completely.

The paragraph expresses our difficulties concisely; it also serves
to illustrate the rather generalized language which musical critics
of this subject too often employ. Critical language should be
specific; it should be possible to state exactly what are the resem¬
blances and differences between tunes, and to do so by meansof
symbols which are intelligible to all. For this purpose it is doubtful
whether staff notation is convenient. Staff notation provides the
best standard of reference, but on points of detail it is costly to
insert fragments of staff notation in a printed page and to do so is
still to remain unintelligible to the unmusical reader. The use of
musical notation involves printing volumes of large size, which are
heavy to handle; and it is not at all convenient to enter on a card-
index which can be shuffled and compared. To compare such
tunes it is necessary to transpose them to a uniform key, since only
then will the likeness of the contours become apparent to the eye.
But it is hard to avoid a suspicion that transposition may injure in
some way a musical text already fixed with some hesitation. There
is, as we have seen, some doubt whether staff notation suffices to
represent the subtler nuances of certain kinds of folk-song. It also
implies—through its connexion with mechanical instruments like
the piano—a greater precision of measurement than is found in the
free narrative style of the human voice. It would be better, then,
having fixed the standard transcription as accurately as possible,
to leave it alone, and seek for some auxiliary notation, less costly
and cumbrous, and more intelligible to the musically uneducated,
more capable, also, of being entered on cards which can be shuffled
and compared.

Such a system has been devised by S. B. Hustvedt, of the Uni¬
versity of California at Los Angeles, 1 to serve to classify the English
and Scottish ballad tunes. There are two things to be done: we
must record the tune and we must characterize it. For purposes
of record it is sufficient to name the notes, give their length, and
add the key and time. A reference to the source of information
will help in later verifications. By using ordinary letters it is
1 See Note A, at the end of the book.

TUNES

TUNES 45

possible to put any ballad tune on a piece of paper not larger than
2 in. by 2 in., that is well within the smallest cards used in card¬
indexing. Thirty or forty such transcriptions can be before our
eyes if we wish to compare tunes, replacing advantageously the
one or two ponderous tomes of staff notation which are otherwise
the most feasible. Also no one who has learned an alphabet can
fail to grasp the notion of notes different in pitch, even if he were
tone-deaf. To characterize a tune, we have to remember that
ballad tunes vary, without losing identity, in key, pitch, rhythm,
and ornamentation. All that is constant is the melodic curve, or
significant parts of that curve. This curve makes a graph on the
lines of the staff notation, but the graph may be represented quasi-
algebraically, by using numbers for rising semitones, and letters
for falling semitones. We need more than ten rising semitones,
since there are twelve in an octave, so our numbers must be drawn
from different fonts of type: roman and italic, or black-ribbon and
red-ribbon on the typewriter. For most purposes of comparison
the melody becomes clear in the first phrase of a ballad tune, so
that it is enough to record it so. Identity of formula is very rare
without identity of tune. The first phrases of the French Roi
Renaud and En passant par la Lorraine happen to coincide in 72D
(i.e. a rise of seven semitones, rise of two, fall of four); the com¬
plete transcriptions indicate differences which seem to exclude the
chance of an association between the tunes. 1 On the other hand,
differences in the melodic contour do not necessarily sever the con¬
nexion between two tunes, since the difference may be super¬
induced on the original tune by some sort of ornament. Such
ornaments are wont to reveal themselves in that they leave sections
of the original tune in the same relative positions, that is, the orna¬
ment returns the same distance that it diverges. So, for two tunes
of Clerk Colvill with formulas BC322BB and BC32BC52BB, we
find the whole of the first in the second, with the addition of an
inserted BC5 (i.e. five semitones down and five up). Of course, as
we are measuring intervals only between one note and another, it

See Th. Garold, Chansons populates des XV‘ et XVI‘ siecles (Bibl. Roman )
Strasbourg, n.d., Appendix 3, 7. Taking the first two lines of the former, which

qU t ra f m f’ and 0mittlng , the refrains of the latter, one may accentuate the
similarity of the two tunes, without erasing the differences, thus •

R oi Renaud En passant.

.... 8U8 __ j fl at 68 U8

ddd/a. .abb/g.. ccc/d. .ccbja.r f.f/c.c/c.c/rf.b/ c.clf.c/c. b/a.g

46 TUNES

does not follow that a given interval refers to the same notes. It
does, when all previous intervals are identical, or when the 'ran -
tions amount to nothing, by cancelling out the new rises and falls.

It is clear from the above instances that we may use these
formulas and transcriptions not merely to say, as the musical
authorities do, that two tunes resemble each other, but precisely
how they resemble and differ; and that without a costly process in
print We can submit the proof to any average intelligence for
acceptance or refutation. We can readily classify and conveniently
compare melodies, and relate the notes as closely as we p ease with
the words The formula for the melodic contour is independent
of key, pitch, time, and accidents, and we can readily discount
ornamentation by noting whether the new intervals alter the rela¬
tive positions of the old. They will not do so if they sum up to
nothing, or if they occupy such places as the beginning and end of

the phrase. ^ .

The broadest affirmation for which the musical evidence can
serve is that European tunes are divided into two classes according
as their basis is the Gregorian or the Byzantine tradition. The
Gregorian chant itself owed something to Byzantium, though it is
difficult to know what, and the notion ‘Byzantine music’ is hard to
define. It is a theory of music which is not entirely congruent with
the practice of the Greek Church, and though there are numerous
points of contact between Greek folk-song and ecclesiastical music,
there seems to be no reason to treat them as identical. 1 What
it amounts to in practice is that the Balkans and Russia form a
musical region in which intervals of less than a semitone occur, and
tunes are liable to arabesque flourishes. Greek tunes are both un-
rhythmic and rhythmic, and use the modes of ‘re la ut’, both
diatonic and chromatic. Concerning the klephtic songs Mme
Merlier informs us that they are unrhythmical; the musical line
is as wavy as may be, and there is similar freedom in execution; the
music does not really admit of being written down. Within this
area the austere recitative manner of the ‘junacke pesme’ as sung
in the Dinaric Alps and Montenegro contrasts openly with the
elaborate melismas of south Serbia and Bulgaria.* The remaining
territories resemble the latter in style.

The existence of this Byzantine area of tunes is associated with

1 Melpo Merlier, Tragoudia tes Roumeles , Athens, 1931, p. k4.

2 W. Wunsch, Heldensdnger in Sudosteuropa , Leipzig, 1937.

TUNES

other common characteristics of the Balkans and Russia: com¬
munity of religious background and legends as shown, for instance,
in the prominence given to Elijah as a saint; negative comparisons
and great freedom in making birds or natural objects speak; a fairy
mythology which is largely common, together with reminiscences
of Slavonic deities; absence of rhyme. In western balladry we
have the opposite qualities to these, together with fixed intervals
and a greater fixity in the melodic contour. The West abstains
from ornament, save in the half-orientalized atmosphere of
Andalusia.

Within these two large divisions it is possible to discover s mall er
ones corresponding to the nations and their languages. These form
‘ethnophonetic’ groups, and the divisions are indicated in Moller’s
Lied der Volker. The reasons for such grouping are not so definite
as to be fully intelligible to me; but they do correspond pretty
closely with geographical facts and those groupings according to
form and substance which we have essayed in an earlier chapter.
Within the nation or community, musicians are able to discrimi-
nate. Sicilian songs have African and Greek colouring, and pass
easily from extreme to extreme; in Emilia and Romagna the tone
is minor, the changes gradual, the extension pentatonic, with a
flowing melodic line and cadence on the tonic; Piedmontese songs
are hesitant in the first part and of an attractive melancholy. 1
Andalusian music has highly imaginative cadences, which contrast
with the European major tonal system in the north and centre.
E. M. Torner gives an interesting demonstration of the changes
suffered by a given ballad (.Pastor que estds en el monte) in its passage
from south to north. 2

These broader considerations have their interest, but some
observations of a more limited range throw light on the develop¬
ment of the ballad in various regions. The simplest musical style
is that to be encountered in Montenegro, in the singing of men’s
songs (junacke pesme). There is not so much a melody as certain

2 p' 1 S°‘^ hlara ’ del P°P ol ° itali ano nei suoi canti, Milan, 1937.

1 VL. I orner, La Canci< 5 n tradicional espanola’, in F. Carreras Folklore v

ZvofrtTTb ESpana U BarCel °?i I931 ’ pp - 29_3 °- There are ^ cadences
volved. The second is an elaborate run of swift demisemiquavers:

T T , 34 U8: /((a.abagf.gagf))/e.

notesl^e: ’ nia ’ ^ Balearics > “d Tetuan the corresponding

9™ . ., J . 38 US: /fe./er

ihere are similar reductions of the ohrase in mnst _

4 8 TUNES

accepted principles of improvisation, which are applied In an
austere chant to the accompaniment of the ‘gusle’. 1 The principal
notes are c d e e /, of which c is slightly above our note, and e e
slightly below. The intervals are thus not exact tones or semitones.
The c serves in the exordium and antepenultimate; the d is normal
or final; the e is penultimate, scornful, and transitional; the / is
heroic. The reciter commences slowly (metronome 58) and be¬
comes more rapid (metronome 88), breaking off in his excitement
for an instrumental voluntary. He may change his time, and speed
up or slow down a line considerably. 2 In any case, the performance
is much more austere than that of the women’s songs of the same
region, and justifies the heroic minstrels in their claim that they
leave singing and dancing to the women.

There is something fundamental about this way of chanting.
The verse itself is probably not so very old, being borrowed from
Italy, perhaps in the fifteenth century; but it is treated in so old-
fashioned a way that critics have looked on the Montenegrin
‘guslari’ as the likeliest approximation to a Teutonic scop or skald
or even a Homeric rhapsode. The problem before the reciter of
the Spanish Poema de mio Cid was probably like that of the Monte¬
negrin singers, save that he had to contend also with irregular
rhythms, like those of Russian ‘byliny’ and Ukrainian ‘dumih In
the singing of these poems the performers improvise on general
principles, but they use a range of notes and variety of pitch
which are unknown in the highlands. This approach towards
developed melody is seen in Bulgaria and South Serbia; in Greece

W. Wiinsch, Heldensdnger in Sudosteuropa , Leipzig, 1937, pp. 24—5. I give
the information as I have received it. The transcription given below uses
different notes.

, .. ,® as Redder Volker, Mainz-Leipzig, ii, Serbian 29 is the famous

ballad The Maiden of Kosovo (Kosovka ievojka). The first four lines run:

2 flats 24 U8: ffcba/cbag/b.a./g.../
bbab/cbab/a.gr/
gbab/cbab/a.gr/

„ _ d.c./b.a./c.b./a.g./a. .(gfg)/a.../

The emphatic treatment of the fourth line, which is the third repeated, is notable.

13 n b ^°f. W ™ n com P ared with a woman’s song, in lyrical deca-
syllables, also collected in Montenegro. No. 32 runs:

4 flats 44 34 U8: ff/g.(gabg)ag/f.c.. . ff/g . (gaba)ag/f.. .r/

44 c..hd.c. I

_ , . c. .dcagf/f. .cc. . ./c.6.a.gf/g.. .g.../

Sir; “hrjd'Xt “ ■ y " ,b1 '" ** ^ -

TUNES 49

the line is subject to ornamental flourishes which are hard to fix
in symbols. 1 In central and southern Serbia the two techniques
overlap.

The musical conditions of the south-western peninsula of
Europe are also interesting. There is a considerable range of tunes,
from the simplest melodic phrases to complex polyphonic settings.
The principal types are thus defined by E. M. Torner: 2

(i) Epic style. Completely unknown ‘at the moment when this poetical
genre began to be formed by derivation from the great epic poems’.

(ii) Recitative. Characterized by an almost straight melodic line,
formed by one sustained note and two or four cadential accents, accord¬
ing to requirements.

(iii) Melodic. The melodic curve is more pronounced, with more
sonorous intervals, particularized rhythms under some time unity; the
declamation is still strictly united to the requirements of the text.

(iv) Expressive. A melodic curve perfectly defined in every way.

The author proceeds to qualify these distinctions. As to the first
style, it is necessary to proceed with the more caution since it is
not actually known that ballads arose from epic texts. That is
simply a theory. There are only two ballads which can be set
beside the words of existing epic texts, and they show differences
as well as correspondences. The epic texts, so far as we can trust
ffiem as transmitted to us, are composed in lines of no fixed rhy thm
There is a medial pause and an assonance at the close. Half-lines
of seven and eight syllables predominate, but there are more
extravagant oscillations also. Some readers, especially those of
Dutch or German extraction, have believed they could detect four
accents in most lines, but that is not according to the normal spoken
accentuation of the words. The declamation of Russian ‘byliny’
shows that there are ways in which irregular texts may be sub-

J l 0T Z ^ tan l e ’ the B ^rian Kroli Marko in W. Wiinsch, Heldensanger,

PP 7lir 8 ’ o he T r T 0 o C decasyllaI ? Ies > but has *is elaborate development: §
Metronome 80 U8: ggggc..ah.(ab)g.. .d. .cc. a.g. .fa. .g

gef(.e)d.f. .efed(.c)d.. .g...
c. ..a.. .g.. .d. .f.. efed. d. • -g...

(not to mention certam grace notes). Greek ballads may be very simple in style
(cf. S. Baud-Bovy, Chansonspopulaires du Dodecanese , Athens, Sideris ioie i
SiLL 3 ): 24 ’ CaStell0riz0 but they are liable to runs like the followi’ng

oSStsSsslT*' U8: “'“ ddi •**&&**

r 2 i n t F ii C % ieT T’ F ° lklore y costumbres, Barcelona, 1931, ii, p . I?0 d t i n „
G. Castnllo, Estudio sobre el canto popular castellano, Palencia, x 9 % S

4615 * * *v

TUNES

mitted to a measure, itself perceptible but indefinite; 1 and this may
have been the epic manner in Spain.

The Spanish ballad may have arisen because of a new musical
manner, as well as through new subjects (civil and frontier wars,
foreign ballads) and a new episodic way of regarding the old sub¬
jects. The unit of musical construction is the phrase for an octo¬
syllable of text; therefore we may say with assurance that the
ballad metre is octosyllabic. The unit was doubled to cover the
sixteen syllables of each sentence, which was closed by an asso¬
nance and a pause. So the form would be
aa aa aa aa, &c.,

since the pauses group the musical phrases in couples. Salinas in
his Be Musica y xv (1577), gives an example of the ‘antiquissimus et
simplicissimus cantus quae Romances appellantur quibus historise
seu fabulas narrantur’ as a tune for Count Alar cos
U8: bbbbc . caa . bbbbc.raa.

The next steps were to introduce more definite time, and to vary
the unit on repetition so as to close the couplet with something like
a cadence. A tune for Gerineldo is

34 U8: ggbagg/b . a... / ggbagf/gg. gr/

The ballad would then run in musical phrases like aa aa' aa' aa'.
This leads naturally on to tunes composed of two phrases: a/3 a/8

a/8. The words and music of such ballads could be written in lines
of sixteen syllables if convenient, and that is how Salinas writes
the music in 1577, and Nebrija the words in 1492; but the practice
does not alter the fact that the unit of construction covers eight
syllables only.

This early austere manner gave way, probably during the later
fifteenth century, to the more developed phrases of quatrains. The
transition may have been effected gradually. Repeating the octo¬
syllables gives the form aafifi aaffi for one tune of The Month of May
(Mes de Mayo). Repetition of the words with variations on the
melody gives us a tune for Alhama of the form aaa% with refrain,
and for Gerineldo aa'fiy. Any of these tunes would still fit Nebriia’s
description of old ballads, since they covered sentences of sixteen
syllables.

It is so that the
1496, only four
. By the end of

See Note B, at the end of the book.

The final step was the quatrain proper: a/ 3 yS.
musician-poet Encina describes Spanish ballads in
years after Nebrija had described them differently

TUNES 51

the fifteenth century the musician expected to set ballads to four
musical phrases. This pattern slowly imposed itself on the texts,
so that the artistic imitators of the ballad style in the later sixteenth
century adopted the quatrain as the normal sentence. But for at
least fifty years after Encina wrote, it would not have been accurate
to describe ballad texts as quatrains; the sentences continued to
end with the sixteenth syllable. 1 Ballads were falling into the hands
of expert musicians, who gave them polyphonic settings for
chamber concert parties. These settings are too elaborate to be
indicated by the symbols we have been using to denote musical
phrases, since the phrases are subject to many slight variations.
Roughly speaking, however, the first sixteen syllables of The Burn¬
ing of Rome (Roma abrasada), as interpreted by Matos Flecha 2 in
the sixteenth century, to cite only the treble part, run like this:
aa a cnrfiypy, where a is the first octosyllable, /? the first words of it,
077 the last words of it echoed variously, and y the second octo¬
syllable, with flourishes. Such a fashion immensely slowed up the
performance of any given ballad, and helps to explain why the
Cancioneros de Romances , when they first appear in the middle
years of the sixteenth century, consist of truncated ballads. There
are many indications that the text of the medieval ballad was
normally longer than that of those examples now accessible to us.

A careful arrangement of tunes will thus help us to divine, if not
to ascertain, the history of the ballad form in Spain between 1350
and 1550. They may suggest, on occasion, some specifically
foreign influence. The Castilian ballad is free from the rule
operating in France and North Italy, by which the halves of a long
line must not have the same kind of ending: masculine and femi¬
nine, or feminine and masculine endings are obligatory in the two
hemistichs. A Castilian tune allowing for this distinction is highly
indicative of foreign provenience, as with a very simple setting of
Rosafresca recorded by Salinas:

68 U8: ee/d.ef . ejd.cc ee/d.ef. e/d.

In Catalonia one encounters such settings without surprise. 3

A simple setting of this sort, for Don Bueso, is quoted by E M Tomer in
^T^folkloricos, Madrid, x 935 , p. *3* - now sung aLtgthTj'sof

• si 8 T^^: P b 66 *“«*- b /.bg, g/ f,d. w

- \ j? in ft ay Briz ’. Cans °ns de la Terra, Barcelona, 1866-87, i, p. ««• Los
Estudtans de Tortosa; m, p. 65: La Escrivana ; iii, p. 111: Los tres tambors ; i, p. 63:

.* TUNES

Into the musical patterns of other countries there is the less need
to inquire, since they are all more developed than those we have
been considering. The traditional song in France took on a new
lease of life in the mid-fifteenth century, and is highly lyrical. It is
characterized by the use of nonsense refrains. Settings for distichs
(or divided long lines) seem to be earlier than those for quatrains;
but the quatrains certainly reach back to the same century. This
kind of folk-song extended into Italy, deeply affecting the song-
books of the early sixteenth century, and they are abundant in
Piedmont and Lombardy. The North Italian tunes are sometimes
more primitive in technique, and the style is more narrative than
in France. 1 In Scandinavia couplets tend to take on the form of
quatrains by inserting refrains between the lines, or by repeating
the last line and a half of the previous couplet; or they are quatrains
with a refrain at the close. These refrains are intelligible clauses,
though not necessarily related to the matter of the ballad. In
Germany the refrain is much rarer, but there is a tendency to echo
the last three or four syllables of a line. In Czechoslovakia there is
a certain disinclination for the refrain, and a cult of the triplet,
while the music often shows a long note marking the middle of the
phrase. So one may go on accumulating notes of this kind. If they
add nothing new, they help to confirm the divisions of European
balladry which we have already inferred from other evidence.
There is no such sporadic scattering of techniques as would be the
natural result, if the ballad was everywhere a product of the soil.
Simple as they are, ballads are an art form and have to be learned.

The tunes must also be scanned by literary students for such
light as they may throw on the diffusion of texts. The texts need
not have the same history as their tunes, since the latter are various
and are capable of migrating alone. When we compare the tunes
of related ballads, we are more often aware of difference than of
identity. None the less, there are cases in which words and tunes
have travelled together, and it may then happen that the tune con-

la Dama d'Arago. The first certainly and the third probably are French, the
second is Proven9al, and the fourth probably of Greek origin, but perhaps of
immediate Provencal provenience.

1 For instance, Donna Lombarda has a simple narrative setting (aj8), and a
more lyrical melody for quatrains, with the refrain thrice repeated (aBaBooo'):
2 flats 68 U8 3 flats 38 U8

fbdjf.df.dlcb eddjc. .edd/ci g(c. .d)/ec.j a(a..e)ldc./ bis
__ ’ edS/cij eiE/cr/ ecrf>£/../cr/ refrain.

Ihe tunes are related as 543 C3CBB to 521C.

TUNES . S 3

tains some association of detail which is not found in the words;
it may, for instance, be found in a particular part of the original
country, or arise at some specified date. Were all collections of
ballad tunes as ample as those of Germany, it is probable that we
should be able to notice more migrations of melodies than at
present appear.

A simple instance of migration (because of its relatively recent
date) is Marlbrough s’en va en guerre. The tune may come from a
seventeenth-century hunting song, but it suddenly sprang into
popularity in 1781 through being sung by the Dauphin’s nurse
and taken up by Marie Antoinette. It spread abroad so rapidly
that Goethe heard it almost everywhere on the road to Naples.
The melody has the curves indicated by the formulas 4B3A in
France (Doncieux), B3A in Germany (Erk und Bohme), and
()BziA in Catalonia (Pelay Briz). The French and German tunes
are identical in all but the first interval, and a detailed transcription
would show that this affects only one note. The Catalan differs in
the first interval, and takes the third in two stages; it also differs in
time and key, and in the arrangement of the lines. In an older song
traditional variations would be more numerous, as in The poisoned
Marchioness. This also is a French song, referring to the hapless
Gabrielle d’Estrees, who died in 1599, under the usual Renais¬
sance suspicion of poisoning. The tune occurs in France in the
variants BB25C, B225BBA, 522, E1ABB22, and in Germany in the
variants BB22, BB223C, BBC73C. In the course of four centuries
considerable changes have occurred in the melody, but one can still
discern the group BB22, or its mathematical equivalents, in most
of the versions. The Shipman (El Marinero) is a ballad of French
origin, and its passage from Catalonia to Castile is evidenced by
the formulas 522BB and 522BBE. In the same way we may trace
La Porcheronne from France to Catalonia, and Lord Randal from
England to Italy. In really old cases of migration the evidence
is often obscure. The migrating tune may have died out in its
original land, or it may have been changed out of all recognition.
So the famous Herr Rihbolt , which is our Douglas Tragedy, has
one Danish tune (IC5214B221) which might be connected with
a Swedish tune (IC212C243B); another two (535A35C and
5E5E5221A) which seem related to the Norse (5E54BG) and also
to a simpler Danish tune (522D2) which resembles one in Norway
(5221AB21A). These last two tunes are the nearest one can bring

TUNES

forward to the English (54BBAB), while there are two others in
English which are connected with each other but apparently not
with Scandinavia. One notes that The Harp's Power appears
in Denmark and Sweden under similar musical conditions
(21CBC912C1A and 1212CB2BBAB); but the very popular Elve -
skud, for which there are a number of related Danish tunes, is sung
very differently in Sweden and in Norway. What we have to
expect with the oldest ballads are variations, and it is never easy
to be sure when variation has gone so far as to destroy all resem¬
blance. It is very difficult to demonstrate migration within the
Scandinavian area, though the words be almost identical. Lave and
Jon, a very popular humorous piece, has three Danish variations of
the same tune (5221C25BC, 52B221C2,5A12B412) connected with
the Swedish (52B22B23) and Norse (52B221C2, 322B2DA12); but
one Norse and two Danish melodies seem unrelated with this or
each other. It seems as if we must credit the Scandinavian min¬
strels with a higher degree of musical, than textual, inventiveness.

Where a considerable number of tunes have been preserved,
they may be gathered into classes, and in the case of a migratory
ballad, we may note which is the class of the melody that has
travelled. The fine German ballad of The Castle in Austria is a
convenient instance. Erk and Bohme offer us fourteen airs. One
group has the formulas 2212C and 2212CD, which correspond
with the Norse 221C52BC; a German variant is B212C, corre¬
sponding to the Danish B25BC. An entirely different class con-
tains tunes with the formulas A4ABA and E8ABA, corresponding
to the Swedish A1E8AD5A. They show there has been a diver-
gence between the Swedish tradition and that of the neighbour¬
ing lands. ^ The Swedish use their tune also for the ballad of
The Sultan's Daughter , which has quite different formulas in Ger-

ma ^ nd the L ° W Countries (5223CAB, 523A, 2127EB, 21ABB,
43BCB4).

, ■ ~* Det Wasser : Zwei Konigskinder and Hallewijn are ballads, pro¬
bably originating m the Low Countries, which have attained uni-
versal popularity. The first two are forms, lyrical and narrative
respectively, of the Leander saga. German and Dutch tunes can
b ^m gr ° U £ ed accordin g to their first intervals, as o (9BB0A2CB

Si 9D f B \f B t 3E >' 5 ( 5 *-A:.C 2I A),'andT fcA&S
21AB2122) while the formula 54122BBA finks the first two

groups. The Czech 225BBA is based on 9BBA. Denmark used

TUNES 55

tunes of the first group (92BBA12BA, 543C1ADAB2), and also
the third (521AB2122, if we ignore the first interval, but this tune
is midway between the second and third groups). The end of the
third formula appears in Sweden (21C21C22122), and the French
5AB212B contains the group AB212 which appears in the third
group of tunes. The relationships are very complicated, but they
clearly appear on analysis. Similarly, in the Hallewijn melodies
there is a grouping of intervals (2212) which appears in a fair
state of preservation in German, Danish, Swedish, and English.
Another German group has a cadence BBA, which is to be found
in the Lusatian (or Wendish or Sorb) tune.

The object of this chapter, which has already gone too far for a
tyro, is to show that the history of tunes is a necessary adjunct to
the discussion of ballad form and propagation. If that be clear, the
application may well be left in more expert hands. The expert I
merely entreat to speak in preciser accents than hitherto, and to
use such signs and reasonings as are readily understood by the
common reader.

[W. Danckert s Das europdische Volkshed , Berlin, 1939, unfortu
nately came too late to be used for this chapter.]

KINDS AND DATES

H ISTORICAL ballads form a class which it is comparatively
easy to separate from all others. They arise immediately out
of the events they narrate, not later than within the memory of
living men. Any corpus which is rich in ballads of this kind offers
an easy and sure chronology; for not only are the events datable,
but there are often other lines of testimony as to the age of the
ballad. It may, for instance, have provided material for a chronicle
or be the subject of an allusion. In the older strata of European
balladry, historical pieces are, like narratives in general, copious;
and the earliest dates they imply are as old as any that can be fixed
for the genre itself. In the course of time historical ballads die out
more rapidly than others, and they are seldom passed on to other
nations. If they migrate (as the French ballads on Francois Fs
imprisonment and Gabrielle d’Estrees’s death have passed to Italy
and Germany), this will generally be due to their novelesque or
emotional interest, or to the charm of a tune. Because they are
perishable and stationary elements of the older epochs of balladry,
they are not abundant in ballad corpora of secondary formation.
Their absence is one of the signs of dependence, as of Bohemia on
Germany and Bulgaria on Serbia.

National themes are encountered among such ballads in episodic
or personal aspects. The ballad does not compete with the epic
poem. More often the theme is partisan or local rather than
national: civil wars and frontier defence. The pieces may cohere
round the name of a national hero, like Marko Kraljevic and Marsk
Stig, and in such a case the historical element may become ex¬
ceedingly tenuous. In Russia there were persons known as
Dobrynja Nikitic, Aljosa Popovic, Vladimir, &c., and that is almost
all that is historical in the Kiev cycle. One may then justly doubt
whether these foyliny’ should be classed as historical at all, or
merely as adventurous. They are accepted as historical by the
hearers, who are not much preoccupied with the accuracy of the
details. On a somewhat lower scale are those ballads which take
for heroes the persons of outlaws, haiduks, klephts, robber barons,
bandits murderers, and plain thieves ; and these, too, have their
descending order of merit. At the one end of the scale we encounter

KINDS AND DATES 57

Robin Hood and Marko Kraljevic, who are regarded as almost
wholly admirable. Old Novak, the haiduk, is less heroic than
Marko Kraljevic, but of invincible independence. There is some
esteem to spare for a robber baron like Lindenschmid, who is cap¬
tured and executed, after a gallant fight; the Catalan Serrallonga
and those of his kidney are used to point a vulgar moral about evil
communications which lead to the scaffold. At last we read
notices of a revolting nature or shocking tragedies. Whether such
have historical backing is not always easy to ascertain; the treat-
ment is so generalized that the event might be merely imagined in
order to point a moral. It is then uncertain whether the piece
should be called historical or not. Fortunately the dilemma is not
an important one, since the base metal of most of these pieces does
not deserve assaying. The descending curve of merit has also its
chronological aspect, since the more admirable heroes belong to the
earlier epochs when the people remained homogeneous, but ballads
of bull-fighters and murderers are plebeian and late. We have to
add to this class those ballads which belong to professions that have
themselves an historical date: professions such as the military,
with its succession (in Germany) of Swiss pikemen, German
reiters and landsknechts, the victims of the corvees of the eigh¬
teenth century, and modern conscripts. Historical data are often
embedded in their songs—more richly in the older strata—though
the permanent subject is the vicissitudes of the profession. With
these ballads also the exact repartition is a matter of tact, and a
piece may be given a double classification.

For historical ballads it is important that there should be some
direct connexion between the song and the event, condition, or
person it represents. A ballad carved out of a chronicle is not
historical in the same sense, though it may be convenient on occa¬
sion to overlook the difference. But we should rather insist on the
difference in the case of epical ballads. There was a Theodorich
of Verona, but the ballads of Dietrich von Bern have no direct
relationship to him, but only to the hypothetical epics and real
saga which lie between. The same is true in Spain of Ruy Diaz
the Cid, Count Fernan Gonzalez, or King Rodrigo. The epics on
the first two and the traditions about the last may be rightly de¬
scribed as historical, but the ballads are dependent on the previous
oral and written literature, and are therefore akin to other liter¬
ary ballads. It is true that the epical and historical ‘romances 7

4615 T

5 $ KINDS AND DATES

combine to present the Castilian with an Impressive summaiy of his
national history; but to class them as historical in the strict sense
makes comparison with other ballad corpora difficult, and induces
errors in criticism. There are here also cases open to doubt. The
exploits of IPja of Murom, in Russia, form something like an epic
plot, but there is no external evidence for such a poem. To include
his ballads among the historical ones of Kiev is to stretch the term
to Its widest limits. In Greece there are extant ballads and epics
concerning Digenis Akritas, and their relationships are open to
doubt. If we class them as epical rather than historical, we do so
for convenience of comparison only.

In a second class we may distinguish those ballads which depend
on a previous literary tradition, whether written or oral. Despite
etymology, the mere writing down of a poem or tradition does not
constitute literature; they may be carried in the memory with
equal precision of outline and detail. Since ballad literature is
essentially oral, the dependence on oral originals is the more im¬
portant for criticism, and it has been explained by various theories
which we must examine later. Here it will suffice to enumerate the
principal possibilities. The original may be a traditional epic poem
or an Eddie fragment or a saga based on the epic poem or tradition.
The more ancient narratives of this type are Germanic, the
youngest are Castilian. The former give rise to speculations of
wide range, full of inevitable hypotheses; the latter give narrower
observations of a maximum certitude. Romantic sagas may also be
the sources of ballads, together with novels and novelettes, such
as those of the Carolingian and Arthurian cycles in Spain. There
are cases in which the novelesque form is obvious (e.g. the Castilian
Count Amaldos ), but there is no known original; and there are
ballad novels, like Axel Tordsson or Henry the Lion or Tannhduser ,
which might be classified as literary or as adventurous. In such
cases one’s choice will be influenced by convenience of arrange-
ment within a particular corpus (Tannhduser may also be deemed
historical), taking due note of any formulas of a definitely literary
sort. These include such well-known conventions as the pastoral,
dawn songs, assignations, debates, and tall stories. In the case of
Tannhduser , the opening is an ‘aubade’ and the motif of Frau Venus
is drawn from courtly literature. There is, indeed, no absolute
line separating any ballad from written literature, and in some
countries, like France and Italy, the most 'popular’ songs are shot

KINDS AND DATES 59

through and through with literary reminiscences. The lower
border of the literary class of ballads must necessarily be left
fluctuating.

A small, but very special, class is composed of those ballads
which depend on classical originals: the legends of Leander,
Pyramus, Helen, and Troilus. They are interesting for the distance
they have travelled from their literary originals, coupled with their
undoubted popularity. The wrestles between first Digenis, and
then various unnamed young men, and Charon are doubtless sur¬
vivals of the old legends of Hercules.

Religious ballads imply a literary tradition of a special type. The
motive for creating or singing a religious piece is not mere enter¬
tainment, but edification (coupled with mendicity). They are
recorded earlier than profane verses, and they show more or less
influence from the original text. Genuinely popular carols 1 have
fragments of Latin or are otherwise bilingual, and are not so
wholly entrusted to tradition as to come within the corners of this
study. The greater number of religious ballads are open to correc-
tion from the scriptures, and those which are of most interest to us
are those which have been so transformed by tradition as to sever
the biblical connexion. This is the case with the Hallewijn ballads,
so unlike the history of Judith, and The Samaritan Woman, con¬
fused with the Magdalene and made into the type of the unpardon-
able sinner, or the Serbian account of Peter’s mother, in the same
function. Other biblical suggestions are the story of Benjamin’s
sack (the Spanish Pilgrim to Compostela, Russian Forty Pilgrims),
Potiphar’s wife and Joseph (The Chaste Servant in Germany),
Dives and Lazarus, and the hardness of Dives’ wife’s heart (a
begging ballad), Samson as a Russian ‘bogatyr’, Solomon tricked
by women, Judith and Holofernes ( Hallewijn ). Then there are
saints legends, especially the Marian legends and narratives con-
cerning St. Catherine, St. George, St. Lawrence, and St. Nicholas.
Finally, there are many ballads which point some religious moral,
particularly those which censure stinginess. The inspiration of
these pieces is everywhere the same, after allowance is made for the
difference between the Greek and Roman churches; coincidence
in religious balladry is, consequently, no evidence of direct associa¬
tion between two pieces. For example, the apocryphal miracle of
the roasted capon, which flapped its wings and crew, is associated
1 E. L. Greene, The Early English Carols, Oxford, 1935.

KINDS AND DATES

with St Stephen in England and the Pilgrim to Compostela in
Spain, but the two ballads are not akin.

In more didactic or lyrical styles the religious class stretches out
to cover small dramas; death, resurrection, and the judgement;
prayers, praise, thanks, consolation, confession, legacies; and the
complete calendar of church festivals. This fullness of material is
especially noticeable in Balkan balladry.

To the remaining ballads the word adventurous may be applied.
They relate some event which is interesting; which is an adventure.
Any more precise term would not cover the miscellaneous con¬

tents of this class, which receives those not otherwise placed. It is
not homogeneous. One group, most ample in Lithuania and
Latvia, is the mythological and superhuman. The gods, Sun and
Moon, appear as actors; or there are superhuman characters drawn
from the fairy world of decadent paganism. Earth spirits—elves,
dwarfs, kobolds—water spirits—nixes, mermen, swans—‘vile* and
nereids, dragons and snakes, changelings and bewitched persons,
rey^iants, death and Charon, the Venusberg and the Earthly
Paradise, are themes of ballads reverend for their age or moving for
their mysterious force. Then there are innumerable love ballads:

encounters, happy love, opposition overcome, sorrow and separa¬
tion, tragedy, prevention of bigamy or incest, reunion, adultery,
murder for love, incest, rape, faithfulness in trial, the sad case of
the nun, bride-stealing, death. It is always a material consideration
m ballad poetry whether a crime be a ‘crime passioneP. There are
ballads on the crimes which seem popularly most abhorrent-
cruelty by step-parents or mothers-in-law, poisoning, murder of
husband or wife, parenticide, and infanticide, the worst of all
These pieces may be reports of real crimes or plebeian ballads of
morbid tendency. The number of ballads dealing with prisoners,

2STr? eSCapeS ’ ^ 3 SCarin § COmment on h ^an

kindness, we find1 the .innocent in prison, who is executed or
escapes, or languishes with only a bird for consolation; the prisoner
who escapes by the aid of the jailer’s daughter, sometimes marry-
mg her, sometimes deserting; the ruses of devoted wives and thdr
self-sacrifice; rescue by force or by counterfeiting death If we

° n ° a sectl0n of ballad novelettes under the literary heading

we^shall have to record here such highly developed adwntures Is
those of Tannhauser, the noble Morinver Blueh^rH u Y

Lion, ,he Count in the plough, Wilhehn Tdl, the girl

KINDS AND DATES 61

wax, and the squire of low degree. These are ballads which arise
from many motifs which have cohered into a fully developed tale,
which travels as a unit away from its original focus.

These types of ballads are variously distributed in the different
countries, and each has been made, in consequence, the basis of a
theory as to the date and nature of ballads. To take the richest
areas only: Spain has a sturdy tradition of historical and epical
ballads, but borrows material from foreign romances and almost all
her adventure pieces. England and Scotland are poor in epic frag¬
ments, mediocre in history, but rich in adventure ballads, with
special reference to the supernatural In Denmark there are these
things, and a sturdier tradition of epos and history. Germany has
very many semi-popular historical poems, 1 of which a considerably
reduced number may be admitted as traditional ballads. There are
a few epical derivatives, less close to their originals than the Danish
viser . The lyi ical element is strong, and in course of time vir¬
tually submerges the narrative; German balladry thus appears
especially songful. If we can accept the cycles of Kiev and Nov¬
gorod as historical, Russian ‘byliny’ are almost wholly historical;
as are also the dumi J of the Ukraine. There are no epics extant
related to these pieces, nor much likelihood that there were. The
lack of adventure ballads is compensated by the heavy proportions
of romance in the so-called historical pieces, so that they have been
described, not unjustly, as wholly ‘sagenhaft’. The mythological
themes are few; lyrical treatment is kept for other kinds of folk
poetry. In Serbia there were no epics, and no epical ballads. The
historical series is exceptionally fine and veracious. Adventure
ballads fall into a debatable land between the heroics of the men
and the erotics of the women. Lyricism is not a quality of the
junacke pesmel In Greece we have epics and ballads in a relation
that has not been made clear. The historical ballads are poorer and
later than in Serbia, and the adventure ballads more important.
The lyrical element in Greek folk-poetry is more prominent, and
is not excluded from narrative pieces. It is in Lithuania and Latvia
that mythological balladry reaches its fullest development. There
are only lyrical ballads, with some narrative intermixed; no ballad
novelettes, and only scattered allusions to historical events. The

* Collected by R. von Liliencron, Die historischen Volkslieder der Deutschen,
Leipzig, 1865-9, and continued by other collectors for later centuries. The
selection of true ballads from this mass has been made by Erk and Bohnie.

62 KINDS AND DATES

chief occupation of the singers in all four east Baltic lands is to sing
lyrics appropriate to domestic occasions. In Finland we have to
take account of an exceptionally rich literature of magical formulas.

As a result of this uneven distribution, critics in different coun¬
tries have exalted now one, now another, of the types as the ballad
‘par excellence’. Andrew Lang, 1 stimulated by the rich super¬
natural element of our poetry, was induced to declare that ballads
Svere the immemorial inheritance at least of all European peoples’,
and that their stock situations and ideas are ‘of dateless age and
world-wide diffusion’. Isolated motifs are, indeed, often uni¬
versal, but their coherence in the extant ballads is by no means
‘dateless’. W. Schmidt, writing in Anglia xlix, considers that our
oldest truly traditional ballads are of the fourteenth century only.
There is clearly a difference between the history of an isolated
motif, which may be traced from Homer and Herodotus through
Greek mythology and German fairy tales, and the history of the
complex utterance which is a ballad. Chodzko drew a similarly
exaggerated inference from the mythology of the Lithuanian
corpus, assigning all such pieces to the pagan era before Vladimir
of Kiev. We know, however, that paganism was rife in the
Lithuanian woods as late as the sixteenth and seventeenth cen¬
turies, so that a mention of Perkunas or Laime need not be so
very old.

If attention be paid to the religious ballads there is a tendency
to slip into a theory of clerical origins. The English Judas is our
oldest recorded ballad, going back to the thirteenth century, and
the German Armer Judas has been often cited since the fifteenth
century, being a rendering of a Latin hymn of the century previous.
In other countries there is no similar priority to be claimed for the
religious poems, which have always one unusual feature; viz. the

ease with which they may be compared to a literary text, the Bible
known to all. ’

Spanish criticism has insisted on the importance of the epical
ballads. Having shown that the ballads are always younger than
the epics the great interpreters have gone on to a theory of ballad
origin by fragmentation of epics. The theory clearly will not apply
to countries which have had no traditional epics, though they havl
ab„»da M ballads. In Scandinavia we find'a pr^e £, Serted
between the epos and the corresponding ballads” viz. the tSSZ
Ballads , Encyclopaedia Britannica, nth ed.

KINDS AND DATES 63

saga; and in other cases fragmentation will not account for the
wide differences between epos and ballad. The same is true of
Germany, where the ballads are even further removed from the
epics. Nor will fragmentation explain the historical ballads of
Spain, which are not fragments of long poems but songs essentially
episodic; nor explain the borrowing of adventurous matter from
abroad. What traditional epics can teach us is, chiefly, the rela¬
tively late rise of the ballad genre. In Spain the ‘cantares de gesta*
were still flourishing in 1344, when the scribes of the Second
General Chronicle copied in prose the text of a second Infantes de
Lara, unmistakably epical. 1 The styles are successive, not con¬
current, 2 so that the mid-fourteenth century is indicated as a likely
moment for the rise of the Castilian ballad. Similarly in Scandi-
navia the Eddie poems come down to the eleventh century, and
they employ a rhymeless, alliterative technique quite distant from
the Vise^ style. The rhymed, roughly syllabic, linguistically
medieval viser are thus clearly not older than the twelfth century,
at the soonest. The interposition of the early thirteenth-century
Thidrekssaga forbids placing a number of the epical Viser* earlier
than then.

The epic evidence of Greece is more difficult to assess. We have
a clear allusion to ballad singers in the tenth century: ‘accursed
Paphlagonians who put together songs about the experiences of
famous men, singing them at an obol apiece at each house.* 3
Digenis Akritas has been identified as an Anatolian turmarch who
fell at Kopidnadon on the Euphrates frontier in 788. H. Gregoire
seeks to carry his epic back to the mid-tenth century, and repre-
sents the ballads as anterior and independent. The argument is
not entirely conclusive, but may serve to carry the ballads of Asia
Minor back at least to the twelfth century, if not sooner. We

1 See R. Men&idez Pidal, Leyenda de los Infantes de Lara, Madrid, 1934.

2 ft- D Atkinson, in the Modern Language Review, xxxii, 1937, and Hispanic
Review, iv, 1936, argues that the epics flourished into the middle of the fifteenth
century, so that the epical ballads may be quite recently derived. One may say
that an epical ballad may be created by any minstrel who has a memory of an
epic text, that is to say, perhaps quite late; but there is no such evidence of a
vigorous crop of new epics in the fifteenth century as in the Segunda Cronica
General of the mid-fourteenth.

3 Arethas of Caesarea (850-932), cited by S. Kyriakides, and after him by
H. Gregoire. The latter’s articles, ‘Le tombeau et la date de Dig&iis Akritas’
and ‘Autour de Dig&iis Akritas’, Byzantion, vi, vii, 1931-2, are the principal
authority for Greek ballad and epic origins.

KINDS AND DATES

discover their 'politic" metre in much use at that date. By way of
contrast, the ballads of the mainland are rather late. Those of an
historical cast open with a fragment on a siege of Adrianople,
which may be the siege of 1361; then follow some laments for the
fall of Constantinople in 1453, and for other disasters of that kind;
but the full flow of historical ballads only begins with the klephts
of the seventeenth century and later. Among the adventure ballads,
also, there is a marked Anatolian priority.

Yugoslav men’s songs (junacke pesme) are older than the
'tragoudia’ of the mainland Greeks, but younger than those of
Asia Minor. The oldest strata concern the disaster at Kosovo in
1389; they are the Kosovo ballads proper, and the ballads of Marko
Kraljevic, who was a contemporary of the battle. M. Braun 1
believes that one ballad report was heard by the Russian pilgrim
Ignatii in that very year. It is, at any rate, virtually certain that the
substance of such ballads was rendered into Italian towards the end
of the fifteenth century, and also into Polish; while there are bal¬
lad traces in Constantine the Philosopher’s life of Despot Stepan
Lazarevic, composed in 1431“2. There is no evidence of 'junacke
pesme before the great disaster. Vv omen’s songs are a century
older, but they also had a definite beginning. As for the neigh¬
bouring countries, existing Bulgarian ballads are younger than the
Serbian, and Rumanian historical ballads began under Serbian
influence in the sixteenth century.

Russian ‘byliny’ are more difficult to date. The Thidrekssaga
acknowledges ‘Ilias jarl af Greka’ among well-known epic heroes
of the thirteenth century, though the reference is not such as to
exclude the chance that his name appeared in poems unlike the
extant byliny’. Il’ja of Murom is unhistoric. There are various
suggested prototypes of Dobrynja Nikitic and Aljosa Popovic, and
Vk Pj ausibIe perished at the rout on the Kalka in 1224.

the ballad Vladimir combines the characteristics of the tenth- and
deventh-century princes of that name, but the wife assigned to
him (Apraksia) was one of Batu’s victims in 1237. In the Novgorod
cycle, Vasiln Buslaevic was, doubtless, the governor Vaska Bus-
“Y- mentl0 f 1 d m chronicIe for 1171. The historical assur-

ooo vearTTh 7 — St f tement “ the baIIad that his fa ther lived
900 years. The principal names mentioned in Russian ‘bylmv’ are

us historic, but their adventures are wholly fantastical,^nd suit

. Braun, Kosovo: die Schlacht auf dem Amselfelde , Leipzig, i 937 .

KINDS AND DATES 65

the fourteenth and fifteenth centuries far better than the tenth.
The unreality of Vladimir’s court offers a sharp contrast with the
genealogical exactitude of the twelfth-century Igor's Raid . What
we have are simply names and hints of character. These might be
obtained from the epinicia (velicanija) which we know were im¬
provised in old Russia after successful expeditions. Round these
names wholly fabulous adventures were woven. The first event
recorded, even indistinctly, is the battle of Kalka in 1224; hut there
are ballads which cause Ermak, the conqueror of Siberia (1582), to
take part in that battle. An additional cause of confusion is that the
entire Kiev cycle has been transferred to the Great Russian area,
while in the Ukraine itself much more modern types of song pre¬
vail. In Great Russia we encounter a veritable historical ballad for
the first time in The Princes of Tver (1327), and after that the series
run on unbroken. In sum, the ‘byliny’ reach back to the thirteenth
century in all probability, and some poem on IFja of Murom may
have arisen in the twelfth, but a greater antiquity for the genre
seems unlikely. Their apogee is marked by the cycles of Ivan the
Terrible and Boris Godunov in the sixteenth century.

The historical ballads of central and western Europe give dates
of a more decisive character, both because the events are better
known, and because subsidiary information from monuments and
chronicles is more abundant. The oldest ballad dates are provided
by Denmark. The ballad of the Battle of Lena takes us to the
year 1208; it is a straightforward report on the event, without
romanesque accretions. There are older ballads which are not so
free from retouching: Erik Errnris Murder (1137), Sir Stig Hvide's
death (1151) and previous marriage, Svend Grade or the battle of
Graahede (1157), Tovelille (Valdemar I’s reign), Esbern Snare
(Valdemars I and II): shortly after the thirteenth century opens
there come the ballads of Queens Dagmar and Bengerd, the good
queen and the bad, expressing the passions of the day. The evi¬
dence is overwhelming that ‘viser’ of the present type existed in
1200 and probably in 1150. On the other hand the history of Saxo
Grammaticus, which draws heavily on poetical texts, uses no
ballads. His information comes from late heroic poems in the
Germanic metres, such as the Bjarkamal ; his Haghard and Signe
is a much more circumstantial text than the ballads we now have.
Saxo shows us, within living memory, a literary epoch when there
were no ballads—assonating poems with refrains—but only the

4615 -rr

66 KINDS AND DATES

older heroic alliterative poetry. He also provides evidence of
changing fashions. ‘Saxon poets’ were in favour at court, and
among their subjects were at least two that invite comparison
with the ballads. One singer gave a warning to Knut Lavard in
1131, before the battle of Haraldssted, by chanting ‘notissimam
Grimildae erga fratres perfidiam’, which is the subject of the
Danish Grimhild’s Revenge ; and in 1157, before the battle of
Graahede, another improvised an invective against King Svend on
behalf of Valdemar Sejr: ‘cantor Germanicus fugam Svenonis
exiliumque cantilena complexus, varias et contumelias, formatis in
carmen conviciis, objectabat.’ This is not the same treatment as
found in the ballad of Svend Grade, but it is evidence showing that
political subjects were being treated in a new manner, a manner
which was coming in from the south. Knut Lavard is said to have
preferred this style: ‘Canutum Saxonici et ritus et nominis aman-
tissimum scisset.’ In the other Scandinavian countries the evi¬
dence as to age is less conclusive. The Battle of Lena concerns the
Swedes and circulates among them, but the first exclusively
Swedish ballad is King Byrge and his Brothers, referring to their
murder in 1317 by the king. Norse themes include King Haakon's
Death (1263) and Lady Margaret, the ‘Maid of Norway’, which is
preserved in the Faeroe Islands after a lost Norse original. Thus
Swedish and Norse balladry appears younger than the Danish,
though of respectable antiquity. Faeroese and Icelandic ballads
manifestly depend on those of the homelands.

Estimating the age of English and Scottish balladry is much more
difficult. The south of England enjoyed a culture of the French
sort among the upper classes, and as in France, the doubt con¬
stantly arises whether a given piece is genuinely in the traditional
manner. Scottish ballads, which have often the closest associations
with those of Norway and Denmark, may often have come from
thence. Binnorie is a case in point. It is not possible to affirm, as
the evidence runs, that our ballads are due to a previous Scandi¬
navian impulse (they are strongest and truest in the Danelaw), or
that they are older, or that they are independent. We cannot
separate absolutely the English and Scottish contributions to the
joint stock, nor yet treat them as entirely similar. The cardinal docu¬
ment as to the date of our ballads is a piece of external evidence. 1

1 S. Grundtvig, Danmarks Gamle Folkeviser, Copenhagen, 13: Ravengaard

og Memering*

KINDS AND DATES 67

William of Malmesbury states definitely that a poem about
Canute’s daughter Gunhild, falsely accused before her husband
the Emperor Henry III, and unexpectedly delivered, was ‘nostris
adhuc in triviis cantitata’ (c. 1140). Brompton (c. 1350) names her
accuser and defender, Roddyngar and Mimicon; Matthew of
Westminster gives us Mimecan. There is no doubt that these
references are to a poem of traditional nature and of content
identical with the ballad of Sir Aidingar. The poem was either the
ballad itself, or some very similar piece in another style which we
gratuitously hypothetize. The early rise of the ballad is assured by
its spread to Denmark and over the whole north. On the other
hand, if we allow the existence of this definitely English ballad
in the mid-twelfth century, we are not bound to go farther and
associate its rise with the reign of Canute in the tenth. The narra-
tive is probably unhistorical. It is an inversion of the Joseph and
Zuleikha story, and had been fitted to St. Cunegunde and other
women, the ordeal being by battle or by fire. In the precise form
taken by the ballad, one notices that very many derivatives con¬
sider the credulous king to be a German emperor, and the names
Gunhild, Roddingar, and Mimecan prevail over all the north.
Elsewhere Mimecan gives place to some hero of the ballad-poet’s
own choosing, such as, in Catalonia, Count Ramon Berenguer IV.
In Catalonia the injured lady is still stated to be Empress of
Germany.

Apart from this ancient romance about an historical figure, the
dates show a certain priority of English over Scottish ballads,
though the Scottish ones have a more authentic ballad style. To
the twelfth century, in theme, pertain Queen Eleanor's Confession
and Robin Hood. The former ballad was rather literary in style;
those on Robin Hood were gradually accumulated between his
supposed date (1198) and the sixteenth century. The long Geste
arose by the union of four ballads before 1400. The ideal date of
Sir Hugh or the Jew's Daughter is 1255, but it is a pious legend and
may have been committed to verse considerably later; Judas is also
a pious piece, committed to writing in the thirteenth century. At
the end of that century we find the Scottish Sir Patrick Spens , if it
refers, as opinion generally concedes, to the death of the c Maid of
Norway’ in 1290. The Norse ballad gives a quite different account
of the same affair, since the poet was an adherent of the pretender
Margaret, who was burnt at the stake a few years later. Some of

68 KINDS AND DATES

the earlier fourteenth-century pieces may be regarded as not wholly
traditional work, but there is no question of the authenticity of
Chevy Chase and Otterburn (1388, earliest versions composed
about 1400), nor of the Aberdeenshire Harlaw (14x1). There is a
steady sequence of ballads from the fifteenth and sixteenth cen¬
turies, in the traditional style on the Scottish side, but with more
ostentatious minstrelsy on the English.

In Germany the attempt to attribute very high antiquity to
ballad compositions encounters similar difficulties, and it is not
until the fourteenth century that poems identifiable as ballads
appear. The Ludwigslied (881) has been considered ballad-like.
There are some phrases in it which have a heroic ring, like that
of the traditional epos; but the sentiments are definitely clerical
and the transmission is by manuscript. The old Hildebrandslied is
an epic fragment, possibly composed as a fragment, but no more
like a ballad than Finnsburg or Waldere. The ballad Hildebrands¬
lied is different in style and content, and probably as late as the
fourteenth century in date, if not later still. More relevant to ballad
origins are the political poems in Latin composed by Saxon poets.
With the Latin they mixed vernacular words, as in the Ottolied
(941), and we have seen that they exercised their trade in Denmark
in the twelfth century, enjoying royal favour. Though this kind of
work was not traditional in style, it could readily take on a more
popular manner. They also bridged the transition from alliterative
to assonating verse. The authentic ring is first heard in Lippold
von Homboken (13x1) and the Churls in Flanders (1323-9), both in
Low German territory. After the middle of the fourteenth century
was passed the historical ballads come thickly upon us: the Swiss
battles (Sempach 1386, Nafels 1388, Birs 1444, Granson 1476,
Murten 1476, Nancy 1477, Bruder Veit 1515, leading on to the
landsknecht ballads of Pavia 1525 and after), German feuds (Busso
vcmErxleben 1372), robber barons (Epple von Geilingentfi, Linden-
schmid 1490), and sea-rovers (Stortebeker 1402).

In the Romance area we find that Piedmontese and Provenqal
popular songs rely on a French technique which is so lyrical that
it is hard to date. A revolution in musical taste towards the year
I 5 °° I to the gradual abandonment of the more complex and
emphatic styles, and corresponded to a simplification of the lyrical

0 1 'flu Garold, Chansons populates des XV‘ et XVI‘ slides avec leurs melodies,
Strasbourg, n.d.» p. vi.

KINDS AND DATES - 69

style which operated from about the middle of the fifteenth century.
With this encouragement men of letters and music took to record¬
ing songs which may have had a considerable previous life in
tradition. The absence of historical pieces, save a small number,
prevents our guessing how long that life may have been. The most
striking piece is The hanged Scholars which goes back to an event
at Pontoise in 1259. It is the source of the English ballad The two
Clerk's sons of Oxenford, and of a Catalan ballad also. But suc^h a
date is exceptional, and we should have to allow that transmission
of the necessary knowledge might have occurred otherwise than by
way of traditional verse. Some French adventure pieces are neces¬
sarily anterior to the Spanish ones based on them, and other
Spanish adventure poems imply lost French originals at least as
old as the fifteenth century. What really causes the difficulty in
estimating the age of French 'chansons populaires’ is the extra¬
ordinarily rich and varied artistic life of medieval France, so
markedly superior to all culture but that of the Italian city-states.
Very few French pieces are untouched by artistic fashions, and on
the other hand the greater part of French literature—'chansons de
geste J , 'chansons de toile’, romances, legends—was familiar in one
form or another to natives of other lands, and so exerted an in¬
fluence on their ballads. We have to deal with the effects of Ogier
de Dinamarche in Spain and Denmark, of Tristan in Spain, Iceland,
and Germany, and of the courtly 'pastourelle , everywhere. On the
other hand, one cannot say definitely that much of the twelfth- and
thirteenth-century matter in Bartsch’s Romanzen und Pastourellen
is traditional in the narrower sense, or that any of it is irrelevant to
the study of traditional poetry in France or abroad.

As for Italy, outside the Franco-Italian area we find a poetry
based on the 'ottava* and so lyrical as not to admit of sure dates.
Those who trace Sicilian folk-song back to the twelfth century rely
on allusions to the Norman kings which any person who had access
to the monuments at Palermo could have made. One needs a
narrative of some complexity before one can hazard a date of com¬
position. The oldest such piece in Sicily is the Princess of Carini>
a story of the year 1565.

In the Spanish peninsula the priority of the Castilian ballads
over all others is indubitable. Gil Vicente (d. 1536 ?) and Camoes
(d, 1580), in Portugal, quote their 'romances’ in Castilian, showing
that their entry into Portugal belongs to that century; the same

7 0 KINDS AND DATES

century doubtless witnessed the invasion of Catalonia by pieces
different in style from the older Franco-Provenpal popular songs.
They are sometimes in Castilian only, but more often in a mixture
of Castilian and Catalan. The Castilian historical romances ,
defined as above, take us back to the Civil Wars which established
the House of Trastamara in the second half of the fourteenth cen¬
tury. One of them has preserved an ancient slander, like a fly in
amber, and either two or three were used by the chancellor Pero
Lopez de Ayala in his chronicles before 1394 * 1 frontier

ballads present a continuous sequence from the year 1407? the
oldest known adventure ballad can be encountered in a mixed
Castilian and Catalan form about the year 1421. 2 The middle
years of the fourteenth century are also the time indicated by a
comparison of the traditional epics and the epical ballads.

In the Baltic regions it is possible to make out a case for the
antiquity of Finnish ballads on the basis of one piece with a twelfth-
century subject, and the use of mythological and magical motifs.
I am unable to estimate the strength of such evidence, since an
ecclesiastical legend does not always imply continuous popular
verse traditions, and the use of pagan motifs depends on the dura¬
tion of practising or virtual paganism among the Finnish country
folk. In Lithuania the mythology may easily be quite modern.
The songs are generally too lyrical to admit of dates, but there is
one memory of the ravages and cruelties of the Teutonic Knights,
which carries us back to the fourteenth or fifteenth centuries.
Soldiers’ songs contain historical allusions from 1700 onwards.
The practice of singing may, of course, be very much older than
these dates, and the word ‘daina’ means merely ‘song’; but ‘dainos’
of the present form and content seem unlikely to go farther back
than the later Middle Ages. The form, I believe, is an import from
Germany, with Lithuanian modifications.

Some of the ballad novelettes also indicate dates. Tannhauser is
not older than Tannhauser himself; indeed, the motifs belong to
the fifteenth century. The noble Moringer uses the name of the
twelfth-century Heinrich von Morungen. The two pieces imply
that the aubade’ or ‘Tageslied’ had already descended from courtly

1 See W. J. Entwistle, ‘Romancero del Rey don Pedro 5 , Modern Language
Review, xxv, 1930. s s

* ?b LeV > * E1 romance Aorentino de Jaume de Olesa 5 , Revista de Filologia
■tspanola, xiv, 1927. &

KINDS AND DATES 71

literature to traditional ballads. The Rumanian Monastery of
Arges uses a name for the master-builder which belongs to a Greek
engineer of the seventeenth century. The Castle in Austria is of
fifteenth-century date in its original home, and so must be younger
elsewhere. Many other cases of the kind could be cited, but these
should suffice.

To sum up, the evidence of the historical ballads everywhere
indicates that ballad-poetry as we know it had a definite date of
commencement; before that date there were songs and epics and
other sorts of literature, but not our ballads. In France this date
cannot be conveniently given because of the special situation of
medieval French literature. In Spain the mid-fourteenth century
is indicated. In Denmark it is the mid-twelfth century, in England
either then or later, in Germany the early fourteenth century, in
Czechoslovakia the seventeenth or sixteenth, in Lithuania the fif¬
teenth. In the east there is a marked priority of the Anatolian
cycles which may go back towards the tenth century. Serbian
ballads arise in the later fourteenth or early fifteenth centuries, and
Bulgarian and Rumanian later. Russian ‘byliny’ can hardly be
older, in any form recognizable as the modern, than the early
thirteenth century; the fourteenth is a more likely date for the
older pieces. Comparison with the traditional epics, when directly
or indirectly available, shows that there was in each country a time
when ballads were not. The diffusion of certain important ballads
also follows a course laid across comparatively recent centuries.
Ballad literature arises at definite dates, and sets, as we have seen,
when the practice of entertainment through reading becomes
general among the better classes. This spread has occurred at
different times, according as the country is one of the progressive
group or not, but the whole span of time between the rise and
decline of the genre can reasonably be called medieval in each land.

Isolated motifs may be Timeless, nameless 5 . For single ballads
it may not be possible to give a date; but they are always dated in
general terms by the limits assigned to the genre in any country,
and they are usually to be dated more closely by comparison with
others of the same class and characteristics. In the second half of
the fifteenth century, and in the sixteenth, almost all the peoples of
Europe were able to enjoy the finest pieces of their repertoires.

HOW BALLADS SPREAD

r-pHE traditional style employs only the tutt

1 and the most A Live estperi.nce,

rnrifthemiginarpoem has no.

ballad will be oeeumentcl ms^‘> r5a ,i ty has led to the

-leto—erLions h tire peoples:

b u f nlhe Stimmen der Volker, as Herder’s editor put it. There
they ar . ris£ of ^ a p a ds so mystical that they cannot be

pursuedby me; theories which dispense with personal authorship,
and^nvisage some undefined sort of spontaneous generation among
the folk Andrew Lang, without denying the fact of personal
authorship, denies its relevance to the criticism of a manner and
motifs which are universal . 1

No one any longer (he wrote) attributes them to this or that author or
to this or that date. The attempt to find date or author for a genuine
popular song is as futile as a similar search m the case of a Mai chen. It
fs to be asked, then, whether what is confessed y true of folk-tales-of
such stories as the Sleeping Beauty and Cinderella- is true a W»-
songs. Are they, or have they been, as universally sung as the fairy¬
tales have been narrated? Do they, too, bear traces of the survival of
primitive creeds and primitive forms of consciousness and of imagina¬
tion? Are they, like Marchen, for the most part, little influenced by the
higher religions, Christian or polytheistic? Do they turn, as the
Marchen do, on the same incidents, repeat the same stones, employ the

same machinery of talking birds and beasts? Lastly, are any specimens

of ballad literature capable of being traced back to extreme antiquity?
It appears that all these questions may be answered in the affirmative;
that the great age and universal diffusion of the ballad may be proved;
and that its birth, from the lips and heart of the people, may be con¬
trasted with the origin of an artistic poetry in the demand of an aris¬
tocracy for a separate epic literature destined to be its own possession,
and to be the first development of a poetry of personality a record of
individual passions and emotions. After bringing forward examples of

1 A. Lang, ‘Ballads’, Encyclopaedia gritannica, nth edition.

HOW BALLADS SPREAD 73

th e identity of features in European ballad poetry, we shall proceed to
show that the earlier genre of ballads with refrain sprang from the same
primitive custom of dance, accompanied by improvised song, which
still exists in Greece and Russia, and even in the valleys of the Pyrenees.

The questions might all be answered in the affirmative (though
Perrault s Cinderella is perhaps not the happiest instance of time¬
lessness), and not lead to the conclusion Lang announced. The
answers would refer to the isolated features of ballad poetry and
traditional literature, features which are to be found also in the
epical works he deemed aristocratic and personal; but the ballads
are not isolated features, but complexes of a definite type. It is at
this point we must remember that the ballad, for the duration of
this book, is a narrative song. Pure lyrics do not offer us a web of
circumstance such that we can recognize repetitions of the same
pattern without great difficulty. Their motifs are universal and
simple, and they are presented without elaboration. There is still
no reason to deny that each such song arose at a particular date
under the hand of a particular author; the fact that we do not know
either date or author does not imply that neither existed. Indeed,
the lyrics may be, though we do not know it, quite young; ever¬
lasting reincarnations of the same simple notions. From the pure
lyric it is not possible to draw conclusions; but the narrative ballad
is quite different. We have already noticed that there exists a vast
mass of testimony as to the date of the genre in different European
countries, and the periods within which some particular pieces
have arisen. It is because these narratives are complexes of motifs
that this dating is possible; the motifs may be universal, but the
union established between them is temporal and distinctive. The
form also is generally defined with precision; one cannot mistake
the form of a ‘romance’, ‘vise’, ‘junacka pesma\ or ‘bylina’. This
form has its historical relation to other forms before and after, and
itself implies a chronology, though we may not know it. The
grouping of the motifs is too close to be explained by fortuitous
coincidence, and not infrequently gives some indication of the
circumstances attending on the original composition or the course
of migration.

Within any given area there is never a doubt but that two ballads
of identical content are forms of the same ballad, and not separate
spontaneous creations. The Three Ravens and the Twa Corbies are

forms of the same ballad tradition, though the one be English and

461s T

HOW BALLADS SPREAD
1 othei Scottish; the sa™ is •»

and the Scottish Sr Hush lore than one

complete list shows me 181 viser num ber of ‘viser’ is about
of the Scandinavian countries, re k ,v L j le ballads

600. In the German area it is not feasible «° ““ the Flemish

of High Germany from Low er, , ri: | ' ' , g ; s f rom the rest,

and Dutch ballads from the mam bod, or the^w. ^ ^

All that one can do is « ~Tw s s tr gi“ and tit other, have
marks of Flemish, German, or Swi g Serbian

continued to be sung within a nanow The 7

coincidental ballads, in w nf Francois I’s im-

Franco-Italian balladry; it is the same baUadofi
prisonment that circulates in France and

Porcheronne Provengal Porcheireto, and Catalan Noble roiquera
^STUriants of the same thing. Portuguese and Castilian
ballads are found to agree, and Castilian and Catalan. Wherever
there is no difficult frontier of language or culture to surmount
traditional ballads are able to travel from mouth to mouth withouL
impediment. The dialect used for their performance takes on
slowly new characteristics as the song passes over the ground, un i
it may reach the limits of the linguistic area. It is not translation
but substitution that occurs. The substitution may be left in¬
complete when the original is sufficiently understood One may
see this state of language in the Castilian-Catalan ballads, or in the
Faeroese Nykkurs visa , several verses of which are in the origina
Banish. Borrowing, not spontaneous creation, in such cases is the

The words of ballads are, in fact, complexes of motifs just as
their tunes are complexes of notes. The chances against fortuitous
coincidence in the words are scarcely less heavy than against for¬
tuitous coincidence in melody. These two aspects of the full per¬
formance have each their history of rise, expansion, and decline.

Tannhauser is a ballad of which the history is singularly easy to
trace, thanks to its literary connexions. The minnesinger so called
lived from about 1200 to 1268 and was a contemporary, as the
ballad asserts, of Pope Urban IV (1261—4). The connexion with
the Venusberg or Sibyl’s Paradise must have been made in the
fifteenth century, since we first hear of the Paradise from Antoine

HOW BALLADS SPREAD 75

de la Salle about 1420. An apocryphal poem detailing a conversa¬
tion between the poet and Venus has served to cement this
association; the poem is recorded in a manuscript of the year 1453.
Therefore, we can affirm that the ballad was made by a High
German in the fifteenth century, and (since the opening is in the
style of an ‘aubade’) one who was familiar with the processes of the
courtly lyric. There is external evidence of its passage into Low
German territory in the eighties of that century. Danhuser leads
to the Flemish Herr Daniel on the one hand, and to the Danyser
which appears in Denmark in 1684. The reasons for believing
The Castle in Austria to be a fifteenth-century piece are of a more
general nature, but equally satisfying. It is one of many ballads on
the theme of the guiltless prisoner, who has set his love too high
and must suffer for it, and it goes back to a common source, in all
probability, in which the prisoner was a poor student of the Uni¬
versity of Paris. However, it represents a new start and has a
history of its own. The ballad is easily recognizable by its first
lines :

There lies a castle in Austria
that is so nobly founded,
with silver and the red, red gold
and marble stone surrounded.

The verse is recognizable even when, for devout reasons, the
Eastern realm (Oosterrijk) is allegorically understood. Starting in
High Germany the ballad descends to Low Germany and spreads
westwards into the Netherlands and northwards into Denmark and
Sweden. The oldest German tune belongs to the year 1480 and is
unconnected with the rest (4D412). What is called the ‘old wise’
is a tune recorded from the middle years of the sixteenth century,
first in the Low Countries. In 1540 its intervals are 73 G, but
there are many variations, and from Rostock Erk and Bohme
reported one in B212C which is virtually that of the Danish ballad
(B25BC). 1 The ballad entered Denmark from the Low German
territory, possibly from Mecklenburg; the date must lie between

1 Souterliedekens, 1540. Rostock Danish

73G B212C B25BC

1 flat U4 1 flat 32 U4 2 flats 68 U4

d:dda.a:c.cf.r a./a. .ag./a. .ab./c.. .c./a... a/a(.a)ga. djd. ca (.a)

g:at.aa:ga.r c.jc. .cc./a. .gf./g./a. c/b.ag.f/ga...

b:cbagfc:a.ad.r J/*f.ca.b/a.g£(.f)

a:gdfg:eed.r a/c.ba(.g)f/ed...

„ 6 HOW BALLADS SPREAD

ihe first appearance of this tune in 1540 and the printed Danish
ballad of 1697, and probably nearer to this later date since Syv did
not admit the piece into his collection (169s). The Swedish ballad
also uses a melody current in the Low Countries and north

There'are a number of other pieces which could be cited as illus¬
trating the spread of German themes into Scandinavian balladry,
the Count harnessed to the Plough, the Heathen Princess (or Sultan's
Daughter) are among the more obvious. Similar expansion into
Slavic territory can be studied in The Shroud (German Per Vor-
wirt), which is found in Lusatia and Czechoslovakia also. It is a
macabre ballad: a dead husband rises from his grave to reproach
his widow for her lightheartedness. The Slavic ballads have been
collected only in the nineteenth century, and so it is scarcely pos¬
sible to give a date for the borrowing; but the fact is confirmed by
the resemblance of some of the tunes. The basis seems to be a
formula 22BB. Der Vorwirt has 522iABB, with a different initial
interval and an inserted iA; the Czech Rubas has 4BBA, 4BB21,
and D22BBAB as three of its five tunes; and the Lusatian Kitel has
43E2BB, in which 43 E is equivalent to 2. There has been con¬
siderable modification, but the resemblance still dimly appears, and
a full transcription would show that some parts of the melodies
are still fairly close.

These are cases in which ballads have crossed a frontier, consti¬
tuted in Denmark rather by ballad tradition than by language, and
in the Slavic countries by language rather than tradition. In their
commerce with Scotland and England the Scandinavian ballads
have had to cross the sea, though they are welcomed on arrival as
pieces in our own manner. We do not dispose of studies so ade¬
quate in this respect as those which have been available for us in
establishing Germany’s relations with her neighbours. The texts

declare their intimate relationship, but have been much changed,
and the melodies are hard to identify. Ribholt and Guldhorg is a good
example. It is essentially a story of taboo. By naming Ribbolt’s
name, Guldborg leads to his death in battle against her seven
brothers. The plot is encountered in the Eddie HelgakviSa II,
and the superiority of the Danish version justifies the belief that
the ‘vise’ arose in Denmark. It has covered all the North: Swedish
Hillebrand or Redemld, Norse Rikeball and Veneros, Icelandic
Ribbalds kvceSi. In Scotland it is called Earl Brand (doubtless a

HOW BALLADS SPREAD 77

variant of Hildebrand) and The Douglas Tragedy, and it is living
in the Appalachians under the titles The Seven Sleepers, The Seven
Brothers, or Sweet William. In a previous chapter some mention
has been made of the likeness and unlikeness of the tunes. The
Two Sisters, Edward, Clerk Saunders, and Hind Etin are other
examples of ballads essentially the same in their Scottish or Scan¬
dinavian forms. Edward is probably Scottish in origin. It is
sung m the Appalachians to two tunes of the same class. The
words have travelled from Sweden to Finland, but the Swedish
and Finnish melodies are different.

In order not to extend this discussion too much, it may suffice to
say that the same kind of migrations occur elsewhere. It is not
motifs that travel, but ballads as units, often accompanied by their
tunes; and so long as the movement is limited and recent, the
evidence of identity is incontestable. The Lament of the Robber’s
Bride is the same thing in the Balkans, the Ukraine, and Czecho¬
slovakia; the Rumanian Marcu and the Turk is a translation of a
Serbian or Bulgarian original; Roman and Olesa is found in Poland
(or at least, Polish Galicia) and the Ukraine; the Murderous Wife
is sung in Czechoslovakia and Poland. The gipsies of southern
Hungary sing ballads of German and Greek origin among others
inspired by the conditions of their own life. Anrus and Death is
their variant of the Greek Digenis and Charon. There is movement
of ballad literature all over Europe, just like the movement of
books. The transmission is from mouth to mouth and not hand
to hand. The linguistic obstacle is overcome by the shading of
dialects into each other, thus effecting a gradual change of language;
and as against ignorance and want of curiosity about the produc¬
tions of other people, oral transmission finds compensation in the
sameness of the ballad style.

The movements which we have considered hitherto have been
of a simple and restricted sort, but there are others of ampler range
which have constituted a veritable commonwealth of ballads.
These great international cycles are, of course, much more subject
to variation than those of narrower orbit; it is much more difficult
to locate their centres; yet the principle is the same; the ballad
rises at some definite place and time and follows definite routes in
its expansion, over Europe. By considering such pieces we get
some impression of the energy of invention which has been shown
by the greater ballad peoples. A maD of Eurone rnnlH Kp +r\

7 g HOW BALLADS SPREAD

show the migration of ballads, and it would be criss-crossed by
lines in every direction; but there would be some emphatic foci:
France, Germany, Denmark, Greece, Serbia. Peripheral nations
export less than those at the centre of the Continent, but even at
the centre there are peoples who more often borrow than lend.

Owing to the nature of its ‘romancero 5 Spain has comparatively
little to put into the common stock. The predominantly historical
trend, admirable as it is as an education for the national conscience,
is not suited to carry the ballads beyond the limits within which
this sort of historical information is interesting. The Spanish
'romances 5 , therefore, circulate among peoples of Hispanic stock
and language. An exceptional case is provided by the ballad of
GuarinoSy in the Carolingian series, which chanced to begin.

An evil fate befell you, Frenchmen,
at the Chase of Roncesvaux;

Charlemagne there lost his honour,
died the dozen Peers also.

The ballad was translated by Karamzin into Russian in 1789, and
in 1834 Baron Erdman encountered it as far afield as Siberia. It
has the honour of being the first literary ballad in the Russian
tongue.

Catalonia, Provence, France, and north Italy form one area with
a common store of ‘chansons populaires 5 , and it is not easy to
assign a particular piece to one division rather than to another. The
Samaritan Woman has been reckoned to the credit of Catalonia in
particular. The woman of Samaria became confused with the
Magdalene, and her meeting with Christ led to her repentance,
and works of penitence. Passing through France as The Penance
of the Magdalene, the omission of the proper identifications led to
the English Maid and Palmer , which has no obvious biblical con¬
nexion. A ballad on the Magdalene is found in Denmark and
Sweden also. Crossing Germany the ballad tended to paint the
sinner’s life in ever blacker colours; she was guilty not merely of
carnal sins, but of wholesale infanticide to hide her guilt. This is
the legend as it appears in Aria the infanticide in Lusatia, the
Sinner (. Hrimice ) in Czechoslovakia, and in Poland. The un¬
pardonable sinner is swallowed by the floor of the church, un¬
shriven. From Provence spreads the ballad called La Escriveta.
The story is older, since it has been plausibly connected with the
epic of Walther of Aquitaine, who escaped with his bride from

HOW BALLADS SPREAD 79

Attila’s capital. The pattern was used for later verse-novels, and
must have adopted a form in which the hero’s name was given as
Gaiffier. This stage of the tradition is represented by the Castilian
‘Carolingian 5 ballads of Gaiferos, ‘who quite forgot his lady free’,
and the escape of Moriana or Julianesa from a Moorish palace. A
more anonymous form is the Provencal Escriveta, which extended
into northern France and Italy (II Moro saracino).

France has supplied to balladry a great deal more than individual
themes. The ‘belle maumariee’ and ‘bergere’, the ‘aubade’ and the
songs of artisans and soldiers, are French patterns which have been
used elsewhere to cover more or less original extemporizations; and
the influence of French ‘chansons de geste’ extends from Portugal
to Sweden. The narrative which Thackeray has popularized as
Little Billee seems to spring from the seafaring population of
Brittany and Normandy. There is a Breton ballad, The Sailors,
on this theme. Extending to the south-west, it becomes the Cabin-
boy of Provence and Catalonia (Lou Moussi , El Grumete, also La
Tortolita), and then the Nau Catarineta of Portugal. The name is
that of a famous vessel which served, among other things, to carry
Princess Beatrice to Savoy in 1519. To the north-east the ballad
spread among maritime folk to become the Seafarers of Denmark,
Sir Peter's Voyage in Sweden, and the Merchant's Tale in Iceland.
The theme is one to attract attention among seamen and travellers,
and is in fact confined to the seaboard of western Europe. There
is no such limitation on another theme which I think should be i
assigned to France, namely, The Warrior Maiden. This maiden
has been known in China also, where it is said that a certain Mu-lan
performed her father’s military duties as an example of filial piety.
Whether that has any connexion with the European ballad, it is not
needful to inquire in a book restricted to European balladry. In
Europe the story has a definite nucleus which persists in all its
varieties, and there is in most of them a tendency to stress the
aspect of sex, which is characteristically French. The point is not
merely that a girl should be a distinguished soldier, but that her
unmasculine beauty should arouse some youth to set her traps
which she skilfully evades. I should like also to think that the
situation owes something to the exploits of Joan of Arc; there
seems no reason to consider the ballad older than the fifteenth
century. If it did not arise in France, the lines of migration suggest
northern Italy as its home. The Portuguese Dom Martinhos,

8o HOW BALLADS SPREAD

Castilian Marquitos, and Catalan Joven Guerrera imply a French
original, extant in the sixteenth century. In the Messm Puymaigre
encountered La brave Claudine. In Italy the ballad appears as La
Guerriera, and crossing the Adriatic on Venetian vessels it reached
Dalmatia and Serbia as Zlatija. The tests of her sex became at the
same time of a more improbable sort, though they continued to
include the essential test of stripping to bathe. Breaking down
into a popular tale, the story became the Silvertooth of the Al¬
banians, the Basilopoula of the Epirot Greeks (who contaminated
with it the folk-tale of the wife who would not speak), and at last
reached Wallachia (Mizil Crai). From France or Italy the ballad
reached Czechoslovakia in the Czech Bojovnice, where the second
half was lost. There is a White Russian ballad verbally connected
with the Czech, and at last this line comes to a stop in the Devuska-
voin of the Ukraine. In this last shape there is a new complication,
taken from another ballad, since the girl is advised not to fight in
front of or behind the main battle, and her neglect of good advice
involves her in further trials.

Donna Lombarda is the most famous of north Italian ballads.
The story how Rosmunda poisoned her faithless paramour Hel-
michis is as old as the Historia Langobardorum of Paul the Deacon.
This history was one of the most popular in the Middle Ages;
there is no need to ascribe very great antiquity to the ballad itself.
Its simple structure, however, shows that it is in the first flight of
the ‘canti popolari’, and this is naturally confirmed by the music.
In France the poem has the same form and content, but in Spain
and Portugal the anti-heroine’s name has been forgotten and
replaced by Moriana, Julianesa, or Ausenia. Mariana is one of the
few ‘romances’ for which we have an older form than the one in
tirades; it exists also as a lyric, and that brings it closer to the
Franco-Italian style. In the other direction, towards the east, this
tale spread and was generalized to cover a variety of poisonings. A
convenient title is the German Schlangenkochin, since, apart from
the tense dialogue proper to the whole group, these derivative
pieces generally describe a menu of fish and snakes, or eels. They
vary according as the lady friend be the poisoner, or the suitor’s
or the fiancee’s mother, or a sister wishing to get her own way.
Thus we have the different varieties of the German Schlangen-
kochin, the Danish Poisoning (Giftblanding ), the Czech Fiancee as
poisoner and Sister as poisoner, the Lusatian Hindrask and Evil

HOW BALLADS SPREAD 81

Mother , Hungarian Gipsy Wicked Mother-in-law , Serbian Stojan
and his Mother, Russian Two Lovers, English Lord Randal, and
Italian Poisoned marts will {Testamenio deW Avvelenato). This last
is the more curious since, though the cycle is indigenous to Italy,
this piece has come from England. It has the form of Lord Randal t
and its tune is a variant. 1

It seems also advisable to associate with Italy the motif of
soporific drugs and feigned death. The most interesting of such
stories is that of the girl who simulated death to escape a ravisher
or unwelcome lover. It exists in France, Italy, and Catalonia. In
Yugoslavia it is entitled Erceg Stepan (the name shows connexion
with the seaboard which was under Venetian influence). A notable
featui e of this ballad is the severity of the tests to which the pre¬
tended corpse is put; they belong more suitably to the haiduk
ballad of Little Radoica who, as a hero, is more fitted to stand the
trials of a snake in the bosom and nails driven into the quick,
though not of dancing-girls. Erceg Stepan is represented as a
Moslem creditor. In the Czech form the ballad is entitled The
Turk s Bride. Boccaccio’s use of the Imogen theme is probably
oldei than any of the ballads of the cycle of Mananson, which may
be either French or Italian. The ingenuity behind the story is
rather Italian than French. In France it is called Marianson or
Innocence proved, in north Italy The Rings . It does not appear in
the Spanish peninsula. In Scotland it is Reedisdale and Wise
William * and The Twa Knights ; and in the latter form it presents
the oddity of agreeing better with the Greek ballad of Maurianos
and his Sister than with its neighbours. (The point is that the girl
keeps her honour by sacrificing that of her handmaiden.) The
Rumanian form is called Iancea Sabiencea. An unwelcome note of
ferocity is present in the Serbian Marko Kraljevic and the Royal

See Countess Martmengo-Cesaresco, The Study of Folk Songs (Everyman’s
Library), London, n.d., pp. 171-80. There is a tune to King Henry (Child 12)
wlneh closely resembles one gathered in Lombardy by Cocchiara:

King Henry (521ABE4BB)
2 flats 68 U8
dga(b.. a)g/gdfe. d/
dga(b, -a)g/gddg.

(i bc)l(dd)edca(2.b )/baga.
(dd)/dga(f>. .a)b/dgfg../

UAvvelenato (521 A)

2 flats 68 U8
d/g-gg-a/ba gg.a/bagbag/a.r

c ld- aa.b/b./b.ar (1st hemistich repeated)

dig * gg • a/ba gg. a/bagbag/a. r
d/e . aabc/c.. b (3rd hemistich repeated)

. a/g.gb.a/g.r (refrain)

The chief difference is in the arrangement of the lines, in which the Italian
version stands closer to what is usual in the Lord Randal series. There is also a
Donna Lombardo, melody used in Romagna with the melodic contour 521C,

g2 HOW BALLADS SPREAD

DeUbaAa, since it is not rings ttot are

;l„g husband unwisely boasts of his wif.-is found also .n tire

W l taXer in the reign of Emperor Frederic II, Nicholas by
name bst his life under circumstances that have been preserved by
Slkd traditions of a much later date. Their immedtate source ,s
“Sy but the ballad of Cola Peace has enjoyed populartty m a
number of countries, since it lends itself toadaptat,on. Theessenoe
is that some one is dared to dive or swim, and usually drowns.
Most ballads make this a test of the lover. It is known in Brittany
and France and in all Italy. The Greek Diver and half a dozen
Lithuanian ballads correspond to this theme the latter having
the German Wager for a go-between. In Serbia there is a ballad
of a girl who will choose the man that dares to swim twice over a
dangerous river, and the Rumanians tell the same story of a certain
Rada, beside the Pruth. But perhaps these pieces are too distantly

related to be ascribed to the one cycle.

The most notable English ballad to have spread to other lands
is Sir Aldingar , which has already been mentioned. It is difficult
to know whether some pieces in our repertoire are of island or
continental origin, not to mention the always obscure speculation
as to the influence the Celtic peoples of the fringe may have exer¬
cised on the Northmen’s imagination. The elfin music or our
balladry is especially poignant, and if the same note sounds in
Denmark, Sweden, and Norway we may have some part in that.
Our islands stand on the circumference of Europe, and so do not
export themes so readily as the central nations. Merit is no more
the sole condition of fame in oral than in written literature, but
other considerations, material and fortuitous, intervene. The case
is demonstrated by such pieces as Tam Lin and Sir Patrick Spens,
which are unsurpassed in power and beauty throughout all Europe,
but have remained solely our own. Our kinsfolk in Denmark,
whatever they may owe to us or have given us, undoubtedly occupy
a central position in Scandinavian balladry. Elveskud is Danish;
at least as a ballad. The youthful bridegroom is compelled to enter
into a fairy dance and is blighted; he returns home to die; his
death is concealed from his bride (sometimes because of her condi¬
tion), but she hears the tolling bells, learns the truth, and dies for

HOW BALLADS SPREAD 83

sorrow. The ballad is the English Clerk Cohill and the Swedish
Sir Olof in the Elves' Dance, Icelandic Olaf Lily-rose and Faeroese
Olavur Knightly-rose. The German equivalent in traditional litera-
ture is The Knight of Straujfenberg; the Danish poem is made per¬
fect in Goethe’s Erlkonig. There are several Lusatian ballads of
The Luckless Wedding, which is also found in Czechoslovakia. The
extension south-westward was probably by sea, since we encounter
a well-preserved form of the ballad in Brittany [Count Nann). In
France the elfin opening has been lost; Le Roi Renaud appears
with a mysterious mortal wound. A sub centre was thus created in
France whence we have the Basque King John, Catalan Widow and
Don Ramon, Spanish and Portuguese Don Pedro, in a south-
westerly direction; and Count Anzolin or the Concealed Demise in
Italy, leading to the Rhodian Greek Constantine or Marriage and
Departure . It has been claimed that the head of the ballad has
been preserved in the Spanish and Catalan Gentle Lady and Rustic
Shepherd, but the claim seems too ingenious. 1

Scandinavian scholars are wont to disclaim for their region the
ballad of the Merman s Bride [Agnes and the Merman). It is found
on the shores of the North Sea, among the Low Germans, and in
Lusatiaj but there seems little reason to accept a further disclaimer
by German critics, who would assign the ballad to the Slavs.
Water-sprites among the Slavs are different from the northern
mermen and nixes; and they play a more prominent part in Scandi¬
navian than in German tradition. Assuming, despite the authori-
ties, a Danish centre of diffusion, we can run rays out to the north
(Faeroese Nykkurs visa, Icelandic Elenar IjoS), to the west (French
Belle Helene, Breton Fiancee), and to the south (German Wasser-
manns Braut, Lusatian Wodny muz, Hungarian Pabeli Antal).
Revenants are a notable feature of northern mythology, and there
are two powerful Danish ballads which use this motif: namely,
The Mother beneath the Mould and Aage and Elsa. The former is a
medieval equivalent of the Eddie Waking of Angantyr and Svip-
dagsmdl. The Svipdagsmdl has direct descendants in Scandinavian
balladry, and the connexion between these and The Mother beneath
the Mould is generic, not direct. The tears of orphans waken the

This is the appropriate place to mention the model of all comparative
ballad studies, S. Grundtvig’s classic Elveskud, dansk, svensk, norsk, feeroesk,
islandsk, skotsk, vendisk, bcemisk, tysk, fransh, italiensk , catalonsk , spansk, bre-
tonsk Folkevise, Copenhagen, 1881. Elveskud is No. 47 in his Danske gamle
Folkeviser.

g HOW BALLADS SPREAD

id ia Provence (The Orphem). In Germany the condus.cn has
been shed doubtless for good theological reasons: the motker
SiofS We are conceded chiefly with Are bitter tears o the
orphans as also in Czechoslovakia and Lnsatia. Orphans and their
Zl croSy considerable portion of the folk-songs of Ltihuama
IIS which are chiefly inspired by dome* joys nd
sorrows, of which the bereaved state is typical There aie seve al
pieces which suggest that Laime, the goddess of luck, is especia y
kind to orphans. The lack of narrative elements makes comparison
very slippery, but one of Rhesa’s pieces, though extremely simp e,
does indicate the tableau of the orphan girl weeping at her mother s

S Two ballads of literary origin, the one classical, the other biblical,
have been referred to the Netherlands : The Two King s Children
and Hallewijn. The former is the legend of Hero and Leander, but
so simplified that the classical link has snapped. From Flanders
it passed into France {Flambeau d’Amour) and northern Italy. In
Germany it divided into two types: the one is narrative {Zwei
Konigskinder), and spreads to the Czechs, Magyars, Wends, and
Poles; the other is lyrical (Zwei Wasser), and is of exceptional
charm. As for Hallewijn it is only in the Netherlandish version
that the heroine returns bearing the head of her would-be seducer.
The detail establishes firmly the connexion between this ballad and
the story of Judith and Holofernes. The Jews of Bethulia have
been forgotten, however. If the heroine goes out, it is because she
has been beguiled by the suitor Hallewijn or enchanted by his
music; and Hallewijn’s crimes are all personal, since he has killed
many a bride. As the ballad pursues its career in Germany we are
presented with alternative solutions: the heroine may be murdered
and avenged by her brothers, or she may escape by quick thinking,
making a pretext to borrow a knife wherewith to stab Hallewijn.
The latter’s name becomes Ulinger, Ulrich, Gert Olbert, &c.,
travelling always farther from the biblical Holofernes. Molndr
Anna is a Magyar derivative of the German versions. In Scandi¬
navia the ballad made contact with others which describe elfin
malevolence, and so it appears in English in Lady Isabel and the

HOW BALLADS SPREAD 85

False Knight and May Collin. The French have Renaud wife-
killer , perhaps from an English intermediary; and so we reach the
Italian Monferrina and the Spanish and Portuguese Rico Franco .

The spread of German ballads in the immediately adjacent lands
has been very vigorous. As examples of wider influence one may
consider those of the Siideli and Moringer cycles. The former
depends on the medieval epic of Kudrun, and has taken several
forms. The gist of the matter is that a brother discovers his long-
lost sister washing by the shore or otherwise ignobly employed (in
later forms she is a servant in an inn); generally he has made insult-
ing remarks to her while still not recognizing her, and the ballads
are sometimes spiced with the motif of incest narrowly avoided.
The Lusatian and Czech versions are of a late type. In France
there is the same theme, or a similar one, in the Hapless Bride
avenged by her Brothers , which appears as the Sister avenged in
Ital y, and Clotildo in Provence. The name Clothilde does not
occur in the ballad, but was given by Arbaud to his version by way
of allusion to the daughter of Clovis, unhappily married to Amalaric
the Visigoth. In Spain the tale is told under the rubric Don Bueso .
The epic Kudrun rests probably on an older epic Hilde y which
belongs to the maritime peoples of the north. The Danish Mer¬
maid may be a relic of the older poem.

The Noble Moringer or the Crusader’s Return uses a plot as old
as the Odyssey. The details, however, are gathered into a well-
knit tale which can usually be recognized at sight. Immediately
after his wedding a nobleman leaves his wife, appointing a period of
years (usually seven) for his return. She is free to marry after they
have elapsed, but she usually waits still longer, and is often con¬
strained to a new match against her will. Then the lord returns in
disguise, makes his way to the banquet, and attracts her attention
by dropping a ring into the loving-cup. The new match is broken
off at once; the new bridegroom being treated with considerable
harshness by most popular poets. The motif is universal, but the
plot is particular. One encounters it first in Germany about the
year 1200 in Caesarius of Heisterbach. The name of the minne¬
singer, Heinrich von Morungen (of the twelfth century), is attached
by the unknown minstrel who modelled the opening of his poem on
the convention of the aubade . In a later form the hero was stated
to be Henry the Lion of Brunswick, whence the Bruncvik of Czech
historical legends. In England the name is also changed, it has

86 HOW BALLADS SPREAD

become Hind Horn ; in Spain Count Dirlos is a ballad with an

elaborate Carolingian setting, and a later Portuguese adaptation

was the legend of Frei Luis de Sousa . The tale appears m Denmark
as Finnekonsier , a debased form, and in Russia in l the fine
of Dobrvnja and Aljosa Popovic. There is a simplified Czech ballad
entitled First Love , and the poem has flourished in the Balkans also
(Serbian Pomorovac Todor and Jankovic Stojan's Imprisonment ,
Greek Constantine the Little , Rumanian Mosneagul ). 1 In all these
instances we are dealing with poems which reproduce most of the
essential details of the Moringer legend, though with changes of
name. There exist other poems, such as the French and German
Soldier's Return, the Polish Return by Night, the Piedmontese
Soldier's Return , and Sicilian Return of the Prisoner , which may or
may not be associated with the cycle. They shade off into a number
of other returns: the soldier or wanderer may not only find his wife
about to remarry, but actually remarried; or it may be a fiancee
who has died, or who has married some one else. Doubtless many
of these simple pieces have been spontaneously invented, but others
show connecting links.

The consideration of the Danish Aage and Elsa has been post¬
poned in order to take it along with the Greek legend of Constantine
and Arete . In either case there is a revenant. But Aage is a dead
lover roused by the bitter tears of his fiancee to take her with him
to the grave; Constantine is a dead.brother who is forced by his
mother’s curse to fetch his sister home from a far country. She
dies, of course, after so terrific an experience; but that does not
seem sufficient reason to identify the two traditions. The power
of tears to disturb the dead is one motif; the might of a mother’s
curse is quite different. Aage and Elsa is a more powerful piece in
Danish than in the otherwise admirable Sweet William's Ghost.
The German ballads of the cycle are feeble, but they (with the
English poem) lead up to Burger’s Lenore and the full flowering
of the ballad-cult. The Czechs and Hungarian gipsies have ballads
depending on the German. On the other hand, Constantine and
Arete is an equally vigorous creation of the Greeks. Its origin
seems to have been in Asia Minor, for it belongs to the older strata
of Hellenic balladry. In Macedonia it discovered a focus for radia¬
tion into Rumania ( Voichitd ), Bulgaria (Lazar and Petkana ),
Serbia (Jovo and his Sister), and Albania (Constantine, the Dead
1 See Note C, at the end of the book.

HOW BALLADS SPREAD 87

Voyager). In some way not yet explained, the theme detached
itself from Greece and became the English Suffolk Miracle, a
ballad which tramples an impressive subject under a pedestrian
style.

The Greeks of Asia also brought into being and began to circu¬
late the ‘tragoudi’ of The Bridge of Adana, later identified with
Arta. It is a masons’ ballad; to keep a bridge or building firm, the
mason s wife had to be buried in the foundations. The focus for
diffusion was again Macedonia, where all the languages meet in a
multilingual region, but the course was not quite the same. The
master-mason was identified with Manuel, a Greek engineer whose
name appears on a bridge of the seventeenth century. By way of
suspense, we are now told that the victim is to be selected by the
chance that she will be the first to arrive. There is a Bulgarian
fragment, Mono mantore , which tells in lyric decasyllables that
Manuel built a tower and ascended to the top; but fell from there
through being dazzled by the glitter of widow Gjurgja’s daughter’s
ornaments. This may be independent of the other tradition, and
even a direct expression of peasant suspicion of the clever crafts-
man Manuel; but it seems to be the source of the curious ending
of the Rumanian Mesterul Manole. According to this poem the
architects are puffed up with pride at their work, ascend to the top
and think they can fly; so they are dashed to pieces and perish.
For the rest, the Rumanian ballad of the building of Curtea din
Arges is the finest of them all in the use it makes of helpless sus¬
pense ; no prayers of the master nor miracle of God suffice to deter
the faithful lady from coming to her doom. The Serbian Founding
of Scutari {Skadar) uses the datum in a new way. Three brothers
Vukasin, Ugljes, and Gojko—agree to sacrifice the first comer,
but two send secret messages home to their wives; this is how the
Albanians tell their legend of the Bridge of Dibra. Professor
Starkie found a Gipsy family singing the Greek ballad in Morocco
a year or two ago. 1 Another Greek piece is The Bridesmaid who
became the Bride. The discarded concubine dresses herself up with
oriental splendour and dazzles the eyes of the officiating priests
and the bridegroom. The details of her toilette are given in virtu-
ally the same words in the Catalan Lady of Aragon and Spanish

W. Starkie, Don Gypsy, London, 1936, p. 118. The Christopoulos family
were Greeks, but the old Paraschiva was a Rumanian. Hence they disputed as
to whether it was a bridge or a house that Master Manoli built.

gg HOW BALLADS SPREAD

St. Simon’s Hermitage, which are mere fragments. The text is

found in Greek as far west as Corsica.

The influence of Serbian poetry has been very strong on the
like-minded peoples of the Balkan peninsula and it is perceptible
in such Czech pieces as The Robber's Bride, The Turks Bride, and
also in Polish Galicia and the Ukraine. An interesting.chain of
ballads connects the Serbian Banovic Strahmja with the Bulgarian
Iskren and Milica and the Russian Mihail Potyk. The husband is
engaged in furious battle with a would-be seducer. They fight to
a standstill, and appeal to the lady-who helps the aggressor. The
husband is tied up, but released by his dog or miraculously, and
kills both the guilty ones. There is clearly only one story involved,
but it is less easy to give it a home; perhaps Bulgaria.

These instances are a few out of many. Particularly easy would
it be to increase the number of ballads which have travelled a short
distance from a richer land to its neighbours, as from Germany to
Lusatia. Beyond these there is a nebulous mass of floating motifs,
which are sometimes bound together in quite precise formulas: the
May Song which stretches across all Europe, and follows much the
same order in the Greek Swallow Song as in France and Spain; the
Power of Song to draw down birds and bring up the fish round
the keel of a vessel, or else bind a nix, smash a bridge, or toss the
sea; Love which will alone make sacrifices, when no parents or
relatives avail; the intoxicating Power of Beauty, the tests of true
love. A picture of the essential unity of European balladry grows
clearer and clearer as we consider such cases. The signs of inven¬
tiveness multiply, and the evidence of readiness to learn and adapt
interesting themes lies everywhere around the student. There is
no vague, formless, ‘nameless and dateless’ bubbling of mixed
motifs; but artistic creation according to prescribed patterns and
the intelligent, even eager, appreciation of such work. Such
appreciation does not involve incapacity to create. When all the
connexions are worked out, each ballad area is left with dozens or
hundreds or thousands of pieces peculiarly its own, according to
the richness of local tradition. There are richer and poorer areas,
as there are richer and poorer written literatures; but there is some¬

thing of interest in the achievements of every people, and even in
the special significance each has given to the common material.

Before leaving this chapter, it would be well to mention that
ballad variants can often be followed over a national map as ballads

HOW BALLADS SPREAD 89

have been over the map of Europe. Where the variants are
numerous and can be associated with the oral tradition of single
villages, they can be connected by lines which will reveal, as in
linguistic geography, how the newer variants have thrust the older
into forgotten pockets. New and old can be seen on the con¬
temporary map. In Finland this method has shown that ballads
have spread from the coast, exposed to Scandinavian and German
influences, to their present fastnesses of Russian and Finnish
Carelia. In Spain a similar investigation of two ballads ( Gerineldo ,
The Interrupted Wedding) showed their expansion from the south
or south-east towards the north-west. 1 Such demonstrations con¬
cern the specialists who study each individual area; but they show-
in detail what this chapter seeks to establish ‘grosso modo’: that
creation is not a spontaneous effervescence of some mystically
conceived people, but the work of humble but genuine artists
whose songs have been admired, acquired, and spread by persons
who, without being professionals, have a special aptitude for the
art of minstrelsy.

1 -R* Menendez Pidal, in the Revista de Filologia Espanola, vii, 1920.

VII

THE DESCENT OF BALLADS

E PICAL ballads have, for more than a century, aroused an
interest greater than their intrinsic worth, since they seemed
to offer a clue to the rise of the great traditional epics. The ballads
and epics which treat of the Cid Campeador have the same hero,
subject, and incidents; they are both forms of traditional literature,
expressed in not dissimilar language and metre; and it is natural
to suppose that the simpler form will have come before the more
complex. The ballad has been described as ‘logically’ the most
primitive form of poetry. On such a supposition Bouterweck 1
opened his account of Spanish literature with the hypothesis that
‘The poetic spirit . . . was doubtless first manifested in romances
and popular songs’. The hypothesis was most attractive where
ballads were not extant, either because it seemed a providential
clue to a lost antiquity, or simply that scholarly ingenuity was not,
in such a case, cramped by too accurate a knowledge of facts. To
surprise the secrets of the Iliad and Odyssey, tracing their origins
in a ‘sequel of songs and rhapsodies’, ‘loose songs not collected
together in the form of an epic poem until about 500 years after’, 2
was a thrilling achievement; and it was followed by the discovery
of the ‘twenty popular ballads, originally handed down orally, but
written down about 1190 or 1200’, which Lachmann had dis¬
cerned in the Nibelungenlied, and the ‘cantil&nes’ which Gaston
Paris and Leon Gautier saw at the roots of the Chanson de Roland .
The Greek klephtic ‘tragoudia’ served to demonstrate the fact of
ballad literature in Greek, and the Akritic epos had not yet
emerged to show how ambiguous such evidence might be. For at
least three-quarters of a century epic criticism was dominated by
the hypothesis of ballad priority.

This hypothesis was formally disproved for Spain by Mila y
Tontanals, 3 who showed that in every case the Spanish ‘romances’
are much younger than the corresponding ‘cantares de gesta’. Pio
Rajna soon after commenced his hunt for Merovingian epics, as

1 J* Bouterweck, History of Spanish Literature (trans. Thomasina Ross),
London, 1842, p. 17.

_ - ^ Homer , Encyclopaedia Britannica, nth ed. (quoting Bentley, and Wolf’s
Prolegomena ad Homerum, 1795). See further ‘Nibelungenlied’.

A*. Mila y Fontanals, De la Poesia heroico-popular en Espana, 1874.

THE DESCENT OF BALLADS 9 i

the sources and prototypes of later epics; W. P. Ker doubted
whether epics could arise by stitching ballads together; and A.
Heusler 1 showed that, in fact, stitched ballads are just ballads after
all, and quite unepic. The question was investigated with the
utmost care by D. Comparetti, 2 apropos of the Finnish Kalevala .
The Kalevala is a poem of indubitable merit, enjoying a certain
unity of tone, and even of subject, and there are popular songs
extant from which Ldnnrot made his selection. The investigation
made it clear that Lonnrot was far from being the sort of rhapsode
supposed by the Wolfian theory, and that he had produced some¬
thing very different from the Homeric poems. ‘Let it suffice’,
wrote Comparetti,

that we have here shown, from the observations to which the Kalevala
has led us, how devoid of foundation is the theory, under whatever form
it presents itself, which sees in the ancient poems we have mentioned
nothing but songs mechanically joined together; and hence authorizes
the decomposition of these poems into the elements from which they
are supposed to be built up. Any attempt at decomposing organic
poems that do not present a variety of written redactions, sets out from
a principle that is arbitrary, is carried through with insufficient criteria,
[and] is and ever will be barren, fruitless toil.

There is also chronology. We have seen that the European
ballad is the result of a social condition which is defined, for each
nation, within precise limits. It opens after the great migrations
and crusades have subsided, and medieval man has settled down
to cultivate his own acre; and it closes when reading draws away
the ballad-monger’s best patronage. Such conditions are, in the
west of Europe, definitely medieval in time; in the east, they are
medieval in spirit and partly in time. Their epoch is part of their
nature. Ballads so conditioned cannot serve to reveal the sources
of Homer’s poems or even of the Chanson de Roland . The pre¬
servation of accurate historical information across centuries in
which writing and reading can have played little part in con¬
serving knowledge remains as puzzling as before. We have to rule
out the ballads as testimony, though we cannot rule out the notion
of simpler songs preceding more complex. For the Germanic epos
these songs are recorded by Jordanes as ‘carmina paene storico

1 A. Heusler, Lied und Epos , Dortmund, 1905 (‘testibus’ Robin Hood and
Marsk Stig).

a D. Comparetti, The Traditional Poetry of the Finns , trans. Isabella M.
Anderton, London, 1898.

9 z THE DESCENT OF BALLADS

ritiT, and he gives some of their heroes: Eterpamara, Hanale,
Fridigem, Vidigoia. He knew a poem on Filimer and the first
wanderings of the Goths, and his information about the great
Ermanaric was coloured by poetical reminiscences. The form and
manner of such poems is beyond conjecture. They may have been
epyllia or short improvisations preserved in traditional prose con¬
texts ; they may have been brief and dramatic like the Edda or long
and circumstantial like the Beowulf. They may have been eked out
by genealogies, with legends entangled among the names. They
may have been epinician odes. In the Russia of Rurik’s successors
it was customary to break out into paeans after any great or little
triumph, and we are told that ‘velicanija’ were sung for Andrei
Bogoljubski in 1149, Mstislav in 1209, and Daniel Romanovic in
1251. The famous Slovo called Igor's Expedition mingles paeans
and dynastic lists among bursts of narrative. Dragomanov has
reproduced several epinicia among his Ukrainian folk-songs, one
of which runs:

Famous, fair, renowned N . . .,

What the deed that brought thee glory ?
Dusk—and on horse the saddle’s laid,
Day—he alights at Tsarigrad,
and fights and fights with Tsarigrad.
Out comes the Tsar so sore afraid,
and burghers hasten, counsel take,
if any gift his ire may slake.

Such poems would suffice to preserve the meagre historical details
of the ‘byliny’. The Germanic songs may have been of this type
or some other; but the ballads proper offer no sufficient clue.

By way of antithesis to the ballad theory of epic origins there has
arisen, on a Spanish initiative, an epic theory of ballad origins.

The people, hanging on the lips of the (epic) minstrel, caused him to
repeat the most striking passages in a long epic poem so as to commit
to memory the happiest verses. They forgot then the insipid common¬
places, the sluggish developments common in these poems of the
decadence, but they faithfully preserved the memory of the culminat¬
ing points of the narrative or the finest episodes which these strolling

rp /V® M 7 lIa y Fontanals theory, applied by M. Menendez y Pelayo, in his

wfint ° S romances vie 3° s > and developed, with variations, by R. Men&idez

P3SSage ? U ° ted 18 from the kst-named’s Mpopee castillam & travers
la litterature espagnole, Pans, 1910.

THE DESCENT OF BALLADS 93

rhapsodes of the last epoch of epic recitations had been unable to alter
or had been lucky enough to imagine in their search for new sensations.
These passages, thus conserved and often repeated by heart, when iso¬
lated by the surrounding people, became the oldest extant ballads.

It is the theory of fragmentation’. In more recent essays Sr.
Menendez Pidal has given this doctrine a less mechanical cast
allowing a greater place to the artistic sense of the ballad-maker.
The application has been to Spanish conditions only. It assumes
(what is not established) the superior antiquity of the epical ballads
in the Romancero, and goes on to assert that these established a
mould into which other sorts of ‘romances’ could be cast. In other
countries we find evidence of contact between epic poems and
ballads, in which the latter are of younger birth. A straightforward
theory of fragmentation would be difficult to apply, since the epic
originals are, outside Spain, almost wholly conjectural. A brief
survey of the whole ground will show that there are points in
common between the Spanish experience and that of other
lands.

There is only one case in all Europe in which we can place a
ballad against its indubitable original in the older style, viz. when
comparing the Danish Tord af Havsgaard with the Eddie Thryms-
kvida. The ballad is known also in Sweden and Norway, but the
Danish version has been known from the sixteenth century, and
is fuller and better. The two poems are of almost equal length,
the ballad being somewhat longer, thanks to its looser struc¬
ture and repetitions. Individual phrases are preserved as well
as the general outline, but there has been a subtle change of the
poetic temperature. The Thor of the Thrymskvida is a god of
Asgard, named by a liturgical name, and acting in the correct rela¬
tionship to Loki and Freya. The Thor of Tord af Havsgaard is a
farmer:

Now there was Thor of Sea-garth,
rides over the fair green lea,
and he has lost his hammer of gold,
was taken so far away.

Thor, he tamefh his foals on the heath.

Freya is a ‘proud young miss’ and Loki is ‘Lokke the jester’, and
the giant Thrym of mythological Jotunheim has become ‘the old
troll-count’ of Norrefjaeld. The divine Hammer has no mythical
powers, and Thor’s eating and beating are merely a huge jest:

THE DESCENT OF BALLADS

Now there was Thor of Sea-garth,
with the trolls he holds good Thing:
full fifteen trolls and forty-five
they lay there in a ring.

Thor, he tameih his foals on the heath.

The jesting note is present in the Thrymskvida , but it is more dis¬
creet. The singer finds these gods, in whom he only half believes,
amusing in their thick-headedness and petty vanities, like the gods
of Homer. The broader humanity and generalized situations make
of the ballad a new creation, different in class, though partly
identical in language and content, from the aristocratic epyllion.

There is no comparison quite so close in Spanish experience.
The epics of Spain consist of the Poem of the Cid , the Rodrigo , and
a fragment called Roncesvalles, The former is extant in a manu¬
script executed, without great care, in 1307, but the original must
have been composed about 1140. Only one of the ballads of the
Cid answers to this poem, namely the ballad of Bucar. The resem¬
blance is not close; Bucar is killed in the Poem , but survives in the
ballad. In the First General Chronicle Bucar is made to survive,
because in fact he did outlive the Cid ; it has been conjectured that
this chronicle used a refundition of the old Poem, and. it has been
noted that the ballads correspond generally with late refunditions
of the epic material. 1 If that be so, the hope of directly compar¬
ing ballad with epic disappears. The Rodrigo has the advantage,
under this hypothesis, of being a late poem; but its state is so de¬
plorable, so unlike poetry, that the single instance in which com¬
parison is possible becomes fruitless. There are two ballads which
may be conjectured to be part of the Roncesvalles poem, but they
do not correspond with the 100 extant lines. If we augment our
material by admitting that the lines reconstructed by Sr. Menendez
Pidal from the Madrid National Library’s manuscript chronicle
(F182) are genuinely epic—as they may well be—we have still to
face a discrepancy between the circumstantial manner of the epic
lament and the elliptic ballad style. 2

1 R- Men&idez Pidal, in his edition of the
Lara, and in his note on the Feman Gonzalez
y Pelayo.

dd, his Leyenda de los Infantes de
cycle in the Homenaje a Menendez

3 The discrepancy has been noted by M. Men&idez y Pelayo, Tratado de los
romances viqos,Antologia depoetas liricos Castellanos, Madrid, 1924, xi pp 276-

THE DESCENT OF BALLADS 95

For these reasons it is impossible to clinch the proof of ‘frag¬
mentation by pointing to any epic fragment which is a ballad The
correspondences are striking enough, despite the differences noted
and fragmentation offers a possible, though not a certain, explana¬
tion. Apart from the instances cited we know Spanish epics not
in their verse form, but as embedded in the prose of general
chronicles of spam. The ballads correspond with the data of later
chronicles better than with the earlier, and it is possible these data
may be due to redactions of the old poems. The ballads which deal
with the destruction of Spain by the Moors and King Roderick’s
death in 711 are based frankly on a prose text, the Cronica Sarra-
cma of 1430. They are not really epical ballads; indeed, it is not
ceitain that there was a Spanish epic on this subject, the surviv¬
ing forms of which are prose traditions reported by Arabic and
Castilian historians. There was a French epic on the theme-
Jlnseis deKarthage. A number of ballads deal with Bernardo del
Carpio, his resistance to the French at Roncesvalles, and his efforts
to release his father from imprisonment. They are not old, and
all but one could have been excavated from the chronicles. * One
of the ballads on Count Fernan Gonzalez, liberator of Castile,
appears to be authentic, the others being examples of mere ballad¬
making. We are here dealing with a ‘cantar de gesta’ summarized
m the Cronica Najerense (about 1160), but not in the Historia
Silense (1109); it was probably assembled between those dates by
the use of common novelistic matter, seeing that all the episodes
are of a romanesque sort. About 1250 the ‘cantar’ was rewritten
in rhyming quatrains by a cleric of no great literary attainments,
and his work was turned into the prose of the First General
Chronicle. The old ‘cantar’ doubtless lingered on, supplying to
later chronicles and the ballad of the ford of Carrion details which
the cleric had ignored. The case hardly requires us to hypothetize
a second redaction of the ‘cantar’, since we have only indirect
access to the first.

The ballads of the Infantes de Lara cycle are more numerous,
vivid, and of primitive passion. It is a family history of revenge and
treachery, and as such did not attract the notice of the earlier his¬
torians who wrote in Latin concerning the affairs of king s The
vernacular historians were thus not tempted to distort the legend
in the attempt to make it square with some preconceived notion
of history, and we can be more sure in this case than in others that

9 6 THE DESCENT OF BALLADS

they have reproduced in their prose the matter of the epic poems.
The assurance is doubled by the copious traces of versification still
visible in the prose; it is almost always possible to pick out the
assonances. The kernel of the legend seems to have been a disaster
in which a Moorish frontier chieftain and his sons perished in 972.
The epic story concerns a Christian family, and presents other
points of difference from the historical facts; but the relations sup¬
posed to be normal between Moors and Christians are such as were
terminated in 975, when Al-Mansur took over personally the con¬
duct of the frontier wars. The original ‘cantar de gesta’ must have
arisen in the last quarter of the tenth or the first half of the eleventh
century; it is reproduced in the First General Chronicle (1289).
The conclusion of the tale is a work of imagination, however
historical the first part may be; it is a tale of revenge which invites
amplification. From one chapter in the First General Chronicle it
swelled to four in the Second (1344); and its aggrandizement con¬
tinued until, in 1834, the revenge became the whole story. 1 The
new details arise out of the old, but are not compatible with them.
They are a new conclusion, and a conclusion in verse, as we learn
from the numerous assonances in new series. It is clear, then, that
for the Infantes de Lara a new redaction arose between 1289 and
1344, more romanesque and circumstantial, less tragic and more
pathetic, with humorous touches and a greater geographical spread.
It is this epic which is represented in the ballads, probably by
Immediate descent.

Lastly, the ballads of the Cid amount to more than two hundred
and cover the ground of three epic poems: the Mocedades or youth¬
ful feats, of which the Rodrigo is a debased form, the Siege of
Zamora (lost), and the Poem or feats of his maturity. For our
present purpose the Cid ballads prove disappointing. Only twice
is it possible to attempt a comparison, and the comparison reveals
differences which force the critic to take refuge in fresh hypotheses.
The theme was doubtless too popular. It invited improvisation,
since it could count on an interested acceptance; new ballads drove
out the old, so that the whole cycle seems relatively modem and
arbitrary.

When we transfer our glance from Spain to the Germanic North,
we encounter more evidence that traditional epics have provoked
the rise of epical ballads. The contact is less intimate than in

1 Don Angel de Saavedra, Duque de Rivas, El Moro Exposito , Paris, 1834.

the descent of ballads 97

Spam. The ballads belong, for the most part, to the fifteenth or
fourteenth century, but the corresponding epics had run their
course by the tenth. Some of them, like the Nibelungenlied and
Kudrun, received new and more ample proportions in the twelfth
and thnteenth centuries under the influence of the courtly epics
of France; but even so, the contact between epic and ballad in the
same cycle is hard to establish.

Our own Beowulf is the only complete survivor of this ancient
epic mode, and it has had no ballad consequences; nor has the
fragmentary Fight at Finnsburg. The adventures of Walther of
Aquitaine, his escape from Attila’s court with his beloved Hilda
and his great fight in the Vosges, survive in two tiny fragments of
our Waldere and in the Swiss-Latin Waltharius (tenth century).
The story should belong to the German tribes on France’s eastern
border, but it appears to have effected a lodgement in Aquitaine
taking the local name Gaiffier. From such a version the Castilian
ballads of Gaiferos descend. Offshoots of the legend are the
romances of Aye d’Avignon and the Comtesse de Ponthieu , and also
the Proven?al ballad Escriveta, with its novel maritime setting,
which has spread to neighbouring lands in the style described in
the last chapter. While there is not much reason to doubt this
genealogy, the fact is that the original poem is very imperfectly
known and each essential link in the evidential chain has been lost.

Germany offers the epic and ballad Hildebrand. The epic belongs
to the ninth century, and is a fragment. It may not be a fragment
of a longer epic poem, since in reality old Master Hildebrand has
no other achievement than this fight with his son; the poem may be
a planned fragment, like Tennyson’s Ulysses and Morte d'Arthur . 1
The battle ended disastrously, and did not please the taste of later
times. In the Norse Thidrekssaga, itself depending on a Low
German Dietrichs Saga, the combatants are reconciled, and return
happily, and even playfully, to the Lady Ute. So the story was
told as the twelfth century gave way to the thirteenth; and so the
story is told in the ballad Hildebrand, with additional developments
of a sentimental kind. It is clear that the saga has intervened
between the ballad and the epic fragment, so that there is no direct
contact between the two. It was probably an older, more epic form
of the story which travelled to Russia and gave rise to the bitter
ballad of IVja and his Son; though the latter might also have been
1 I owe this suggestion to Professor Norman, of London.

9 S THE DESCENT OF BALLADS

inspired by Firdausi’s account of Sohrab and Rustem, or the Greek
Tsamados and his Son. The battle of father and son is a theme as
old as the Telegoneia , and its medieval forms are too scattered to
be reduced to one original. In addition to the Hildebrand , Germany
offers the curious Ermanarids Death , which reads more like an
exercise in the medieval epic style than a ballad. It also appears to
arise from the lost Dietrichs Saga. Between the Wolfdietnck and
the Hunter from Greece ( Jdger aus Griechenland) there is a con¬
nexion, though distant.

The Nibelnngenlied and Kudrun offer special cases of ballad and
epic contact, since they existed in both primitive and medieval
forms. The story of the Nibelungs appears in the Edda , the
Vdlsungasaga y and the medieval German poem still extant. The
testimony of the Edda is twofold, since it relies on versions already
established in the north, and on later German information. (The
account given by Beowulf lies outside this connexion, since the
author lived before Sigurd-Siegfried was invented.) The legend
has given no German ballads, but several in Scandinavia. Once
again, the Thidrekssaga intervenes, through the incorporation of
the story of the Volsungs. The Faeroese Regin the Smith goes
back, apparently, to a lost twelfth-century original, distinct from
the Eddie Reginsmdl; the Brinhildar-tdttur depends on the Thid-
rekssaga, Volsungasaga, and some other source; and the Hogna-
tdttnr on the Thidrekssaga, through a west Norse song. 1 The
Danish Grimhild's Revenge may owe something to this song, but
has suffered the influence of the twelfth-century Nibelungenlied.
It is one of the ballad-class which aims at summarizing in the
traditional style the material of some more aristocratic romance.
Brynilds Vise differs from the Eddie treatment and from the
Nibelungenlied . The Norse Sigurd Svein maintains an association
with the Eddie Reginsmdl and Fdfnismdl , which is not evident in
the more developed Danish Sivard Snarensvend.

Kudrun stands next to the Nibelungenlied for popularity and
merit, as the Odyssey does to the Iliad. An older form of the legend
must have been current under the name of Hilde , the plot of which
is summarized by Snorri at the end of his prose Edda. It is this
Nordic form which is conjectured to be at the base of the Danish
Havsfrmiy but the German ballads may derive from Kudrun itself.
The epic is composed of conventional materials, and has only one

1 H- de Boor, Die fdrdischen Lieder des NibelungenzykluSy Heidelberg, 1918.

THE DESCENT OF BALLADS 99

memorable scene. It is the moment when the bullied girl goes out
to wash clothes on the sea-shore, and two strangers approach,
treating her as if she were a servant. They prove to be her fiance
and her brother; she throws her washing away, and they concert
an escape. In the ballads there is no place for the fiance, and the
incident is 1 educed to the discovery of a sister in distress by her
brother. The detail that she is washing clothes is preserved in
some ballads, but not in all. In late German derivatives she is
represented as a drudge in a hostelry, exposed to the insults of
travellers. Thus the whole plot of Kudrun is extraordinarily
simplified, it is reduced to the one striking episode, and that is
stripped of all confusing detail. Hence the German Siideli,
Meererm, Starling and Bathtub (the tub is all that remains of the
original laundry), Rediscovered Sister, Rediscovered Princess. The
Czechs have a late form of this cycle ( Nalezena Sestra), in which
the girl is found to be a barmaid. In the Spanish Don Bueso 1 we
have a more heroic variant, and in the French Maumariee vengee
par sesfreres, Provencal Clothilda, if they belong to the same cycle
we have versions with almost all the characteristic lines effaced.’

The Thidrekssaga, which we have seen affecting the ballads of
the Nibelung cycle, also lies behind a number of the best Danish
ballads: the Tournament (or They were seven and seventy), Kin?
Diderik’s raid into Beriing’s Land, Viderik Verlandsen’s fight with
Longbone Giant. King Diderik and the Lion is the Wolfdietrich
legend, which is connected with the French Yzuein and the German
Heinrich der Lowe. The pattern of these adventures serves for
other ballads of free composition, such as Berner Rise and Orm
Ungersvend (unless this is based on some lost original) and Diderik
and Holger Danske. The ballad of Samsing belongs to the prole¬
gomena of the saga. We are told that the old Norse saga, of the
early thirteenth century, rests on a German original, and this need
not have been very much older. It is true there was a Theodoric
of Verona, the great fifth-century Ostrogothic king. Deor men¬
tions him in his lament :

Theodoric held thirty winters

the Mayings’ burg; that was to many known.

He is a dignified recipient of Gudrun’s laments, and so provoked
Attila’s jealous reproaches in the Eddie Gudrunarkvida III, and he

* , R ' Men&dez Pidal, ‘Supervivencia del poema de Kudrun’, Revista de
r ilologia Espanola , xx, 1933.

100

THE DESCENT OF BALLADS

was the chief representative of the beaten German peoples at the
Hunnish court. But, unlike his ancestor Ermanaric, he had no
‘geste 5 of his own until the late twelfth century, when he suddenly
became the centre of a sub-Arthurian court, and the representa¬
tive of all Germany. The sagas of Weland the Smith and Widga
helped to give the proper introduction to his feats, and the Vilkina-
saga, Volsnngasaga, Biterolf and Detlieb, Waltharius, Hildebrand,
Ermanaric"s Death, and other heroic pieces were annexed to provide
the necessary incidents. His own part is chiefly to sally out against
the most reputable enemies, for no reason at all, and prove his
superiority. It is on this model that the Danish ballads are based,
both those which come from the text of the saga, and those which
are free. His confidence is overweening:

King Diderik sits in Brattingsborg
and out he looks so wide:

‘No man know I in all the world,
that can be deemed my like.’

Then answered Brand Sir Vifferlin,
for he had wandered so wide:

‘E’en will I show you a champion good,
that well with you dare strive.’

That is how the raid on Berting’s Land opens; as Berting’s Land
seems to be Bretagne, his achievements prove the superiority of
the German Round Table over the Breton one. An unknown
minstrel supposed he made an attack on Denmark in the same
spirit.

King Diderik sends King Holger word,
and thus he bade him say:

‘What, whether wilt thou strive with us,
or wilt thou tribute pay ?*

Holger refused, and won his fight, as he had previously won his
duel against the giant Burmand, who was also thought of as one of
those powerful Germans who perpetually threatened the southern
frontier of Denmark.

The saga is interesting also for the notice it takes of the Russian
hero Il’ja of Murom. He is called ‘Il’jas af Greka’ and is supposed
to be the bastard son of Hertnit ( =Ortnit), whose eldest son
Osantrix ruled in Lusatia, and his second was Valdimar or Vladimir
of Pulinaland. ‘Ilias von Riuzen’ is also mentioned in Ortnit\ he

THE DESCENT OF BALLADS I0I

was therefore an epic hero of Russia whose fame had spread to the
German peoples m the thirteenth century, though the account they
gave of him was different from that of the surviving ‘byliny’ Con¬
versely, the ‘byliny’ of Il’ja bear traces of the legend of Detlev
Danske (or Detlieb), in the motif of his powerless early years, the
manner of his leave-taking, and the early skirmishes with robbers.

bandk°Sigurd SOlOVei ^ S ° mething t0 Detlev ’ s °PP™ent, the

We have already mentioned the ballad adaptation of the Eddie
Tkrymskvida, but the indebtedness of the ballads goes further
than that. Only one of the supernatural pieces is used, and in the
heroic ones of the Nibelung cycle we find that the ballad poets
have drawn on newer German material by preference. There
remains the similarity of plot between the second Helgi lay and
Ribboli and Guldborg (our Earl Brand or The Douglas Tragedy )
and that between the two poems which make up the Svipdagsmdl and
the ballad Ungen Svejdal. In this latter case there need be little
doubt that we are dealing with two forms of the same tradition-
but the interval between the Eddie and ballad styles is immense’
i here is a ritual solemnity about the Eddie poems. The waking
of Groa m her grave is only less impressive than that of Angantyr
and the riddles propounded by Fiolsvinn to Svfpdag, while they
are of maddening triviality, have a serious magical import. All
this is loosened in the ballad style, where magical gifts are accepted
with the naivety of a fairy-tale and riddling is a mere contest of
wits. In this case also there is an evident contrast between the
aristocratic manner of the Edda —mythological or heroic—and the
broad, though naive, humanism of the ballads. Alf i Odderskjser
is arehandling of the Hervararsaga, which reckons among the
Eddica minora, and Regnar and Kragelille with the Fight -with the
Dragon depend on the post-heroic saga of Ragnar Lodbrok.

Yet one more important Germanic epos to be encountered in
Scandinavian balladry is the tale of Hagbard and Signe, the Tristan
and Isolt of the far north. Saxo Grammaticus has reproduced the
egend in prose and verse, basing his narrative on the older, more
circumstantial epic which has been lost. The Danish ballad is one
of the finest of all the ‘viser’. The central incident is Hagbard’s
disguise in woman’s clothes in order to penetrate to Signe’s apart-
ment. It survives, probably, in the German and Dutch Prince
a-wooing or Disguised Margrave's Son .

102

THE DESCENT OF BALLADS

Later sagas are used freely in the formation of Norse and
Faeroese ballads, and the Faeroese Ormurin langi (from St. Olaf's
saga ) shows that the process of hewing ballads from such sources
has continued to our own times.

The French ‘chansons de geste' have had no consequences in the
‘chansons populaires', which originated in the lyric only, nor in
Italy, Germany, or England; but the Spanish ‘romances' and the
Scandinavian ‘viser’ have been deeply affected. In Spain there
circulated a local version of the Roncesvalles epic, in a late form
which gave Reinaud de Montauban more importance than Roland.
Reinaud has the same prominence in King Marsiris Flighty a ballad
which is manifestly a fragment. Another ballad, that of Lady Alda ,
described in pathetic detail the foreboding dreams of Lady Aude and
the receipt of the news of Roland's death. This piece is complete
in itself, but corresponds in general to the pathetic amplification
given to this incident by decadent epics. By a curious develop¬
ment, Roland’s sword Durendal became a person in Spanish
balladry, and there is a pretty little cycle describing Durandarte' s
love for Belerma and his dying messages. They recall, by way of
contrast, the grimly tragic Sword of Vengeance (.Hsevnersverdet ) of
Danish and Norse tradition, which leaps of its own accord to kill
a banesman and his child, and can hardly be stopped from killing
its holder also. There also is a sword that speaks, though still a
sword, fitted into an imaginative tableau of epic circumstances.
There is a small fragment of a Norse ballad of Roncesvalles
{Rolandskvsedi) as well as the Faeroese Runsivals Strid, a summary
based on the Karlamagnussaga.

Excavations conducted in this saga have given the Faeroese a
number of not very interesting Carolingian ballads: the Geipa
tdttur which reproduces the gabs and incidents of Charlemagne’s
pilgrimage, Emunds rima on Roland's youthful feats, 6 dv aid's rima
relating his fight with this pagan, Flovants rima, Runsivals Strid,
and Oluvu kveedi, corresponding to the Icelandic Landres-rimur.
The history of the Ogier ballads in Denmark and Spain is more
remarkable . 1 The cycle is divided into the thirteenth-century
Enfances, which describe Ogier’s combat with the monstrous

1 W. J. Entwistle, ‘Concerning certain Spanish ballads in the French cycles
r Awl (Montesmos), and Ogier de Dinamarche\ Studies presented to

i.. E. Kasiner, Cambridge, 1932. S. Grundtvig, Damnarks gamle Folkeviser,

the descent of ballads i03

fhTtwelfS t0 ? V£ ?° riande for his friend Karaheut, and
and oli/r^T 7 Chem l erie retaiIin S Charlemagne’s injustice

has simnlified ^ In each CaSe the baIlad P° et

timeT 1 3 u 0m .? eX CP1C situation ‘ The Dane cannot spare

hZ Sr I K ar eut d0es not fi g flt for his °wn bride, nor

?amo?s g he ^ 3 PnS ° ner 3nd ? et a friend t0 of his

T h P e r’Jv erelySUpp0Ses that there was cowardice somewhere.

CharSP r ^ the Wh ° le feudaI «yrt«n, which prevents

Charlemagne from offering his son Chariot to Ogier’s vengeance

but C ause S the emperor to consent at once, instead of Zr tin
housand lines, and to hurry on a farouche trial. The name Holger
Danske was enough to make this hero famous as a defende/of
. Danish liberties, and the refrain ‘Holger Danske, he won Hew

2 B , U / mand seemed worthy to be carved on a Swedish chand
m the fifteenth century. Burmand represented the German menace
with poetic vagueness; to make the point clearer, there arose the
companion ballad of King Diderik and Holger. I was from such

iTthSef ^ d f enderS f Ae Da “ eVerk t00k ^

fr °m r U r 3Sthe m0dd f ° r D ° n Quixote ’ s sally.

the Sna rnku ° r a cka P boob served as immediate model of

epic p^rwHct ^ U T t3in ‘ ThC baIkd summariz ^ the whole
p plot, which is one of injury and vengeance, but omits all tho-e

nsiderations which make for epic length. Had Charlemagne

concluded ^amarche wodd hSe

rendtred to' n^ IinCS; at lon § Iast Chariot is sur-'

rendered to Ogier as the price of his assistance in a time of

desperate need. The ballad poet does what the epic poet Sd^

domg the latter wished to spin out a long tale, the former to give

A" fh +h te ^ atter ' The same treatment is given to the French

Spani3h ^- "allads, to AiZlt

chapbook, as m the Valdovmos series and Guarinos. Gerineldo one
of the most popular of Castilian ballads, is vaguely Camhngkm

it h S a rdT neCted Wlth the legend of E § inhard ’ s love for Emma g but
on^tlie themeP)f t phe m0^ ^ , *** ff b& j 3d sba P e ’ rh 311 an improvisation

ePrnfinP 3qmre ° f ] ° W de g ree ’- °n the other hand, the

ballad f a? convention is such m Spam, that common international
ballad stuff is given a Carolingian setting. This is the case S

io4 THE DESCENT OF BALLADS

Count Dirlos, a derivative of the Noble Moringer theme, and Count
Clams of Montalban. The latter employs the motif of the lover
who brags of his real or imagined successes, a motif that appears
separately in Florencios and Count Velez bragged , and also in the
Russian Youth and Prudent Woman ; it goes on to describe lovers
separated by imprisonment, whether of the young man (corre¬
sponding to the French Pernette) or the princess (La Fille du Roi
Loys). The elaborate Carolingian setting of Count Dirlos resembles
the equally elaborate Kiev atmosphere of Dobrynja and Aljosa.
The maker of the Gaiferos ballads was also careful to collect
Carolingian heroes to form an entourage for his escaping lovers,
though there is no such background in the Provengal Escriveta.

The French romanesque influence is only second to the epical.
Among the Arthurian legends, those of the Holy Grail proved to
be too mystical to attract the ballad-monger’s attention. There is
a Castilian ballad, certified ancient by Nebrija in 1492, on one of
Lancelot’s adventures, which is akin to the Lai de Tyolet and an
episode of the Dutch Lanceloet. Another ballad on this hero seems
rather to be a free imitation of the romance in its general tenor. A
ballad of the death of Tristan, also of ancient date, is composed of
a few pregnant phrases taken from the last three chapters of the
novel of Tristan de Leonis. There is an effaced relic of the same
matter in Germany (Liebestod), and in Iceland and the Faeroes
there are popular ballads (Tristrams kvaedi, Tistrams tdttur) based
on the alternative conclusion of the romance. It is in England that
the Arthurian cycle is most developed in pieces largely independent
of the great romances: The Boy and the Mantle , King Arthur and
King Cornwall, The Marriage of Sir Gawain. They are ‘ballads of
minstrelsy’; that is, the evidence of personal composition is so
strong that one may doubt whether these pieces are truly tradi¬
tional. Floire and Blancheflor gives rise to German and English
ballads, Amis and Amiles to one in Spanish, and the romance of the
Castellan de Corny is the source of the German and Dutch Brem-
berger. Similarly, the English Hind Horn is based on a romance
which has been preserved in three shapes; 'fat Count of Rome is
founded on Alexander von Metz , and has spread from Germany to
Scandinavia; and Herzog Ernst , though included in the standard
German collections, is rather ballad-like than a ballad.

Before leaving the French area we must notice the effect of con¬
ventions laid down by literature. They have been mentioned at

THE DESCENT OF BALLADS 105*

the close of the last chapter, but have their place here also. There
is the ‘aubade’, with its lovers who part; the Spring-song, with its
sequence of set phrases; the ‘pastourelle’, degenerating into any
frivolous encounter between a man and a maid (and so giving such
urban consequences as La Belle Barhiere ); the ‘maumariee’ and the
ampler convention of taking matrimony in vain; scorn of the timid
lover; assignations by night and the consequent adventures. Once
laid down, these conventions form moulds for new inventions
which require little effort to imagine. They are an inexhaustible
fount of Gallic wit, and operate more strongly in Flanders than in
Germany, in Catalonia than in Castile, for in Castile marriage is
taken seriously and its infringement is tragic. These are not cases
where the student may compare an offshoot with an original, since
there need be no direct original, but only a pattern. The pattern,
however, is definitely of literary origin.

In the east of Europe we can hardly command such accurate
documentation as in the west, and the relation between epos and
ballad is less easy to define. Greek ballads were first known in their
late European forms, chiefly the klephtic ‘tragoudia’ of Epirus
describing guerrilla warfare against the Turks in the eighteenth and
early nineteenth centuries. It was much later that the great store
of Cappadocian ballads from Asia Minor began to be opened, and
the name and deeds of Digenis Akritas swam into our ken. The
Greeks of Cappadocia have lived without much knowledge of their
compatriots in Europe, under an alien rule. Only as a result of the
disastrous Graeco-Turkish War and the consequent population-
shifts have these isolated communities been brought into contact
with Hellenic scholars. The collection and comparison of Akritic
ballads is thus still under way; one hardly knows how far the name
extends or what other cycles should be marked off from those of
Digenis. A kind of epic poem on this hero was published by Sathas
and Legrand in 1875, from a fifteenth-century manuscript of
Trebizond, and other manuscripts have appeared more recently,
of which H. Gregoire relies on that of Grottaferrata. 1 Apart from
Digenis’s defence of the Euphrates we have to allow for a ‘geste’ on
the defence of Amorium, and for songs concerning the Apelates
Philopappos, loannikis, and Kinnamos. The ballads and the epics

1 C. Sathas and E. Legrand, Collection de Monuments pour servir a Vetude de la
Langue neo-hellenique, nouv. s€r, 6, Paris, Maisonneuve, 1875; and H. Gregoire’s
articles in Byzantion, v, vi, vii, 1929-32.

I0 6 THE DESCENT OF BALLADS

employ the same ‘politic 5 metres, but are different in style and con¬
tents. Which came first and how they are related seem to be
matters still ‘sub judice 5 , since the ingenious arguments of H.
Gregoire have not been subjected as yet to any second examination.

Current opinion ascribes priority to the ‘tragoudia , while con¬
ceding great age to the epos. Basileios Digenes Akritas is a poem
which uses material from Genesius’s biography of Basil I, with his
feats of strength and prowess as a hunter. The date of Genesius is
approximately the mid-tenth century. The Grottaferrata text
states that the Holy Face relic is to be found at Edessa. It was
removed to Constantinople in 944. Before this can be accepted as
a ‘terminus ante quern 5 , we should have to be certain that no
antiquarian information was accessible to the poet. In the Andros
and Trebizond manuscripts there appears a certain ‘Aaron 5 , who
can be identified with the third son of the last Tsar of western
Bulgaria; he entered the Byzantine service and fought against
the Seljuqids about the years 1047-59. This is taken to indicate
a revision of the epos sometime after the middle period of the
eleventh century. The Trebizond manuscript is of the fifteenth
century, and it is such a poem as one would hesitate to place along¬
side the Cid or Chanson de Roland . The Greek employed is of a
literary cast, and the author quotes his Iliad , Quintus Curtius,
Aldelaga and dope (an erotic novel), and several books of the
Bible. On the other hand, he lacks the characteristic merits of the
traditional minstrel, for he cannot tell a story, and mins several
promising situations. He alludes to ballads on the Apelate chief¬
tains, which his poem is intended to supersede.

The ‘geste 5 of Digenis is connected with neighbouring literary
works: with the Armenian epos of David of Sasoun, the history of
'Umar al-Nu'man in the Arabian Nights (which Gregoire terms the
‘Geste de Melitene 5 corresponding to the part of the epos he terms
‘Geste de l 5 Emir 5 ); the Arabic chivalrous romance Ddt al-Himma
wa’l-Battal; the Turkish metrical romance of Sayyid Battal; and
the Russian Devgenit, Philopappos was, it seems, the original hero
of the ballad Syropoulos (or Scleropoulos) steals Kostantad Bride,
later transferred to the Akritic cycle; Andronikos’s Son was origin¬
ally independent; and the song of Amouris (or Amouropoulos) was
originally connected with the defence of Amorium in 838, and not
with Digenis, turmarch of the Anatolians, who perished at Kopid-
nadon in 788. In addition to the 'Geste de Melitene 5 , Gregoire

THE DESCENT OF BALLADS 107

discovers an ‘Epopee commagenienne’ (taking Philopappos as
Antiochus Philopappos, the last king of Commagene, whose tomb
was erected in front of the Athenian Acropolis in a.d. 114-16).
This Commagenian epos would link the Byzantine to the Persian,
since Kinnamos was a rival of Artabanes III, whose son Gotarzes
or Goderz is a personage of the Shahnama.

We can indulge in such remote conjectures only at our peril.
They are not unlike equally hardy theories about western epics,
which have been discarded in favour of more restricted statements.
We may retain chiefly the impression that the epics and epical
ballads of Greece are somehow linked, as they are in the west, so
that there is no absolute separation between oral and written
literature.

Finally, there is the evidence of Russia. This also is inconclu¬
sive. We have already noticed that certain Germanic sources seem
to be intimately related with the Russian ‘byliny’. Detlev and
Hildebrand are heroes who resemble IPja of Murom; he, in his
turn, is the centre of a cycle of stories which might have had, at
one time, a formal unity. To the compilers of the Thidrekssaga and
of Ortnit , Il’ja is a hero of epic mould associated with Vladimir of
Kiev; but he is also connected with a Hertnit and an Osantrix quite
unknown to the Russians. Some knowledge of French literature
may have penetrated as far as Russia, since the rich Indian king¬
dom of Djuk Stepanovic reminds us of Prester John’s realm and
Charlemagne’s pilgrimage; while Vasili! Buslaev’s evil-doings are
like those of Robert le Diable. The Greek Amouris and Tsamados
are analogues of Sam Vanidovic and IVja and his Son. An eclectic
criticism 1 would go farther and enumerate, as well as those given,
Yugoslav, Caucasian, Iranian, and Finnish elements in the ‘byliny’;
but in these remoter fields it is not always easy to feel confident as
to the relationship of cause and effect. Some international ballad
themes have also penetrated as far as medieval Russia: the rings of
Marianson or Imogen, and the Noble Moringer’s return home.
There are too many unknown quantities to allow definite conclu¬
sions, but the evidence seems to indicate that the Russian ‘byliny’
arose with some knowledge of Germanic poems, whether German
or Scandinavian, and that western and Balkan elements continued
to penetrate so far.

1 V. A. Keltujala, Kurs Istorii Russkoi Literatury, St. Petersburg, 1913, i,

pp.'824-5.

ioS THE DESCENT OF BALLADS

On the other hand, the Tyliny 5 are strangely unlike the surviving
exemplar of old Russian epic poetry, the Slovo o Polku Igoieve.
The nucleus of this piece is an account of Igor of Novgorod’s raid,
in conjunction with his brother, against the Polovcy. It was an
unsuccessful affair, but the poet favours Igor, who is represented
as the pattern of headlong daring. He is still alive (he died in 1102).
There are some fine descriptive passages, such as the one which
compares the flight of the nomads 5 arrows to a storm at dawn. The
nomads are pagans, but the Russians still believe in Stribog,
Dazd’bog, and Div, the wood-demon. The hyperboles are many,
and negative comparisons are used, but there is otherwise little
resemblance to the ‘bylina 5 style. Intermixed with the main theme
are interpolations: one in praise of Svjatoslav of Kiev, who success¬
fully defeated the Polovcy under Kobjak, can hardly be by the
partisan of the unsuccessful Igor. Vladimir is mentioned, but only
as a remote memory. On the other hand, the poet is skilled in the
genealogies of more recent sovereigns in the various petty states.
Not only is his work unconnected with any ‘bylina 5 , but it repre¬
sents a technique and field of knowledge incompatible with their
style. One must admit, however, that the unknown poet claims to
be an innovator; lie refuses to write in the style of the eleventh-
century Bojan, and lie even parodies his exordia. Bojan, therefore,
may have treated other subjects in another fashion; but he can
hardly have been so vague as to dates, times, and persons as the
extant ‘byliny’.

Finally, to close this chapter on the literary antecedents of
European balladry, we should not forget the biblical and classical
ballads which are by definition literary. They gave rise to the
stories of Joseph and Potiphar’s wife, Benjamin’s cup, Samson,
Holofernes, episodes of the life of Christ, the Samaritan woman,
St. Stephen’s martyrdom, the miracle of the roasted capon which
crew, Judas’s remorse, and also the ballad progeny of Orpheus and
Eurydice, Leander and Hero, Heracles and Charon, Helen, Pyra-
mus, and Troilus. There is no need to insist on them, since once
the difficulty of recognition in some cases is overcome, the relations
between original and copy are not hard to define.

VIII

THE ASCENT OF BALLADS

T HE humble folk who listen to ballads with all their ears say
little about their merits. 'That is a pretty ballad’ or 'that is
sad’ are their comments, and they are even shy at mentioning them
in the presence of strangers. They may suspect, as Vuk Stepan
Karadzic did, that the townsman’s curiosity is the preamble to
some taunt; 'love songs’ seem to them less worthy than other tales
and songs; and they would rather display their acquaintance with
some urban novelty. When they can be coaxed into singing tradi¬
tional ballads, or surprised in the act, it is their absorbed attention
alone that shows their esteem. They may discuss the event
narrated or compare other variants of the same poem, but they
have no adjectives to spare for the manner of balladry, and the
wisest of critics have followed their example. The late W. P. Ker,
that model of delicate and judicious appreciation, describes the
ballad not by formula but by example.

In spite of Socrates and his logic (he wrote) we may venture to say, in
answer to the question 4 What is a ballad?’— 4 A ballad is The Milldarns
of Binnone and Sir Patrick Spens and The Douglas Tragedy and Lord
Randal and Childe Maurice, and things of that sort.’

Sir Arthur Quiller-Couch, 1 who quotes these words, goes on to
define the ballad style by examples, for which the following may
be substituted: a ballad is

There came winging then two coal-black ravens,
feathers dripping blood upon their shoulders,
white the foam that from their beaks was dropping;
in their talons bring they a hand of hero,
on the finger is a ring all golden,
and they cast it in the mother’s bosom.

and

God destroy him Vuku Brankovicu!
traitor to his kinsman at Kosovo.

Lazar that day was by Turks confounded,
when there perished all his goodly army,
when there perished seven and seventy thousand:
all was holy, all was honourable,
all was as our loving God appointed.

1 The Oxford Book of Ballads, Oxford, igio, 1932.

THE ASCENT OF BALLADS

It is
and

and

The daylight dawns to eastward,
it lightens over all.

There lies a castle in Austria
that is so nobly founded,
with silver and the red, red gold
and marble stone surrounded.

The dance went down by Ribe street,
there danced the knights so glad and fleet.

The dance went Ribe bridge along,
there danced the knights with fair slashed shoen.

And first there danced Sir Riber-Ulv,
he kept so well his faith and truth.

Abenamar, Abenamar,

Moor above all Moors to me,
on the day thou wert begotten,
mighty signs there were to see,
for a calm was on the ocean,
and the moon was full and free;

Moor beneath such sign begotten,
hath no taint of falsity.

The accent is unmistakable, whether it be high:

See him, see him, where he cometh,
cometh the avenging prince.

Then spake the sword in weary mood:

4 Now lusted I for thine own blood/

The king, he rules the cities
and over all the land,
and over so many heroes bold
with naked sword in hand.

Let bonders mind their dwellings,
the courtier his horse!

The king, the king of Denmark,
he ruleth town and force.

or low

He let strike iron for hand and foot,
for he would wend from that land out.

THE ASCENT OF BALLADS n

And took his pilgrim-staff in hand,
and far he journeyed from that land.

He journeyed o’er the Middle Sea,
till he the Holy Grave did see.

And never healing message got,
but the iron sat on hand and foot.

and

A linden stands in yonder vale,
above ’tis broad, and under small.

or merely flat and pedestrian:

They fought one day, they fought for two,
the third till even fell,
they sat them down upon a stone,
the heroes fain would dwell.

The accent is always unmistakable, and like no other.

It is when epithets are sought for this strange ballad charm, this
davjiaoTov tl xprj/j.a (as one of the ancients called Sappho), that the
voices of critics seem unable to avoid a condescending note. The
ballads are ‘wild’ or ‘vulgar’ and constitute only a ‘sort of’ poetry.
Perhaps the earliest thoughtful reference to the genre—that of the
fifteenth-century Castilian Marquess of Santiliana—may be taken
as typical, for he declares that the lowest sort of versifiers are ‘those
who without order, rule or count, make these ballads and songs in
which persons of base and servile condition take pleasure’. 1 The
base and servile did indeed find opportunity in the ballads for
entertainment, but so did also men of rank and taste, though they
had no way of justifying their emotion. Ballads affected the noble
Sidney’s soul; he was ‘moved more than with a trumpet’, though
his schooling led him to note the absence of the ‘gorgeous elo¬
quence of Pindar’. Sidney^ sought for a ‘perfect Poesie’, like the
‘pure’ poetry which certain modern critics have demanded. But
this was not an objective which entered into the calculations of the
‘blind crowder’; it was enough for him to deliver his song to his
audience in a manner unknown to Pindar, and where Pindar’s
gorgeous eloquence would have been as unintelligible as thunder.

The publication of Percy’s Reliques stimulated a new interest in
these verses, by revealing to men of taste ballads which had been

2 Letter t0 Don Peter, ed. Prestage and Pastor, Oxford, 1927, p. 74.

P. Sidney, Defence of Poesie, London, 1595, sig. F.

112

THE ASCENT OF BALLADS

buried beneath the degenerate songs of the eighteenth century.
The critical aberration swung to the other extreme. In the Scottish
ballads Herder caught the authentic accents of a people, and he
demanded the same of his own German people. The Stimmen der
Voiker , the ballads, could do no wrong; they had the supreme
virtue of being natural. Confounding poetry of different kinds or
failing to discriminate between those that were similar, he included
Gongora’s polished songs among his ballads, and declared that ‘the
greatest singer of the Greeks, Homer, is also the greatest of folk
poets". Herder praises the naivety and childlike accent of the
ballads, their firmness, truth, liveliness, and assurance, their shud¬
dering tragedies and passionate music, together with their innate
nobleness: ‘the older they are, the more popular and lively, and in
the same measure the more bold and striking. 5 High and noble
their speech; great and mighty the folk who sang them. And then
he launches his impassioned appeal to the German folk—that vast
kingdom, the kingdom of ten peoples: ‘has the voice of your fathers
faded away, is it silent in the dust? 5 Herder’s passion sounded a
loftier note than his matter demanded; he exaggerated the achieve¬
ments of ballad poets. The inclusion qf heteroclite poems in his
collection led to such misprisions as F. W. Newman’s: ‘the style of
Homer is direct, popular, forcible, quaint, flowing, garrulous: in
all these respects it is similar to the old English ballad. 5 The
sentence stirred the formidable anger of Matthew Arnold. 1 He
declared that Homer is above all, noble 5 —a quality incompatible
with

Now Christ thee save, thou proud porter,
now Christ thee save and see,

and

While the tinker did dine, he had plenty of wine.

For this reason (he wrote) the ballad style and the ballad measure
are eminently /^appropriate to render Homer. Homer’s manner and
movement are always both noble and powerful: the ballad manner and
movement are often either jaunty and smart, so not noble ; or jog-trot
and humdrum, so not powerful.

Andrew Lang and W. P. Ker are critics who have not shared this
indignation against the ballad, but the general course of eulogy has
been turned aside. A writer of our own day has said:

I believe that the aesthetic value of popular poetry, if we are to
1 On translating Homer,

THE ASCENT OF BALLADS 113

measure by an absolute poetic standard, is habitually overrated, per¬
haps mainly from reasons of sentiment. It is forgotten that the master¬
pieces, like Sir Patrick Spens or Chevy Chase , are exceptional and rare
—how rare and how different to the ruck the perusal of a single volume
of Child should convince any unbiassed person. 1

‘If we are to measure by an absolute poetic standard’—and if,
perchance, such a standard exists. The classic and the critic
demand the application of such standards; but the search for a
poetry that shall be pure dissolves even the greatest works of art
into unrecognizable fragments. The Divine Comedy , considered
absolutely, becomes a cento of lyrics set in a long theological novel,
and the Lusiads will always remain one of the world’s greatest
poems by reason of its magnificent lyric flights’. The lyric alone is
short enough to sustain such scrutiny, and it would appear that the
modern aesthetic has just skill enough to appreciate that which our
waning poetical genius just suffices to produce—the artistic lyric.
The poetry of the commonalty we neither produce nor admire.
Those who, like Kipling and Watson, seek to express what many
men feel are the more lightly esteemed for their art; while, on the
other hand, the tribe of those who refine and polish and sublimate
their art is left to prophesy before a void. It is not so that much
of the world s greatest poetry has been composed, and it is not
the way of the ballad. The greatest poets have written neither to
extrovert their personalities nor to comply with the demands of
taste, but to voice the common thought of masses of men. So
Homer has been, in a sense, the voice of the Hellenes, and Virgil
of imperial Rome; Dante, Cervantes, and Shakespeare were each
the fruition of an age ; Camoes unburdened the ‘illustrious Lusi-
tanian breast’, and Spenser and Milton gave utterance to a Puritan-
ism, either sweetly reasonable or embattled and dogmatic. There
have been great poems which can be assessed as ‘pure poetry’, such
as the Orlando Furioso ; but to the men of the sixteenth century
Ariosto’s masterpiece seemed wanting in substance or seriousness.
We cannot be sure that it expresses something clamouring for
utterance—

Glory and generous shame,

the unconquerable mind, and Freedom’s holy flame.

But these are the themes of a ‘God-gifted organ-voice’, and these,
rather than absolute perfection, assure survival in literature. And

W. R. Halliday, Folk Studies , London, 1924, p. xi.

4615

114 the ascent of ballads

common, fundamental, moving themes of this kind inspire the best

of ballads.

The ballads are, and that is their best justification. Despite
changes of fashion and of language, they have clung to the people’s
memory with strange tenacity. This is not to claim for their genre
the quality of immortality; on the contrary, we have seen that they
have their begetting, and that printed literature is the sign of their
end. But they have clung to life, sometimes during four to seven
centuries, and that without any aid from courtly society (which has
favoured romances and troubadour lyrics), nor from the schools
(who have adored the ancient classics and are now embalming the
moderns), nor from official literature, contemptuous of such wild
snatches. Unwritten and traditional as they are, they owe their
survival to each individual singer and each individual audience.
They live only in the moment of performance. It has simply been
worth while to recreate them time and again throughout the cen¬
turies, since on each single occasion they have signified something
to their unlettered hearers and have moved them more than with a
trumpet. It is a glory not often achieved by the great artistic poets,
and when achieved, it is through some partial endowment of the
generous ballad simplicity. In Tasso’s case, for instance,

the fishermen’s wives of the Lido ... sat along the shore in the evenings
while the men were out fishing, and sang stanzas from Tasso and other
songs at the pitch of their voices, going on till each could distinguish the
responses of her own husband in the distance. 1

But how much of the Gerusalemme could be recovered from their
memories ? How faithfully modern ballad singers have preserved
versions of Danish and Castilian ballads unknown to the collectors
in the sixteenth century! In praise of Burns’s and Gil Vicente’s
gift of spontaneous song, it has been said that one cannot know
where the received traditional matter ends and the new creation
begins. The ballads have survived, fragile and imperfect as they
may seem, so long as the society for which they were created has
endured, and longer. The critic’s business is not to apply ‘a priori’
standards, but to look for the qualities which have justified their
amazing survival.

Ballads are to be accepted as true. Truth is, perhaps, not a'
quality demanded by the aesthete, but it is the necessary leaven of
1 Countess Martinengo Cesareseo, The Study of Folk Songs, London (Every-

THE ASCENT OF BALLADS 115

traditional narrative poetry, whether epic or ballad; and whether
in Plato s thought or that of Alfonso the Wise, the discovery that
the poet has indulged in fabling causes a sharp feeling of exaspera¬
tion. The audience for which the minstrel composes, and of which
he is a part, is not critical in the factual sense, and applies its own
peculiar standards of verisimilitude. One may note these standards
in the treatment of the supernatural: almost wholly absent from the
Castilian romancero’, this element is abundant in the ‘viser’ and
the Balkan ballads, because that is how nature seems to each. The
presence of nixes in a Castilian ballad or their absence from the
Scandinavian repertoire would seem equally untrue to nature; it
is part of our northern experience that winter and storm and raging '
waters are spirits bitterly hostile to human life. As evidence of the
truth of their ballads, the singers are wont to identify the places
where the events occurred. The Douglas Tragedy took place ‘way
back in Mutton Hollow’, as one old gentleman remembered,
through having been aware of it at the time. The notes to Grundt-
vigs Danmarks gamle Folkeviser have many such identifications.
One can describe, with some accuracy, the Jutish kingdom ruled
y Holger the Dane, and defended against Burmand and Diderik.
It does not matter that Holger was a figment of a French imagina-
tion and the Douglas Tragedy an offshoot of the Edda ; the point is
that as ballads, both were accepted as true. This sort of truth
reaches its maximum in the historical pieces, and gives them their
special importance. A sturdy and ancient balladry generally
springs from a stout historical trunk. When they travel from one
land to another, which is but seldom, they go as simple adventures;
ut at home theirs is a solid veracity, which educates the people.

1 heir tale has a definite importance. To the Castilian they explain
the chief mutations of Spanish history: the Moslem conquest, the
rivalry of Leon and Castile, the progress of the reconquest down to
its close. The Montenegrin, listening to the heroic story of Kosovo
and comparing it with the songs of outlawry, must have felt some¬
thing like the exaltation Tennyson attributed to Ulysses:

Though much is taken, much abides; and though
we are not now that strength which in old days
moved earth and heaven; that which we are, we are;
one equal temper of heroic hearts,
made weak by time and fate, but strong in will
to strive, to seek, to find, and not to yield.

n6 THE ASCENT OF BALLADS

One need not repeat the demonstration across Europe; nor repeat
the words said in a former chapter on the relation between balladry
and nationalism. The community Is defined and made conscious of
itself In the ballads, but not chauvinistic or exclusive. There are, of
course, certain simple antitheses which seem inevitable: Christians
and Moors, Scots and English, Serbs and Turks, holy Russians and
assorted pagans. But though there are enmities taken for granted
in the ballads, they are not exacerbated, and as they belong to the
order of things, they may pass when the order changes. There Is
no political programme involved in the definition of the communal
personality; the Influence of ballads depends on their mirroring,
undistorted, truth as that community knows it, both as to facts and
as to sentiments.

We must avoid the exaggeration of believing, however, that all
the popular mind finds an outlet in ballads. They are a form of
literature, though spoken, and they depend on precedent and
fashion. In different countries they have arisen at different times,
and in those whose ballads are late, we find often restrictions which
affect their quality or content. Theirs may be perceptibly more
vulgar than those of an older ballad centre, since the working of
humanism may be already evident. We note, for instance, the
absence of the high heroic vein of Kosovo in Bulgarian balladry;
in Bulgaria there are current Serbian ballads of Marko Kraljevic,
but the native ballads start from the lower level of haiduk song.
The Czech ballad of the Kudrun cycle is a tavern adventure; we
find this tavern level reached in Germany in the songs of lands¬
knechts and reiters, as a perceptible descent from the knightly
level of Linden,schmid, which is itself beneath the heroic level of
Kudrun. In the Ukraine, if we suppose that the Kiev ‘byliny’ were
ever indigenous to the region, there has been a complete change of
manner in the lyrical songs of to-day. In Lithuania the ballad is
lyrical, with a minimum of narrative, and its themes domestic and
sentimental. There are Lithuanian narratives, but their appointed
form is the prose Marchen. Convention requires the Frenchman
to sing of the ‘belle maumariee* and treat marriage as a jest; happy
marriage is not non-existent in France, but it is not a subject for
song. There is much that is unexpressed in a people’s balladry.
These songs strike an average of experience, but men are capable
of more than the average. The great poets and thinkers are also
voices of the peoples, and their highly personal inspirations also

THE ASCENT OF BALLADS 117

enter into tradition, in a different way, and shape the destinies of
the folk. One has only to think of those phrases of the Bible or
Shakespeare which leap to the mind and shape our reactions to
many situations; no ballad is more omnipresent than they. The
traditional ballads of Europe contain less than the full diapason of
its peoples’ voices; they lack the finer notes together with some of
the dominant chords.

When due allowance has been made, however, we return to
admire the ballads for their truth and universalism: c quod semper,
quod ubique, quod ab omnibus’. They have often descended from
older literary works, but they have transformed their models into
something more generously human. The traditional epics and the
Eddie poems imply a society precisely graded and bound by stiff
conventions. We may admire the words of Byrhtwold:

Thought shall be harder, heart the bolder,

courage the greater, as our might faileth ;

but we cannot do so within his own framework of relations proper
to lord and vassal. The scaffolding is taken down for ballad poetry.
Everything is generalized; there is the king and the lovers and the
enemy, and so on for all the typical actors. The situations and
emotions are all of the most general sort; such as can be taken for
granted. To survive in tradition the ballad must say what every
one would feel about situations which any one could expect, and in
language that all men understand and use. The complex imagin¬
ings of men of letters are thus simplified and universalized in
ballads, and new simple themes have been created. The whole
mass is direct, fundamental, moving, and vividly suggestive. When
men of letters have turned, in these last two centuries, to the ballads
for inspiration, they have found suggestions of action and conduct
free from cramping conventions and ready to glide into new
moulds.

The ballad style is unbound and universal, and the form is
singularly free. We cannot weigh and measure texts as we do with
written literature, since any given ballad may exist in many diver¬
gent texts. The study of variations is of supreme critical impor¬
tance, and no edition has scholarly value which does not resign the
pretension of restoring an archetype. It is not that the different
versions are of the same emotional value. Felicity resides in single
expressions and specified versions, and the love of ballads is best

THE ASCENT OF BALLADS

fed by selective editions such as the old collectors used to make,
and such as Sir Arthur Quiller-Couch has offered in the Oxford
Book of Ballads. Still, the ballad is not to be identified with its
happiest examples any more than with some supposititious arche¬
type, for Its real life lies inward. It lives in all its variants, exquisite
or crude, as an urge and pattern of creation. ‘Mens agitat molem.’
In some countries, such as Russia, the essential ballad is hard to
discern; it Is little more than suggestion of adventure and character,
a sketch which the singer fills out with his traditional resources.
He wishes not to innovate, but to remember; and it is seldom, as
Principal Halliday has remarked in the passage quoted above, that
any version is through and through admirable. The music falters;
it falters and lingers unfinished in the memory, which seeks the
chord that would close the song. Ballads are the ‘Capelas Imper-
feitas’, the unfinished chapels, of literature and challenge the great
word-builders to complete them. They have been completed by
the greatest, who have composed definitive 'ballad-like poems that
no one ventures to refashion; yet the broken accents and lisping
charm of the ballads remain, inviting new creation.

Free as the ballads are, however, in content and execution, the
Ballad is Form. In some countries this form is precise, invariable,
and almost tyrannous. It becomes a metrical pattern, like the
Castilian octosyllable, which imposes itself on all sorts of material,
extinguishing its predecessors. A country which develops a
romancero, a corporate sense of its ballads, usually develops a
precise metrical and stylistic technique. But apart from this
uni onnity to the eyes, ballad poetry has its inward formalism.
Human life becomes a pattern with an expected design running
through; experience has its proper rhythm. The adventure the
poet has to relate follows a prescribed course. If It be a Russian
a venture, it will as likely as not begin with a feast in Golden
adimir s court, reaching the point when the guests are flushed
with wine, and some hasty words are said which provoke the sally

Tf thCre al " VayS be remembe red forms to continue to
guide the minstrel in his narrative. You cannot fight any sort of

battle m a ballad, but only the ballad-kind of battle:

The first stroke that’s given, Sir Aldingar,

I will give unto thee,
and if the second give thou may,
look then thou spare not me.

THE ASCENT OF BALLADS 119

If one is to be married in Montenegro, the affair must be made to
look like bride-stealing; if one scuffles with a Turkish gendarme,
he must be a ‘black Arapin’.

Wine were drinking two good boon-companions . . .

Wine was drinking Kraljevicu Marko . . .

Wine were drinking three voivods of Serbia . . .

These are the traditional commonplaces, and there are many others.
They serve, doubtless, to ease the strain on the reciter’s memory
and to make fresh composition easy; but they also set the tempo for
adventure. The lines are beautiful, or at least efficient, in them¬
selves, and there is no need, nor is it desirable, to invent new
expressions for recurring experiences; it would be as unprofitable
as to write out in full a recurring decimal. By their recurrence they
impose a rhythm on the narrative like that which the Homeric
commonplaces give to the Odyssey. How much of the charm of
the Odyssey lies in the inevitable lines: ‘thence we sailed yet further
with our hearts full of grief, glad to have escaped our death, though
with loss of our dear companions’ ? They are not padding, but the
story itself. The joys and sorrows of the home-coming Greeks are
shaped to the pattern of the line

So sitting there in order due we smote the sounding furrows.

But the final word must be left to the poets, and their testimony
is abundant. The reticence and the frankness of the ballads
delight them; the unfinished music echoes in their minds, and they
are ever ready

to call up him who left half told
the story of Cambuscan bold.

They leave their own work rough-hewn, to achieve a like felicity.
Since the coming of Romanticism the debt of literature to the
ballad has been comparable to that of the Renaissance to the Greek
and Latin classics; the Renaissance demanded enrichment of style
or thought, Romanticism brought rejuvenation. The reckoning is
the same in so many countries that this chapter would become
unduly long, and the reader perplexed, if half the items were to be
entered. It must suffice to work out, by examples taken here and
there, the consequences of two publications: the Antwerp Cancio-
nero of circa 1545, and Percy’s Reliques in 1765.

A stout little volume emerged from the famous press of Plantin
at Antwerp, about the year 1545, containing the text of a number of

120

THE ASCENT OF BALLADS

Spanish ballads. The versions were, for the most part, cut short,
since the book would be used as the libretto of polyphonic con¬
certs. The nameless editor could not have anticipated that two
other collections would appear, a lustrum later, in Spain, and that
his venture would precipitate a landslide ending in the vast
Romancero General Nor did it stop there. The work of the six¬
teenth century was to collect and preserve the traditional poetry in
these volumes and on flying leaves, and to augment the mass by
verses carved by Timoneda, Sepulveda, the 'Caesarean Knight’,
and others from the general chronicles of Spain; but in the last
twenty years of that century and the first twenty of the next poets
of the highest order adorned and cultivated the genre. Gines Perez
de Hita poured new wine into old bottles, and fashioned the image
of Granadine chivalry, which dazzled Chateaubriand and Washing¬
ton Irving. Into sentimental ballads Lope de Vega poured his
abundant vein; Gongora, more artful, was 'simplex munditiis’ in
his Angelica andMedoro and the Spaniard in Oran; Quevedo made
the ballad malicious and jocose. With 'romances’, 'serranillas’ and
many kinds of songs, Castilian prosody could not fall entirely under
the yoke of Italy. The art forms of the Middle Ages were aban¬
doned readily; but beside those imported by Garcilaso and his
successors, there flourished the naive and spontaneous poetry of
the romances . Inspiration and emotion could fill many moulds.
Emotion is in constant ebb and flow in the drama, and when this
was mirrored by changes of metre in the Lopean Theatre, there
were roundels for conversation, and 'romances’ for the reports of
messengers; there were songs in simple stanzas, and only when the
poetic temperature rose steeply was it necessary (and natural) to
have recourse to the metres of Italy.

The Lopean Theatre leaned heavily on the ballads, which were
the sum of the nation’s memories. 1 The national epics, long since
extant only in the prose of forgotten chroniclers, survived frag-
mentarily, but universally, in epical ballads. Ballads of the
Carohngian and Arthurian cycles provided Cervantes and Calderon
with the better part of their knowledge of chivalry. Ballads of
mer the Cruel outlined the most tragic character known to Spain •
those of the Cid distinguished between the self-willed youth and
the prudent senior, who were the Don Juan and Don Quixote of

Paris?i^o nfedeZ Pida1 ’ L ' ep0pie Castillane d travers 1 “ Literature espagnole,

THE ASCENT OF BALLADS 121

the Middle Ages. Frontier ballads, splashed with intense colour,
represented the dramatic antithesis of Moors and Christians.
Morris ballads, chivalrous and sentimental, expressed the single¬
ness of so many amorous hearts that the sound of

Listen, Zaide, as I advise you,

don’t go strolling down my street

would have prevented any real Zaide from strolling in any of the
streets of Madrid. Once Juan de la Cueva had drawn on ballad
material, though not ballad style, for his first national drama, the
Siege of Zamora , the transfusion of ballads into drama proceeded
rapidly. Lope de Vega built up an entire dramatic history of Spain
from chronicles and ballads, interweaving ballad phrases into his
verses in such a way that the original cannot be discerned from the
new. He could do so the more readily because the best Castilian
ballads cut short the narrative, leaving only the dialogue. The
breathless dialogue between Ruy Velazquez and the avenging
Mudarra is more dramatic than the more diffuse treatment of the
same incident in Lope’s Bastardo Mudarra. Calderon had not the
same appreciation of the epico-dramatic element in balladry; to
him ballads seemed to provide a simple conversational metre like
the roundel. He slips from one to the other as his topics change.
The conversational use of the ‘romance’ increased so much that
Leandro Fernandez de Moratin used it as the unique metre of his
Old Man and Girl, towards the end of the eighteenth century. To
do so was a notable impoverishment of the resources of poetical
drama, which he abandoned in time; but it was eloquent evidence
of the vivacity of ballads.

The ballads of the younger Cid worked over materials taken from
the lost epic of the Mocedades. The Mocedades had spoken of the
feud between Diego Lainez and the father of Jimena. The boy
Ruy Diaz had taken on himself his father’s quarrel, killed the
count, and harried the lands and washerwomen of the orphan
heiress, until she could see no way of protecting her patrimony
save by marrying her invincible enemy. She appealed to the king,
and the king commanded the marriage. Ruy Diaz, in his flaming
indignation, swore not to honour his wife until he had won five
pitched battles, which the straggling epos proceeds to recount. But
one of the ballads put a different complexion on the matter. Mind¬
ful that Ruy Diaz had been brought up with the Castilian princes

4615 „

I22 THE ASCENT OF BALLADS

and therefore should hurt none of them, the ballad depicts th
Princess Urraca in the act of reproaching him for leading he
brother’s army against her stronghold of Zamora. Such was thei
intimacy, she says, that he might have married her and gained
high estate, had he not preferred Jimena and wealth. The Cid ha<
chosen Jimena! But if so, the Cid had killed the father of the lad
he loved. The ballads do not speak of the conflict between lov
and honour, but they provide the datum which Guillen de Castr
was prompt to seize. His Mocedades del Cid is, in places, a cent
of ballads, and he is compelled, for verisimilitude, to repeat ballad
which are not allied to his main theme. The latter is the conflict c
love and honour indicated by the ballad. Honour triumph
speedily in Ruy Diaz, though it tears his heart; for honour was a3
obligation upon men which could not be gainsaid. With womei
it was rather different, since their reputation was defended by thei
males, and only the most virile of women were expected to take uj
the obligation of washing out dishonour in blood. Jimena is
virile woman; but she is a woman none the less, so that with hei
after a decent resistance, love prevails. The idea of the dramati
does not rest, for Spain, in such conflicts. The Mocedades was
sport of Castro’s peculiar temperament, but in passing into Frenc
as Le Cid of Corneille the plot, made more intellectual and abstracl
resolved itself into the conflict of duties which became the formul
of Racinian tragedy.

Cervantes also was a ballad-lover, and in his ‘romances’ was a
near as ever he was to being a poet. He knew no more of the Arthur
ian legend than three ballads told him; and though he had rea<
Italian epics and Castilian chapbooks full of Carolingian matter, th
themes nearest to his mind were those contained in the ballads of th
Marquis of Mantua, Durandarte, Gaiferos, and Montesinos. Whs
more natural, then, than to sit down and begin a short ‘exemplar
novel’ on the chivalric mania by depicting a country squire ridin
out to redress wrongs like Ogier from Mantua ? It seems only t
have been the afterthought that no knight lacked his squire th*
caused Don Quixote to turn back from his first sally. The Knigt
of the Sorrowful Countenance returned to the midst of one c
those didactic chapters with which Cervantes, the greatest of peda
gogues, could not help interlarding his works. A criticism of mis
cellaneous novels and romances served to deflect the course c
the parody; and when Sancho joined his master, their adventure

THE ASCENT OF BALLADS 123

were patterned on those of Amadis de Gaula. The ballads, no
longer the main butt of Cervantine irony, persist through the great
novel as a lingering echo; and it is out of the ballads that Cervantes
fashions the central episode of his Second Part. 1

Shakespeare, like Cervantes in so many ways, enjoyed his ballad.
He quoted

Some men for sudden joy do weep,
and some in sorrow sing,

from the 'Godly and virtuous song or Ballad made by the constant
member of Christ, John Careless, being in prison in the King’s
Bench for professing His word; who, ending his days therein, was
thrown out and buried most ignominiously upon a dunghill, by the
adversaries of God’s word 5 . Making the fullest possible use of the
amatory, journalistic, satirical, and pious ballads that poured from
Elizabethan presses, 2 Shakespeare, not having the fortune to be a
Scot, rarely laid hands on such fine traditional matter as Child
Roland to the dark tower came? A mass of traditional and semi-
popular poetry was gathered into a folio manuscript during the
reign of Charles I, but not published. In 1765 Thomas Percy pub¬
lished extracts from this folio along with Scottish and Northumbrian
ballads which were communicated to him, thus suddenly revealing
the wealth of the tradition buried under the journeyman produc¬
tions of the seventeenth and eighteenth centuries. The effect was
electrical. It was felt first, and most strongly, in Germany, where
Hanover was at that time part of the English domain. 4 In the
university of Gottingen, Holty indoctrinated his pupils with a new
conception of poetry, basing his style on Percy. This constituted,
according to Kayser, 'the foundation of the serious ballad 5 in Ger¬
many, and a coping stone was soon found in Burger’s Lenore
Despite the excessive literalness of its opening verses and the
sensationalism of its 'hurre, hurre, hopp, hopp, hopp 5 , this trans¬
formation of a Low German folk-song under the influence of
Sweet William 1 s Ghost had no small measure of the forthrightness
and vigour of the ballad, together with its atmosphere of super-

1 E- Menendez Pidal, Un Aspecto en la elaboration del Quijote, Madrid, 1920.

2 Sir C. Firth, ‘Ballads and Broadsides’, Essays , Oxford, 1938; ‘Ballads that
illustrate Shakespeare', Percy’s Reliques, Book the Second.

3 It is an offshoot of the Scandinavian Rosmer Havmand.

4 This matter is fully discussed in W. Kayser, Geschichte der deutscken
Ballade, Berlin, 1936, and in F. Arnold, Das deutsche Volkslied, Prenzlau, 1027
(4th ed.).

124 THE ASCENT of ballads

natural dread. Denote dike wildfire swept across Europe, from
Scotland to Poland and Russia, from Scandinavia to Italy . l
Burger devoted himself to the ballad, and drew from his studies
far-reaching conclusions. He found in folk-poetry an overflowing
of the heart (‘Herzensausguss’), needed to redeem German verse
from its false erudition. The supreme panacea was contact with
nature and the folk; poetry is a gift the poet offers to all men, not
merely to an elite. Folk-songs are the essence of poetry, and the
more elevated lyrics are justified only when they are Volkstiimlich’.
He insisted on the importance of fantasy and invention, and
demanded a German collection to be placed on the same shelf as
Percy.

The German collection was already in the making, since Goethe,
with Herder’s encouragement, w T as busy in Alsace collecting songs
and gaining for himself the gift of song. It is because of its charac¬
teristic freshness and even naivety that the German lyric transcends
other forms of German literature, and in Goethe, even when most
artful, there is a living fountain of spontaneous song. Meanwhile
Herder was compiling his Volkslieder , issued in 1778-9, and later
entitled Stimmen der Volker in Liedern. The bulk of the book con¬
sists of translations from Percy, including some Shakespearian
songs, together with Spanish ‘romances’ of the Morisco type
(under Percy’s influence) and seven lyrics from Gongora. The
German songs of his fifth book are not authentically traditional,
but rather showed what might be done in this field with a closer
study. The Scandinavian North is represented by Scaldic and
Eddie verse, with four Danish ‘viser’, and there was enough to call
attention to the Lithuanian ‘dainos’ and the ‘Morlakian’ songs of
the Serbs. Seventy ballads on the Cid (Der Cid nach spanischen
Romanzen, 1805) completed the range of examples on which
Herder based his impassioned appeal for a new German poetry.
Herder internationalized Percy. The appeal for German verse was
splendidly answered by Goethe’s Erlkonig (based on the Danish
Elveskud, which is also the source of Leconte de Lisle’s Les Elfes)
and his Komg in Thule. The perfect phrasing of these pieces would
not outlive traditional variations, but in every other respect they
are ballads: in speed and felicity, abrupt dramatism, and the magic
of words unspoken. The honest Schiller was less happy. He aimed
at a more elaborate style, in which short narratives taken from
1 J* G. Robertson, History of German Literature, London, 1931, p. 308.

THE ASCENT OF BALLADS 125

antiquity or more recent eras, received their just rhetorical develop-
ment. His Handschuh is happier than his Toucher , since it is the
less pompous. The elaboration and theatricality of the latter are
an ill-fitting frame for a very simple anecdote. Goethe cast an eye
on ‘Morlakian’ balladry also, and Wilhelm Muller’s Lieder der
Gnechen (1821-6) made the style of the ‘tragoudia’ accessible to
Germans. 1

The appearance of Arnim and Brentano’s Des Knaben Wunder-
horn in 1805-8 opened a new era, by giving a greater precision to
the notion ‘ballad’. Despite a few foreign pieces like Herr Olof
(Elveskud again), the collection is genuinely German. Herder and
Goethe’s internationalism has gone, and in its place there are a
greater number of traditional German songs, with fewer erudite
intrusions. The emphasis is on narrative. The Wimderhorn thus
comes to define the notion ‘Ballade’, as a short narrative poem
touched with lyrical imagination, and so distinct from the purely
narrative ‘Romanze’. A third term, ‘Lied’, indicated a third tradi¬
tional influence on the German artistic lyric. The influence of
Brentano and Arnim was all the more effective as, like Scott and
Almeida Garrett, they were not primarily concerned to produce
faithful texts. Their pieces had the simplicity of ballads, without
their imperfections. Thus a schism arose between the recon¬
structed and the palaeographic way of presenting this matter, like
that which had already arisen with Ritson in England. The
demands of scholars were more exacting, especially after Grundt-
vig’s great Danish collection began to appear, and the critical style
of editing appears tentatively in Uhland, to be confirmed by Brk,
Bohme, Liliencron, John Meier. On the other hand, the ballad as
poetry has descended through Uhland and Heine to Blunck in our
own age. Uhland’s Balladen und Romanzen y mostly composed
between 1805 and 1815, were of a more pronounced narrative cast
than those of Goethe, more objective and dispassionate. They
do not sing themselves so well, but they have learned to avoid
the sensationalism of Schiller. The styles employed are the
lyrical narrative and the straight narrative, the latter without any
second intention. It is in supplying this second intention, in Die
spanischen Atreiden and other pieces, that Heine raises the
‘Romanze’ to its height of perfection. Under the surface of a

1 A convenient anthology illustrating these points is J. T. Hatfield’s German
Lyrics and Ballads , New York, 1900, 1924.

126 THE ASCENT OF BALLADS

narrative, apparently of the strictest neutrality, Heine clears a
channel for his lambent irony. The poem and its congeners are
not ‘volkstiimlich’; it is not right that poetry not destined for
traditional preservation should ape all the characteristics of the
traditional style. There must be a new creation into something
better and more definitive, whether in form or intention. Heine
discovered that the neutral manner of the ‘Romanze’ is unrivalled
as a means of letting the terrible things about men and life express
themselves. More recently, G. Duhamel has done the same. Tor¬
mented by the sight of useless suffering in the military hospitals
he frequented during the War, Duhamel could find no epithets or
rhetoric which did not diminish the horror of the real thing. The
stark objectivity of his couplets in the Ballade ofFlorentin Prunier , 1
without ornament or sentimental disguise, rivets our horrified
gaze on the tragedy itself. Any adjectivation would have afforded
a relief which the poet was determined to refuse. So too with
Heine, though he is more prone to disgust than to horror. In
Heine, on the other hand, pure song reaches a second perfection,
as in his Pine and Palm, inspired by similar allegorical folk-songs
about trees.

It is not capricious to take Germany on our road to considering
the effect of Percy on English poets, since our Romantics were
affected by the German enthusiasm. Southey saw in The Ancient
Manner a ‘Dutch attempt at German sublimity’—to such an
extent had a sublimity characteristically English (or rather Scottish)
come to appear German. Scott, though he collected the Border
Minstrelsy , was also the translator of Burger’s Lenore , Goethe’s
Erlkonig, and the traditional Sempach and Moringer. It was as a
collector and imitator of ancient ballads that Scott set up business
as a man of letters, developing thence into an author of ‘lays’, and
so into a novelist; but the incidental verses in the novels show that
he had not lost his first love. If the ballads have, in a measure, sug-
gested the historical novel, they also, in another measure, initiated
the Romantic revival of the lyric. The Lyrical Ballads owe little
to specific folk-songs, but they aim at attaining two of the excel¬
lences to be found in balladry: the imaginative presentation of
ordinary experience, and the humanizing of the supernatural. The
influence of individual pieces is much less in England than in
Germany, though one may mention Lockhart’s renderings of
1 Elegies, igzo. M. Duhamel mentioned these poems to me himself.

THE ASCENT OF BALLADS 127

Spanish ‘romances’, Byron’s Alhama, , and Arnold’s Forsaken Mer¬
man. The ballad manner has been imitated, at a distance from the
authentic style as revealed by Child, in Campbell’s political odes
(notably Hohenlinden ), Tennyson’s Charge of the Light Brigade and
Revenge , Macaulay’s Armada (and Chesterton’s Lepanto ), and
Kipling’s barrack-room ditties. Swinburne is an unexpected
example to the contrary effect. It is hard to imagine any style more
remote from the plain-speaking ballad manner, yet Swinburne is
very closely attentive to ballad form in such pieces as After Death
and The Sea-swallows , while in Laus Veneris he inserts a paraphrase
of Tannhduser:

Then came each man and worshipped at his knees
who in the Lord God’s likeness bears the keys

to bind or loose, and called on Christ’s shed blood,
and so the sweet-souled father gave him ease.

But when I came I fell down at his feet,
saying, Father, though the Lord’s blood be right sweet,
the spot it takes not off the panther’s skin,
nor shall an Ethiop’s stain be bleached with it.

f Lo, I have sinned and have spat out at God,
wherefore his hand is heavier and his rod

more sharp because of mine exceeding sin,
and all his raiment redder than bright blood

‘Before mine eyes; yea, for my sake I wot
the heat of hell is waxen seven times hot

_ through my great sin.’ Then spake he some sweet word,
giving me cheer; which thing availed me not,

Yea, scarce I wist if such indeed were said;
for when I ceased—lo, as one newly dead
who hears a great cry out of hell, I heard
the crying of his voice across my head.

‘Until this dry shred staff, that hath no whit
of leaf or bark, bear blossom and smell sweet,
seek thou not any mercy in God’s sight,
for so long shalt thou be cast out from it.’

In French literature it is possible to make a case for the interest
of poets from an early epoch in the ‘chanson populaire’; 1 but the

1 De Beaurepaire-Froment, Bibliographie des Chants populaires franpais,
Paris, 1910, gives quotations from authors ranging from the thirteenth to the
eighteenth centuries.

123 THE ASCENT OF BALLADS

consequences of the ballad revival for nineteenth-century literature
are less important than in Germany or England. There is, of
course, the exuberant Hugo. He has his Odes et Ballades , of which
the fifteen poems under the latter title are neither ballades nor
ballads, but rhetorical lyrico-narratives in Victor Hugo’s own
manner. La Chasse dn Burgrave , a tale cast in the mould of the
Renaissance echo songs, even inspired some hesitation in its author,
as being perhaps ‘somewhat too Gothic in form’. Les Orientales
(1829) contains matter more to our purpose. Dedicated to the
cause of Greek independence, these pieces mention Botzaris and
other heroes whose feats are celebrated in ‘tragoudia’, and, after
Byron, he retails the Mazeppa legend, for which there are popular
parallels in the Ukraine. The direct imitation of ballads, however,
is limited to La Bataille perdue and Romance mauresque , both of
them offerings on the altar of his peculiar devotion to Spain. The
former is the ballad on King Roderick’s defeat, made more
rhetorical; the latter is a flaccid version of the dramatic ballad of
Mudarra’s vengeance. The savagely intense Castilian song has
been converted into an arabesque, and the local colour is so thickly
splashed that it sometimes cries out. At least to a Spanish ear it
must seem strange that Mudarra, ‘who commands a frigate of the
Moorish king Aliatar’, should have been chased from Alba to
Zamora in the central Meseta. The devotion of Gerard de Nerval
to the ‘chanson populaire’ was both more national and more
natural than this. But in France, as in England, there is one sur¬
prise. Leconte de Lisle 1 exercised his Parnassian genius for pure
form on the matter of the Danish Elveskud, the Swedish Sorrow's
Might (Christine) and half a dozen Castilian ‘romances’. 2 From
these ‘rudes poesies’, as Vianey dubs them, the Parnassian could
learn little but his own doctrine of objectivity; apart from that,
their contents were mere raw materials, on a par with Greek
myths and Vedic hymns. They gave him, however, the vigour
and initiative which throbs beneath the classic perfection of his
polish.

In Spain the ‘romance’ flourished among poets once its prestige
abroad was sufficiently understood. It produced an offshoot, the

1 J. Vianey, Les Sources de Leconte de Lisle, Montpellier, 1907.

* From Damas Hinard, Romancero general, Paris, 1844. The poems are La
Tete du Comte, based on the Cid’s vengeance for his father, Ximena , L*Accident
de Don Inigo (also a Cid ballad), Les Inquietudes de Don Simuel, La Romance de
Don Fadrique , and La Romance de Dona Blanca.

THE ASCENT OF BALLADS 129

‘leyenda’. The ‘legend’ is a poem in ballad lines, but of a more
definite structure; it differs from true ballads chiefly in the eager-
ness to give ‘local colour’, which the traditional ballad always takes
for granted. Other countries with rich balladries inevitably show
signs of their influence. A bird’s-eye view of that aspect of Danish
literature can be obtained readily in anthologies or by scanning the
works of Oehlenschlager. 1 In Serbia we have the poems of Njegos
and adaptations of the ‘junacke pesme’ like Stojkovic’s Lazarica.
The publication of folk-songs opened a new epoch in both the
poetry and the music of Hungary, while in the Baltic states they
revealed the very possibility of a literature inspired by national
sentiment. We must pass over these things, which would delay our
discussion too long; but we must not pass over the case of Russia,
both because its novelists have raised its literature to the front rank,
and because the history of ballad influences has run a peculiar
course. 2 The impulse came at first from abroad, chiefly from
Germany. The first of the ballads was actually the Castilian
Guarinos , translated by Karamzin in 1789, and published in 1792;
but the most potent single influence was certainly Burger’s Lenore .
Lenore appeared as the Ljudmila of Zukovskii in 1808, and as his
Svetlana in 1813; as Katenin’s Ol’ga in 18x6 (a plain translation)
and Lermontov’s Ljubov ’ mertveca in 1840; while Anna Turcani-
nova rendered into Russian our Margaret’s Ghost ( Vil’jam i Mar¬
garita) in 1800. Percy’s Reliques offered to Russian poets the
themes of Edward and the Three Ravens. In addition to these
sources, Goethe’s Erlkonig and Konig in Thule were well known,
along with Schiller, Heine, Scott, and Campbell. Herder’s Cid
provided Spanish information, and there was direct contact with
Denmark and also with Greece (Maikov’s Borba so smert’ju, the
Greek battle of Digenis and Charon). Russian prosody was settled
on the basis of rhyme and measured lines, and Russian taste
educated in western types of balladiy before the revelation by
Rybnikov and Gil’ferding of the rich store of indigenous ‘byliny’.
In the earlier period the matter of Kiev is deemed either comic or
more appropriate to prose genres. It is thus only at a compara¬
tively late moment that the Russian ‘byliny’ come into their own

1 The Oxford Book of Scandinavian Verse unfortunately omits the ‘viser*. The
comparison can be made on consulting O. Vig, Sange og Rim, Christiania, 1854.

z F. W. Neumann, Geschichte der russischen Ballade, Konigsberg und Berlin,
* 937 *

4615

I3 o THE ASCENT OF BALLADS

kingdom, with Aleksei Tolstoi’s Aljosa Popovic and Sadko (1872).
By contrast, Puskin had much earlier drawn on Ukrainian songs
for his Kozak (1814), and the music of the Russian folk has been
employed by Mussgorskii, Rimskii-Korsakov, and Borodin. There
is perhaps a closer affinity between the gorgeously extravagant
‘byliny' and opera librettos, than with the more disciplined taste
of modern poetry. If that be so, the apotheosis of the genre is
Borodin’s Sadko, where the most musically worded of ‘byliny’
becomes opera.

The past tense has been used so often in these pages that it may
be supposed the ballads have no present. The present day is
analytical and introspective; it has had a surfeit of romantic spon¬
taneity and would willingly reimpose the criterion of poetic artifice.
An example may suggest that the broken arches of the ballads will
bear the new superstructures we would like to rear on them. It is
the Spanish Count Arnaldos. Count Arnaldos, strolling along the
sea-shore one midsummer morning, saw a wonderful boat ap¬
proaching and heard its steersman singing a song which brought
the birds to rest on the mast-head and the fish to swim on the
crests of the waves. Deeply moved the Count cried out:

Tn the name of God, I pray you,

teach me, shipman, what you sing.’

But the steersman would not; the ballad breaks off abruptly. What
was that song? Longfellow’s answer is: The Secret of the Sea —

‘Wouldst thou’, so the helmsman answered,

‘learn the secret of the sea ?

Only those who brave its dangers
comprehend its mystery?’

The answer was worthy of a New Englander, with the hardy
sailors of the clippers for his compatriots. Azorin, in Al Margen
de los Clasicos, is content with hints. He says that the sailor’s voice
‘speaks of contentment, enfranchisement, geniality, health, hope’;
it may be he comes from beyond the infinite and stormy ocean to
lead us to dreamlands of illusion. I do not know whether A. E.
Housman had read the Spanish ballad or Longfellow’s or Flecker’s
rendering. 1 He used, however, exactly the same picture to dash our
hopes of an answer. The time is evening; the place a headland in

1 Collected Poms, London, 1935 (‘Lord Arnaldos’): a direct translation.

THE ASCENT OF BALLADS 131

Biscay. The poet is not alone, but accompanied by Grief. The rest
is in conformity with the tradition. The ship is of gold—

gold of mast and gold of cordage,
gold of sail to sight was she;—

lacking the softness of silk which mitigates the gold in Spanish.
The sailor, too, has lost the magic of Arion and Orpheus, but comes
a-hailing:

On the golden deck the steersman
standing at the helm of gold.

Man and ship and sky and water
burning in a single flame;
and the mariner of Ocean
he was calling as he came:
from the highway of the sunset
he was shouting on the sea,

‘Landsman of the land of Biscay,
have you help for grief and me?’

When I heard I did not answer,

I stood mute and shook my head:
son of earth and son of Ocean,
much we thought and nothing said.

Grief and I abode the nightfall;

to the sunset grief and he
turned them from the land of Biscay
on the waters of the sea.

(Last Poems, London, 1936.)

BOOK II

BALLADS IN PARTICULAR

ROMANCE BALLADS

i. France , Provence, North Italy , and Brittany

T HE ballad In France is the narrative aspect of lyrical poetry.

It has no form of its own nor does it make a corpus to which
the word Romancero (borrowed from Spain) might justly be
applied. The narrative pieces merge into the more general
‘chansons populates’, from which, except when borrowing from
abroad has taken place, they are scarcely distinguishable. Lyrical
effusions arose from situations which were constantly repeated—•
an encounter with a shepherdess, amorous intrigue, separation and
reunion—and these situations required some words of introduction.
They are not domestic and personal; they do not, as in Lithuania,
follow a girl through the ritual prescribed by custom for each stage
in her development. On the contrary, the matter of the ‘chanson
populaire’ is the same set of conventions which occurs in the liter¬
ate effusions of the troubadours and trouveres. Contact with
written songs Is felt in all the folk-songs of France in a greater or
less degree; there is no possibility of setting a dividing line between
the literate and the oral. What we possess of the latter is more
recent in date than the former, and we need not hesitate to suppose
that it Is modelled on the conventions of the courtly poets. The
troubadours, however, had their antecedents, which are unknown
to us. About the year noo there was a quickening of the spirit
along the dividing line between the ‘langue d’oc and the ‘langue
d’oiT, and Provencal poetry burst into green leaf and full bloom.
The causes of this sudden efflorescence are still concealed. What
is certain is that the Provencal lyric appears at once with an
elaborate technique and equipment of conventions, far removed
from any poetry that might be called primitive. On the other
hand, long before noo we read of the choric songs of rustic
women, either for mere entertainment or on some momentous
occasion. They perplexed the pious by their ‘diabolic songs’,
though doubtless innocent enough, from the sixth century

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 133

onward. So little is known of their themes and style that the
citations serve equally well to supply the prehistory of the epic,
ballad, and lyric genres. The repertoire must have been created
and preserved orally, and must have exercised some influence on
subsequent artistic poetry. It was an oral, traditional literature
of women; sung by women and doubtless shaped for them.
When the lyric at last appears in written form, the predomin-
anee of women’s interests is strongly marked, though the poets
are men. This preponderance descends throughout all French
literature, and has spread from there into the other literatures of
Europe.

Love is the single chord of the modern lyric, and woman the
focus of attraction. Other themes arise in literate verse, though
always in a minority of examples; but in the folk-song the other
interests, apart from religious and drinking songs, are so few as to
be negligible. The chief distinction of types is according as we
picture the women dancing or sitting over their embroidery. The
former gives the dancing-songs (‘chansons a danser’), equipped
with refrains which the dancers, halting, could sing in chorus;
the latter gave the ancient ‘chansons de toile’, for which, more
recently, scholars have preferred the enigmatic term ‘complaintes’.
The complainte’ is without refrain, and is sung by a soloist, not
a precentor and chorus. The word covers, more or less adequately,
many different types of songs named from the convention which
they employed: there are ‘pastourelles’ or ‘bergeries’, songs of the
faithless wife or ‘chansons de mal mariees’, ‘reverdies’ in which
mention is made of the viridescence of Spring, songs of villeins,
songs addressing the nightingale (rossignolet) as the messenger of
love, songs beginning with the mention of some female operation,
&c. These conventions, as they affected the semi-popular poets
of the twelfth and thirteenth centuries, are seen exploited in
Bartsch’s collection of Romances et Pastourelles . 1 The doubt arises
whether many of the pieces have been preserved traditionally.
Mere anonymity is no proof; the output of the named poet Aude-
froi le Batard is noticeably more popular in tone than much of the
anonymous verse. It is not the single poems which need be sup¬
posed traditional, but the types; types which are repeated not only
4 ad nauseam’ in the French song-books, but also in all the ballad¬
ries of western Europe.

1 Leipzig, 1870.

134 ROMANCE BALLADS

They are the moulds and patterns of traditional narrative
lyrics in many lands. The ‘pastourelle’ opens with a glimpse
of the knight or squire riding along some particular stretch of

road:

Twixt Arras and Douay
outside GravelJe,

I, riding on my way,
found Perrennelle.

He makes his offers of love; the shepherdess pertly replies. He
presses his suit, and she yields or evades him, or, in songs which
may be supposed to appeal to the vulgar, calls in some villeins to
beat the gentleman off. Sometimes the damsel tricks him into
doing a service, such as killing a wolf, and repays him with no more
than good words ; at others she frightens him or holds him off until
it is too late, and then mocks him for not taking advantage of his
chances. The early songs were satisfied with these personages; but
in the fifteenth century and later it was licit to mark the lady’s
lowly status by assigning to her a trade, as in La Belle Barbiere.
In French verse and prose, as distinct from French life, courtship
and marriage are treated in a lightly sardonic vein, and infidelity
is accepted as the norm. Here enters the vast tribe of ‘chansons de
la mal mariee’. Married to a rich, twisted old man who is impotent
in love, or to a boor who beats them, the young matrons are advised
by servants and friends to look for a young lover; they complain
to some sympathetic squire, give assignations, open their windows
or a side door at night, exult in infidelity and defy the consequences.
To be a nun is, from the standpoint of popular verse, to be un¬
happily unwed; we have therefore the frequent pattern of the nun
who laments her solitude. The dawn-song or ‘aubade’ is yet
another form of the same convention. It is because the lady is
already ill-matched that dawn forces the lovers to separate. The
convention is complete in all its details: the birds’ songs announce
the dawn or the watchman blows his trumpet, the lovers com¬
plain that the night has been short, and they abuse The jealous
one—the unfortunate husband. In Spain the dawn breaks
and brings ingenuous lovers together; but that is not the
convention of the ‘aubade’ nor of the German ‘Tageslied’.
Tannhauser, the Noble Moringer , and the Pyramus ballad are
notable instances of songs cast in a traditional mould formed
in France.

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 135

The ‘reverdies’ are less apt to give narrative details. They are
characterized by the beginnings only:

In the fresh sweet month of May,
when the fields are green and gay,
beneath a bush I heard a tale
sung me by a nightingale.

Saderala bon!

How sweet to slumber on
beneath the bushes in the dale.

(Bartsch 27.)

There follows an idyllic scene of a pretty girl in a garden talking
to the nightingale, and perhaps also to thrush, finch, lark, and
merlin:

and other birds, the shrubs within,
heard I singing in their Latin.

(Bartsch 30 a.)

It was not so much-in the lyrics as in the epics that there developed
conventional description of the coming of Spring which has
ciiculated to other parts of Europe. Spring, for the epic minstrel,
is the season when kings go forth to war; he feels it must be duly
marked.. An order of development was established and used with
permissible variations: the grass grows green, buds appear, birds
build nests, bulls and horses frisk, and the lover seeks his lass.
The gambit may be raised to exquisite poetry, as by Chaucer in
the Canterbury Tales, or left in its significant simplicity, as in the
Castilian ballad of The Month of May; the order and details are
respected. The convention spreads as far as Greece, where the
modem Swallow Song is of this type (a description of the burgeon-
ing of nature). In the Ukraine the Spring song (vesna) is an
established type of folk-song, but the details are adapted to the
severer conditions of a Russian spring. The part allotted to the
birds of love makes the ‘rossignolet’ a synonym for the lover.
From this point it is a short step to the delightful little allegories
we find in German traditional poetry ( Vogelhochzeit ).

The narrative lyrics of Audefroi le Batard are especially rich in
traditional themes worked out in a semi-popular style. By feigning
death Belle Ysahiaus contrives to join Gerard; the theme is much
used in later balladry, and the device seems to be Italian in origin.
Belle * Ydoine is an imprisoned princess, rescued by her lover Count
Garsile; this is the later Pernette. Belle Beatrice was snatched from

136 ROMANCE BALLADS

an unwelcome bridegroom by Hugues, a young Lochinvar; Belle
Argentine , as sorely abused as any medieval wife and mother, was
justified at long last. The Count de la Marche exploits the theme
of the £ shepherdess of joyous heart 5 who mocks the timid gallant:

Oh how much of worth you miss,
since without one little kiss
I’ve eluded you!

(Bartsch ill. iii.)

Anonymous tales of this sort open Bartsch’s collection. Belle
Eremhors resumes her old affair with Raynaud; Belle Aiglentine is
rewarded for her true love to Count Henry by marriage; Belle
Doeite hears that Doon is dead, refuses a comforter, and enters a
nunnery (Marlbrough s'en va en guerre is the same story in another
style, and the various tests of ladies’ faithfulness may be expansions
of the same theme). There are encounters at fountains and any
number of £ mal marieesh Different in style from what are later
admitted to be ‘chansons populaires 5 , more literary and perhaps
set to more intricate music, these poems contain the seed of many
future ballad developments in Spain, England, Holland, and
Germany.

In the second half of the fifteenth century, perhaps due to an
uprush of national and popular sentiment at the close of the
Hundred A ears War, 1 a current set in in favour of simplicity of
music and rhythm, and we encounter for the first time poems of
the same texture as those preserved in actual oral tradition.

Ballads and rondeaux give way gradually to livelier and lighter pieces.
To pompous and precious phrases more natural and realistic turns and
expressions are preferred. The style of composers also changes: com¬
positions for voices accompanied by instruments no longer predominate *
the vocal cl capella style is established with Ockeghem and his school. 5
In the sixteenth century, in particular, the form is simplified, and,
perhaps under the influence of the Italian frottole , lyrics approximate
to the forms of the popular dances, or even to the songs of the twelfth
and thirteenth centuries. 2

This simplification of style is applied, characteristically in France,
to artistic lyrics. There remains the same doubt as before concern¬
ing the traditional nature of a large part of the albums in which

1 In several songs (as Le Capitaine et la Belle)

2 Th. Gerold, Chansons populaires des XV e
Romanica, 190-2), Strasbourg, n.d., p. vii.

the villain is an Englishman.
et XVI e si&cles (Bibliotheca

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 137

these simple songs occur; and there is no absolute line of division
between the literate and the oral styles. Certain forms define
themselves for the first time, and their spread reveals the com-
munity of France, Provence, North Italy, and the older Catalan
stratum in respect of subject and technique. An important group
of sources for our knowledge of the ‘chanson populaire’ is consti-
tuted by song-books printed in northern Italy during the earlier
years of the sixteenth century; notably O. Petrutius’s Harmonice
mtisices Odhecaton ("Venice, 1504). This new manner is more
closely defined by the fact that it takes no account of the innova-
tions due to humanism. Humanistic lyrics have attained a real
popularity of their sort, but they have not entered, nor are suited
to enter into oral tradition; and consequently collections made in
the nineteenth century show no such admixture in the ‘chanson
populaire’. 1

t There is a considerable variety of stanzas employed by the
chanson populaire’, but some simple types are of special impor¬
tance. The poems may be divided into ‘chansons a danser’ and
complaintes , according as they have refrains and are suited to
choric performance with a precentor, or are fit for recitative singing
only. What seems the older and simpler manner may be repre¬
sented in this translation of La Perronnelle , one of the most popular
of all such songs :

Haven’t you seen Miss Perronnelle
whom the troops have ta’en away ?

They’ve bedecked her like a page boy:
his to pass through Dauphine.

Now she had three loving brothers,
who have gone to seek that may,

and they’ve sought her and they’ve found her
by a fount in a meadow gay.

‘God be with you, Perronnelle!

Please come home with us to-day.’

‘O no, no, my bonny brothers,
never to France return I may.

But commend me to my father
and my mother dear, I pray.’

(Paris et Gevaert, xxxix.)

1 See Note D, at the end of the book.

I3 g ROMANCE BALLADS

There is assonance in the alternate lines, and it is a rule that the
cadences of the two lines making the couplet must be different (in
this case feminine endings, followed by masculine endings with
the rhyming word). Doncieux has preferred to arrange these
pieces, therefore, in long lines, so as to allow the assonances to
follow at the end of each long line, making a tirade. Nigra in
Piedmont and Mila y Fontanals in Catalonia have adopted the
same arrangement. The rule is then given that the hemistichs of
a traditional line must have different cadences. An appeal to music
shows that there are phrases for each line, which might justify the
arrangement adopted in the translation; but it also shows that these
songs have a different form as sung from that which they appear
to bear as read. La Perronnelle is sung as a quatrain, with musical
phrases arranged as afiaa'. The use of nonsense refrains in the
dancing songs, and the repetition of the single lines or hemistichs,
gives to some pieces quite an elaborate stanzaic form as sung. The
almost equally popular La Pernette has a tune arranged as agg'afiyp ;
the first line is sung, then the nonsense refrain, then the first
repeated, and the second sung three times over, using a different
musical phrase the second time.

Whether these poems be composed in long divided lines or in
couplets may be open to doubt; but they are in effect related as
couplets to the slightly later quatrain style. The quatrain is rarely
used for dancing, and is more appropriate for narrative. The
rhymes are frequently arranged in the simplest manner (aabb), and
the music in four phrases (ajSyS). This is the metre of Jean Renaud,
King Loys* Daughter , La Belle Barbiere, The Torch of Love , The
Drowned Diver, and other famous narrative songs, a notable pro¬
portion of which must be considered importations into France
from Germany, the Netherlands, and even Scandinavia. A
genuinely French song, The Sheep saved from the Wolf (which is
one of the mocking ‘pastourelles’) arranges its rhymes alternately
(. abab ), and observes the rule as to change of cadence; this rule,
however, does not generally affect the poems in quatrains. There
is a tendency for poems in couplets to pass into quatrains, as one
may note in comparing La Perronnelle with its derivative Fanchon,
or the Girl and the Dragoon. Yet the couplet remained alive, and as
late as 1781 was the form taken by the highly popular song about
Marlborough.

The French Romancero, as arranged by Doncieux, is poor in

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 139

historical matter. The oldest event recorded is to be found in The
Hanged Scholars . It is the foundation legend of the Hotel-Dieu
at Pontoise, built by St. Louis from the fine paid by Enguerrand
de Coucy in 1259; he hanged three scholars of St. Nicholas’s
Abbey, Laon, who had hunted in his forest. The facts correspond
to those of a Dutch and German ballad, My Lord of Brunswick and
the little Boy . In the French ballad the crime is inspired by the
more convenient theme of amatory intrigue, as in our ballad of
The Clerk's Two Sons of Oxenford (Child 11). It is not certain at
what time the ballad arose, since the legend would always be
available at Pontoise. In some details, perhaps by chance, the piece
resembles the High German Castle in Austria , which is of the
fifteenth century. The word Pontoise was, in the course of tradi¬
tion, replaced by Toulouse, and it is under this style that It spreads
to North Italy and Catalonia (Nigra 4 and Mila 208). Sometime
before the return of Francois I to Paris (17th March 1526) from
his captivity in Madrid, a ballad was composed, not wholly in
his favour. It spread later to Italy, where the imprisonment of
Louis XVI seemed a sufficiently close analogue to justify the
change of names. It is one of several ballads inspired by the fight
at Pavia in the imagination of the contestants. The Spaniards
remember the victory in a prosy and circumstantial piece which
gives credit above all to Antonio de Leiva. To the Germans it
seemed rather an opportunity for exalting the prowess of the
landsknechts against their professional rivals, the Swiss pikemen;
their hero is Georg von Frundsberg. So, though one event has had
ballad consequences in different countries, Pavia illustrates the
truth that historical ballads seldom travel beyond the frontiers of a
single homogeneous ballad area. It is somewhat surprising, there¬
fore, to encounter in Germany a version of the French Poisoned
Marchioness, complete with tune, which gives voice to the suspicion
that Henry IV’s mistress Gabrielle d’Estrees was poisoned in
1599. The English Marriage refers to Henrietta Maria’s marriage
in 1625; but the celebrated Marlbrough s’en va en guerre, which
has spread to Germany, Italy, Catalonia, the Asturias, and
Portugal since 1781, has no more connexion with our general
than the name. The story is that his death was reported to his
faithful wife—which is the old plot of Belle Doette . The tune (a
singularly sprightly one) may have been used for a hunting song
in the seventeenth century. A comparison with other collections

140 ROMANCE BALLADS

does not increase the historical content of French balladry, though
there are allusions in plenty to English ravages in our losing fight
to retain a hold on France in the fifteenth century. The death
of Olivier Basselin (c. 1450) and the misconduct of Louis XFs
garrisons (c. 1465) are also mentioned, and serve to confirm the
general period assigned to this later medieval phase of popular
poetry.

Doncieux prints seven religious pieces, but the bulk of his collec¬
tion consists of amatory adventures, which can be grouped round
the name of the heroine. Some of these ascend to the fifteenth
century, and were then already welcomed by many. There is the
brief Perronnelle, already mentioned, which was later transformed
into The Girl and the Dragoon . It is really a professional song of
the troops, like the late pieces, The Jolly Drummer and The
Soldier-Husband's Return. Sailors also have their ballads: The
Embarkation of the Singing Girl , The Corn Ship , and above all
The Short Straw. It would be convenient to place the origin of this
piece on the coasts of Normandy or Brittany, whence it would
radiate outward to Provence, Catalonia, and Portugal on the one
side, and to Scandinavia and England on the other. There seems,
however, to be no reason other than geographical probability for
placing this ballad to the credit of a French minstrel. Thackeray
has used it in his Little Billee, which one would reckon as a jocose
ballad of the period of decadence, despite the respectability of its
authorship. Similarly there is a professional element in La Belle
Barbiere, The Orange Girl , and The Orange Seller. Outside
Doncieux’s collection we find a strong tendency to connect ballads
of the sixteenth century with the professions. The composer gives
his or her status by way of signature in the last verse. The practice
extended to Germany, and was normal in landsknecht and reiter
ballads which purport to convey information about contemporary
events.. We have the ballad of Sickingen , for instance, on the
authority of a landsknecht who had just come (to somewhere
unnamed) from Landstuhl, after witnessing the hero’s fall.
Ballads about prisoners, the Prisoners of Nantes and Pierre de
Grenoble , are, if not professional, at least of interest to a peculiar
social class.

There remain a few songs which deal ingenuously with love (Les
Princesses au Pommier doux and Claire Fontaine) or with lovers’
tragedies. In The Hapless Marriage, the lovers, separated by the

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 141

force of their parents’ will, die together in church. The theme is
too simple to be located in any one country. Scarcely more de¬
veloped is Beau Robert (not in Doncieux), the youth who returns
to find his fiancee dead and passionately addresses her corpse on the
bier. She who feigned death to guard her honour , or to avoid a dis¬
tasteful marriage, is another well-known ballad theme. The fond¬
ness shown by Italian novelists for this motif of apparent death,
either as an effort of wdll or under the influence of some narcotic,
is so pronounced that one is inclined to assign the germ of all such
tales to Italy. Feigning death to avoid a distasteful marriage is the
motif of the English Gay Goshawk (Child 96), the Serbian Erceg
Stepan , and other poems discussed in Chapter VI of the First
Book. It is used in King Loys ’ Daughter , perhaps as an after¬
thought. The ballad is one of a pair. A princess, in love with a
gentleman of low degree but high merits, is prevented from marry-
ing him either by her own imprisonment, as in King Loys’
Daughter and the older Belle Ydome , or by his imprisonment and
execution, as in La Pernette and La Belle se sied an pied de la tour
(v T hich is the more lyrical variant of La Pernette). In the former
case a happy ending arises either by the lady’s escape, through
feigned death or some other show of ready wit, or by the lover’s
valour; in the latter there may also be a happy ending, but a more
usual conclusion is that the lovers should die and flowers or trees
intertwine above their graves. Both stories are represented in the
Castilian pseudo- Carolingian cycle of Count Claros of Montalban,
and the latter in Conde Olinos. As for The Swine Girl (La Porche-
ronne), which has a wide European range, the chief reason for
considering it French in origin is a tradition which links it with
the name of Guilhem de Beaucaire.

There are a number of pieces in Doncieux’s collection which
show that, however much other nations have owed to suggestions
from French literature, the narrative ballad in France is largely due
to a reflux from abroad. Movements within the Franco-Italian
area itself are only to be expected. Provence is the focus of
UEscnveta , which has spread to northern France and to Catalonia.
From Italy came Dame Lombarde , the poisoner theme, The
Drowned Diver , and probably Marianson’s Rings. The first is the
history of Rosmunda and Helmichis, probably excavated by some
felicitous poet from Paulus Diaconus. The diver was Cola Pesce
who lost his life diving for the amusement of the Emperor

x 4 2 ROMANCE BALLADS

Frederic IL The range of this story is from France to Greece,
with Schiller's Der Toucher as an offshoot. It is not easy to deter¬
mine where it was first made a ballad; it may even have been exca¬
vated from Belleforest in the sixteenth century. As for Mariansoris
Rings , the priority of Boccaccio’s use of the motif in the Decameron
(ii. 9) is undoubted; but again there is no clear indication as to
where it first took ballad form.

From the Low Countries came the Holofernes ballad of Renaud
the Woman-Killer and the Leander song called by Doncieux The
Torch of Love. The soldier’s return home, which has already been
mentioned, is either a fresh creation on a commonplace event, or
a worn-down variant of the Moringer ballad. Somewhat more
travelled are the pieces entitled Marguerite or the White Beast ,
Belle Helene or the Dancing Girl who was drowned, and King
Renaud or Jean Renaud . They are all Scandinavian: the first is a
tale of bewitchment; the second belongs to the cycle of the nixes;
the third is an offshoot of Elveskud. It lacks the beginning—the
encounter between the hero and the elf-woman—and so is less
perfect than the corresponding Breton poem of Count Nann. For
this reason, no doubt, Doncieux considered that the Breton piece
must have intervened between the French and the Danish; and
that may have been the case. The evidence as to tunes does not
wholly confirm this opinion. Child reproduces two variants of one
tune in connexion with his Clerk Cohill , and it is clear that they
are versions of the melodies A, B, and C in the Udvalgte Danske
Folkeviser of Abraamson, Nyerup, and Rahbek (vol. v, melodies for
No. 35). A fourth tune, recorded by Kristensen from oral tradition
in 1891, differs considerably from these, but it is not unlike that
which Bujeaud found circulating in the Angoumois. There is also
an enigmatic Catalan tune, which might be associated with either
of these patterns. The Breton melody also might be of Scandin-
avian origin. On the other hand, the bulk of the melodies current
in Franee are of quite a different sort. They adopt various equiva-
lents of the melodic formula 72D, which is the one actually chosen
for reproduction in Doncieux’s Romancero . I have encountered
what appear to be analogues of this type only in Czechoslovakia;
if the connexion be genuine, then it suggests that the prevalent
French tune is more recent than those of Angoumois and Catalonia,
and is an intruder, perhaps from Germany. It would not be at all
surprising that the oldest and correct melody should be the one

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 143

with least support. 1 La mal mariee vengee par ses freres, which
Arbaud entitled Clotildo, is more likely to be a derivative of the
German cycle of Kudrun ballads than an historical notice of the
sorrows of a Merowing princess.

The ballads of northern Italy belong to the same group as those
of France and Provence: they have the same form, many of the
same subjects, and a related language. 2 The Maritime Alps are not
a linguistic obstacle, and the changes which mark the difference
between French and Italian take place so gradually that a traveller
would not be able to encounter a fixed frontier in all the distance
from the Alps to the Apennines. 3 The Franco-Italian ballad area,
however, extends only to the limits of Cisalpine Gaul, not on
account of the Celtic substratum of the Gauls, but partly for
political reasons and partly through opposition to the ‘lingua
toscana’. This tongue—standard literary Italian—is not identical
with the speech of any particular place, even of Florence. It has
always been, to some extent, a convention of educated men, a
‘volgare illustre’; and since the middle years of the thirteenth cen¬
tury it has been the vehicle of a rich and brilliant literature. This
literature, and the whole culture it implies, has enthralled Central
Italy, leaving to the dialects a restricted field of peasant improvisa-
tions; it is only in regions linguistically and (at one time) politically
remote that original creation in the local speech has found room
to develop. The Sicilian dialect in the far south supports a flourish¬
ing balladry, though even this has been influenced by, or modelled
on, the Tuscan literary tradition. In the far north, Piedmont and
Lombardy employ vernaculars equidistant from standard French

1 Elveskud 5A1221AD2B 1 sharp 24 U8 d/g.fg/a.bc/b.ga/g.

5A14BBE5A14B 1 sharp 24 U16 d./g..fg..b/b..ag..d/g..fg..b/a..-
52BAB212C1AD 1 sharp 38 U8 d/g..(agf)/efga(afgf)/d.

C, Colvill 525BE2 34 U8 dgg.g./a.ad.,/cga.

(d..g)/g...g./a...a./£..eg/a...

Elveskud 5AD5234B2B 1 flat 24 U8 (cc)/f.ec/f.gb/<f.<tf/£..

Renaud 5A1A1A 1 flat 68 U8 cf.e/f.ef.ef.

Bona Viuda 122 38 U16 e(.f)g.g./a...a./

a Nann 25BE25B 2 flats 44 U 4 f.(g..c)/b..r(f)/g.c./c.br/

Nest'atnd

svat'ba III D9B 1 flat 34 U8 aaa.f .jddc...!

Renaud 3ABB 68 U8 aaa/a..cba/g..

32B22 68 U8 aaac../<fofe../

3BABB 1 sharp 68 U8 bbb/di.dba/g..

2 See especially C. Nigra, Conti popolan del Piemonte, Turin, 1888 (with
some tunes), and G. Ferraro, Canti popolari Monferrini, Turin-Florence, 1870.

3 H. Schuchardt in Hugo Schuchardt-Brevier, ed. L. Spitzer, Halle, 1928, p. 167,

i44 ROMANCE BALLADS

and standard Italian, and until the later years of last century
constituted a sphere of French political influence. The French
ballad style therefore advanced over northern Italy until it en¬
countered the resistance of the ‘volgare illustre’ south of the Po.

The presence of ballads dealing with French history is, as we
have seen, evidence of unity within the area; The imprisoned King
transferred from Francois I to Louis XVI, Gabriella and The
English Marriage (Nigra 5, 36, 46) are our witnesses. The other
historical ballads circulating in Piedmont are, for the most part,
glorifications of the house of Savoy, commencing with the March
of Prince Tommaso di Savoja (Nigra 137), a soldier active in the
second third of the seventeenth century, and descending to the
Napoleonic era. Somewhat earlier than those pieces is The
Marquis of Saluzzo’s Will (Nigra 136), but it is not certain whether
this ballad is to be considered contemporary work. The Siege of
Verrua (Nigra 135) is still more difficult to date.

The castle of Verrua it is so fair a hold,

upon that rock ’tis planted that guards the ford of Po.

The beauty at her window has cast a look below;
has seen approach a vessel with armed men loaded low.

Their arms are all a-glitter, it seems they are of gold.

-The beauty throws a missile, and down the boat must go.

Had she not cast that missile, ta’en were Verrua’s hold,

ta’en were Verraa’s castle and Monferrato also.

The missile was, in fact, a stone, and I have used the word merely
for the sake of the trochaic rhythm. ‘La Bella’ has not been identi¬
fied, so far as I am aware; I suppose she was a cannon, but she
might have been some older engine for throwing stones; or, of
course, she might have been some person. As for the sieges of
Verrua, there is an embarrassment of riches; we can choose
between those of 1377, 1387, 1625, and 1704. The earlier dates,
preferred by Nigra, seem incompatible with the general date given
to this ballad style.

In other respects also the themes circulating in Piedmont
approximate closely to those of France. Almost all the ballads
included in Doncieux’s Romancero have their equivalents in Nigra’s
Canii popolari del Piemonte and in Ferraro’s booklet, both those
which are purely French and those which we have seen reason to
think borrowed from the Germanic peoples. Not only is this
agreement close in theme; it descends to the order of sentences and

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 145

to the words. The Piedmontese prosody is also that of France and
Provence, with the same characteristic alternance between mascu¬
line and feminine endings in what may be described as either
couplets or long lines divided into hemistichs. The points of
difference are only of minor importance. The assonances are often
in e or e plus an unaccented vowel, as in French, but <2, t, u, 6 are
also accepted in assonance as equivalents of e. The niceties of
prosody are also less scrupulously guarded. This impression may
arise in the mind partly owing to an accident of editorship, since
none of the French texts have been reproduced with the austerity
of the Italian editors; but it must also be due in part to the absence
of contact with literary models, since these employ the quite
different ‘lingua toscana’. It is possible to accept the greater
number as genuine traditional poems, without being perplexed by
semi-erudite elements. The line tends to be somewhat longer, as
the North Italian dialects have suffered less from the loss of final
vowels than French has done. The style also is more narrative.
In this matter we have to discount once more the preferences of the
modern editors, which have been normally lyrical in France and
epical in North Italy; but the mass is sufficiently large to justify
the observation.

In details we may sometimes note a closer connexion between
the Italian and some southern French redaction. Thus Marianson’s
Rings (Nigra 6) take for hero Trinsi Raimund* and for villain the
‘duca d’Ambo’, while the scene is given as Lyons. The Hanged
Scholars is associated with Toulouse; and II Mom Saracino is an
offshoot of the Proven£al Escriveta. We may suppose also that
there would have been traffic in ballads between Genoa and Barce¬
lona, such as would account for The Princess (Nigra 8), which is a
very short, dramatic equivalent of the Spanish Count Alarcos , the
Catalan Conde Floris. The Poisoned Man's Will (Nigra 26) is a
rendering of Lord Randal . A version encountered at Pisa by
Alessandro D Ancona has the same form as the English ballad; in
others the metre has been accommodated to the pattern of the long
ballad line. Similarly, Ambrogio and Lietia , the theme of the cruel
husband who compels his wife to travel fast while with child, stands
closer to Child Waters than to other forms of the same tale.

To distinguish what is original to Piedmont is more difficult,
and perhaps only Donna Lombarda can be referred to an Italian
minstrel without dubiety. The Italian versions of this piece are

4615 jj

146 ROMANCE BALLADS

more numerous and veridical than those of other lands; but, of
course, there is no need to suppose a poetical tradition going back
to the time of Alboin, seeing that Paulus Diaconus’s text was one
of the best known sources for history in the Middle Ages. One
supposes also, though without assurance, that the use of narcotics
is a characteristically Italian novelistic device, and that ingenuity
has more admirers among the Italian folk than elsewhere. This
difficulty of asserting the Italian origin of many poems does not
warrant an imputation on the originality of the Piedmontese
singers. It is due to the absence of characteristic marks on ballads
of adventure, which forbids our localizing very many of them. As
a high-road for ballad commerce North Italy is important for
the history of those which have spread from the European
North and West Into southern Italy (and thence sometimes to
Greece), to Venice and Dalmatia (and so to Serbia, Bulgaria,
and Rumania), and perhaps also, in the Hapsburg armies, to
Czechs and Hungarians.

At the other extreme of this Franco-Italian ballad area, the folk¬
songs of Brittany form a group defended by their peculiar language
against the comprehension of strangers, though themselves open
to influences streaming in from France. 1 The ‘gwerziou’ show no
sign of the ancient connexion with our islands; they enshrine no
heroic traditions comparable with those of the Gaelic poems, nor is
there any Indication of a bardic school. The English appear as the
enemy, with only this difference from French balladry that it is
chiefly the English seamen who are feared. There is a considerable
amount of international matter, and many ballads deal with the
more popularly revolting crimes in the manner characteristic of
decadent balladry everywhere. The verse forms are those em¬
ployed in France: assonating couplets and quatrains (often with
the simple rhyme-pattern aabb), with or without refrain. The
narrative element is more pronounced in these poems, and the
literary admixture slight. The Breton ‘gwerziou’ doubtless form
a greater proportion of the total production of the Breton imagina¬
tion, than do the ‘chansons populates’ in France, and so would
have played a greater part in the moulding of thought. In this

Breton is not one of the languages I read, and I am dependent on collections
m which the original has been faced by a French translation. I follow F M
Luzel’s Gwerziou Bretz-Izel , Lorient, 1868-74, with confidence, and H de la
V illemarque’s Barsaz-Breiz , Paris, 1867 (6th ed.), with hesitation.

FRANCE, PROVENCE, NORTH ITALY, AND BRITTANY 147

loGal sense the religious and moral ballads would acquire real im¬
portance, especially those which deal with Breton saints, like St.
Mathurin and St. Henori, ballads of witchcraft (such as The Wax
Child , Luzel, i, p. 142), and those that reprove criminals in a
dramatic manner. One may cite the case of Marie Till (Luzel, ii,
P* 5*8), who concealed the Holy Wafer in a pot, intending to insult
it; the piety of a horse, which kneeled, revealed the sacrilege, and
after execution, the useful moral is reached that boys should not
drink nor girls learn Latin!

The historical ballads descend in a regular series from the six¬
teenth century. The debauched Bishop of Penanstanc (Luzel, i,
p. 424) has been identified as Francois de la Tour, who died there
in 1593; Count Des Chapelles (i, p. 456) was a certain Francis de
Rosmadec, decapitated in 1627; a ^d in Les Aubrays a poet has
sung the duel between Koat-ar-Ster and Jean de Lannion, who
died in 1651 (Luzel, i, p. 286, La Villemarque’s Lez-Breiz). The
Siege of Guingamp (Luzel, ii, p. 40) is concerned with the siege of
I 59 I ? with which most of the details are In agreement; but the
mention of the heroism shown by the cannoneer and his wife may
reflect that of Rolland Gouyquet and his wife in the siege of 1488.
Another early date is provided by Marguerite Charles, dealing with
the capture of a celebrated female bandit in 1598. In a number of
cases it is not certain who are the persons of these little dramas,
which are represented chiefly in their domestic or seignorial aspect,
but there is no reason to believe them earlier than the ones cited*
La belle Catoise (Luzel, ii, p. 348) is a ballad of 1812, and the
Marquis de Locmaria mentioned in Clerk Lammour (Luzel, ii,
P* 466), was an acquaintance of Madame de Sevigne’s. These dates
throw into high contrast the suggestion made concerning The Sea-
wolves (Luzel, i, p. 72). In quatrains rhyming aabb , with a refrain
ccc, this piece tells how privateers attacked Gueodet and were
beaten back in a three days’ battle. The event has been identified
with Hasting’s attack on Roz-Gueodet, which he destroyed in 837;
if this be correct, the ballad must have had some literary source
or other embodiment in local tradition, since the verse form is
relatively modern, as we have seen above, and the genre does not
appear to go back earlier than the sixteenth century in Brittany.

The external affiliations of Breton ballads are chiefly with
France, but also with Scandinavia in some notable instances. As
the best English ballads belong to the regions north of the Trent,

I4 S ROMANCE BALLADS

and even north of the Border, It Is not possible to discern any
English influence on the formation of the Breton corpus. The
international matter preserved by Luzel contains some pious pieces
which have attained wide celebrity and can hardly be assigned with
assurance to any particular focus. There are those which reprove
Dives and his wife for their hardness of heart: The Famine (I, p.
76), The Poor Widow (i, p. 80, miraculously saved from killing her
children by the intervention of the Virgin), and The Two Sisters
(ii, p. 508). The innocent maiden who cannot be burned at the
stake appears in Anne Cozik (i, p. 218), and there is the apocryphal
miracle of the roasted capon which crew, associated with the name
of Marguerite Laurent (i, p. 210). Specifically French are the
poems entitled She who went to see her Mistress in Hell , Marivonnic
or death before dishonour, The Knight and Shepherdess and Robert
the Devil , the latter probably taken from a chapbook (i, pp. 45,
350, 194; ii, p. 24). So also are La petite Franpise et le petit Pierre
(ii, p. 26), which Is a version of King Loys ’ Daughter , The Sarracens
—a variant of Escriveta —and probably also The Short Straw (ii,
pp. 20, 182). The Hallewijn theme appears under the title of
Rozmelchon and the Kudrun cycle as Brother and Sister (i, pp.

202). Among the ballads of Scandinavian origin, the best known
is Count Nann (i, p. 5), which has already been discussed. The
text is fuller than that of France, since it contains the first part of
the complete ballad; the tune may also be of Danish origin. A very
curious piece, The Tailor and the Dwarfs (i, p. 134), relates an
attempt to steal the dwarfs’ treasure. The robber is compelled to
dance to death among them. Apart from the motive for the visit,
this resembles the Danish Eheskud and other ballads describing
the dances of trolls and elves, and it has no French equivalent.

2. Central and Southern Italy and Sicily

Central and Southern Italy are regions subject to the dominion
of a great artistic literature which has given masterpieces to the
world since the thirteenth century; in Sicily, while the dialectal
differences suffice to withdraw traditional verse from the over¬
whelming supremacy of the Humanities, the poetry of the people
still takes its rise from the hendecasyllable of the cultured poets,
the authors call themselves ‘pueti’, and indulge in a modest pride.
Some names, even, are known, and some details of their ingenuity
as practitioners. Their art is essentially lyrical. The art of Pied-

CENTRAL AND SOUTHERN ITALY AND SICILY 149

montese ballads was, as we have seen, noticeably narrative, though
the impression was sharpened by the personal preferences of the
collectors. Nigra preferred, and gave first place to, narrative verses,
admitting only a small fraction of the effusions in the lyrical style,
the £ stornelli’ and £ rispetti\ Consulting other collectors, such as
Marcoaldi, 1 we find that these forms are abundant in the North
also. On the contrary, epic pieces of the northern type, the exist¬
ence of which in the southern districts has been formally denied,
are abundant. Barbi gathered 116 versions of Donna Lombarda in
Sicily, and Cocchiara announced a form of the Belle mal mariee
from the Basilicata and another from Romagna, the French
Captain and the Beauty from Emilia, The Diver from Umbria, as
well as offshoots of Elvesknd (via France) in Venice and probably the
Abruzzi. 2 There is no trench separating the north from the centre
and south, and yet the differences are very marked in their sum.

The chief difference is that poetry south of the Po supposes a
background, not of French medieval literature, but of the Italian
Humanities. The hendecasyllable *is the established measure, and
is used with virtuosity. The great authors are not too far removed
from vernacular usage to fail in popularity; stanzas from Tasso are
sung by the sailors' wives of Venice. For the purposes of narrative
a middle style has been devised by the £ cantastorie’ which is at best
semi-popular.

Even at this day here and there in the cities of Italy, one meets many
‘cantastorie’, worthy heritors of Daphnis, who sing their songs (usually
strings of octaves) at the corners of the squares, telling stories of knights
who fight not for a lady but for a kingdom, the adventures of lovers,
satires against boors, contests of mothers anxious to marry off their
daughters and debates between brunettes and blondes about the primacy
of beauty. These are, in fact, all the themes of the ancient popular
poetry in flower, clad with new motifs and modem traits: particularly
in those pieces in which we perceive a detailed exposition of local facts
and notorious events. The name changes; at Naples the singer is called
a ‘rinaldo’, and at Palermo £ orbo\ The ‘cantastorie’ is, however, the
successor of the old rhapsodes. 3

1 O. Marcoaldi, Canti popolari inediti; umbri, liguri , piceni, piemontesi, latini,
Genoa, 1855.

2 M. Barbi, ‘Scibilia Nobili\ Pallante, i, p. 14; G. Cocchiara, VAnima del
Popolo italiano nei snoi canti, Milan, 1927, p. 92 Fior-d* auliva and p. 189 Virgu-
leina , p. 184 La Bella Ingrese and p. 230 Cecilia (Nigra 3), p. 136 UAnello 0 il
Pescatore, p. 206 II Conte Angiolino and p. 112 Nucenzie (?).

3 G. Cocchiara, L'Ammo del Popolo italiano nei snoi canti, Milan, 1927, p. 24.

I 5 0 ROMANCE BALLADS

So the presses of the fifteenth and sixteenth centuries produced
such works as the Vaunts of the Paladins, the Story of Milon and
Berta , Bradiamonte, and other chapbooks in verse, 1 which suppose
some literacy in the singer, but very little in the audience.

The demand for formal narrative being met in this way, there
remained only the lyrical genres for spontaneous improvisation.
Two moulds were fashioned. The one, a mere lyrical cry, is that
of the 'flowers’ (Sicilian ‘ciuri’, Tuscan 'stornelli’, and other
names). It opens with the name of a flower: flower of the lupine,
pomegranate, lime, of flowers, &c. Then follows a free line
(generally), and one other rhyming with the ejaculation:

Flower of the beech,

Above all beauties thou dost bear the prize,

and unto you the sun’s first beams do reach.

Flower of the vine.

Thy charms are things that last through endless time.

The gambit is used in Rumanian poetry in the form 'green leaf of
...’ (Frunza verde), to introduce poems of considerable extension;
but in Italy the two or three lines suffice. A variation on this style
is that in which no flower is named, but the same technique is
observed:

Starry eyes!

As many stones as go to make a bridge,
so many send I ardent lover’s sighs.

The other mould is that of the Sicilian ‘canzuni’, Tuscan 'rispettih
These are hendecasyllables assonating alternately or in pairs; the
number of lines varies, but there is a strong tendency to limit the
poem to eight lines. They are less fleeting and more substantial
than the 'stornelli’, and they have some of the plasticity, though
not the neat workmanship, of the classical Greek epigram. The
themes are mostly erotic. The lady challenges all nations with her
beauty (and especially France), she defies the skill of painters with
her blonde tresses, or writers to describe her aright; the singer
longs to be a bird and visit her, a necklet to clasp her neck, or to
lay a hill of gold at her feet; he holds a dialogue with the water that
has mirrored her face or with an intimate thing like a handkerchief;
the Holy Father is ready to pardon the pleasant sin of loving; the

1 G. Barini, Cantdri cavaliereschi dd secoli XV e XVI, Bologna, 1905.

CENTRAL AND SOUTHERN ITALY AND SICILY 151

lover protests his faithfulness and sends his heart. Or he is angry
or jealous or glad to end a base affection:

Dear God, good day,

now all my love has fled away;

or he scorns the advances of an amorous old hag. Religion and
proverbial wisdom, misfortune, imprisonment, satire, civic rivalry,
and other topics find expression in these songs; and in verses of
looser structure we find the usual assortment of games, lullabies,
prayers, riddles, and 6 airs’ (arii).

The £ arii’ and a few ‘canzuni’ provide all the narrative verse of
Sicily, together with ‘legendi o storii 5 . There are mere allusions
in the ‘canzuni 5 : to the public baptism of the Saracens by William
the Good (1166-89), and even to Count Roger I (d. 1101), to the
Sicilian Vespers, and to hostility towards the French. 1 These
references have led to the inference that these poems go back to the
twelfth century; the inference is supported by the present tense in
Now that King William wears the crown

(Pitre 568)

and by Pitre’s belief that the ‘contrasto’ of the Two Lovers (968) is
a prolongation of Ciullo d’Alcamo’s thirteenth-century Contrasto.
But the allusions are too brief to warrant such conclusions,
especially as all these improvisations are touched by the higher
literature and show considerable virtuosity. It is not that there
were no songs sung in Sicily in those early years—far from it. But
the songs are not likely to have been these, with their dependence
on a much later development of artistic literature.

When we get narratives of some length we find that they belong
to a comparatively recent epoch. The oldest and best of the
Sicilian pieces is the Baroness (or Princess) of Carini (Pitre 918).
It is a confused, semi-literary bulletin in verse, in octaves and
irregular stanzas, describing the murder of Caterina Talamanca-
La Grua on the 4th December, 1565. Her lover, Vicenzo Veraa-
gallo, actually escaped to Madrid and entered a religious order; but
the poet, perhaps influenced by Dante, sends him to Hell to see
the dead lady and the traitor who betrayed them. Monsii Bonello
(Pitre 919) corresponds, perhaps directly, to the first tale in the
Decameron. The poet dates the event precisely: 26th February
1399; but when he wrote Geneva was already a Calvinist city.

1 G. Pitre, Canti popolari Siciliani, Palermo, 1891, 2 vols,, with bibliography.

152 ROMANCE BALLADS

The Devil’s Bride (Pitre 927) appears to be of Spanish origin. It
concerns a girl who gave her soul to the devil, and was only saved,
in the Calderonian fashion, by her early devotion to Our Lady of
Mercy. The level of taste is below that of the best epochs of ballad
composition; below that of Piedmont. There is a calendar of
storms, floods, and pestilences afflicting Palermo and Messina
between 1630 and 1850, much hero-worship of bandits like
Antonino Martino or Antonino Catinella, surnamed ‘Jump-the-
vines’. There are murders: how an adulteress murdered her
child and husband, and a bad baron procured the death of his
daughter’s lover, and how pretty Giuseppa of Palermo was cut to
bits. To such ingenuous horrors the ‘pueti’ oppose ingenuous
marvels of piety, honouring Our Lady of Carmel and of Mercy, and
Saints Joseph, Andrew, Anthony of Padua, Antonine, Catherine,
Lucy, Rosalia, and Genevieve. There are naive recapitulations of
the Gospel story, and a few international narratives. The Soldier-
husband s Return has sunk to be The Prisoner’s Return (from
Pantellaria). We meet also the North Italian ballads of the Lover
as Confessor and The Lovers’ Flight. Such pieces may have come
directly from Liguria by sea, or have followed the land route
through Central Italy, where they are also found.

3- Spain, Spanish Jewry, Portugal, Ibero-America

Linked to France by the subject-matter of literary and adventure
ballads, independent in form and national spirit, the Castilian
romances are unsurpassed in Europe for their number, vigour
influence, dramatic intensity, and veracity. Unified by their em¬
ployment of the octosyllable assonating in alternate lines and by a
common technique, these poems form a body, a ‘romancero’,
which works towards a total effect. It is true that the ‘romancero’
has never been present to the mind of any particular audience as
it is to a reader of Duran or Wolf and Hofman,- and it is true even
that the word is not traditional. The earliest collections, in the
middle years of the sixteenth century, were published merely as
song-books’, ‘cancioneros’; but the practical convenience of
distinguishing between these and other song-books led to the
syncopation of ‘cancionero de romances’ into ‘romancero’ The
word corresponds, however, to a real thing, to the influence exerted

1 See Note D at end of book.

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 153

by a homogeneous balladry through its mass; and it has passed,
more or less appositely, to denote such collections as Doncieux’s
Romancero and Heine's Romanzero. Either usage is an admission
of a thing most perfectly exemplified in Spain; a uniform body
of narrative verse, severely objective in manner and capable of
traditional survival Similar masses of verse exist in Denmark and
Serbia, but neither the 'viser' nor the 'junacke pesme' have given
rise to a collective term covering all their ballads.

The Spanish 'romances' are narratives in form and intent, and
are thus sharply to be distinguished from the Franco-Italian songs,
which have a lyrical origin, and also from those of northern and
central Europe which have the form of a lyric even when their
matter and manner is narrative. The Russian £ byliny' and the
Serbian junacke pesme' are also wholly narrative, and the latter
are often historically true. The 'romances' differ from them by
cultivating a more austere objectivity. Not merely is the lyrical
element slight or excluded altogether, but the slight subjectivity
involved m the minstrel's words and presence is eliminated in some
ballads, leaving the characters in direct colloquy with the auditors.
They^ aie intensely dramatic in a way rarely attempted by the

'junacke pesme', more often by the 'viser', and not at all by the
'byliny'.

King Don Sancho, King Don Sancho,
never say you've not been told,
from this city of Zamora
has gone forth a traitor bold;
he calls himself Vellido Dolfos,
son of Dolfos, traitor old;
four the treasons he’s committed,
with a fifth the tale is told.

If the sire was a foul traitor,
fouler is the offspring’s mould.'

Shouts break out in the encampment:

'Sancho's hit and lying cold:
murdered by Vellido Dolfos,
by the traitor overbold.’

Hilled the king, he's fled for shelter
by a postern in the hold,
through the streets of fair Zamora
have his shouts and clamours rolled:

'Time it is to pay, Urraca,
if your promise you would hold.’ {PvwiciveTa 45)
x

4615

r 54 ROMANCE BALLADS

This is not the first state of a ballad. The ‘romances’ must have
been lengthy pieces like the extant Count Alarcos and Count Dirlos,
but the needs of the concert chamber in the sixteenth century have
caused many to be cut short, and discretion has pruned away from
the best of them every detail that could be spared. The characters
are left to act out their own drama, and so the impression of
veracity is heightened. The dominant impression is one of strict
historicity: such and such things did happen because they actually
are taking place before the eyes of the public. They did happen,
also, because a vast mass of Spanish ballads refer to historical per¬
sons and events, and because they do so in the most factual mann er
possible. The supernatural and the marvellous are almost wholly
absent from Spanish balladry. It seemed to Southey, who was a
good judge, that this is a defect of the ‘romancero’ which sets it
beneath the ballads of our own land. The ‘romances’ are inferior
in the fruits of the imagination which broods on things that are not;
they offer no escape from life, either as a source of horror or relief;
but their stark humanity powerfully stirs the human in us.’
‘Mentem mortalia tangunt.’

The form of the ‘romance’ being precisely defined, it follows
that there are many topics excluded from the ‘romancero’ which
we have to notice in dealing with the ballads of other countries.
At least two lyrical moulds were formed in the earlier Middle
Ages, serving for the domestic occasions and typical encounters
which bulk largely wherever the ballad is half lyrical. The older
of these was the ‘cossante’; 1 it was at home in Galicia, whence it
extended into Portugal, and its prestige was such that Castilians
also composed courtly verses in the Galician dialect. The ‘cos-
santes were in distichs with refrain (generally very brief). Their
chief aesthetic feature was their immobility; that is to say, lack of
either narrative or psychological development. For the space of
two, three, or four distichs the poet simply repeated the same
situation, the same emotion of longing, anticipation, joy, desola-
tion, restlessness. The clauses naturally tended to run parallel
with little more change than required by the assonance. In some

thatlf AT? m 7 p“ f P ° etiCal form ’ the most a »ractive being

number of examples can be found in the Oxford Book ’of PortJues^V^e Tn
rom cantigasdeamigoto“cantigasdeamor”’ Revue deLittfa-ntnrar A .
*vm, I93 8, pp. 137-s*, I attempted to amend

of poetry, and to show how it affected subsequent lyrics in thf Provenfal s^e

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 155

of the later ‘cossantes’, which have been too readily accepted as
typical of this art, the parallelism was made more rigorous, and
progress (such as there might be in such simple expressions of
feeling) was made by linking the lines. These poems, then, gave
expression to the domestic occasions of courtship and separation,
and were for the most part, though running under the names of
male troubadours and minstrels, regarded as ‘women’s songs’.
Their form suggests the round dance guided by a precentor.
They raise the question of feminine priority in the lyric for
Portugal as the ‘chori feminarum rusticarum’ do for France and
the zenske pesme’ for Yugoslavia. In their feminine and domes¬
tic setting, their delicate parallelism and exquisite candour, the
cossantes’ resemble the ‘damns’ of Lithuania, which we shall
have occasion to consider later. During the fifteenth century
and early sixteenth, they gave way to another popular form, the
quatrain or ‘quadra’, just as the lyrical ‘daina’ of Lithuania gave
way in Latvia to the epigrammatic ‘daina’, more often than not a
quatrain. At the present day improvisation among the peasants
of Portugal and, to some extent, of Galicia uses the mould of the
‘quadra’.

In Castile, on the other hand, there flourished a narrative lyric
called the ‘villancico’. Its early history is hard to unravel, but it is
at least plausible to believe that the pattern was known to the
Spanish-speaking subjects of the caliphs and sultans of Andalusia.
In the fifteenth century the form is exemplified by the ‘serranilla’
of the- Marquis of Santiliana. The theme is announced in an
opening quatrain or phrase :

No prettier grows a
dear maid on the Border
than she, the cow-warder
of La Finojosa.

The poet then gives, in a sequence of verses ending with this phrase
or part of it, a narrative setting for his emotion. He states where
he met the maiden, what he was doing, and how the encounter
proceeded. These pieces correspond to the ‘pastourelles’ of France,
but they are excluded from the Castilian ‘romancero’ by the differ¬
ence of form and by their lyrical manner.

The Spanish ballad style is thus singularly uniform. To be a
ballad the poem must be in one sort of verse—the octosyllable
with alternate assonance—and in one style—the objective narra-

[56 ROMANCE BALLADS

ive. 1 In Spain, as elsewhere, and perhaps more than elsewhere,
>allad is form. But this form is sometimes secondary. There exist
yrical equivalents for the ballads of Moriana (the Donna Lomharda
ballad), the lament for Prince Afonso of Portugal in 1491, and The
jentle Lady and Boorish Shepherd. On the other hand, even
:opical news might be expressed otherwise than by a ‘romance’, as
vith the fall of Alora and the song of the Comendadores de Cordoba ,
foundation of a play by Lope de Vega, with its lyrical outburst:
You Comendadores
to my hurt I’ve seen.

A number of different lines of evidence converge on the middle
fears of the fourteenth century as the date of the rise of Castilian
balladry. The historical ballads, which relate recent events either
is news or as propaganda, give us a steady succession from the
epoch of the civil wars in Andalusia between Pedro the Cruel and
his brother of Trastamara. Some of them are clearly older than the
chronicle of Pedro Lopez de Ayala, which dates from 1394, 2 and
it is likely that they go back to dates close to the events. The oldest
important incident is the murder of Prince Fadrique in 1358.
Ballads claiming to relate episodes from Spanish history before
that date are either based on epics or on chronicles, and in either
case are of literary origin. The events of the reign of Alfonso XI
down to 1344 were celebrated in a long poem, resembling a ballad
in some respects, but not in the ‘romance’ form. The chronicles
of medieval Spain show that the epics were still vigorous in the
year 1344, when new versions of some of them were recorded for
the first time in prose. The epical ballads, which arose out of those
epics and at last supplanted them in popular favour, would thus
probably belong to a later date than 1344 when the epics were still
in bloom. It is in the nature of things that adventure ballads should
not be readily datable \ but one occasionally encounters wisps of
external evidence concerning them. One is that the Gentle Lady
existed in mixed Spanish and Catalan about 1421. Those which
depend on French models must be younger than those models,
which, as we have seen, were not very old.

1 Some critics describe the form as embracing sixteen syllables with medial
pause and final assonance, and connect it with the epic line of the ‘cantares de
gesta\ This point has been discussed in the First Book, Chapter IV. A very
few pieces in the narrative style but with lines less than octosyllabic are also
included in the ‘romancero*.

3 S ee m y Roniancero del Rey Don Pedro*, Modern Language Review, xxv, 1930.

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 157

The state of Castile in the third quarter of the fourteenth cen¬
tury was such as to favour the emergence of political verse. Never
had faction been so envenomed. The ballads of King Pedro the
Cruel’s cycle (65-9) polarized this hatred by riveting on the king
responsibility for three murders, unpardonable in their atrocity.
Brief and anonymous, like editorial articles, they presumed to be
the expression of a common judgement; but, more efficient than
editorials, they were so framed that they became each reciter’s own
possession. The king was condemned for the murder of his
brother, his cousin, and his wife. The first and second he did, in
fact, kill; the third probably died of plague, aggravated by neglect.
Yet the force of the ballads concerning Queen Blanche was such
that not only were they received into the Chancellor Lopez de
Ayala’s history of the reign—complete with some assonances—
but they persuaded the very partisans of King Pedro. Their
reply was not to refute the baseless charge, but to insinuate a new
scandal: they insinuated (67)—

it is noised among the people,

whispered, not as something known—

that the brother and queen might have been lovers, and so deserved
their fate.

King Pedro came to the throne in 1350 as a child and without
friends. His brothers would, according to Castilian dynastic pre¬
cedent, have provided centres for disaffection under any circum¬
stances ; but their hostility was aggravated by their bastardy, which
was a slur to them and a reminder to Pedro of many years of neglect
in his father’s court. But Pedro had grown up with a mind patho¬
logically rigid. He would make no concession, resign no claim,
forgive no fault, nor thank any one for a service. His arrogant
pretensions disgusted his neighbours on all sides, alienated the
nobles one by one till he was left with but a handful of adherents,
and led on to those black crimes which made reconciliation with
his brother Enrique impossible. Paying no heed to his father’s
example, he complicated his position by taking for paramour
Dona Maria de Padilla, a compassionate lady with greedy relatives.
For her he neglected his queen, Blanche de Valois, from the day of
her marriage. On her untimely death he married Dona Maria.
The marriage legitimated her children, and especially Constance,
who became later the claimant to the crown; but the children of

* 5 § ROMANCE BALLADS

such a marriage were eyed with suspicion, and it was to the interest
of the enemy to allege that the marriage had only come because of
murder. The career of Pedro the Cruel went from bad to worse,
till he fell at Montiel in a fratricidal struggle (1369). Enrique
possessed the kingdom by right of conquest and consent, despite
his bastardy; a condition he imputed in self-defence to the late
king s daughters, one of whom was the wife of John of Gaunt.
Mith John of Gaunt the civil wars entered on a new phase,
culminating in his invasion of Castile (1386-8), and closed by the
treaty of 1390. The hatred distilled into the ballads of this cycle
is thus not only that of the last years of King Pedro’s reign,
but also that of the years in which his daughters were a deadly
menace.

The ballads deal with these events as three murders, two
prophecies, and some miscellaneous news. The murders were
those of the Master of Santiago, Don Fadrique, in 1358, and of
Don Juan, Lord of Biscay, in the same year; Queen Blanche’s
death in 1561 was set down as a third murder. By the first killing
Pedro was branded a fratricide. The poet increased the horror
of the event by charging the account to Doha Maria (though
we know—and perhaps he did also—that she begged for the
Master s life), whom he represented as a modern Herodias-cum-
Salome;

’Twas Epiphany in the morning,

’twas the first feast of the year;
when both damisels and matrons
from the king a guerdon seek;
some for cloth of silk petitioned,
for brocade some, fine and sleek,
some petition him for favours
in the cause of lovers dear.

Lady Mary, first of many,

comes a-begging with a tear,

begs the head of the Grand Master

of St. James’s chivalry. r x p

The minstrel who first sang of Don Juan’s death let the victim
speak in the first person. With dramatic pantomime he could
excite the lively indignation of the Biscayans and other Basques
whose territories Don Ennque’s armies had repeatedly to cross’
The device must have proved successful, for it was extended to the

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 159

other ballad also. In an amended version which was generally
accepted Don Fadrique also relates his own taking off:

There was I in fair Coimbra,

city that mine arm had won. (65)

The words are untrue. They may have been suggested by the
Portuguese war of 1384-8, when Coimbra was several times in the
news, and the daughters of Dona Maria were carrying out, as
brides of powerful English princes, an invasion of Castile.

But the crowning horror was an Imagined tragedy: the murder
of Innocent Queen Blanche. A ballad-poet came out with a cir¬
cumstantial account of the event, contrasting atrocious brutality
with angelic patience :

Came the slayer to the princess,
found her at her orisons.

On the dire mace-bearer gazing,
saw she her unhappy doom.

Said the knave: ‘My queen and lady,

5 tis the king who sends thee word,

set thee now thy soul in order

with its Maker, even God,

for thy hour is come upon thee,

nor can I delay it more. 5

‘O my friend, 5 then spake the princess,

‘this my death I pardon you;
as the king my lord commandeth,
even so thou’rt bound to do;
but deny me not confession,

pardoned unto God to go. 5 (69 a)

These are strong words, which have become history.

A device of the political journalists of that age was to dress up
vituperation as prophecy. The Chancellor Lopez de Ayala crowns
his chronicle of Pedro the Cruel by inventing (or discovering) a
certain Benihatin of Granada who prophesied evil concerning the
king and then explained his prognostications by the event. In the
Baladro del sabio Merlin there are alleged prophecies of both earlier
and later date; indeed, Merlin was the bookman’s prophet, but
there were many others. The ballads introduce a clerk who
prophesies before the battle of Najera in 1360, and Lopez de Ayala
repeats the tableau. The circumstances do not show clearly whether
song or story came first; but it is virtually certain that the shepherd

i6o ROMANCE BALLADS

(pastorcico), who prophesied to the King just before the murder
of Queen Blanche, belonged to the ballad (66) rather than the
chronicle. It leads up to the murder, but the murder is not of
history but of balladry. The 'pastorcico 5 , furthermore, is a super¬
natural figure; he makes his appearance out of a 'black mass’ which
the royal huntsmen saw approaching from the direction of
Medina Sidonia, and he easily escapes arrest. In rationalizing the
episode the chronicler falls into contradictions.

The ballad of Don Pedro’s death at Montiel is savage, but
pedestrian. In the preceding year the royalists had made a
military promenade through the Guadalquivir valley, during which
they had momentarily appeared before the gates of Baeza. The
readiness of the governor, Ruy Fernandez, scared them off after
a brief skirmish. This event was celebrated in a ballad, a few lines
of which were gathered by Argote de Molina in the year of th~
Armada (Lx, p. 196). It is remarkable in two ways. Referring a
it does to one of the fortresses of the Guadalquivir valley, it i
virtually a frontier ballad (romance fronterizo), and was readih
confused with the oldest piece of that kind, which was also a Sieg<
of Baeza (1407) (71). Thus the civil wars merged into the borde:
wars, and the series of ballads is unbroken. The other featur<
is that the ballad is assuredly the work of an eyewitness. Th<
skirmish was of too little consequence to be mentioned by the
historians; but the ballad gives precise information as to persons
and places, and by calling the king Pedro Gil it gives tongue to the
contemporary slander that he was not son of his father but of Dor
Juan Gil de Alburquerque, his mother’s faithful supporter. Suet
a slander was in the interest of the Trastamaran bastard only sc
long as the claims of Pedro’s house were not united to those of the
usurping dynasty; as they were by the marriage of Enrique III and
Catherine of Lancaster in terms of the treaty of 1390. A jolly ballad
of The Prior of St. John (69) represents Pedro as a baffled tyrant,
and has not the bitterness of the contemporary pieces. It must have
existed by 1455, when a note concerning the alleged event was
inserted in the Fourth General Chronicle of Spain.

The siege of Baeza, attacked this time by Yanegas and defended
by Pedro Diaz, opens a long series of ballads which extends from
the year 1407 to the date of Don Alonso de Aguilar’s death (1501).
They are divided into two series. The earlier is of isolated forays,
triumphs and disasters (71-83), without any co-ordinating design!

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA x6x

In the later series (84-96) the background is the War of Granada,
but the heroes disport themselves in a merely chivalresque fashion.
There Is an element of the conventional in the later pieces, with
idealized Moors and polished Christians. It has fallen also under
the Influence of a singular genius, Gines Perez de Hita, whose
romanticized history of that war shines through Washington
Irving’s prose for English readers. He is our leading source for
several such pieces, and he was not incapable of writing a quite
plausible 4 traditional 5 ballad of his own Invention. It is rather in
the earlier group that we find the classic conditions of balladry.
These are the songs of small communities, intensely preoccupied
with their own immediate dangers and successes. Local names—
Sayavedra, Bishop Gonzalo, Fajardo, &c.—are those that effectively
matter. The nation is taken for granted along with the division
of religions, but these are not tales of a national drive. As in other
lands, the national question is reduced to its simplest form: the irre¬
ducible antithesis of Moor and Christian. Kings and great nobles
appear in the ballads only when they happen to be on the spot;
otherwise royalty is a part of the remoter background. The ballad
public was completely homogeneous, but it was no plebs. It had its
leaders, and the poets sang of and for these leaders and the gentlemen
who lent their swords. A Swedish ballad says of similar raiders:

Them shall men praise
in courtly lays

mid knights and dames.

Such courtliness is to be found in the Castilian frontier ballads:
instinctive good-breeding in the songs of persons well-bred.

An average ballad of the series—not the dashing Verdant River
(96 a), which Percy found in Perez de Hita and made famous in
Europe—concerns the fall of Antequera in 1410 (74). It begins:

Fled the Moor from Antequera
three long hours before the day,
carried In his hands his letters
praying earnestly for aid;
blood in place of ink was written—
not that ink was wanting there.

Moor that bore the hasty missive,
doubly sixty years of age;
white his flowing beard as silver,
shaven pate that shone like day,

*62 ROMANCE BALLADS

with his turban wound about it,
precious turban, richly made,—

’twas embroidered by a Mooress
whom he kept as well-loved dame—
over all a scarf he carried,
silken-tasselled, fine and gay;
riding on his mare for swiftness,
of his charger was not fain.

Lone he travelled, having only
for companion a page;—
not for any lack of squirelings,
many in his palace stayed !

Seven the ambushes they set him,
horsemen bringing him to bay;
but that mare was light and nimble,
and through all she made her way,
through the fields of Archidona;
shouted he, and thus did say:

‘Good my lord and king, if only
knewest thou my tidings grave,
thou wouldst tear thy locks in handfuls,
tear thy beard in dire dismay.’

The finest pieces are sometimes those that recount Christian
disasters °ne such occurred on the Rfo Verde in 1448, when the
bold soldier Sayavedra fell in by chance with an enormously
arger Moorish expedition. The event was confused with the
similar death of Don Alonso de Aguilar in the same region in x501
(96). Another confusion affects two expeditions which set out
from Jaen: one, of three hundred

boys of honour only greedy,
nay, more truly, boys in love,

who were trapped at Montejfcar on 10 May 1410; the other, of
our hundred who sallied forth under the guidance of the fighting
bishop Gonzalo at a date not precisely known (82). Incidents on
the frontier were liable to repeat themselves, and were seldom of
such importance as to deserve other witness than the ballads. The
poems constituted a stirring, but perishable, record; yet there were
fortunately some historians, like an anonymous chronicler of

ST Si t° t0 bdkdS t0 enrich their ^"nation. Thus it
is that the chronicle gives us the very day of the conversation which

makes up the magical ballad of Abendmar (78). It was on the ^th

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 163

June 1431, in full sight of the longed-for city, that King John II
conversed with the Moorish suppliant prince Yusuf ibn-Ahmar.
The ballad has touched the scene with the gold of poetry. The
Moorish prince is aglow with mysterious portents. The city lies
unfolding its beauty: Alhambra, Alijares, Generalife, Torres Ber-
mejas. The king is suddenly moved to address the city as a bride
apparelled for her marriage; but the city replies she has already
chosen her bridegroom:

Up and spake King John unto her,
well you’ll hear the word he said:

‘Wert thou willing, O Granada,

I and thou today should w^ed:

Cordoba and all Sevilla
give thee for thy bridal bed.’

‘Nay King John, for I am married,
wedded wife, not widowed;
by the Moor who me possesses
lovingly am cherished.’

A vignette of a governor of Murcia who played chess for cities
(83) is another example of Castilian sobriety enlivened by Oriental
imagination. Just as the metaphor of the city as a bride is Oriental,
so the staking of cities or kingdoms on a game has its parallels,
notably in the famous game played between Ibn Ammar of Sevilla
and Alfonso VI in the eleventh century, which, according to Al-
Marrakasi, led to a complete withdrawal of the Christian forces
from the kingdom of Sevilla.

With the final campaigns directed by the Catholic Monarchs
from 1481 to 1492 there comes into the frontier ballads a greater
monotony. The genre had had its day, and was no longer in a state
of pristine freshness. A succession of Moorish champions issue out
of the city gates: Barbarfn, Albayaldos, Alatar, Muza. They are
defeated, one and all, in single combat with Christian champions.
In this atmosphere of chivalry war seems a tournament; its
tragedies are ignored. Perez de Hita saw how the convention could
be exploited, and he produced round and through these ballads
the atmosphere of sentimental chivalrous melancholy—with its
Abencerrajes and Boabdils—which the name of Granada now
inevitably suggests. All that is needed is that the champions who
die in the Vega should be hopelessly in love, and, for preference,
each the last of his line. The Chateaubriands and Washington

164 ROMANCE BALLADS

Irvings have done the rest. Hopeless, romantic passion (and not
the savage faction fights of history) has become the mark of the
Moor in literature. In the ‘morisco’ ballads of the seventeenth
century the pretence of historical accuracy was dropped, and any
Spaniard could pour out tender reproaches in the streets of
Madrid by simply calling himself Zaide, Tarfe, or Muza. This is
not the atmosphere of the ballads of the Granadine War, though it
can hardly now be separated from them. They were, despite their
conventions, more sturdy, and one of them, in Byron’s spirited
rendering, has become world-famous (85):

The Moorish King rides up and down,
through Granada’s royal town;
from Elvira’s gate to those
of Vivarambla on he goes,

Woe is me, Alhama !

Letters to the monarch tell
how Alhama’s city fell:
in the fire the scroll he threw,
and the messenger he slew.

Woe is me, Alhama !

He quits his mule, and mounts his horse,
and through the street directs his course;
through the street of Zacatin
to the Alhambra spurring in.

Woe is me, Alhama !

It is one of few ballads composed for the defeated party, and one of
the few with a refrain. The style is nervous, but not dry; every
action is personal and vivid, and is expressed with more swiftness
in Spanish than in English. A Moorish faquih levels reproaches
at the King and the sense of disaster thickens, until the ruin of the
kingdom is made to appear so close as to be irrevocable:

By thee were slain, in evil hour,
the Abencerrage, Granada’s flower;
and strangers were received by thee,
of Cordova the chivalry,

Woe is me, Alhama !

And for this, oh King! is sent
on thee a double chastisement;
thee and thine, thy crown and realm,
one last wreck shall overwhelm.

Woe is me, Alhama 1

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 165

There are hundreds of historical ballads which arose after 1501,
serving as news-bulletins for the great events of the sixteenth cen¬
tury and for more vulgar occurrences later; but they belong to the
age of print and do not enjoy the welcome into oral tradition of
the older pieces. Among these, in addition to the themes of civil
and frontier warfare, a few miscellaneous ballads were put into
circulation. They are interesting for several reasons. Isabel de
Liar (103-6) has been explained as a castilianization of the tragic
history of Inez de Castro, celebrated by Garcia de Resende and
Camoes, and, together with the ballad of the Duchess of Braganza
(107) and the Duke of Guimaraes (108), is evidence of the interest
that Portuguese affairs began to rouse in the neighbouring land.
Other Castilian ballads refer to Aragon and even Naples (98-102).
The Duke of Gandia's Death (27 July 1497, ix, pp. 205, 207) was
the subject of a bulletin in verse which was treasured by the exiled
Spanish Jews and forms part of their oral wealth; it is proof that
all contact with their homeland was not immediately severed. The
Death of the Prince of Portugal (ix, p. 204) concerns the premature
decease of Dom Afonso, heir to the Portuguese and Castilian
crowns, In 1491. Fray Ambrosio Montesinos, who later became
a bishop, was instructed by the Catholic Queen to make a ballad
on this subject, which shows what latitude must be given to the
term ‘popular 5 In dealing with balladry. Another form Is more
lyrical, and begins with a cry of grief, which is repeated after each
four octosyllables:

Ay, ay, ay! what bitter sorrow!

ay, ay, ay! what bitter grief!

It is not certain which form is the older. The second is one of
many instances In which ballad and lyric coincide. 1 Jovellanos dis¬
covered another in the famous ‘danza prima 5 of the Asturias, and
other Instances are I was going , dear Mother , to Villareale (ix,
p. 269), The Comendadores of Cordoba (x, p. 372), The Serrana de
la Vera (ix, p. 209), Yes , yes , Antequera ’s fallen (xii, p. 134), and
Moriana {Romances y Baladas , p. 7). Some of these are parallel-
istic; they are all evidence of the process of incorporation whereby
the ballad form engulfed material previously treated in other ways.
Thus the uniformity of the Spanish Romancero, one of its most

1 R. Menendez Pidal, Romances y Baladas, presidential address to the Modem
Humanities Research Association, printed in M.H.R.A., 1927.

166 ROMANCE BALLADS

distinguishing characteristics, proves to be in part secondary, due
to the power of oral tradition to remodel its matter.

The word ‘historical’ has been used by the Spanish authorities
to cover ballads of historical themes previous to 1350; but the
usage is ambiguous and has led to confusion of thought. There are
ballads, it is true, which purport to relate episodes from Spanish
history from the fall of the Gothic kingdom in 711, and these
accounts were either true or believed to be so. But the ballads do
not, like historical ballads properly so called, arise directly from the
events which occurred or were imagined. The proper form for
traditional narrative verse previous to 1350 was the epos. Extant
are a Poem of the Cid (composed about 1140), The Cid’s Youthful
Feats (of the late fourteenth or early fifteenth century, replacing an
epic of the thirteenth) and one hundred lines of Roncesvalles. The
Seven Infantes de Lara is reproduced in two prose forms: the older
in the First General Chronicle of Spain (1289), the second in the
Second General Chronicle (1344). The poetical text is followed with
such closeness that it is easy to reconstruct entire tirades; and the
best authority is satisfied that he can reproduce for us some two
hundred lines from the two chief scenes substantially as the reviser
left them. Difference of plot and of assonance prove the existence
of two poems, the ballads corresponding to the second one. With
similar fidelity a tale of the Siege of Zamora has been preserved to
us since 1289, and has taken its place among the sources of ballads
on the Cid. Count Fern an Gonzalez, who died in 970 and was the
■virtual liberator of Castile, was the subject of an epic poem which
was remodelled in rhyming quatrains by the author of the Poem of
Feh nan Gonzalez about 1250, and it was this rhyming poem which
was used by the chroniclers in 1289. Some ballads are independent
of the rhyming account and correspond better with the prose of the
chronicle of 1344. It is fairly certain that they rest on the traditional
epic, now lost. An epic of Bernardo del Carpio existed and was
followed by the chroniclers of 1289; but it was followed only in
part, since the chroniclers preferred to rely, whenever possible, on
the Hispano-Latin historians who were their immediate prede-
cessors. Again it is probable that some ballads give us the epic
text though few of them are really old. As for the oldest matter—
the fall of the Gothic monarchy in 711—it was developed through
the pages of Hispano-Arabic, Hispano-Latin, and Castilian
chroniclers until it reached its ultimate form as the historical prose

SPAIN, SPANISH JEWRY, PORTUGAL, IBBRO-AMERICA 167

romance called the Cronica Sarracina of 1430. All the ballads of
the cycle rest on this work. There is very little reason to believe
that there ever was a Castilian epic poem on this subject, though
there are one or two epic touches in the prose of chroniclers and
romancers. It did, however, give rise to a French epic, Anseis de
Karthage of the thirteenth century.

It is not necessary to discuss the earlier dates which may be
assigned to these epics, since it is only their later metamorphosis
into ballads (. Primavera , 2-62) that concerns us. The chroniclers
of 1344 used the texts of still lively ‘cantares de gesta 5 . It has been
argued, though it is less certain, that the chroniclers of the fifteenth
century continued to draw directly on the epic texts. In either case,
it is true that the contact between epos and ballad is closer in
Castile than in any other land: the epics were at their best in the
twelfth century, and still luxuriant in the first half of the fourteenth;
in the second half of the fourteenth there were ballads. There is
not evidence, however, to show that the metamorphosis occurred
at once, so that the epical ballads were older than any other ballads.
On the contrary, all those dealing with the fall of the Goths are
necessarily later than the year 1430. The change may well have
been a slow one, and epos and ballad may have overlapped for a
considerable period. The style of the two genres is very similar.
Ballad-makers could use most of the epic lines, assonances, and
technique; they could also use the epic language, when not too
archaic, and they necessarily presented pictures of an older age.
Thus there are reasons for calling the epical ballads ‘old’, not
merely in the sense of Nebrija in 1492 (who meant no more than
£ as old as his own childhood 5 ), but in a more absolute sense. Their
unselfconsciousness contrasts with the ostentation of a simple art
in the ballads of the Carolingian cycles, which are considered to be
‘ballads of minstrelsy 5 (juglarescos). The logical antithesis to this
term, however, is not ‘old ballads 5 (viejos). The critic who opposes
‘romances juglarescos 5 to ‘romances viejos 5 begs the question of
age; he is opposing a style to a time, and there is no real antithesis.
The ‘old 5 ballads of epical origin may indeed have arisen later than
some of the ‘ballads of minstrelsy 5 .

The association between epics and ballads is close. The Spanish
school insists on drawing the bonds as close as possible. The
ballads are, according to them, fragments of epics; indeed, they go
further and consider that fragmentation is the essence of the

i68 ROMANCE BALLADS

‘romancero’. There is no doubt that the ballads, as we have them,
are more fragmentary than once they were. The compiler of the
Cautionero de Amberes consistently gave his readers truncated ver¬
sions, probably because the fashion of singing them at social meet¬
ings had led to overcrowding in the programmes, overcrowding
much intensified by polyphonic settings. A fortunate discovery
among the Jews of Morocco has proved that the Count Arnaldos
poem was at least twice as long (and not a tithe as effective), and
the cycles of the Marquis of Mantua and Montesinos were probably
at one time single long ballads. In their present form Spanish
‘romances’ are the shortened and broken relics of older and longer
poems; but in most cases those poems were also ballads. This is
necessarily true of the historical pieces we have discussed in former
paragraphs: they never had any other form than that of ‘romances’,
apart from which they would not have worked their proper effects!
It is also true of those Carolingian ballads which summarize, but
do not reproduce, the plots of French epic poems. In the cycles
of Aimer i de Narbonne, Montesinos, and The Marquis of Mantua,
the narrative takes certain short cuts to a ‘denouement’, decimating
the lines and episodes of the epics. To derive the ‘romances’ from
epics by a process of fragmentation is therefore an explanation, if
valid, of no more than a small part of the whole ‘romancero’.

Even in respect of that small part it has not been possible to
reach certainty. The greater number of epical ballads have to be
written off as ballads of minstrelsy’. Those concerning the Gothic
defeat (2-7) all fall within this class, and we further know that they
are based on no epic poem, but on the Crdnica Sonatina of 1420.
This does not prevent them from boasting one of the most vivid of
romances , that lament which Lockhart versified with such spirit as

Last night I was the King of Spain, this night no king am I. (5)

In other cases we are not able to carry the argument into detail
because of the loss of the early poetical texts. That is true of the
ballads of Bernardo del Carpio (8-14), Ferndn Gonzalez (15-18),
Infantes de Lara (19-26), and Siege of Zamora (35-54). An ex-
ceptional argument can be urged in favour of the Lara group viz
that the prose of certain chronicles is a close approximation to
poetry, and that the ballads of the cycle bear all the marks of
authentic epical transmission. However that may be, the absence
of the epic originals prevents our reaching an absolute assurance.

SPAIN, SPANISH JEWRY, PORTUGAL, IRERO-AMERICA 169

On the other hand, though one old and one decadent epic exist
concerning the Cid, by far the greater number of ballads must be
written off as 'ballads of minstrelsy 5 , or even sheer inventions. In
the few cases where the two texts can be placed in parallel columns,
there are as many divergences as similarities. Fragmentation alone
would not account for the only two which can be rigorously com¬
pared; the material and the style have been reshaped. 1

By their contact with epical poems the Spanish ballads gained
many things. They gained in gravity, energy,' imagination of a
literal kind, dramatism, and nationality. The old Castilian
‘gravitas 5 freed them from the triviality which is a prominent
feature of other balladries; even their international pieces have a
more sober mien. As a repository of all that concerned the Castilian
spirit and as a veracious account of Spanish history the 'romancero 5
attained a unique authority. It was able to shape a considerable
amount of the classical literature of Spain, especially the drama,
and thereby to extend its influence beyond the frontiers. This
authority, exerted through one style instantly recognizable though
dispersed among thousands of ballads, has caused the Spanish
corpus to be the most perfect example of a 'romancero 5 —of short,
oral, traditional narratives in verse, collectively forming one whole.

Precisely on account of his popularity, the Cid has not proved to
be a conservative figure in the 'romancero 5 . The ballad-mongers,
as they receded farther and farther from the Castilian heroic age,
lost sight of other heroes and saw only the Cid. The 205 ballads of
the Romancero del Cid (edited by Carolina Michaelis, Leipzig,
1871) are mostly late and arbitrary. In the other cycles many
ballads have to be considered comparatively late medieval produc¬
tions, from the end of the fifteenth century or the beginning of the
sixteenth; but those of the Infantes de Lara (19-26) stand out for
their exceptional apparent fidelity. It is from them, after all, that
we must glean our best idea of the old 'cantares de gesta 5 : they give
us, as prose cannot, the rush and fury of poetry. Their language
is starker, fitted to the stark ferocity of the action. It is in them that
we hear the actors speak out their jealousies and hates. In two
supreme scenes the language of the chroniclers is poetry thinly
disguised: the episode of the heads, and the conclusion. The
episode of the heads gives a terrible ballad (24). The first words
show that the poet is following the story as it is contained in the

1 S. G. Morley, Spanish Ballad Problems, California, 1925.

461s z

1 ~° ROMANCE BALLADS

revised epic (prosified by the chroniclers of 1344). They also con¬
tain what may be an actual historical date from the lives of the
infantes ; the day on which the Castilian invasion became known
to Almansur at Cordoba. The Moor Alicante rides into the capital
with the heads of the seven ‘infantes’ and their tutor in a kerchief.
Almanzor, according to the epics and ballads, took the cloth to the
prison and left it to be unrolled by their father Gonzalo Gustos
The old man’s grief was terrible. One by one he took up the heads
and lamented for his sons, remembering incidents of their young
manhood These references are more precise in the epos as
recorded by the chroniclers than in the ballad. The epical style is
circumstantial and leisurely; the ballad style summary and
dramatic. The distance between the chronicler’s text of the
denouement’ and the last of the ballads (26) is much greater. As
it now stands, this piece has omitted everything but the words
spoken when Mudarrillo surprised his enemy Ruy Velazquez. Two
great poets have traversed the same ground and have been worsted
by the anonymous minstrel (or by his effort as sublimated by
tradition). Lope de Vega, in his Bastardo Mudarra, has avoided too
close a contact; his treatment is romanesque and torpid, and the
ballad-poet proves to have been more dramatic than the dramatist
more tragic than the tragedian. Victor Hugo’s Don Rodrigue est d
la Chasse Orientates 30) is a bit of Asiatic rhetoric, not devoid of

,' dlt f y due t0 “ lsaid ‘ local colour’. Where two such poets have
fa ed, it is not likely that any rendering will succeed; but the

Gone a-hunting is Rodrigo,
and of Lara is his name,
in the heaviness of midday
underneath a beech-tree laid,
heaping curses on Mudarra, ’
son unto the Renegade—
that his soul (could he but seize him)
from his body would he tear.

Thus that lord so fiercely muttering,
Mudarrillo doth appear.

Gentle sir, may God be with you
underneath the beech-tree laid.’

Welcome are you here, good squire,
unto you I wish the same.’

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 171

‘Gentle sir, now tell me truly,
how it is men style your grace ? 5

‘Men do call me Don .Rodrigo,
and of Lara is my name,
kin am of Gonzalo Gustos,

Lady Sancha’s brother am,
who begat my seven nephews,
seven who from Salas came.

Here I wait for Mudarrillo,
son unto the Renegade,
and his soul (were he before me)
from his body would I tear . 5

‘If they call you Don Rodrigo,
and of Lara is your name,

Fm Gonzalo’s son Mudarra,
born unto the Renegade,

, son of good Gonzalo Gustos,

Lady Sancha’s step-son made,

who begat my seven brothers,

seven who from Salas came,

traitor! whom thou soldst to slaughter

in Arabiana’s vale;

here thou leavest thy foul spirit,

if but God come to my aid ! 5

‘Nay, but tarry, Don Gonzalo,
let me take my coat of mail . 5

‘Such a tarrying as thou gavest
them who erst from Salas came;
here thou perishest, Sir Traitor,

Sancha’s enemy, today ! 5

The ballads of the Cid, on the other hand, if late and seldom
authentically epical, have exerted a powerful influence through
their mass multiplied by their popularity. It is from them that the
figure of the youthful Cid arises in opposition to the grave states¬
man of the old Poem of the Cid. The young Cid is all fire and dash.
He is untameable. To do homage to his liege lord seems to him a
humiliation; to show respect to foreign king or pope an outrage to
be at once avenged. He is not a lover, since love would curb his
unbridled will. As the embodiment of reckless force of will the
young Cid is the Don Juan of the Middle Ages; the Don Juan of
the classical period is inferior to the young Cid inasmuch as his field
of conquest is so much more trivial. He resembles those great

x 7 2 ROMANCE BALLADS

nobles of the later Philips, full of great ambitions without adequate
outlet owing to the jealous royal policy. If the young Cid is Don
Juan the old Cid is Don Quixote; but a Don Quixote not frus¬
trated, not tilting against windmills. He seeks justice, but he
obtains it; he releases not galley-slaves, but loyal vassals unjustly
suspected. The two immutable portraits of the Spanish mind thus
emerge from the romancero’ of the Cid, less complete than when
later sketched with all the increased literary power of the Renais¬
sance, but more masters of their fate.

Ci f “ n0 ^. a , lover; y et k is from the ballads that there
rises the idea of the Cid as a lover, and so of the tragic conflict

couTap 2 q I s dUty - He had been educated * the royal
thus hi! h- UnC f Sa . ncho - Urraca > who was somewhat older, had

W P yma ?’ and When he appeared in arms before

er fortress of Zamora, she reproached him with forgetting his

thin k ! ndlmeSS - ?° far the text of the chroniclers tak!s us.Sut

r ba f P ° et , g0eS further and i ma giues that the Cid’s match with
Jimena implied passing over the princess-

Twas my father gave you armour,
and your horse my mother gave,

I did gird your spurs upon you,
that more honour you might have;
for I thought we two should marry,
but my sins have said me nay,
and you wed Jimena Gomez, ’
daughter of the County Gay;
with your wife you’ve won you money,
but from me you’Id have estate;
noble, Roderick, is your marriage—
you m ight have had a princely mate! (69)

ppTeS

for her sake he refused the finest match in Castile Tn m f ’ ! I
with the obhgation in honour to kill Jimena’s father was theTork of
^^il^. en d^ Castro; to universalize the conflict of love and honour to
tragedy on alternatives of duty, was Corneille’s finishing touch

tZ since Reynald of Mo ^ auba " ^ aa

S aS R0land ‘ The moment preserved is that in which

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 173

Charlemagne returns to the field to scrutinize the dead. These
characteristics of a thirteenth-century epic are to be found in the
ballads of King Marsh's Flight (ix, p. 245) and its derivative (183).
It is a fragment of some longer poem, and the moment is the rout of
Marsilie’s forces. Roland dare not blow his horn In case he excite
the contempt of his rival Reynald. The narrative is hispanized, and
includes a technical term of Peninsular warfare unlikely to be
known In France; In this piece, therefore, we have probably a
second fragment of Roncesvalles or a derivative poem founded on it.
Lady Alda (184) may be another fragment. It is a romantic and
pathetic piece in which Lady Aude’s heart breaks to slow music,
as in the rhymed Roncesvaux, Other ballads (180-6) ostensibly
relating to this battle are not of epical origin. Some one made
Roland’s sword into a hero, Durandarte, 1 and endowed him with a
romantic affection for Belerma. He dies, with tender reproaches
on his lips, in the lost battle:

Oh Belerma, oh Belerma,
for my sore affliction bom!

Seven years I truly served thee,
but from thee have nothing won.

I unhappy (now thou lovest)
perish in this rout forlorn.

Not my death is such affliction—
though death call me all too soon,—
but it grieves from thee beholding
and thy service to be torn. (181)

The ballad of Guarinos y Admiral of the Sea (186), is a tale of escape
from captivity, of uncertain origin. It uses many of the common¬
places of the chivalresque epos, and the name is that of Garin
d’Anserine, a deuteragonist in many "chansons de geste’. The first
words connect these adventures of his with the battle of Roncesvalles:
Evil fortune saw ye, Frenchmen,
in the Chase of Roncesvaux;
where your Charles has lost his honour,
died his dozen peers also.

1 In Ogier de Dinamarche one meets the line (describing Courtain, Ogier’s
sword):

qui moult ce tint valt mains que Durendal.

The comparison is between two swords, Durendal and Courtain; but, according
to the syntax, it might be a comparison between two heroes, Durendal and the
wielder of Courtain. Some such confusion may have originated the Spanish
Durandarte, if he is not an arbitrary creation of fancy.

i 74 ROMANCE BALLADS

These words were encountered in Russian, in distant Siberia in
1834, thanks to Karamzin's translation executed in 1789, which
became, on this account, ‘the birth-year of the Russian ballad’. 1

Epical ballads form the most important class of ‘romances’
dependent on previous works of literature, both because of their
intrinsic merit, and because they continue the central tradition of
Castilian thought. Other literary suggestions came from France;
particularly the Arthurian and Carolingian novels. The former are
a small closed group of three ballads. They arise directly out of the
prose texts of Lancelot and Tristan, probably in the Spanish transla¬
tions executed at the turn of the thirteenth and fourteenth cen¬
turies. That can be affirmed positively concerning the ballad of
Tristan's Death (146). It is a laconic summary of three chapters of
the still extant novel, reproducing some of the phrases. In later
versions this ballad became entangled with the superstition of
the herb of pregnancy, which belongs properly to Doha Ausenda
(x, p. 105). The resurrection of dead lovers as entwining trees is a
motif present in this ballad as also in Count Olinos (x, p. 72). When
these accretions are removed, we are left with the substance of the
old romance. Two ballads deal with Lancelot (147,148). They are
probably parts of the romance translated as Lanzarote de Lago, of
which only the second section survives in an unpublished manu¬
script. It would be between the close of that part and the commence-
ment of the Grail that we should expect to encounter the adventure
of the White-footed Stag (147), which is akin to the French Tyolet
and to an episode of the Dutch Lanceloet . The assonances change
either three or four times in the course of the ballad; the extra
assonance was recorded by Nebrija who, in 1492, mentioned the
ballad as an old one. These changes, since they cannot refer to
laisses in an original which was in prose, probably correspond to
chapters of the original. Formerly the ballad must have been longer
than it now is, because the introductory lines are quite inexplicable.
This ballad speaks of a Dona Quintanona or Lady Centenarian,
who is a character of the other Lancelot ballad. In this piece she
stands in place of Lady Malehaut as the lovers’ go-between. For
the rest, the ballad reproduces no specific incident, but a general
impression of the challenges, excursions and returns to Guenevere’s

1 vP as J a ^ r ist also das Geburtsjahr der russischen Ballade, allerdings
der Ubersetzungsballade. ’ F. W. Neumann, Geschichte der russischen Ballade ,
Komgsberg, 1937, p. 27.

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 175
affections which constitute the stuff of the Lancelot story.
Cervantes particularly cherished the opening lines:

Ne’er was knight so nobly tended,
nobly served by gentle dames,
as was Lancelot the goodly
when from Brittany he came.

Matrons stood and waited on him,
damsels waited on his jade,
and that Lady Centenarian
skilfully his liquor strained,
and Queen Guenevere the lovely
by her side the hero laid. (148)

The Carolingian ballads are far more numerous and elaborate.
They cannot be made into a closed group, because of the Castilian
custom of attracting into this cycle all romantic narratives. Some of
them should thus be termed semi-Carolingian, rather than Carolin¬
gian, and even so there remain others uncertainly attached. The fact
is that almost all the adventure ballads current in Castile were of
foreign origin. The genius of the land was to form veracious
historical statements, or at least such as could lay reasonable claim
to historicity. Mere fabling was at a discount. But in France there
was a lively imagination at work to produce effects which, even if
they had some historical sanction for Frenchmen, were mere novels
to Spaniards. The Castilians made no effective discrimination
between the pseudo-historical adventures of Charlemagne’s peers
and those of anonymous heroes. The poets knew France as a land
with one city (Paris) and one emperor (Charlemagne). It was also
a land where proper names commonly ended in -os, as Oliveros,
Montesinos, Guarinos, Gaiferos, Calafnos, Carlos: a curious sur¬
vival of the Old French nominatives in -s. Now it was from French
originals and intermediaries that they knew all, or almost all, they
learned about the balladry of the rest of Europe. Signs of the
im portation are that the action of such ballads is often said to be in
France (though Aragon may be mentioned as a land lying between
Castile and France and probably actually traversed by these ballads
in their extension to Castile), that proper names are of a French
sort, that there may be mention of an emperor, and that there is
some tendency to conform to the French rule about the unequal
cadences of hemistichs in a long line. While it is not impossible
for an adventure to be wholly castilianized, romantic adventures

ROMANCE BALLADS

under Spanish names and In Spanish places are both rare and
short.

A sub-group Is formed by the cycles of the Marquis of Mantua
(165-7), Montesinos (175-6), and Almerique de Narbona (196),
because in each case we are able to identify the remote epic original,
though there was probably some nearer source in chap-books.
These ballads descend from the French Ogier , Azol , and Aimeri de
Narbonne ; in each case there is some datum in the epics which
makes for length, and which has been suppressed by the ballad-
poets. In Ogier it is that, when he demands satisfaction for the
murder of his son, Charlemagne refuses it. That is the source of
the long war in Italy, ending with Ogier’s flight and imprisonment.
According to the ballad-maker, Charles at once set up a court to
try the offender, who was his own son Chariot. Aiol had to
avenge his parents on a powerful faction at court, and consequently
devoted his efforts at first to conciliating the king and gaining sup¬
porters : Montesinos goes to the palace (which he can see from the
Pyrenees!), overlooks a game of chess, accuses his enemy of cheating,
and kills him on the spot. He kills him with a chess-board, thus
taking a hint from Doon de la Roche, along with the villain’s name.
In the third case, the epic siege of Narbonne was complicated by
the capture of Aimeri, whose release was purchased by the sur¬
render of the town, though the citadel remained in French hands;
there was thus no ultimate surrender of the fortress for the sake of
the hero, and the ballad-maker goes straight to the point by causing
him to reject every proposal for exchange. By these means it was
possible to summarize for the populace the essence of long epic
poems, doubtless already abbreviated in intermediate sources. The
same kind of correspondence links the poems of Reinaldos de
Montalbdn to the prose romance about his Empire of Trapizonde
(187-9), Calainos to Fierabras (193-4), Melisenda to Amis et Amiles
(198), and Valdovinos to the Chanson des Saisnes (169). Gerineldo
(161) appears to rest on the tradition of Eginhard’s amours at
Charlemagne s court. It is one of the most popular of all Spanish
ballads, and commonly goes united to the international motif of the
Prevented Remarriage (135).

The semi-Carolingian group includes such important ballads as
Couni Alarcos (163), Count Dirlos (164), and Gaiferos (171-4).
They sire international ballad themes thrust into a Carolingian con¬
text. The first depends on the motif of the wicked princess who

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 177
forces a man she admires to put his wife to death. So stated, there
is material for only a short poem, and so it appears In Piedmont
(Nigra 6); but delaying and elaborating the incident the Spanish
minstrel has contrived to make It one of the longest that survives
from ancient date. The longest is Count Dittos , and for the same
reason. It is merely the Noble Monnger motif, of German origin,
but elaborated with all possible Carolingian pomp of names. As
for Gaiferos , who

quite forgot his lady free,

the case is more complex. Essentially we are concerned with an
escape from imprisonment by a hero and a lady: that is the theme
of the old Germanic epos of Walter of Aquitaine . In that poem,
however, Walter and Hildigund are both hostages at Attila’s court,
but in the ballads of Gaiferos he is free and at his ease in Paris
when the action opens. People remind him that he has a lady-love
languishing in prison in Sansuena (which Is both Zaragoza and
Saxony in the ballads); he makes his way thither, recognizes the
lady, leaps the walls with her when the Moor closes the gates, and
fights a running battle all the way to the French frontier (effica¬
ciously aided, at Maese Pedro’s puppet-show, by the redoubtable
Knight of la Mancha). So it is in three ballads (171-3), but in a
fourth it is Gaiferos himself who is escaping from prison:

On the very stroke of midnight,
when the cocks began to cry,
very secretly Gaiferos
issued from captivity. (174)

This is somewhat incoherent, and belongs to the traditions of
Walter. The other adventure is the stuff of the Catalan-Proven?al
ballad of L’Escriveta, but served earlier for the French romances of
Aye d’Avignon and Comtesse de Ponthieu. The deeds of Gaiferos
are not directly derived from any of these three sources, but
probably from one now lost. His name suggests a Gaiffier of
Aquitaine, substituted (with deference to certain noble persons) for
Gautier, the Germanic Walter.

A more recent German epos to penetrate into Spain is Kudrun.
This gives the ballad of Don Bueso (the name is French), a deriva¬
tive of the ballads, perhaps, rather than the thirteenth-century
epic (x, p. 56).

As we approach the domain of narratives which have never been

178 ROMANCE BALLADS

other than simple ballads, we have still to take note of French in¬
fluences, and even of those of literary convention. The convention
of ‘amour courtois* is found in The gentle Lady and boorish Shep¬
herd (145) and the Infantina (151). In the former one critic has
thought to see the beginning of Elveskud, but for this there seems
little justification. There is no evidence that we have to do with
elves, and the strongest evidence that we have an inverted ‘pastour-
elle\ The boor does not recognize his luck when he sees it. An
amusing parallel to the Spanish ballad is the English doggerel
ditty:

There was a lady loved a swine,

‘Honey, 5 quoth she.

Tig-hog, wilt thou be mine ?’

‘Humph! 5 quoth he.

‘I will build thee a silver sty,
honey, 5 quoth she,

‘and in it shalt thou lie . . . 5
‘Humph! 5 quoth he.

‘. . . pinned with a silver pin,
honey, 5 quoth she,

‘that thou mayest go out and in. 5
‘Humph! 5 quoth he.

‘Wilt thou now have me,
honey? 5 quoth she.

‘Humph, humph, humph! 5 quoth he,
and away went he.

It is also the theme of Henryson's Robin and Makine , which has a
conclusion more in keeping with the second ballad. This is con¬
cerned with the tardy gallant who makes his proposals too late,
and is mocked by the lady, as in the French Occasion Manquee.

The ‘maumariee 5 motif appears in Blancaniha and Albaniha
(136), but it is treated with intense emotion. The Spanish ballad-
poet does not think adultery amusing; persistent cruelty explains it,
without justifying it. Consequently, there is no immunity granted
to the lady in this ballad; but after some swift questions and
answers her guilt is evident and she herself asks her husband to
kill her. Allied in theme and treatment is the ballad of Bernal
Frances {x ii, p. 502). It occurs also in Piedmont (Nigra 30). The
identification with Bernard of Septimania, executed by Charles
the Bald in 844 on suspicion of intercourse with Queen Judith,

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 179
seems adventurous. The Italian name 4 re Inardi resembles
‘Bernaldino’ or ‘Bernardin’ closely enough to guarantee the
identity in all Important respects of the Spanish and North Italian

traditions.

More fugitive conventions are represented by the May theme
and the power of song. Gil A icente gives us a short rhapsody,
which Is the May song in a very simple form:

This is the May, the May is this,
this is the May and all aflower.

But the Spanish tradition, which appears in the ballad of the
Prisoner (114) and elsewhere, depends on the set descriptions of the
season which are to be found interspersed In French 'chansons de
geste 5 , as well as in the prologue to the Canterbury Tales and In
Greece (Politis 235, &c.): that is, It proceeds by stages from the
warm season, through flowers, birds, animals, to human lovers.
This Is not the manner of the medieval lyrical ‘reverdies’, but more
epic in tone. It serves as prologue in Spanish to a narrative about
some prisoner which has, fortunately, been truncated. The con¬
trast is thus the more poignant:

Oh J tis May, the month of May,

when the season’s heat is high,

and the larks above are singing

and the nightingales reply,

and all lovers are a-running

on love’s errands far and nigh;

all but me, afflicted, wretched,

that in prison-house do lie;

neither know r I when day cometh,

nor w r hen night is passing by,

were it not for one wee birdie,

singing when the dawn is nigh:

but an archer slew my birdie—

may he earn God’s curse thereby! (114)

The other motif appears in Count Olivos (x, p. 72), which is a
fine ballad, but yet inferior to Count Arnaldos (153), which many
would reckon the flower of the ‘romancero’:

Fain would I have had the fortune
ocean’s rolling waves upon,
fortune that befel Arnaldos
on the morning of St. John!

i8o ROMANCE BALLADS

On his wrist a hooded falcon,
was the Count a-hunting gone,
when he saw a stately galley
just about to reach the shore.

Silken all the sails she carried,
and of sendal rope and thong,
and a sailor who was steering
chanted, as he sailed, a song:
song that hushed the sea to stillness,
quieted the wild winds down;
fish in the abysses swimming
it beguiled to swim above,
birds in heights of air a-flying
charmed to rest his mast upon.

Then outspake the Count Arnaldos,
you shall hear his word anon:

‘In the name of God, good sailor,
teach me, teach me, this your song.’

Answered thus to him the sailor,
thus to him has answer done :

‘No man teach I what I’m singing,
save he sail with me along.’ 1

The charm of Count Arnaldos is produced entirely by a most
fortunate forgetfulness. The end of the story is a commonplace
matter; the sailor is a pirate and seizes the Count, but then recog-
nizes in him his long-lost lord, and all ends happily. The cut,
applied in the Cancionero de Amberes, has added just that salt of
lyrical emotion which is needed to make the perfect ballad; for
though we have defined ballads as narratives, yet such narratives
only reach exquisiteness when they have this lyrical touch. It is
m Mudarra’s Vengeance and Blancanina , in Count Olinos and The
Prisoner , and in each case it comes from the same refining process,
which has cleared away prosaic detail, and left us an emotion in
purest poignancy. The same touch is found in Rosa fresca and
Fonte fnda (u S , 116), which would doubtless have been at one
time banal histories of adultery. The latter must have existed early
in the fifteenth century, for its wording influenced Jaume de
Olesa’s copy of the Gentle Lady, which must be dated about 1421.
Fontefrida might be rendered thus:

Cooling fountain, cooling fountain,
cooling fountain, fount of love.

1 See Note F.

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 181

Thither little birds are winging,
all there consolation know,
save that drooping in her sorrow
broods the widowed turtle-dove.

Thither hasteneth a traitor,
comes the nightingale along:
and he fashioneth his discourse,
and of treason is each word:

4 Wert thou gracious to me, lady,

I would be thy servant now,’

4 Out away, thou art a traitor,
false, sly, tricky didst thou prove.

Never more I on green bough settle,
nor in meadow make my bower.

Turbid is the fountain’s water,
that, when found, did limpid show.

Never may I marry husband,
never children bear of love:
pleasure comes no more from children,
nor does consolation come.

Out away, thou sorry traitor,
false, sly, tricky dost thou prove,
never will I be thy lady,
never be thy married love.’

The full story, did we know it, might be that of the Lady of Weissen-
berg reversed. Nufto Veto (168) is another of the same kind. It uses
the motif of a false report of a husband’s death. Some of its words
have crept into the Tristan ballad, thus attesting its own early date.

The power of love is a motif of all balladry, and it is the principal
moral left by the romantic portion of the ‘romancero’:

that the faults that lovers make
worthily forgiveness gain.

It is the indulgent moral of the classical £ comedia’, which uses love
as the universal motive. The notion of the lover who alone pays
ransom is not known in Spain; but the triumph of love over death,
exemplified by the springing of trees and flowers as in the French
Thorn and Olive , is a feature of Tristan and Count Olinos . As for
the power of beauty, it finds expression in an admirable fragment,
St . Simon’s Hermitage (143)- A fuller version is current in Cata¬
lonia under the title of The Lady of Aragon . It was probably
borrowed by the Catalans, directly or indirectly, from the Greeks

182 ROMANCE BALLADS

of Corsica or Southern Italy, since it exactly corresponds to a part
of Bridesmaid into Bride (Politis 83). A ballad which might appear
to be written in praise of the power of love is Angela or the Dead
resurrected (x, p. 136). It is based, however, on a real event in the
sixteenth century. A girl, married against her will, died and was
buried. She revived when her true lover opened her grave, and a
lawsuit was begun before the tribunal at Valladolid to determine
whether death annulled the marriage.

To return to the more important narrative ballads, the Re¬
marriage prevented (135), also known as Count Sun , describes how
a slighted lady comes and claims her husband before he takes to
himself a second wife. In England this motif is used in Young
Beichan and Susie Bye . The Spanish version arises by amalgama¬
tion with an independent motif, The Signs of the Beloved, which
appears in reports of the husband’s death: true reports in the case
of Belle Doette, false in Unter der Linde. Another theme current
in England—and this time of definitely English origin in ballad
literature—is Ramon Berenguer and the German Empress (162),
which is the Spanish representative of Sir Aldingar. The Spanish
poem is late; it is later than the castilianization of Catalonia. Its
source is the chronicle of Desclot, where a decasyllabic heroic poem
on this subject has been reduced to prose, not greatly deviating
from the probable poetic text. Narrative verse is rare in Old
Catalan literature, so that this instance (like that of the probable
heroic poem on the conquest of Majorca) is of high historical
interest. The immediate source of the Catalan poet was probably
Toulouse, which city was a distributing-centre for this and other
legends of remoter origin. Our Child Waters and the Piedmontese
Ambrogio and Lietta (Nigra 35) are among the closest parallels to
the Spanish Dona Arbola (x, p. 93). Definitely Italian in origin is
the ballad of the poisoner Moriana (x, p. 98), which is an offshoot
of Donna Lombarda. Rico Franco (119) is a worthy representative of
the Dutch cycle of Hallewijn , since it is swift and concise; and with
it is grouped a rather similar tale of a murderer outwitted by a
woman s ruse, Marquillos (120). The Danish Elveskud , passing
through France and becoming Le roiRenaud , lost its elfin opening.
So it comes to Spain, where it is the ballad of Don Pedro and Dona
Alda(x,p. no).

These ballads of remoter origin bear marks of their passage
through France. Those of French origin are more numerous and

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 183

more developed. One with a pseudo-Carolingian setting is Count
Claros of Montalban (190—2). The fief of Montalban related Claros
to Reynald, and the affair is said to pass at Charlemagne s court,
with the usual mention of other peers; but it is In reality an anony¬
mous intrigue, or rather two. For the story is told in two forms.
In one Claros boasts that he has conquered the affections of the
princess, and he is apprehended and led off to be executed, but
escapes by her intervention; In the other, it Is the princess who is
arrested and kept in prison with water up to her waist until Claros
comes to set her free by his valour. The former version corresponds
to the fragmentary Count Velez and Florencios (ix, 189, 312, and
Primavera 138), in which other heroes boast of their triumphs
unduly, and to the Russian ballad of Nastasja Politovskaja
(Rybnikov 33). The Asturian Galanzuca (x, p. 42) is a somewhat
simplified version of the first adventure. In the French Romancer0
the two imprisonments for love are represented by King Loys
Daughter (imprisoned princess) and La Pernette (or La Belle se sief
an pied de la tour ); but in the former her escape Is due to her own
ruse of shamming dead, as in our Gay Goshawk , and in the latter
both lovers die. The first adventure of Count Claros thus does not
correspond to the songs now accepted as constituting the French
‘chanson populaire’, but to an older age: it corresponds to Aude-
froi le Batard’s Belle Ydoine , which itself was based, doubtless, on
a traditional story. The second is also different in detail from the
corresponding French poem, as well as being more amply narrative
in manner.

This tableau of high jinks In high society is found also in A Child
is born to the Princess and Galvan and the Princess (160,159). More
dourly licentious is Delgadina (x, p. 126), reproducing the Manekin
legend. The Princess and the King of France's Son (158) Is a tale of
intrigue or ravishment, similar to the Catalan Manner (Mila 199,
201, 207), though it is not stated in the Castilian form that the
ravisher is a sailor. It is as a sailor that the king’s son appears in
northern Italy (Nigra 44); but this description would not suit the
vaguely Carolingian setting of the Castilian version. Another
maritime ballad has not made a lodgement in Castilian, but has
been restricted to the coastal regions. It is the Portuguese Ship
Catherine (x, p. 258), with the name of a particular ship of the early
sixteenth century. In Catalonia it is called The Cabin Boy (Mila
215), and in France The Short Straw (Doncieux 17). Found also

^4 ROMANCE BALLADS

in Brittany and Scandinavia, its absence from English balladry is
notable, for we must wait for Thackeray’s Little Billee to adduce a
parallel.

Stories of vengeance in a vaguely Carolingian setting are The
Avenging Prince and The Palmer (150, 195). The former would be
banal enough but for the swiftness and excitement of the first lines,
which bring the action vividly before the auditors 5 eyes :

See him, see him, where he cometh,
cometh the avenging lord !

Riding with his stirrups shortened
on his war-horse swift and strong,
mantle twisted round his left arm,
all his ruddy colour gone,
in his right hand firmly grasping
a javelin both keen and long,
such that one might cut a ploughshare
with that javelin’s keen point;
seven times was it attempered
in a furious dragon’s blood,
seven times more was it sharpened,
so that it might cut the more:
twas in Franee the steel was hammered,
cut the shaft in Aragon:
still he whets it, as he hastened,
on the wings of his falcon.

The Warlike Maid or Don Martin of Aragon (x, p. 119) has parallels
m all the lands of Europe, and even in China (the legend of Mu-

D77 ™ S0 7 urce m S P ain must be eitb er a poem akin to the French
Bette Uaudine or the Italian Warlike Maid (Nigra 48).

In addition to these there are a few pieces for which a foreign
origin need not, so far as one knows, be conjectured. In Espinelo
(152) we have a conventional adventure story starting from the
superstition that of twins one must be the fruit of adultery:
Espineb is therefore set adrift by his mother, the Queen of France,
drifts to a Moorish land, is adopted by the sultan, and ends his life
as a great monarch. In Bovalias the Pagan (126) nothing occurs;
the sole point made by the ballad is that his tent was crowned by
a ruby which gave light to the camp. This belief in carbuncles or
rubies is found generally throughout Europe, and was used more
than once m French chansons de geste’ which were known to the
Spaniards. Peranzules, King Bikar, Sevilla, Alfonso Ramos are

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 185
titles of other similar pieces too brief to give away the secrets of

their origin.

My father came from Ronda,

my mother from Antequera ( 1 3 1 )

sings a Christian captive girl in a ballad which should, perhaps, be
included among those of the frontier; it is historical as to customs,
if not as to facts. On the other hand, Moraima (132), where a pretty
Mooress is tricked into admitting a Christian who pretends to be
her uncle, is mere fiction. Don Manuel de Leon (134), a ballad of
the sixteenth century, contains the legend of the glove thrown to
lions and recovered by a brave knight, together with the punish¬
ment of the flighty lady who exposed him to the risk, which is the
legend of Schiller’s Handschuh. Schiller’s source was French, but
the Spanish version appears to have been older, since already in
1557 Don Manuel de Leon w r as famous for his successes with the
ladies, in battle against the Moors and over lions. A group of
ballads in general imitation of the Gaiferos series is entitled Moriana
or Julianesa (121-5) and contains some striking phrases. The best
of them describes an unwearying pursuit, and some genius has cut
away from it all the entangling narrative:

For my ornaments are weapons,

I in battle take my ease,

stubborn rocks they are my mattress,

keeping vigil is my sleep;

dens of beasts surround me darkling,

roads untrod by human feet,

and the heavens, forever changing,

take delight in hurting me,

as I roam from hill to hilltop

and by beaches of the sea,

seeking, seeking, if safe fording

through my -wretchedness may be.

But for sake of you, dear lady,

all endure I cheerfully. (125)

Don Garcia (133), who defended Urena by dressing his dead in
armour and manning the walls, depends probably on the chap-
books of Ogier , where a similar stratagem was employed. The
Skull's Invitation (x, p. 209) is the germ of the Don Juan legend.

This is the picture of Castilian balladry in its hey-day, or when
but slightly impaired, but it represents only a small section of the

4615 B b

186 ROMANCE BALLADS

total wealth of Spanish ballads. Production has gone on unceasingly
since 1500. Its mass has been swelled, and its quality grievously
diminished, by print. Towards the middle of the sixteenth century
the first collection appeared, the anonymous Cancionero de Amberes.
It had an instant success, and several other collections of traditional
pieces were issued simultaneously. These collections gave the
standard for ballad-texts, and have exercised a decisive influence
in favour of curtailment. At the same time the printing-presses
poured out a flood of fly-leaves (pliegos sueltos) for the use of
street-singers. Collections were made also of particular cycles
such as the ballads of the Infantes de Lara and the Cid. Into them
were swept not merely traditional pieces, but also modern inven-
^ 0ns ri ’ for the mteres t aroused by the ‘romances’ tempted versifiers
like Timoneda and ‘the Caesarian knight’ (Mejia?) to chop up the
chronicles into verses. Such ballads are customarily called learned
(eruditos) by Spanish critics. They are not usually happy.

In this way a new wave of invention set in. What the ‘learned’
had begun was continued by the great poets (romances artisticos).
Lope de Vega’s love for the genre was so great that his plays are a
principal source for discovering lost versions or whole ballads.

a urally the texts are accommodated to his requirements as a play¬
wright; but so skilfully that it is scarcely possible to determine
what is new and what traditional. Like Burns, Scott, and Goethe,
°" e .° f th ® § reat P° ets who has completely absorbed the
til k“ qUe ‘ a P art from P ie ces in, or associated with, the

Ifeirf chliT’ Pe dC ^ C ° mp0Sed ballads on his
hterarv ? ? 7 ° f “ “I? 017 nature ‘ Jt was in this direction that
hteraqr fashion was tending. The sentimental Moor had been in¬
vented by Gines Perez de Hita, with some help from the Granadine
senes of frontier ballads, and all Spaniards who felt sentimental

Tk . obl :f dt vr in ^ verses (romances moris e c n o tai

,. blS 1S klnd °ihalladmwhichGongoraexcelled,thoughsome-
. eS - W1 ,. a ™ 0re Vlnle accent than is common in the genre. His
tfrri ° mn 15 a u SP l endid Piece ’ the balIad technique receiving

HerdeTtofrivl th* 1 ^ ° f G6ng0ra ’ s verses >~ned te

SomLe iJl S3m u qUallt f S 38 tbe ano ^°us ballads, and to
longed enrichment for the German poetry for which he

fulwSi biTi y ° Unger G6ngora ’ was not a success¬
ful writer of ballads. It is true that he made but one attempt, but

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 187

his plays show that he had lost sight of one essential element: the
narrative value of the 4 romance’. There was general decadence In
all the types of Spanish literature, and when the eighteenth century
opened, the ballads had been divorced from good taste. They had
become plebeian in subject and style (romances vulgares). Though
the one quite successful lyric of the age was an attempt to retell an
episode from the life of the Cid, the measure employed by Nicolas
Fernandez de Moratm was not the ballad measure. They came
back to favour with the Romantic movement of the early nine¬
teenth century, and were gathered into the vast Romancero of
Agustin Duran. In this form, however, there was little distinction
between good and bad, and the 'strengthless heads’ of late and
artificial ballads were found pell-mell among those which spoke
'winged words’. The Romantics, furthermore, were digesting
all their past national traditions simultaneously. Rivas’s Mow
Exposito (1834) owes more to dramas of the decadence than to the
fine cycle of the Infantes de Lara . His epos illustrates also the
Romantic preoccupation with 'local colour’. It is this concern
chiefly which prevents the 'leyendas’ of Rivas and Zorrilla from
being true ballads. They self-consciously attempt to produce an
atmosphere, which the veracious ballad-poets took for granted.

Both artistic and vulgar ballads continue to appear in our own
time. The lamented Garda Lorca aimed at bringing back to the
Spanish Parnassus not only their traditional style, but also their
music, restoring to poetry once again its old outer clothing of
melody. Other modems, such as Don Salvador de Madariaga,
have used the ballad metre for their own purposes. On the other
hand, the deaths of bullfighters, tragedies, and (in countries of
Spanish speech, when they exist) bandits, are deemed subjects
profitable to be sung round the streets of Madrid and other cities
by blind beggars, for whom a tribe of beggarly rhymesters works.
The civil war has already produced its Romancero, though of little
merit. Humble as most of the vulgar ballads are, there are some
not devoid of wit or vigour. The eighteenth-century Miller of
Arcos suggested to Pedro Antonio de Alarcon his Sombrero de tres
Ricos ('three-cornered hat’), which Russian dancers have carried to
most parts of Europe.

The old historical ballads treated of civil wars in the Guadal¬
quivir valley or frontier skirmishes on the Granadine border: they
were thus specifically Andalusian The epic themes belonged

188 ROMANCE BALLADS

originally to Old Castile, and it may have been Old Castile that
transformed them into ballads. The original focus of Spanish
balladry was thus Castile in the widest sense, the whole area em¬
braced between Burgos and Sevilla. It was from here that the
national themes broadened outward. Study of the melodies shows
a certain dichotomy between the more European style of the north
and the more Oriental luxuriance of the south; it shows also that
southern tunes have penetrated northward. As for the ballads
which are not indigenous, a geographical study 1 has shown how
they advanced in waves of versions and variants from the south
or south-east to the north-west, isolating older forms and confin¬
ing them to the remoter regions.

Hence the importance for Spanish ballad study of the versions
still preserved in the remoter regions: in Catalonia and Portugal
(which require a place apart), in the Asturian mountains, in Spanish
America, and among the exiled Balkan Jews. 2 The fire has burnt
out in the original hearth, but the sparks are tended wherever any
semblance of the old close communities still persists. The ballads
of the Asturias are exceptionally rich in authentic texts; texts which
may be used to emend the versions preserved for us by sixteenth-
century collectors. There has, it is true, been some loss in topics
and style. The lyrical element has increased, the narratives have
become predominantly novelesque. Historical ballads are perish¬
able, and if they survive it is for their entertainment value alone.
One frontier ballad is known in the Asturias, none in the Balkans,
but five among the highly conservative Jews of Morocco. The
murder of the Master of Santiago is still retold in the Asturias and
Morocco, but the others of the series have gone. On the other
hand, the deaths of Prince Afonso and the Duke of Gandfa are
still remembered. The ancient literary ballads derived from
Spanish or French sources have been severely thinned out by time,
but adventure stories are numerous, together with those of Biblical
and pious origin.

Amid all these losses, one is impressed by the faithfulness of
allad tradition. The Jews are exiles, with every reason to hate the

1 R.Men&idez Pidal, ‘Sobre geografia folkldrica’, Revista de Filologia Espa-
nOia y VH, 1920. . *

Pida1 ’ ^ 1 I°-? anCer ° J ' udl '°- es P aH ol’ and ‘Los romances en
’ 19275 J - de Cossio and T. Maza, Romancero

MaSd,f 91 i. ’ Santander > ‘ 933 - 4 ; R - Gil, Romancero judeo-espanol,

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 189

land which rejected them; they follow a different religion, and their
language has been corrupted by the intrusion of Balkan and
Plebrew words. They have a limited number of pieces for use on
ritual occasions, and their own ballads about Moses, Adam, Dinah,
Goliath, Tamar, Absalom, and Solomon; but there has been no
systematic hebraizing of their inheritance. The ballad persons go
to mass. The Cid is a hero, the Moor an enemy, even to people
who have been sheltered by Moors and rejected by Christians. So
it is among the Jews of Morocco that we encounter the end of the
ballad of Count Arnaldos , though unwelcome, together with a new
ballad of the frontier, Poriocarrero . From the Asturias we learn
how the ballad of Don Fadrique’s death originally opened. The
Balkan Jews tell us enough about Count Velez bragged to let us
know that it belongs to the same family as Marianson's Rings and
the story of Imogen and lachimo. Conservative as they are, these
regions also illustrate the traditional freedom of the minstrel. The
versions of old ballads rise to scores, variants of texts to hundreds,
and the total store of ballads—still mostly uncounted—mounts
to many thousands.

The ballads of Portugal 1 differ from those of the Spanish regions
only by the change of language. It is not so much a change as a
transformation. The two tongues are so closely similar that in most
cases a Portuguese word has the same prosodic value as a Castilian,
and the Castilian term need only be pronounced in a Portuguese
manner to become Portuguese. This was the first step in transla¬
tion : substitution. There remain some incompatible forms. Such
forms tend to survive in both Portuguese and Catalan ballads, at
least for a while; but they are naturally more infrequent in Portu¬
guese texts than in Catalan, which has a rhythm so different from
Castilian. Lisbon is the centre of a new scattering of ballads. Just
as those of Castile have spread to the circumference of the Spanish
sphere, so some of the richest gleanings in Portugal have been
made in the conservative region of Traz-os-Montes and the distant
Azores. The texts are consistently more modem than in Spanish,
and have suffered some new applications. These may prove mis¬
leading. Almeida Garrett, encountering a piece which began
I came by sea from Hamburg,

and describing a Christian captive in Moslem hands, supposed
1 T. Braga, Romanceiro geral portuguis , 3 vols., Lisbon, 1906-9.

I 9 0 ROMANCE BALLADS

that the theme referred to some raid by Sallee Rovers in the seven¬
teenth or eighteenth centuries; but, in fact, the words merely adapt
to modern uses

My father came from Ronda,
my mother from Antequera.

For various reasons an erroneous opinion is current, at least
among English scholars, that the Portuguese Romanceiro belongs
to the Franco-Italian-Catalan group of lyrical balladries, distinct
from the severely epical manner of Castile. It was natural that
Almeida Garrett, the Walter Scott of Peninsular balladry, should
be impressed by the novelty of the versions he collected, while yet
the ‘romancero* was insufficiently appreciated in Spain. Teofilo
Braga devoted his bizarre erudition to explaining the international
affiliations of his collection. He came to some queer conclusions,
such as the supposed existence of a Portuguese ‘Odysseian’ cycle
about a supposed ‘Nausicaa’; but his work had a sound core in the
use he made of Nigra’s Canti popolari piemontesi. Hence Braga’s
Romanceiro offers a serious study of the international adventure
ballads current in Portugal, and such ballads are of French origin.
Only perfunctory attention was given to this matter by Menendez
y Pelayo in his Tratado de los Romances Viejos. Spanish scholars
have been concerned with the historical and national matter of their
‘romancero’, which is still not entirely explained; the theory of
epic origin has turned their eyes away from poems which have
evidently no root in national epics. But this national and epical
matter is, as we have seen, the most perishable. It cannot easily be
exported to another country. It is the adventure ballads that sur¬
vive and travel, and these adventure ballads are, as we have seen,
borne on a stream of French influence. But they are borne into
Spain as much as into Portugal, and the Portuguese versions
are identical in form and content with, and later in detail than,
the Castilian. They are, therefore, still constituent parts of the
Castilian-Portuguese sub-group of Romanic ballads.

Penetration began in the sixteenth century. Political poetry
existed in Portugal from a much earlier date. In the War of Libera¬
tion X1384 ff.) we hear of short lyrical snatches in scorn of the
Castilians or in praise of the Holy Constable, but no ballads. They
are a popular and vulgar development of the courtly satires
formerly composed by such troubadours as Airas Nunes de

SPAIN, SPANISH JEWRY, PORTUGAL, IBERO-AMERICA 191

Santiago. The death of Xnes de Castro provided an admirable
ballad theme, but no ballads. Garcia de Resende worked up the
incident into the best poem published in his Cancioneiro (1516),
but in stanzaic form. It seems to have been left to Castilians, with
dim memories of the event and the persons, to reshape this episode
into the cycle of Dona Isabel de Liar. The Balkan Jews have a
ballad of Gian Lorenzo , dealing with the humiliation of Joao
Louren?© da Cunha in the fourteenth century; but their ballad is
in Castilian. The same observation holds for those on Prince
Afonso’s death in 1491.

It is possible to watch the introduction of the new form and the
acclimatization of the ‘romance’. Gil Vicente, writing between
1502 and 1536, a poet of exquisite sensibility, normally employs
the Castilian language for his ballads, though with occasional lapses
into Portuguese. The court at which he worked was undergoing
an intense process of castilianization. Camoes, active between
1546 and 1570, quotes ballads in Spanish, but writes none. The
first place where we encounter a number of Portuguese ballads is
in Jorge Ferreira de Vasconcellos’s Memorial of the Second Round
Table . Issued in 1567 but probably written about 1553, the book
is a neo-Arthurian romance, composed after the Spanish model of
Amadis de Gaula . At the date of its composition the principal
Spanish Cancioneros de Romances were enjoying a wide circulation.
From this date onwards the composition of ballads in Portuguese
becomes an established literary manner.

The gift of Portugal to the Peninsular ‘romancero’ was Gil
Vicente. 1 Whether he writes In Spanish or Portuguese, he writes
with a simple directness so undefiled that his ballads are as familiar
as those of oral tradition. A distinction cannot be made between
his style and the best traditional pieces, whether these be Castilian
‘romances’ or Portuguese ‘cossantes’; his use of material available
to hand pervades his work, so that it is hard to say when he is
remodelling traditional verse, when pouring out a new wealth of
music. He reproduced the May song in his Flerida and Duardos,
isolated from King Bucar, the Moorish lament for the loss of
Valencia, which alone gives that ballad its charm, composed an
exquisite ballad barcarole, and uttered ballad compliments in a
tone of popular rejoicing. His ballads are only a part of his lyrical
output, which is distinguished by the same exquisite sureness of

1 Lyrics of Gil Vicente, edited and translated by A. F. G. Bell, London, 1925.

i 9 2 romance ballads

touch. Lope de Vega could do the same when he was willing, but
he was less willing; the ostentation of the Philippine age creeps
into his 'romances’, which he cannot allow to speak for them¬
selves. But the style of Vicente, like Burns, is of the same cloth
as his sources, and goes to prove that the presence of a great name
does not prevent poems in the appropriate style from becoming
part of the traditional store.

4. Catalonia

The Catalan language is closely related to Proven 9 al by most of
its traits and actually extends into Roussillon, north of the modern
border; on the other side it merges into the Aragonese dialects of
Spanish by a process of gradual change. During the Middle Ages,
Barcelona served as the capital of a kingdom which pursued a
Mediterranean policy of its own, and so held somewhat aloof from
the development of Spanish unity under the aegis of Castile. The
fifteenth century witnessed a movement of peaceful penetration by
Castilian policies and culture, and at the close the kingdoms were
brought together by the marriage of their sovereigns. Since 1516
Catalonia, Valencia, and the Balearics have been integrated into
Spain, without losing the sense of regional differences.

Thanks to these and other considerations Catalan balladry 1
requires a double classification. In its oldest texts it is a portion
of the Franco-Provengal group; in its later forms it is Spanish.
Between the two phases lie those ballads in mixed Spanish and
Catalan which represent, at least in most cases, the penetration
into Catalonia of the Castilian ‘romancero’. Mila was disposed to
carry back the history of the genre as far as the fourteenth century.
There is no quite convincing evidence to support so early a date,
since the first sure fact is that Jaume de Olesa possessed a mixed
Catalan-Castilian version of The Gentle Lady about the year 1421. 2
The legendary date of the indigenous Comte Arnau is 1017; the
ballad itself was recovered in the nineteenth century by Piferrer;
but there is nothing to show when it arose, though on general
grounds it is impossible to believe it as old as its matter. Count

1 Lhe best collection is M. Mila y Fontanals, Romancerillo Catalan, 2nd ed.,
Barcelona, 1882, reprinted in his Obras Completas, vi, Barcelona, 1895. Mila’s
titles and explanations are in Spanish.

2 E. Levi, ‘El romance florentino de Jaume de Olesa’, Revista de Filologia
Espanola, xiv, 1927; W. C. Atkinson,- ‘The Chronology of Spanish Ballad
Origins’, Modem Language Review, xxxii, 1937.

CATALONIA *93

Arnold is, according to the ballad, an unforgivable sinner; in the
dead of night he appears to his wife, and though she defends her
children and her household, she cannot stop his grasping her and
haling her away to hell with his burning hand. The legend is
firmly fixed in the local traditions of Ripoll, and the count’s house
may be seen at Parnau on the road to Candevano. Apart from this
case the dates to be inferred from ballads making historical allu¬
sions are relatively modern. King Francois s Imprisonment (Mila
80) is the French ballad on this subject, and shows traces in the
language of its origin. The Capture of Nice (Mila 79) goes back to
the same period, but the bulk of the historical ballads recorded by
Mila lie between the separatist revolt of 1642 and the Carlist wars
of last century. A similar range of dates lies behind the ballads of
banditry, extending from the days of Serrallonga (£L 1632-7) to
the late eighteenth century. They are sometimes quite stirring
adventures—as in the Servant-girl (Mila 114), who detected and
captured disguised brigands in the inn—but there is a tendency to
utter platitudes about the importance of a godly upbringing as a
way of ultimately avoiding the gibbet.

We have to assign to Catalan ballads a history covering three
epochs: from the early fifteenth century to the mid-sixteenth those
of French cut prevail; from 1550, when the Castilian printed
collections were in every man’s hand, there was a period of intense
castilianization; then come the vulgar and plebeian ballads of
the eighteenth and nineteenth centuries, not to be distinguished
from those of the rest of Spain. Correspondences between Mila
and Doncieux and Nigra are to be associated with the ballads
of the first period, especially when supported by formal criteria.
There are poems in metres based on the French decasyllable and
alexandrine, and they obey the ruling that cadences of hemistichs
must be different. In these pieces French words are not infrequent:
words like £ xivall’, £ arjant’, Toy 5 , £ aymar 5 for £ chevaF, £ argent 5 ,
£ bois’, £ aimer 5 . There are poems in the assonance e, corresponding
to the French infinitive er, and the incorrections in usage show that
these are foreign. One notes, too, that the word proper to these
poems is £ canso 5 (chanson), whereas £ romanso 5 means properly a
ballad printed on a flying sheet. (Compare the Appallachian distinc¬
tion between love song’ and 'ballet’). This is an acknowledgement
of the way in which the Castilian 'romances’ penetrated Catalonia.
The method of singing, according to Mila, transforms these poems

i 94 ROMANCE BALLADS ,

into couplets, since the first two are sung, then the second with the
third, the third with the fourth, and so on. In Castilian pieces, or
original work formed on the Castilian model, the rule as to cadences
is ignored, and the verse is octosyllabic. There are, however, many
poems which chance to have a masculine assonance while the other
octosyllable is feminine. They include some of the most interesting
international pieces to be found in the Primaveray Flor de Romances
as well as in Catalonia, and their exact scansion has been a moot
point since the days of Wolf and Hofmann. The modern method
of making unequal lines equal in singing is to lengthen the accented
syllable over two notes (as ‘Gibraltaar); the older method, attested
in manuscripts of the Castilian epics, was to add a ‘paragogic e*
(as ‘Gibraltare’), and this would serve for the ballads in the
Primavera. On consulting the music, however, we find that some
of them were intended to be sung as syllabically unequal; that is
to say in a manner foreign to Castile. 1 This circumstance, which
should be further investigated, stresses the role of Catalonia in the
transmission of foreign material to Castile, not merely from
France, but also directly from Italy, and perhaps from yet farther
afield.

1 Pelay Briz, Cansons de la Terra, Barcelona, 1866, i, p. 105, shows this
difference of cadence in the music of The Students of Toulouse, which is certainly
French:

1 flat 34 U8: aa jd.xdejf {.e)d. ^jd.xdejc.v See.

Salinas (1577) has a very simple tune for the Castilian Rosafresca with the same
peculiarity:

68 U8: eejd.ef.eld.ee eejd.ef.ejd.x

From present-day tradition in Andalusia, E. Tomer collected a setting of the
same kind for The Gentle Lady {Folklore y Costumbres, Barcelona, 1931, ii,
p. 29):

34 U8: e/aag/a.g/ag a/bdb/a.g/(fgf)e./er &c.

These lines are made equal not by the use of a paragogic e, but by the musical
rest.

NORDIC BALLADS
i. General Considerations

T HE ballads of the north and centre of Europe belong to one
vast store of methods and themes, and in their oldest forms
they have the same background of ancient Germanic heroic poems
and German-Latin political verse. The composers of these verses
were often Saxons. Saxon court poets are said by Saxo Gram¬
maticus to have been active in the Danish civil wars of the twelfth
century, where they made verses about the battle of Haraldssted
(1131) and the flight of King Svend (1157)- They sang, no doubt,
in Low German. It is possible to carry back the dates of historical
Danish ‘viser’ to the year 1208 with certainty (. Battle of Lena), and
to the middle of the twelfth century with some hesitation (the death
of Erik Emun, sung in one ballad, occurred in 1137). It is clear,
therefore, that at the moment when the new ballad style developed
in Denmark, Saxon singers were giving performances of a similar
nature under the patronage of such kings as Knut Lavard; but it
is not possible to say how the foreigners affected the viser , or even
to assign to them an assured priority. In Saxon England there was
a ferment of creation, of which the ballad of Sir Aldingar is one
fruit. We are told that

Merie sungen the muneches binnen Ely
tha Cnut ching reu ther by.

‘Roweth cnites noer the land
and here we thes muneches saeng’.

The form is almost the form of a ballad, yet it is not a ballad. We
are on surer grounds when we learn from William of Malmesbury
that a song corresponding to Sir Aidingar w r as sung at the cross¬
roads of England in the middle years of the twelfth century. As for
Germany, Flemish and Low German political pieces seem to have
priority over those of Switzerland and Austria by something like a
quarter of a century. They are not nearly as early as might have
been expected. The unmistakable ballad note is not heard before the
beginning of the fourteenth century, when Lippold defended his
castle of Homboken against the Bishop of Hildesheim and Duke
Otto of Brunswick.

NORDIC BALLADS

196

In 1311

From Spiegelberg fast riding came
Lippold, a mighty warrior’s name;
his sword, two ells and one half long,
was sharp, his harness bright and strong.

His helm weighed seven and half a pound,
with pearls bespattered and with gold,
his golden-gleaming shield was round,
he bestrode his horse in hosen bold.

He came to Brunswick, to the town,
at the Golden Lion gat him down;
he met his company, his own,
to every man was full well known.

‘From Spiegelberg have I come here,

I bring you news of goodly cheer!

On us new-fangled fines they lay!’

They all thought good what he did say!

(Liliencron 6.)

The common centre of these Nordic ballads, if there were one
centre, would thus seem to have been the northern parts of
Germany, and the first area of diffusion to have been the rim of
land round the North Sea from Jutland to East Anglia. To the
above indications we may add that the Low German Dietrichs Saga ,
now extant in an Old Norse translation of the early thirteenth
century, is of peculiar importance for the transmission of epic
material both to Scandinavia and to Germany. The similarity of
style between Danish and German ballads is such as to suggest a
common fount and origin, and there was undoubtedly at a later
date a considerable influx of German ballads (such as Tannhduser,
The Count of Rome, and The Castle in Austria) into Denmark and
Sweden. What the Saxon poet sang to Knut Lavard in 1131 was
‘Grimhild’s notorious treachery to her brothers’, that is, the matter
of the Nibelung epos. The same matter is repeated in the Danish
ballad of Grimhild’s Revenge, but as a much later borrowing from
Germany.

On the other hand, there are considerable difficulties in the way
of tracing the whole Nordic ballad effort to one original centre of
inspiration. The chief of these is the lateness of the indubitably
German ballads. The style declares itself in Denmark fully a cen¬
tury before it is certainly to be found in Germany. In the latter,

GENERAL CONSIDERATIONS J 97

such a poem as the Bohmenschlacht (1298), though a political poem
addressed to the widest possible audience, is not suited to tra 1-
tional oral transmission and is not composed m any ot the ballaa
metres In England, evidence for the existence of known ballads
in their present state is withheld until the fourteenth century, apart
from the Judas manuscript of the thirteenth. The great bulk o
soncrs held in common with Norway and Denmark is Scottish, and
can°be accounted for as a flow of ballads from Scandinavia to the
ports of Aberdeenshire and Fife. When we compare the standard
collections of Child, Erk and Bohme, and Grundtyig, the priority
in time and merit of Danish balladry is indubitable, and it would
be rash to set up against this weight of testimony the inferences
drawn from debatable texts. Confining ourselves to the evidence,
we find two great groups of Nordic ballads: the older Ang o-
Scandinavian group which is best represented in Denmark, and
the somewhat younger German corpus. .

Some scholars, while admitting the relative modernity ot the
extant ballads, have sought to carry back their history to a much
earlier date by means of a genealogy running up to the primitive
epics and the political songs of the ninth and tenth centuries^
There is the old Hildebrandslied of the ninth century, and the a.
Hildebrand of much later date. They differ in one important point
namely that the more modern poem has a happy en mg. C
happy ending occurs in the chapter of the thirteenth-century
Thidrekssaga which tells the same story. Two explanations are
clearly possible: either the old Hildebrandslied continued to be
sung and to undergo modifications until it became the modern
ballad, the Thidrekssaga witnessing one of the intermediate forms
or the Thidrekssaga (in its lost Low German form) took up and
modified the old poem, possibly from a manuscript, and was itself
in turn the source of the younger Hildebrand. To the latterexplana¬
tion I incline because of the undoubted influence of this saga on
all the Danish ballads concerning Theodonc; the saga, m fact, is
a literary source of the ballads and interrupts the epic tradi om
Among the older political poems, the piece most relevant tobala
origins is the anonymous Ludwigslied. It was written by a Frankish

author to celebrate Ludwig Ill’s victory at Saucourt in^and
before the king’s death in 882. The metre is that used fo^ous
themes by Otfrid, and the style is defu ait f \ d f n “ l ; JJ*
wishes to draw a moral, that sins lead to defeat and repentance

i9 8 NORDIC BALLADS

victory. In the excitement of combat, however, his tone becomes

epical in its directness and vigour:

Song was sung, battle begun,

blood shone on cheeks, spieled there the Franks.

The same event gave rise also to the French epic of Gormond and
Isembart. So far as the Ludwigslied is concerned, it is not a ballad
since its transmission is not through tradition nor is its style ballad¬
like; its affinities with secular poetry are rather with epics than
ballads. Still less ballad-like are the Modus Ottinc and Be Henrico
and other poems of the same sort. In Denmark the Bjarkamdl and
Starkadsmdl , in England the Battle of Maldon and (to a less extent)
Brunanburh, are epical compositions of late date and style which
show how far the epics had evolved immediately before the rise
of the new ballad manner.

A metrical argument for connecting the Nordic ballads with the
preceding epics rests, I believe, on a misunderstanding. Through¬
out all this area there is no way of applying syllable-count to ballad
verse. It is measured by four accents, two on each side of a pause,
which may be the end of a line or of a hemistich. Thus it corre¬
sponds to the four-accent principle of heroic Germanic verse, and
differs from the syllabic manner of Romance balladry. Lines are
bound together by rhyme, or rather assonance; but whereas in
Romance territory the assonances are adequate, in Nordic balladry
they are fleeting and elusive. The most distant similarities suffice.
Sometimes there is no correspondence of sound at all. This
elusiveness is so omnipresent that we must suppose it a source of
pleasure to the hearers. Listeners did not want to hear the clang of
identical sounds at regular intervals; they preferred echoes. They
preferred to echo not only the ends of words but also the begin¬
nings. Nordic balladry is full of alliteration, though this does
not occur according to the fixed rules of the older epic tech¬
nique. While all this is true, however, it does not amount to
proof that the ballad and epic metres are essentially one. It is
merely evidence that when the new rhymed or assonating and
measured verses came from France, the shift to a new technique
was made in England, Denmark, and Germany with caution.
Poets were slow to surrender the charm of alliteration, though
ready to drop its fetters, and they preferred approximate to rigid
equivalence in their lines. The lines have, none the less, a new

GENERAL CONSIDERATIONS 199

sort of regularity and a new principle of cohesion, which is
assonance.

The two great Nordic groups—the Anglo-Scandinavian and the
German—are fundamentally lyrical. It is to the lyric that they owe
their stanzaic form, which marks them off sharply from the con¬
tinuous narrative verse of Spain, Greece, Serbia, Rumania, or
Russia. It does not mark them off from Franco-Italian ballads
which, as we have seen, also rise out of the lyric. There is consider¬
able common ground between the ballads and folk-songs of France
and those of Germany, as well as a direct influence of troubadour
themes and style in such poems as Tannhciuser, which is formally
an ‘aubadek The specifically English parts of Child’s collection
are also heavily in the debt of France. The lyric, then, has given
the stanzaic form of Nordic balladry, but, especially in older
medieval instances, the stanzas are used for objective narrative.
This is especially the case with Danish ballads; in Germany, on
the other hand, lyrical treatment becomes more and more promi¬
nent as the centuries pass, and the present-day soldiers’ ballads
are almost entirely subjective. A formal distinction between the
German and Anglo-Scandinavian ballads is that the former were
composed to be sung or recited, probably by one person entertain¬
ing a group, but the latter were designed to be danced while a
precentor chanted the narrative verses. A refrain or double refrain
characterizes Amglo-Scandinavian ballads, and is the part sung by
the dancers as they pause between two movements. So character¬
istic is this in Scandinavia that absence of refrain in a Danish ballad
is good ground for suspecting German provenience. Our own
practice fluctuates. Specifically Scottish ballads have often Scandi¬
navian affinities and are adorned with a refrain; though it is not
certain that they were always danced. In England a recitative style
seems to have been customary.

All these northern balladries have for common background the
Germanic epos and the older achievements of French literature,
both epical and lyrical. The relationship between epos and ballad
has been discussed in an earlier chapter, but it has its place here
also, and demands attention even at the risk of some repetition. We
know from Jordanes, writing early in the sixth century, that c songs
of an almost historical sort* were current among the Goths. He has
indicated some of the heroes and subjects. There were songs about
Eterpamara, Hanale, Fridigern, and Yidigoia, of whom the last-

200 NORDIC BALLADS

named is the Vidga of the Thidrekssaga , and the Vidrik Verlandsson
of Danish ballads. He knew an important poem about Filimer and
the wanderings of the Goths, and what he tells us concerning
Ermanaric is coloured with poetry. Similarly, Paulus Diaconus
reproduces poetical legends of the Lombards, and Saxo Gram¬
maticus those of the Danes. The Italian ballad of Donna Lombarda ,
with the whole dependent cycle of ballads of poisoning, derives
probably, directly or indirectly, from Paulus Diaconus, who was
a favourite historian of the middle ages. In all this activity
curiously little corresponds to the Germans of Germany proper.
There is ambiguity in the use of the adjective ‘deutsch’ by German
critics, since it has the two senses of what is nationally, and what
is racially, German. These ancient epics were the property of all
the Northern peoples, and particularly of those who were most
violently agitated by the great migrations. The more important of
these peoples penetrated into Roman territory and became latin¬
ized. Those on the northern fringe—the Saxons, Danes, and
Norsemen—remained Germanic in language and part of their cul¬
ture, and it is among them that we find the most authentic sur¬
vivals of the old poetry.

What form may have clothed Jordanes’ songs we cannot know.
The oldest, and only complete, ancient Germanic epic is the Saxon
Beowulf which, in its present form, bears marks of composition in
the seventh century. 1 The Beowulf preserves authentic traditions of
the fifth and sixth centuries. What is peculiar to the seventh-
century author is the tinge of Christianity and the elegiac mood; in
other respects, and particularly in style and form, he may have re¬
mained true to the epic manner of his predecessors. It has become
customary to accuse him of longueurs, to contrast his style with
the more rapid manner of the Finnsburg and Waldere fragments.
But the author of Beowulf is not always dilatory; in passages that
describe action he can be as rapid as any. The fragments called
into comparison are, as it happens, descriptive of episodes of
intense activity; they are not necessarily conclusive as to how the
same authors would treat scenes of high revelry or incidental
episodes. What we learn in the Beowulf concerning the complete
plot of the Finnsburg poem suggests that this may have had a con¬
siderable length—in the neighbourhood of 2,000 lines more or less.
There was evidently a great wealth of episodes. Similarly the
1 R. Gixvan, Beozvulf and the Seventh Century, London, 1935.

GENERAL CONSIDERATIONS 201

Latin Waltharius , of the tenth century, is a poem of some length.
The style is delayed at places by the use of Latin cliches, but in
other places it is sober and bare where one would expect Germanic
cliches. The Anglo-Saxon evidence, therefore, tends to indicate
that the 'songs of an almost historical sort’ were not ballads but
short epics, suited to recitation in the long evenings of winter in the
great banqueting halls. So Beowulf himself is represented as
entertaining the court of Hrothgar in Heorot.

Of these pieces only the poems of Walter have given rise to
ballads, thanks to their coalescing with various old French novelistic
poems. The German Hildebrandslied is a younger poem, dating
from the ninth century and now extant in a mixed dialect.
Hildebrand is a hero of only one achievement; he fights with, and
conquers, his son. That is a very widespread motif, one which only
requires to be capped by names. There is little to compel us to
believe that the song of Hildebrand preserved any authentic tradi¬
tion of the Goths of Verona, or that it was other than a composition
by some ninth-century German poet. The episode was in¬
corporated in the Low German Dietrichs Saga, of the late twelfth
century, where it has the optimistic ending later to characterize
the German ballad. While the evidence is not conclusive, it is at
least arguable that there is no direct contact between the epic frag¬
ment and the ballad in this case, but that the episode came to the
ballad-poet through the mediation of the saga. This saga was trans¬
lated in the early thirteenth century into Old Norse, and in that
form is the undoubted source of the Danish ballads of Diderik and
Vidrik Verlandsson, some of the most stirring in the collection.
Rather than an authentic Germanic tradition, the legend of
Dietrich von Bern is an encyclopedia of Germanic adventure,
assembled in the late twelfth century, not without assistance from
the Carolingian and Arthurian models.

In the Beowulf we find a sketch of the plot of the Nibelung story
before the invention of Siegfried-Sigurd. Sigurd is the hero of a
number of lays in the older Edda which are either episodic or
summary. They resemble ballads in these two respects, and in
their anonymity, but they differ in their aristocratic appeal. They
are the property of trained reciters who will respect their texts, not
of amorphous tradition. We have, in these lays, specimens of a last
epoch of northern alliterative verse, approximating to the ballad
style which was to rise in the twelfth century in Denmark, but later

4615 d d

203 NORDIC BALLADS

in the neighbouring countries. Appeal has been made to the Eddie
lays to determine the form of the primitive Germanic epos. They
are brief, nervous, and highly dramatic. But these qualities may
well be secondary. The Eddie poems are not self-explanatory; on
the contrary the prose introductions are designed to set each of
them in their context in some larger traditional narrative. 1 wo
traditions, indeed, are mentioned, the one as current in the north,
the other as from Germany. What the poets have done is to pick
out salient episodes of the pre-existing tradition, and to realize to
the full their dramatic possibilities; they have also, in the Atlantal
and Atlakvida , offered summaries of the intensely dramatic con¬
clusion of the poem. These Eddie lays, alone 01 in combination
with sagas ( Thidrekssaga, Volsungasaga ), have given rise to Danish
and Faeroese ballads of the Nibelung cycles. Similarly, the Eddie
Svipdagstndl is the source of the Danish Ungen Svejdal, the poetical
fragments of the Hervararsaga are the source of the ballad Alf af
Odderskser, and there is a resemblance between the second Helgi
lay and the tragic poem of Ribold and Guldborg {The Douglas
Tragedy). One Eddie poem of a religious cast has had ballad
consequences. This is the Thrytnskvida. The Danish Tord af
Havsgaard is so faithful a transposition of this lay into the ballad
style, that the transposer must have had the very words of the
Edda ringing in his ears. The older Edda is, therefore, a mediator
between the primitive epics and the medieval ballads.

Other mediators were the refashioned and amplified German
epics of the twelfth century. Their authors have profited by the
technical advances of the French jongleurs. They show a new
amplitude, a new skill in weaving episodes, a new eloquence, and
a new decorative art. The Nibelungenlied and Kudrun are many
times longer than their originals, thanks to the accretion of new
material and to the more developed art of story-telling. The former
affects the Faeroese ballads of this cycle, modifying the details
offered by the Edda; it is also the direct source of the important
Danish ballad of Grimhild's Vengeance, which is in the debt of the
Thidrekssaga as well. As for Kudrun, its essential episode is the
recognition of a lost sister by a brother while she was washing
clothes beside the sea. Variously misrepresented, this episode
figures in the German ballads of Sildeli and others, and in the
Spanish Don Bueso. The epic itself is the most likely source,
but the evidence is not close enough to exclude the possibility

GENERAL CONSIDERATIONS 203

that the ballads may rely on the sources of the epic. Other
reconditioned German epics have had ballad consequences in
Germany: the Hugdietrich and Wolfdietrich. Biterolf and others.
The influence of Ortnit and Hugdietrich , and possibly of the
Thidrekssaga , has extended also into Russia.

In the Scandinavian North also, ballads have drawn heroic
material from literary works in a relatively recent manner. These
are the sagas. The Hervararsaga has already been mentioned.
Ragnar Lodbrokssaga, though late and somewhat puffy, was a
favourite with ballad-makers in Denmark and Norway. Norwegian
balladry shows a considerable number of pieces taken from the
less-known sagas, and in the Faeroe Islands this method of aug¬
menting the people’s repertoire was well established. One such,
Qrmurin langi, was rough-hewed from St. Olafs Saga last century,
though the style is wholly that of old traditional ballads.

Summing up, then, we notice that the primitive Germanic epics
form the background of both German and Scandinavian balladry,
but they are not immediate sources. The immediate sources are
in Scandinavia the older Edda and the heroic sagas, while more
recent sagas affect the ballads of Norway and the Faeroes. In
Germany the immediate sources are the lost Dietrich compilation
(probably for both the Hildebrand and the freely composed
Ermanarids Death) and the reconditioned epics of the twelfth and
thirteenth centuries. The Danish ballad of Hagbard and Signe is
not easy to place. There does not appear to be any mediating
saga. Saxo Grammaticus gives a number of Latin renderings of
a Danish poem which was more circumstantial than the extant
ballad.

English balladry is quite without these epic contacts. From the
tenth century England was assimilating the Romance culture of
France, and her epic traditions were fading from memory. In the
eleventh she was overrun by the romanized Normans. Her writers
contributed to French literature, and during the whole Middle
Ages much French literature was composed by subjects of the
English Crown. It is therefore not surprising to note that, in place
of Germanic epical traditions, Child’s collection shows, in the
specifically English pieces, a double dose of ballads dependent on
French romance. The Arthurian motif is prominent: The Boy and
the Mantle , King Arthur and King Cornwall , The Marriage of Sir
Gawain, King Henry. Then we have Hind Horn , King John and the

204 NORDIC BALLADS

Bishop , King Estmere , Sir Cawline , Lady Diamond, Blancheflour
and Jellyflorice. The first is related to the German Moringer cycle;
among the others there are some which may be romances invented
by the ballad poet. The style of all these ballads is only semi-
popular. The poet is consciously displaying his powers of inven¬
tion and narration, and the whole group is scarcely to be distin¬
guished, save on the ground of length, from the Tail-rhyme’
romances, like Athelston and Chaucer’s burlesque Sir Thopas,
which formed the staple of public entertainment in England. These
romances are much less prevalent in Scottish balladry, which is
distinguished for its close association with Denmark and Norway.

The influence of French romance was exerted strongly on Flan¬
ders and the Netherlands in the German ballad area, and from
thence spread to the rest of the territory. There is the usual Flos
and Blancflos. The troubadour biography of Guilhem de Cabe-
stanh, which occurs also in the French twelfth-century romance of
the Castellan de Coney, gives material for the Flemish and Low
German ballad of Brennenberg. The Tristan legend is reflected in
the German Liebesiod. One notices also, especially in Flanders, the
influence of the later French popular songs. There are many
which reproduce the cynical French view of marriage, describing
amorous intrigues in which the offender is held up for admiration,
or in which he is a monk or other religious person. The motif of
the vivandiere, always popular in France, is also encountered in
the Netherlands. It is a military motif. Among what may be called
songs of the professions—that is, songs in which the composer is
stated to be of some particular profession or status—Germany has
greatly developed those of the soldiery. First the songs of reiters
and landsknechts in the sixteenth century, then the songs of the
religious wars, and later the conscript lyrics of modern times.

The influence of literary conventions fixed by the troubadours
and trouveres is also noticeable, particularly in the ballad use of
the ‘aubade’. Both Tannhduser and The noble Moringer have the
opening proper to an ‘aubade’, and that is true also of the ballad
refashioning of Pyramus and Thisbe {Abendgang). This is the less
surprising, since it is clear that both Minnesang and Meistersang
have affected the composition of love ballads. Tannhauser and
Heinrich von der Morung were poets about whom ballads were
composed, and Duke Ernest, The Knight from Steiermark, The
Count of Rome, &c., are c master-songs’ turned ballads. The literary

GENERAL CONSIDERATIONS 205

influence of High Germany on the whole ballad production of the
nation is as marked as the specifically national ballads of the Swiss,
Scandinavia, lying at a farther remove from the focus of western
civilization, was less strongly and directly affected by French in¬
fluences, though we must not rule out the possibility of immediate
contact by sea. The sea route doubtless brought to Denmark the
matter of The Short Straw (La Courte Faille ), represented in Eng¬
land only by Thackeray’s Little Billee , but known In France,
Catalonia, Spain, and Portugal. More often, however, the French
material w r as conveyed through Germany or England. Through
England came the Carolingian compendium, the Karlamagnmsaga ,
and the story of Tristan. There is a Norse Roland og Magnus
Kongin and a fragment on Roncesvalles, and six Faeroese ballads
in Hammershaimb’s collection systematically excavated from the
prose compilation. Ballads on the death of Tristan are known in
Iceland and the Faeroes. More important, however, was the know¬
ledge of the Enjances Ogier de Dinamarche which seems to have
reached Denmark by way of Low Germany. This gave rise to a
most popular ballad, Olger Danske og Burmand , which reproduced
the material in free fashion. Olger was felt to be a compatriot;
Burmand a foreigner, and probably a German. That the ballad
signified something important was made clear by the fact that its
refrain was cut into the wood of a Swedish church in the fifteenth
century. When Olger’s enemy was stated to be Diderik af Bern, in
another ballad freely modelled on the style of the Thidrekssaga, the
superiority of the Danish champion to the representative of Ger¬
man might was made patent. A primitive epic had sung of the
defence of that border by Offa; in the fourteenth century, the
historical ballad of Niels Ebbeson is to the same effect. But some¬
how, the legendary feat of Olger Danske was felt to be more sym¬
bolic, and this was the ballad chanted by the Danes who manned
the Danneverk in 1864.

With this knowledge of the common background of Northern
balladry and its different perspectives, we may go on to state what
are the relations between the various subdivisions. The unity of
the Scandinavian group is at once apparent, as also the priority
of Denmark in general. The historical series Is particularly well
developed in that country, and Sweden shares in some of these
ballads. Denmark and Sweden thus form a particularly close
group. Norwegian ballads stand farther off; they contain many

206 NORDIC BALLADS

independent reminiscences of the sagas. There are also Noise
ballads like the famous Axel and Valborg which have extended over
the whole Scandinavian area, including Denmark. In Iceland the
‘viser 5 are a relatively late importation, which has not proved
entirely congenial. The stories arrive in evolved forms from Den¬
mark or Norway. They have then to withstand the competition of
the local Timur 5 , excavated from written sources like some ballads,
with their elaborate alliteration and kennings. This sharpness of
wit in the Timur’ was truly congenial to the Icelanders, so that
the naive c viser 5 remained few and insignificant. The Faeroese
ballads seem to have arisen by the example of Iceland, but they
firmly established themselves in popular favour. One unfailing
source of supply has been Icelandic sagas, the prose of which has
been cut by Faeroese improvisers into verse length. They drew
also on Denmark directly, as may be seen in such bilingual pieces
as The Nix's Ballad (Hammershaimb, Anthologi , 3). Scandina¬
vian settlers in America have composed new ballads in the old
manner.

With the ballads of Scandinavia go our own ballads. Our ballads
are divided between England and Scotland, but it would not be
practicable to form two collections. They are, when compared to
those of Denmark, late and deficient in some categories, such as
the epic. The ballads of Scotland are the most closely associated
with Scandinavia, those on general themes are normally Norse or
Danish ballads acclimatized in our island. They are marked by the
use of refrain, though one cannot be certain they were danced. In
England, on the contrary, there is a strong French influence,
associated with a more definitely narrative technique. Yet Sir
Aldingar is English, but is one of our most striking ties with Den¬
mark; and the ballad cycle of Robin Hood is exactly paralleled by
that of Marsk Stig. The old stores of English and Scottish balladry
have travelled with their tunes to the United States, where they are
collected by modem investigators among the mountaineers of
Virginia and the Carolinas. These settlers have created new ballads
in the old style, which is represented also by such offshoots as the
ballads of cowboys and negroes.

German ballads form a separate group. They are stanzaic in
form, but seem always to have been designed for recitation, and
refrains are exceptional. There are numerous instances of borrow¬
ing and lending, but there is no such identity of details as exists in

GENERAL CONSIDERATIONS 207

the different forms of a Scandinavian ballad or even .between
Scandinavia and Scotland. The influence of France is differently
exerted, and more directly; the epical ballads depend on re¬
fashioned German epics, not on sagas or the Edda. The ballads
of Low Germany are older and more international; from High
Germany a powerful literary influence was exerted by the Minne¬
singers and Mastersingers, and Switzerland is the home of patriotic
military narratives. The use of narrative is more marked in the
Middle Ages; later ballads become more lyrical, and personal or
domestic. "These later characteristics mark the ballads of countries
dependent on the German creative urge, which was exerted in
three different directions towards the east. In a south-easterly
direction the German ballad type of construction and theme pene¬
trated Czechoslovakia and Hungary about the fifteenth century;
beyond Hungary the Rumanians learned to assonate in couplets,
though they owe inspiration chiefly to their Balkan neighbours. In
an easterly direction, the ballads of the Lusatian Mends are Gei-
man in form, partly in music, and (in all International ballads) in
content and exposition. Rhyming couplets and quatrains are found
in Polish Galicia, in the narrative verse of the Ruthenians. Their
kinsmen of the Ukraine adopted this same style in place of the more
irregular verse of the ‘byliny* and c dumi. It was a substitution of
the western assonance and themes for the indigenous tradition, and
a break with the old poetry of Kiev, which has since survived only
outside the Ukraine. To the north-east the German influence Is
notable in the almost purely lyrical verse of Poland proper. It
extends into Lithuania and Latvia, where themes are lyrical and
personal or domestic; where assonance is abundant, but too spora¬
dic to be a principle of versification, and where the construction is
(thanks to parallelism) definitely stanzaic. In Latvia these stanzas
are often reduced to one, so that there is no repetition to show that
the construction is stanzaic; the length of the line is normally of
eight syllables. In Finland and Esthonia, popular traditional poetry
employs eight-syllable unrhymed lines upon themes generally
domestic or personal, with casual assonance and abundant allitera¬
tion. Parallelism is used to form groups of lines of approximately
equal numbers, though there are no stanzas. The singers are now
peasants and fishers, as in Lithuania, but there is abundant evi¬
dence that their art came at some earlier date from the coastal
towns, where German influence was strong. In Finland these

2oS NORDIC BALLADS

towns have been occupied for centuries by people of Swedish
descent. There is a younger sort of ballad poetry current m that
country, using typical Scandinavian verse-forms. _ Even in the
older balladry the word ‘runo’ recalls the Scandinavian runes, and
the principle of alliteration is probably Scandinavian. Other words
used for Finnish poetry are also of foreign origin such as vnsi
which is the German ‘Weise’ or Swedish ‘visa’, and virsi which is
the Latin ‘versus’; the native word is ‘laulu’.

In this chapter, therefore, I shall attempt to describe the ballads
of the Northern nations in the order of their age and relationships.
The oldest extant are the Danish, with the close y associated
balladries of the Scandinavian lands. More remotely connected
with these come next the English and Scottish ballads, with their
important extension in the United States. Thirdly, t^ eie & re t e
ballads of Germany, and then those of the lands inspired by her:
Czechoslovakia and Hungary, Lusatia and Poland with the Ukraine,
and the Baltic lands in their two linguistic groups.

2. Scandinavia

Danish ballads are the most copious and authentic examples of
a class which extends over the whole Scandinavian North, and
which is fully represented in the balladry of Scotland.^ The word
most commonly used to describe these ballads is Vise , though it
is not in any special sense appropriate. In Germany the word
occurs chiefly when one thinks of the tunes (Weise). In Norway
and Iceland the preference is for equivalents of the Danish ‘kvaede
‘lay, chant’, and the refrain of the ‘vise’ in Denmark is called the
‘omkvsed’. These terms, therefore, do not signify more than
chanting, or a manner of so doing, while the refrain is a round-
chant. Nearly all ‘viser’ are adorned with refrains, which are but
loosely attached to the subject-matter of the ballad. They may
occur at the end of the stanza, whether distich or quatrain, or they
may be double and inserted between lines of recitative. It is sel¬
dom that these refrains amount to more than lyrical ejaculations,
though in the late balladry of the Faeroes they are occasionally
extended to several lines. Another device frequently employed by
the ‘viser’ is the repetition of the narrative by the chorus of dancers.
The precentor sings a distich. The chorus repeats the last word or
two of the first line and all the second, and then the precentor sings
a second distich. A description of the performance has been given

SCANDINAVIA 209

in the appropriate chapter of the first part of this book. Some one
who is famed for memory and voice sings the narrative lines of the
ballad, while the dancers move in a chain or ring, holding hands.
They halt and chant the refrain before moving in the other direc¬
tion. The scene of these dances, before they descended to humble
bams, was the noble garth. They were a fit entertainment for the
gentry, who led the movements, and a spectacle for kings. They
were, however, and remain in the Faeroes, essentially local and
communal in character. The subjects might be international or
national, but the treatment and outlook were invariably circum¬
scribed by the interests and requirements of a community closely
linked with the soil.

Collection and publication of Danish ballads began in a good
epoch. The hundred which A. S. Vedel presented to Queen
Sophia, at her request, in 1591, belonged to a period when the
medieval treasures of song were still unimpaired and when creation
was still active. The husband of VedeFs patroness lives in the
‘viser’ for his exploits in Dittmarschen. The collector, it is true,
was not above correcting his originals; yet VedeFs emendations
have themselves become traditional. A second hundred were
gathered by Peder Syv in 1695, while reprinting VedeFs pieces.
Taking manuscript evidence into account, Abrahamson, Nyerup,
and Rahbek reproduced 222 ballads and a number of tunes in
1812-14. Theirs was still the epoch of amateur editing. The whole
science of ballad study was altered by Sven Grundtvig in the great
collection which began to see the light in 1853. Grundtvig would
neither 'improve 1 his originals nor yield to the temptation to make
a ‘critical* edition. He realized that every version and variant of
a tradition has its own authority. Mere variants—small verbal
alterations—could be grouped in footnotes, with appropriate
acknowledgements; but each version of a ballad—involving
changes of some moment—must stand intact beside the other
versions. The editor copied manuscripts with the strictest faithful¬
ness, and drew both on oral tradition and on that of the other
Scandinavian peoples. He himself collected the folk-songs of Ice¬
land and the Faeroes. For Sweden he could rely on the work of
Afzelius and Geijer and that of Arwidsson; for Norway, on Land-
stad. To each ballad he prefixed an essay ranging over the whole
of the relevant European evidence. Curiously enough one of the
most fruitful sources for improving Grundtvig’s work was one very

2I0 NORDIC BALLADS

near to hand: the oral tradition of his own countrymen. E. T.
Kristensen found many an old ballad still alive m Jutland, still
sung to an ancient melody. With this material and his own expert
knowledge tael Olrik expanded Grundmg's work from five to

WTobSS'some idea of the state of poet,? immediately ante-
cedent to the ‘viser’ by consulting the prose of Saxo Grammaticus.
His Historia Danica was commenced, at the instigation of Arc: -
bishop Absalom, about the year 1185, and completed after *
patron’s death, about 1208. As sources for his work Saxo drew on
many poetical compilations, some of which he r epr° d u ces in Latiri
verse. The glance of a historian is directed backwards. We nee
not suppose that strictly contemporary verses received a ready we -
come from Saxo, so that his silence as to the viser is not neces¬
sarily evidence against them. He certainly does give an inventory
of the older verses bearing on history and mythology which were

accessible to men of his time. The rise of the viser corresponds
with the fundamental change in prosody from alliteration to asso¬
nance, and this change must have taken place during the twelfth
century. But much of the older alliterative poetry continued to be
heard,'and with such frequency and acceptance that it has been
transformed into ‘viser’. It is not necessary to believe that such
ballads are the oldest of their kind. The change could occur at any
date at which the old heroic poems were still remembered, i hey
are of the oldest lineage, however, and contain embedded m them

archaisms of thought and language. . .

The poetic Edda circulated freely. The extant Codex Regius is
a somewhat fortuitous collection of such poems, and we have to
take into account not only the pieces contained in that manuscript,
but also others often grouped by critics as Eddtca minora. I he
mythological poems, however, were already outmoded. . 1 hough
highly dramatic, they are not well suited to become narratives, and
their paganism had become an offence. The ballads draw only on
the Thrymskvi&a, the most human of all the Eddie mythological
poems The god is transformed into a heroic farmer, and his
associates might almost be bonders. Names of ritual significance
are not remembered or are confused, and the divine or gigantic
aspects of the action are not underlined. So the ballad of Tord af
Havsgaard (DGF 1), well known in Denmark, Sweden, and Nor-
1 See Note G, p. 387.

211

SCANDINAVIA

way, survives because of its broad humour, having shed the
subtlety of the original poet, whose pleasure it was to shoot barbed
arrows of wit against his gods. A pagan appeals to ‘Oden Asagrim’
in the Swedish Proud Sir Alf( Arwidsson 2), but otherwise the gods
are unknown to ballad-mongers .in Scandinavia. The vaguer under¬
world of elves, nixes, trolls, and sprites persisted in the imagination
of men, and gave to the ‘viser 5 their shuddering dread of the super¬
natural. In Norway particularly, where the conditions of life are
peculiarly forbidding, ballads of malicious trolls and kobolds are so
numerous as to constitute a distinguishing mark of the region.

If the first half of the literature represented by the Codex
Regius had little influence on the ballads, the second entered almost
intact. The adventures of Sigurd and Gunhild have given rise to
traditional ballads which are doubtless but the residue of a more
copious older corpus. In the Edda the poets already handle that
saga episodically; the ballads follow suit, using the same episodes.
In matters of detail the German Nibelungenlied brought correc¬
tions and variations which were used to modify the tradition in and
behind the Edda. The second lay of Helgi runs close to a ballad,
the Smpdagsmal is assuredly a source of Ungen Svejdal (DGF 70),
and the Waking of Angantyr (among the Eddica minora) of Angelfyr
and Helmet Kamp (DGF 19). Saxo tells this story in his fifth book,
and in the seventh he devoted many paragraphs to the tragic
history of Hagbard and Signe, rendering the vernacular verse in
various Latin metres. Saxo gives us the proper dynastic pomp and
circumstance; the ballad poet (DGF 20) seizes on the picturesque
detail that this unwanted wooer should have penetrated to Signe s
chamber in a female garb, and on the tragic tableau of his death,
hanged before her eyes. Known also in Sweden (Bergstrom 22),
the story was familiar in a highly generalized form in Germany
(Erk-Bohme 140) and Flanders (Hoffmann von Fallersleben 14).
Saxo relates old legends of Skjold and Hrolf Kraki which have not
given rise to ballads. He puts into hexameters the Bjarkamdl ,
which was sung in the vernacular to St. Olaf on the morning of the
battle of Stiklastad (1030). That poem, and the later StarkaSsmal,
belong to an age of transition. They are concerned with heroic
defences against desperate odds. Bjarki, who is the counterpart of
our own Beowulf, defended the hall of Lejre while it blazed behind
him. He displayed the narrow heroic virtues and spoke the heroic
language, but he is not a great wanderer like Sigurd or the sagamen,

2I2 NORDIC BALLADS

and his history is episodic. Under other circumstances he might
have been immortalized by a ballad.

Amid all this which is old, there are suggestions in Saxo of a new
spirit at work. He tells us that a Saxon singer, wishing to warn
Knut Lavard that treachery was being prepared against him
(Haraldstad, 1131), ‘quod Canutum Saxonici et ritus et nominis
amantissimum scisset’, sang him the famous song of Grimhild’s
treachery towards her brothers. It is the matter of the Nibelungen -
lied in some older form, and also of the Vise 5 Grimhild s Vengeance
(DGF s). Somewhat later, in 1157, a German singer composed a
partisan poem against King Svend Grade, full of abuse. It is
clearly different from the extant ballad of Svend Grade (DGF 118),
which sang of the murder of Knut Magnusson by King Svend.
Scarcely two months later, at the battle of Gradehede, a singer rode
between the lines to kindle the enthusiasm of Valdemar’s troops
by singing 4 parricidalem Svenonis perfidiam famoso carmine . The
song can hardly have been other than the vise , tecovered fiom a
seventeenth-century manuscript. To achieve its effect the song
4 must have been in Danish.

'We are thus brought to the evidence of the historical ballads
themselves. They form a well-nourished group among the viser ,
and thanks to their uniformity of style and sentiment they are the
cement of the whole collection. The oldest belong to the era of
nation-building under the two Valdemars 1202—4 1 )*

There is a new doctrine of royalty implied by an ancient lyric
embedded in one of the Diderik ballads:

The King he rules the city,
he rules over all the land,
and over so many bold heroes
with naked sword in hand.

But the King he rules the city.

Let the bonder rule his dwelling,
let the courtier rule his horse;
the King, the King of Denmark
he rules both city and force!

But the King he rules the city. (DGF 8.)

Such a king is no longer a distributor of gold rings to a retinue of
purely personal followers. The heroic hall has ceased to be the
place of first importance; its place has been taken by the boroughs,
backed by the farmers, whose defence is the duty of the military

SCANDINAVIA 213

and courtly class. This is the attitude of the poet who composed the
ballads concerning Valdemar IFs queens Bengerd (or Berengaria
of Portugal) and Dagmar. Bengerd has been singled out for
condemnation. As a morning-gift after her marriage to the king
she demands the imposition of heavy taxes on the bonders (DGF
139); she is thus contrasted with her predecessor Dagmar, the
merciful (DGF 133.) A beautiful ballad entitled Queen Dagmar’$
Death (DGF 135) increases the antithesis. Somewhat later, Sir
Tidemand’s Murder (UDV 60) is concerned with the opposition by
the countrymen to the unpopular ploughpenny tax, probably in
the reign of Erik Ploughpenny (1241-50). The frontier battle of
Lena (1208), an attempt by the Danes from Skane to recover
control over Sweden, is the subject of a ballad which the competent
authorities do not hesitate to class as contemporary (DGF 136).
It is encountered in Sweden too (Arwidsson 153), as are also the
earlier ballads of Esbern Snare, Sir Stig, Tovelille, and also Queen
Dagmar’s Death. The fact that so many historical ballads have
migrated from Denmark to Sw T eden is one proof of the exceptional
intimacy existing between those two lands. The oldest indepen¬
dent ballad in the Swedish series is King Birger and his Brothers
(Bergstrom 94), which is also known in Denmark (DGF 154). This
refers to a royal family tragedy of the years 1317-18. In Norway
the oldest indigenous historical ballad seems to be King Haagen
Haagenson’s Death (DGF 142), preserved in Danish, and it must
have been soon followed by the Norse originals of the Faeroese
Frugvin Margreta and Eyduns rinia (Hammershaimb FA 19, 18,
DGF iii, pp. 921-3). There is some record in the sixteenth century
concerning this ballad in its Norse form (Marittepd Nor dues). The
theme is the obscure fate of that Maid of Norway w r ho brought to
an end the old Scottish dynasty and gave cause for the fine ballad
of Sir Patrick Spans . According to the Scottish tradition she lies
where

Haf ow r re, haf owre to Aberdour,
it’s fiftie fadom deip.

The Norse singer, however, is a partisan of the woman who claimed
to be the Maid, and was burned at Nordnes in 1301.

It is clear from these examples that the composition of f viser 5 was
active in Denmark from the earliest years of the thirteenth century.
In Sweden the fashion was one of Danish origin, but was fully
naturalized in the fourteenth century, and in Norway ‘viser 5 seem

2I4 NORDIC BALLADS

to have arisen midway between these two dates. To carry the
Danish ‘viser’ back to the twelfth century is something not beyond
a doubt. Those purporting to relate twelfth-century events are
somewhat novelistic, and the facts cannot be ascertained without

peradventure. Yet the ballads are relatively numerous. They begin

before the accession of Yaldemar I, with Erik Emun and Sorte Plog
(DGF 116) Sorte Plog murders the king to avenge his brother;
the details are not as in Saxo, and another authority supposes
that Sorte Plog was avenging his father. The Sir Stig Hvide who
perished at Yiborg in ii S i (DGF 117) is probably the same person
as the Sir Stig Hvide who enchanted with his runes Rigisse,
a king’s sister (DGF 76, Bergstrom 46). In so far as it concerns
rune-casting the ballad belongs to the supernatural group; the
description of Sir Stig’s palace, which is all that the Swedes have
remembered of this tradition, is a set-piece of baronial magnificence,
like the description of Digenis Akritas’s paradise on the Euphrates.

Then when they came to little Stig’s port,
there played a hind and danced a hart,

So joyfully.

Then when they came Stig’s gate within,
there played a hart and danced a hind.

Then when they came within the hall,
the gleam of gold was over all.

The roof it was with gold bedight,
the floor was spread with brass so bright.

The oven was of marble-stone
and all the walls of elfin-bone.

The board and stools of reddest gold,
but the tablecloth of acre’s wool.

So joyfully. (Bergstrom 46.)

In Saxo, as we have seen, there is almost a contemporary reference
to the ballad of Svend Grade (DGF 118); a quotation would have
clinched the proof. Valdemar I is remembered for his domestic
misfortunes, which may be mostly imaginary. The role of a jealous
wife is given to Queen Sophia, who is alleged to have burned the
paramour Tove (DGF 121, 122, Bergstrom 43) in a bath house,
to have embroiled Yaldemar with his sister Little Christine
(DGF 126), and to have perished in a similar attempt to ruin her
daughter (DGF 127). As to the truth of all this, one cannot say
more than that Valdemar had an illegitimate son named Kristoffer,

SCANDINAVIA 215

for whom he felt great affection; and that doubtless the marriage
with Sophia was an occasion for setting aside his mother, by
name Tove or whatever it may have been. The opposition of
gentleness to furious passion in the ballad is cause enough for
the success of Tovelille throughout all the North.

Thanks to a small, but very well defined, group of ‘viser’ (. DGF ,
145) the name of Marsk Stig has been added to those of Robin
Hood, the Cid, and Marko Kraljevic, among the heroes of the
peoples. The cycle has served, along with that of Robin Hood, for
one of the most important contributions to the theory of epos and
ballad. 1 There is a long ballad of Marsk Stig, as there is one of
Robin Hood, stitched together from shorter fragmentary ballads.
The effect, however, is not at all epical in either case: they remain
simply long ballads. Thus the whole process by which the
Homeric and other epics were supposed to have been stitched
together by rhapsodes, working with the materials of ballads, is
evidently unfeasible. One does not produce epics in that manner,
but only longer ballads. Marsk Stig is the champion of basic
human rights. His exile is a protest against domestic tyranny, and
his conduct a pattern of honourable independence. In 1286 King
Erik Gipping was murdered at Finderup; in 1287 Marsk Stig was
outlawed. The ballads suggest he was outlawed because of the
king’s unhallowed longing for his wife. Like a new Bathsheba, she
was exposed to solicitation while her husband was at the wars.
According to the long Vise’, which harmonized the episodes of the
shorter ones, Stig had an evil dream, which was confirmed when
Erik’s messengers arrived to summon him to lead an expedition.
Shortly afterwards the King arrived and attempted to corrupt Fru
Ingeborg. She repelled his advances. Stig returned secretly, and
hearing of this affair, swore not to rest till he had avenged his lady.
He killed Erik at Finderup, and rode to court with this news, in
order to reproach the queen with her lasciviousness. The new king,
Kristoffer II, immediately banished the hero.

The defence of the frontiers gives rise to a few historical pieces
of merit, beginning with The Battle of Lena in 1208. The finest of
these is Niels Ebhesen , of 1340 (DGF 156). The grounds of his
quarrel with the Holsteiners are, in the ballad manner, personal;
but, though not explicitly recognized, there is an element of Danish
patriotism in his deeds as in those of the epical Olger Danske.

1 A. Heusler, Lied und Epos, Dortmund, 1905.

2i6 NORDIC BALLADS

Frederick II’s incursion into Dittmarschen concerns the same
frontier, and there is a stirring ballad on the defeat of King b
by Queen Margaret, when he attempted to seize the Danish
(DGF 159). Like the coast of Germany, the Danish harbours were
molested by sea-rovers in the later Middle Ages: the names o
of Tonsberg and Jon and Lave Rimaardson are preserved in ballads,
to warn their imitators. For the rest, the histones preserve the
memory of many private tragedies— murders, insults, and iavis '
ments-the details of which are hard to verify. They are to be
found equally in Sweden and Denmark. A notable subdivision
contains the ballads which describe assaults on cloistered nuns.
The most elaborate is that concerning Sir Karl (UDV 2I2 > Be ^'
strom 24), in which the hero entered the cloister as a pretended
corpse on a bier. The line of historical ballads continues in both
countries, persisting longest in Sweden. As ate as *739 ^

Scottish captain Malcolm Sinclair became the subject of a ballad
in the true traditional manner (Bergstrom 104).

In lands where the ballads are of relatively recent importation,
the historical class tends to be wanting. That is true of Iceland an
the Faeroes. In the Faeroes, however, the memory remains of the
old Norse verses concerning the ill-fated Maid of Norway, and in
quite recent times, thanks to the vigorous life of the genre, local
happenings have been versified. In late ‘viser’ a mocking note can
be heard which is not characteristic of the best age, and horror

takes the place of tragedy. . . ,

Among the ballads of literary origin, those derived fiom the
Edda and the Thidrekssaga are of the greatest importance. From
the Edda come Tord af Havsgaard , Ungen Svejdal and, perhaps,
Sven Vonved , Sivard Snarensvend, Gramvold Kongesen, an
Framdehevn (these three belonging to the Nibelung legend), and
the famous Ribald and Guldborg (DGF 82), which is probably to be
associated with the second lay of Helgi. The connexion between
lay and ballad is not often close, and in this last instance it is by no
means obvious. Ribold dies because his snatched bride pro¬
nounces his name, thus breaking a tabu. The ballad is one of
magic, and in that respect parallel to many others freely invented
by the poets. It is one of the most eerie of its kind, and has taken
so firm a hold on the imagination of the Scandinavian and Scottish
people that it is still alive and still receives new settings. In the
American mountains the ballad is called The Seven Sleepers , and

SCANDINAVIA 217

the events are supposed to have occurred within living memory.
The excitement of bride-stealing and battle, which opens the piece,
contrasts with the pathos of the last scenes, when the wounded
lover, trying vainly to hide his sores, comes home to die. Grimhild's
Revenge (DGF 5) derives, not from the Edda, but from the Nibe-
lungenlied , but the Faeroese ballads of the cycle combine more than
one source. 1 The Norse ballads (Landstad 9-11) are also character¬
ized by their independence. The battle with the dragon in Ragnar
Lodbroks Saga was a popular feature, re-created in the new manner
by the authors of Regnar and little Krage and The Dragon-fight
(DGF 22-4). It was also freely imitated in Riholi's fight with the
Dragon and Aller (DGF 27). The poetical fragments embedded in
the Hervararsaga are evidence of the great age of this tradition.
The sword Tyrfing—named after the ancient Tervingi—is in¬
vincible indeed, but brings death to him who holds it. It causes
two brothers to fight together in Samso, and a father to avenge the
one who falls (Alf i Odder skjasr or Angelfyr and Helmer Kamp,
DGF 19). A fine symbolism sublimates the ballad of The Avenger's
Sword (DGF 25). The sword is inspired with a rage for vengeance;
having killed, it lusts to kill again.

Now lay thee still, thou shining brand,
now lay thee still in Our Lord’s name.

Then spake the sword in weary mood:

4 Now lust I for thine own heart’s blood.

Hadst thou not named me by my name,
right now should I have been thy bane/

Hagbard and Signe (DGF 20) is a relic of the old Danish epos of
that name. It is one of the finest of all ballads, and one frequently
imitated. Corel's Daughter and Count Henrik and Sir Carl and
Lady Rigmor and Alvar LeSiesak in Norway (Landstad 32) are
related to it, and in Germany and Flanders the most characteristic
episode is a ballad motif. There is tenderness in the tragedy, and
the irresistible power of love; Hagbard and Signe are the Tristan
and Isolde of the Scandinavian North.

A striking group of ballads derives from the Thidrikssaga , trans¬
lated from Low German at the beginning of the thirteenth century.
There is the gathering of the German paladins in They were seven
and ten times seven , the expedition against Brittany or Berting’s
Land, the fight between Svend Ungersvend and Berner the giant,

1 H. de Boor, Die far oischen Lieder des Nibelungenzyklus, Heidelberg, 1918.

4615 p £

2I g NORDIC BALLADS

Diderik’s adventure with the lion, and the feats of Samson (DGF
6-q ii). Olger Danske appears as one of the German paladins,
but’he is also represented as a king of Jutland, and as such meets

„d routs Diderit's champion (DGF 17). H «^
French Osier, and his early prowess rs the subject of the Carol-
inman ballad of Olger and Burmand ( DGF 30). There is not other-
wfse much use made of the Carolingian traditions. In Norway
there remain fragments of a ballad of Roncesvalles, and m the
Faeroes there is a group carved out of the Karlamagnussaga ra
quite recent times. Olger fitted into the viser because of the
chance that he bore the title of ‘Denmark 1 and because he is one of
many giant-killers. His resistance to Didenk is on hne f ^ < ? own
bv the Thidrekssaga : Diderik boasts his unmatched might, he ears
that there is a possible rival in Denmark; he advances to challenge

him ~~ stout Diderik sent word to Olger the Dane,
and did this word indite:

Or would he give him tribute-gold,
or on the marches fight;—

but, contrary to precedent, Diderik was routed. „ ,

Minor sagas, like those of Hromund Gnpson, Oim Stoiolf s
Sons IllugC Hermund illi, and Asmund, were also used as sources
for ballads, notably in the Faeroes. The age of such adaptations
varies. As late as the first half of last century St Olaf s Saga gave
the Faeroese Ormurin langi (Hammershaimb FA 35 ) 111 best
traditional style. Among the Arthurian legends those of Tristan
alone appealed to the Scandinavian taste. Tristan s death is sung
in Iceland and the Faeroes (Grundtvig 23, Hammershaimb
27). There is evidence that this ballad was once known m

JO enmark ,

The religious balladry of the north is also full of good things
(DGF 06-113). The usual divisions occur: biblical extracts, saints
lives, moral tales. The Magdalene is, as elsewhere, confused with
the Woman of Samaria. St. Olaf (DGF 50, 51)*the patron of the
north, and his fights with trolls are just like those of other heroes,
but more edifying. Innocence is triumphant, even on a blazing
pyre (DGF 108,109), and guilt is suitably pumshed. Many of these
ballads are international: there are Danish representatives of the
St. Catherine ballad, St. Stephen and Herod or the miracle of
the capon, Little BiUee, St. George, the Sultan’s daughter, and the

SCANDINAVIA 219

journey to Eastland "a lo divine’ . Hallewijn has, of course, its
representatives in Scandinavia and Scotland, and it was here that
the lady-killer became conflated with the nixes which waylay
damsels beside the torrents, so that the legend takes on a super¬
natural tinge. The sentiment of these old pieces is uniformly
Catholic.

The religious poems (Legendeviser) and the heroic ballads
(kaempeviser) are either based on previous written or oral litera¬
ture, or formed according to the same patterns. The adventure
ballads of free invention are divided by experts into ‘trylleviser’
and ‘ridderviser’; the former are magical, the latter normal. The
distinction is an acknowledgement of the singularly important
place held by the supernatural in the Nordic imagination. The
ancient gods, it is true, had vanished before the ballads flourished;
but the ancient fears had not abated. Nature still sent a shiver
down men’s spines. The sea swallowed ships in storms, and the
rivers devoured maidens in spate. Every bridge was unstable, and
beneath every bridge was a murderous nix. Mermen and mermaids,
dwarfs and elves, trolls and dragons beset these men of the North,
who v T ere not such sound Christians as to defeat the enemy by
making the sign of the cross. Runes were mightier than the Chris¬
tian sign, and a powerful harpist might make the chthonian powers
disgorge their prey. The ‘trylleviser’ are thus a distinguishing
merit of Scandinavian balladry, which is shared with that of Scot¬
land. The atmosphere of mystery and dread which fills so many
great ballads is unknown to the brilliant South, and rare in
Germany. The Russians and Bulgarians have much to sing about
malicious supernatural beings, but they have not the same cower¬
ing dread. The Latvians and Lithuanians remember their old
pantheon in detail, but Perkons and Laime are friendly little
domestic creatures. It is not so in the Scandinavian lands. The
twilight of Asgard seems to have left men without any friendly
protectors to face the same terrible Nature, accompanied by fog,
storm, ice, and rock.

The ‘trylleviser’ occupy the space from Nos. 33 to 95 in Grundt-
vig’s collection (DGF) and are abundantly represented in all other
compilations. Germand Gladensvend (DGF 33) is among the finest
of these pieces, and might be a survival of some heroic poem.
Devoted, at the moment of birth, to a monster of the sea, Germand
was brought up by his mother till he was almost a man. He sought

220 NORDIC BALLADS

a bride in England, but was caught and marked by the merman, who
drank half his blood; on his return journey he vanished altogether.
The fact that he makes his trips in a coat of feathers indicates that
the whole legend is pre-Christian. Matthew Arnold has made
English readers familiar with the ways of a merman and a maid.
The group of ‘viser’ using this motif comprises one of the nix’s
wooing, and another of his married infelicity. He woos a maiden
at the church door (DGF 39), taking the form of a knight; he reveals
himself only as he drowns her in the waves. The nix’s wife begs
to revisit her home (DGF 38); she forgets her husband there, until
he drops an apple in her lap. These ballads occur in Denmark,
Sweden, Norway, the Faeroes, and Iceland. In Sweden the mer¬
man has become a hill-sprite (Bergstrom i), but the story is the
same. Both legends are firmly located in Low Germany, and in
one case identified with the Jade Estuary (Erk-Bohme 1). Haupt
and Schmaler (i, 34 Wodny muz) reproduce the second ballad as it is
known among the Lusatian Wends. In France the first one is
probably the source of La belle Helene , who is drowned while
dancing over a bridge; but a moral twist has been given to the story.
Olrik and Grundtvig held that the Danish ballad was translated
from the German in the latest years of the Middle Ages, but Erk
and Bohme thought it might be Slavonic. There does not seem any
convincing reason for removing from the North Sea coast a pair
of ballads so obviously at home there. Rosmer (DGF 41), the ballad
cited by Shakespeare as Child Roland to the dark Tower came , is one
of romantic adventure, with a tinge of the supernatural. He wins
his bride back from a Greenland ogre by a trick. In the Faeroes
the hero is called Rolf the Ganger. Mermaidens are less notice¬
able than mermen. A certain Sir Peder or Hillebrand (Bergstrom
77) or Villfar (Landstad 55) recovered his sister from the power of
one of them.

Elves, trolls, hill-sprites, and dwarfs give cause for many ‘viser’
in all the Scandinavian countries. Goethe’s Erlkonig is based, in
the words of one commentator, ‘on a rude Danish ballad in which
a young nobleman falls in with the elves and thereby receives his
doom to death’. ‘Goethe’s treatment (he goes on to say) is far more
imaginative and subjective than that of the original.’ Yet one may
go back from Erlkonig to Elveskud (Elf-shot, DGF 47) with a
sharpened appetite. Not only is the narrative used by Goethe
somewhat different, but his use of atmosphere and dramatic sug-

SCANDINAVIA 221

gestion are palpably literary devices which contrast with the un-
emphatic directness of the original. The original’s merits depend
not on a form of words—essential in the case of Erlkonig —since the
words change with each version; the appeal is made by the inherent
mysterious dread of the events staring starkly through the im¬
personal expression. The ballad is at home in Denmark. It has
travelled over all the North: the exact affiliation of the Faeroese
ballad is shown by the name given to the hero Olavur Riddararos
('knightly-rose’), which derives from the Icelandic Liljuros (lily-
rose’), as that from the Norse Liljukrans (lily-wreath’). In
England and Scotland it is Clerk Colvill , in Lusatia and Czecho¬
slovakia The luckless Marriage (Haupt and Schmaler i, 3, Susil 89),
in Brittany Count Nann (Luzel i, p. 5), in France Le roi Renaud
(Doncieux 7), in Italy Sore-wounded or Count Anzolin (Nigra 21,
22), in Catalonia Count Ramon (Mila 204, 210), and in Spain and
Portugal Don Pedro (Menendez y Pelayo, Antologia x, p. no,
Braga, Romanceiro i, p. 627). In Germany, apart from the modem
translations by Herder (accepted in the Wunderhorn as an original
German folk-song) and Grimm, the motif was used for the adven¬
ture story of the Ritter von Stauffenberg. It is the most travelled
of all the ‘viser’. In the lands to the south, where elfin superstitions
were not rife, the interest is excited by the concluding tableau: the
young knight, mortally wounded, who conceals his death from his
newly wed or pregnant bride. The human tragedy of this poem is
as keen as its ghostly dread.

Elf-hillock (Elvehoj) has travelled less, but is scarcely inferior
(DGF 46), followed by Sir Bosmer , Malfred , Duke Magnus, &c.
The Icelanders have created an unusual monster, the 'stafro 5
(Grundtvig 9), which appears to be a combination of elf-queen and
hind; she bewitches Kari so that he forgets his runes.

Then there are the revenants. Two of these are of exceptional
interest: the Lenore-motif in Aage and Else (DGF 90) and the step-
mother-motif in Moderen under Mulde (DGF 89). In the waking
of Angantyr and of Groa, Old Norse poems embodied the belief
that strong spells could rouse the dead to aid the living; and the
strongest of such spells, in ballad poetry, is love: the love of a
mother for a child or of a lover for the beloved. The bitter weeping
of Else wakens Aage from his rest, so that he comes and rides with
her to a common tomb; the bitter weeping of an injured child
wakens her mother to rise and reprove the cruel stepmother. Both

222 NORDIC BALLADS

ballads have spread very widely. The motifs might be deemed
universal, were it not that in Germany and the south credulity does
not go so far as to suppose the dead able to rise. The stepmother
ballad thus reduces itself to a lyric: the useless tears of orphans.
In our own Sweet William's Ghost (Child 77) we feel the full force
of the Northern superstition of Aage and Else , which has spread
through Germany to Czechoslovakia also. The stepmother ballad
is to be encountered in Denmark, Sweden, Norway, the Faeroes,
Iceland, Germany (Erk-Bohme 190), Lusatia (Haupt and Schmaler
i, 132, ii, 102), Czechoslovakia (Susil 159), Hungary (Wlislocki 3),
Lithuania (Rhesa 24), Latvia (Endzelin, Lett, Lesebuch 33),
Esthonia (Hurt i, 68), and also Poland and the Ukraine. Sir
Morten of Fogelsang (DGF 92) is also an impressive figure,
compelled, like King Hamlet, to revisit the glimpses of the
moon.

Lastly, there are various kinds of magic: tabu, witchcraft, runes,
the power of music. The breaking of a tabu on a name is the main
feature of Ribold and Guldborg (DGF 82) and of Blok og Ravn hin
brune (DGF 62). The malicious powers of witches, who were often
also stepmothers, sufficed to change boys and girls into ravens,
nightingales, hinds, werewolves, snakes, and trees. Some of these
have travelled: Jomfru i Hindeham (DGF 58) becomes the French
La Biche blanche (Doncieux 16). They may not all be indigenous
to Scandinavia, since superstitions of this kind are universal; but
what is characteristic is the presence of so many examples of real
power in a single ballad corpus. The use of runes, both for
bewitchment and to compel love, is certainly a Scandinavian motif
(DGF 73-81). Three ballads show the power of music. Sir Verner
(DGF 383) escapes from prison, thanks to his singing; the same
motif is familiar in France. The Harp's Power (DGF 40) tells how
Sir Villemand recovered his drowned bride from the power of a
troll at Blide-bro.

Sir Villemand took his harp in hand,
he went before the stream to stand.

He played his harp so soft and low,
there moved no fowl upon the bough.

He struck his harp so sore and hard,
the sound was heard in all the garths.

The bark it sprang from the oak tree,
the horn sprang from the well-fed kye.

223

SCANDINAVIA

The bark it sprang off from the birch,
and all the studs from Mary’s church.

He struck his harp to hurt and harm,

—The strings are of gold —
and struck his bride from the foul troll’s arm.

So fairly he played for his lady .

The power of music to disturb and disrupt appears also in the
Russian ballad of Sadko and some Greek songs. In another song
the action is performed by a harp, so that the harp is often named
in the title. We know it rather as The Twa Sisters (Child 10), and
that title is also used by some Scandinavian collectors. The speaking
Harp or Chords (DGF 95) tells how one sister murdered another
and made her relics into a harp; the chords of the harp revealed
the crime.

The Tidderviser’, or adventures without the supernatural, are,
as elsewhere, chiefly histories of lovers: how they w-oo and are con¬
stant or unfaithful, how they test each other’s affection, how they
survive or succumb to opposition, and suffer from the force or
wiles of rivals, how they elope, how brides are stolen or ravished,
and brothers and sisters may unwittingly commit incest, how love
leads on to death and murder, and how vengeance is exacted. Some
ballads adopt a scoffing tone before love; others are concerned with
romantic material in which love plays only a small part. Some of
these adventure ballads extend to a considerable length, as Axel
and Valb org, Knut i Borg , Malfred and Mogens , Edmund and
Benedik. They are short and simple epyllia, and their number was
increased, towards the sixteenth century, by similar long ballads
imported from Germany, such as Tannhdmer and the Count of
Rome. The great majority of ballads, however, are of a more epi¬
sodic cast, and consequently much briefer. Sometimes it is evident,
as in the case of Lady Ingefred Torlufs Daughter (UDV 118), that
the ballad (a variant of Marianson) is of foreign origin; others have
an air of being records of fact; but in general the question of
originality hardly arises, since little more is attempted than the
rearrangement of a few simple common motifs.

There is charm in Proud Ingeborg's Disguise and The Utile Foot-
page ( UDV 185, 186), since one of the lovers is no more than a
stable-boy, though later revealed to be of higher degree. In many
cases the course of true love is interrupted by long absences,
resembling those of the Moringer cycle. Of such are Thor and

234 NORDIC BALLADS

Thure (UDV 54, 55, DGF 72), Lovmand and Thor (UDV 199), and
others. Opposition to lovers is more than once offered in the form
chosen by David, the king sending the unwanted suitor to the front
of the battle. In the fine Norse tale of Axel and Valborg (UDV 143)
the difficulty arises from the fact that the lovers are related in the
third degree. The king’s son Haagen contrives to have their pact
annulled, with an impressive ritual which is adequately described,
and to marry Valborg himself. In a war with Sweden which
follows, Haagen is killed. Axel avenges him; Axel dies victorious,
and Valborg enters a convent. There is less merit in Torkild
Trundeson (UDV 200) and in Malfred and Mogens (UDV 217,
DGF 49), which also seems to have been of Norse origin. The
rivalry of concubine and wife gives rise to the powerful Sir Peter
and his Concubine (DGF 210, UDV 158), and to other ballads of the
same sort. In Fair Anna (UDV 177), the rivals are sisters and
recognize each other on the bridal night. One of the most popular
ballads in the north is Sir Lave and Sir Jon, distinguished for its
mocking tone. Sir Jon is the Lochinvar of the north.

Trickery and force are used against rivals in a number of ‘viser’,
of which perhaps the best is Ebbe Skammelson (DGF 354, UDV
120). While Ebbe was at court his brother spread a false report of
his death, and so contrived to wed his fiancee; returning, and
being unable to persuade the bride to run away with him, he kills
her and his brother, and wounds both his parents. The ballad is
marked by a sense of actuality.

There are many elopements. It is evident that the custom of
bride-stealing was not far from the experience of those who com¬
posed ballads, and that a stolen wedding was the most interesting.
Such exploits led to fights with the young lady’s brothers and with
her father; and to these encounters a number of different solutions
are given. In Ribold and Guldborg, so often cited, and Hilde's
Sorrow (DGF 83), the bride makes an appeal for one last surviving
relative, naming her lover, who at once receives his deadly wound.
In others (The Lady in the Wood, Sir Helmer Blaa, &c.) the battle
ends in a reconciliation, or all agreement is refused and the fight
ends in death. The most dramatic of these pieces is Nilus and little
Hilde (DGF 325, UDV 121). Caught by a storm while eloping,
Nilus is forced to seek shelter with his lady’s brother—a bitter
enemy. The brother welcomes his sister, but not her lover, who
has to fight to the death:

225

SCANDINAVIA

Up then stood Sir Nil us,
and his bright sword he drew,
and as I tel! you verily
right manfully did hew.

They played a play , and the play was all of anger ,

Now there was Sir Niks,
and hewed he all so fast,
so long until his goodly sword
in the hilt asunder burst.

He parried with the cushions
and with the bolster blue,
and out before the bedroom door
received his deadly wound.

Then spake the bold Sir Niks,
in pain and hurt he cried :

‘Now come you forth, proud Hilde,
for now his time to ride!'

Now there was Sir Niks,
he sat him on his horse;
so rode he on to Hedingsholm,
for better or for worse.

They played a play , and the play was all of anger .

There are many tragedies of vengeance in the ‘viseri, some of
them commonplace and prosy (like Edmund and Benediktf but
others intense. A fierce motif descending from pagan days is the
need to destroy even children, lest there be a subsequent reprisal.
This is the moral of Ung Villum and Liden Engel ( UDV 126, 127),
in which a babe is hidden by its mother so as to grow up as an
avenger. The avenging sword, in the ballad cited in a former para¬
graph, slays the babe in the cradle before thirsting after the blood
of its owner. Daughters avenge their Father {UDV 171, DGF 193)
has the peculiarity of showing that the passion for vengeance could
be felt by women as well. There are many ballads, also, of wild
crimes, possibly- records of particularly revolting occurrences,
though they are now hard to identify: Sir Palle , Olaf Strangeson ,
Sir Jonas , &c. In Sir Truer $ Daughters (UDV 164), which is known
also in Sweden, Iceland, and the Faeroes, we have essentially the
same theme as the Russian Nine Brothers and their Sister : the out¬
law brother or brothers attack a girl or girls, who prove to be their
own flesh and blood. So the religious awe attaching to incest is
added to the horror of violence. In other ballads incest is used to

23 6 NORDIC BALLADS

whet the emotions, as in Sir Salletnand, Sir Sverkel, The Foundling ,
and the Icelandic Tale of Margreta.

Taken as a whole there is a sombre tone in the themes and the
verse of the North. Despite some mocking ballads (skaemteviser), one
does not encounter the light touch of French folk-song. It is not a
convention that love and marriage should be targets for wit, despite
the Tricked Suitor (UDV 122, DGF 229). The tone of the whole
Northern balladry is earnest, and its language and. concepts
intense. The form tends to diffuseness and lyricism; it is remark¬
able how the poets contrive to pour into their essentially lyrical
mould a content which is strictly narrative, and even dramatic.
The unity of style prevailing throughout the whole corpus is one
of its most remarkable features. The ‘viser’ must have been an
absorbing passion. The graver emotions are exploited with in¬
genuous mastery: love strong as death, jealousy, revenge, treachery,
and the formless, viewless horror of an evil world ever beside
man’s elbow.

For the sake of convenience this account of Scandinavian
balladry has been based on Danish collections, not merely because
they are on the whole older and better than the others, but also
because Danmarks gamle Folkeviser is, in some sort, a summation
of all other evidence on the subject. The survival in Danish of a
given ballad, however, is not necessarily proof of its Danish origin.
On the contrary, we have seen evidence of Norwegian creative
activity, which has enriched the common fund. The regional
differences of the ‘viser’, though interesting for the specialist, are
rarely such as to demand notice in a general statement like the
present. The Swedish songs resemble, when viewed as a corpus,
those of Denmark most intimately. Arwidsson and Bergstrom are
complementary collectors, so that, for instance, we find in Arwidsson
precisely those epical ballads which we miss in Bergstrom. Danish
is spoken in one of the Swedish provinces—in Skane—and the two
languages are so intimately allied that transference is almost im¬
mediate from the one to the other. On the other hand, there are
signs of specially close relations between Norway, Iceland, and the
Faeroes. The Norse ballads recorded by Landstad show consider¬
able originality. The activity of trolls and monsters is more con¬
stant and more formidable, and there is a greater use made of minor
and late sagas as the source of ballad stories. A very curious piece
in Landstad’s collection is DrmmekvxSi (7). Olaf Astason or

SCANDINAVIA 227

Orknison dreams of Hell and Paradise. He mentions Christ and
St. Michael, but he also mentions the Gjallarbrui—the bridge
spanning the abyss in old Norse mythology. The basis of the

dream is, in fact, heathen superstition; and for a considerable part
of its length the poem runs parallel to the minor Eddie poem
SolarljoS . It is supposed that Astason may be a corruption of
Aasgardson, and so connected with Ansgarius, the apostle of Nor¬
way, who died on 3 February 865.

The Faeroese ballads have the interesting feature of being still in
active life. They came late to the islands, but have proved wholly
congenial. Those of Iceland, though earlier, have not flourished.
They came into conflict with the local ‘rimur’, and were worsted
in the contest. 1 The Timur 5 appealed to the same public by means
of similar themes. The Icelanders have perforce relied on their
own resources for entertainment, both because of their distance
from European foci of literary fashion, and because the hard condi¬
tions of life keep the social units small. The 4 rimur 5 have circulated
by means of recitations in such circles continuously since the four¬
teenth century. They are, like the Tdser’, distinctively medieval.
They are narratives. There is no pretence of originality of inven¬
tion. Sir William Craigie has reproduced the Danish prose text
of the Gowrie Conspiracy, from which Einar Gudmundsson carved
his Skotlands Rimur somewhat later than 1620. There is no refrain
in these pieces, and full rhyme is used as against assonance. At
first sight the verse is simple: quatrains, tercets, &c., used without
variation throughout the piece. They were undoubtedly popular.
In all this the Timur’ resemble ballads, but in other ways they are
very different. They are subject to very complex conventions of
alliteration and rhyme, and they are grouped together under rules
which forbid the repetition of the same devices. Alliteration is
present in the ‘viser’, where it often gives pleasure, and it is a
prosodic principle of the songs of Finland and Esthonia; but it is
nowhere used with the precision demanded by the Icelandic
Timur’. The poet is generally known; there is no traditional
anonymity in his art. His object is not to contribute to a common
fund, but to display his art before connoisseurs. Relatively un¬
interested in the matter of the song, the Icelandic people were, and
are, acute critics of the form. In the long winter nights there is

1 See Sir William Craigie, The Art of Poetry in Iceland, Oxford, 1937, and
Skotlands Rimur, Oxford, 1908.

22S NORDIC BALLADS

time to discuss art in the minutest detail: in fact, prosodic study
took the place of the general conversation which often follows a
recitation in the American mountains, though the latter turns on
the subject-matter of the ballad. Only by research could the
elaborate, but quite conventional, kennings be understood, and to
serve this research the records of older literature were^ carefully
preserved. Scholarship is greatly indebted to the ‘rimur for their
by-products. If it is scarcely possible for a European taste to
esteem these poems for their own merits, though we may coldly
admire their intricacy, the ‘rimur’ serve to remind us that there is
no inevitability about the ‘popular’ in poetry. A ‘people’ can be
a people of connoisseurs. Indeed, an aesthetically enlightened
people would display connoisseurship in some respect of the ballad
art. The case of Iceland goes further, however, since it shows an
undoubted ‘people’ interested in the niceties and subtleties of an
advanced art, to the exclusion of those ready appeals to the under¬
standing and senses which are normally supposed to be popular .

3. England , Scotland , America

F. J. Child’s English and Scottish Ballads is designed to make
easy the comparison of the British and Scandinavian material.
The annotations draw deeply on those of the great Svend Grundt-
vig, whose criteria of admittance rule throughout the collection,
and who was actually consulted in forming the canon. The collec¬
tion contains, inevitably, much material which is doubtfully
traditional. Some English medieval pieces, retained in deference
to the Percy manuscript, seem hard to justify on any other ground.
On the other hand, the criterion of anonymity favours the balladry
of Scotland at the expense of its southern neighbour. Scottish
Border balladry rises to its zenith in the sixteenth century. English
balladry was not one whit less active in that century, but it had
taken a new direction. The invention and application of printing
was causing ballad-mongers to print their wares on broadsides; the
censorship forced them to register their pieces and names at
Stationers’ Hall. Thus we hear of William Marten, Richard
Beard, William Forrest, Henry Spooner, John Bradford, William
Elderton, and the rest of the motley company, whose wares have
been gathered into the Pepys, Wood, Shirburn, Bagford, and other
collections. Under other criteria some of this material might have
been included in the standard collection of our ballads. They were

ENGLAND, SCOTLAND, AMERICA 229

undoubtedly popular. Shakespeare hummed them, and expected
his audiences to pick up an allusion or parody in the space of a line
or two. Under other criteria a number of English carols might
have been included. The student of comparative balladry is, in
fact, somewhat under the thumb of the collectors, who have not all
worked with the same intentions. That which, in Child’s work,
helps us to make Scandinavian comparisons, hinders comparison
with Erk and Bohrne and Liliencron. At the same time it has to
be admitted that a printed text does not permit the full range of
traditional variations, and that the Tudor and Jacobean ballad-
mongers functioned rather as journalists than as entertainers. They
sought an immediate response of a pious or political sort, and they
were less scrupulous as to the means employed. There is already
apparent in them a decline in popular taste. The intuitive delicacy
of traditional verse is unknown to them, and by their prints they
contrived that it should be forgotten by the public also. It sur¬
vived in Scotland; and in Aberdeenshire, where society is still
notably self-centred and self-sufficing, the older traditional ballad
persisted to our own day. It was these that Child desired to in¬
clude, and his collection has a remarkable predominance of Scottish
pieces. 1

The Scottish and English elements of Child’s book cannot be
separated entirely. Sir Aldingar (59) is undoubtedly English, but
two of the significant versions are Scottish. Thomas Rymer (37) is
Scottish, but has the support of English manuscripts, and it is
clear that Thomas of Erceldoune’s reputation stood equally high
on both sides of the Border. The Battle of Otterburn and The
Hunting of the Cheviot (161, 162) are both English and Scottish,
though the sentiments are mostly English. The approximation of
the two countries is the more marked because the best ballads come
from that 'other English nation’, northern England. Northumber¬
land and Cumberland, Newcastle and Carlisle, are foci of Border
balladry, and those of outlaws —Adam Bell, the Robin Hood cycle
(115-54)— are located north of Trent. One may credit southern
counties with distinctively literary ballads, in all probability, but
with not many others. Their exiguous share in this work is partly
due to the drying up of balladry in and near London thanks to the
diffusion of broadside ballads in the early sixteenth century. The
pressure of the new style was particularly severe in the south. The
1 See Note H, p. 388.

23 o NORDIC BALLADS

south, however, was subject to considerable pressure from written
literature during the whole Middle Ages, from which the north,
sheltered by its dialect, was relatively free. It is really remarkable
how slight a trace has been left on traditional balladry by the wars
in France—the great historical theme of the southern counties.
Yet there Is no want of political poets of the calibre of Minot or
Drayton, masters of a semi-popular style. One is justified in con¬
sidering the dearth of southern ballads to be original, so that the
ballad England is north England as the ballad Scotland is the
Scotland where Tnglis 5 was spoken. It was from Bristol, Plymouth,
and southern ports that settlers went to New England and Virginia,
taking with them authentic traditional tunes and ballads. Though
apparently not indigenous, ballads did not fail to find acceptance
in the south.

The national elements of the English and Scottish Ballads,
then, cannot be separated; but they can be broadly distin¬
guished. English ballads are typically (i) the Robin Hood pieces
from middle England, (2) the historical narratives of the type of
Chevy Chase and Durham Field , and (3) picaresque and romantic
pieces. The Scottish ballads are, distinctively, (1) supernatural
narratives, (2) tragic love ballads, (3) Border ballads, and the later
ballads of feuds in Aberdeenshire and round about. 1 The English
ballads are recitatives; the Scottish ballads have a single or double
refrain in many cases, and are closely associated with Scandinavian
‘viser’ designed to be danced. Whether the Scottish pieces were
themselves danced there seems not sufficient evidence to deter¬
mine. The Scottish ballads are linked to Scandinavia, and the
ports of Aberdeenshire and Fife must have been marts of the traffic;
the English ballads typically show reliance on France, when they
have international material. That is true of the romantic ones, and
of the Christian ballads, The Maid and the Palmer , St. Stephen and
Herod , and Judas (21—3). They also offer means of comparison
with Germany and the lands dependent on German initiative, at
points where these touch the French tradition.

The age of British balladry is hard to determine. Collection
began, in the Percy Folio, at too late an epoch. By the reign of
Charles I we may feel sure that the Tudor broadside poets had
wrought considerable damage; it is evident in the damaged condi-

1 The distinction is made by W. Schmidt in the Neuphilologische Monats-
schrift, viii, 1937, p. 86.

ENGLAND, SCOTLAND, AMERICA 231

tion of the old ballads taken to America at that time or a little later.
There must have been heavy losses of medieval material. As no
distinction was made between new and old, doubtless there are
apparently old ballads (such as some of the Robin Hood cycle)
which are really quite modern inventions. Discriminating collectors
were unknown until we pass the date of Percy’s Reliques, and even
then they were unsystematic. The evidence offered by Child’s
collection is more debatable than that of Danmarks gamle Folke -
viser, and it is more hazardous to come to any negative conclusion.

The entire group of Scottish Border ballads, from Johnnie
Armstrong (169) to Farcy Reed (193), belongs to the sixteenth cen¬
tury, together with Mary Hamilton, Edom o’ Gordon , The Bonny
Earl o’ Murray, The Laird 0’ Logie, Willie Macintosh , and perhaps
Outlaw Murray (173, 178, 181-3, 305). They are followed by
thirteen ballads on Scottish battles and tragedies of the seventeenth
century (194-206), and three from the eighteenth (208-10). There
are other Scottish ballads, of a more domestic nature, which can
be referred to the seventeenth and' eighteenth centuries. Against
these we can set only a handful of early pieces: Sir Patrick Spens
(58), Glide Wallace (137, based on Blind Harry, c. 1460), the
Scottish Otterbum ballads (of uncertain date), and Harlaw (163).
Only Harlaw gives a reasonably secure date. The event occurred
in 1411, and it was of immediate local interest to Aberdonians.
The ballad is of the sort which arises directly out of the experience
it narrates, and we are, in any case, certain that it existed in 1549.
Sir Patrick Spens is, in its style, the finest of the Scottish ballads.
The words work on the imagination right from the conventional,
yet ominous, opening:

The king sits in Dumferling toune,
drinking the blude-reid wine.

He chooses a captain for his ship, and sends him his written com¬
mands :

The first line that Sir Patrick red,
a loud lauch lauched he;

The next line that Sir Patrick red,
the teir blinded his ee.

‘O wha is this has don this deid,
this ill deid don to me,

to send me out this time o’ the yeir,
to sail upon the se?’

233 NORDIC BALLADS

The ship is, as he dreaded, totally wrecked:

0 our Scots nobles were richt laith
to weet their cork-heild schoone;

bot lang owre a’ the play wer playd,
their hats they swam aboone.

In the ballad, however, there is nothing to say when this disaster
occurred. Some versions give Norway as the destination, and it is
a conjecture that the ballad speaks of the loss of the Maid of Nor¬
way in 1290; and it is only conjecturally that we can assign this
ballad to the early years of the fourteenth century. The Scottish
evidence thus indicates that there were only a few ballads before
the sixteenth century; the apogee of this genre in Scotland is to be
placed in that era. A late date for Scottish balladry is also com¬
patible with the circumstance that the adventure ballads held in
common with Denmark and Norway are more satisfactory in their
Scandinavian forms, and consequently almost all must be reckoned
imports into Scotland.

The English evidence allows for a higher antiquity, but is more
difficult to follow because of possible literary interference. Child
has a sturdy group of sixteenth-century ballads extending from
Andrew Barton to King James and Brown (167-80, from a.d. 15 ii-
78), and he reproduces four sea-songs from that era and the next
(285-8). This is the principal group of historical ballads, and it
would have been vastly increased had his editorial conscience
allowed him to include the Tudor political pieces by named
authors. The sixteenth century was the apogee of the historical
ballad in England as in Scotland. But there are also a considerable
number of English ballads of much earlier date. Excluding, for the
moment, Sir Aldingar (59)’ which refers to persons of the tenth
century, we find that Robin Hood was probably a real personage
of the twelfth century (dying in 1198), and that four ballads were
stitched into the Geste (117) before 1400. The circulation of
Robin Hood pieces is attested in 1377; nothing prevents the sup¬
position that they were composed much earlier. Queen Eleanor’s
Confession (156) and Sir Hugh (155) are not ballads to corroborate
the evidence of early date, though they refer to personages of the
twelfth and thirteenth centuries respectively, since the former may
be of literary origin, and the latter is a religious legend. In the
fourteenth century we encounter Sir Hugh Spencer (158), Durham
Field (159), Otterburn (161), and in the fifteenth century we find

ENGLAND, SCOTLAND, AMERICA 233

Robyn and Gandeleyn (115, manuscript of 1450), Agincouri (164),
Sir John Butler (165), The Rose of England (166, an allegory). Most
of these pieces are classified among the * ballads of minstrelsy’, that
is to say, they are such as to lie open to the caprices of self-conscious
art. Their testimony is not conclusive, but, taken cumulatively, it
seems to establish the existence of English ballads in the fourteenth
century, while leaving open the question of an earlier date. The
Otterburn ballads, referring to an event of the year 1388, were
probably extant shortly after 1400. Sir Philip Sidney is a witness
for Chevy Chase in 1559. Child took his version of Judas (23) from
a thirteenth-century manuscript; his St. Stephen and Herod (22)
is from a fifteenth-century manuscript.

Sir Aldingar (59) is the most important of our ballads, and is in
other ways one of the most interesting. The name given by Child
is that of the villain, Aldingar or Roddingham. Others have been
attracted by the name of the hero, Mimecan, which appears in the
Danish Ravengaard og Memering (DGF 13); in some English ver¬
sions this name is conventionalized as ‘Sir Hugh le Blond’. The
story is a sort of inversion of the legend of Joseph and Zuleikha; in
this case the man is the tempter, and he falsely accuses the innocent
wife. She is condemned to the flames by a credulous husband, but
surprisingly delivered by a champion whom no one could expect.
The reason for this unexpectedness, in our ballad, is that the
accuser is a giant, the champion Mimecan a dwarf. In other ver¬
sions, while there is some similar disparity of natural force, the
champion is made to come from a long distance, viz. from the
poet’s own country. The plot is probably not private to the ballad.
Paulus Diaconus tells the same story in a few words concerning the
Lombard queen Gundiberga in the seventh century, 1 and, in view
of the remarkable popularity of the Lombard History, there seems
no reason for refusing to consider that this text is the original germ
of the ballad. William of Malmesbury (c. 11 50) related the plot of
the ballad, insisting on the smallness of the champion (puerulum)
and the great bulk of the accuser (delatori giganteae moliminis
homini), and he adds the significant remark that this was a subject

1 Gundiberga . . . quum de crimine adulterii apud virum accusata fuisset,
proprius eius servus, Carellus nomine, a rege expetiit, ut cum eo, qui reginae
crimen ingesserat, pro castitate suae dominae monomachia dimicaret. Qui dum
criminatore illo singulare certamen inisset, cum cuncto populo astante, superavit.
Regina vero post hoc factum ad dignitatem pristinam rediit (De gestis Lango-
bardorum , iv, cap. 49).

4615 Hh

2 34 NORDIC BALLADS

for public recitations (et nostro adhuc sseculo etiam in triviis
cantitata). The language of this song was doubtless not that of the
sixteenth-century ballads we now read; the plot may have been
more fully expounded, but it cannot have extended to more than
the single significant episode. The poem was of unknown age even
in the middle years of the twelfth century (adhuc etiam). Mimicon
and Roddyngar are named by John Bromton in the fourteenth cen¬
tury, and Mimecan by Matthew of Westminster. From the time
of our first quotation there are fixed points in all versions of the
legend: the victim is an empress of Germany, married to a Henry;
she is condemned to be burned if no champion will appear, no
champion dares to fight the accuser save one who is a dwarf or
comes from a distance; the champion belongs to the poet s own
land. The Danish ballad is identical with the English one in all
essentials; Gunhild’s name persists in the Icelandic, Faeroese, and
some Danish versions.

Professor R. W. Chambers has indicated, in a lecture delivered
in Manchester, that the function of this legend was consolation.
The ruined Anglo-Saxons comforted themselves with a tale of
their lost dynasty, in which a Saxon hero vanquished an oppressor
seemingly as all-powerful as the Norman lords. If this be so, the
ballad is all the more pertinent to the Norman dynasty. Its later
development is not entirely confined to balladry, since it passed
into romance, with the Earl of Toulouse ; to the historical novel in
Gines Perez de Hita’s Guerras de Granada ; to the moral apologue
in Don Juan Manuel’s Conde Lucanor; to the opera in Wagner’s
Lohengrin. When Ariosto talks of the ‘aspre legge di Scozia’ as
condemning peccant wives to the flames, he bears witness that the
old English motif has become anybody’s plaything. It is not, as we
have seen, original to the English poet; but it was he who gave the
story a definitive form, and sent it rolling among his connexions
over all Europe. He also created, as it seems, the oldest English
ballad.

The cycle of Robin Hood and his merry men is historical in so
far as he was or was believed to be a real person. Some of them,
particularly the older ones, are expressions of a doubtless historical
emotion, the feud of Saxon and Norman. Those who oppress are
Normans: the sheriff of Nottingham and the richer clergy. The
oppressed are Saxons or associated, through their poverty, with
the Saxon lower classes. They find a typical representative in Sir

ENGLAND, SCOTLAND, AMERICA 235

Richard of the Lee. The king, however, Is regarded as no foreigner,
and the straggle Is more of classes than of races. For the rest, the
ballads of this cycle have little foundation in fact. As with the Cid
and Marko Kraljevic, the ballad-poets have created a type of hero
and of adventures, and the latter can be readily multiplied accord¬
ing to pattern. The hero is distinguished by his sense of fair play.
He seeks a rough justice for all, including himself, and he takes the
readiest means to that end. His sense of humour is lively, if not
subtle. Robin Hood can take a beating without malice, and he does
not lose his power to impose respect. Apart from the particular
history of his feud with the sheriff and abbot concerning Sir
Richard's property 7 , the episodes tend to repeat the tableau of good-
humoured cudgellings, and it is with one of the series that the
cycle Is closed.

A notable feature of the English and Scottish ballads Is the com¬
plete absence of contact with the ancient Germanic epos. The only
complete example of that epos is the Saxon Beowulf , and Fmnsburg
and Waldere are important fragments; Widsith is the completest
inventory. The existence of such documents must be due, in part,
to contact with the literate Romance area; but for the same reason
the traditional epics were the sooner eclipsed. In exchange for
these, English balladry includes some interesting Arthurian pieces:
The Boy and the Mantle , King Arthur and King Cornwall , The
Marriage of Sir Gawain (29—31). King Henry (32), a Scottish
ballad, is notable for its connexion with Celtic poetry on the one
side and Scandinavian on the other. It is a tale of the disenchant¬
ment of a hag into a beautiful princess. That is the motif of The
Marriage of Sir Gawain , at the close, but also of the West High¬
land tale of The Daughter of King Under-waves , and of the saga of
Hrolf Kraki. Disenchantment by a kiss is used In Kemp Owyne
(34), which is associated with the saga of Hjalmter ok Olver and
more remotely with Libeaus desconeus. In Denmark It Is The Maid
in Dragon-form (DGF 59).

Other ballads of a semi-literary type are Hind Horn (17), and
The Kitchie Boy (252), modelled upon it. This Is yet another varia¬
tion on the theme of the wanderer’s homecoming, of which the
Noble Moringer is the outstanding example. More literate forms
are the poetical romances of Horn ei Ry menhUd and Horn Child and
Maiden Rimnild. King John and the Bishop (45) is one of Burger s
sources. King Estmere, Sir Cawline , and Sir Lionel (60, 61, 18) are

236 NORDIC BALLADS

all English and ‘ballads of minstrelsy 5 ; while Blancheflour and
Jellyflorice (300), though Scottish, is a popularization of the well-
known French sentimental romance. An interesting piece of
Classical origin is King Orfeo (19). It is an example of the Classical
themes which can become completely popular. The persistence in
it of refrains in the Norrone dialect of Shetland is unique.

Child included few religious pieces, but they are good. The
Samaritan woman, judas, Dives, the miracle of the capon, and
the miracle of the cherry-tree (21-3, 54, 56) are English and
Scottish representatives of wide ballad families. In Lady Isabel
and the Elf-Knight (4), a Scottish ballad, we possess a fine repre¬
sentative of the Hallewijn family, in which not the slightest
reminiscence remains of the original Holofernes. The ballad is
at home in Holland and Flanders, but it has not come from there
to Britain. On the contrary, it has come from Norway or Den¬
mark, along with other imported pieces, and has acquired its elfin
character on that journey.

The use of literary formulas, so frequent in France and Ger¬
many, is rare in Scottish balladry. But The Grey Cock (248) is
an example of the ‘aubade 5 .

There are no mythological ballads in the English and Scottish
corpus. The lapse of time between the fall of the gods and the rise
of balladry was too great to be bridged. Good use is made, how¬
ever, of the lesser paganism of elves, revenants, and changelings;
less abundantly, perhaps, than in Scandinavia, but with effective¬
ness. The greater number of such ballads, though not all, are
Scottish. The elfin ballads are Thomas Rymer, The wee, wee Man,
Hind Etin, and Clerk Colvill (37, 38, 41,42). The last-named is the
Danish Elves hid, recommended not only by its words but also by
the air. 1 Thomas Rymer is based on the Scottish verse romance of
Thomas of Erceldoune, written before the middle of the fifteenth
century. The personage is historical, since Thomas seems to have

1 Clerk Colvill (first line)

(i) 525BE2

34 U8

dgg.g./a.swL/cga.

and

(ii)

(d. .g )/g.. .g./a.. .a .Id. ,xg/ a...

Elveskud

(i) 5A1221AD2B

J sharp 24 U8
d/g.fg/a.bc/b.ga/g.r

(ii) 5A14BBE5A14B

1 sharp 24 U16
d./g..fg. .b/b. .ag..d/g. .fg. .b/a.. .r

(iii) 52BAB212C1AD

j sharp 38 U8
d/g..(agf)/efga(afgf)/d

ENGLAND, SCOTLAND, AMERICA 237

prophesied In the later years of the thirteenth century; but the
ballad is separated from the person by the intervening romance.
It Is a case of popular verse which, by the working of tradition, has
become a ballad. The story of his ride to Elf-land, in company with
the Queen of the fairies, Is made memorable by one supremely vivid
stanza:

For forty days and forty nights

he wade thro red blude to the knee,
and he saw neither sun nor moon,
but he heard the roaring of the sea.

In Tam Lin (39) also we are concerned with the elves and fairies,
but the hero is himself human and a changeling. The ballad is
uniquely Scottish, and would take Its place in the first half-dozen
in Europe. Everything conspires to hold our interest: the love
of a mortal for a supernatural being, and especially for one of
the damned, cannot fail to move our hearts; the fatalism and
courage of the girl who redeems her lover; the fantastic trans¬
formations of the captured Tam Lin and the fruitless anger of
the Queen o’ Fairies are expressed in words uniformly direct and
poignant.

A number of ballads are concerned with the making and undoing
of wicked enchantments: Kemp Owyne, Alison Gross , The laily
Worm (34-6), Broomfield Hill (43), Willie's Lady (6), The Mother's
Malison (216), Burd Isabel and Earl Patrick (237), The Knight's
Ghost (265), and The Earl of Mar's Daughter (270). This group Is
almost entirely Scottish. Its first member has Scandinavian con¬
nexions, and The Mother's Malison makes use of a motif employed
also in Piedmont and Rumania. Our ballads have no share in the
merman cycle common to Scandinavia and North Germany, but
one encounters the Shetland superstition that seals become human
and have human children ( The great Silkie of Side Skerry , 1x3).
It can hardly be a medieval ballad, since the deserted lady
is to marry a 1 proud gunner’, who will kill the silkie with his
first shot.

Ballads concerning revenants are found in both countries. Sweet
William's Ghost (77) belongs to the Lenore tradition, and is a source
of Burger’s poem. It would be hard to fix the focus of radiation for
this legend. The Suffolk Miracle (272) Is a version of the Greek
ballad of Constantine and Arete or The Dead Brother's Return ,
brought directly to England by sea during the eighteenth century.

238 NORDIC BALLADS

The twin legends of the lover’s and the brother’s return from the
grave have often been deemed to be one. To identify them appears
to me wrong, both because the motivation is different, and because
the lines of their diffusion indicate foci at opposite ends of Europe.
Excessive grief interferes with the repose of the dead in The Unquiet
Grave (78) and The Wife of Usher's Well (79). To this class of
ballads we have to assign also Proud Lady Margaret (47))
Margaret (74), the late James Harris, the demon lover (243)—more
impressive in the superscription than in the text—and Willie's
fatal Visit (255).

Child’s first three ballads and Captain Wedderburn (46) are
riddling ballads, belonging to the same tradition as the Eddie
Alvismdl and the Danish Ungen SvejdaL

The use of the supernatural is, thus, a characteristic mark of
English and Scottish balladry when contrasted with that of France,
and it is also a principal link of union with Denmark and Norway.
The bulk of the corpus, however, consists of those stock amatory
adventures and conspicuous crimes which fill most European ballad
books. The themes come from all sides; they are almost as often
southern as northern. The Clerk's twa Sons of Oxenford (72) is the
French Scholars of Ponthieu, and The gay Goshawk (96) is the
French King Louis' Daughter, which seems to have Italian ele¬
ments. The twa Magicians (44), concerning the pursuit of a loved
one by a lover who will persist however she transform herself,
reproduces a widespread theme which is ultimately connected
with the Greek myth of Proteus; another use, and a finer one, is
made of transformations in Tam Lin . Reedisdale and Wise William
and The twa Knights (246, 268) are independent Scottish versions
of Marianson's Rings, which has its centre probably in Italy.
Curiously enough the march of the story in The twa Knights
resembles more intimately the Greek Maurianos than any of the
intervening versions, though there is no reason in this case to infer
a direct contact between a Scottish and a Greek ballad-monger.
The baffled Knight (112) is English, and belongs to a tradition begun
by France; it happens also to correspond with a Danish ballad.
On the other hand, a Scandinavian origin must be postulated for
Fair Annie (62) and Lord Thomas and fair Annet (73), Earl Brand
or the Douglas Tragedy (7), The twa Sisters (10), and a considerable
number more. A notable feature is the absence of German or
Dutch ballads among us. Arising in the fourteenth century, and

ENGLAND, SCOTLAND, AMERICA 239

still not verv numerous in the fifteenth, German ballads began to
expand across the frontiers in the sixteenth century, following the
south-eastern and north-eastern ways. But in the sixteenth century
England had become a land of broadside ballads, with many
original composers and a dying oral tradition. Scotland, separated
from Holland and Germany by the length of England, was m con¬
tact with France by its alliance and with Norway and Denmark by
its trade relations. Hence Hallewijn seems to have reached us, not
directly from the Low Countries, but from Scandinavia. We have
to wait until Scott translated Lenore , 2 he noble Moringer , Sent -
pack , The Fire-king , and The Erhking , and then the contact is not
of oral tradition.

There are cases in which, whether a tradition be indigenous or
not, it is connected with us by the superior vigour of our versions.
Lord Randal (12) is such a one, and has spread—words and music
together—to Italy, which borrows so little directly from Britain.
Edward (13) is even more distinctive. The theme of fratricide is
common in all balladries, and the particular treatment given it here
is found in the Scandinavian countries (as in the Swedish Sven 1
Rosengdrd ), and even in Finland ( Werinen Poika— one of the
modem type of imported ballads). The method of sharp question
and evasive answer is also commonly used: for an adulteress in the
Spanish Blancanina or a profligate girl in the Danish II itty Answer,
for a husband’s murderess in the Czech Murderous Wife , and an
infanticide in the Polish Marcisia, Many of these are fine ballads,
but none is full of such breathless anxiety, none so stark, as the
Scottish Edward. The questions come hotly, but the delaying
refrain allows for the embarrassment of the mother and the evasive
boy as in

‘Why dois your brand sae drap wi bluid,

Edward , Edward?

Why dois your brand sae drap wi bluid,
and why sae sad gang yee O?’

c O I hae killed my hauke sae guid,

Mither , wither,

O I hae killed my hauke sae guid,
and I had nae mair hot hee Od

A similar claim could be made for Child Waters (63), the cruel
husband, a ballad with closer analogues in Italy and Spain than in
Scandinavia. It is both English and Scottish, as is The three

240 NORDIC BALLADS

Ravens or Twa Corbies (26), which has travelled as far as Russia in
recent times. The number of ballads of really first-rate quality in
Child's collection is high. One may cite 'honoris causa' The Flower
o' Northumberland , Clerk Saunders , Leesom Brand, The Lass 0’
Rock Royal , Young Beichan , Little Musgrave, Young Hunting, and
Babylon. It is not only that the story is well handled, but individual
verses are often of surprising power and beauty, proving how
great a reserve of true poetical skill was available for the composi¬
tion of our ballads. By contrast with the pedestrian style of
such recent pieces as The Lord of Lorn, The Suffolk Miracle, The
famous Flower of Serving-men, and James Harris, we form some
estimate of the harm done to style by the broadcasting of ballad
journalese.

It is not necessary to go into the history of ballad journalism in
England, since it has no international significance. From the Tudor
ballad-mongers onwards the broadside ballad was the most direct
means of impressing the common man for some political purpose.
Their value for historians was recognized by Selden, and it was
doubtless of them that one wrote that he cared not who made a
people’s laws, if he made their songs. Superseded by the news-
sheet in due course, the broadside ballad did not wholly lose its
influence until the spread of compulsory education deprived it of
its public. It was still possible for boys to stop Macaulay in the
street as he was reading a paper, and to ask him to recite the
supposed ballad to them. Though the whole public is now a
reading public, the demand for oral entertainment still exists, and
is satisfied by the lamentable products of the music-hall and of
crooners.

English settlers in America took with them the debris of the
national ballads. The versions encountered in the Appalachians
and elsewhere are of definitely English ancestry, and in no case do
they contain variants superior to those of Scottish tradition. In
general they show a certain deterioration of taste, due to the cor¬
rupting influence of broadsides; yet the damage is not so very con¬
siderable. Many texts are old, reaching back to the seventeenth
century. The use of melodies in the old modes, employing the
pentatonic and hexatonic scales, is a guarantee of authenticity.
In general, it appears that the melodies have been well preserved
or even improved in quality. The collection of Appalachian ballads
by Campbell and Sharp contains no less than thirty-five which are

ENGLAND, SCOTLAND, AMERICA 241

in Child’s collection, together with eighteen unknown to him.
Ballads have been encountered in more than half of the United
States and in the English provinces of Eastern Canada; in Quebec
there are traditional French ballads, Scandinavian communities in
North America have their "viseri, and the Spanish and Portuguese
elements of South and Central America their e romances’. The love
songs’ (as they are called by those who sing them) of English
origin stand as typical of all the balladry of the New World*
since they alone have been extensively collected and intensively
studied.

Among those which survive from medieval tradition the greatest
popularity attaches to the sentimental Barbara Allan , the adulter¬
ous Little Musgrave, and the tragic Douglas Tragedy. Innumerable
versions of these pieces seem to be obtainable in all parts. Their
survival is part of a certain limitation of interest. The modem
singer prizes above all a love song, and the historical and super¬
natural ballads have receded from his ken. In addition to these
pieces there are many more modem ballads of British origin and
plebeian cast, such as The Butcher's Boy y Botany Bay , The Keys of
Heaven , The Waggoner's Lad , of sentimental or horrific Import.
Such ballads seme as the models for new creations on the American
continent, which are of interest as springing from a kind of neo-
medieval society. Some of them convey news in the broadside
fashion. So Springfield Mountain , which appears to be of the eigh¬
teenth century, tells how John and Molly Bland died, he of snake¬
bite, she of poison in the attempt to suck the wound; Frankie and
Albert is a "crime passioneF in the pedestrian style:

When Frankie shot Albert, he fell down on his knees,
looked up at her and said, ‘Oh! Frankie, please,
don’t shoot me no mo’, don’t shoot me no mo’.

"Oh, turn me over doctor; turn me over slow,

turn me over on my right side, ’cause the bullet am hurtin’ me so.

I was her man, but I done her wrong.’

It is a strange relative of Donna Lombardal

The highlanders of Virginia and the Carolinas are a remote and
self-contained folk not unlike the ballad "people’ of the Middle
Ages. For another reason the American negroes form a modem
ballad community. Living among the whites, they are cut off by
social and racial distinctions which force the coloured folk to keep
to themselves. They are endowed with a keen musical sense and
4615 I i

24Z NORDIC BALLADS

a unique gift of improvisation; but they are not suited for a more
reflective art. Their gifts are displayed, apart from the famous
‘spirituals’, in numerous ballads, peculiarly fluid in style. What
constitutes any one ballad is a certain coherence of inner structure,
but not any contexture of words. Almost any verse can appear in
almost any ballad; improvisation and tricks of memory deform the
tales almost out of recognition, and throw them together in strange
confusion. John Henry, a gigantic steel-driver, and John Hardy,
a murderer, have been confused in this manner. Rude and plebeian
as they are, there is some suggestion of the old economy of effect
in negro ballads, as in Stagolee:

Stagolee was a bully man, an’ ev’ybody knowed,
when dey seed Stagolee cornin’, to give Stagolee de road.

Stagolee started out, he gave his wife his han’:

‘Good-bye, darlin’, I’m gain’ to kill a man.’

The cowboys on the western ranges had also their peculiar
interests, which have found expression in ballads. The type has
disappeared with the annihilation of the herds of bison that once
roamed the plains, but while the horsemen continued they had to
find such recreation as they could in singing beside their camp¬
fires or while riding. The Old Chisholm Trail is a catalogue of cow¬
boy woes, not unlike the Blighty of 1918, and ending in the same
hope of release:

Goin’ back to town to draw my money,
goin’ back home to see my honey.

With my knees in the saddle and my seat in the sky,

I’ll quit punching cows in the sweet by and by.

Coma ti yi youpy, youpy ya, youpy ya,
coma ti yi youpy, youpy ya.

The occupation was seasonal, as was that of the lumberjacks, who
had their own heroes Paul Bunyan and Young Munroe, but no
ballads of notable merit.

4. Germany and the Low Countries

The supreme merit of the German ballad collectors has been to
realize to the full the importance of ballad melodies. It is not that
they anticipated other scholars in this respect. On the contrary, the
collections made between the dates of Herder and Uhland—the
collections which, in fact, have exerted a powerful influence on the

GERMANY AND THE LOW COUNTRIES 243
modem German lyric—were of texts alone. Ludwig Erk and Franz
Bohme, however, amassed a vast number of tunes from popular
traditional singing and from old books. So numerous are theyand
so securely anchored to specified times and places, that it is
possible only in the Deutscher Liederhort to embark on com¬
parative studies oi tune variations comparable to those of the texts.
These authors themselves assembled the melodies, without more
exactly studying them. John Meier, in a work still in progress,
restricting his view to the more famous narrative ballads, conducts
a systematic examination of the tunes. He is able to draw on the
less regularly collected material of other countries. Tunes to be
compared are written in staff notation, the one immediately above
the other, so that one can follow at a glance, with more or less
accuracy, the editor’s comments . 1

To the musician there is little to be gained by separating ballads
from other forms of traditional singing; indeed, the melodies of
popular lyrics are often superior to the less venturesome narrative
airs. Erk and Bohme’s collection, when compared with those of
Child and Grundtvig, is markedly lyrical; only one half of their
materials are relevant to the present study. The connotation of
the words 4 Lied 5 and 4 A olkslied’ is very vague. One may distin¬
guish more precisely ‘Volkslied’, ‘Ballade’, and ‘ Romanze as lyric,
lyrical narrative, and narrative; but the distinction is liable to break
down in practice. Firstly, the German ballad is stanzaic ; it has a
lyrical origin, and there are lyrical elements present at all times.
It is true that the narrative element is marked in Lindenschmid ,
Agnes Bernatier , Stortebeker , and other masterpieces of the fifteenth
century. But this tradition also tends to become more and more
lyrical In the ballads of landsknechts and reiters in the sixteenth
century, the narrative is used to suggest an emotion. A fine piece
like Franz von Sickingen exists not merely to describe the hero s
fall, but also as an outlet for Protestant grief. In modem military
ballads, such as Zu Strassburg auf der Schanz and Sedan there is
little more narrative than is required to describe a pathetic situa¬
tion ; the rest of the ballad is an appeal to a commonly felt emotion.
The German corpus is therefore both formally and intrinsically
lyrical, and it has exerted this kind of influence upon its imitators.
The neighbouring lands have lyrics in abundance or lyrical narra¬
tives, but few pure narratives. The relative weakness of the narrative
1 See Note I, p. 388.

244 NORDIC BALLADS

element makes German balladry so much the less corporate.
Verse forms are very different from one to another, and connecting
threads of interest are relatively slender. The richness of German
folk-song is undoubted, but it is dispersed, just as the nation itself
was dispersed into many principalities. It escapes, perhaps, some
of the monotony of the Spanish 'romances’, Danish ‘viser’, and
Serbian 'junacke pesme’, but it can hardly have exerted the same
massed force. Though no medieval people enjoyed the advantage
of such collections as we now have, those who expected the
repetition of a certain type of narrative in a certain type of verse
must have formed some collective idea of the whole. As a whole,
the ballads, especially historical ones and others believed to be
true, must have moulded the Danish, Spanish, and Serbian minds
to a uniform pattern; the dispersed German ‘Volkslieder’ corre¬
spond to a dispersed nationality.

By way of compensation, German balladry (which extends far
beyond the political frontiers of Germany) displays a sturdy
regional variety. The songs of Swiss pikemen from Sempach
(1386) to Marignano (1515) built up a professional tradition, which
was taken over by their professional rivals—the landsknechts and
reiters—once the victory of Pavia (1525) had showed the superiority
of the new methods. These landsknechts and reiters formed the
armies of the Protestant princes, so that the stirring ballads of the
Thirty Years’ War and its predecessors belong to the same style
and model. The landsknechts added, however, ballads on the
private aspects of their professional life: the devil-may-care free¬
dom of their expeditions, carousing in taverns, debt and beggary,
robbery and the rope. These are ballads of recurrent situations and
emotions, comparable with the domestic ballads of civilians during
the same epoch. They pass readily into the camp songs of the
modern conscript armies.

Austria and, in general, Upper Germany, was subjected to an
indigenous literary influence which was not immediately sensible
elsewhere. It is from the South, therefore, that ballads come
coloured by the personalities and styles of Minnesingers and
Mastersingers. Tonnhduser and Moringer are legends which have
congealed upon the fame of Minnesingers; the Count of Rome is a
product of the Mastersingers. Traces of literary convention, how¬
ever, extend in the ballads beyond those of definitely literary origin.
In modern times the distinctive Bavarian dialect occurs in some

GERMANY AND THE LOW COUNTRIES 245

soldiers 1 ballads to express humorous contempt. So we have the
Bavarian Schnadahufl of 1870, which begins

’Twas Bismarck spun the fabric,

’twas Moltke cut it square,
and for those poor old Frenchies ’twas
a miserable affair.

More obvious is the satiric intention in another:

Napoleons One and
Two are under sod,

No. Three is in clink,

No. Four—help him God!

Plattdeutsch had been used, with the same satiric intention, in
ballads of the War of Liberation.

In Flanders and the Low Countries one marks the proximity of
France. There is a more witty and cynical turn to the narratives,
more criticism of the lower religious, and the use of typically
French situations. A similar attitude seems to characterize the
Rhinelands. Cologne is a capital towards which Flemish eyes turn.
In other pieces there is some special affinity with Hanover and
Brunswick. The ballads of the Low Countries are, in fact, an
inseparable part of the balladry of Low Germany. The majority
of pieces are held in common, and it would be extremely difficult
to say where any one arises. One notes, however, the strong civic
sense of the Flemings, which sometimes finds expression in their
songs. The sea-robbers of the North Sea and Baltic give rise to
some fine ballads which are North German ; so are the ballads of
mermen. North Germany and the Low Countries were equally
affected by the tide of High German ballads, when this began to
flow strongly northwards in the sixteenth century, and it is across
this area that tunes and words reached Denmark and Sweden. The
mingling of Low and High German ballads is more complete than
those of England and Scotland. Separation would be utterly im¬
possible, nor Is it feasible to indicate any general distinctions.
Throughout the whole area there is no suggestion of the danced
ballad, apart from an example from Dittmarschen, close up against
the Danish border.

Political verse, in Latin and German, flourished for some cen¬
turies before the rise of the true ballad, and examples are given by
most anthologists. At what moment we can identify oral traditional
narratives it is hard to determine. The authentic note is not heard

2 4 6 NORDIC BALLADS

in the pieces, reproduced by Liliencron, on the alliance of Bern and
Freiburg in 1243, the defeat of Ottokar II in 1278, or the battle of
ollheim in 1298. The second of these begins in a spirited manner:

The banners and the host

began to take up post,

beat drums and trumpets blare,

troops, moving everywhere,
advance unwavering.

On the field a song men sing
‘In God’s name now march we’.

Many wept—credit me—
their sins and sinful life,
their children and their wife.

It is spirited and semi-popular, but it is not a ballad. It is not ever
m the form of a ballad, which is, in Germany, stanzaic. Th<
necessary qualities are present in the few lines concerning Lippolc
of Homboken (13 n), which come from Low Germany, and in the
song of the Kerels (1328-9):

Now shall we of the Kerels sing,
folk of a wretched sort,
the lords they’ll in subjection bring!—
a long, long beard they sport.

Their shoddy clothes are rent and tom,
upon their heads a hood to match,
askew their hats and caps are worn,
hosen and shoes show many a patch.

00 curas and whey and bread and cheese
they eat and eat and eat all day;
and so the churl gets in a daze,
who eats and eats more than he may.

The great series of Swiss victories in the fourteenth and fifteenth
centuries gave rise to a series of stirring songs which had genuine
popular acceptance, as they celebrated Sempach (1386), Nafels
Birs, Granson, Murten, Nancy, and the Swabian War of moc-s’
They built up a sense of nationhood, of the invincibility of the pike-
men, and of professional soldiering. The Battle of Sempach (Erk-
Bohme 231) is one of the earliest of German ballads. It originated
m a little drogue in couplets (Spruch) composed, apparently,
while the war still lasted. The Austrian invader threatens to hang
tiie insurgents, and the mavnr nf S Am tao ^ *j_1 _ t •

GERMANY AND THE LOW COUNTRIES 247

Lucerne, Schweiz, Uri, and Unterwalden. The Steer of Uri Is
mentioned in this ‘Spruch’. Hard on its heels came a ballad in
which the speakers are the Lion (for Leopold) and the Steer, and
the poet took liberties with the event. He scornfully supposes
that the lowlanders have come to confess to a Swiss pastor:

‘Pray tell us where’s the parson,
that we our sins may tell?’

‘He sits at home in Schweiz, sirs,
and hell confess you well,
good penance give to you;
from cutting halbards’ edges
you 11 get his blessing too!’

In the last elaboration, the ballad becomes long and circumstantial,
setting forth the legend of Winkelried’s brave feat of gathering the
enemy spears Into his own bosom. This last phase is due to
Halbsuter of Lucerne, composing In or about 1480. The legend of
William Tell is international in substance, since it is found inde¬
pendently as far away as Norway; but it is now inseparable from
Switzerland (Erk-Bohme 32).

Religious disturbances in Germany gave rise to a number of
poems, which are scarcely to be reckoned among ballads. Extrava¬
gant penances came as an emotional reaction from the terrors of the
Black Plague, and troops of Flagellants roamed the land, performing
their rites of self-torture. Their songs were, no doubt, popular and
traditional In kind, but were primarily devout lyrics, lacking any
developed narrative. The Schism of the Church, somewhat later,
perplexed the consciences of Christians, and produced political
verse designed for the widest possible circulation, but never to be
dissociated from the name, or at least standpoint and personality,
of the author. Of this nature are, for instance, Thomas Prischuch
of Augsburg’s 1,860 couplets on The Council's Foundation , and the
100 couplets of Johann Engelmar addressed to the Council of
Constance. The Hussite movement, occurring at this time, has
produced a Czech lyric, but no ballad.

The ballad style found inspiration In one of its perennial themes:
the doings of celebrated malefactors. In the south these were
robber barons, of whom Gotz von Berlichingen has become, under
Goethe’s moulding hands, the eternal type; in the north there were
also sea-rovers. The moment of inspiration is the moment of their
fall. Later ballad-mongers, with decadent instincts, work a trite

2 4 ^ NORDIC BALLADS

moral into such pieces; but the earlier ones are more objective.
They relate some incidents of wild audacity together with the final
scenes of arrest and execution, so that the general impression is the
more gallant. Fine examples are Epple von Gailingen , whose head
was put between his legs’ in 1381, Lindenschmid (c. 1490),
Schuttensam (1466), and Fritsche Grad (1430) (Erk-Bohme 230’
246-7, 242). Thijsken van den Schilde (Fallersleben 23, Erk-
Bohme 249) is a ballad of the same sort, from the Low Countries,
but with rather less valiant tone. The robber-baron was a concept
so purely German that there Is nothing quite similar in other
balladries. The Lusatian Handrias and Rajsenberk (Haupt and
Schmaler, i, 14) is evidence of the almost complete Germanism of
these Slavic ballads. The tune of Lindenschmid was new, and served
to shape the stanzas of many new ballads.

In the fourteenth century the state of Denmark was much per¬
plexed. Valdemar IV had launched an attack on the Hanseatic
towns and made the seas unsafe, and the succession of his daughter
Margaret was disputed. The uncertainty of trade favoured the rise
of piracy, especially when Margaret saw in the freebooters a pos¬
sible source of strength against the Hansa. The proclamation of
peace in 1383 let loose a flood of marauders upon the commerce of
Hamburg, Liibeck, and other trading towns; and in 1395 a re¬
conciliation between Sweden and Denmark set adrift the marauders
who had made Wismar and Rostock their lairs. Three bands
were organized for the systematic pillage of the Russian coasts,
Friesland, and the Spanish sea. The leaders were Godeke Michel,
Wichmann, Wigbold, and Claus Stortebecker. The expedition
organized from Hamburg against the last of these in 1402, and its
success, were recorded in a most vigorous Low German ballad,
Stortebecker (Erk-Bohme 233):

Stortebecker and Michael Godecke,
they were two robbers equally,
by water and not by land, sirs,
till God was angered grievously,
and brought them bitter shame, sirs.

They took them to a paynim lord,
and with the heathen traffic hold,
his daughter will they marry,
they rend and scream, like bears so bold,
and Hamburg’s beer drink gladly.

GERMANY AND THE LOW COUNTRIES 249

Spake Stortebecker on his own:

‘The Western Sea to me’s well known,
some beer from thence FLI cany,
and Hamburg’s burghers, money-blown,
will pay our merry party. 5

Yet another topic for historical ballads in Germany was found in
the endless private and public feuds between nobles and cities.
Busso von Alvenskben of Erxleben (Arnold 20) records a raid on
Altmarkt in 1372; the narrative is firm and vigorous, and shows
knowledge of the scene of action. Sometimes small groups
of such ballads cohere into little cycles, such as those which
describe the feud between Soest and the xArchbishop of Cologne in.
1446-7. Better than these, however, are some isolated pieces, such
as the Flemish Jakob van Artevelde:

It fell upon a Monday,
on a Monday afternoon,
folk came together stormiiy
Sir James to set upon.

From suburbs came the troopers,
peace to the folk to bring,
but traitors they rushed inwards
and threw down everything.

Into the house they hurried,
the ground was red with blood:

Sir James he died for Flanders,
his soul went up to God.

The pearl of medieval German ballads, however, is The fair
Bernatierin (Erk-Bohme 92). Agnes Bernauerin, of humble origin,
gained the love of Albrecht of Bavaria, Duke Ernst’s son, and con¬
sented to marry him after a seemly resistance. The Duke, like
Afonso IV of Portugal in a similar case, took the way of murder.
In 1435 he had the unhappy woman thrown into the Danube. She
might have saved herself by swimming had not men forced her
under with long poles; the ballad heightens the pathos by recording
her appeal to the Blessed Virgin, and by causing her to die at her
own request, refusing the way of safety which was open to criminal
women willing to marry the hangman.The poem was not composed
before Duke Albrecht had instituted an eternal mass for her soul in
1437 (renewed in 1447). The kidnapping of the princes of Saxony
461s K k

250 NORDIC BALLADS

in 1455 is cause for the half-allegorical ballad of Kunz von
Kaufungen (Erk-Bohme 239).

In the sixteenth century the German historical ballad reaches its
apogee. The century opened with the rivalry of the landsknechts
and pikemen, expressed in the scornful Brother Veit and Brother
Heini (Erk-Bohme 261), a propos of the battle ofMarignano (1516).
Brother Veit is the landsknecht; Heini appears in another piece
beginning

How now, you knavish Switzers,
you Heinis, oh so bold!

Yet another sings

Hear how the Switzer yokels
landsknechts have vilified.

This rivalry reached its height in the battle of Pavia, where the
German landsknechts and Spanish infantry broke the once invin¬
cible pikemen. A new 4 tune of Milan' arose—apparently a variant
of the old Hildebrand melody—and was used for a five-line poem
on Pavia; hence it is frequently known as the ‘Pavia tune'. There
was also an eight-line Pavia melody composed and first sung by
Hans von Wurzburg. The battle developed in a zoological garden.
The artillery breached the wall, and the Swiss fled before Georg
von Frundsberg and his landsknechts. A lyrical piece ascribes the
whole victory to him:

Frundsberg’s Sir George,

Frundsberg’s Sir George,
he has Pavia’s battle won,

Pavia’s battle won in a park,

in eight hours and a half won land and folk.

The King of France,

The King of France,
he has Pavia’s battle lost,

Pavia’s battle lost in a park,

in eight hours and a half lost land and folk.

(Arnold 32.)

Apart from their victories the landsknechts had to sing the varied
misfortunes of their way of life.These are historical by their circum¬
stances, though they do not rise normally out of particular events.
The narrative is important chiefly as leading to the expression of
some emotion: delight in a swashbuckling career, freedom from

GERMANY AND THE LOW COUNTRIES 251

restraint, contempt for labouring men, the hardships of the service,
the profits of summer and miseries of winter:

The wintry rime and chilly snow
do our poor riders grievous woe,
what easement have we got ?

How may we then our reckoning close,
when ride the roads we cannot, ay cannot ?

In summer time the sheep we drive,
and comely maids do troop behind,
and on my grey I’m swinging,
in green shaw up and down we ride
and hear the birdies singing, ay singing.

(Arnold 65, 2.)

Landsknechts and reiters waged war on the settled community.
They drank up their savings and were robbed by the taverners.
"When crippled they had to beg, and when starving they turned
highwaymen. The latter end of the discharged landsknecht was
the gallows, and under their shadow he was already dubbed < black-
neck’. All these things are expressed In dare-devil songs with some
admixture of narrative.

In addition to the war with France there was the war with the
Turk. The colossal disaster of Mohacz (1526) has found in The
Queen of Hungary (Erk-Bohme 276) pathetic expression. It was set
to a love-tune: a lament by the young queen for the youth who is
more lover than king. This dramatic treatment makes the piece
superior to straightforward narratives, which are also encountered.
The victory of Sankt Gotthard (1664) and the relief of Vienna by
Sobieski (1683) are other landmarks of Turkish frontier warfare.
The series is closed with the famous Prince Eugene (Erk-Bohme
324), celebrating his triumph at Belgrade in 1717.

The Reformation gave to Germany one language—the language
of Luther’s Bible—and a magnificent body of religious lyrics.
Though their texts are not liable to alteration by traditional
performances, they are Indubitably popular. Fine ballads arose
from the same movement. Ulrich von Hutten (Arnold 29) is a grave
monologue serving to voice the same passion as may be en¬
countered in such political lyrics as Praise God , ye pious Christians
(Arnold 27). The movement soon took a military turn, once the
reformers were compelled to defend themselves against persecu¬
tion; the soldiers available for its defence were the landsknechts

252 NORDIC BALLADS

and reiters who had already their own professional balladry.
Sickingen (Erk-Bohme 266) is thus a ballad of the reformers and of
the landsknechts simultaneously, and is, like many of its kind,
signed by the author (that is, by his classification as a landsknecht),
and given for something new:

He who for us did newly frame
this song, a landsknecht is by name,
and well indeed he’s singing :
the news—and well he knows the same—
from Landstal is he bringing, ay bringing.

The date is 1523; it is the personal aspect of Franz von Sickingen’s
fall, the death of a landsknecht chief, that the poet sings. Luther’s
death in 1546 coincided with a sharp ebb in the Protestant fortunes,
which is expressed in The Saxon Maiden of 1548 (Arnold 34):

O God our Father in Christ Jesu,
the orphan’s only Father Thou!

From my heart’s ground I Thee beseech,
and cry aloud in this my speech.

At first the speaker is anonymous. Later she declares herself to be
the Saxon Maiden, terrified for her honour before the Spanish
soldiery, and ready to deem him her true love who will free her
from shame :

And whosoe’er best aideth then
to me shall be the best of men,
e’en be he young or be he gray
or poor or twisted as he may.

Such man is one of real worth,
his shall be praise in all the earth;
a garland put I on his hair,
entwined by my hands so fair.

The tune used was a hymn tune: Oh God y uphold us by Thy
word .

The Saxon Maiden was, or came to be, the maiden fortress of
Magdeburg. Its heroic resistance in 1551 called for a song To the
tune of the city of Milan’ (Erk-Bohme 293):

O Magdeburg, now hold thee fast,
thou nobly builded house!

There comes full many an outland guest,
that wills to drive thee out.

2 53

GERMANY AND THE LOW COUNTRIES
The city declares her integrity and confidence:

‘The Maiden City is my name,

well kent in every land,

I trust the heavenly Christ will aid
me with His mighty hand’.

This confidence, circulated on a flying-leaf in 1629 when Tilly was
pressing his siege, greatly intensified the city’s resistance. The
metaphor could be used of other towns also, though without the
paronomasia, as for Stralsund when besieged by Wallenstein in
1628. Fine as they are, they lack the specific magic of the Castilian
Abendmar , in which a political statement is transmuted into the
purest poetry.

The quantity of ballads does not diminish with the Thirty Years’
War, though the quality is lower. The storm broke amid lampoons
on Cardinal Esel, with jibes about Esel ‘donkey’. The farcical reign
of the ‘Winter King’ in 1618 gave cause for Imperialist laughter,
and the death of the first Protestant leader, Mansfeldt, was finely
lamented (Arnold i, p. 188). A quinquennium of Imperialist
triumphs lies behind the heroic Stralsund and the tragic lament for
Madgeburg : a bitter Consolation for the pitiful slaughter of many
thousand Christians in Magdeburg (1631). The stages of the Protes¬
tant counter-attack—Leipzig, Breitenfeld, and Liitzen—are duly
recorded in songs, together with the death of Wallenstein. Other
pieces express the lawlessness of the times, the abasement of the
coinage, the destitution of the soldiery and of the peasants. The
close of the war left the land exhausted but not at peace. The
Swedes terrified northern districts until the Great Elector crushed
them at Fehrbellin, on which there is a ballad. Oxenstjerna was a
bogy who would make children pray, and

The Swedes are here,
they’ve swept all clear,
the windows they’ve shaken,
the lead they’ve taken,
and bullets have got
and peasants have shot.

Yet one poet, writing of the peace of Saint Germain en Laye in
1679, had the courage to say:

All signs cannot deceive us,
some day brings happier lot.

254 NORDIC BALLADS

The happier lot came with the War of the Spanish Succession,
in which there arose a cycle of ballads, of Austrian formation, about
the heroic figure of Prince Eugene. Such poems were primarily of
military inspiration. Soldiers have cultivated and propagated the
‘volkslied’ throughout the eighteenth and nineteenth centuries, but
in moments of crisis the soldiers have been the effective 'people 5 .
The principal wars covered have been the Seven Years’ War, the
Napoleonic, and the Franco-Prussian of 1870. Some of those
dictated by the War of Liberation are exceptionally fine, especially
Ferdinand August’s

With man and horse and waggon,

so has our God them stricken!

Others, like Father Bliicher , are humorous, though intense. In that
period the line between written and oral verse is hard to trace,
since many poems of literary style, but popular passion, gained
universal currency. The German Rhine , The Watch on the Rhine,
and Deutschland ilber Alles are cases in point. There are Austrian
barrack-room ballads of real merit, such as The Tenth Jagers and
By Sarajevo on the Hill, and the transfer of ballads from one setting
to another is a frequent phenomenon. The pieces indicate a general
situation or emotion. A soldier’s last words are topical both at
Custozza and Sedan, though the latter battle imposes itself 'a
fortiori’. These are conscripts’ ballads. In the eighteenth century
soldiers were impressed rather than conscripted, and there was an
evil traffic in cannon-fodder among the lesser German princes.
The fears and hates engendered by this practice find outlet in
O Strasshurg, In Strassburg on the Rampart, and The pitiless
Captain. That manner has passed. The conscript ballads of the
Austrian armies exist not only in German, but also in Magyar and
Czech, and to some extent in Rumanian, since all these different
nationalities were subjected to the same discipline in the same
camps. They account for a remarkable uniformity in this section
of the modem balladries circumjacent to Austria.

A small group of ballads is based on the medieval German
national epos. For the ‘younger’ Hildebrand (Erk-Bohme 21, 22)
an older model might be claimed. The epic fragment of the same
name dates from the ninth century. Between ballad and epos there
are no verbal resemblances of note, and the younger poem is
altered in a fundamental way, a way first found in the lost saga of

GERMANY AND THE LOW COUNTRIES 255

Dietrich von Bent which was translated into Old Norse in the early
thirteenth century. The theme is the battle of two heroes, father
and son. The epic ending is tragic; that of the saga and ballad
paints a family reunion.The particulars are fuller in the ballad than
in the saga, so that the former would seem to be a later production.
The fame of the Hildebrand tune in the early sixteenth century is
proof that the ballad is medieval, dating perhaps from the thir¬
teenth or early fourteenth century. A Low German King
Ermanarids Death (Erk-Bohme 23) has been preserved by a flying-
leaf of the year 1560. It is a free composition based, no doubt,
on the lost saga or some derivative. The poet, caring little for
tradition, describes Dietrich’s journey to Breisach, his fight in the
palace against a vast retinue, and his killing of Ermanaric—a
narrative which will not fit into any of the Gothic legends, but
distantly resembles the episode of Osantrix in the saga.

The name and fame of Dietrich pervade the vulgar Heldenbuch .
There is there a certain Wolfdietrich, so called from having been
protected as a baby by a wolf. His adventures include intervention
on behalf of a lion against a dragon, imprisonment in a dragon’s
lair and escape, fights with giants and dragons, and the capture of
a supernatural wife. There is a German and Dutch ballad, The
Huntsman from Greece (Erk-Bohme 24), which deals with the
last matter. Wolfdietrich’s father was Hugdietrich, emperor of
Constantinople. His achievement was that he disguised himself as
a woman to win the princess of Salonica without her father’s
knowledge. The motif probably survives from the Danish epos.
In Hagbard and Signe it is used, both in the epic and the ballad, to
lead up to an impressive ‘denouement’. In Hugdietrich and the
German and Dutch ballads (Erk-Bohme 140) the motif has become
a commonplace, and is part of no momentous action. By way of
preface the Heldenbuch recounts the adventures of Otnit or Ortnit,
the Lombard king of Garda. By the help of his father, the dwarf
Alberich, he won for himself a heathen king’s daughter; he ex¬
tended his power at the expense of the Greeks, but perished in
battle with a dragon. His adventures have no ballad consequences
in Germany. 1

1 I have used A. von Keller’s edition of Das Deutsche Heldenbuch , Stuttgart,
Litterarischer Verein, 1867, which contains Otnit, Wolfdieterich (with Hug-
dieterich), Der Rosengarten su Worms, Der Heine Rosengarten oder der kleine Konig
Laurin,

256 NORDIC BALLADS

No trace of the Nibelungs 5 adventures is found in German or
Dutch balladry, but there are several pieces which recount the
characteristic incident of the other great poem, Kudrun (Erk-
Bohme 178-81). The heroine is a serving-maid at an Inn. The
whole group, therefore, belongs to the epoch when the ballad was
no longer knightly, and taverns had replaced palaces as the normal
scenes of adventure. The epic is of the thirteenth century;
the ballads, as they now survive, are of the sixteenth or seven¬
teenth. The landlady’s ill treatment of the heroine is Implicit,
and the contemptuous attitude of the man who encounters her
at the inn is also implicit. In the epic both are fully accounted
for. The man recognizes her, as in the epic, for his long-lost
sister; the ballads do not trouble us with the figure of her
lover Ortwin, for they simplify the story and prune away every
detail that can be spared. Recognition in one case is by a bath¬
tub which the girl possesses. It is an unlikely object, but a re¬
miniscence of the scene in the epic in which she is encountered
washing clothes on the sea-shore. A much older ballad, how¬
ever, is Die Meererin (Meier 4), which reproduces the epic
episode:

How soon afoot is the young sea-maid,
the beautiful, the young sea-maid !

In the morning early goes she out,
she goes to launder the fine cloth

To the wide, wide sea, to the deep sea.

She begins to wash, she washes clean.

At sea there swims a ship so small,
and sit therein two youthful lords:

‘Good morning, beautiful sea-maid,
thou beautiful, thou young sea-maid. 5

‘Now thank you, thank you, youthful lords;
of good mornings my store is small V

He from his finger took a ring:

‘Thou fair sea-maiden, have this thing ! 5

‘Nay, sir, I am no fair sea-maid,
for I am but the laundry-maid ! 5

Into their ship the maid they guide
and sail away over ocean wide.

257

GERMANY AND THE LOW COUNTRIES

‘Thou art indeed the fair sea-maid,
the beautiful, the young sea-maid F

She takes a kerchief in her hand,
and sails the sea, and comes to land.

When they were come unto that place,
they began to kiss and to embrace

and there they kissed her, that sea-maid,
that beautiful, that young sea-maid.

It is in this earlier and more authentic form that the ballad crossed
France and became the Don Bueso of the Spanish corpus. Meier s
third ballad, The Wooing, is also, in all probability, a highly
generalized version of an episode in Kudrun.

There are also reminiscences of French narratives of a romantic
or epic sort. The Flemish Flos and Blancflos (Erk-Bohme 81) had
less success as a ballad than had the romance in prose, but in
Brennenberg or Dev Bremberger (Erk-Bohme ioo) the Germans and
Netherlander found a satisfying equivalent for the French Castel¬
lan de Coney and the Provencal legend of Cabestanh. Probably a
modern invention is the Flemish Roland and Godehnde (Erk-
Bohme 91), composed to attract interest to the nuns of Godelinde’s
cloister at Ghent. The restlessness of a princess at night, causing
her to go to a hayloft to sleep with an ostler, is the motif of The
Nutmeg-tree (Erk-Bohme 141). Its author had been three times in
France. In Spain the motif is used in a Carolingian setting.

The influence of Minnesang and Meistersang on the ballads pro¬
duced some new legends which, though undoubtedly preserved by
oral tradition, have the perfections of written literature also. The
greatest of these are The noble Moringet (Erk-Bohme 28), and
Tannhduser (Erk-Bohme 17, 18). Each opens with the separation
of lovers at dawn, that is, with an ‘aubade’; each uses the name of
a Minnesinger for its hero. In the first, Moringer (Heinrich von
Morungen, wdio flourished round about 1200), on the morrow of
his wedding, leaves his home; he appoints a tryst of seven years
with his bride, but stays away longer than that. At last he returns
when she is about to be forced to accept a new husband, makes him¬
self known by a ring dropped in a cup, and displaces his rival. The
story is older than its hero, and is found in the same complex form
under other names (Duke Heinrich in Germany, Hind Horn in
England, Count Dirlos in Spain, &c.), having a tendency to attach

461 s l!

258 NORDIC BALLADS

itself to national heroes (Marko Kraljevic in Serbia, Dobrynja in
Russia). As The Husband's Return it appears destitute of most of
its details in French folk-song and in Germany (Erk-Bohme 49,
93, 110, 191). There are many possible 'denouements 5 : the soldier
may kill his rival or find the match already complete; if only a
suitor, he finds his fiancee has been fickle; there may be an accom¬
modation with this rival, such as the offer to him of his sister to wife,
or the lady may be considered guilty in a greater or less degree, and
the ballad end with a blood bath. But in the best and more de¬
veloped ballads there is a precise contexture of incidents, as des¬
cribed above, which it seems right to associate with The noble
Moringer itself. Tannhauser is a still finer ballad, in both High and
Low German and later in Danish. It is of the fifteenth century,
and has grown out of the songs by and attributed to the poet him¬
self. The hero’s heart is tom, like that of so many medieval men,
between beauty and salvation. The Church, represented by Pope
Urban IV, denies him salvation, and he turns back with passionate
despair to his cult of sinful beauty.

Henry the Lion (Erk-Bohme 26, 27) exists in the form of ballads
and mastersongs. It is a rambling romance which makes use of the
old Dietrich legend for the hero’s intervention to defend a lion
against a serpent, and of the Moringer tradition for his absence and
home-coming. Though unworthy, it has passed northward as far
as Denmark, and has enriched Czech legendary history with a king
Bruncvik. Duke Ernst (Erk-Bohme 25) is of great length, and
should rather be deemed a romance than a ballad. The Count of
Rome (Erk-Bohme 29) is undoubtedly a ballad, and as such has
spread to Denmark and Sweden as well as to Holland. Its source
is the mastersong of Alexander von Metz oder der Graf im Pfluge ;
and its characteristic motifs are the enslavement of the count, and
his wife’s heroic coming in disguise to release him. She comes
disguised as a minstrel, and asks no reward from the Kaiser but the
freedom of the wretched prisoner. These particulars, apart from
the yoking of the count to the plough, are found in the Russian
ballad of Stavr Godinovic , which is thus probably of German
origin.

Two fine ballads inspired by Classical legends are The two King's
Children and The Evening Walk (Erk-Bohme 83-5, 86-8). The
former is the legend of Hero and Leander, derived from Ovid’s
Heroides through a Middle High German poem. The ballad is

GERMANY AND THE LOW COUNTRIES 259

equally common in Low and High German; the normal version
seems to have been fashioned in the Netherlands. It has extended
into Scandinavia by way of Denmark, and as far as Piedmont and
Catalonia by way of France. The theme is popular in Lusatia,
Czechoslovakia, Poland, and Latvia. Though the ballad is fine,
finer still is the lyric which has sprung from it:

‘O Elsie, darling Elsie mine,
how glad were I by thee!

So are there two deep waters
that sunder thee from me/

"That brings me bitter sorrow,

dear comrade of my heart!

From all my heart declare I,

it is a bitter smart P

‘Hopei time will bring an end, dear,
hope! joy will yet be thine;

all things will change for better,
heart’s dearest Elsie mine!’

The Evening Walk is the legend of Pyramus and Thisbe with
acknowledgements to the conventions of the ‘aubade’. In its
medieval dress, with tower and watchman, and most of the classical
details gone, the ballad is not unlike that of Tristan’s Death (Meier
9). The Clerk in a Basket (Erk-Bohme 144, 145), cynical and anti¬
clerical, is the legend of Vergil, suspended, according to medieval
tradition, in a basket beneath the window of a lady whom he made
senile efforts to dishonour. A mastersong on this subject exists,
and Luther knew the ballad.

Ballads of Biblical origin are interesting chiefly when their
material has been so transformed that the source is no longer
apparent. The German dinger (Erk-Bohme 41, 42, 195) is one
of these. Older than the German is the Dutch version, Hallewijn
(Hoffmann von Fallersleben 9, 10), in which the resemblance of the
name to the Biblical Holofernes is more apparent. According to
this ballad a girl is beguiled from a castle by a knight who reveals
himself as the murderer of her sisters. He orders her to prepare to
die. There are various sequels to this order: in some she dies and
her brothers avenge her death, in others they arrive in time to save
her, and in others she gets a weapon by some trick and kills her
persecutor. This last is the authentic version, and it has, in the

26o NORDIC BALLADS

Dutch ballad, a sequel not found elsewhere. She encounters
Hallewijn’s old mother. The old woman asks for news of her
son, and the girl replies that he is dead and his head is on her
own lap:

‘Hallewyn your son is hunting gone,
no more you’ll see him your life long;

Your son Sir Hallewyn is dead,
here in my lap I bear his head,
with his heart’s blood my apron’s red.’

To her sire’s gate when that she came,
her horn she blew like any man.

Her father, when he heard the same,
joy filled him for that home she came.

Down sat they to a fair banquet,
the head was on the table set.

These are the lines which identify the legend with the story of
Judith and Holofemes. Without them, in the German versions for
instance, the ballad appears to belong to the saga of Bluebeard, but
for the fact that it contains no tabu. There is no closed room not
to be entered; the villain is inspired by mere insensate blood-lust.
It has spread outwards in all directions over the greater part of
Europe, though probably not itself older than the fifteenth century.
In France it has become Renard the Woman-slayer, and has de¬
scended thence to Italy and Spain (Rico Franco). In the Scandin¬
avian lands it encountered the indigenous superstition about nixes.
The hero became a malevolent spirit, a supernatural, and as such
he appears in the Scottish Lady Isabel and the Elf-Knight. This
superstitious increment is late, and there is no reason to include
the ballad among those of supernatural adventures. As for the
Bluebeard saga and the history of Gilles de Retz, the common
features are probably due to contamination with the then existing
ballad. Another Biblical ballad of interest is The chaste Serving-
man (Erk-Bohme 76), which is, under a knightly disguise, the
history of Joseph and Potiphar’s wife. Inverted so that it is the
man who tempts and falsely accuses, it is The slanderous Chaplain
(Erk-Bohme 55).

These histories, though Biblical, have been secularized. Others
have remained religious in intention, though they have drawn away
from the Biblical text. A lyrical fragment on Judas has survived

26 i

GERMANY AND THE LOW COUNTRIES
from the fifteenth century, at least, when it was proverbial for any
act of treachery:

Miserable Judas,
what is this thou’st done,
that thy Lord, thy Master,
thus hast thou foredone!

Therefore must thou suffer,
in hell's torment lie.

Lucifer’s companion
be for ever and aye.

Kyrieleison !

The presentation in the Temple, the Nativity, the Crucifixion, and
the Resurrection give cause for some beautiful pieces, which
approximate to carols, and some carols have become indubitably
traditional. The Heavenly Lime and the Winter Rose are exquisite:

There is a Rose has blossomed
out of so sweet a root,
as men of old have told us,
from Jesse came the shoot;
it grows a bud so bright
in midmost cold of winter,
just at the dead of night.

The rosebud that I mean, sirs,

is that Isaiah said,

To usward brought it only
Mary, the cleanest Maid:
by God’s eternal might
has she a Child begotten
just at the dead of night.

The saints are not forgotten. The songs in their honour are
sometimes narratives, sometimes appeals for aid. St. George, St.
Lawrence, St. Catherine, St. Kilian, &c., are patrons much invoked.
Some of the Marian legends have more originality and interest;
for example, St. Mary and the Shipman (Erk-Bohme 2064). The
ferryman attempted to violate his passenger, but as they approached
the midmost of the strait the bells began to ring in her honour :

They rang and rang, both great and small,

They rang and rang together all.

St. Mary kneeled upon a stone,
the shipman’s heart it sprang in two.

262 NORDIC BALLADS

Moral tales and pious legends also have some exemplars
worth quoting. Some are of souls rescued from condemnation at
heaven’s gate, such as The pardoned Dancer, Three Sisters at
Heaven's Gate , and The poor Soul The Sultan's Daughter (Erk-
Bohme 2127, 2128) turns a sentimental history to pious uses; and
a group of ballads exploits the matter of the old French fabliau of
Maquerel, the devil who rode a priest’s concubine to hell (Erk-
Bohme 218, 219). The best of the group is The Smith's Daughter
(Erk-Bohme 11) which isolates one tragic moment. It does not tell
us why the smith’s daughter should have been bewitched into a
mare. We only know that the smith was forced to shoe her, and
drew blood with the nails till she cried out who she was in her
anguish. It is a moment of unspeakable sorrow in the lives of two
sinners. Few ballads better illustrate the truth that often the half is
better than the whole.

German ballads of supernatural adventure are less numerous and
less eerie than those of Denmark, Norway, and Scotland, but they
include some notable examples. The two merman legends (Erk-
Bohme 1, 2) have been already discussed. Common to Denmark
and the German seaboard, they have been claimed as German, and
are in any case Germanic. In one version the location is given as
the Jade Estuary; it may well be original. In some the girl of the
second ballad is said to have been drowned in the Rhine, meaning
thereby the sea. The Riibezal (Erk-Bohme 3) is a hill sprite of
German nationality, and there are some entertaining ballads about
the humblest denizens of the other world 1 kobolds and poltergeists
(Erk-Bohme 4-6). The metamorphosis of damsels into trees and
flowers (Erk-Bohme 8-10) is a gracefully pathetic conceit. There
is a changeling (Erk-Bohme 12), and a Crowned Snake (Erk-
Bohme 13), who is held to her earthly lover by his possessing
her crown, just as Bulgarian samodivas are the servants of those
who keep their clothes. Tenore (Erk-Bohme 197—202) is the most
famous ballad of revenants, on account of its consequences, not its
merits. The Danish Aage and Else and Scottish Sweet William's
Ghost are superior by far. There are German equivalents for
the Orphan at her Mother's Grave, the child who cannot rest
in death because of its mother’s weeping, the lover’s talk with
his fiancee in her tomb, &c. The damp Shroud or Der Vorwirth
concerns a neglected dead husband who demands from his
wife a new shroud; it is the source of the Czech Rubas, possibly

GERMANY AND THE LOW COUNTRIES 263

the most impressive of Czech ballads. Sundry ballads concern
themselves with dreams, generally of an allegorical sort, and one
shows evildoers successfully cited before God’s judgement seat.
The pied Piper of Hamelin, The fiddling Hunchback, and The
Minstrel" $ Son (Erk-Bohme 14-16) are wonderful adventures told
to the credit of the profession. Erk connected the first of these
stories both with the character of Woden, as inferred from the
relics of Germanic mythology, and with the historical fact of the
twelfth-century 7 migration of Saxons to Siebenbiirgen. Tannhduser
is, of course, a ballad of the supernatural, with its use of the learned
superstitions about the Sibyl’s Paradise and its figure of Venus:

‘Frau Venus, noble lady sweet,
thou art a devil-woman/

The bulk of the German corpus is formed of ballads of adven¬
ture, the greater part of these being love-songs. The remainder
concern horrific crimes: parricide, infanticide, rape, incest, poison¬
ing. Many of these latter may refer to events which really occurred,
and indeed ballads generally offer that sort of assurance, even when
they are fictitious; but the circumstances are forgotten now, and
the appeal of the ballad is to the morbid-minded. Some of the
crimes belong to cycles of fiction. The guiltless prisoner is a figure
which the German singers have especially delighted to honour.
The whole group of adventure-ballads is dateless, and some may be
ancient. The general tone, however, is somewhat vulgar, especially
in the criminal pieces, and the favourite ‘mise-en-scene’ the tavern.
It corresponds to the decline of taste in the later sixteenth century,
when the ‘reiter’ took the place of the Titter’, the swashbuckler of
the knight. It is work of this quality which has overflowed into
the neighbouring lands, whose balladries rarely recall the purely
medieval aspects of the Volkslieder.

When love is the theme there is no absolute distinction between
song and ballad. The theme is immemorial. Already in the
twelfth century we have good work done, as in the untranslatably
simple Thou art mine, I am thine . 1 Its simple metaphor—the heart
that is a coffer safely locked and the key thrown away —is heard in

1 Du bist min, ih bin din:
des soltu gewis sin.

Du bist beslozzen
in minem herzen:
verloren ist daz siiizzelin:
du muost immer drinne sin.

264 NORDIC BALLADS

quatrains that live to-day in Switzerland, Alsace, Carinthia, and the
Tyrol. In such pieces we have the originals of Heidenroslein and so
many other of Germany's best artistic lyrics; we find also the secret
of Goethe and Heine’s music. A naive felicity breathes in such
lines as (being translated):

Shine on us, lovely sunshine,
give us your brightest ray,
shine on two loves together
who fain would meet to-day.

The songs of parting lovers are full of such felicities; desertion
and death fill folk-poetry with an emotion both tender and

deep.

I heard a sickle rustling,
a-rustling through the corn,

I heard a maiden crying,
of her true love forlorn.

'So let it rustle, loveling,

I care not how it go:
new mistress here I’ve won me
in this green sod below.’

'Hast thou a mistress won thee
in that green sod below,
then stand I here so lonely,
my heart is full of woe.’

This tenderness, sometimes degenerating into sentimentality, is
universal in these expressions of the joy of true love, of longing,
of wooing and winning, of opposition, separation, loss, and the
ecstasy of reunion. The German minstrel sees these chapters in the
immemorial romance simply. He has none of the austerity which
breaks through in Spanish ballads; still less has he the light cyni¬
cism of the French. The French, unfortunately for their poetry,
have not chosen to give due representation in their verses to the
claims of simple emotion. The popular singers of the Middle Ages
inherited a tradition from the troubadours, which turned love into
an intrigue irrelevant to marriage. The result is a great deal of
amusing and witty verse; one touches delicately, carelessly, upon
illicit love, furtive meetings, ludicrous contretemps, and the
cuckolding of husbands. There is a good deal of this in the corpus
of German ballads also, and it is easy to recognize the foreign

GERMANY AND THE LOW COUNTRIES 265

leaven. A heavy debt for entertainment has been contracted by
Germans in France. But the truly German manner is different.
It is simpler, more immediate; the sentiments are fewer but deeper,
and the lines are warmed by human kindliness. Neither is adultery
a model nor are husbands ridiculous; on the contrary, faithlessness
leads to sorrow, tragedy, and crime. Thus the German ballads
form a spectrum extending from the rosy idylls of the one extreme
to the blood-stained sordidness of the other.

The Ransom (Erk-Bohme 78) is a simple history, which is found
in Rumania, Czechoslovakia, Denmark, Sweden, the Faeroes,
Lusatia, Esthonia, and Finland. The German version is geo¬
graphically central, whether it be original or not. A young girl,
captured by a pirate, appeals to father, mother, sister, and brother,
but only her true-love is willing to pay the price for her. Her peril
is not always the same: in the Ukraine and Bulgaria she is repre¬
sented as swimming for her life in the Danube or another river.
In the Rumanian Girl and Cuckoo , she has lost her way; the cuckoo
will save her, not to become her honorary cousin or brother, but
only to be her lover. The ransomed Slave (Erk-Bohme 79) is quite
modem. It is a tale of love at first sight. There is something
pathetic and childish about the Swiss Dursli and Babeli (Erk-
Bohme 80), though Dursli is old enough to join a regiment. It is
thus a soldiers’ poem, despite its appealing simplicity. The
landsknecht was glad to represent himself as true in love, and so
much to be preferred to the rich townsman. That is the moral
of Two Companions (Erk-Bohme 70). It was a theme easily parodied
‘a lo divino’, and such parodies exist from the fifteenth century.

A number of ballads are no more than tableaux (Erk-Bohme
71-5). There is a girl and a clerk or ensign or shipman or soldier,
and that is all. Usually maidens are not unduly coy, though there
was one Incorruptible (Erk-Bohme 74); and consequently there
are many ballads of maidens betrayed (Erk-Bohme 112-35).
They make haste to elope with tipplers who pawn their clothes, or
with landsknechts, ferrymen, officers, fiddlers. These songs are
often connected with some particular rustic ploy, such as cutting
grass or com.

When the history is of bride-stealing or elopement, there is
often a more pronounced narrative element. There was The
Lady of Kerenstein (Erk-Bohme 34), for instance. She ran away
to her lover, with the connivance of the watchman; her father

4615

266 NORDIC BALLADS

consoled himself with the thought that it was by her own consent and
the watchman on the ramparts
sang clear the dawning-song—

which must have been an unexpected sign of hilarity, since he had
just been condemned to die. The conventions of the ‘aubade’
pervade such pieces. Song's Reward (Erk-Bohme 35), a Dutch
piece from the fifteenth century, even goes so far as to use a
‘senhal 5 , in agreement with Provencal traditions:

I see the ‘Star of Morning’,
my love’s clear countenance.

The lady did not consent in the tragic Palatine or the bloody
Marriage (Erk-Bohme 39). The Palatine killed her relatives one
by one and carried her off; he agreed not to molest her on the first
night, and she was dead of sorrow by the next morning. A certain
princess Margaret committed the imprudence of eloping with proud
Henry (or Syburg), thinking him a rich man (Erk-Bohme 40).
When she learned that he was poor

What took she from the scabbard ?

A golden sword so red;
and down she kneeled before him,
and struck, and fell down dead.

‘So if my father ask thee
why I have stayed behind,
then tell him I have perished
in a far foreign land.’

The false Bride (Erk-Bohme 211) is a commonplace and meagre
treatment of the theme of the Scottish Daemon Lover.

The aubades’ dealt conventionally with the partings of lovers,
and so are akin to the more popular parting songs (Abschiedslieder),
which occupy a large portion of the corpus. They have necessarily
some narrative elements, at least a note of place and indication of
the persons. More fully expanded, the parting song leads on to the
happy or unhappy return, and so to complete histories like The
noble Moringer. A simple and delightful version of this history is
contained in Unter der Linde (Erk-Bohme 67):

A linden stands in a deep vale,
above ’tis broad and under small.

Beneath, a loving pair is set,
rejoicing all their griefs forget.

267

GERMANY AND THE LOW COUNTRIES

‘Now, dearest love, we two must sunder,
for seven years have I to wander/

He, of course, overstays his leave, and returns with a tale of his own
fickleness. As the lady does nothing but wish him well, he at last
reveals himself as her true-love. These tests give the ballad its
alternative title (Liebesprobe ?), and, apart from the final declaration,
we have here the matter of the old French Belle Doette, The
German ballad has, however, an elusive, wistful charm of its own,
and its own power to expand into Poland and other lands. There
are other typical returns (Erk-Bohme 49, 93, 112, 191, 201),
according as the lady was dead or indifferent or remarried.

Then there are ballads of separation which have a more tragic
cast. Several very fine German and Dutch pieces tell the history
of the knight who was ambushed by his lady's friends, and how
she found his body and broke her heart over it (Erk-Bohme 94-6,
Todtenamt , &c.). Brunswick is given as the scene of the tragedy,
which may have been an actual event. Parental opposition might
go to any extreme. The most powerful was the mother's malison
(Erk-Bohme 192-4: Blaublumlein , &c.). More novelesque is the
sort of opposition seen in The cruel Brother (Erk-Bohme 186),
known also in Denmark, and The King of Milan (Erk-Bohme 97).
In both the lady has been imprudent, but that is no reason why
her kin should murder her. In one ballad the King of Milan
rescues her and all ends happily; in the other the King of England
arrives too late. He kills the brother,

and took the baby in his arm:

‘No mother have we, God keep thee from harm!'

A kitchen-boy was the lover in another case, and one does not know
who killed her murderer:

Now Kirstie was buried beneath the rose-tree,
the young Margrave they broke upon the wheel.

For Kirstie all the bells were rung,
the Margrave to the crows they flung.

Other deaths are accidental. Count Frederick's Wife (Erk-
Bohme 107) was accidentally pierced by his sword; her relatives
avenged her, but a sign from heaven justified the count. The
Miller's Daughter (Erk-Bohme 108) was caught in the mill-wheel,
and The young Margravine (Erk-Bohme 109) died in childbirth at

268 NORDIC BALLADS

the age of twelve. Three Roses (Erk-Bohme 203, 204) is a deserted
maiden’s cry of sorrow, and in the Low Country King's Daughter
(Erk-Bohme 99) poetical justice is rather heavily done by making
the faithless seducer return to beg bread at his lady’s door.

Cynical ballads of intrigue are quite numerous in the German
corpus (Erk-Bohme 127-56). They are almost all known also in
the Low Countries, and they betray a French inspiration, though
they are rather picaresque than witty. James I knew a Scottish
ballad of the Man in the Hay type (Erk-Bohme 150), and the
minnesinger Gottfried von Neifen is responsible for another two
(Erk-Bohme 130, 138). The ballads are sometimes made at the
expense of certain classes of persons supposed to be especially
amorous: millers and their wives, students, monks, and parsons
(Erk-Bohme 137, 152-6). On the other hand, nuns in love are
treated tenderly, whether fortunate or unhappy (Erk-Bohme 68,
69, 89, 90). The pieces have had a wide circulation, and in their
lyrical form (Nonnenklagen) they are especially affecting; for them
there were precedents in French. Few ballads also have such fine
airy beginnings as

I stood upon a hill-top,

I looked in a deep mere,

I saw a skiff a-swimming,
three young earls in it were.

The device of pretended death, used of a nun in a Swedish ballad,
is used in Pretended Death as Match-maker (Erk-Bohme 111); for
which there was a French precedent in the old Belle Isabiaus.
Stories of resurrected wives and fiancees (. Richmode von Adocht)
may have had a foundation in fact, like the Spanish Doha Angela .

The Lady of Weissenburg (Erk-Bohme 102—3) * s one of the most
fully developed of the ballads of passionate crimes. The event was
historical and recorded in chronicles. In 1065 Friedrich of Saxony
was murdered by the Landgrave of Thuringia for the sake of
the fair Adelheid, his wife. The ballads date probably from the
fifteenth century. They are hostile to the lady, whom they accuse
of laying the plot for her husband’s murder; in some versions she
seeks to reward the slayer, but he remorsefully throws away her
ring, in others he confesses and dies, and in others he dies return-
ing to her the thirty ducats of the betrayal. The ballad-mongers
of the Low Countries connect the event with the duchy of Luxem¬
burg. Hans Steutlinger, recorded in print in 1544, is a history

GERMANY AND THE LOW COUNTRIES 269

modified by reminiscences of the older ballad. Other sanguinary
adventures are recorded in Mutschelbeck and Knight and Squire
(Erk-Bohme 104, 105), the former possibly historical, the other a
rather pointless fiction. The most famous of adulterous criminals
was Rosamunda, but the version given by Erk and Bohme (106)
is a modern rendering from the Venetian of Donna Lotnbarda. The
poisoner-motif was associated not with intrigue but with family
hatred in the famous Schlangenkochin (Erk-Bohme 190), which is
a form of the Donna Lotnbarda tradition. The evil-doer is a step¬
mother, aunt, grandmother, or, in one version, a lover. In Degener
and Lussewine (Erk-Bohme 46) we have a black tale of revenge.
Lussewine lures Degener into a bedroom so that her brothers may
kill him for his murder of their father. The Infanticide (Erk-
Bohme 56), imitated by both Burger and Schiller, The Raven-
Mother (Erk-Bohme 212), The Wise Woman (Erk-Bohme 213), and
The hatef ul Deed (Erk-Bohme 66) are crimes that involve the per¬
version of the best instincts recognized by the ballads, namely, a
mother’s love for her child and brother’s for brother. A miscellany
of crimes (Erk-Bohme 48-54) includes murders for jealousy, and
others, like the Sold Miller’s Wife (Erk-Bohme 58), were attended
by special circumstances of horror. The criminal in this case was
broken on the wheel in 1596.

Tyrannous crimes are recorded in The Lord of Brunswick (Erk-
Bohme 64, 1886), which is concerned with the game-laws, and in
The unmerciful Sister and The unmerciful Youth (Erk-Bohme 209,
210). In the former a rich woman refuses her poor sister a loaf, but
at last gives her one with a stone baked into it; in the latter the
refusal of bread drives a poor widow to kill her children and
herself. The rapes and robberies committed by landsknechts are
also tyrannous crimes and give rise to many ballads; and there are
some of a vulgarly doctrinal sort which trace malefactions to bad
companions and bad upbringing.

Ballads concerning guiltless captives form a sort of cycle by
themselves. The best is The Castle in Austria (Erk-Bohme 61),
not for its contents, which are vaguely pathetic, but for its striking
first verse:

There lies a castle in Austria,

so well and truly founded,
with silver and the red, red gold
and marble walls surrounded.

270 NORDIC BALLADS

But that verse seems to have been a second thought. The verse
makes the ballad easily recognizable in Denmark and Sweden; and
as the ‘East-realm' might be Austria or the East from which our

religion comes, this ballad was promptly adapted to religious ends.
Peter Unverdorben (Erk-Bohme 60) is, perhaps, too explicit to
leave the same impression on the mind; and even The Lord of
Falkenstein (Erk-Bohme 62), though a noble tableau of wifely
importunity and devotion, is not equal to The Castle in Austria .
The Knight and Shepherd (Erk-Bohme 43) is fiction. Seized by a
robber baron, the shepherd is set free in return for his parents'
gold, his daughter's snood, or his acceptance of the baron’s
daughter in marriage.

A particularly delightful group is composed of ballads dealing
with birds, beasts, and trees. Some are simple allegories, others are
not so much parodies as etherealizations of human activities. Such
are the animal marriages (Erk-Bohme 163-5 : Vogelhochzeit , Kdfer -
hochzeit, Tierhochzeit ), and the amusing Hare's Complaint and Song
of two Hares (Erk-Bohme 167-70). An amusing skit is The Swabian
Round Table (Erk-Bohme 142), in which the heroes gather to attack
a hare, but are routed by a frog. It is based on a mastersong of the
sixteenth century. There are naive allegories and parables in The
Squirrel and his Wife , The Owl and Eagle , The Girl and the Hazel ,
and The Fir (Erk-Bohme 172-6). The latter has given a lyric of
exceptional beauty.

In the above catalogue I have not attempted to discriminate
between Low and High German work or to give special treatment
to ballads which may have originated in Holland and Flanders. It
was hardly practicable to do so. As the songs from these regions
were incorporated by Erk and Bohme in the Deutscher Liederhort
at appropriate places, it seemed better to mention them as they
occurred. Yet one must not conclude that the indebtedness was all
on one side or that Dutch balladry lacks its independent life. Such
famous pieces as Hallewijn and The two King’s Children probably
originated in the Netherlands, and the same may have been the
case with others. The repute of Dutch folk-poets may have been
unduly depressed in England by a remark of the late W. P. Ker’s.
As an example of the art of sinking in balladry he picked on the
lines

And then she lit three candles,
three candles at twelve to the pound.

GERMANY AND THE LOW COUNTRIES 271

That is better thrift than poetry, and there is a good deal of it. But
there is also much bathos In English balladry, as Matthew Arnold
knew too well. xAs against these we must place many ballads fairly
conducted in Dutch, and some fine beginnings like
The day it springs from eastward,
it lightens over all,

and possibly, though with a German hint, the spendthrift hyper¬
bole:

If every mountain was of gold
and all the waters wine,
yet far more full of joy I’d be,
fair maiden, wert thou mine.

Dutch ballads sometimes represent traditions which have been
submerged in Germany as High German has overflowed the Low
German area. Those of Flanders contain original traditions, as in
Genoveva (Erk-Bohme 82) and Roland and Godehnde . They also
show French influence more clearly than in Germany. The French
Vivandiere’ type of ballad appears to be well represented in The
Captain's Daughter (Hoffmann von Fallersleben 41). It has the
right amoral touch. The girl wants to join his company, but he will
not allow her. She disguises herself, and when he proceeds to make
love to her she is able to threaten him with pains and penalties at
home; and so she joins the troop:

‘O maiden, pretty maiden,
an you with me will go,
then all the clothes you carry
with thread of silk Ill sew.’

£ 0 captain, great commander,
such thing can never be !

Bethink you of your wedded wife,
how angry she would bed

4 Now would indeed my wife at home
here at my feet lay dead,
and you and I, my pretty maid,
were plucking roses red V

And then he gave her wine to drink,
cool wine out of a glass,

’twas then he first began to think
his daughter that she was !

272 NORDIC BALLADS

The Low Countries have, naturally, their own tradition in his¬
torical balladry, though they have given easy acceptance to those
of German origin. That the Flemish series is older than the
German we have already seen to be probable. The oldest-known
subject of a Dutch folk-song is the murder of Count Floris V by
Gerhard van Velsen in 1296 (Hoffmann von Fallersleben 3). At
the beginning of the thirteenth century the civic disturbances in
Flanders produced some popular political verse scarcely distin¬
guishable from balladry, and in the fifteenth century Thijsken van
den Schilde and Claus Molinaer (Hoffmann von Fallersleben 23,
Erk-Bohme 249, 250) were robber barons whom the Dutch, unlike
the Germans, were unwilling to idealize. Holland had her own
songs of the Reformation, and her national anthem:

Willelmus van Nassouwe
am I, of German blood,
to the fatherland I am faithful
and true until I am dead.

A Prince of Orange, I,
so free and unafeared,
the great King of Hispania
at all times I revered.

Nothing could more clearly indicate the conflicting loyalties of the
Dutch Revolt. The tune was more popular than the words, and
was a frequently used model for German songs.

5. Czechoslovakia, Hungary

Many lines of evidence point to the modernity of the ballads
now sung in Bohemia, Moravia, and Slovakia. They contain no
reminiscences of the old historic legends of the Czech people:
Premysl the ploughman-king and the wise Libusa, the clever horse
Sermik, or Bruncvik, to whom were attributed all the adventures
of Henry the Lion. An Alexandreis existed in the language in
the thirteenth century, yet neither this popular literary tradition
nor any other has been turned to ballad uses. The religious
disturbances which convulsed the region from the advent of John
Hus to the fall of the Winter King are recorded in no ballad,
though there exists a Hussite hymn which was undoubtedly
popular. The allusions of soldiers’ songs take us back to the period
of the French Revolution and the Seven Years’ War, but not earlier.
In topics and style they are typical conscripts’ songs, exactly like

CZECHOSLOVAKIA, HUNGARY 273

those of the Austrian and Hungarian contingents which made up
the Imperial armies. Lyrical poems vastly outnumber the narrative
verses ; the actors are humble, and the action tends to take place in
inns or other plebeian surroundings. Though landsknechts and
reiters are not mentioned, much space is given to stories of crime.
Taken as a whole, the Czechoslovak corpus resembles the later
German, balladry as it had become after the middle of the seven¬
teenth century. It is from Germany that the more interesting sub¬
jects derive, together with the technique and some of the music. 1

German ballads or ballads which have crossed Germany give
rise to over a score of those current in Moravia. The saga of Hero
and Leander appears in two forms (185, 187), and, owing to its
lyrical degeneration, it is hard to mark off from other drowning
ballads, which may be its derivatives or be based on actual events
(90, 129). The Danube takes the place of the Hellespont, and
serves as well. The Samaritan woman is identified with the
Magdalene {The Sinner 3) in a ballad which, though crude, is im¬
pressive in its ruthlessness. The Murderer (189) is an offshoot of
Hallewijn; it belongs to the group in which the girl is actually
killed, and later avenged by her three brothers. All Czech ballads
of supernatural adventure owe either their origin or their trans¬
mission to Germany: they include The Shroud and—a variant—
The Dead Man (110, 160), The Orphan at her Mother's Gram (159),
The Girl at her Lover's Grave and at her Mothers Grave (86, 102),
The enchanted Daughter (146) who has been metamorphosed by a
curse into a tree, and another Dead Man (1x2), Janosko, who is
Sweet William to Kacenka’s Lenore. The Shroud is more gloomily
suggestive in Czech than in the surviving German versions of Der
Vorwirthy and some of the other pieces are not unworthily ren¬
dered. The luckless Wedding (89) seems to be a rationalized version
of Elveskud. Herman is the hero. He goes for his bride, but
robbers set on him and kill him; his bride is brought home in
ignorance of his death, but she hears bells tolling and kills herself
on his bier. Under the same number Susil included a ballad con-
substantial with the German Count Frederick , in which a bride is
accidentally killed by her husband's sword. The soldier's return is

1 F. Susil, Moravske ndrodnz Pzsne, Bmo, 1868-74. Some idea of the older

Czech literature may be obtained from W. Szegeda, Tsckeckoslotoaktsche
Anthologie von den Heldensagen bis ssur Gegenwart . Numbers in the text corre¬
spond to Susil’s numeration.

274 NORDIC BALLADS

a theme occurring in two versions: The Homecomer (144) includes
his return and the tests he sets his fiancee, and First Love (135)
contains the more circumstantial Moringer legend. The poisoner
theme of Schlangenkochin is variously rendered as the Poisoner
Mother (157), who kills her son in error for his bride, but is success¬
ful in another ballad (92), and The Poisoner Fiancee (166), which
stands nearer to the archetype of Donna Lombarda. The Infanticide
(158) is also of German origin, and The long-lost Sister (175) is a
tavern variant of the Kudrun story.

Suggestions have come from other sides also. The appearance
of the legend of Lear and his three daughters (119) is remarkable.
Susil gives two versions of the ballad, which is entirely traditional
and bears no indication as to its source. One cannot feel sure that
French folk-song has influenced the Czechs apart from what has
been mediated by Germans. The Biter bit (108) is an intrigue con¬
ducted with the light touch usually characteristic of France. The
Amazon (109) is a ballad which probably began its career in France,
spreading outwards from thence to Spain on the one hand and to
Northern Italy on the other. The Czech version is of a western
character and unlike those which circulate in Yugoslavia, Greece,
and Rumania; it stands between the western ballads and the some¬
what modified Ukrainian legend. One may suppose that it travelled
northward from the Italian possessions of the old Empire. In one
version the Kaiser of this ballad is called John, but in others he is
anonymous; the enemies fought by the Amazon are Turks, so that
the importation of the ballad may have taken place at any time
during the Turkish threats to Vienna (1532-1643). In contrast
with the paucity of Romance influences, there is the presence of
unmistakable Serbian motifs in Czech balladry. The Turkish Wooer
(128) is the old French ballad of the girl who feigned death to avoid
dishonour, but dressed in the Serbian manner, and Affianced to a
Turk (147) is wholly Serbian. Maruska and Janosek (107) is akin
to the ballads in which a girl is made to sing and her voice brings
robbers to attack her escort; the story is told in Yugoslavia of
Marko Kraljevic and other heroes. There is also The Poisoner
Sister (168), Ulianka, who gives her brother a stone for a couch and
a snake for fish. The transference of such themes was probably
direct, since in the Austrian armies Czech and Croat regiments
must have mingled, and their languages stand closer to each other
than to German or Magyar.

CZECHOSLOVAKIA, HUNGARY 275

A similar account should, no doubt, be given of Hungarian
balladry . 1 Its originality Is probably more musical than narrative.
The Magyars have stubbornly adhered to their language which,
being non-Aryan, cuts them off from communication with all other
Europeans. What they create in this language is their own peculiar
treasure, and remains veiled before the curiosity of western
scholars. They cannot easily lend, but they can borrow through
the languages (especially German) which their Isolated position
compels them to learn. Hungarians are of necessity bilingual or
polyglot, and they acquire for themselves the advantages possessed
by their neighbours. Their ballads thus are part of a continuous
pattern with those of Czechoslovakia and Austria, though I have
encountered no instance of borrowing from, the former. They are
of late German type in both technique and topics, but they show
also some novel aspects. The life of the Puszta, with Its cowboys
and cattle raids, Is unlike that of any other part of Europe, and It is
celebrated In songs, chiefly of a lyric kind. A strong air of freedom
blows through them. They are defiant of poverty and of restraint:
I am an orphan unafraid,

I get along with what Fm paid,
my herds are work done by my hands,
my two hands serve all my commands,
and the wide plain's my home.

Love lyrics occupy the greater portion of the Hungarian collec¬
tions. They are often sprightly, and touched with Imagination.
Humour and satire, drinking and dancing songs show equal verve.
At the other extreme there are some notable religious songs, such
as The Magyar Galley-Slaves 7 Song , which dates from the persecu¬
tion of Protestantism by the Hapsburgs in the seventeenth century.
The religious reform produced one of many situations in which
Hungarians opposed Austrians. Rdkdczfs March celebrates a more
purely political occasion in the eighteenth century, though its music
dates only, as It would seem, from 1809. Soldiers* songs are
numerous, and are those of the conscript type. The situations may
be universal, but the settings are modem, such as the watch on the
frontier in 1866 and the attack on Bosnia in 1878. The emperor
Francis Joseph is mentioned not without affection; though emo¬
tions of distaste and repugnance are usually evident in references
to the necessity and conditions of military service.

1 See Note J, p. 389.

276 NORDIC BALLADS

A general description of the Magyar repertoire by J. H.
Schwicker brings out clearly its links with Germany and the rest
of Europe:

In more recent times (he wrote) popular ballads and romances have
dealt principally with social conditions, events and types of daily life.
They are chiefly erotic narratives, and, with few exceptions, are of a
tragic nature. There is the youth who murders his fiancee out of
jealousy; the murdered boy who cannot be aroused by his father and
mother, but only by his true-love; the youth who dances to death his
false love; the tricked maiden or bridegroom; the sultan’s daughter who
frees two Hungarian knights from prison and flees with them; the test
for wives, when a prince despises the rich and takes to wife the poor
maiden; the bride who contrives her husband’s death; the faithless wife
who fondles another man as her husband lies a-dying; the man who
surprises his wife and her paramour, and kills both; the complaint of
a girl forcibly betrothed to a robber; the false wife who yet cannot
desert her children; the mother who surrenders her treasures for her
children; the daughter who leaves her mother to languish in prison;
the mother who kills her son’s fiancee and so drives him to kill himself;
the master-builder who builds his wife into the foundation of his walls
to secure their stability.

With some additions and subtractions, this paragraph might almost
serve as a description of Czech balladry also. It is not that the
catalogue is quite unoriginal. There are many ballads on the power
of love, contrasting the lover with the more tepid affection of
parents and friends; but one does not encounter elsewhere the test
that love should bring the dead to life, which is prescribed in The
dead Boy . Kadar Kata , the most-quoted of Hungarian ballads, is
a simple-minded story of true love and parental opposition, the
reported death of the lady and despair of the lover, and two graves
which give soil to intertwining ‘chapel-flowers’. The lines are
ragged, but the ingenuous history is carried to a swift conclu¬
sion by life-like dialogues. Fair Lilia is an offshoot of the
French King Louis's Daughter , and Kerekes Andros is the soldier
who returns, like so many, to find his fiancee married to another
man. The music, with its use of ecclesiastical tones as well as
Gypsy rhythm, shows the ballad to be old in Hungary. The
derivative of The Two King's Children (Gacsaj Pesta szep fiu vaut)
is highly lyrical, and its pentatonic structure is also a guarantee of
age. Susanna of Homlod is the Magyar representative of the
Schlangenkdchin y and Molndr Anna is the heroine of a Hallewijn

CZECHOSLOVAKIA, HUNGARY 277

ballad. The North German and Danish ballad of the Merman is
represented by the Magyar Palbeli Antal Clement the Mason's
Wife is of interest because it is due in the first place to the Ruman¬
ians who formed the basic population of Hungarian Transylvania.
Obtained by them from their kinsmen the Wallachian Rumanians
(who call the legend Master Manole), it is a ballad which started its
wanderings from an original home among the Greeks of central
Asia Minor.

Greek and Balkan elements mingle with German in the songs of
the Hungarian gypsies. 1 Their songs are mostly lyrical, but they
have some narratives, including some that may be considered their
own, either as composed by gypsies or as appropriated to the gypsy
life. A girl is advised to prefer a fiddler to any other suitor; death
meets a gypsy woman; and Black Vodas murders his wife. In
other pieces the themes are the roving life and irresponsibility in
love. They are of no high merit. The ballad technique at its best
is simple but delicate, and is not readily gained by a roving horde.
Roving leads rather to the picking up of unconsidered trifles.
Wlislocki’s collection shows many such borrowings. The battle of
father with son is told in Father and Son (13) rather in the Greek
manner of Tsamados than in the German of Hildebrand, and Anrns
and Death (1) is certainly the Greek ballad from the Akritic cycle.
In Hasten, hasten, Mother (6) the test to which parents and lover
are put is the Serbian test, namely, to snatch a viper from one’s
bosom. On the other hand, the ballads of German origin are more
numerous, for we are still within the sphere of German influence.
There is the inevitable Schlangenkochin (2), the Lenar e motif (3),
The hungry Child (11), The Maid and the Forest (12, the German
Tannenbanm), the Captured Lady (17), who must be ransomed and
only a lover will do so. There is, of course, no trace of literary or
historical tradition in gypsy balladry. Their tunes, or rather their
manner of playing tunes, have left a deep mark on Hungarian songs.

6. Lusatia, Poland

The districts of Upper and Lower Lausitz or Lusatia extend
from within the eastern border of the Saxon kingdom northward

1 H. von Wlisloeki, Volksdichtungen der siebenbiirgis chert und siidungarischen
Zigeuner, Vienna, 1890. He gives translations only in this volume. In his
Haidebliiten, Leipzig, 1880, there are texts and translations; but they are merely
epigrams.

278 NORDIC BALLADS

for a distance of about 50 miles in the valley of the Spree. The
inhabitants, numbering about 200,000, are of Slavonic stock and
speech. They call themselves Lusatians or Sorbs (Serske, Sarske),
but to the Germans they are known as Wends—the old name
applied to any representatives of the Slavonic stock. Their ballads
have been collected by Leopold Haupt and Johann Ernst Schmaler, 1
and are of special interest for two reasons. Firstly, the collection
is, in a peculiar sense, definitive. Apart from an abortive literary
movement in the nineteenth century, there has been no other
expression of Lusatian creative genius. The ballads are their all.
The collectors gathered texts from the whole district, together with
the surviving melodies; and they have added an illuminating intro¬
duction and full notes. Secondly, the people are a Slavonic islet
wholly subjected to German hegemony. At least fifty of their
ballads are palpably German, and there is a marked German strain
in their melodies. Though enclosed within Germany, they are not
out of touch with their Slavonic kinsmen. For a period Saxony
and Poland formed one kingdom. The 'king’ of their ballads is
usually the Polish king. Lusatia thus forms a bridge between
German balladry and that of Poland, Galicia, and the Ukraine. It
shows correspondences also with Czechoslovakia, into which
similar German influences poured, though probably from another
direction.

In the opinion of the collectors, Lusatian tunes stand on a footing
of equality with the best in Germany, and by their use of ecclesias¬
tical modes, they give evidence of respectable antiquity. They held
that specifically German traits are late. ‘The greater number (they
wrote) have the characteristics of Slavic folk-song; but some, which
resemble German types of song more closely, betray their later
origin/ It is possible the collectors may have begged the question
of date. They wrote at a time when music and song were considered
innate endowments of all men, so that a Slavonic folk would be
supposed first to sing Slavonic tunes. The point seems to lie open
to debate, since tunes collected only in the nineteenth century can¬
not but be of undetermined age. Save in dance-tunes, the enuncia¬
tion is leisurely, and there is much use of tremolo and trill. Verses
frequently begin, as in Polish Galicia and the Ukraine, with an
exclamation (ha or haPe), and this ejaculation may be used to

1 L. Haupt and J. E. Schmaler, Volkslieder der Wenden in der Oher- und
Nieder-Lausitz, Grirnma, 1841-3.

LUSATIA, POLAND 279

replace a missing syllable. The music shows that the lines are to
be considered syllabically regular, despite the minstrehs occasional
failure to keep the count. It covers two, three, or four lines, but
most often three. The texts, however, as in Czechoslovakia, are
usually couplets in this case, and the third line is obtained by
repeating one of the others, though It occasionally appears to round
out the sense of the stanza. Couplets are also the texts for several
pieces of quatrain music. There are lyrical cries, usually formed
of unintelligible syllables, and repetitions of words, as in German,
but no proper refrains, as in Scotland and Scandinavia. Repeated
lines naturally rhyme with themselves, and the treatment of the
odd line is wont to be arbitrary. Assonance is the principle of
construction, and is obtained from the simplest grammatical devices
(diminutive suffixes, flexions of verbs, &c.) ; but It Is so often absent
that It is evident the principle was not held in high esteem.

The Lusatian ballads have often the dignity and discretion which
Is attached to genuine medieval balladry. Evidence that they reach
back to a good epoch Is afforded by Handrias and Rajsenberk (I. 14),
which belongs to the ballads of robber barons. This anti-hero was
Christoph von Ressenberg, a belated practitioner of the gentle
trade, who was associated with Siegmund von Kauffung, a rogue
beheaded in 1534. The sixteenth century Is indicated also by a
piece of more doubtful reference, The poisoned Lord (i. 57). The
lord Is said to have been a king of Hungary, and he may be the
ill-fated Ludwig who perished at Mohacs in 1526, According to
the ballad the young lord drank poison in a cup offered him by a
Turkish woman, but his queen saw his soul go up to the bliss of
heaven. The evil Robbers (I. 2) are said to have been Tatars; in
Polish Galicia they are termed Tatars, and this may be due to the
fact that the Tatars have remained in the Ukraine until modern
times. In the Czech equivalent ballad they are said to have been
Turks. The name, therefore, Is not necessarily proof of high
antiquity. In view of these identifications, it would appear hazar¬
dous to follow the collectors of these ballads in their interpretation
of two others: Komm * muter, mein Gretchen and The Sorbs J
Victories (i. 31,4). They take these up to the remote tenth century,
to the epoch of the German wars with the Wends ; but the first
piece, with its German refrain, describes a Lusatian girFs choice
between three German pretenders, and the second celebrates three
victories over the Germans—themes which are equally interesting

280 NORDIC BALLADS

if merely fictitious. We cannot carry Lusatian ballads to a remote
date; but the sixteenth-century pieces are well enough marked
for us to consider the fifteenth as a probable starting-point. In any
case, Lusatian balladry contains examples of older styles than are
to be found in the neighbouring countries encircling Germany.

German counterparts can be found for about fifty of the pieces
reproduced by Haupt and Schmaler. These include the counter¬
parts of the Kudrun and Wolfdietrich ballads (i. 5, 82) in good states
of preservation, of the Leander {Two King's Children, i. 52, ii. 1)
piece, and of the mastersong ballad of The Count of Rome (i. 21).
The supernatural ballads include Hilzicka (i. 1), which corresponds
to the German Rubezal, with elements of the Hallewijn cycle, The
Merman (i. 34), The Shroud (i. 58), The Orphan at her Mother's
Grave (i. 132). There are several derivatives of Unter der Linde
(i. 15, 43, 47, 82, 134; ii. 15), and of the ever-popular Schlangen-
kochin (i. 77,147). The minor German ballads are also represented,
such as The Grass Girl, The Beggar from Hungary, The grim
Brother, The Nobleman in the Sack, and the highly typical pieces
in which natural allegories are used: The Girl and the Hazel, The
Owl and the Eagle. It is of interest to note that the Lusatian
ballads, though Slavonic, differ from those of Moravia in showing
no traces of Yugoslav traditions; nor is there any trace of the free
verse of Russia or Bulgaria. With their nearer neighbours, how¬
ever, the Sorbs maintain closer relations, so that the verbal resem¬
blances between their versions and those of the Czechs and Poles
are notable. One may see this in The Orphan at the Mother's Grave
and The Shroud. In Kitty of Niedergurig (i. 11), the heroine has a
German name, but the Don Juan who attempts to dishonour her
is called Wujerjowski panik ‘Lord of Unworth’, an evident pun on
Wuherskipan ‘Lord of Hungary’, which echoes the Czech distrust of
Hungarians. There are also some pieces of strictly Slavonic origin,
such as The evil Robbers, already mentioned, and Animal Life (i. 85).

More texts seem to have been borrowed than tunes. Several
Lusatian ballads are grouped under one tune, whereas German
ballads frequently have several tunes. The narrative detail was
rendered more precise, or perhaps retained the medieval precision
it may have lost in the land of origin. The objective, unlyrical
manner of the Lusatian ballads corresponds to older and better
models than those now available for several German pieces, or is
a reaction in favour of the older manner.

LU8ATIA, POLAND 281

These proofs of indebtedness should not blind us to the original
merits of the Sorb baiiads. They are divided into significant
classes, which correspond with the round of peasant occupations.
So there are songs for the fields (psezpolna), apophthegms and
saws (ro.nc.ka), dances (reje), improvised rounds (wuzenenja),
marriage ditties (kwasne spjewv) and songs for begging (stonanje),
as well as religious legends (podkurlusje). The apophthegms give
maidens a chance to show their wit before the dancing starts ; the
rounds enliven proceedings with personal allusions; one begs a
morsel when any one kills a pig or bakes a loaf. The field-songs are
sung when passing along the fields from the townships to the
country, and their length is as the length of the road. There are
no historical cycles of ballads in a district which has had no
independent history, and no direct literary reminiscences. There
are religious ballads on Jacob and Rachel, Joseph’s chastity, the
Samaritan Magdalene, St. George, St. Nicholas, and various moral
tales, generally such as occur also elsewhere. The bulk of the
collection consists of love-songs which cover the usual classes:
wooing and winning, seduction and remorse, the vigilance of
chaperons and warnings to girls, trickery, tests of affection, separa¬
tion and reunion, desertion, opposition, tragic accidents, death, and
passionate crime. There is comparatively little of the goliardic and
anticlerical satire of the Low Countries and Rhine provinces, nor
is there a Newgate Calendar of vulgar crimes. Criminal ballads
are a well-known symptom of decadence, and their absence is a
sign that the Sorbs have retained in their songs their pristine
innocence.

Lying farther to the East, Poland is a land .rich in folk-songs, but
poor in true ballads. The ballads, indeed, seem to be not so much
Polish as Galician, that is, they belong to the same Ruthenian-
speaking people as cultivate narrative oral poetry in the Ukraine, and
their evidence is used in forming Ukrainian anthologies also. 1
Representatives of German ballad cycles appear in Polish in highly
lyrical forms. The lyrics, on the other hand, flourish vigorously;
"they accompany our folk from the cradle to the grave*. They are
adapted to the main family events, such as betrothals, marriages,
christenings, and burials; they celebrate the red-letter days of the
calendar, as Easter, Midsummer’s Day, and Christmas Eve (koledy)
and their language is the utterance of farmers, herders, huntsmen,

1 See Note K, p. 389.

2 8 2 NORDIC BALLADS

miners, fishers, raftsmen, artisans, and soldiers. Fully half the
repertoire consists of love-songs on the lips of women.

All these lyrics are danced as well as sung. ‘A Polish song’, said
an eighteenth-century observer, ‘makes the whole world dance.’
Every genuine folk-song is a dance, and every dance a song. Lively
music quickens the feet; the notes of the bagpipe, shawm, and
fiddle are piercing and sprightly, and the rustic picture is com¬
pleted by the swirling, colourful popular costumes. These are
elements added by the Poles themselves to the technique of
Germany, for their verses stand in close relationship to those of
Germany, and not to the danced ‘viser’ of Scandinavia. They are
generally quatrains or couplets, though more elaborate stanzas also
occur. The lines are tolerably regular and are assonated or rhymed;
but the assonances are so elusive and so simple that one might
doubt whether they constitute a principle of Polish prosody. There
is not much use of refrains, nor of triplets. The rhythms of the
dance-songs, especially in Galicia, are those which characterize
Ukrainian traditional verse, thanks to the powerful cultural in¬
fluences which extended from Poland into Southern Russia from
the sixteenth century. Conversely, there is no free verse to corre¬
spond to the indigenous ‘dumi’ of the Ukraine or to the Russian
‘byliny’. The Polish folk-song, though Slavonic, belongs to the
occidental tradition.

Tur nin g to the ‘men’s songs’ (m^skie piesni), which, according
to Zaleski, ‘describe events referring to the whole country or to
particular individuals’, it is disconcerting to find among them some
which express frank dislike of the Poles. They are Ukrainian songs
dating from the great Cossack revolt. There are others which do
not concern Poland at all, but refer to Muscovite victories in
Turkish lands. Their evidence is of doubtful value in an attempt
to determine the age of Polish balladry, since so many of them are
not Polish but Ruthenian. For what it is worth, the evidence
points to a quite modem date. Chotim (Zaleski AA 3) refers to the
defeat inflicted on the Turks in 1739 by the Russians. A swiftly
moving ballad entitled Cossack Nyczaj (Zaleski AA 1) tells how he
was surprised and captured by the Poles, then hanged and
quartered. It should probably be associated with the revolt of
1651, though there is no compelling reason to prefer that date to
another. The discomfiture of a certain Drewicz or Derewicz
(Zaleski AA 2), who was led in chains to Cracow, is in the more

LUSATIA, POLAND 283

lyrical Polish manner, but Kozubaj (Zaleski AA 4) is again a his¬
tory of one who fought against the Poles. Bows and arrows are
mentioned as weapons, and the poem may go back as far as the
sixteenth century. In short, the harvest of historical ballads is
meagre, and their sympathies generally Ruthenian. The use of
full rhyme, though irregularly maintained, is a most modern
feature. A poem entitled The Turk's Bride (Zaleski BB 31, Walter,
p. 41) is the same as the Ukrainian Mother-in-law Prisoner to her
Son-in-law (Dragomanov 64). An old woman is captured and set
to work by her captor’s wife; she rocks the baby, crooning to it
that she is its grandmother, and so a recognition takes place.
Dr. Kamienski refers this tableau to the epoch of Turkish Wars
with Poland, which raged from the fifteenth to the seventeenth
centuries; but it is not certain that the piece is Polish in origin.

With the Ukraine are shared the ballads of Roman and Olena
(Zaleski BB 7, Dragomanov 65), a modem, sentimental ballad, The
dying Cossack's Farewell to his Horse (Zaleski BB 8, Dragomanov 58),
The Falcon on the Poplar or the lover pays the ransom, (Zaleski
BB 15, Dragomanov 34). The Robber's Bride (Walter, p. 52), is
probably of Yugoslav origin; it is found also in the Ukraine and in
Czechoslovakia. With this neighbouring country one has to com¬
pare Marcisia (Zaleski BB 9) or The Infanticide (Susil 158), The
Falcon on the Poplar , The Murderess (Zaleski 26, Susil 98), and
the inevitable Samaritan Magdalene (Walter, p. 85). These corre¬
spondences do not amount to the identity observable in the case of
Ukrainian ballads. Resemblances between Polish and Lusatian
songs are frequently noted by Haupt and Schmaler. These
instances are usually ballads which have a German origin, or at
least Germanic analogues: The Mermaid , The Soldier's Home¬
coming, The Test of Affection (Liebespmbe), Leander , The Ravisher's
Reparation , The Poisoner (Walter, pp. 6, 15, 8, 17, 28, 32), together
with some of the themes held in common between the Poles and
the Czechs. Comparatively few of these have extended into the
Ukraine. In comparative balladry, therefore, Polish Galicia is a
frontier of the occidental style. Few of the themes, but more of the
technique, have penetrated farther, into the Ukraine; and there
has been a less conspicuous reflux from the Ukraine into Poland.
In the centre and north of the country, however, the predominance
of the lyric makes Polish popular verse a natural transition from
the German to the Baltic manner.

284 NORDIC BALLADS

Even in borrowed poems there is room for a certain originality.
The Polish ballad of Leander contains no more than the final scene
of Two King's Children, and shows the withdrawal of Leander’s
body from the waves. It is the most poignant moment of the
pathetic history, and well suited to treatment as a song. Another
ballad may be based on Eheskud (Walter, p. 15) or on another
cycle. It begins:

There rides a rider with steel bedight,
he rideth home from the awful fight.

His family receive him, as in the Danish ballad, and put him to
bed; but then the story takes a new turn. He learns that his mother
has died, and he dies of grief:

'To thee, dear Son, what can I give,

I that so long have ceased to live.

Take a small room so dim and lone

’twixt worm and root and the grey, grey stone.’

‘Oh heaven, oh earth, oh heavenly grace !

Oh dearest mother in grisly place !’

The rider rides with death and night,
comes no more home from the awful fight.

The rendering of The Castle in Austria (Walter, p. 11) has a similar
light touch. In Sir Saw a (Zaleski BB 23) we find an effective
history of what may have been an actual event. Sir Sawa returned
home to find that his wife had given him a son; but no sooner had
he descended to the cellar for wine to celebrate the event, than
Cossacks attacked him and killed him unarmed. The tragic
emotions are here, as elsewhere, those affected by the ballad-
poets, as in the more vulgar Lord Kaniowski, Jasia, and Nastyna
(Zaleski BB 13, 14, 2, 1).

7. Lithuania, Latvia

One might say that the colour of Lithuanian ‘dainos’ is a trans¬
lucent green. It is not merely that the colour is so often mentioned
in the poems. It is rather that the language and manner have the
qualities of early Spring; that there is something strangely delicate,
pellucid, and yet vivacious in the songs of the Lithuanian peasants.
They are songs rather than ballads. The predominance of the lyric
which is notable in later German work, and augmented in Poland,
1 See Note L, p. 389.

LITHUANIA, LATVIA 285

becomes absolute in crossing this further linguistic frontier. All
Lithuanian work is lyrical. There remain, however, certain
narrative suggestions, of a faint and elusive quality. The situation
0.11 which the emotion is founded is outlined in one or two strokes,
like those of a Chinese painter of birds in flight. One learns some¬
thing about place and circumstance, as one does also in the "cos-
sautes’ of Portugal. The emotional appeal is heightened by repeti¬
tion. As in old Portugal and Galicia, so in Lithuania, the subtlest
suggestions are made by means of parallel phrases and tableaux;
but there is greater variety in the types of Lithuanian parallelism.
These "dairies’ are so often exquisite, so normally happy, that they
must be held to be the most perfect type of those ballads which
have been described earlier as narrative lyrics. They lie at one
extreme end of the ballad spectrum.

Lines are of various length, and without rhyme or assonance in
principle. In practice it is hardly possible to avoid assonance,
thanks to the incessant use of beautiful liquid diminutives, and to
identical flexions in the verbs. Alliteration, also, is not a prosodic
principle, but a constant occurrence. We have not to suppose that
the singers are unaware of these f eatures of their songs, but that, on
the contrary, one of the principal sources of their pleasure is the
gossamer of elusive echoes which floats about their poems. Lines
are, however, grouped into stanzas of simple, but precise, struc¬
ture. It is the function of the melodies to group the lines, but they
cohere also in sense. The parallel construction of phrases and
distribution of line-lengths produce stanzas which are readily
recognized as such even in print. In short, this is a quality held
in common with the whole Nordic group, and the absence of dance-
refrains is a specific feature of the German sub-area. We have seen
that the intermittent assonance of German balladry shows signs
of failing altogether in Poland; Lithuania’s "dames’ are but one
remove farther on the same road.

The music scans the lines. The ordinary prose accents of words
are not obligatory in verse, and the latter can be "read by the
melody’ (ant balso skaityti), 1 with an arbitrarily regularized

1 Nesselmann quotes as an example:

Normal Melodic Translation

Ant tiltuzio stowejau, Ant tiltusio stozvejdu, I stood on a bridge,

su mergyte kalb^jau $u ?nerg$te kdlbejdu I spoke to a maid

The grave and acute accents represent minor and major stresses. (The dots over
certain vowels and consonants are orthographic.)

286 NORDIC BALLADS

enunciation. These musical phrases are built into distichs, triplets
(usually with the last line long), quatrains (with or without refrain),
and stanzas of seven or more lines. The recurrence of the same
phrase or of its twin brother marks the divisions of the thought,

and there is an ampler parallelism in many songs, which consists
of placing together two complete tableaux to make a diptych. The
listener is asked to notice the same situation in two aspects; the
emotion is not elaborated, it is shared. So we find many ballads
divided between, for instance, a youth and a maiden, as in the
following case:

Early rising in the morning,
to my garden-plot I wandered,
and I gathered me green merplants
and among the rue I bedded.

Then there came, there came a stripling,
sought to waken me, the maiden:

‘Wake, oh waken, gentle maiden,
much too humble is your chamber.’

‘Let it grieve you not, my stripling,
good enough for me my chamber,
with the green rue for a mattress
and my garland for my pillow.’

Early rising in the morning,
to my horse’s stall I wandered,
combed and curried there my sorrel,
and beside the horse I bedded.

Then there came, there came a maiden,
sought to waken me, the stripling:

‘Wake, oh waken, gentle stripling,
much too humble is your chamber.’

‘Let it grieve you not, my maiden,
good enough for me my chamber,
with the green hay for a mattress,
saffron saddle for my pillow.’ (Nesselmann 298.)

In Latvia the ‘daina’ becomes epigrammatic. The number of
stanzas is reduced drastically in all cases, but the great majority of
songs contain no more than one. If the Lithuanian poems corre¬
spond to the ‘cossantes’, the Latvian correspond to the modern
Portuguese ‘quadras’. Lines may be of various lengths, but there
is a strong preference for the octosyllable. Stanzas may be of any

LITHUANIA, LATVIA 287

length, but the preference is markedly for the quatrain. In the
absence of other verses, the stanzaic structure is no longer apparent,
and the Latvian ‘dainas’ have the appearance of unrhymed, un-
strophic, octosvliabic verse, with elements of assonance and allitera¬
tion corresponding to no fixed plan. Such a description would fit
the folk-poetrv of their neighbours, the Esthonians, though the
‘laulud* of the latter are normally longer than a quatrain. Phrases
of the Latvian Llamas’ are made to run parallel,, and a similar
parallelism of phrase in the longer folk-songs of Esthonia tends to
form strophes of unequal length. Such divisions are optional, and
not ahvavs present. They occur in the Finnish ‘runot also, which
have the Esthonian characteristics, together with a more scrupu¬
lous use of alliteration,, which becomes a principle of versification
in Finland, though not reduced to a mathematical norm. Thus by
a sequence of minor changes the technique of German folk-song
can be placed at the end of a continuous band, of which the other
extreme lies in Finland. That is, I have no doubt, what actually
occurred ; and the Latvian ‘daina* is a form derived from the
Lithuanian by compression. This does not imply necessarily
Lithuanian priority, but only that the Lithuanian form is the older
as now extant.

The unity of c dainas 5 and ‘dainos* is attested not only by the
words, and by the close kinship of the two languages; but also by
the use of the same mythology, the same domestic range, the same
motifs and divisions, and even the same fables. The Owl at the
Sparrow’s Wedding , The Orphan at her Mother’s Grave , The stolen
Sister (Barons 2546, 3944, 13646), are pieces with Lithuanian
parallels (Nesselmann 13, 67, 204). Correspondences of this sort
are the more remarkable, since the Latvian quatrains are essentially
improvizations. In a sense, they have no permanent existence;
they can be called into being at any moment in accordance with an
accepted formula of composition. Under such conditions there
would be an inevitable loss of narrative particulars which were
already elusive in the longer Lithuanian ‘dainosk On the other
hand, the fact that the Latvian songs spring spontaneously on all
occasions, and not only on such as are naturally fitted for poetic
elaboration, causes them to follow much more intimately the joys,
sorrows, and thoughts of the singers. Their documentary value is
high ; their aesthetic value is much lower than that of the Lithuanian
songs.

288 NORDIC BALLADS

Indebtedness to the West is evidenced by the first two of u,
pieces cited, and by several others. The class of animal histort ''
as we have seen, peculiarly German. Thev are f ra , , to ! s ’
Lusatia, Czechoslovakia, and Poland—lands subject to ST
influence. In both the Baltic states they are quite m,™ 1

wholly charming. The Sparrow's Wedding recalls the Ger ^
Bird's Wedding ; The Wedding of Bear and Wolf (NesselmamTiA
is a satire hke the Lusatian Animal Life (Haunt and Sd™

83) a»d ta Polish congeners. Eight of LLlXn Vd wti
the story of the girl who lost her ring in the water and Z J
recovered by her ,over (83-9, 9S ), whhh is the are? ’j£ tae,

K W ?° l “ PeS “■ The wooer who pretends to ^eat
wealth but is detected in his lying, appears in one T^Jg

/Srp 1 '’ 8 ; 1 ”™ l 53 ", “ Wel1 as ,he Swedish Seven golden
Hills (Bergstrom 64) and its German congeners. Mother father

sister and brother are asked to perform a dangerous service which

Swfss D W f° (NeSSdmann I2 5 ); a boy enlists, like the
Swrss Dursh, to make money for his wedding (Nesselmann r,„v

nS co ^ >ei ' r0nnelle ’ g° es off with the soldiers and Mil

mannTsvWeL d CSSe m fT 2 ° 4 ’ 357); Nutm ^ s andPink * (Nessel-
Tn altotf 7 d ® US ° fthe same combination in a German song

Jbar n 11686 f 3868 thC S ™ phfication of detaiIs makes for the loss d
that nexus of events which constitutes the identity of a ballad Th

orphans tragedy is simply her loneliness, and there are many

beSTidlntifi “° m ° re ' Rhesa ’ s twenty-fourth ballad has

mth j^Graiin l VerS ; 0n r 0f the widespread Orphan at her

eouId be &ZZ f T’ ^■ 7 ’ NesseImann 67); but nothing

could be farther from the circumstantial narrative of Denmark than

I, poor house-drudge,

I, poor orphan,
to drudge am bounden
the weary daytime.

How I long for
my mother only,
my intercessor!

Long she’s tarried
in high hillock;
o’er her barrow

rue drips dew-drops
glittering brightly
like pure silver.

LITHUANIA, LATVIA 289

In a ballad of bride-stealing we are told that she has become a
Tatar (Nesselmann 204,1; the detail is the more to be treasured
because it is so rare ; though one hardly knows whether to Infer that
the sone Is ancient, or that It has wandered In from the Ukraine.

Any attempt to fix the age of the Latvian epigrams Is defeated
by their brevitv. There are more shreds of evidence attached to
those of Lithuania, thminh they are not easy to clarify. One piece
collected by Juskevic [Lie!. Dai nos 11624 entitled Five Brothers ,
relates the sufferings of prisoners driven Into Tilsit and flogged to
death with green willow withes. There are no notes as to arms and
accoutrements, but the tyrants are called crusaders. That being
so, they can hardly be other than knights of the Teutonic Order,
whose power was shattered at the battle of Tannenberg in 1410.
It is presumptive evidence that the 'daina 5 was already an estab¬
lished form in the early fifteenth century, though we must admit
that the case rests on a single word. Apart from this Instance, when
allusions can be caught, it is to events of the eighteenth and nine¬
teenth centuries: to an attack on Danzig, which may have been
that of 1734 (Nesselmann 23), in which ravens are depicted bring¬
ing back, as in Serbian songs, the white hand of a slain hero; to
Kiistrin or Kunersdorf in 1759; to the battle of Rosieny (Juskevic
102), and to Kosciusko’s son (Nesselmann 52). The Russians are
regarded as compatriots, and the foreigners are Germans and
French. The former are usually regarded with sympathy, the latter
as enemies:

The French king’s
a mighty robber,

the Prussian king’s

a mighty hero. (Nesselmann 28.)

The historical situation is that of the Napoleonic Wars, from Tilsit
to Montmartre. One singer, unusually generous of detail, tells us
that Yorck faces Bertrand at Wartenburg, Kleist faces Vandamme,
Billow is In Berlin, and Bliicher directs (Juskevic 1150). Pats
Blukerisy ‘Father Bliicher’, is as familiar to Lithuanians as to Ger¬
mans, and the situation Is summed up that ‘were Bliicher not alive,
the Prussian land were destroyed’. One seldom encounters a ballad
that praises the French at the expense of the Germans (Juskevic
1106), though that may accord with the sympathies of Lithuanians
in the days of Jena, while there "was still small risk of the invasion
of Russia. A similar inference as to date may be made from the
4615 p p

290 NORDIC BALLADS

conscripts 5 ballads which, if they mention a master, it is the King
of Prussia, and if they mention an enemy, it is France. With
identifications such as these, the Five Brothers seems singularly
isolated in the fifteenth century.

The ninth century has been claimed for the antiquity of ‘dainos’.
I do not know on what evidence the claim is based, nor whether
any attempt has been made to distinguish between the word in the
general sense of ‘song 5 and the poems of precisely the modem
mould and texture. The claim cannot be supported, at least, by
reference to the numerous pagan ‘dainos’, which refer to the
thunder-god Perkuns, Luck and Ill-luck (.Laime and Nelaime ), the
Moon-god and Sun-goddess, Zemyna who governs earth, and the
deities of the sea. The bigamy of the Moon-god is thus recorded :

Moon brings the Sun, a bride, home,

’twas in the first of Springtides.

The Sun arose so early,
her had the Moon deserted.

Alone the Moon did ramble
and wooed the Star of Morning.

But Perkuns, filled with fury,
he swung his sword and cleft him.

‘Why’st thou the Sun deserted ?
why wooed the Star of Morning,
by night a lonely rambler?’

(Rhesa 27, Leskien, Lit. Les. 1.)

According to another version (Nesselmann 4), the god split an oak
and sprinkled blood on a maiden’s frock; the blood could only be
cleansed when nine suns rose on a happy morning. These whimsical
pranks of the deities give a childish charm to the ‘dainos’; and it is
a pretty touch, in some orphan ballads, to declare that the orphan’s
only kin are Moon, Sun, and Stars.

The Latvian ‘dainas’ offer us more intimate glimpses, though
tantalizing. A vast amount of detail is preserved in Barons’ col¬
lection in nearly twelve hundred songs; but as they are so
short, one seldom gets more than a stray note. Some are charming
vignettes:

The Sun at even goes to bed

and she bedecks the forest’s summits:

she gives the lime a golden crown,

LITHUANIA, LATVIA 291

she gives the oak a crown of silver,
and on. the tiny willow-tree
she slips a lover’s ring of silver.

(Barons 33879, Jonval 203.)

Sometimes the information has the merit of oddity:

The heavenly Forger toils in heaven,
his sparks fall glittering Dvina-wards;
he forges for that girl a crown—
for that girl who has nine brothers.

(Barons 33722, Jonval 453.)

Perkuns has nine black horses,
the nine have stones for fodder,
and silver drink for water,
of shining steel the trough is.

(Barons 33705, Jonval 429.)

At other times we hear of the sendees rendered by the gods.
Perkuns rides the sea and blasts oaks, but he it is who causes grain
to swell in the barley. Still more beneficent are the Sun and Laime,
since they concern themselves with human joys and are especially
kind to orphans:

White, so white, a spirit comes
at eventide at set of sun.

Nay, ? tis not a spirit white,

5 tis the orphan’s loved Laime.

‘Come, good evening, little orphan !

Time, they say, for your betrothal.’

‘Whither can I go, dear Laime ?

Not a thing have I that ’s needed !
not a glove, and not a stocking,
neither have white woollen blankets. ’

‘Go, dear maiden, cease to tarry !

I must give you what is needed,

I must give you gloves and stockings,

I give you white woollen blankets.’

‘But no bullocks, but no oxen
have I, nor a sorrel pony.’

‘Ill give bullocks, 111 give oxen,

I must give a sorrel pony/

Come, all people, come and see then
how the wooers court, the orphan.

292 NORDIC BALLADS

how they herd the bulls and oxen,
how they herd the sorrel ponies;
and her trousseau—gloves and stockings
packed between white woollen blankets.

(Endzelin 36, Barons 4976.)

These poems are indeed interesting and tell us much concerning
the intellectual background of the ‘dainos’ and ‘dainas’; but they
do not give their age. The Baltic peoples remained pagans officially
until the opening of the great Jagellonian period, and Chaucer’s
£ verray parfit gentle Knight’ endeavoured, with blows, to change
their beliefs:

Ful ofte tyme he hadde the bord bigonne
aboven alle naciouns in Pruce.

In Lettow hadde he reysed, and in Ruce,
no Cristen man so ofte of his degree.

Jagiello’s politic conversion did not affect the mass of the people.
In the fields and forests the ancient deities still commanded belief
and affection, and as late as the seventeenth century a missionary
was informed that there were as many gods as places, persons, and
wants. The singing of ‘dainos’ was considered a heathen practice,
as that of ‘laulud’ in Esthonia, until as late as the middle of the
nineteenth century; and pious ‘giesmes’ circulated in opposition to
them, but with a sharp trench between. The epoch of universal
heathenism, then, reached to the late fourteenth century, and that
of popular heathenism almost to our own times; the pagan spirit
of the dainos and dainas cannot be used in evidence for their
antiquity.

For the rest, the poems of the two small Baltic peoples, being
essentially lyrical, must be reduced to a small summary in a book
devoted to narrative verse. They are poems of ploughmen and
fishers, who have no history beyond the needs of the day and
season, and no delight in war and ambition. They speak of the
sorrows and devastation of wars, of the pain of parting, when the
conscript rides away saying ‘Nevermore’, to leave his body on the
field and a riderless horse as a token. Maidens are advised not to
follow the troops, since ploughmen make more constant husbands.
Better be a yokel than a trooper, or even a rich man, artisan, or
boyar. Then follow ballads of domestic occasions. Over and over
again we follow the round of falling in love, wooing, wedding,
marrying, crooning lullabies, dying. Toil quickly ages the peasant

LITHUANIA, LATVIA 293

woman.. Courtship is the one happy moment; for marriage drags
her away to a strange land and folk, and her life becomes an un¬
ending burden. But the maid is happy and expectant. She sports
her wreath of rue, and makes pretty fancies revolve about it. But
it may be withered by scandal or suspicion, or lost by carelessness.
She may lose a ring also at the fountain or in a wood, and its
recovery is the equivalent of a proposal of marriage. The youth, for
his part, is a ploughboy or fisherboy. He sleeps in a stable and
rides a-wooing on a brown horse. The encounter may be anywhere,
but the proposal must be in due form to the assembled family.
Then there is bustle about the trousseau, which should have been
gathered in the years of spinsterhood, but may not yet be ready.
The old folk lose a spinner or a farm-hand, and feel the economic
strain keenly. A wagon takes away the gathered trousseau, and if
its wheels break, that is a sign of bad luck. Then on the wedding
morning the girl wakes brightly enough, but is soon crying in her
parents’ arms and lamenting her long voyage. Sometimes she has to
reproach them for sending her so far away: two hundred miles is
as bad an estrangement as the circumference of the globe. Or she
may complain that she has been paired off with a knave. Then
there are the heavy songs of toil, aggravated by the habits of the
husband. He lolls in taverns and sings songs in praise of hops, rye,
and barley. Then there are the liquid cradle-songs, and at long
last there are dirges in measured prose. The song may be about
the song itself, which is a kind of companion always at the singer’s
side :

Singing bom, a-singing grew I,
singing all my life must pass,

singing goes the soul within me
in the garden of God’s sons.

(Barons 3, Endzelin 1.)

8 . Esthonia, Finland

In the previous chapter attention has been drawn to the common

elements of ballad technique which prevail in the two Baltic
countries of Indo-European speech and the two which speak
varieties of Finnish, 1 The Esthonian tongue is as intimately con¬
nected with Finnish as are the dialects of a West European speech;

the ballad style, the themes, and the quasi-epic associations are

1 See Note M, p. 390.

294 NORDIC BALLADS

those of Finland also, though less perfectly exemplified. There is
but one ballad manner in Esthonia and Finland, the pattern of
which is to be found in the songs of Eastern Finland or Karelia,
collected by Elias Lonnrot. In these songs the line is of eight

syllables. The spoken accent is free, but the music imposes a
trochaic rhythm. There is no rhyme or assonance, save what may
occur by chance. Parallelism is a conspicuous feature of the style,
and it tends to break up lyrical passages into approximately equal
blocks; but the stanza is not achieved, and in narrative pieces there
is practically no such division of the material. Thus far there is no
essential difference between the Finnish-Esthonian style and that
of their immediate neighbours in the south.

The difference arises in the use of alliteration. Alliteration is
frequent in the folk-songs of all Nordic peoples, but in Finland
(and to a less extent in Esthonia) it is so prominent as to become a
prosodic norm. The accentuation of the first syllable of every word
encourages the use of this device, and the hint may have come
originally from the Swedes. In Eddie verse and in the modern
Icelandic ‘rimur’ the recurrence of certain initial letters follows a
fixed pattern. There is no fixed pattern in Finnish alliteration. It
may affect all the words of a line, and generally does affect two at
least; but these words are not required to occupy a fixed position.
On the other hand Finnish alliteration really affects more than an
initial letter. The language is relatively poor in consonants, and
the vowels also, divided into three classes by the laws of vowel-
harmony, are relatively few, but well defined. The alliteration is
satisfied, indeed, by its occurrence in the first letter of words, but
the following vowels usually correspond, either as being identical
or as being associated in the regular permutations, and there may
be several other consonants and vowels yielding echoes . 1 They are
due, as in the Timur’, to the application of a high technical stan-
dard to the composition of popular verse. It would be hard to
decide whether this standard was original or elaborated in more
recent times. In Esthonian verse it is more arbitrary and fluctuat¬
ing. Though the surviving forms in Finland are the older, they are
encountered in regions which were not the home and focus of
Finnish balladry, but have suffered displacement towards the
periphery of that culture.

The Finnish ballads have been pressed back from the coastal
1 See Note N, p. 390.

ESTHONIA, FINLAND 295

lands into the eastern interior. They flourish now in Karelia, and
it is the Karelian dialect which has come to be considered the lan¬
guage of poetry, though the modern literary tongue is that of the
western towns. The study of ballad variants has proved this move¬
ment. One notices also that the scene of action is usually the coast
—the fortress of Viipuri or Viborg, and the province of Hama.
There are cases of ballads which have ascended, in all probability,
from Esthonia. Such a one is Kanteletar 4, a version of the wide¬
spread Dives and Lazarus , in which Lazarus is identified as an
Esthonian slave. In Esthonian folk-songs there is emphasis on the
norms established by the coastal towns, especially by Riga. The
contents of such songs show that they belong to peasants and
fishermen; but their inventors insist that the matter has come from
the towns. They insist, moreover, that the matter is German ; to
be German is a guarantee of merit:

Folk-songs are no songs for babies,
songs for babes or scolding women,
folk-songs are fine German ditties,
cradle-songs for our young lordlings,
drinking-songs that please the junkers.

(Vana Kannel , i 3.)

The great city is Riga, a city of Latvia under the control of German

barons and merchants. These allusions are evidence that the ballad
came to Esthonia from the south-west.

To be more precise, the Germans referred to in the ballads are
‘ Saxons’. Such indications accentuate the importance of the pieces

which have international connexions. The Murderess of her Husband
{Vana Kannel , i. xxi) belongs to the series of ballads on this crime,
which are current also in Poland and Czechoslovakia. It contains
even the swift set of searching questions which are to be met with
elsewhere. The Bride Murderer (Vana Kannel , i. 109) is more
difficult to place. It resembles the Hallemjn ballads in its ‘leit¬
motif’, but in some of the details it recalls our Edward. Edward is
directly represented in Finland by the modem Bloody Son (Verinen
poika , Arwidsson, ii. 87), a translation from the Swedish. The un¬
charitable Relatives (Vana Kannel , i. 103) is a ballad of useless
appeals to relatives for help—on this occasion, for the means of
avoiding conscript, service. The tableau is a familiar one in those
ballads which end by the generous sacrifices of a lover; but in this
poem no lover appears, and the recruit leaves bitter legacies to Ms

2 9 6 NORDIC BALLADS

people. It is, therefore, not certain that this Esthonian ballad
belongs to the international cycle indicated. No doubt attaches to
The Daughter at her Mother's Grave (Vana Kannel 68), which exists
in several versions and stands especially close to the Lithuanian
ballad. In Finland, Inkeri's Suitors (Kanteletar : Virsi-Lauluja o)
is a ballad of the cycle of the crusader’s return. Lalmanti the great
reiter leaves home; Eirikki falsely reports his death to Inkeri;
Lalmanti returns. The names suggest Swedish provenience. The
Sea-Suitors (Kanteletar: Virsi-Lauluja 38) is one of the Merman
eyde. Among religious pieces Lonnrot printed one on St.
Catherine, a Dives and Lazarus with somewhat original details
and a Magdalene (. Kanteletar : Virsi-Lauluja 3, 4, 5). In a general
way, all Christian ballads are evidence of the influence of Central
and Western Europe upon Finland and Esthonia. So, Maid Mary’s
Verse (Kanteletar : Virsi-Lauluja 6), the longest of all these pieces,
must be included among ballads indicating a foreign influence. It
certainly originated in the Christian towns of the western coast,
and not in Karelia, where it is now encountered.

The names of the Finnish and Esthonian types of folk-verse also
beai witness to the same effect. The indigenous word is ‘laulu’
‘song’. Two terms are current for ‘narrative song’, and of these
Lonnrot allows ‘virsi’ ‘verse’ as more appropriate to the tradi¬
tional ballad. In Esthonia one encounters ‘viizi’ ( Vana Kannel, i.
6), which is the German ‘Weise’ or Swedish ‘visa’. The word
‘runo’ does not prevail in Esthonia, but only in Finland, where it
is a sign of Scandinavian guidance. Apart from the word ‘laulu’,
therefore, all the nouns involved indicate German or Swedish
models. Naturally, this evidence is more cogent when we consider
the new stanzaic songs which, in the coastal cities, have taken away
the fame of the Finnish ‘runot’. Lonnrot printed twenty-four
such pieces in the preface to his Kanteletar. They are in a variety
of metres. couplets with and without refrains, quatrains, verses of
five, six, seven, and eight lines. The length of the individual lines
also varies, and among the terms used to describe this sort of folk-
poetry we find ‘romantsi’, the Spanish word which had been
acclimatized in Germany in the early nineteenth century.

Historical ballads give some indication of the age of the ‘laulud’
also. The allusions in Hurt’s Vana Kannel are few and vague.
There are conscript-ballads, and they refer to service in the Russian
armies. A couple of songs refer to a raid on the island of Oesel, off

ESTHONIA, FINLAND 297

the mouth of the Gulf of Riga (Van a Kannel 100, 101). The
enemy are Swedes, and the date would seem to be the seventeenth
or early eighteenth century. In the Kanteletar , No. 10 refers to an
attack by Ivan of Russia on the frontier fortress of Yiipuri (Yiborg)
in the sixteenth century; No. 11, Jakko Panins, tells of an attack on
Riga, which was held by insurgents in the last years of the sixteenth
century; and Charles's Army (No. 12) is an account of the Swedish
disaster at Pultava in 1709, ending with the return of Charles XII
to Sweden. It is a counterblast to the Swedish ballad of Narva.
Earlier than these is Elina's Death (No. 3 ). It is probably the best
of the Finnish narratives—a grim history of the mutual hate of two
women, of adultery and of ferocious murder. The foundation of the
ballad is an actual occurrence of the fifteenth century; in modem
times the ballad has been made into a drama and represented on the
stage. An older event still is the killing of Bishop Henry in 1157.
Bishop Henry, the apostle of Finland, was an Englishman. In the
ballad we overhear his interview with the Swedish king (Erik), and
the defiance which he suffered from the pagans. It is a spirited
ballad, but we are not compelled to believe it the work of a con¬
temporary. A martyr’s reputation was, of necessity, in the hands
of the clerks, and one of them - could confect a ballad out of the
available materials at any date whatsoever. The religious ballads
are Catholic in sentiment, and so represent a point of view which
no longer prevailed after the sixteenth century.

A high antiquity, as we have seen in the neighbouring lands,
cannot be attributed to Finnish and Esthonian ballads on the
evidence of their paganism alone. Lonnrot grouped pagan and
Catholic ballads together as if concurrent, and the founder of
Finnish literature, Bishop Agricola, drew up a list of deities still
reverenced by rural Finns and Carelians in 1551. In eastern
Esthonia pagan practices survived into the nineteenth century, and
pious Lutherans identified all folk-song with superstition. These
beliefs were doubtless older than Christianity, but without other
evidence we cannot say when they were embodied in folk-songs,
and particularly when they took the precisely defined form of the
extant ballads. They.are the gods and heroes of the Kalevala and
Kalevipoeg, and the wonder-workers to whom the magical loitsu-
runqja’ are addressed. There are Ukko the All-Father, also known
in Esthonia as Taara, and Jumala, whose name was taken by the
God of the Christians; Pikne, the god of thunder, Tuoni, of death;

298 NORDIC BALLADS

Vanamoinen, a water-god and god of song, who is the hero of the
Kalevala, along with the divine smith Ilmarinen and the sprightly
Lemminkainen; the giant son of Kalev, who is the hero of the
Kalevipoeg ; and many another. An indigenous name for Pikne is
Aike; Pikne, or Pikker, is the Perkunas or Perkuns of the Lithua¬
nians and Latvians and the Perun of the Russians, and he has also
the attributes of Thor. One Esthonian earth-spirit, Lijon, even
owes his name and fame to the Bible; he is the ‘Legion’ of Mark v.
9 an ^ Luke viii. 30. These deities are shape-shifters and wonder¬
workers ; they are shamans of vast power. The wizard has skill not
merely to deal with devils, to ‘conjure up spirits from the vasty
deep ; he alone can approach the gods and have power to constrain
them. Hence the practice of witchcraft was so firmly rooted in
Finland and Lapland as to arrest the attention of all superstitious
Europe. For all moments, embarrassments, and occupations there
were spells, which have remained to this day. The method is to
flatter an enemy, such as a bear, even after death 5 to know its
names and its composition, so as to be able to dissolve and dis¬
arm it. There are samples of such incantations in the Kalevala,
since Vanamoinen, god of music, was pre-eminent in wizardry.

The very act of singing was magical. Between chant and incanta¬
tion there was no frontier:

Let us go and frame our ditties,

knot the ancient words together,

set in order gifts of friendship,

float them with the river’s current,

roll them with the waves of ocean,

rear them with the precipices:

what from Gold they’ve listened unto,

what from Silver comprehended,

what revealed from Wanemuine,

what have learned from Taara’s wisdom. 1

The singers hold hands and sway rhythmically as they chant. The
practice is, doubtless, all that remains of a dance, but it is also a
method of incantation. There is wizardry also in the choice of a
magical stone for the singer to sit on, and in the old practice of
singing naked. _ With the beginnings of song in Finland and
Esthonia these incantations have much to do j but the present

1 F. Kreutzwald and H. Neus, Mythische und magische Lieder der Ehsten,
St. Petersburg, 1854, p. 24. ‘Wanemuine’ is the Esthonian form of Vanamoinen.

ESTHONIA, FINLAND 299

inquiry is more narrow. Folk-songs of one kind or another have
doubtless existed for miilenia, and they may even be, as Lonnrot
observed, as old as the human race; but ballads of the still extant
types have had a definite beginning. They may have been power¬
fully influenced by pre-existing magical runes., though it would be
almost impossible to prove this by dated examples. They have
also exerted a reciprocal influence on the forms of the incantations.
But the only sure evidence of their date and provenience are the
marks they still bear: the dates indicated by their historical allu¬
sions, the correspondence of their plots with those of neighbouring
lands, and the resemblances we have noted in their prosody.

These magical ballads throw light on the incantations in the two
national epics, and there are narratives also which relate epical
episodes. It is thus that, in the far North, we encounter again the
problem of epico-ballad relationships, and that under conditions
that seem to promise a definitive solution. This hope attracted to
their study one of the keenest minds of Europe, that of Domenico
Comparetti. 1 The first ballad reproduced by Lonnrot is Suo¬
metar 9 s Wooers. Suometar is bom from a duck's egg, and the
Moon, Sun, and North Star come to woo her; she refuses the Moon
as a wanderer and the Sun as capricious, but weds the North Star.
In this form the ballad is somewhat suspect, for it seems too tidy an
allegory to be traditional. Suometar means, being interpreted,
Finland's daughter; and, apart from the difficult mythological
business about the egg, one understands only too easily why Fin¬
land's daughter should marry the North Star. In the Kalevala
(xi, line 21 ff.) we have a more mythological presentation of the
same events: Kylli(kki) refuses Sun, Moon, and Star, but she is
not herself an allegory. Still more convincing is the episode of
Salmi’s wooers in the Kalevipoeg (i. 177-450), where Salmi is a
goddess, not a representative figure, and accepts the North Star
after refusing the Moon and Sun. Her experience was repeated
in that of her daughter Linda (recited in the same canto), who had
many suitors before she took Kalev, the father of the hero.
Similarly, the second of Lonnrot's ballads, Lyyiikts Snow-running,
corresponds to the forging of snow-shoes for Lemminkainen by
Kauppi in the Kalevala (xiii), when Lemminkainen wished to hunt
Hiisi's elk. In this way the scholar has both the epos and the ballad

1 D. Comparetti, The Traditional Poetry of the Finns, fcransl. Isabella M.
Anderson, London, 1898.

3 °° NORDIC BALLADS

still surviving in authentic forms, and the way appears to be open
for a definite judgement as to their relations.

Such a judgement cannot be given. The Kalevipoeg had to be
left out of the reckoning, according to Comparetti, because of ‘the
liberty which Kreutzwald allowed himself in the composition of the
poem (he even went so far as to versify prose stories and sagas) and
the ugly fact that he burned his manuscripts’. We are left with the
Kalevala, but we find that this is not an epos in any of the accepted
senses of that term. The Kalevala was constructed out of the
ballads by a scholar who was also a very fine poet; with the lightest
° f . t0Uctes he harm °nized many things said and sung about
Vanamoinen and Lemminkainen, and he found a place for many
incantations and bridal songs. With consummate tact Lonnrot
gave his poem a uniform texture, consecutive episodes, and a fine
Gotterdammerung’ in the last canto; but the Kalevala has no real
unity of action or hero or traditional transmission. The Kalevala
is definitely un-Homeric. Andrew Lang was overjoyed to discover
that Comparetti’s investigations gave no support to the theorists
who supposed the Iliad to be stitched together from pre-existing

T 1 , 1 ?, Rather ’ the instan ce of this poem shows that even the most
skilful stitching does not produce an epic. Contrariwise, of course
the comparison of Finnish ballads and epos will not serve those
who may hold the theory that ballads are epic detritus.

Jakob Hurt’s collection of Esthonian folk-songs shows that in
general they revolve around the same themes as those of Lithuania
and the later German balladry. They are chiefly songs of domestic
occasions or adapted to particular occupations. Hurt’s divisions
are into songs about songs, for girls, wooing, marriage, man and
wife, counsel, the unfortunate, work, jests, conscription, some
narratives, and miscellanea. Those who sing them are evidently
peasants and fishermen, as in Lithuania. A girl is advised to prefer
a ploughman husband to any one else, and an idle husband or wife
is a crowning misfortune. To offset this impression is the import¬
ance which the singers attach to the towns as providing the stan-
dards of their art; particularly notable is the place assigned to the
Latvian port of Riga. Marriage is, for peasants, the beginning of
a kmd of slavery. The husband and his folk are, as with the
Lithuanians, ‘strangers’; in one he is called ‘the bad one’. A
woman may even marry the Evil One himself (Vana Kannel 53,
54 )* Orphans arc assigned a sad pre-eminence in ballads of woe.

ESTHONIA, FINLAND 3 01

In all this there is great similarity between the various Baltic
balladries. If one should note the dearth of mythological matter
in Hurt’s collection, a reference to that of Neus corrects the
wrong impression. Compared with their neighbours to the south,
the Esthonians have a more narrative technique. There is no trace
of the epigrammatic Latvian style, which is probably of quite
recent development. There is parallelism, but not so ornamentally
arranged as in Lithuanian "dairies’; nor does the poet spare precise
details. The work is still, however, lyrical, save in the case of a
handful of narratives, since the details do not cohere into a story.

The art of the Finnish ballad is both more exact and more epical.
Lonnrot’s divisions are into "Songs for everybody 5 , "Songs for
different people 5 , and "Ballads and Romances 5 . These three sec¬
tions make up the Kanteletar , and they show the predominance of
domestic and occupational interests which we have noted in the
other countries. To these have to be added his other collections;
the Lottsurunoja or magical runes, and the Sananlaskut or proverbs.
Yet the ballads, though outnumbered, are well-articulated narra¬
tives in a technique that shows no sign of ageing. With the Finns of
Karelia one creative impetus reaches its utmost bound: along the
shores of the White Sea. Beside the Arctic Ocean the Lapps have
an equal reputation for wizardry, and they sing ballads in a ruder
style, but employing a technique often recognizably Finnish.

Ill

BALKAN BALLADS
i. Asia Minor, Greece

ARETHAS of Caesarea (850-932), commenting on the word
. ^ a gy rtes . noted that it means ‘mountebanks’, such as, for
instance, the accursed Paphlagonians who contrive songs on the
experiences of distinguished persons, and sing them, at an obol
apiece, round the houses. 1 It is in this casual manner that ballads
cross our view not merely for the first time in Greek lands, but in
a Europe. The pedant, picking up straws of information and
. °, m them like a jackdaw, has preserved unsuspectingly a
jewel. The themes of these songs, the humble style, and the man¬
ner of performance indicate unmistakably ballads in existence in
the early tenth century. The minstrels are Paphlagonians, or, as
we might now prefer to say, Cappadocians, since the word
Cappadocian is attached by modern critics to all Akritic verse.

hey were not Greeks of the European mainland, still less were
they the cultured citizens of Constantinople. In Constantinople
the ancient literary culture had not died, though it had dwindled.

e anthologies continued to receive epigrams in classical elegiac
couplets which ignored the change in the vernacular from tone-
accents to stress. There was no want of grammarians in the Greek
capital to check deviations from the established norms of speech
New themes and passions forced their way into verse, and the
politic measure arose for the use of chroniclers and court poets •
but its use was controlled in the capital. It was only on the far-
ung frontiers that an entirely traditional poetry was possible: in
Paphlagonia and Cappadocia and on the bend of the Euphrates.

On the frontier medievalism had begun while the great cities
still clung to the skirts of the ancient world. A kind of feudalism
had been set up. The Saracen peril was ever present; the borderers
could not wait for the cumbrous mobilizing of the imperial armies,
but had to defend themselves against sudden forays, often by
initiating raids of reprisal. The remote intrigues of the capital were
not their affair, though dynasties might rise and fall. Their
independence was accentuated by the sectarian revolt of the

Grego"irf yiiakldeS ’ <EUenike Lao Sraphia, Athens, 1923, p. 81. Quoted by H.

ASIA MINOR, GREECE 3 c

Paulidans in the eighth century, which the emperors drowned i
blood only alter a tradition oi line generalship had been set up b
the Paulician commander Corbeas. Only strong emperors—th
two Basils ana John 1 zimiskes—appeared on this borderland, th
Percies of which were the powerful Doukas family. On the othe
side also, it was not from the distant caliph of Baghdad that peac
and war were to be expected but from the more imminent power o
the emirs of Edessa. No general plan governed the war. Then
were daring razzias, brilliant single-handed achievements in wiiicL
a man might owe everything to his armour or his horse, ambuscades,,
and sudden eclipse. Such were, no doubt, the ‘experiences of dis¬
tinguished persons' which the accursed Paphlagonians touted front
door to door, at an obol apiece, in the first years of the tenth century.

To all such poetry it is customary to assign the epithet ‘Akritick
Digenis Akritas was a historical personage who perished in battle
some ten years after Roland and his peers died at ,Roncesvall.es.
H. Gregoire 1 has recognized his name in that of Diogenes, tur-
march of the Anatolians, who perished at the skirmish of Kopid-

. ^don in 788. The site was probably a defile (kome Podandos) of

the chief pass leading from Ciliciai nto Cappadocia. Popular tradi¬
tion assigned to him as a place of burial the tomb constructed for
one of the Commagenian kinglets on an abrupt hillock overlooking
the Cappadox or Gok-sii, which flows into the Euphrates near
Samosata. His mother was of the house of Doukas, and his father
an emir. According to Gregoire, this emir must have been the
Abu-Hafs, grandson of e Umar al-Nxriman (who perished gloriously
in 863), who carried his tribe of the Banu-Habib over to the Greeks
in 928. So, for Diogenes, we get Digen.es (pronounced Digenis or
Dienis) Tom of two races". The hero’s personal name thus became
a nickname, and in the epics he is called Basileios. A further by¬
name was Akritas 'the frontiersman", a title which he shared with
such other champions as Porphyries, the Farfurius of the Persian
epics. Scarcely distinguishable from these defenders of the
frontiers were the Apelates (Philopappos, Kinnamos, Ioannakis),
who were rievers and robbers. Their names appear in the ballads,
but their exploits were in a large measure annexed to the glory of
Digenis. His feats of strength were those historically associated

1 H. Gregoire, ‘Le tombeaix et la date de Digenis Akritas* and ‘Autour de

Digenis Akritas*, Byzantion v-vii, 1929-32; S. Kyriakides, 'O Digenes Akritas
(Syllogos pros diadosin ophelimon biblion, 45), Athens.

3°4 BALKAN BALLADS

with the name of the emperor Basil I, of titanic memory; and his
weapon—a club—may have been that of Hercules. Another name
which appears in the Akritic ballads is that of Tsamados, otherwise
unknown; there is a Kimiskes derived from the emperor John
Tzimiskes (969-76), a certain Skleropoulos or Syropoulos (i.e. a
Syrian), and Amouropoulos, a hero of Amorium (captured in 8^8)
Above all these names, however, that of Constantine prevails in the
ballads, taking the place of the Sweet William of Scotland or the
Stojan of Bulgaria. He is generally called Constantine the Little
Mikrokonstantmos, or Kostantas or Kosta. ’

Digenis, a hero of both ballads and an epos, 1 was so well known
in one or the other capacity that the emperor Manuel Comnenus
(^ i 43 _ 8o) could be addressed as ‘the newAkritas’. Asthepanegy-
rist, Theodore Prodromos, goes on to mention the hero’s club, his
allusion is satisfied rather by the epic poems than by the ballads.
Akntas club spoils many a promising encounter; he bludgeons his
enemies before his admirers can savour the piercing emotions of
doubtful battle. His crudity is the more strange because the poet
has read some of the best epic models. He quotes textuaily from
the Iliad and the Bible; he knows either his Quintus Curtius or one
of the Alexander-romances; he has enjoyed Greek novels, and cites
Aldelaga and Olope. The Greek is tolerably accurate, especially in
the Trebizond manuscript, and the whole poem is sub-literary
rather than oral and traditional. Yet its conduct is a series of
failures; the poet continues to outline dramatic situations, which
he continues to foozle through sheer lack of gift for nar rative. It is
the most disappointing of popular epic poems; for there is no doubt
of its popularity. Its influence has spread to Russia, Arm enia, and
the Syrian Arabs, and the Turkish hero Sayyid Battal is a Wm . n
m spirit of Digenis Akritas. 2

The first three books detail the romance of Digenis’s birth: his
mother was carried off by the emir Mousour during an absence
from home of her brothers; they followed the raiders, upbraided
1 BaHleios Digenes Akrites, ed. C. Sathas and E. Legrand (Collection de
AtwT P T S£rVlr j r€tUde de Ia lan & ue nSo-hellenique. nouv. ser. 6)
Thfmt PanS ’- I f 7 ?’ A re T° duCeS a fifteenth-century manuscript from Trebizond.
tree t SCn a of ^ros is allied to this by their inserting in the hero’s family
AeGrotfefr 11 ldentlfied “ a duke ° f Edessa in the eleventh century. In

‘ ArS™ manuscript Digenis’s maternal grandfather is said to have been

e^bthThe^ r K ™l d6n ’ A VariorUm editi0 * is nee ded in order to
establish the relationship of the various manuscripts.

21 See Note O, p. 391.

ASIA MINOR, GREECE 305

the captor, agreed to a match and reconciliation, and so provoked a
spectacular conversion to Christianity of the emir and all his tribe.
This section may be described as the ‘geste’ of the emir. Two
books then describe Digenis’s prowess as a hunter, modelled on
that of the emperor Basil I as given in Gesenius, Liber Regum. The
date of this work is not quite certain; it does not appear to have
been available until the second half of the tenth century, and so gives
a date ‘a quo’ for the formation of the epos. The hero’s success¬
ful elopement with Eudoxia Doukas is also given, in this section.
There follow two books of the hero’s reminiscences: an adventure
with the daughter of Haplorrhabdis, and the defence of Eudoxia
against the amorous Philopappos and the Amazon Maximo. It was
a grievous mistake to relate these things in the first person. An
elaborate account of the castle and paradise built by Akritas on the
banks of the Euphrates occupies the eighth book; which is followed
by an account of his mother’s illness in the ninth, and his own death
(with Eudoxia’s) in the tenth. Though not a ballad sequence,
this disjointed narrative may well be the result of attempting to
reproduce the episodes of a ballad cycle in the grand manner.

The priority of the ballads over the epos is taken for granted by
modem scholars. They contain incidents not to be encountered in
the longer poems, and they relate the common episodes in quite
another fashion. They cannot be fragments mechanically pro¬
duced; nor have they been stitched together by rhapsodes. It
would appear rather that they have been harmonized, rationalized,
and subjected to capricious selection and alteration. This treat¬
ment was meted out to them, according to Gregoire, before the
year 944; since the Grottaferrata speaks of the Sacred Face as being
at Edessa, whence it was transferred to Constantinople in that
year. The argument seems doubtful. There is no reason why a
poet should not place in Edessa a relic known to have been there in
his hero’s days; and the date cannot be squared with the use made
of Gesenius. A cautious scholarship would be satisfied with the
twelfth century as the ‘terminus ad quem’, guaranteed by the
compliment paid to the emperor Manuel Comnenus. One
complication is that the names of ballad heroes permutate, and the
same action is current under several names. Even in the epic it is
clear that one of the author’s hopes is to eclipse anything that may
be said or sung of the leading Apelates, who doubtless had their
own ‘gestesh H. Gregoire has offered identifications. Philopappos

3°6 BALKAN BALLADS

he would take to be Antiochus Philopappos, the last king of Com-
magene, whose tomb was erected at Athens between a.d. 114 and
116, and what was said of him constituted an ‘epopee commage-
nienne . If this identification seems adventurous, there is at least
some evidence that the original hero of a ballad of bride-stealing was
Philopappos, not Digenis. The exploit is simply transferred by the
epic poet to his own hero; in some ballads Chiliopappos (who is
Philopappos) figures as a go-between, and so as a friend of the hero
whose constant enemy he is in the epos. Kinnamos is identified as
the rival of Artabanes III of Parthia (Josephus, xx, 3, 2), whose son
Gotarzes is a hero of the Shahnameh. Farfurius of the Shahnameh
is the Porphyrios of the Greek ballads. All these matters are very
obscure, but they amount in sum to evidence of a considerable
burst of creative energy in an Akritic age of Greek literature.

The epos helps us to set in some kind of order the surviving
Akritic ballads, 1 though not to make absolutely certain identifica-
tions. The brave Girl and the Saracen (Politis 72 a) is a stirring
ballad of chase and capture: the girl is an Amazon chased by a more
powerful Saracen; she hides in an icon of St. George, and only
surrenders when her lover turns Christian. The Saracen takes the
name of Kostantes and says their son will be called Iannes. If this
Amazon is to be identified with Digenis’s mother, one cannot but
note the discrepancy between the ballad and the epic treatment.
A closer parallel is the ballad entitled Andronikos’ Son, of which
the Trebizond version is entitled Porphyrios (Sathas and Legrand,
Basileios Digenes Akntes). This concerns a youth who grows
prodigiously. His mother has been stolen by an emir when
enceinte; she is said to be the wife of Andronikos, and yet crypti-
cally declares the child to be the emir’s son. He grows prodigiously:

One year of age he takes the sword, at two a long lance seizes,

and when his years were close on three, was a pallicar renowned.

1 S. Kyriakides, 'O Digenes Akrites, Athens, no date; N. G. Politis, Eklogai
apo ta Tragoudia tou 'Ellenikou Laou, 3 ed„ Athens, 1932 (which omits some of
the most characteristic); S. Baud-Bovy, Chansons du Dodecanese, i, Athens,
1935 . n, Paris, 1936 (with valuable notes); P. Arabantinos, Sylloge demodon
asmaton tesEpeirou, Athens, 1880; E. Legrand, Recueil des Chansons populates
grecques (Coll, de Mon. pour servir a l’etude de la langue neo-helldnique),
Athens-Paris, 1874; Lucy M. Garnett and J. S. Stuart-Glennie, Greek Folk-
Eoesy, 1, London, 1896. A. Passow’s Carmina popularia Graeciae recentioris,
Leipzig, i860, contains no Akritic pieces, though it is otherwise a standard of
reference. Fauriel’s Chants populaires de la Grece moderne, Paris 1834-8, is the
pathfinder to these studies.

ASIA MINOR, GREECE 3°7

The Saracens cannot bind him, and he sets out to seek Ms father.
At the end of his first month Porphyrios was singing in the streets
of the capital:

O I have loved a maiden free, I love the king’s dear daughter—

which caused considerable annoyance at Constantinople (or Nicaea
according to another version): this affair probably had nothing to
do with Digenis, whose youthful feats are copied by the epic poet
from Gesenius’s De Regum.

His suit to Eudoxia was opposed, according to the epos, by her
family. A ballad (Politis 74) tells how Mikrokonstantinos sent
Chiliopappos, Phocas, Nikephoros and Petrotrachelos (or Tro-
mantacheilos) as his go-betweens, only to be rebuffed by Liogen-
nete’s mother. He penetrates to her chamber in woman’s clothes,
like Hagbard, and the episode ends in a double tragedy. In the
epos, the family gave chase, and were taught by hard blows to
recognize in Digenis a worthy son-in-law. In the epos, Philo-
pappos makes determined efforts to steal Eudoxia, both by his own
might and by invoking the aid of the Amazon Maximo. In the
ballad the hero speaks in person (Politis 75) :

Now as I drank and banqueted beside my marble tables,
my coal-black steed began to neigh and sword in scabbard rattled,
’twas then my heart bestirred itself to court my tine beloved.

The thought was well timed, because only the swiftness of Ms grey
mare enabled the hero to reach her before she married another
husband. Some versions call the hero lannakis, and this ballad is
one of the Moringer cycle, like the Serbian Marko frees his Beloved
and other ballads in Serbian, Bulgarian, and Rumanian. In Syro-
poulos or SMeropoulos (Kyriakides) we encounter a youth who sets
out to steal his uncle Kostantas’s bride. He persists despite the evil
omens; putting his uncle out of action, he seizes the lady and flees.
Kostantas picks his swiftest horse, obtains news from a shepherd,
cuts his way through an army on the frontier, and cuts off the
ravisher’s head. This would appear to represent the genuine
Akritic tradition. The victim was probably lannakis. A somewhat
similar pursuit is related in the ballad of Amouris or Amouropoulos
(Kyriakides). The young Amouropoulos begs his mother to allow
him to pursue the Saracens. She agrees, if he can wield Ms father’s
arms. He is able to do so, and hurries to the Euphrates. Arrived

3 °S BALKAN BALLADS

there, he would have had to suffer the taunts of the Saracen without
retaliating had not an angel helped him to leap the stream He
destroys an army, but a spy steals his horse. The horse and trap¬
pings are recognized by his father, who is a prisoner in the Saracen
camp at the Cilician Gates and is forced to write a letter for¬
bidding his son s advance. Amouropoulos answered only with
threats, and the emir preferred to be reconciled to so doughty an
adversary by offering him his daughter. The ballad has been ex¬
plained as a sublimation of the sack of Amorium in 838, showing
both the furious Greek reaction to the defeat and the policy of ap¬
peasement on the frontier. It may be so, though the historical
touches are slight; but it is undoubtedly an old ballad that con¬
templates a frontier on the Euphrates or at the Cilician Gates.
Beauty's Castle (Politis 73)—its vain defence by a princess—is
also vaguely Akritic.

Legrand was able to quote a brief fragment, Akritas built a Castle
which corresponds to the eighth book of the epic. It is a character¬
istically eastern ballad, and there are versions in the shorter lines
usual in Pontic poetry. Then follows the hero’s death (the ballads
have nothing to say concerning the death of his mother). It is in
two parts or aspects: The Wrestle with Charon and Eudoxia’s Death.
The former is a famous piece, and occurs both in Akritic guise and
in later generalized forms (Politis 78, Sathas and Legrand, Legrand
214-16, &c.). As the medieval Hercules it befitted Digenis to
emulate the greatest achievement of the Greek god. Digenis was
victor but soon lay on his bed in agony;

So Digenis does agonize upon his iron bedstead,

and, books in hand, the doctors take their stations round about him.

And now uplifts his dying head, and calls for his beloved:

Sit here beside me, pretty one, sit here beside me, darling;
for of the years IVe lived on earth the tale is three and thirty,
but now the messenger has come to reave from me my spirit/

He grips her by her two fair hands, a thousand kisses gives her,
he grips, he squeezes in his arms; she is crushed and suffocated.

So died Eudoxia. In the more generalized versions of the first
ballad current in Europe (Politis 214-16), the youth is a peasant
or soldier, who trusts to his strength and is speedily overthrown by
Charon. A more poignant counterpart to this scene is Eugmoula
( 0 ltis 2x7), who thought because she was young and just married
she need not fear death:

ASIA MINOR, GREECE 309

J Twas Eugenoula, pretty child, so recently she married,
she boasted as she went her ways that Charon little feared she;
so high the roof-tree of her house, so valiant her husband,
and all her bretheren nin.e so stout, the vanquishers of castles,
they every castle overcame till every country trembled.

But Charon, when he heard that word, so deeply did it grieve him.
To a black bird lie changed his shape, like unto a wild swallow,
he sent his arrow from his bow against that lady lonely,
he struck her on her finger fine, upon her fine ring-finger.

Three other ballads remain to be cited in this Akrltic group,
though they have no connexion with Digenis. They are Tsamados
and Ins Son, The Bridge of Art a and The Dead Brother's Return or
Constantine and Arete (Politis 77, 89, 9a). In the first we are told
that the brutal Tsamados descended the hill and forced a deadly
wrestle upon his son. The legend belongs to the same class as
Ulysses and Telegonos, IPja and his son, Sohrab and Rustum,
and Hildebrand; it stands closest to the Russian version inasmuch
as it condemns the father for a tragic situation generally regarded
as caused by blind fate. The second legend was first told concern¬
ing the bridge of Adana, in Asia Minor, and only at some latex date
transferred to Arta in Acarnania. It is a masonic legend. The
bridge will not stand, however the builders toil, until the master-
builder's wife is enclosed in it alive. The oldest form of this ballad
is Cappadocian, but the best development is in the Rumanian
Master Mamie or ike building of Arges , where it is connected with
the name of a seventeenth-century Greek architect and a Rumanian
shrine of much earlier date. The Rumanian version is superior,
because it conveys supremely the sense of helplessness in the face
of fate ; no prayers or entreaties, not even rain and storm, will pre¬
vent the devoted girl from advancing to her doom. The third is
a powerful ballad, which has spread as far as the gypsies of Hun¬
gary in one direction and England in the other (The Suffolk
Miracle). Arete was an only sister of nine brothers. When a wooer
came from a far country", her mother was unwilling to let her go,
and only consented when Constantine promised to bring her back
when needed, at all costs. Plague swept away the nine brothers
and left the mother sick to death. Constantine rose from his tomb,
compelled by his oath, and rode to seek Ms sister; he brought her
back, and after so tremendous an experience both she and her
mother fell dead. The odour of death rests on all the poem; the

330 BALKAN BALLADS

listener sympathizes anxiously with the girl’s restless inquiries
which her dead brother does so little to calm and satisfy :

Along the roads as they passed by the little birds were twittering,
but twittered not with voice of birds, nor yet like unto swallows,'
but twittered they and spoke in words as men do speak together:
‘Now does she know, the pretty maid, that a dead man conducteth? 5
‘But do you hear, my Constantine, what things the birds are saying?'
‘Just birds, so let them twitter on; mere birds, so let them chatter.'
And further, as they went their way, yet other birds addressed them:
‘Surely it is a crime, a wrong, a strange and mighty wonder,
that living men should go their ways in company with dead ones!'
‘But do you hear, my Constantine, what things the birds are saying ?
how living men do go their ways in company with dead ones?’

°Tis April now and so they talk; 'tis May, and nests a-building.'

‘I fear me greatly, brother mine, with unguents you’re anointed.'

‘Nay yester even as I went into St. George’s temple,

the reverend father sprinkled me with overflowing unguent.’

And on they passed and on they went and other birds addressed them:

‘Now all the world is gone awry, and ’tis a mighty wonder,

that such a pretty maid should be she whom the dead conducteth.' 1

Such ballads are Akritic because they go back to the Akritic age,
which is the oldest stratum of European balladry: an age when the
Greek frontier was on the Euphrates and the Saracens were their
enemies, before the Seljuqid Turks have appeared. They are Cappa¬
docian, because found in that province, or Paphlagonian, as Arethas
preferred to say, or Pontic. Removed so far from European Greece
they escaped the eyes of the first collectors, who found only the
later klephtic ballads of the seventeenth century and onwards;
and they have only in our own day been brought within the easy
reach of scholars, thanks to the Greek migrations which accom¬
panied the triumphal advance of Kemal Atatiirk. Old modes of
music and unexplored variations of the words enchant the col¬
lectors, who are actively gathering the remains of this ancient
poetry in the Greece of to-day.

The Akritic ballads are separated from the klephtic by the depth
of the Turkish invasions and the breadth of Byzantine literature.
The Greek capital was hardly favourable soil for the growth of
balladry. Peculiarly literate, the flood of classical manuscripts
which fecundated the Renaissance in Italy represented but a back-

1 J. E. Flecker’s Bryan of Brittany is a beautiful adaptation of this ballad in
the English style.

ASIA MINOR, GREECE ?n

wash of those the Greeks habitually enjoyed. The political fac¬
tions, it is true, required verses for their own purposes, but such
were rather "vulgar than traditional. It was probably in the capital
more than elsewhere, however, that the new accentual prosody
arose, and in particular the ‘politic’ metre. This metre might be
represented by the line

The King was in his counting house, counting out his money.

The line is crudely garrulous, but the rhythm is easy and lends
itself to improvisation. Its emergence is not easily explained. The
iambic tetrameter catalectic, scanned by quantity, was less used by
the ancients than many other verse forms, so that its sudden rise to
popularity is surprising. The passage from quantity to stress would
not depend on the rare cases in which the ancient accent coincided
with the quantities, but no other intermediate stages have been
observed. In the Comnenian age we know that the metre was used
for lampoons circulating in the streets of the capital, but its chief
employment was for chronicles and panegyrics, for wdiich it was
suited by reason of its colloquial ease. This w r as not the only metre
of the Greek ballad-mongers. An accentual iambic trimeter
appears in many; dimeters and many other verse forms were at
the disposal of song-writers in a more lyrical vein. All this poetry
is unrhymed, but rhyme is occasionally encountered in late work
from the eighteenth century and after, probably due to Venetian
teaching.

A four-line fragment survives to satirize the flight of Alexios
Comnenus in 1081, and six lines of politic verse refer to some
siege of Adrianople (Politis i):

The Anatolian nightingales and all the birds of Westland,
they cry at dawn, they cry at eve, they clamour in the noontide,
they cry for Adrianopolis, that is so sorely battered,
that enemies assaulted her during three holy feast-days:
upon the day of Jesu’s birth, and thence unto Palm Sunday,
and bright Easter Sunday too, when Christ was resurrected.

This is the oldest ballad of European Greece, so far as we know,
though it is little more than a cry of distress. The occasion is un¬
certain. The Bulgarians thrust their forces into the city in the
Easter of 1205; in 1353 and 1361 the city fell before the Turks, but
the precise days are unknown. The concensus of critical opinion
is that the ballad refers to Amurath’s siege of 1361. The great

312 BALKAN BALLADS

disasters of the fifteenth century are recorded in laments of this
sort: the fall of Constantinople in 1453 is mirrored in Constantine
Dragazes Death, which depicts his heroic despair, and in a ballad
on the saving of the relics in St. Sophia (Politis 2). The fall of
Trebizond and Poliokastro in 1461, and the heroic defence of
Kordyle by a woman, follow hard upon these pieces. The conquest
of Greece was not completed until a century later. Skanderbeg
held out in Epirus and Albania, and a ballad speaks of a general
Epirot rising against the Turks in the decade following 1565. The
period was brought to an end by the battle of Lepanto in 1571.
The Sea-fight and the Slave speaks of Don John’s triumph (calling
him Regas) and the death of Ali Pasha. The battle saved the west¬
ern Mediterranean lands from Turkish invasion, but the attempt
to carry the war into the eastern basin resulted in a stalemate. The
hapless Greeks were abandoned by the westerners, and the Turks
had leisure to devour them.

The Greek resistance, no longer national but personal and local
was the work of Epirot klephts. The word means no more than
‘robber’, so low had freedom sunk. The importance of the
klephtic ballads as the nurse of Greek independence cannot be
over-estimated. Their poetical worth is often slight. They were
little more than bulletins giving news of the revolt of some leader,
a successful raid, a lucky escape, and the inevitable capture and
death of the klepht. But so long as such news circulated, Greeks
could not forget that their mountains at least were free. The
oldest piece of klephtic verse is probably that which celebrates
Malamos, who revolted at the instigation of the Venetians in 1585.
In the seventeenth century klephtic ballads remain rare: one hears
of a certain Bishop Serapheim of Phanari, who was unjustly
executed in 1612 on suspicion of complicity with the insurgents,
and of Nikolas Tsouvaras, who made an attack on Louro in 1672
and carried off the local commander. From the last quarter of the
century the succession of klephtic ballads extends without interrup-
tion down to the War of Independence, and we hear the names of
Metsolsos, old Mpoukovalas, Demakes, Liazes, Ntritzas, Tsoulkas,
young Mpoukovalas, Milionis, Kolokotronis, Botzaris, and many
others.

As the eighteenth century drew to a close the klephtic resistance
took more ambitious forms. The decadence of the Sublime Porte
was evident; the rise of Russia brought a new great power into the

ASIA MINOR, GREECE 313

eastern Mediterranean. Relying on the promised aid of Catherine
II* a certain Daskaloiannes (Master John) raised 0 revolt in Crete
in 1770; but he was deceived in his hopes, taken and put to death.
In the ballad the Turks taunt him with his Greek patriotism. The
operations on both sides became more elaborate as the turn of the
century approached. Ali Pasha of Iannina, himself of klephtic
descent, undertook punitive expeditions at the head of Turkish
regulars and Albanian spahis, and the resulting conflict could be
deemed a war (the Souliot War). He was surprised and defeated
by Botzaris in the defiles near Souli on 20 July 1792, and the Greek
ballad is a shout of triumph;

Then Botzaris upraised his voice, his right hand shook his falchion:

Stay, Pasha, stay; why sneak away? why flee among the fleetest?

Come, turn again into our town, turn back to empty Kiapha,

and set you up your royal throne, and make yourself a sultan!’

(Politis 5.)

There is a longer ballad on the defence of Missolonghi and others
follow down to the arrival of King Otho and the Treaty of Berlin.
New ballads have arisen at even later dates In the islands, especially
in Crete. The Cretan Alidakis relates his feud with the local
Turkish ruler in some thousand lines of vigorous narrative, but
pedestrian verse.

The klephtic cycles were once quoted for the light they might
throw on the composition of the Iliads based on hypothetical
ballads of Troy. Their irrelevance in this respect is now generally
recognized, and they have lost caste in the eyes of scholars because
of their patent modernity. Yet they have their niche in ballad
history. They fanned the flicker of national spirit in the most
desperate age. They link up with the haiduk ballads of Serbia,
Bulgaria, and Rumania, both in themes and in sites. Those of
Rumania derive from Serbo-Bulgar models, and the best of these
belong to the western mountains which extend southward into
Epirus, the favourite haunt of the Greek klephts. On the Albanian
frontier, Greeks, Slavs, and Albanians mingled in feud or amity.
Metsoisos, Ali Pasha’s. great-grandfather, revolted in 1690 with
three thousand klephts and Albanians; his ballad is Greek, but it
is popular in Albania. A Greek ballad commemorates the prowess
of the Albanian Liazes, who led a mixed company of Bulgarians,
Albanians, and Ylachs into Macedonia, Thessaly, and Epirus. On
the other side, the Turks relied on Albanian renegades to crash the

3*4 BALKAN BALLADS

revolted Greeks. It is not surprising, therefore, that the Mephtic
and haiduk ballads of all the Balkan peoples should form virtually
one corpus, uniform in inspiration, episodes, and style. Constitut¬
ing a homogeneous mass of verse, it is they that give each of these
countries a ‘romancero’.

Their literary interest, however, lies not so often in datable
bulletins as in those of a more general sort which describe the
vicissitudes of an outlaw’s life. It is a defiant and a romantic pose:

O let the plutocrat have gold, the poor man have good humour,
let others praise the pasha’s state, let others praise the vizier;
but I will only praise the sword that slaughters Turkish foemen,
for youthful courage has its boast, the klepht his reputation.

(Politis 20.)

The klepht is free from every tie. Only Nature bounds his freedom,
since he can roam the mountains from May to August, but must
descend to the plains when the snows creep down the flanks of
Olympus and Kissabos, risking death or torture. Nature sympa¬
thizes with the klepht, encourages his revolt, and mourns his death:

Olympus—see!—and Kissabos, two mountains in a quarrel:
which shall pour down the heavy rain, which shall pour down the
snowstorm.

’Twas Kissabos that sent the rain, Olympus sent the snowstorm.
Olympus turned his mighty head, to Kissabos thus spake he:

‘Nay, scold me not, Sir Kissabos, by feet of Turks betrampled,
for I am Olympus full of years, in all the world renowned,
and forty-two my summits are, and sixty-two my fountains;
and on each peak a banner free, ’neath every branch an outlaw.’

(Politis 23.)

Not only the hills are at one with the klephts, but so are all free and
noble birds and beasts.

There was an eagle seen afar, there was an eagle youthful,

who was a klepht. Three partridges foretell his chances, and three
clouds rest on Karpenisi for his sake. Devices of this kind become
monotonous by repetition, but they must have been invented by
some one with an eye for nature, and they sometimes appear in
entirely adequate contexts. So we see the klepht at ease under his
plane tree. We are present at rapid night-marches and halts during
the day, at careless feastings, dances, and wrestling-bouts. The
approach of bitter winter is only less awesome than the oncoming

ASIA MINOR, GREECE 3 i 5

of helpless old age. The klepht must say farewell to the mountains;
the trees,, twigs, springs, and dens mourn for the captured outlaw,
and he dies In the midst of cruel enemies or on a naked hillside.
The happiest death was that which came sword in hand; and lest
dishonour should come even after death, klepht and Turk struggled
over the corpses in the Homeric fashion:

Androutsos cried a bitter cry, a bitter cry, empoisoned:

Come, children, grip your trusty blades, and cast away your muskets,
lest Turkish dogs should carry off the head of our own Vlachos,
an outlaw old and gray was he, among the klephts a captain.’

(Politis 52.)

The ALritic and klephtic poems embrace all the historical ele¬
ment in Greek tragoudia’, together with fictions and occupational
songs of a certain kind. There remain for discussion the adventure
ballads of a more modem type and other occupational or occasional
ditties. Some of these show foreign influence. Manrianos’ Sister
(Politis 81) belongs to the cycle of Mariansori s Rings and the
betrayal of Imogen. The Greek text shows a number of Jatinisms
(mandata, foro, furca, kadena) which, though common enough, are
pointers towards the Italian origin of the ballad. In handling the
story the Greek ballad stands closer to The Twa Knights ( Child°268)
than to any other version. The Sicilian legend of Cola Pesce
appears in Greek dress as The Diver (Politis 90); and The Infanti¬
cide (Politis 91) is a horrid tale of adultery and murder which is cur-
rent in Sicily (Pitre 910), but might aswell be Greek as Sicilian.
The process of commingling Italian and Greek ballads is most
thoroughly carried out in the Greek-speaking townships of
Southern Italy. Greek tragoudia are current there and as far
west as Corsica, 1 but the majority of pieces are improvisations
in the South Italian style of ‘canziuni’ and ‘ciuri’, though in the
Greek language. One may follow more exactly the course taken
by The Girl who went to War. It is a folk-song of Venice, as well as
other parts of Italy, and crossed with Venetians to Dalmatia. There
it was taken up by the Yugoslavs, who fashioned a ballad of Old
Ceivan's Daughter, and by the Greeks, who formed the folk-tale of
The Princess who went to War. The Albanian Silver-tooth 2 is a

1 Professor W. Starkie has encountered Greek gipsies who relate the Bridge
of Aria legend in Morocco. The singer, however, was a Rumanian by birth,
and her source seems to have been the Rumanian ballad of Master Manole .

2 I* G. von Hahn, Greichische und albanesische Mdrchen , Leipzig, 1864, 10
(Greek),. 101 (Albanian); cf. A. Dozon, Contes albamis, Paris, 1881, No. 14.

3 i6 BALKAN BALLADS

further development of the same story in the same geographical
area.

On the other hand, we have to allow for the force of expansion
of the Greek inventions. In the Balkan peninsula the province of
Macedonia has especial significance as the meeting-place of Greeks,
Bulgars, Serbs, Albanians, and Vlachs. It was also the region in
which the war-like descendants of the Paulicians were settled by
medieval emperors after the overthrow of their power in distant
Cappadocia. The Cappadocian ballad of The Bridge of Arta , once
located in Macedonia, begat the Rumanian ballad of Master Manole
at some date after the end of the third quarter of the seventeenth
century. The essential difference in this version is that the identity
of the victim is not revealed, but left to chance. The name Manole
first appears in Greek Macedonian ballads. Mano or Manoil is the
hero of Bulgarian ballads (Miladinov 424) from the Rumanian. In
Serbia the choice of victim was left to fate by three brothers, but
two of them conspired to keep their wives from coming (The
Founding of Skadar, Karadzic, ii. 25), which shows the influence
of the Rumano-Bulgar version; and the Serbian plot appears in the
Albanian Bridge of Dibra (Dozon, Contes albanais, Supplement:
‘Le Pont du Renard a Dibra’). On the other hand, Constantine and
Arete seems to have been borrowed by Rumanians, Bulgars, Serbs,
and Albanians separately; they do not pass on names or incidents
from one to the other. In Bridesmaid into Bride (Politis 83) we have
a ballad which has effected a lodgement in the Spanish Romancer0}
It is encountered in Catalonia earlier than in Castile; so that it is
relevant to remember that the Catalans of the Middle Ages lived
beside the Greeks in Corsica, Sicily, and Southern Italy, and they
also for a while—though probably too early for the borrowing of
this ballad—ruled the principality of the Morea. The main feature
of this ballad is the dazzling description of the bridesmaid’s beauty:

The chapel looked upon her face, from end to end it shivered.

The priest he looked and lost his place, gaped open-mouthed the
deacon,

fumbled the psaltery the choir, the choir and the precentors.

But when the bridegroom saw that maid, he fell to earth a-swooning;

said: ‘Priest, ring backwards all the bells, begin again the singing,

take back again the marriage-crown, to set it on the bridesmaid.’

Other contacts between Greece and the west include The

1 See my article on ‘La Dama de Aragon’ in Hispanic Review , vi, 1938.

ASIA MINOR, GREECE 3*7

Love/s Return (Politis 84) and the tests he makes of his wife’s
fidelity, and The Luckless Marriage (Arabantinos 471), which Is a
distant offshoot of Elveskud. From the Ukraine In the eighteenth
century came a version of Mazeppa’s ride entitled by Legrand
The Husband’s Vengeance . A ballad of Captain Manetas, In 1780,
describes how he went to the Black Sea ports on hearing a report
that his wife was unfaithful, and how he sent back orders for her
murder. The ballad Is evidence of Greek contact with the Ukraine;
at the same time it betrays the Influence of foreign (Italian) aesthetic
preferences, since It is a Tima 5 in twelve-syllable couplets.

The greater number of the Yragoudla’ have no obvious attach¬
ments outside Greece. They cover the usual grounds of folk-
poetry : love, tragedy, crime, adventure. There is tripping of feet
In the charming Zerbopoula (Politis 99), or Syropoula, as others say.
A king comes unexpectedly upon a round of dancing girls, and he
loses his heart to the fairest of them:

The golden damsels danced before, the brown girls danced behind
them,

and In the midmost of them all, fair Zerbopoula tripped It.

They danced the Thoms’, which is the Bulgarian *horo\ the
Rumanian ‘hora’, and the Serbian £ kolo’. The Greek ancestry of
all these dances is evident in the words; but it is also possible that
the Thoros’ is the source of the old French ‘carole’, and so of carols
and danced ballads. The simple movements are carried out by
women, who leave to men feats of virtuosity. In Greece the dance
does not lead to a distinction of kinds in folk-poetry, as it does in
the opposition of women’s songs to men’s songs In Yugoslavia;
but it does give us yet another glimpse of the part women seemed
to have played In the restoration of love-poetry to Europe.

The Greek love-poems are very numerous, and they indulge In
pretty fancies. They speak of trees that flourish just so long as
lovers remain faithful to the vows spoken under their branches*
A handkerchief or a mirror Is a girl’s counsellor and confidant*
Lovers’ secrets are never inviolate, but the star tells the sun, the
sun tells the sea, the sea tells the oar, the oar the sailor, and the
sailor all the world. There are metaphors from fowling and hunting
applied to the sport of love. Lovers are tested: they are sent down
wells to get a ring or must lift a marble block In a garden. The
lover pursues his girl through many transformations of shape.
Then follow the changes of fortune; quarrels, desertion, the over-

318 BALKAN BALLADS

bold suitor who lives to repent his rashness. The deserted girl
sings a magic song that bursts bridges and drowns ships at sea;
she rouses the elemental from his bed in the river, and he complains
that he can find no peace. The supernatural is less developed in
the ‘tragoudia’ than in prose folk-tales. One encounters ele-
mentals, dragons, and charontes in the ballads, but in the folk¬
tales also sirens, fates, lamiae, and nereids. The nereids are not
restricted to the waves, but live in hills and woods, and in their
characteristics resemble the Serbian ‘vile’ and the Bulgarian
‘samodive’. The shining loveliness of Helen still stops men’s
hearts on that Asian shore; for when Emirissa (Politis 97) went
down to bathe

the shore flashed light, it lightened on the beaches.

There are the usual reproaches and tragedies, as when the tender
Theonitza dies of love.

Almost equally numerous are the dirges (moirologiai). They are
sung by professional women, and are adapted to typical occasions;
the individual case they fit more or less appositely. The ballad of
Kyr Borias (Politis 98) is used for deaths by shipwreck, though it
is a narrative of a particular tragedy caused by a miscreant Jew.
Few dirges are so specific. They employ pathetic comparisons:
death is a bitter potion, the house of mourning lies under a cloud,
the earth has a ravening maw, death is a long journey without
return, the garden is withered, the agonizing soul is entreated to
wait at least till summer come. These dirges pass easily into the
songs about Charon. Charon has been multiplied, like Cupid: one
encounters Charontes and even a Charontissa. The story of
Admetus undoubtedly lingered among Greek folk-tales, and in the
wrestle of Digenis with death it took a new life. Young shepherds,
klephts, or soldiers suddenly encounter death, and the pretty rich
girl, newly betrothed, does not escape her doom. Death has his
feast, and the girl who dies may still feel keenly the forgetfulness of
those who live.

Both the love-songs and the dirges are reduced to their simplest
form, and become distichs. These distichs are the final stage of
Greek folk-poetry; for when the power of composition is lost, as
in the Dodecanese, only distichs continue to appear. There is
thus a curious parallelism between the ‘tragoudia’ and the history
of classical Greek poetry. Of classical verse the epigr am was the

ASIA MINOR, GREECE 319

last ^ to be cultivated with grace and freshness. The traditional
distichs have not the formal perfection of the best epigrams, but
the\ maintain, a level much above the worst. They slip easily over
the tongue, since their rhythms are those of normal speech. They
are unemphatic, direct, and full of poignancy or charm. Were we
to count them not by lines but by hemistichs, they would resemble
the quatrains of modern Portugal and Latvia, which are also reduc¬
tions of older, more varied verse types.

There are ballads of crime—typical crimes, rather than particular
ones. The Evil If 'ife (Politis 80) relates how one brother slew
another to take his wife, but, overcome by remorse, slew her too.
An evil mother drives her son into exile, and an evil step-mother
bewitches her children. The special motifs of piracy and' brigand¬
age are a sauce for old situations: a girl proves to be her captor’s
sister (Politis 86), and a count is forced by misery to sell his wife
to one who proves to be his brother (Politis 86 b). A girl offers a
kiss for her lover s ransom, but is refused; the Emtrissa is seized by
pirates when bathing; the captive women of Chios lament, and so
do Greek prisoners in the Barbary states and a galley-slave in a
Turkish ship at Lepanto. There are stirring fights between three
monks and a Turkish privateer, or between Tsoulakis and a pirate.
In one poem the Greeks go down defiantly resisting, in another it
is the ship that will not yield to a storm. The best of these sea-
pictures—a special merit of Greek poetry—is the ballad of Kyr
Borias. A storm overtook his vessel, on which was a Turk and" a
Jew. The Turk was ordered to pray in his own fashion, the Jew
to recant. So the wind abated. But the Jew changed his mind
again, blasphemed Christ, and roused the wind afresh so that the
ship was lost.

To complete the picture of Greek folk-verse, we find many types
that are not strictly relevant to this work. They serve to mark the
unity of Balkan folk-song, for they are generally found in other
lands in much the same form. The stages of funerals are signalized
by special dirges, and there are songs for each moment of the
bridal feast: betrothal, the bridegroom’s pride, stuffing the beds on
a Saturday, riddling, leading the bride, reaching the new home,
scattering bramble-straw and ivy. So also there are the stated
seasons of the year: Noels or ‘kalanda’, Palm-tide songs or
‘baitika 5 , with appropriate compliments for master and mistress,
.priest and heir. Shepherds and other workers have their songs,

320 BALKAN BALLADS

and gnomic pieces are storehouses of rustic prudence. The Mill
Song (Politis 234) must be one of the oldest in type, since Herodotus
speaks of one sung in his day. The modern Swallow Song has only
one line in common with that used by children in ancient Athens;
it is used to greet the spring, and enumerates the signs of the
opening year in a manner like that of the Old French May-songs,
for which there are several other parallels. Songs of exile are a
peculiarly Balkan genre. In such sterile lands young men are com¬
pelled to seek work far from their mountain homes; their regrets
are embodied in a whole class of lyrical ballads.

2. Yugoslavia , Bulgaria

The Serbs and Croats occupy the ridge of the Dinaric Alps
which separates the Dalmatian coast from the interior valleys of the
Save, Morava, and Vardar. All this land is theirs; consequently
theirs is the watershed between Greek and Latin civilization
together with the outposts of each. Farther to the east, Bulgarians,
speaking an almost identical language, are wholly Balkan. To the
north the Yugoslavs maintain contact with the outposts of
Germanic culture, which is, in its best features, an offshoot of the
Latin. Illyricum and Pannonia belonged, under Diocletian’s rule,
to the Western Empire; but with the fall of Rome their fortunes
were identified with the East. The infiltration of the Slavonic
tribes impressed a new character on this region. Stepan Dusan,
in the fourteenth century, extended his short-lived empire in the
direction of Bulgaria and Macedonia, while the coastal towns—
Ragusa-Dubrovnik, Spalato-Split, Zara, &c.—were falling under
the influence of Venice. The Serbs and Bulgars embraced the
Greek faith, the Croats the Roman. With the faith went the
alphabet, so that two are still current in Yugoslavia. The Croats
speak the same language as the Serbs, but on the coast men spoke
Dalmatian Romance. They are the ‘Latins’ of the bride-stealing
ballads of the fifteenth and sixteenth centuries, which show how the
Christian alliance against the Turks was hampered by the mutual
suspicions of head-strong mountaineers and crafty townsmen. The
sailors of both shores of the Adriatic lived the same life and under¬
stood each other well, but an abrupt change of ideals and manners
occurred in the course of the five-thousand-foot climb of the
Montenegrin wall.

Thus Yugoslavia is a land of transition from one civilization to

YUGOSLAVIA, BULGARIA 321

another, and it has, like other regions of the same kind, developed
a sturdily independent culture of its own. This culture is expressed
supremely in its ballads. W ith the fall of the medieval empire and
the desperate struggle which continued through five centuries, there
was no time to devote to a sheltered written literature. Inspiration
was immediate and local; a form of verse was found w'hich proved
to be a satisfactory vehicle for hasty improvisations. The art
was rude and simple, but it was cultivated by the best minds of the
nation. The principle of hardy resistance was embodied in the
petty Montenegrin chieftains or in the outlaws who held the moun-
tam defiles ; beside the chief stood the poet, often a blind fiddler on
a single string, able and ready to put deeds into u'ords, to praise a
hero, condemn a traitor, and lament a disaster. The ‘guslar’ was all
the history and posterity that his leaders could expect. His gift of
oral traditional expression was so highly cultivated that, in our
own time, deputies of parliament have had a sufficient store of tags
to put a whole debate into extempore verse. But his gift was better
applied than in this modern instance, and the fame of those he
praised ‘volitabat per ora virum’. We have the less reason, there-
fore, to be surprised at the singular nobility of Serbian verse. It is
of ballad simplicity, but of epic seriousness; critics find it hard to
kncnv to which branch it belongs. There is a sad, simple, and
beautiful grandeur in the cycle of Kosovo which causes that group
of poems to stand out as a supreme moment of European balladry.
If the same height is not held by those of Marko Kraljevic, and if
there is a perceptible descent into savagery' in the haiduk pieces,
the Serbian ballads are seldom cheap and vulgar in their most
degenerate epoch. In those of Bulgaria, which are all recent, crime
and superstition occupy a greater place, and prosy language is
almost universal. The neck of the Bulgar was immediately under
the heel of the Turk, and his despair is not so much expressed in
his ballads as to be inferred from their meagre aspirations. There
are Serbian ballads of the same kind; but the survival of fierce
liberty in the Montenegrin mountains and some divine instinct
keeps old tags vivid and old emotions still keen . 1

The Serbian music is old Byzantine . 2 An extremely simple

1 See Note P, p. 391.

3 W. Wiinsch, Heldensanger in Sudosteuropa, Leipzig, 1937. Bulgarian, music
is treated.by C. Obreschkoff in Das hulgarische Voikslied , Bera-Leipssig, 1937.
Genera! questions of performance are noticed in W. Wiinsch, op. cit. , and in
M. Murko, La Poesic popuiaire epique en Yougoslavie au debut du xx € stick,

4615 T t

322 BALKAN BALLADS

monotonous chant is heard in the men’s songs of the whole moun¬
tainous region. The range of notes is narrow, and their intervals
not those of the west. In Bulgaria a new, more ornamented style
occurs, more akin to the modern Greek ballad chant. There are
the characteristic irregular intervals, and quarter-tones and third-
tones as well as semitones. But it is particularly in flourishes and
rapid improvisations that the two styles differ. They are found
together in the valleys of the Serbian rivers, especially in the
Vardar basin. Two waves of Greek music have followed the same
course from east to west, but the second has not reached the l imi ts
of the first. The men chant in the high mountains, and reckon to
leave singing to the women; to the women belongs also the sole use
of the round dance.

This distinction between women’s songs (zenske pesme) and
men’s or warriors’ songs (junacke pesme) is fundamental in
Serbian balladry. It was made by V. S. Karadzic, and serves to
mark off lyrics from narratives, and love-songs from mar tial
adventures. It is the men’s songs that concern us in this study,
since they are the ballads which our definition seeks. But the
distinction is not absolute: there were love-narratives which
Karadzic himself did not know how to classify, and indeed the
distinction in some cases is rather by religion than by sex. Bride¬
stealing is one of the main themes of the men’s songs, and is
associated with the most august names of Serbian history; but a
ballad of bride-stealing is a kind of love-song. The harassed
mountaineers could hardly woo in any other form; but the peaceful
Yugoslav Moslems—the Omer of the ballads and others like him—
could woo and wilt in a more feminine way. Between the men’s
narratives and the women’s lyrics there thus lies a middle region of
love-tales suited to either sex. None the less, the broad distinction
between men’s songs and women’s holds throughout Yugoslav
oral verse, and it justifies the attention given by students of the
ballad to the ‘junacke pesme’ in isolation.

A very large number of metres are used for women’s songs,
including the heroic decasyllable of the men. There is a lyric

Paris, 1929. The whole social background is etched by M. Braun, ‘Zur Frage
des Heldenliedes bei den Serbokroaten’, and M. Braun and T. Frings in ‘Helden-
lied’, both articles published in Beitrdge zur Geschichte der deutschen Sprache und
Literatur, lix, 1935. F. Saran discusses the prosody of Serbian verse in Zur
Metrik des epischen Verses der Serben, Leipzig, 1934, referring to the decasyllable
only.

YUGOSLAVIA, BULGARIA 3.23

decasyllabic, with caesura after the fifth syllable, and the use of the
variety with pause after the fourth is associated, even In women's
songs, with narrative. The men’s songs use the heroic decasyllabic
with pause after the fourth syllable. Under more modem and less
lofty conditions the octosyllable is found, especially in Bulgarian
verse; and there are some seventy-six surviving samples of an ir¬
regular metre older than the heroic line. This older verse is called
‘bugarsticab The word must be connected with ‘Bugar* ‘'Bul¬
garian 5 ; but one is deterred from associating the thing with
Bulgaria, both because of the derivative character of Bulgarian folk-

poetry as we now have it, and because the irregular lines surviving
in Bulgarian have quite a different character. The second half of a
‘bugarstica’ line shows a fairly regular succession of trochees:

. . . hero Marko Kraljevicu
. . . and his brother Andrijasu
. . . glittering sabre bright and golden.

These are the second halves of trochaic tetrameters, though they
are not unlike the second halves of ‘politic’ lines also. It is to the
first hemistich that no exact rule was applied.

But indeed they two were not, oh my friends, a pair of paupers,
for of twain the one was hero Marko Kraljevicu,
hero Marko Kraljevicu, and Ms brother Andrijasu,

warriors so young! (Duric iv. i. 1.)

If we allow the name to influence our hypotheses, we must grant
that a Bulgar empire arose before the Serbian, and in its striving
for culture laid itself open to Byzantine influences. The Bulgarians
may have transmitted to the Serbs some notion of the Greek
trochaic tetrameter (with a more precise conception of the second
hemistich than of the first); or they may have passed on the
politic metre of the capital, adapting it to the trochaic rhythm, as
characteristic of their language as is the iambic of Greek, In any
case, this oldest of Yugoslav traditional metres witnesses by its
name to its eastern origin; and other eastern characteristics of
Serbian verse are the absence of rhyme, the use of certain set
phrases, the negative comparisons, the speaking birds, and many
leading motifs.

The ‘bugarstice 5 are found in documents of the sixteenth cen¬
tury, but from the seventeenth and onwards the heroic decasyllabic

prevails. It may have been used to renovate lost ‘bugarsticeh On

32 4 BALKAN BALLADS

the other hand it may also be as old as the men’s songs themself
since ( oug 1 manuscript evidence is wanting) some early prose'
that incorporates the sense of ballads of Kosovo gives unde/r
construction, heroic decasyllablesd The resemblance ofTcoLZ
Serbian verse to the old heroic line of the French ‘chansoiSV
geste 1S striking, both because of the lofty intonation, andTelse
of the marked pause after the fourth syllable:

Rose so early Maiden of Kosovo,
rose so early on a Sunday morning,
lrst arose sbe e’er the sun had risen.

The difference is only that the French line is iambic (when it W
a demonstrable rhythm), but the Serbian is trochaic. The French

svllaH U h b \ h °^ eV T’ WaS converted int0 ^e Italian hendeca-
steadofthe‘m 'T 6 , ex P edient of taking the ‘feminine’ line in-
f ! ^ T lme aS norma1 ’ and in folk-verse this hendeca-
y lable tends to have a fixed pause after the fifth syllable The
Italian line is felt to have a trochaic rhythm. % supnr ess fo^
unaccented first syllable, the Serbian decasyllable adses Them
eems little reason to doubt the western character of this line and

? ITmTid*/ by way °?r* “ d *•

7, , a P f h marked down, as we shall see, by the traces of western

O ne entered V pemed int0 Yugoslavia and Greece.

of Kosovo t! J T ke PeSm thl '° Ugh the ma gnihcent portal
Kosovo. Traces of song can, of course, be followed through

allusions to a remote date, and there is even evidence that women’*

songs presumably of a lyrical cast-were ali^ady^S^S

mately the modern manner in the thirteenth century. °But his the

men s songs that constitute the true ballads of this region and the,'

iTsT lThas°bT S aI , m0St K , abrUptIy With the nationaI disaster in
J* 9 '. L + u h P IausibI y represented, though it is by no means

sh o r of SeTJck ? TT* ^ ^ reached the

heard K fu 7 S f before the y ear out, and was there

iai lnt h” 11 u gnat iL and induded in his R ^^ian travel

Philosopher (writing J ° ^ S f r Lazar ’ s son b y Constantine the

construction nnt g “ we enc °unter the same religious

obstruction put upon the event as we find in the ballad of The

3 S 'afd e 9 4 he 6SSayS ^ reCOnStrUCtl0n ^ M. Braun, Kosovo , Leipzig, I937) pp .

I937 Tj he texts are tfven in Helen
’ 9 2o, accompanied by an excellent verse translation.

YUGOSLAVIA, BULGARIA 325

Fall of the Serbian Kingdom. A Yugoslav Janissary, Constantine
by name, writing in Polish after 1463, follows ballads which he
must have heard before his capture by the Turks in 1455; and the
text of an anonymous Dalmatian translator into Italian of Don¬
bas s chronicle uses language that may be readily retranslated into
berbian verse. He wrote before the end of the'fifteenth century.

. e ore I ^ 00 > therefore, this cycle was virtuallv complete; its most
important members must have circulated widely by 1450; and one
may belong to the very year of the great disaster. "

Serbian balladry thus begins in the form of a news bulletin. The
precedent was set not so much by Greek ‘tragoudia’, though these
were already in existence and probably known to some Serbs, but
m the Political verse of the Greek ‘iitterati’. They were accustomed'
to compose panegyrics on events of the dav and to circulate
information at court; and it was probably at court that the
hrst Serbian narratives arose. Their milieu is courtly; their poise
aristocratic. The style of the Kosovo cycle is one fully elaborated;
its elaboration presumably occurred at the court of the emperor
Stepan Dusan, whose military' achievements clamoured for
consecration in verse. There is, indeed, a ballad of his marriage
(Karadzic, ii. 28), but it has so many inaccuracies that one can
hardly accept it as contemporary' work. He is said to woo Roxanda,
aughter of King Mihail of the Latins. His champion Milos pene¬
trates the country of the Latins disguised as a ‘black Bulgar’.
Doubtless there is a reminiscence here of the emperor’s Bulgar
marriage, confused by memories of Montenegrin alliances with
the princelings of Dalmatia. Ballads which profess to refer to the
Nemanjid dynasty, of still earlier date, all show signs of late com¬
position or re-editing. The art of poetry' may have been elaborated
at Stepan Dusan’s court upon themes supplied by his dangers and
triumphs; but it was only with the disruption of the kingdom
that such poems were launched upon oral tradition. The Kosovo
ballads had, perhaps, antecedents; but they are unquestionably the
oldest surviving narratives. They report an event of breath-taking
importance to the Serbian folk. But they also insinuate remoter
and immediate causes, interpret character, and view the event
sub specie aeternitatis’. In the moment of overwhelming disaster
they fortify the national spirit. They make it invincible since it is
pure spirit, independent of places and persons, of strength or
weakness, but living in words and rhythms that could not be

3^6 BALKAN BALLADS

forgotten. It is in these ballads that the Serbian nation truly lives;
and the peasant-soldiers who crossed the plain of Kosovo with
reverent awe in 1912 were completing a ritual act begun by the
self-sacrifice of Tsar Lazar.

The remoter causes of the disaster are stated to have been the
treachery of the south Serbian King Vukasin and the Laodiceanism
of the northern Tsar Lazar. The ballads are thus originally from
North Serbia. Vukasin’s treacherous character is seen in The
Founding of Skadar (Karadzic, ii. 25), which is borrowed from the
Greek Bridge of Aria, but used to calumniate that king, and again
in Uros and the Mrnjavcevici (Karadzic, ii. 33). Marko Kraljevic,
though son to Vukasin, decides against his claims in competition
with the legitimate successor of Stepan Dusan. These ballads are,
perhaps, quite recent. Tsar Lazar’s fault, as described by The
Founding of Ramnica (Karadzic, ii. 34, 35) is more venial. He has
neglected to build monasteries. When reminded of his fault he is
even over-anxious to make amends, but the warning voice tells him
it is too late. The immediate cause of the Serbian disaster is, how¬
ever, the quarrel between the hero Milos Obilic and the traitor Vuk
Brankovic (Duric, iv. 8, 11). The Tsar insulted the hero and
favoured the traitor. In this ballad we are entering the area of
songs contemporary with the event. In actual fact Milos pene¬
trated the Turkish camp before the battle and killed Murad. It
was a feat of immense daring; but it was rendered vain by the
promptitude with which Bajazet suppressed all news of the disaster,
so that the Turks remained undismayed in face of the Serbs. The
ballad was presumably composed by some poet bitterly hostile to
the powerful family of the Brankovici, who later bore the brunt of
Turkish attacks on the Serbian valleys. He insinuates a cause for
Milos’s action which may not have been historically true; but the
feat itself is historic, and appears in the earliest reports. A ballad
of Queen Milica and Vladeta (Duric, iv. 7) reinforces this contrast,
and one of Milos among the Latins (Karadzic, ii. 36) relates other
evidence of his prowess; in a visit to Ragusa he tossed his club over
the roof of a church.

The battle is described in The Fall of the Serbian Kingdom
(Karadzic, ii. 45). Waves of myriads of Turks roll forward, are
checked by a hero and a small band, and routed—but the hero also
lies dead. The nine sons of Jug Bogdan perish along with their
father (there is a ballad entitled Milica and the Jugovici which

YUGOSLAVIA, BULGARIA 327

shows how every one of them refused to blench before his fetal
duty), and last of all Tsar Lazar is overwhelmed. The battle is
stylized; it is almost a piece of ritual. And rightly so; for the
poet has chosen to represent Kosovo as a national expiation.
A prophet offers Tsar Lazar the victory upon earth or in heaven,
and he chooses the latter. His death is as the Lord appointed, so
that the end of the ballad answers to the beginning;

Came a-flying, came a grey-backed falcon,
from the temple and from Salem’s city,
and he carried in his beak a swallow. *

Nay, ’twas not so, not a grey-backed falcon,

rather was it the seer, great Elijah;

and he carried not with him a swallow,

but a message from God’s holy Mother,

to the Tsar he brought it, to Kosovo,

placed the message on the Tsar’s own knee-cap,

and the message thus the Tsar bespeaketh:

‘Tsar Lazare, worthiest of people!

which now would’st thou choose between two kingdoms—
whether would’st thou have an. earthly kingdom,
or would’st rather win a heavenly kingdom?

If thou wouldest win an earthly kingdom,
saddle charger, girth and trappings tighten,
gird upon thee thy heroic sabre,
furiously on the Turk make onrush,

Turkish warriors every one will perish;
would’st thou rather win a heavenly kingdom,
consecrate thou a temple in Kosovo,
whose foundations build thou not of marble,
silk and scarlet build thou of the purest,
purify there, order there the army;
all must perish, perish all thine army;
in the midmost, ’tis thy doom to perish . .

So there perished Serbia’s Tsar Lazare,

and about him perished all his army,

for their number seven and seventy thousand:

all was holy, all was honourable,

all accomplished as the Lord appointed.

It is to this conception of holiness that the later ballad of The

Finding of Tsar Lazar's Head (Karadzic, ii. 52) corresponds. Its

tone is hagiological.

328 BALKAN BALLADS

The poets give us glimpses of the tragedy from several stand¬
points. It is seen most terribly in The Death of the JugovieTs
Mother (Karadzic, ii. 47), for she perished when a raven brought
her the hand of her dearest son, Damjan, together with the news of
the ruin of her house.

Then that mother takes Damjane’s hand up,
turns and turns it, turns it o’er and gazes;
still and softly to the hand she whispers:

‘Hand the dearest, green and lovely apple!
where begotten, where hast thou been severed !

There begotten, in my loving bosom,
there wert severed, on the plain Kosovo I’

Swells with anguish Jugovici’s mother,
swells with anguish, till her frame was shattered,
for the dear sake of nine Jugovici,
and the tenth one—aged Jug-Bogdane.

But there were others of less exalted category, whose sorrow was
softened by pathos. Just as the young hero Music Stepan (Karad¬
zic, ii. 46) perished in a manner more romantic than tragic, so there
is a softer side to the grief of The Maiden of Kosovo (Karadzic, ii.
50). Early in the morning she rises and goes to the battle-field; she
wipes the blood from dead faces and turns over the heaps, till she
comes upon her own dead lover. With the discovery the joy is
eclipsed from her own young life, and she ends lamenting her
doom to barrenness:

Down her white face streams of tears descended,
homeward goes she to her own white homestead,
from her white throat wailing like a cuckoo:

‘Who more wretched ? O my fate is bitter!

Wretch, embracing such a fair green pine-tree,
but in green youth has my pine been withered.’

These ballads do not constitute an epic, though in their uniform
dignity they have tempted the skill of modern epic poets. 1 The
disaster is never presented as a complex whole. It is glimpsed in
sections, as it affects various people: Queen Milica, the mother of
the Jugovici, an unnamed girl, Milos, the Jugovici and Lazar. Yet
some strange instinct would seem to have been at work to see that
nothing entered the Kosovo cycle but what was worthy of a place

1 For instance, the Lazarica Hi Boj na Kosovu of S. J. Stojkovic, in 24 cantos.
It is in places a centon of ballads.

YUGOSLAVIA, BULGARIA 329

within it. The language is unapproachable, and the phrases have
become the resource of all subsequent poets. The ritual solemnity
persists in poems on less solemn themes; all Serbian folk-verse
makes life a pattern that repeats itself through eternity.

Next to the cycle of Kosovo stands that of the Serbian national
hero Marko Kraljevic, or Marko the King’s Son. 1 The cycle is
historical in so far as Marko was indeed a prince of Prilep, a fortress
commanding the pass from the Vardar valley to the plains of Mona-
stir in southern Serbia. He seems to have taken, part in the battle
of Kosovo, but on the Turkish side. The rain, of the old kingdom
was largely due to the dissensions of the Serbian despots, and the
ballad poets, while condemning Marko’s father, Yukasin, on this
account, have exonerated the son. His independence had been
forfeited by the loss of south Serbia in 1370 as the result of the
battle of the Marica; but Marko remained in semi-independence as
lord of Prilep. So he is represented in the ballads: he receives
orders from the Turks and serves as their vassal, and yet he main¬
tains a solitary personal independence. He is typical of the hero¬
ism which was still possible at that date in southern Serbia, and
his cycle is opposed to that of Kosovo as southern to northern
and semi-independent to unsubdued. He is also a plainsman as
opposed to the Montenegrin highlanders, and as such he is as well
remembered in Bulgaria as in Serbia. The Miladinov brothers
collected twenty-four Bulgarian ballads of Marko Kraljevic, but
none of Kosovo. As a strictly historical and national group the
Kosovo poems have not the power of migration or survival that
belongs to the more novelesque Kraljevic cycle. They are un¬
represented in the Erlangen manuscript, collected by an unknown
German in Hercegovina in the eighteenth century, though he
heard several relating to Marko Kraljevic. It is one of the respects
in which the Bulgarian corpus resembles a late stage of Yugoslav
balladry.

The cycle is novelesque. In forming the hero’s character it

seems impossible to doubt that some ballad-poets had in mind the
character already assigned by the Greeks to Digems Akritas. Both
are club-heroes, both are solitaries, and both are ambiguously

1 D. H. Low, The Ballads of Marko Kraljevic, Cambridge, 1922, contains
these poems in translation. An important topic is discussed by M. Budimir in
£ Digenis und Marko Kraljevic*, Actes du quatrieme Congres internationals des
Etudes byzantines, 1937.

4615

33 ° BALKAN BALLADS

associated, in a mixture of truce and war, with the Moslem enemv
Both are styled ‘the widow’s son’. Both die defying the super¬
natural, since Digenis wrestled with Charon himself, and Marko
ascended his last mountain despite the warnings of a Vila/ while on
another occasion he fought against the ‘vila’ Ravijoila and com-
pelled her to give back the life she had taken from his foster-
brother Milos Obilic. The material of The Bridge of Arta is used
as we have seen, to the disadvantage of his father Vukasin His
faithless wife resembles the Greek Magda (Erlangen 151). On the
ot er and, his name could easily be used for legends that had
come from the west. He is the hero of one variant of the Morinser
t eme (Dune, iv. 1, 8), which is also the subject of the ballad of
lodor of Pomorje, outside his cycle. Marko Kraljevic and the royal
Dehbashaj Erlangen 139) is a sanguinary derivative of Marianson's
Rings ^and Marko and Tuka Senkovic' (Erlangen 151) resembles not
only the Greek piece mentioned above, but also the Italian Avens-
mg Husband. In Marko Kraljevic and the Arab King's Daughter
(Dune, iv. u, 1) we have the old story of the prisoner and the
gaoler s daughter; having escaped, like Young Beichan, by her
aid Marko neglected her, and later suffered remorse. A model
might have been found in the incident of Haplorrhabdis in the
Akritic epos. Owing to the mingling of so many stories, Marko’s
matrimonial life has become too complicated to unravel. His

”, S .° was not sim P Ie ; in King Vukasin's Marriage
(Karadzic 11. 24) we are told he seduced the wife of Momcil of
Pirlitor and caused her to kill her husband. Rightly inferring that
she would in time be equally ready to kill him, Vukasin spurned
her, and married the noble Jevrosima, Momcil’s sister:

so begat he Andrijas and Marko,
but was Marko liker to his uncle,

to his uncle, to the voivod Momcil.

. old , motif of the Minotaur’s tribute of damsels is exploited
Jtos adnnhge.in Marko cancels the Wedding-tribute
(Karadzic 69). As Ogier of Dinamarche saved Gloriande, so
^ ark ° v Sa 7 ^ d , a sult “’ s daughter from marriage with a black Arab
( ara zie 5), and like the Ogier of the Moniage, he was liberated
rom prison as being the only champion ready and willing to fight

^ Sa Arbana f (Karadzic 66). Probably the best of the
Kraljevic ballads is Marko's Ploughing (Karadzic 72). His mother,

YUGOSLAVIA, BULGARIA 331

weary of Ms vagabondage, urges him to follow a useful pursuit. He
decides to plough; but ploughs up the public highway and beats
the Turks when they protest:

Wine was drinking Kraljevicu Marko,
with his ancient mother Jevrosima.

When of drinking both of them were sated,

Marko s mother thus began her discourse:

‘Oh my sonny, Kraljevicu Marko I
Make an end, son, of your raids and scuffles,
for of evil never good ensueth;
your old mother grievously is troubled
washing bloody gobbets from your clothing;
take you rather to your plough and oxen,
plough and furrow then the plains and ridges;
you are, sonny, a youngster white and lovely,
yet you nourish neither self nor mother. 5

Then did Marko listen to his mother,
and betook him to his plough and oxen,
but he furrowed not the plains and ridges,
rather ploughed he up the royal highways.

Thither hurried Turkish janissaries,
bringing with them three great loads of treasure,
and admonished Kraljevicu Marko:

‘Come, come, Marko, don’t plough up the highways.’

‘Come, come, Turkeys, trample not my ploughing.’

Come, come, Marko, don’t plough up the highways.’

‘Come, come, Turkeys, trample not my ploughing.’

Now when Marko grievously was troubled,

Marko lifted high his plough and oxen,
and he slaughtered Turkish janissaries;
so he gathered three great loads of treasure,
brought them homeward to his ancient mother:

' ‘Here I bring you this day’s wage for ploughing.’

The humour is of a broad Slavonic kind, but it is not the huge
guffaw of Russia. The Serbian art is one of measure and restraint.
As a prodigious ploughman, however, Marko invites comparison
with the Russian Mikula, and this is one of the few pieces which
seem to link the southern and the eastern Slavs.

Apart from the last ballad of his death, Marko is the centre of
a cycle which has not the elevation or seriousness of the Kosovo
poems. He is an average Serbian raised to a greater stature. He is
a solitary, satisfied with a harsh independence; his family is his

332 BALKAN BALLADS

interest, but not clearly his nation. He is generally good-humoured
and slow to anger, but he can be abrupt and bloody. He rights
wrongs like a Robin Hood, but does no more for those he has
saved. He is not exempt from moral and mental weaknesses
which he recognizes without false pride; and his deeds are always
greater than his words. J

In a long losing battle the fragments of the Serbian kingdoms
after Kosovo were defended by many petty despots. There were
the Brankovici at Smederevo on the Danube, the Jaksici in Bel¬
grade, and the wild Crnojevici on the Black Mountain. Their lives,
as simplified by the ballad poets (their closest retainers), revolved
around two moments: marriage (zenidba) and death (smrt). The
marriages follow a fixed patriarchal pattern of bride-stealing. The
suitor collects a train of supporters (kupiti svatove), to the con¬
ventional total of a thousand. Among these are persons deputed to
special offices at the ceremony, who are ‘ipso facto’ officers of his
little army. They set out across the great watersheds and descend
to the Dalmatian coast. The tricky Latins are frightened at the
Jow of force, and try to.get the Serbs to lay their arms aside.
They try to get the champions dismissed, on the ground that they
are quarrelsome in their cups. If the champions have been dis¬
missed (as in Dusan’s Wedding) the wooer is only saved from his
imprudence by an unexpected succour; if not, the ballad will end
with the Serbs cutting their way out of an ambuscade (as in Jiirju
of Smederevo’s Wedding, i.e. George Brankovic, d. 1458). Marko
and the heroes of Kosovo are given prominent roles in all the early
weddings. Then the brides have to be led home. A proxy takes
chaige, and a gust of air may lift her veil. Then the proxy falls
violently in love, and so commits the worst offence known to
ballad-poets: the breaking of a foster-brother’s fidelity. A ballad
of Marko Kraljevic has that pattern (Karadzic, ii. 55). What there
may be of historical truth in such stories it is scarcely possible to
imagine. A modern Montenegrin, if he thinks of singing his prow¬
ess as a wooer, will cut his tale to this pattern without reference to
the real facts. The only difference is that he may adopt a tone of
comic exaggeration.

, Heroes die, for the most part, overwhelmed. The ballads of
their deaths thus serve to mark the stages of the Turkish advance
up^the Morava valley; an advance which cut Serbia in two and
isolated Montenegro. A brilliant example is Voivod Prijezda’s

333

YUGOSLAVIA, BULGARIA
Death (Karadzic, ii. 83), which dates from the mid-fifteenth cen-
tury. The Turk Memed (Muhammad II) orders him to surrender

his horse, sword, and wife. He refuses, and a vast Turkish army
advances on his castle of Stake. For three years the invaders are
kept at bay. Then one day Lady Jelica sees that the waters of the
Morava are running muddy, and she infers that the Turks have
mined under the river. They descend to the cellar and see the first
Turks arriving. After a fruitless resistance Prijezda kills his horse,
shatters his sword, and leaps from his battlements into the river,
grasping his lady. Another way of accounting for the Serbian
losses has been the invention of an evil queen, namely, Jerina of
Smederevo. According to one ballad (Erlangen iS), failing to
seduce Damjan Sajnovic, she tricked him into killing his young
wife. So she destroyed good counsellors and good voivods in the
interest of a sultan to whom she had given her voung sister in
marriage. Her brother was a would-be kidnapper, and Jerina
forced his hand on the fiancee of a better man. A third cause of
disaster were the quarrels of the Jaksici and other leaders, quarrels
which amply account for the fall of Belgrade.

Interesting figures in the background of these poems are the
Hungarians John Hunyadi and Matthias Corvinus (1458-90).
Already Marko Kraljevic has countered with ruthless actions the
‘gabs’ of the Magyar Philip (Karadzic, ii. 58), and the later Hun¬
garians are treated as uncertain allies, ‘Janko of Sibin’ and
Matthias are regarded with sympathy, and there is a considerable
group of folk-songs which mention their capital of Buda with
admiration. The friendliness had some interruptions, as in the
rivalry which led up to Voivod Kaiccis Death (Karadzic, ii. 80).
On two occasions the Erlangen poems show John Hunyadi in the
moment of defeat. King Maiyas (Erlangen 75) deals with his
elevation to the throne in a romantic manner; and according to The
King of Buda (Erlangen 73), Matthias was the only survivor of a.
battle which may have been Mohacs (1526). He was not a king but
a ‘servant’, and he ran to tell the news of the Hungarian king’s
death to the queen, who promptly married him. In a rather jolly
drinking-song, Matthias rebukes Peter of Varadin for Ms heavy
drinking, but Peter retorts that the wine of the tavern is worth all
Buda and Pesth;

Then retorted Pete the stroller, lord of Varadin;

‘Stop your prating, King Matthias, our land’s sovereign.

334 BALKAN BALLADS

Were you guesting with the landlord in whose inn am I,
greater wealth and far more ducats would you spend than I;
all your Pesth and all your Buda would you squander too—
girls in Pesth are far too pretty ... a murrain on the crew!’

(Erlangen 178.)

With the collapse of resistance in the sixteenth century the free

Serbs were restricted to Montenegro, or to small bands of outlaw
‘hajduci’ (conventionally composed of thirty persons in the ballads)
or raiders from the coastal cities (uskoci). Hundreds of ballads
speak of the forays and misfortunes of these persons who stand
midway between heroism and criminality. A score of names have
been handed down to posterity: Ivo of Senj, Janko of Kotor
Stojan Jankovic, Mihata, Ilija Smiljanic, Milja the Shepherd, Vuk
Mandulic, Erko the Latin, and others who occupy many pages of
the Erlangen manuscript. The models for such pieces were the
existing ballads of bride-stealing and heroic death, imprisonment
and escape, skirmishes and feuds, modified by some touches taken
from the actual experiences of the outlaws. Highest in esteem was
Old Novak. In Old Novak and Radivoja (Buric, vii. 2) we learn
how much better the old leaders were than the new. The injustices
that drove men into the mountains are illustrated in Old Novak
and Bogoslav (Buric, vii. 1). Novak m Prison (Erlangen 67) is a
ballad built on the old heroic lines. Captured and abused by the
sultan, Novak elects to run the gauntlet of the Janissaries as the
most heroic mode offacing death; with the aid of a ‘vila’, his sister, he
kills three thousand men, and escapes. The old lion roamed over
the Rudnik hills, and was more intimate with nature than with men:
Keen reproaches from Rumanian mountains:
f Curse thee, curse thee, aged outlaw Novak;
all our leafage is stripped by hungry eagles,
all our tree-tops crushed by wing of ravens,
downward swooping on young warrior corpses.

(Erlangen 96.)

Ivo of Senj’s marriage and death were both remarkable. To get
him a bride his lieutenant penetrated into Udbinja, the home of his
worst enemies, and kidnapped the fair Hajkuna (Burid, viii. 6); his
death occurred in the pursuit aroused by one of his forays (Buric,
viii. 7). What made his exploits the more hazardous was that his
opponents were Slavs like himself, especially the doughty Ljubovic,
and they had the same courage and manner of life. His own he was

YUGOSLAVIA, BULGARIA 335

ready to stake by attending a Moslem round-dance, and he went
through the usual vicissitudes of imprisonment, skirmishing and
quarrelling in his cups. Deep drinking led on to ‘gabs’, and when
sober again the heroes found themselves committed in honour to

the wildest adventures. So, in Bojcic Alija and Gliimac Asanaga
(Duric, viii. 8), the Moslem Serbs undertake in their cups a raid
on the coastal cities which they have to execute when sober; they
ravage the coast successfully, and it is only a wanton crime that
brings down vengeance upon them. Formal challenges to single
combat are stock tiiemes, and there are lively representations of
stratagems. In Gruica and the Pasha from Zagorje (Duric, vii. 3) the
conventional thirty outlaws disguise themselves as girls and kill, off
an amorous pasha with his retinue. Imprisonment and death end
the career of many a band; and in such a case silence was the virtue
most esteemed (Duric, vii. 7: Old Vlijadifi). Ingenious or courage¬
ous escapes furnished a more congenial theme: Milja the Shepherd
got away by a clever ruse, Stojan Jankovic beguiled the sister of his
jailer, and Young Radoica shammed death, not flinching though
they laid a snake on his chest and drove nails under his finger-nails;
if his eyelid fluttered when Hajkuna danced round at the head of
a bevy of pretty girls, that was a weakness Hajkuna could conceal
and forgive (Duric, vii. 8). A certain Vukosav escaped because his
fiancee dressed like a Turkish soldier and pretended to bring an
order for his release. The free life was hardly less harsh. Juris of
Senj kept his band together after a lean season, and only after hope
of profit was abandoned did they strike a rich convoy, and in other
ballads we have glimpses of the band dispersed for the winter and
dependent on the loyalty of their friends. The adventures of
‘uskoci’ and ‘haiduci’ are too numerous to follow in detail; and we
need only add that there was at sea a certain Erko the Latin, who
was in the habit of shanghaiing hillmen to form crews for his raids
against the Italian coasts, and it was in a sea-fight that Djurja
Danilic and Simo the Latin met their end.

There is a short and undistinguished ballad in the Erlangen
manuscript (91) on the battle of Belgrade in 1717, the only interest
of which is that it forms a Slavic contrast with the German Prince
Eugenius. With the nineteenth century a more national spirit
entered into the war. The ballad-poets continued to organize the
Serbian guerrilleros, and in some way they had their finest hero in
Kara George. But the time for the best ballads was passed, and

336 BALKAN BALLADS

that of the great collector Vuk Stepan Karadzic was mature. It is
not so much a loss of invention that one notes—for the devices
were always common property—but a kind of ageing and stiffening
of the ballad-style, a restriction within limits and a failure to sur¬
prise. A certain collection of Popular Songs concerning the first
national Insurrection (Belgrade, 1914), though adorned with photos
of fiercely whiskered bravoes, opens with cliches in almost every
poem:

1. In a vision dreamed a pretty maiden . . .

2. God of mercy! What a mighty marvel! . . .

3. Fluttered thither two black-coated ravens . . .

4. Captain Kulin quickly massed his army . . .

5. Wine were drinking three young Serbian voivods . , .

17. Pens a letter a black-visaged Arab . . .

18. Screamed the vila high on Rudnik's mountain,
by the streamlet, slender Jasenica,

o'er Topola’s plain she sent her summons,
called upon him, Petrovidu Djoko . . .

These had been fine things once, but they had lost a good deal of
their truth. The full-mouthed phrases could be strung together
with too much ease, and they slipped into garrulous and prosaic
contexts. Yet this art, though declining, is still, in Montenegro, the
chief recreation of a simple, sturdy race, and sets the pattern for
their lives.

There are some curious and interesting religious ballads in
Yugoslavia, in which the authentic version of the Bible or hagio-
logy has been transformed by a naive use of the imagination. They
are collected in Buric’s first volume, and open the second of Kara¬
dzic. Thus, in Christ's Birth (Buric, i. 1), we learn that Mary was a
sheep-farmer and John and Simeon were hired men; Jesus is taken,
like any Serbian baby, to be baptized at once, and the only differ¬
ence is that the Jordan and Judaean hills do him prodigious
reverence; a Yule feast is ordered when the hired men return. In
Rich Gavan's Lady (Karadzic, i. 207) the parable of Dives and
Lazarus is turned to the disadvantage of Dives 5 wife. It is a beggars 5
ballad, along with Deacon Stephen and two Angels (Buric, i. 3).
St. Peter’s mother has a bad reputation and is sent to hell; and
Fiery Mary (Buric, i. 4) saw in hell various sinners, including one
who had been a bad daughter, bad wife, and bad mother, and so
was altogether bad, like the Samaritan Magdalene of western ballad

YUGOSLAVIA, BULGARIA " 337

tradition. A sinner whose sins excite the sea to a storm in the
Erlangen manuscript (190) was doubtless one of Jonah’s kinsmen,
a cousin of the Russian Sadko and the Scottish Brown Robin

(Child 57).

St. Helen is probably the heroine of Faliends Boasts (Erlangen
42). She boasts that she will accomplish some remarkable feats of
magic; so that the tsar sends for her and marries her. Constantine
the Great is not in high favour. We are told that he tried to
murder a deacon who had imposed a heavy penance, and fire from
heaven came and destroyed all but his hand, which had done some
deeds of kindness; he was also a destroyer of churches, only to be
restrained by the apparition of three great saints. These are
Michael, Nicholas, and Elijah'—a typically eastern group. A very
remarkable ballad is entitled Diocletian and John the Baptist
(Buric, i. 8). Diocletian or Dukiijan and St. John play a game for
which the stakes are an apple and a crown, and the saint loses. He
flies to heaven to get permission to swear one false oath; and armed
with this permission he returns to his game. He contrives to induce
Dukiijan to dive for the apple, while he himself steals the crown.
He freezes the sea and flies heavenward, but a 'cursed bird’ nips his
foot as he goes in:

Weeping sorely, John approached his Maker,
bright the sunshine he restored to heaven,
deep compassion felt God for the Baptist,
for the insult foul the tsar had done him.

Then the Lord God words of comfort uttered:

‘Never fear thou, good and faithful servant!

Even measure shall I give to others.'

So it happened: to our God be glory.

There are also ethical ballads which reprove unpopular vices.
Sons who expel their mother from home are turned to stone; the
archangel reproves Stepan Dusan’s pride; a church grows from
innocent bones; lightning strikes a Bulgar who has transgressed
the law of foster-brotherhood. This relationship is held more
binding than kinship. In The Foundling Simeon (Buric, i. 17) we
have a variant of the'Oedipus story, taken from oral tradition.
Mocked by his playmates, the foundling goes to Buda and there
wins the love of the queen; but he comes to know she is his mother,
and ends his life with a long penance of snakes and water.

The great bulk of the ‘junacfee pesme* consists of realistic

4615 xx

33§ BALKAN BALLADS

narratives which have to be classified as in some sense historical
even when they reproduce well-known fables. The class of adven¬
ture ballads, without attachment to historical facts, is thus contained
chiefly among the ‘zenske pesme’ of narrative cast. They are also,
not infrequently, the work of Moslem ‘guslari’. The Moslem Serbs
had no struggle for independence to sing; in compensation they
were better acquainted with the peripeteias of a civil polity. The
metres employed for such narratives are the heroic decasyllable,
the octosyllable, occasionally the lyric decasyllable with pause after
the fifth syllable, and rarely the trochaic tetrameter.

The ballads show no remembrance of the greater figures of
Slavonic mythology, and, with regard to the lesser daemons, the
Yugoslavs seem to possess a less fertile imagination than the
Bulgarians. The sun and moon are actors in their narratives; they
know vile and samovile’ who live in the hills and fountains,
dragons and snakes, and the use of magic. A single apparition of a
ghost is due to a Greek model: Brother and Sister (Karadzic, ii. 8),
like the Bulgarian Lazar and Petkana (Miladinov ioo), is a version
of Constantine and Arete or the dead brother’s return. The ‘vile’
are not always harmful. One, who is also the Morning Star, makes
a plain girl beautiful and becomes her foster-sister. But generally
they are mischievous. A girl should not bathe in a fountain if she
knows a ‘vila’ is there; a youth perishes in the elemental’s arms on
their wedding night; and they are wont to spirit away husbands.
In The fiery Dragon's Love (Karadzic, i. 239) it is the girl who is
spirited away to the banks of the Danube; but she refuses help to
come home. The ballad is from Dubrovnik. One works magic by
means of flowers and scripts dropped into fire; the effects are
various, but one of the most malevolent, as in Scandinavia, is to
defer a pregnant woman’s delivery. Particularly attractive are the
ballads of sun and moon, which form a small cycle. The Sun and
Moon woo a Maiden (Karadzic, i. 229) and The bright Moon's
Wedding (Karadzic, i. 230, 231) remind us of the Esthonian and
Lithuanian pieces, though they are quite independent. The
maiden prefers the Moon, with so many stars for relatives, to the
solitary and fiery Sun; in the other it is the Day Star which marries
the Moon. The Sun has a sister, whom it is dangerous to woo;
one tsar did so, but was glad to release her with gifts; a pasha tried
to capture her, but she hurled three apples and three thunderbolts
and destroyed all his forces (Karad2i<5, i. 233, 232). It is presum-

YUGOSLAVIA, BULGARIA 3 39

ably this lady who is the heroine of The Tsar mid the Maiden
(Karadzic, i. 234). The maiden is determined not to spin or
embroider, but to build a church. The sultan tries to beguile her
by messages, and then sends a vast army of Arabs, Tatars, and
Janissaries; but he is defeated, captured, blinded, and forced to
wander over the mountains like a wild bird.

The Bulgarian songs, if my authorities are truly representative
in this matter, display a more lively interest in the lesser super¬
natural. The Miladinov brothers reproduce twenty-six ballads of
this sort, and Dozon has a minimum of sixteen. To the Vile 5 corre¬
spond the Bulgarian ‘samodivek A ‘juda* is a ‘samodiva’, but with
a New Testament name. There are also Lamias, dragons,
Charontes, Fates, a Pest, snakes, dragons, the Sun and Moon, an
unchristian God, and talking birds and animals of all sorts. The
bold youth defies the ‘samodive’ to a contest on the flute, tricks
them by enchantments, and steals their magic clothing. The last
is an important point; the clothes may be stolen while the elemental
is bathing; she will then be fully domesticated and bear children,
but only so long as the clothes are kept from her. The Sun marries
human girls. In Grozdanka (Dozon 13) we encounter the wide¬
spread story of Dummy. Grozdanka is virtuous. For nine years
she keeps the silence befitting a young bride, to the great annoyance
of the Sun, who is about to take another bride when she speaks and
shames the pert new fiancee. In a Greek folk-tale this motif is
united with that of The Girl who went to War, to make a double
plot. The Christian supernatural mingles with the pagan. It was
the curiosity of St. John the Baptist which gave one ‘samodiva* the
chance of recovering her clothes and escaping. Then there are
magic and transformations. A snake may come from a magic flute
and bite the player; a girl may challenge a nightingale in song as
well as a ‘samodivak To escape his mother's reproaches a boy
would like to become an eagle and fly to Malamka’s garden; two
lovers would like to become forest trees, so as to be sawn into
planks for one bed. The fountain is the rendezvous, and a ring and
'girdle are guardians of virginity. Curses are a "wasting sorrow to
many : to the girl who is doomed to be infertile till fishes sing, and
one who falls ill for nine years, to the husband who has no children
by his wife, and to Koico who was married for only two days.

The numerous lovers' ballads shade off imperceptibly into true
lyrics. This is especially so in ballads of momentary situations.

34 ° BALKAN BALLADS

such as the first meeting and the incidents of courtship. I n
A Lover s Wreath (Karadzic, i. 334 ) we learn merely that a girl
gave her wreath to a boy, and he went to arrange the wedding.
There are some interesting tests for lovers: a contest of reaping'
swimming a river, or a mountain scramble (Karadzic, i. 252, 738'
730). The Yugoslav form of the ballad which shows how much
better is kind than kin is entitled Pavle Zec'anin (Karadzic, i. 289)
Pavle finds a Turkish woman’s pearl necklace and hides it in his
bosom. He returns home and says there is a snake there. His
family refuse one by one to take it out, and only his fiancee dares.
One young man goes through the well-known diving test, and
another has to get things that are impossible. In Sister tests Brother
(Karadzic, i. 301) the sister asks to be ransomed from the Turks;
when she is refused she says (to our surprise) that she is a queen.’
Then there is the girl who sends falcons to three kings and picks
the lover whose answer is most ardent; this ballad has also drifted
Into the Marko Kraljevic cycle.

In ballads of seduction the characters often bear Moslem names.
One of the most favoured motifs (Erlangen 55, 130, 191) is that of
the young nobleman or prince who falls sick for love; various ladies
visit him, but he is not cured until the lady of his choice comes.
He seizes her, and kisses her for three days and nights. There are
two denouements. In one the lady outwits the trickster and forces
him to marry her. In the other there is an aggrieved husband,
Ali Pasha, who complains to the prince’s father:

‘God be with thee, good my lord the Sultan,
still thou sendest not my lady pasha,
one who seemeth like a quail for beauty/

But the Sultan this reply inditeth :

‘Cease thy folly, good my lord the Pasha,
and thy seeking this thy quail for beauty
now enarmed by my grey-green falcon/

A clever woman finds means of evading the dishonest proposals of
her brother-in-law, and a bride contrives to kill the escort who
would take advantage of her on her bridal journey (Karadzic, i.
743 )* As^ for ballads of adultery, the most fortunate has been
Asanaginica (Erlangen 6). There are several versions of this piece.

ne came to the notice of the great Goethe, and passed from him
to the hands of Sir Walter Scott. In this way it played an im¬
portant part in the diffusion of the fame of ‘Morlachian’ balladry.

YUGOSLAVIA, BULGARIA 341

Justly or unjustly, Asanaga suspected his wife, beat her and sent
her home; then he languished for love of her, but she would not
return. The version known to Goethe opens finely with the sick¬
ness of Asanaga:

Say, what whitens on the grassy hill-side ?

Is it snow-drift? are white swans a-flying:

V ere it snow-drift, snow long since had melted;
were it white swans, swans away had fluttered.

Not a snow-drift, not white swans a-flying,
but pavilions of Aga Hassan Aga,
where he sickens, where is sorely wounded.

To his comfort mother comes and sister,
but Ms true love shame forbids to cheer him.

So, with gashes healed and wounds a-closing,
he unbending ordered thus his true love:

‘Ne’er attend me in my whitened mansion,
in my mansion, nor among my people. 5

Repudiated in this fashion, she soon has new suitors, and the
bridal train passes Hassan in his house. Her heart cannot bear the
thought of her two sons:

Mere spectator then was Hassan Aga,
summoned softly to his side two children:

‘Hither come ye, oh my luckless orphans!
your misfortunes cannot win compassion
from your mother, from a heart so stony. 5
Stood and listened Hassanaginica,
white of feature to the ground she stumbled,
by the roadside yielded up her spirit,
slain by anguish, looking on her orphans.

(©uric, viii. 10.)

A group of ballads concerning a false wife punished for her
falseness should probably be centred on Bulgaria and dated from
an epoch before the extant Bulgarian ballads arose. The tableau
is generally the same: a warrior is travelling with his bride when
he is attacked by enemies, with whom she sides; some chance gives
him back his weapons, and he takes due revenge. This is the matter
of the Yugoslav ballad of Grujo Novakovid and Popovic Stojan
among those of bandits (Erlangen 117, 71), and of Banovid Sira -
hinja among those classed as historical and ancient (Karadzic, ii.
42). The Bulgarian Iskren and Milica (Dozon 34) is of the later
haiduk type, and notably ferocious; and in Russia this tale is related

342 BALKAN BALLADS

in the imaginative ballad of Mihail Potyk and his faithless super¬
natural bride, the white swan Marja Lebedaja. As for passionate
crime and incest, the usual types are current: murders for jealousy,
unnatural motherhood, poisoning, incestuous proposals, and the
like. In The King and Queen of Buda (Karadzic, i. 615) we have
the motif of the Curioso impertinente. Wishing to know which of
her three husbands the queen has loved best, he learns that he is
not only the last but the worst.

The tragedies of love are so numerous that they can only be
indicated generically; Trouble arises from the opposition of
parents or the intervention of powerful rivals. The simplest,
naivest tableau is that of the youth who is forced into a hateful
marriage, so that he dies and his girl with him ( Omer and Merjem ,
Erlangen 65). Another is hanged, another decapitated; the fiancee
follows with the same death. In Latin Andro and maid Marica
(Erlangen 56) the scene is rather more original. When Andro died
Marica’s grief was so deep that she put off her suitors for years.
Even when she had at last consented to be remarried, she could
not pass Andro’s grave without a sigh; which roused the disgust
of her new husband so that he killed her on the spot. The true
lovers’ hands met beneath the mould and a green apple was in
them; from Marica grew a rose and from Andro’s side flowed cool
waters. The ‘maumariee’ theme also is well known in Serbian
balladry.

A remarkable version of the theme, originally perhaps French, of
the girl who feigned death to save her honour, or at least an unwel¬
come suitor, is entitled Hercegovinian Stepan (Karadzic i. 727, 728).
It is the ruthlessness of the tests, which are like those of Young
Radoica, that distinguishes this ballad from its western congeners,
but link it to the Czech piece previously mentioned. Neither fire,
serpent, nor tickling with a moustache can make the pretended dead
girl quiver. As a sign of its provenience one may note that in other
ballads Hercegovinian Stepan is said to be a Latin. There are also
several more playful ballads of baffled suitors, in which a youth
tries to waylay a girl, but she rides off unharmed. One girl dresses
like a newly wedded woman, another like her brother, another
like a young prince and demands the ravisher’s services in her
retinue. Marko Kraljevic, Todor of Pomorje, and Stbjan Jankovic
were named as heroes of the adventure of the warrior’s home¬
coming, and there are comparatively slight differences in the telling.

YUGOSLAVIA, BULGARIA 143

A pleasing feature of Stojan’s return is that he reveals himself bv
singing a little allegorical ditty:

‘By her dear nest watched the pretty swallow,
truly watched it til! nine years were ended,
but this morning she begins to rend it;
comes a-guesting here a grey-green falcon
from the high seat of the honoured sultan,
and he wills not that the nest be scattered. 51

In the treatment of crimes the Yugoslav poets have maintained
generally the discretion of the best periods. They speak of typical
horrors rather than of the Newgate Calendar. One such, entitled
Woeful Janja (Karadzic, i. 668), has spread as far as Czechoslovakia,
where it is entitled The Robber's Bride . A special feature of Balkan
poetry’ is the record of Turkish extortions. On one occasion a
couple of tax-gatherers are shown as trying to collect poll-tax from
a cock, on the ground that the cock is" probably a Christian. The
Sale of Bogdan's Wife uses this motif of extortion for preamble to
the tale of the man who sold his wife to one who proved to be her
brother (cf. Politis 86 a). How Turk preys on Turk is shown, in
Asanaga's Palace (Erlangen 142).

There are the usual surprising recognitions, as when the queen
recognizes in her slave her brother. A feminine bandit, who had at
last been reduced to enter a pasha’s harem, rebukes the other
women for their wantonness, and Old Ceivaris Zlatija (Duric, vii.
4) is a good ballad of the girl who went to war.

What has been said of the ballads of Yugoslavia is generally true
of those of Bulgaria, though with some differences of detail. The
collection brought together by the Miladinov brothers closely
resembles in style and content the Hercegovinian ballads of the
Erlangen manuscript. There is the same absence of the nobler and
more archaic poetry of the Montenegrins, the same prominence of
Marko Kraljevie in his more novelesque appearances, the same
weakness in history. In the Miladinov collection there are
variants of The Founding of Ravantca and Busan's Wedding , the
latter transferred to the Bulgarian Tsar Sisman (47, 57), who has
another ballad also to his credit. It is not impossible that in some
such case the Bulgarian form may be older. Eight ballads prove
that the Bulgar interest in medieval Hungary was as keen as the
Serbian, and that John Hunyadi and Matthias Corvinus were their
typical heroes. But there is no solid core of ballads definitely

344 BALKAN BALLADS

important for the nationality of Bulgarians. In this respect the

Bulgarian corpus is characteristically late. It is late also in respect
of a certain coarsening of fibre and exaggeration of the cruel and
horrible. To cover the anonymity of many heroes the poets repeat
the single name Stojan. Stojan suffers all the adventures, and has
no personal attributes that would interest us in a man of flesh and
blood. The thin line between tragedy and horror is not sufficiently
observed. Ballads are often cruel; but in those of the best epochs
the cruelty is not italicized. The modern Bulgarian ballad-monger,
especially those known to Auguste Dozon, is wont not merely to
kill, but to butcher. A stale smell of blood hangs round too many
of his wares. To some extent this must be laid at the door not of
the Bulgar but of his pitiless oppressors. Injustice and extortion
were rampant in the villages, leading to savage revenge. The girls
were abused by young Turks, and the boys imprisoned and hanged
when idle khanums happened to ogle them from the harem lattices.
The gendarme is all the government they know, and his actions
are entirely arbitrary; he is satisfied so long as he discovers some
victim to punish. Christian women file past us in the ballads in
long chains of prisoners, nursing a bitter hatred against the Turk.
Yet what is particularly depressing in these ballads is the attitude
of acquiescence, showing an impoverishment of the soul. Even the
outlaws dream of no more than ultimate subjection. Neither
personal liberty nor national spirit inspires them so much as thirst
for blood and booty. To rob is to augment the family property;
all boys and some girls pass through a few profitable years of
brigandage. When captured, the outlaw concentrates his efforts
on making a good appearance on the scaffold.

Lying between Serbia and Rumania and Greece and Rumania,
Bulgaria lies in the track of several international ballads. Its
Master Mano (Miladinov 424) is undoubtedly related in an
intimate way, as source or copy, to the Master Manole of Rumania,
and its Lazar and Petkana (Miladinov 100) to the Rumanian
Voichitd . Sick Doicin (Miladinov 88) corresponds to the Rumanian
Donald (Alecsandri 29) and to the Ivo Karlovic of the Erlangen
manuscript (no). The Kraljevic ballads which have extended to
Rumania have also crossed Bulgarian territory. A number of
Bulgarian poems are concerned with surprising recognitions,
particularly of sisters by brothers who have felt an innate love for
them (Miladinov 76, 87, no, 135,120); and there is a considerable

YUGOSLAVIA, BULGARIA 345

cycle of ballads about prisoners. One of the former, Pavel and his
Sister (Miladinov 135) is related to a Ukrainian ballad reproduced
by Antonovic and Dragomanov (63). The supernatural ballads
concerning the Moon and Sun, which are somewhat richer in
Bulgaria than Serbia, form a link with Rumania, where one of the
best is on the love of these two bodies {Hernia Cosanzeana ,
Alecsandri 9). Among the classes of ballads the authorities on
Bulgaria include one of shepherds’ songs (ovcjarski) alongside
those of haiduks, and this prominence given to shepherds prepares
us for the remarkably pastoral nature of the Rumanian pieces.

The Bulgarian metres are also intermediate between those of
Serbia and Rumania. The heroic decasyllable is less used than the
later and more lyrical octosyllable; Rumanian lines are octosyllabic
or shorter. There is no trace of rhyme in Bulgarian work. An
interesting feature is the use of irregular lines. They are mostly
to be found in ballads of the supernatural from eastern, Macedonia
(Dozon’s appendix). Some of them represent merely failure to keep
the eight syllable count, but in others the irregularity seems to be a
prosodic principle (The Fountain of the Samomle , Dozen, App. 10).
It does not correspond to the archaic ‘bugarstice’ known to the Ser¬
bians, but is rather a wildness that can only be regulated by the chant;
that is, the sort of normalized unevenness which we encounter
in the Ukrainian ‘dumi’ and the Russian ‘bylinyk There seems no
frontier dividing the music of Bulgaria from that of South Russia.

3. Rumania

Rumanian balladry began its career under the impulse of the
burst of Serbian creative genius in the sixteenth century .The £ gus-
lari* were the fathers of the flautari’, and many of the ‘balade’ are
derived from Serbo-Bulgar 'pesme* or are of the same general type.
The correspondence is closer with Bulgaria than with Serbia. In
versification, for instance, the Bulgars already show a preference
for short lines rather than long, and this preference is absolute in
Rumania. A number of special circumstances, however, have con¬
spired to give a peculiar flavour to the songs of the Rumanians.
They are a Latin-speaking people hemmed in between Slavs. In
the 'green leaves’ formula (by which poems or sections of poems
begin with an appeal to the green leaves of some tree or plant) we
find something remarkably like the appeal to flowers in Italian
‘stomelli’. The Istro-Rumanians are indeed in contact with Italian

4615 y y

34& BALKAN BALLADS

culture. The Macedo-Rumanians in the south are to a large extent
nomad shepherds and mingle with Greeks and Slavs. They have
given the word ‘Vlach’ to their neighbours’ vocabularies. A Greek
‘Vlachic song’ (Politic 236) is composed from the standpoint of a
herdsman threatened by the klephts. The pastoral tinge is the most
characteristic feature of all the best Rumanian ballads, and is their
special note in the concert of European folk-poetry. The shepherds
(ciobani) have for neighbours the haiduks, as in Bulgaria; the
seasons govern them both with equal severity, and any chance may
cause them to change from the one class to the other. A third
group of Rumanians—those of Ardeal or Transylvania—took the
lead in written literature under the influence of the Reformation
and of their neighbours the Saxons of Siebenbiirgen. Thanks to
them the Rumanians were exposed to western cultural influences
of a German type, and it was doubtless also due to them that the
‘lautari’ learned to rhyme. In the last place, the separation between
lyric and narrative, which had been set up in Yugoslavia and ob¬
served fairly well in Bulgaria, was obscured in Rumania. The
‘balade’—or rather, to give them their popular name, ‘cantece
batranesti’ 4 old songs’—are narratives impregnated with lyricism.

The clue to Rumanian folk-poetry is thus provided by the lyrics,
and these may be much older than narrative verse. The favoured
lyrical form is the ‘doina’. In one case, it is true, we may see how
a narrative has degenerated into a lyric; that has been the fate of
the finest of Rumanian ballads, Mioritd , in Transylvania. Such a
transformation would not necessarily prove that the narrative man¬
ner is, in all other instances, the older. Existing ‘doine’ express
timeless, nameless emotions, and give no hint of the causes of their
begetting. The bulk of them are doubtless quite recent im¬
provisations. A few short lines suffice; the singer unburdens his
heart by the mere act of singing:

The wind of Spring blows on the moors
as I sing doine out of doors;
to soothe my soul my song avails
mid flowerets and nightingales.

When winter comes tempestuous,

I sing my doine shut in house,
to fill the daytime with delight,
the daytime and the weary night. 1

1 C. Tagliavini, Antologia Rumena } Heidelberg, 1923, No. 95 (a).

RUMANIA 347

So the £ doina’ has the spontaneity of the Lithuanian £ daina\ The
emotions released are those that come from solitude, unconditioned
by company and without definite source. A sense of longing with¬
out objective is the Rumanian's "dor’, and when this passes to dis¬
content it is his ‘urat’. The only other presence is that of Nature,
and Nature—wooded hills—appears in incidental notes. He gives
a vague background, by addressing the green leafage of the nearest
trees, either at the beginning of his song, or in its pauses :

Green leaf of the blackberry—

no boy kindly thinks of me!

Green leaf of the shamrock green—

no such girls as we are seen.

So Trunza verde* or £ foaie verde’ is a refreshing background to all
their songs.

In addition to these general lyrics there are those for specific
occasions. Those devised for weddings are of special charm. The
' bridegroom's friends come with pomp to the bride’s home, saying
they have been sent by the 'young emperor’, who is out hunting; he
has seen the spoor of some fair game that has taken shelter in that
house, or perhaps he has seen a bird and sent his huntsmen to take
it. There are songs to dry the bride’s tears as she leaves her home;
she, who cannot return to her flower garden, will herself become
a garden of flowers. These songs invoke £ Lado\ the Slavonic Eros;
the name of the god of luck survives in the £ leru, lerom’ of the
‘colinde’ or Yule songs. They wish luck to the person feted, and
beg for cakes or refreshment. The influence of the Nativity causes
some of them to embark on a short biblical narrative, which is
usually interrupted. The 'bocete' are laments screamed by pro¬
fessional mourners. The Greek round dance is accompanied by
the artless Tore’, in which the words are few and the sense
negligible, while the shouts of the dancers are sometimes formal¬
ized into a refrain. There are songs for rain-making and for the
ceremonial crowning of the last sheaf in a field, and. a mass of
charms against disease or to favour truel ove. There are even
rudimentary lyrical dramas—the £ irozii’—which have sprung from
the story of the Twelfth Night.

This lyricism invades the narrative poems also, obscures the
.. exposition and infuses the whole with an indefinite emotion. It
goes naturally with the other chief characteristic of. Rumanian

34§ BALKAN BALLADS

balladry: its pastoralism. 1 Many of the best ballads have shepherds
for heroes, and many more are touched incidentally by pastoral con¬
cepts. There is an exquisite dialogue between some bystander and
a dead shepherd, quoted by Densusianu:

Leafage green of flowers three,
little shepherd of the sheep,
where has death encountered thee ?

‘On the summit of the hill,
which the winds with lashings fill
and the firs are never still. ’

By what death, say, didst thou die ?

‘By a lightning-crash on high.’

And who raised the funeral cry?

‘Little birdies chirping by,
raised for me the funeral cry/

Who was there the corpse to lave ?

°Twas the rain that earthward drave,

’twas the rain the corpse did lave/

Who laid thee out in seemly wise ?

°Twas the moon, when she did rise,
laid me out in seemly wise. ’

Who has set the candles there ?

‘’Twas the sun that rose in air,
he has set the candles there/

Who has brought thee to the tomb ?

‘Three big pine-trees overthrown,
they have brought me to the tomb/

Where is now thy flute so fine ?

‘Up there on a branch of pine,
when the thrashing winds are strong,
then my flutelet sings its song,
and together flock my sheep,
over me a vigil keep/

The Dead Shepherd is a thing perfect in its kind. The supreme
example of the pastoral ballad is Mioritd ‘the ewe-lamb’, of which
Densusianu has printed twenty versions, together with another
twenty reduced to the status of a lyric. The variations are great,
and it is hardly possible to single out any version as itself perfect;
but they all convey, in greater or less degree, the essential poetry
of this creation. The subject is the murder, real or imagined, of one

1 D Densusianu, Vieata pdstoreascd in Poesia noastrd populard , Bucharest
1922-3.

RUMANIA 349

shepherd by two others who envied him his fine flock. Making use
of the superstition that sheep can be second-sighted, and that some
of their cries portend death, the poet or Ms successors have been
able to remodel the narrative, reducing it to its quintessential pathos.
It becomes a dialogue between the doomed shepherd and Ms lamb:
‘At set of sun,
there’ll be murder done
by the Ungurean
and that Vranceand
‘O lambkin mine,
so wondrous fine,
must I be killed
in fennel field,
tell the Vrancean
and the Ungurean,
my corpse to hide
in a grave beside
the fold for the sheep,
for aye to sleep
by the sheepcot here,
that my dogs me heard

So the short, sobbing, monotonous verses go on, with no thought
of resistance to an inevitable fate. The genesis of this piece may
be ascribed to the south-western comer of Moldavia, where,
especially in the sixteenth and seventeenth centuries, there were
frequent battles between the shepherds as they sought new pastures
in the Carpathian hills. It spread outward in all directions, but the
versions gathered on the circumference are less moving than those
of the centre.

So it comes about that the Rumanian ballads are habitually un¬
heroic. They are so unlike those of Serbia in tone,, that a stirring
note is almost a sure sign of foreign origin. The ballads of shep¬
herds are timeless and nameless. Nature is destiny, and often
cruel, but it is not tragic, since there is no complaint or repining
against fate. The two great ballads —Miorita and Master Mamie —
express an unrepining helplessness in the face of a disaster which
is seen but cannot be prevented. They are melancholy, but not
sceptical or angry; they are optimistic, because there is a funda¬
mental belief in naive goodness. A wistful, momentary grace falls
on songs in which the simplest words are the most moving. 1 The
1 L. Rusu, Le Sms de Vexistence dans la Poesie populaire routnaine , Paris, 1935.

35 ° BALKAN BALLADS

shepherd works in the clearings of the forests on the Carpathian
ridges. Winter drives him to his hovel for shelter, but in spring
his soul is uplifted in unfettered freedom. The same experiences
affected the haiduk, lurking behind the leafage of the forest trees
His sentiments are those of the shepherds. The two classes were
liable to clash (as in the ballad Fulga ), but the outlaw was hardly
more than an unfortunate shepherd, and the shepherd was looked
upon as a potential robber by the governors in the cities. In this
way the poems of shepherds and outlaws form but one mass, and
that is the largest in Rumanian balladry.

These things being premised concerning the essentials of
Rumanian folk-poetry, we may proceed to an account of its rise
and development. 1 The lyrical genres, naturally, do not lie within
the corners of this argument. The narrative style was not available
to commemorate the feats of Stephen the Great (1457-1504), but
it was probably heard shortly afterwards in the court of the Serbian
wife of Neagoe Basarab (1512-21). It was by the munificence of
INeagoe and Despina that the exquisite monastery of Arges arose,
the gem of Rumanian architecture. Popular tradition has preferred
to attribute the building to Negru Voda, the founder of the
principality. As for the manner and circumstances, they have been
supposed to be those of the Greek Bridge of Arta, following one of
the Macedonian versions in which the name of the architect is
Manoli (who flourished about 1659). It is only under strict
reserves that we can include the ballad of Master Manole or The
Monastery of Arges among historical ballads. In the early years
of the seventeenth century the Yugoslav ‘guslari’ were welcomed in
the halls of Polish lords and by the Cossacks of the Ukraine, and
with them went Rumanian ‘lautari’, ‘qui valachica lingua patrium
carmen pleno gutture cantabant’. Their style was parodied in the
preface to Dosofteiu’s Psalter in 1670, and a good many of
Neculce s anecdotes in the early eighteenth century seem derived
trom ballads. The genre was established firmly, therefore, in the
seventeenth century, having taken its rise in the preceding one.

'T~ e ® rst series of historical ballads extended down to the ruin
oi Michael the Brave in 1601. In Boy Mihu (Alecsandri 23) he
goes singing through the wood, and his song attracts outlaws to
attack him. They are Hungarians and led by Ianos, i.e. by John
Hunyadi. The hero is evidently borrowed from Yugoslav and

1 See Note Q, p. 392.

RUMANIA 351

Bulgar balladry, as also the whole unsophisticated tableau. It is
used again, without historical attachments, in Vidra and Faunas
(Alecsandri 27, 8), and has come from the cycle of Marko Kra-
ljevic. The same John Hunyadl is the hero of Jancu Void (Teodo-
rescu, p. 490), which is based on a Serbian, original. The horrid
fate of Constantine Brancoveanu in 1714—slaughtered after his
three sons had been decapitated before Ills eyes—-is the subject of
a ballad that rises to the height of the occasion (Alecsandri 52). In
geneial, however, tne historical element in balladry belongs less to
the -tacts than to the setting. Under the Turkish yoke the themes of
poetry could hardly be national, especially as the principates had
ne\ei yet been brought into union. But .the Moldavians had to
struggle against the Nogai Tatars of the Steppes, and those who
dwelt by the Danube were exposed to the scandalous exactions of
Turkish tax-gatherers. These are vividly portrayed in Tudor el
(Teodorescu, p. 675) and in the ballad entitled Ilincuta Sandului
(Teodorescu, p. 635). Roman Grue Grozovamd (Alecsandri 25) is
an exciting ballad of escape from the Nogai nomads.

Toma Ahmos (Alecsandri 24) portrays the lawlessness which
such oppression brings. It is one of the 'best of Rumanian ballads,
and is characteristic in the passivity, at least at the beginning, of
its hero. Toma is resting at ease when a certain Manea creeps up
on him. Mortally wounded, Toma contrives to push in again his
entrails and pursue and kill his murderer. The poem is akin to
those of bandits, of which the perfected examples are Golea
(Teodorescu, p. 584) and Corbea (Alecsandri 35).

As in Bulgaria, so in Rumania there are some excellent ballads
of a supernatural character. The best is Sun and Moon or Ileana
Cosdnzeana (Alecsandri 9). It exploits, in the interest of aetiolog-
ical myth, the well-known ballad motif of the avoidance of incest;
the poem explains why two heavenly bodies, which always regard
each other, can never meet. The Sun’s outrageous wooing and the
swooning terror of the Moon lead up to a simple situation of
tragedy which only God’s word can resolve. Ralaurul (Alecsandri
4) is about a hero sacrificed to a dragon by his mother’s malison.
The Cuckoo and Turtledove (Alecsandri 3) would be supernatural,
since it attributes speech to dumb animals, were it not rather
to be considered as allegorical. While these superstitions are
doubtless old, the ballads need not be considered the oldest of
their class.

352 BALKAN BALLADS

There are many Rumanian ballads with foreign analogues.
Master Manole and Vmchita (Tocilescu, pp. 18, 139) are un¬
questionably Greek, and probably borrowed directly from the
Greeks in Macedonia. There are more Serbo-Bulgar contacts, due
to the borrowing either of ballads or of the general pattern of plots.
So Faunas and Vidra (Alecsandri 8, 27) employ, as we have seen,
the Serbian motif of the careless hero who goes singing in the woods
and so attracts dangerous enemies to attack him. Doncild (Alec¬
sandri 29) is the Serbo-Bulgar ballad of Sick Doicin, who arises
from his pallet to defend his sister and punishes those who would
seek to profit by her distress. Badiu (Alecsandri 32) is a Bulgar
ballad of escape from Turkish oppressors. In Rada (Alecsandri 31)
the tests of a lover’s fidelity are those exacted in Serbia by a maiden
of Senj. The Rumanians have three ballads of Marko Kraljevic
(Candrea-Densusianu, pp. 87, 120, Flori alese 185). Voinicul
Oleaz (Candrea-Densusianu, p. 73) is the story of the bride sold to
pay the poll-tax to the Turks, which we have already encountered
in Greece and Serbia. Old Novak is a hero to the Rumanians as to
the Serbs (Alecsandri 36, 37). In the Cadi’s Daughter he intervenes
to protect his son Iovita who has stolen the girl from her father,
remarking that boys will be boys!

These are ballads of the Balkans, but there are also a few which
have travelled more widely. The Old Man (Teodorescu, p. 616) is
the Rumanian offshoot of the Monnger saga, Iencea Sabiencea
(Teodorescu, p. 639) of Marianson’s Rings, and Mizil Crai
(Tocilescu, p. 126) of The Girl who went to War. Oancea (Tocil-
rescu, p. 45) concerns an adulteress who instigated her paramour
to kill her husband, but horrifies him so that he kills or abandons
her; thus it runs parallel to the German Lady of Weissenberg and
the Serbian Vukasin’s Marriage. The tests of affection applied to
his family and his fiancee by Petrea or Mircea (Candrea-Densu¬
sianu, p. 102) are those of the Serbian ballads: he pretends to have
concealed a snake in his bosom, and only his fiancee dares to take it
out. The cruelty of a stepmother, as displayed in the French La
Porcheronne, appears also in the Rumanian Alimon {Flori alese 192).
The Ring and the Veil (Alecsandri 7) makes use of the superstition
that intimate apparel can fade when its former possessor is
oppressed or unfaithful. Looking on the veil his bride had given
him, an unnamed youth bethought him to return home; where he
found that his lands were wasted and his bride drowned in a pond.

RUMANIA

He died too. When the pond was drained, the lovers were found
iasped m an embrace. A pme and a vine grow from their dust,

resn^kf 1 ^ ^ eVen The St0ry ?ecni5 inc %enous to Rumania in
respect of the drowning of the girl; but the miraculous elements
are common property throughout Europe.

RUSSIAN BALLADS
i. Great Russia

R USSIAN ballads—‘byliny’ or ‘stariny’—differ from those of
- . the rest of Europe in form, style, and theme. They are
not entirely separate, since there is some evidence that western and
southern elements have migrated into the Russian sub-continent-
yet even in telling the same story the characteristic Russian
divergencies are such that the identification is hedged with doubt.
Nowhere else is there so bold a reliance on the normalizing power
of the melody, allowing the length of lines and disposal of stresses
to show the greatest anomalies. The plots and characters of these
poems are singularly vague. A Russian ballad hero has no more
than an indeterminate temperament, so that the improvisers attach
to each one stories m keeping with this temperament. Dobrynia is
courteous, Il’ja mature and sage, Aljosa is rash, Curilo is young
and gay, Djuk is rich, and so on; but it is not impossible that the
story toM of Aljosa may also be told of Curilo, or that Dobrynia
and II ja should be credited with the same feat. Having satisfied
this modest requirement of decorum, the ballad-poet and even the
reciter may expand or contract the ballad almost indefinitely. He
may simplify the incidents or slow up the narrative by inserting
long tirades of conventional verses. The bounds of the several
stories are unfixed, and divisions made by scholars in the interests
of precision are in fact arbitrary. A strong antiquarian interest
demands that almost all adventures should be brought into relation
with the old principality of Kiev under the rule of Prince Vladimir.
It has preserved many authentic names famous in the eleventh,
twelfth, and thirteenth centuries; but the adventures related are
almost entirely fabulous. Some of the technique of narration and
the basic principles of melody have doubtless been imported into
ussia from the Eastern Empire; but in neither this nor any other
respect is there any such dependence on Greece as to justify group-

Wkh the <tra g° udia ’- The ballad-poetry of Russia
the byhny of Great Russia and the ‘dumi’ of the Ukraine—
demands treatment in a section apart from all others.

radition has caused all the oldest ballads to be grouped in the

GREAT RUSSIA 355

cycles of Kiev and Novgorod, and, by endowing the heroes with
historic names, has coloured the whole genre with an elusive
historicity. The Novgorod cycle contains three pieces and is un¬
connected with \ ladimir of Kiev. The remaining adventures are

supposed to take place in the Kiev of the tenth or eleventh century
even when they are palpably modem or when they bear internal
evidence of having originated in Galicia, Volynia, or even Novgorod.
At Kiev they are now unknown. Kiev is in the Ukraine (or Little
Russia), and the ‘byliny’ are a Great Russian genre. Their hero is
V ladimir^ but there ^ is nothing to show whether the Saint or
Vladimir Monomah is intended, since the actions described are
not appropriate to either. Dobrynja bears a historical name, and
so does Prince Roman, but there were several wearers of those
names in the period in question, and there is nothing in Dobrynja’s
or Roman s actions that corresponds to one model rather than to
another. At other times scholars are able to suggest expeditions
or sieges as possible bases for ballad incidents, but the identifica¬
tion remains vague. The ‘byliny’ tell us almost nothing that is cer¬
tainly true of the old principate, and they tell us much that is
palpably fabulous; they have all suffered revision in the light of
Muscovite prejudices, and their common background is memories
of the Tatar domination from Batu’s invasion to Eraiak’s conquest
of Siberia.

In these circumstances it is well to fix our eyes first on some
undoubted Kievite poetry, so that we may know in what manner
actual events would have been treated by contemporaries. Prince
Igor’s Expedition 1 must be our guide. It is not a ‘bylina 5 , and it is
not quite an epic. The author is fully conscious of his authorship,
which he parades by parodying the style of an older poet named
Bojan. 2 Igor, who died in 1202, was alive when this poem was
made, since its author speaks of the Kievite principate as lasting
from old Vladimir’s time to present-day Igor (ot starogo Vladimira

.Eugene Liatzky, Le Dit <FIgor t Prague, 1934; Siovo 0 poiku Jgoreve (with
three Russian, modernizations), Moscow, 1934* G. Storm’s rendering has

numbered lines.

‘Nacati ze sja t&i pesni po bylinam sego vremeni, a ne po zamysleniju
Bojanju’ seems to mean ‘to begin the song on the events of his own time, and
not in Bojah’s manner’. About i860, scholars took the word ‘bylina’ to be the
name of a genre, and this use of ‘bylina’ is now established, though the peasants
say ‘starina’ ‘old song’. If we remember its rather stupid origin as a technical
term, we shall not be tempted to beg the question of ballad or epic priority, as
its originators did.

356 RUSSIAN BALLADS

do nynesnego Igorja). The nucleus is an account of a raid executed
by Igor and his brother Vsevolod against the nomad Polovcy. W
is the poet’s type of heroic valour; we would rather call it fool
hardiness. He determines to carryout his raid despite unfavourable
omens. With his brother’s help he advances to the Don and skir
mishes on the banks of the Kajala. The Polovcy rally after their
discomfiture, completely surround the Russian forces, and rout
them despite Vsevolod’s valiant efforts. Igor is taken prisoner but
escapes before the poem closes. The language is picturesque' and
v ™~> aS m description of a storm at dawn which is the flight
of Polovcy arrows. Their leaders are Gza and Koncak, and they are
heathens; but the Russians, who are Christians of a sort still
believe effectively in Stribog and Dazd’bog and in Div, the wood-
demon. Round this nucleus there are apparently discordant ele¬
ments. One is evidently an interpolation; it is a hymn in praise of
Svjatoslav of Kiev who actually defeated the Polovcy under their
leader Kobjak, and so is compared with Igor to his disadvantage
Another passage is used to upbraid the Russian princes for their
disunion There are paragraphs dealing with other campaigns
against the Polovcy, and laments for Igor and other rulers Thanks
to these accretions as well as to the work of the first author, the
poem is strikingly wealthy in precise information about persons.
Vladimir I falls just outside the poets’ view, but scores of his suc¬
cessors are named, so that an elaborate genealogical table is needed
lor the understanding of the references.

All this is very different from the style of the Kievite ballads.
In them all the history of Kiev is telescoped into the reign of a
single Vladimir, without our being able to say which he is of two
(one died m 1015, the other in n25). His consort is Apraksja,
corresponding to the Eupraxia who fell into Batu’s hands in 12^7.
Igor s soldiers fight against the Polovcy, and there is not a whisper
of those Tatars of the Golden Horde who have completely taken the
place of the Polovcy in the ‘byliny’. The geography of Igor’s
Expedition is precise, and one can fix the frontier of the Kievite
mg om a little to the east of the Sula, a tributary of the Dnepr;
but the byliny’ have Kiev as an imaginary centre and the unlimited
steppes for a circumference. It is plainly to be inferred that
nothing corresponding to the extant ballads would have been
imaginable in the time of the author of Igor. Though, on his
testimony, the style of Bojan was different, and apparently (since

GREAT RUSSIA

he offers one or two parodies) more floridly imaginative, it is even
less possible to believe that the notes of time, place, and person
^° u a V een anachronistically applied bv a contemporary of
the Aladimirs. The Kiev cycle, if it represents the literature of the

court o Kiev In any wav, must be regarded as the fiotsam and
jetsam of a foundered poetry, recovered In a later age, pieced
ogether and expanded with the utmost licence of creation.

i wo things, however, may be affirmed concerning this Kievite
court poetry. The Expedition of Igor springs partly out of eoinician
odes addressed to the heroes, as we know”otherwise to have been
an established custom. Narrative details were not necessary, since
the event was fresh in the memory of the troops which improvised
t em; the important thing in them was the name of the victor.
Laments for the dead also preserved the memory of a name.

ence, when the by liny came to be composed, there were at hand
many famous names and surnames, unaccompanied by precise
details of their lives. Those who bore the same name would be
confused m one personality, since the ballad poet had neither the '
means nor the inclination to make distinctions. Dobrynja, the
uncle of Vladimir I, and Dobrynja of Rjazan who perished on
the Kalka in 1224, make one ballad Dobrynja; Aljosa Popovic was
one of the seventy champions who died on the Kalka; Roman in
t e ballads Is a compound of two different Romans separated by
the length of the thirteenth century; Gleb Volod’evic is probably
derived from the names of two allies who attacked Korsuii in 1077,
namely, Gleb Svjatoslavic and Vladimir II Vs tvolodoznc^ though the
ballad also contains the name of Marinka, taken from the wife
of the Pseudo-Demetrius, the seventeenth-century usurper. Simi¬
larly, in the Novgorod cycle, Vasili! Buslaev was probably the
posadnik’ who died in 1171, and Sadko was the rich Sodko or
Sudio who built the first stone church in the city in, 1167. Such
identifications are open to dispute, and the scholar Is warned not
to pursue them Into too meticulous detail; but, taken globally, they
make it certain that many of the names occurring in ‘bylinv* are
historical and belong approximately to the epoch of the Vladimirs.

That such names should survive through encomia and laments
is In keeping with the habits of Rurik and his companions, who
were accustomed to similar improvisations from their skalds. The
Scandinavian practice doubtless stimulated or blended with the
native ‘velicanija’. Addressed by a retainer to a chief, such out-

358 RUSSIAN BALLADS

pourings were careful to preserve the name, and to indicate the
ground for praise or sorrow, but not to go into narrative particulars.
They would adequately account for the presence of so many
authentic names and the absence of authentic narratives. A Ukrain¬
ian piece, with the name left in blank, though late, may be repre¬
sentative of the genre:

Famous, fair, renowned N . . .,
what the deed that brought thee glory ?

Dusk—and on horse his saddle’s laid;

day—he alights at Tsarigrad,

and fights and fights with Tsarigrad.

Out comes the tsar, so sore afraid,
and burghers hasten, counsel take,
if any gift his ire may slake. 1

The poem goes on to describe the offer of a tribute of horses or
gold, but the hero refuses to accept any other gift than that of a
princess under her wedding crown. The editors conjectured that
this piece might be a reference to Svatoslav Igorevic; it is one of
several which have to do with sieges.

The other fact about which we may rest assured is that there
developed at Kiev, before the end of the twelfth century, a con¬
siderable heroic legend concerning the champion Il’ja. It is as the
representative hero of Russia that Elias von Russen appears in Ortnit
and Iljas of Greka in the Thidrekssaga. The latter takes no im¬
portant share in the action, but is associated with Vladimir (Valde-
mar) as well as with a certain Osangtrix (Oserich), who is Dietrich’s
principal antagonist in the east. Oserich of Russia is a figure of the
Biterolf; Hertnid (Ortnit), of Novgorod, transferred to Lake
Garda in Italy, became the hero of his own epos, with Elias, his
mother’s brother, for his chief courtier and officer. The acts of
Elias correspond to none of those in the ‘byliny’, but his name and
fame stood high among neighbouring Germans in the early thir¬
teenth century. Traces of his fame have been found in the Ukraine
and in White Russia 2 of the sixteenth century—neither a district

1 V. Antonovic and M. Dragomanov, Istoriceskija Pisni Malorusskago Naroda,
Kiev, 1874, i. 4.

2 White Russian folk-verse is of a highly lyrical character. It is exemplified
in P. V. Sein, Bilorusskija narodnyja Pisni, St. Petersburg, 1874, and discussed
amply by E. Karskii, Geschichte der weissrussischen Volksdichtung und Literatur
(Grundriss der slavischen Philologie), Berlin, 1926. I have some references to
narrative poems in this dialect, especially to those contained in Bezsonov’s

GREAT RUSSIA 359

where ‘byliny 5 are at home. He was then known as Il’ja Morov-
lenmn , which stands close to the Finnish form of the name,
Muurovitza ; but the surname in the ballads is always Muromec.
Though there have been conjectures that the German Elias or
Eligas may rest on Oleg or VoVga, and that Muromec may derive
through Murmanjanm from ‘Northman 5 , yet Il’ja is the con¬
spicuous case of a hero of ‘byliny 5 whose person has resisted all
attempts at historical identification. His legend, as known at the
end of the twelfth century, must have been heroic fiction, and
more likely than not in verse. The adventures attributed to him
are numerous, and though some are demonstrably late and others
are discordant, they do adumbrate a kind of poetical biography, 1
that is, some sort of epos.

The narrative and prosodic art of Igor's Expedition resembles
that of the ‘byliny 5 , without identity. The lines of both are free
from rhyme and syllable-count, being measured only by the chant.
While it is risky to dogmatize in the absence of adequate studies,
the music of the ‘byliny’ and Ukrainian ‘dumi 5 is apparently
Byzantine, and a continuation of the style practised in the eastern.
Balkans. The device of negative comparison is established in the
‘byliny’, in Igor , in Serbo-Bulgar poetry, and in the ‘tragoudiak
There are fixed epithets in the Kievite epic, but there is no such
use of reduplication or delaying devices as in the ‘byliny’. On the
other hand, the author makes use of the Scandinavian trick of
apposition, which is not favoured later. The poet indulges in
characteristically Russian hyperbole, though he has no occasion for
the titanic humour of some ‘byliny’. In short, the essentials of the
later style are mostly present in the courtly epos of the late Kievite
epoch, and they are to be found also in the folk-verse of Greece
and the Balkan Slavs; but the manner used in the ‘byliny 5 was not
then complete in all its details.

We may thus address our minds to the ‘byliny 5 themselves, 2
poems which belong wholly to Great Russia, though their oldest

subjects are associated with Kiev and Novgorod, lying outside
Great Russia. Whatever relics of history may be concealed within
them, the ‘byliny 5 suppose a framework which is not that of the

Belorusskija Pesni, a work I have not been able to consult. I have not attempted
in this book any account of White Russian folk-song, but content myself with
offering the reader these references.

1 The cycle is separately studied in O. Miller, IPja Muromec £ Bogotyrstvo
Kievskoe , St. Petersburg, 1869. 3 See Note R, p. 392.

3 6 ° RUSSIAN BALLADS

historical Kiev The centre of all adventures is that city, ruled bv
a dateless Vladimir. There is a ballad to tell us how Dobrynia
brought a wife for Vladimir (Rybnikov 9, Speranskii, DairyJa-
svat), and it is a fact that Vladimir I’s uncle Dobrynja was his
proxy with Rogneda, daughter of Rognvolod (Ragnvald) of Polock
before 980; but in the ballads Vladimir’s wife is always Apraksja or
Opraksja who is to be identified with a certain Eupraxia wlm fell
into Batu s hands about 1237. It is supposed that Vladimir and
his heroes vanished from history with the downfall of his kingdom
before the Tatars. This occurred at the battle of Kalka in 1224
w ich gives the first shadow of a historical fact in the ‘bylinv’'
The event is remembered in a highly novelesque manner in IV ia
and Kahn (Rybnikov 7, &c„ Speranskii, IV ja i Kalin-car), where
the enemy s name may be derived from the battle. The Tatars
unexpectedly made no immediate use of their victory, and it was
iett to Batu to consummate their triumph in 1238-40 Batu’s
prowess is remembered in the ballad of Dobrynja’s tribute-carrying
(Rybnikov 8, Speranskii, Dobrynja i Vasilii Kazimirov), in which
however, Dobrynja’s comrade bears the name of a personage
famous m the Moscow-Novgorod war of 1470-80. In some
accounts of the disaster the enemy leader is said to be Mama!, who
ought 1111380 (Speranskii, IV ja i Mamai) ; and in others a leading
role is played by Ermak (e.g. Rybnikov 7), the conqueror of
oiberia in 1581-4. Muscovite notes and customs are seen through¬
out the byhny’, and indicate creation or reformation during the
fifteenth, sixteenth, and seventeenth centuries. Allowing for all
t at may be due to forgetfulness or wilful innovation, it remains
true that the picture offered by the ‘byliny’ is not that of the
historical Kiev (whatever elements of an older epos may be em-
bedded m them), but of an imaginary Kiev and more real Moscow
under the weakening domination of the Golden Horde, The Horde
is frequently mentioned, as are also the pagan Lithuanians; never
he historical Polovcy. In all these poems the Russian heroes are
Christian and have quite forgotten the old pantheon, though they

SpeClfically Christian impulses. 1 They are collectively
styled bogatyri in deference to the Tatar ‘batyr, bagadyr, bahatur’,

colleSs°of °bvlm P v> em A 3re C3lled ‘l Uh ,° Vnye StiH ’ 3nd are not “ cIuded in
Chadwick’s TraJ/wV accou S t °( 1116111 ls 8 lven in Professor and Mrs.
in this book as ^ °1 L j ter j ture - For the sake of brevity they are not discussed

formallv disrin, 3 ’,;^ ^r ’ ^ W ^ eea ° mitted eIsewhe re except when not
iormaily distinguished from the secular poems.

GREAT RUSSIA 361

since it was the Mongols who developed most fully the professional
champion-at-arms. The word used at Kiev had been fhrabryf
brave. One of the champions bears the name Saur , which is

Tatar for ‘bull’.

Independent as they are, the ‘byliny’ have been open to sugges¬
tions from the west and south-west. Constantinople ( Tsarigrad )
is, for them, a greater city than. Kiev, and its songs are finer.
Dobrynja played his lyre in the manners of Kiev and Constanti¬
nople, when he returned to find his wife about to be married to
another. He was disguised as a ‘skomoroh* ‘juggler, jongleur*; the
‘skomorohi’ correspond to the Byzantine ‘skommarchoi* or masters
of the revels. This is the oldest word for a Russian minstrel, though
it has, like ‘juggler*, sunk since to the level of buffoonery. The
word kalika’ is also used in the sense of minstrel, but it properly
means pilgrim (as in the ballad of the Forty Pilgrims), a class
whose interest in the entirely secular ‘byliny’ can only be deemed
secondary. From kalika’ to ‘kaleka’ ‘cripple’ is an easy stage, and
justified by the traditional association between minstrelsy and
bodily affliction; but neither of these terms is supported by the
text of old ballads, as is the term ‘skomoroh’. Negative com¬
parisons and delaying devices in syntax are also a feature of Greek
and Balkan balladry, held in common with Russia.

The Christian faith, when it came to Russia, was Greek in type.
Michael, Elias, Nicholas are favourite saints of the Eastern Church
and of the ‘byliny’. These make use of Biblical narratives, but they
go beyond the Biblical warrant, using the legends of Joseph,
Samson, Bathsheba, and Solomon as they have been elaborated by
Jewish tradition. 1 Curilo Plenkovic’s beauty is that of Joseph in
Jewish apocrypha:

His locks of gold he tossed them all,

his locks of gold he shook them,

like scattering pearls he scattered them down.

Little girls looked and burst the fences,

young women looked and made their windows tingle,

old hags looked and preened their mantles round them.

(Girferding 223, with lines from Kireevskii, Pesni Sobrannya,
iv. 2.)

. 1 For all that relates to external influences see A. N. Veselovskii, Jiiiko-
Russkija Byliny , St. Petersburg, 1881, and Slavjanskija Skastamja 0 Soiomone i
Kitovrase , St. Petersburg, 1872, together with the summary treatment in

Keitujala.

4615 . ~ A

362 RUSSIAN BALLADS

So the author of the Morisco Poema de Jose describes how women
lost their senses at sight of Joseph’s beauty, and Theodore Pro¬
dromes (Rhodanthe kai Dosikles) describes his hero’s stroll in the
capital. Samson has become one of the Russian champions, and
has three ‘byliny’ to his name, of which the second is the Biblical
story of his wedding. The other two show him usurping the place
of the giant Svjatogor. He plays a part also in the early chapters
of the Thidrekssaga. David’s procuring the death of Uriah is
allegorized in Dobrynja’s shooting an arrow at Marina’s ‘dove’, who
Is the dragon Tugarin (Speranskii, Dobrynja iMarina), a story which
is not older than the seventeenth century. Solomon is, as usual, the
type of the sage beguiled by women. These matters receive a more
lush development than the Bible warrants, and so resemble (it is
said) Talmudic tradition. Critics have thought this due to the
Khazar Jewish khanate established in South Russia, but it may be
unnecessary to argue In favour of a specific and ancient Jewish
influence on the ‘byliny’, since, as we have noted, these develop¬
ments are found In Greece, Germany, Spain, and elsewhere.

A definitely Greek influence is proved by the use of the name
Etmanuil Etmanuilovic in some ballads, since this hides the name
of the emperor Manuel Comnenus who conducted active intrigues
in Galicia about the year 1196. He boasted his descent from the
Doukas family, of which Digenis Akritas was the most famous
scion. The epos exists in a Russian translation (. Devgenii ), but
apparently has no importance for the study of Russian ballad
origins. Doukas, however, becomes Djuk, a ballad hero; and his
surname of Stepanovic seems to be an acknowledgement of St.
Stephen of Hungary’s prowess. The Indian kingdom of Djuk
Stepanovic, which so far eclipsed that of Kiev, corresponds to the
kingdom of Prester John, in which Manuel showed much interest.
From the Alexander legend the ballad-mongers perhaps took the
idea of giving the magician Vol’ga a snake for father; Nizami
reports a conversation between Alexander and a ploughman of
exceptional beauty, which offers an analogy to the conversa-
tion between Vol’ga and Mikula (Speranskii, VoVga i Mikula).
Veselovskii devotes a whole essay to working out the correspond¬
ences between the ballads of Saur Lebanidovic and the Cappadocian
ballad of Amouris. The Tugarin, who was overcome by Aljosa
Popovic (Speranskii, Aljosa i Tugarin ), has been identified as the
Tugorhan, chief of the Polovcy in the late eleventh century, who

GREAT RUSSIA 363

became reconciled to the rulers of Kiev; this piece has strongly
influenced a similar ballad concerning II’ja .Muromec (Speranskif,
IIjo, 1 IdoMce), of which the earlier group of versions set the dud
in Constantinople. Il’ja’s battle with his son (Speranskiy Bo! IHa
s synoni) calls to mind the Greek ballad of Tsamados , as also the
history of the Esthonian Kawi-ali, of the Persian Rustem, of the
German Hildebrand, and of the ancient Ulysses and Telegonos,
without it being possible to assign a precise source for the Russian
version.

The Greek stream crossed the lands of the Southern Slavs. In
some particulars the Russian minstrels resemble those of Bulgaria.
The importance they assign to snakes and to swan-women, who
are witches, is perhaps common Slavonic; and the same mav be
the case of the approximation between the marvellous ploughman
Mikula and Marko Kraljevic, who ploughed up the sultan's high¬
way. The Czechs also had their ploughman-hero: Pfemysh The
ballad of Mihatlo Poiyk Ivanovic is a complex of many motifs, but
the most important have Bulgarian parallels. 1 he name belongs
to the St. Michael of Batak in southern Bulgaria, who fought with
a lake-infesting dragon who exacted a tribute of children (in fact,
who borrowed a feat from St. George, as St. George had done from
Perseus). His relics worked miracles, and they were removed in
1206 to the Bulgarian capital at Tirnovo; the date must serve as
the 'terminus a quo’ of his reputation, in foreign parts. One of
Mihailo Potyk’s adventures is thus a dragon-fight; the other motif
is that of the faithless wife who sides with her husband’s enemy.
She is a Solomonic and Samsonic character; but the working out
of the Russian ballad is closely akin to the Bulgarian Jskreti and
Milica and the Yugoslav Banovic Strahinja.

We have seen that the Germans who composed Ortnit and the
Dietrich romances were interested in Russian oral literature to the
extent of learning that II ja and .Vladimir were representative
figures. It is the same group of romances which offers a number
of parallels to the fictions of the ‘byliny’, and though the affinities
are not so close as to be decisive, they sum up to the probability of
a definite German influence on their formation. With Ortnit go
in the Heldenbuch also Hugdieirich , IVolfdietrich, and the Rosen -
garten ; the whole group of poems is preoccupied with the legendary
history of Greece and Italy (though Ortnit is so only by the
accident that Holmgard, the Old Norse name for Novgorod, had

3 6 4 RUSSIAN BALLADS

been supposed to refer to the Garda Lake). With the adventures
of Dietrich or Thidrek, in that saga, there go also several other
romances or strings of adventures: the affairs of Vilkinaland and
Pulaland, i.e. Lusatia and the Ukraine, are a main interest.
Samson’s history is drawn into the Dietrich complex, together with
that of Detlieb or Detlev the Dane, who grew very slowly into
heroism, and then, taking leave of his parents, set out to conquer
many robbers, especially the bandit Sigurd. Among the ‘byliny’
we find Samson, and the slow-developing Il’ja, whose first exploits
are against robbers, especially the great highwayman Solovei the
Nightingale. Hildebrand is a leading figure in the Thidrekssaga
next to the king himself; his one great adventure is repeated in the
career of Il’ja of Murom. Svjatogor, who cannot lift a small purse
which in reality contains the weight of the earth, or Vol’ga in other
versions, resembles the Scandinavian Thor who could not lift the
Midgard Cat. The ballad of Dobrynja and Aljosa is certainly re¬
lated to the German Moringer ; and the ballad-novelette in which
he casts doubt on the chastity of the Petrovici’s sister (Speranskil,
Aljosa i sestra Petrovicei) reproduces in part the Imogen story, of
Italian origin; it is to be found also in the Greek ballad’ of
Maurianos, so that there is no way of determining how it entered
Russia. Its foreign origin is, at least, assured.

Into another set of parallels I am unable to enter. Among the
Tatars and Georgians of the Caucasus tales have been encountered
precisely corresponding to the plots of leading ‘byliny’. The
instances are given by Keltujala, but he does not inform us why
we should consider them models rather than copies of the Russian
poems. Concerning these latter we can affirm that the genre is old,
even though we may not be able to assign dates to individual pieces.
The byliny undoubtedly reached their apogee in the sixteenth cen-
tury, when the ballads of Ivan the Terrible were composed in the
best style; they were substantially complete by the seventeenth.
The Caucasian parallels were gathered after the ballads had been
assembled by scholars in the nineteenth century, and we have little
cause to deem them old. They are, one might say, too close; too
close to allow for the divergences which lapse of years brings in
oral tradition. Their existence may, therefore, be but one phase
of the dispersal of the ‘byliny’.

For the ‘byliny’ have been dispersed from their centre, like the
ballads of other European countries, but their migration has

GREAT RUSSIA 365

covered a far wider area and is thus more imposing. In. Kiev and
the Ukraine, the original home of their heroes and of some of their
motifs, they are quite unknown. A new kind of ballad-poetry, of
western origin, reigns in their stead. There may never have been
Ukrainian ‘byliny*, but only a more courtly kind of verse. The
ballads are Muscovite, but they are not now to be found in any
numbers in Moscow. The institution of serfdom seems to have
been, almost as fatal to them as education has been elsewhere. It
drew the classes apart; the rich sought other amusements, the poor
invented vulgar ballads of buffoonery. The ‘byliny’—so flourish-
, ing in Ivan the Terrible’s capital—now live vigorously to the north
of the great forested belt, where no great accumulations of wealth
are possible, but a hard and healthy struggle for livelihood occupies
all men equally. There, in the long winter pause, the visit of the
blind singer brings an audience from all the district. It is not a
paid profession, since the singer generally has some other; but the
entertainer is rewarded with gifts, and still more with universal
respect. So it is by the shores of Lake Onega, in Olonec, Archangel,
Perm, and Siberia that the best ‘bylinv’ have been gathered;
byliny which record the names of personages dead eight centuries
previously, together with some particulars of the geography and
history of the Ukraine. Other ballads have been gathered on the
line of the Volga, and the tales circulating in the Caucasus complete
a magnificent arc drawn from the ancient capital of the Vladimirs.

The historical school of interpreters, identifying names in the
4 byliny’, see in these poems a storehouse of ancient facts. Another
two schools study the plots, which are fabulous; they differ accord¬
ing as they find analogies in the west or the east. A fourth group
was encouraged to apply mythological methods by the enigmatic
figures Volh or VoPga, Mikula, and Svjatogor. Volh or VoPga
(Rybnikov 3, 38, &c.) was bom to a young princess and a snake
father in a green garden when the sun first shone. He was a prodi¬
gious child:

Now when Volh was seven years old,

his little mother sent him out to learn,
and in a twinkling learned he grammaiye;
she set him down to write with a pen.,

and in a twinkling writing came to him.

Now when Volh was ten years old,

by that time Volh had learned the highest cunning;

366 RUSSIAN BALLADS

the first craft he had learned
was to enclose himself in a glittering falcon;
the second art he had learned
was to enclose himself in a grey wolf;
the third art Volh had learned

was to enclose himself in a wild bull with golden horns.

So when Volh had reached his twelfth year,
this Volh began to gather himself a meinie;
he gathered his meinie three years,
he gathered to his meinie seven thousands;
he himself—Volh—was in his fifteenth year
and all his meinie was fifteen years old.

(Speranskii, Volh Vseslav'evic i IndSiskoe Carstvo.)
He went to, Kiev, and there learned of the threat of invasion by
the Indian’ tsar. Shifting his shape, he went to the tsar’s tent,
overheard his plans; shifting again, he gnawed his weapons so that
they were useless; and, in another disguise, safely returned to
Russia. The same shape-shiftings were used by Prince Roman
Mitreevic in defeating a Lithuanian invasion (Rybnikov 45 ). In
another ballad Vol’ga (Speranskii, VoVga i Mikula) encounters the
prodigious ploughman, who can plough faster than a horse can
gallop and who encloses the weight of the world in a small bag.
The same bag is found in a ballad of Svjatogor (Speranskii,
Svjatogor i tjaga zemnaja), who is the hero of his own marriage’
of an encounter with U’ja of Murom, and of death by burial
alive.

The historical expert sees in Vol’ga the historical Oleg, who
made a great mid on Constantinople in the tenth century. It was
a naval expedition, and so utterly unlike that of Vol’ga, who is
evidently concerned with the defence of Kiev against the Tatars;
nor are the enemy precisely Tatars, since their ‘Indian’ Empire is
that of Prester John borrowed from the ballads of Djuk Stepanovic.
As a shape-shifting hunter he corresponds to Vanamoinen in the
Kaleyala and to the whole scheme of sympathetic magic implied
by Finnish hunting songs. His name of Volh seems to be connected
with the Volhov river, and with the worship of the thunder-god
Peiun. In one of his exploits he resembles Thor; in his general
character he seems to be a fire-god like Loki. As for Mikula, his
name is borrowed from St. Nicholas, but his qualities are those of
a Slavic Triptolemus. Svjatogor is obviously ‘Holy Mountain’;
but which? If it be a place-name from the Black Sea coast, near

GREAT RUSSIA 3 6 7

the mouth of the Donetz, it would lie in Khazar territory. Svjatogor
and Samson tend to change roles in the ‘bylinv’, since both are
giants, the one a pagan and the other a Hebrew. The historical
interpreters go so far as to see in Svjatoeor a type of the Jewish
Khazar kingdom (‘obraz hazarskago carstva’, as KeltujaSa says).
But such identifications are very doubtful. Svjatogor might as
easily be a personification of Mount Athos.

Older critics were inclined to see in these poems an older
stratum of the ‘byliny’, since the heroes are only loosely related to
Kiev. II ja inherits some of the mountainous strength of Svjatogor,
and Dobrynja marries Mikula’s daughter Nastasja. This view is
now abandoned. There are doubtless old elements in the ballads,
but they themselves may be comparatively modern. That of Volk
and the Indian Empire must be younger than Djuk , and Svjatogor’s
association with II ja comes after that hero's rise. A hero named in
the ballads is Kolvvan, who has the qualities of Svjatogor-Samson.
He is evidently the Esthonian Kalev or Kalev’s son, and the name
is derived from Finnish ‘kallio’ ‘rock, cliff’.

The cycle of Il’ja of Murom, apart from his encounter with
Svjatogor, contains some seven main branches. First comes his
cure. For thirty-three years of his life he had been unable to walk.
Cured by three pilgrims (as Martvn was cured by St. Gleb and
St. Boris), he took leave of his parents, giving them proofs of his
heroic strength. He made three forays against robbers, and then
attacked the great Nightingale, Solovei. In this part of his career
the hero has had the same history as Detlieb. A famous duel is that
he fought against Sokol’nik, or, in other versions, against his own
son; a second he fought against the giant Idolisce, in Tsarigrad
according to the older group of ballads. Unjust treatment at
Vladimir’s court caused Il’ja to withdraw in indignation, but he
fought a tremendous battle against Kalin and the Tatars, when
most of the Russian ‘bogatyrs’ succumbed. The robber Nightin¬
gale was associated with the stream called Smorodina (where there
were place-names Solov’inoe and Deyjatidub’e) in the region of
Chernigov. IFja is clearly not an inhabitant of Kiev, and his
attraction to the Kievite cycle may be a reflection of the close
association of these two cities in the twelfth and thirteenth
centuries.

Dobrynja Nikitic and Aljosa Popovic are two heroes who are
represented as younger than the ‘Old Cossack’ of Murom, and

368 RUSSIAN BALLADS

somewhat inferior in strength. Dobrynja is pre-eminent for
courtesy, but Aljosa has a less reputable career, perhaps owing to
the prejudice against priests’ sons. The two names were associated
by their death at the battle of Kalka in 1224: Dobrynja of the
Golden Girdle came from Rjazan and Aleksandr the Priest’s Son
from Rostov. There was another Dobrynja, uncle of Vladimir I,
who may also be commemorated under this name. He acted as
Vladimir’s proxy in his marriage to Rogneda of Polock, just as
Dobrynja does with Apraksja in the ballads. Some eight feats stand
to the credit of Dobrynja; he is, indeed, the centre of a ballad-
cycle as complete as that of Il’ja. First he leaves home to fight a
serpent; then he is intercessor for Vladimir with Apraksja’s father;
thirdly he kills the witch Marina’s ‘dove’ and suffers enchantment
into a bull; then with Vasilii Kazimirov he takes Russian tribute
to the Golden Horde, to Batu, and performs remarkable feats of
strength in the presence of the Tatars; fifthly, he returns to find his
wife about to marry Aljosa Popovic, and prevents the match; then
he himself marries the Amazon Nastasja—it is not quite certain
whether Dobrynja had one or two wives, since the ballads are not
congruent; and seventhly and eighthly, he fights duels with Il’ja
and with DunaL Aljosa, for his part, fought against the snake-hero
Tugarin and made him a friend, and he cast doubts on the virtue
of the Petrovici’s sister, when the brothers boasted about her in
their cups. Tugarin is apparently the historical Tugorhan of the
eleventh century, who was reconciled to the rulers of Kiev and
entered into a marriage alliance with them; but Batu is of the thir¬
teenth century, Vasilii Kazimirov of the fifteenth, and Marina of
the seventeenth. The marriage of ‘bogatyr’ and amazon is a theme
encountered in Caucasian legends, and also in Greek ‘tragoudia’.
The killing of Marina’s lover corresponds to what Jewish tradition,
rather than the Bible, says concerning David and Bathsheba’s hus¬
band. The dragon-fight may be borrowed from Aljosa’s chief
exploit. The Moringer and Imogen motifs are exploited in one
episode of each hero’s life. A Ukrainian ballad uses the Jonah
motif for Aleksei Popovic (Antonovic and Dragomanov 44). Curilo
Plenkovic and Djuk Stepanovic are associated figures. They are
both Galicians, the one supreme for beauty, the other for wealth.
When brought into competition the victor is Djuk. Some memory
of the former remains in Galicia (Antonovic and Dragomanov 18)
in an enigmatic lyric. Zurilo is richly attended by maidens:

3^9

GREAT RUSSIA

Goes Zurilo out of gates,
after him three hundred maids:

Wait, Zurilo, wait, my lord,
where’s your faithful troop and ward ?

Then follow verses that seem to have wandered in from another
song, but Zurilo appears at the close, accompanied by a certain.
Ksenja, keeper of the stack-yard :

Everywhere Zurilo goes,
barley com. springs up and grows,
underneath fair Ksenja’s feet,
sprouting up are ears of wheat.

The ‘byliny’ dwell on the prodigious display of wealth whereby
Curilo bought his place at Vladimir’s court, and the beauty by
which he rose too rapidly in Apraksja’s favour. His death came at
the hands of a certain Bermjata, earned by an, intrigue with the old
man’s wife. As for Djuk, the Kievite court were inclined to laugh
at his display of wealth, and Curilo was particularly pettish; but
an embassy to the ‘Indian’ Empire discovered that Djuk had
spoken no more than the truth. The ballads of these two heroes
took form probably in the fifteenth century.

There are many minor heroes in the Kievite cycle, some of whom
have been attracted from other towns to Kiev. In Suhman , Dunai,
and Don and Nepra there is an aetiological element, since the events
recorded explain the origin of streams. They are otherwise
associated with Kiev; the third is an imitation of the second, and
the second is usually worked into the cycle of Dobrynja ; the first
appears to be based on the battle against Marnal and his Tatars in
1380 on the Kulikov plain, by the river Neprjadva. Saur Lebani -
dome (also known as Mihail or Ivan Danibmc) is not actually
attached to the Kiev cycle, and has wandered in from Greece
{Amour is). Vasilii the Drunkard , who saves Kiev from the Tatars
by suddenly leaving his potations, bears some relation to the
siege of Kiev in 1240 by Batu, but has suffered revision at Mos¬
cow in the fifteenth or sixteenth centuries. Mihatlo Potyk comes
in part, as we have seen, from Bulgaria. Prince Raman, who
cannot be exactly identified, but was one of two princes of that
name reigning in Galicia and Volynia in the thirteenth and four¬
teenth centuries, has three ballads to his credit. Other heroes are
GISb. Volod’evic, Mihail Kazarinov, Ivan ‘the merchant’s son’,
Hoten, and Stavr. Stavr (Rybnikov 14) came to Kiev and fell into

. 4615 .3 b

37 ° RUSSIAN BALLADS

boasting of his wife, so that Vladimir detained him and would not
set him free till she came in person. Being wise, she concealed her
sex. She came in the character of an envoy from Greece, wrestled
with all the heroes and overthrew them. She then demanded to
wrestle with Stavr, and after conquering him she made herself
known. Her strength was due to her race, for she was Vasilissa
daughter of the prodigious Mikula. Stavr of Novgorod is known
to have lived in 1118. The adventure that befell him is that of the
German Alexander von Metz, who is called in the ballads of Ger¬
many and Scandinavia The Count of Rome. The Russian, however,
was not yoked to a plough, and his wife did not disguise herself
as a minstrel. In the Danish ballad Sister frees Brother (UDV 170,
172, Arwidsson 97) there is no mention of ploughing, and the lady
relies in the Russian fashion on brute force. These stories are
probably related, but it would be hard to say how. The legend of
Joseph and Zuleikha, followed by that of Benjamin’s cup, gives us
the pilgrim’s ballad of Forty Pilgrims (Rybnikov 13, 40, &c.,
Speranskii, Sorok Kahk s Kahkoju) ; and there is a strange history
of Solomon in Tsar Solomon and Solomon’s Father’s Dream
(Rybnikov 49, 50, 94), of Jewish origin.

Owing to the attraction to Kiev of heroes like Stavr, the
Novgorod cycle has only two names: Sadko the Rich Merchant,
and Vasili! Buslaev. Both have been surely identified in the twelfth
century; the one was an administrator, the other left a permanent
record of his wealth in the stone church he built. But the atmo¬
sphere of their adventures belongs to the city at the height of her
commercial prosperity in the fourteenth and fifteenth centuries, or
even to her decline in the sixteenth. Their ballads are more satisfy¬
ing than those of Kiev inasmuch as the plots are firmer. Sadko,
the Rich Merchant, was at first a fiddler who gained his wealth
from the tsar of Lake Ilmen, he then equipped a splendid trading
fleet for a supreme venture. His sins, like Jonah’s, raised a storm,
and he was cast into the sea. By his marvellous playing on the lyre
he kept the Tsar of the Sea dancing so long that he had to release
his victim. In prose tales Sadko learns the value of salt. The ballad
exalts the worth of the minstrel’s profession above that of rich mer¬
chants the more effectively for being set in commercial Novgorod.
Sadko’s early poverty contrasts brilliantly with his later wealth, which
is described with oriental hyperbole; but his wealth was gained by
his playing and later preserved by it from total loss. The opening is:

37*

GREAT RUSSIA
And so it was in famous Novgorod,
and such a player on the lyre was Sadko,
but to him there was no store of boundless wealth,
but he merely attended noble banquets,
so made he merry for merchants and for boyars,
he gave them pleasure in their noble banquets.

For one whole day Sadko was invited to no noble banquet,
for a second day also he was not invited to a noble banquet,
and for a third day was he not invited to a noble banquet.

So now this is what occurred to Sadko,

Sadko betook him to Ilmen, to the lake,
and sat him down on a blue burning stone,
and began to play upon his clear-sounding lyre,
and played from mom till eventide. " (GiTferding 70.)

When we remember that the Finnish magicians were accustomed
to sit on a special stone to work enchantment, we are the less sur¬
prised that the tsar of the lake should appear to Sadko and make
him the richest man in Novgorod. Collecting this wealth for a
great trading venture, Sadko took ship with his companions, and
a storm burst upon them. To calm the tempest, Sadko was cast
overboard with his ‘gusli’, and sank to the temple of the Tsar of the
Sea. The Tsar asked him to play, and Sadko’s playing is the
Russian version of the widespread motif of the power of music:

And so began Sadko to play on his clear-sounding lyre,
and so now began the Tsar of the Sea to dance in the blue sea,

and all the blue sea went a-capering with him,

and a wave swept down on the blue sea

and began to shatter so many black ships on the blue sea,

and many rich merchants perished on the blue sea.

As for Vasilii Buslaev, he stands for the civil strife endemic in

Novgorod. An overbearing noble, he gathered a meinie (druzina),
and insulted the rich citizens at his banquets. A revolt against him
was led by his godfather, a monk from the Andronova monastery".
The description of this personage, wearing the huge monastery-
bell for a helmet, and of the encounter with his godson, is an
excellent sample of Russian ballad humour:

Then seized he a bell, four hundredweights heavy,
seized it and clamped it on Ms head,
and sat him then upon his goodly steed,
so went he out to the little stream the Voikov,

372 RUSSIAN BALLADS

to that bridge that spans the Volhov
and he called loudly on the bridge of the Volhov,
and he clamoured in a thunderous tone:

‘Ho thou Vasil’jusko Buslav’evic,
thou little child, do not gallop far,
take not the road straight ahead.’

So then Vasil’jusko made reply:

‘Ho thou then, ho my godfather,
in thy old age art thou so noisy!

I gave thee no egg on Christ’s Easter-day,
but now I give thee an egg on St. Peter’s day.’

So swiftly Vasil’jusko leapt in under him,
and smote him with his club upon the bell,
and shattered the bell into two fragments;
and the next blow he let fly at his head,
and beat him down into the Volhov stream,
and with him beat down his goodly steed.

(Gil’ferding 54.)

In another ballad, Vasilii makes a pilgrimage to Jerusalem, com¬
pleting it successfully; but he obstinately insists on jumping over
a great stone, and kills himself thereby. A suggestion made is that
this pilgrimage and death may have been modelled on Ivan the
Terrible’s pilgrimage to St. Cyril’s monastery in Novgorod in
1570—an event perhaps celebrated in some strictly historical poem.

In addition to the novelistic matter in these two cycles of Kiev
and Novgorod, there are a few narratives among the ‘byliny’ which
are not given a quasi-historical setting. The untold Dr earn (Rybnikov
35) is pure folk-tale. A son dreams that his father will drink his
bath-water, and is so shocked that he refuses to explain to his father,
to a baron, or even to the tsar, being finally imprisoned; then when
the tsar goes on an expedition, the youth saves his life and marries
the princess. Nine Brothers and their Sister (Rybnikov 37) has a
plot akin to the Scandinavian Herr TrueTs Daughters (UDV 164).
Nine robbers attack and dishonour a girl who proves to be their
sister. Nastasja Politovskaja or The Youth and the prudent Woman
(Rybnikov 33, 47) is distantly related, perhaps, to the Sp anis h
Count Velez and Count Cloros, since the ballad hinges on the boast-
ing of a youth in a tavern. He claims to have made a conquest of
the princess, whose chamberlain he is, and it is only by her inter¬
vention that he escapes execution. Mitrii Vasil’emc and Domna
Alexandrovna (Rybnikov 92) is a private tragedy of a type common

GREAT RUSSIA 373

In the west of Europe but rare In. Russia. Domna freely mocked
her wooer, but was ready enough to marry him, despite the bad
omens her mother perceived. The omens proved to be true; she
died as Mitri! approached the house, and her mother died with
her. Mitri! could only kill himself, reproaching (for some reason,
or other) his sister for being the cause of the disaster. There are
a number of other odd ballads of this sort, some of them being
frankly vulgar and full of buffoonery. They represent the decay
of the genre in central Russia.

Thanks to the fact that romantic matter has been taken up by
the cycles of Kiev and Novgorod and given a setting which Is, In
some part, undoubtedly historical, the notion of a historical ballad
is somewhat ambiguous in Russia. There are, however, many
important byliny’ which arise directly out of the events they relate,
so that not only the names or some details can be verified, but the
whole transaction. As in other countries, these historical pieces
give us the chronology of ballad composition. The "byliny 5 of the
older cycles, we have seen, could not have arisen before the down¬
fall of Kiev, though they may rely on ‘velicanija’ or courtly narra¬
tives of an older age; nor do the majority of them reach their present
development before the fifteenth or the sixteenth century. This
same development is seen in the historical pieces. The first we
encounter Is The Princes of Tver (Speranskll, Scelkan Dudenfevic ),
referring to an event of the year 1327. The style of such poems is
more simple and literal than in the old cycles, but the Influence
of these upon historical matter is evident in the three pieces on
Prince Roman of Galicia, which have become entirely novelesque.
Avdofja of Rjazan (Rybnikov 182) doubtless refers to the Tatar
sieges of 1365 or 1377, though the city was attacked also four times
in the fifteenth century. It relates that Bahmet the Turk destroyed
Kazan (meaning Rjazan) and took forty thousand prisoners.
Avdofja resolved to save at least one of them, but her wit charmed
the sultan into releasing them all.

It was with the great tragedies that marked the reigns of Ivan the
Terrible and Boris Godunov that the historical * byliny y reached
their highest development in style and theme. These pieces are
worthy to stand beside the best work of the older cycles. The cap¬
ture of Kazan in 1552 opens this series, followed by the more
domestic themes of the tsaritsa’s death (1559), Ivan's remarriage
(1561),. the attempted slaying of the tsarevich (1.581), and the

374 RUSSIAN BALLADS

Terrible’s death (1584). The flight of the Crimean Khan in 1572
and Ermak’s conquest of Siberia in 1581-4 are the most important
external events remembered by the ballad-poets. The attempt on
Fedor’s life (Rybnikov 19, 31,55) is the most impressive of all the
historical ballads, uniting as it does the utmost horror at the
depravity of the crime with breathless suspense as to whether it
will be actually perpetrated. Credit for stay of execution was given
by some later revisers to the Romanov family. What appears to be a
transformation of this ballad is the remarkable Sunflower Kingdom
(Rybnikov 36), which only Rjabnin among the modem reciters
could remember; it may, of course, be an independent myth, but
the names suggest that it is history reduced to romantic legend.
Tsar Vasilii Mihailovic, by the engines given him by two artisans,
flies to Sunflower Land and there has his secret amours with the
princess, who is Marja the White Swan (known from the ballads
of Mihailo Potyk and others). Thus Fedor Ivanovic is born—
according to the name, his father would be Ivan, not Vasilii: in his¬
tory Vasilii was Ivan’s father. Fedor returned to Russia bearing
the insignia of his royal birth, and entered the taverns to carouse.
At this time the tsar was about to marry Anna Dmitrievicna, and
had issued a curfew order to discourage brawls. Under this order
Fedor was apprehended, and about to be executed, when the tsar
recognized his insignia, remembered his old love, and took to him¬
self the Sunflower Princess, leaving Anna to Fedor. The name of
Mary or Marina became the conventional name of a witch in this
period, being derived from Marina Mniszek, the False Demetrius’s
wife; it intruded itself into the oldest cycles. The age of Ivan the
Terrible was not completed until the downfall of Boris Godunov,
and the line of great ballads includes Grigorii Otrep’ev, or the
False Demetrius, of 1606, one on Suisldi in 1610, and a touching
lament in the mouth of Ksenja, Boris Godunov’s daughter.

After this great epoch the line of historical ‘byliny’ extends with¬
out interruption to the middle of the nineteenth century; but they
have not the old power. In the seventeenth century they celebrate
the capture of Azov (1637), the parliament which saved Smolensk
(1653), the expedition to Riga (1656), the rebellion of Stenka Razin
(1670), &c. In the eighteenth century the reign of Peter the Great
is covered by ballads as numerous as those of Ivan the Terrible,
but of a more vulgar tone. Though a monarch suited by temper
and achievement to be a hero of popular tales and songs, Peter was

GREAT RUSSIA 375

responsible for introducing those western and literate influences
which split the community of rich and poor who had delighted in
the byliny of an earlier age. His birth, suppression of the Streltzi,
his victory of Poltava (also sung by a Finnish minstrel), and his
end, are the principal points of his cycle.

After Peter the Great the wars of the eighteenth century con¬
tinued to furnish material for ballad-journalists; for instance, the
Swedish war of 1743 an< 3 the first Turkish war of 1769. The cen¬
tral figure is that of Catherine II, who was also the first Russian
ruler to extend the web of her intrigues into the Balkan peninsula.
There Is a somewhat naive Bulgar ballad about her (Dozen 42), in
which she is said to have defeated seventy-seven discourteous beys,
who did not let her have time to do up her hair. To these pieces
correspond also the numerous Ukrainian political poems of the
same period which have been collected by Dragomanov. 1 The
ballad-minstrels were still active when Napoleon I launched his
great attack on Russia. The "dog of an enemy’, king Napoleon,
is represented as collecting an army and ships, and sending a letter
to the tsar to demand a lodging in the Russian palace; the tsar
might have acceded to the demand, but for the indignant refusal
by Kutozov; Kutozov promises to prepare delicacies of bombs and
cannons for the French guest.

2. Ukrainia

An account of Russian narrative verse cannot be closed without

some mention of the "dump and other ballads of the Ukraine, 2 If
only because of the strange paradox that overshadows them. The
Ukrainians are the original Russians (Rus), though the name has
been commuted in part of the territory to ‘Rutheniansk Thirty
million strong, they have their own Little Russian dialect, and
occupy not only the south of Russia but also parts of Poland, and
the former Austrian Galicia. In particular, they hold all the ground
consecrated by the Kievite cycles of Great Russian ‘ byliny \ The

1 M. Dragomanov, Politicni Pisni Ukrajinskogo Narody^ xmii-xix sh, 2 vols.,
Geneva, 1883.

2 References are to V. Antonovic and M. Dragomanov, IsUmcesMja Pisni
Malonisskago Naroda , Kiev, 1874. M. Dragomanov also published two volumes
.of historical poems dating from the eighteenth and nineteenth centuries, which
have been already mentioned. There are some useful notes on Ukrainian folk¬
song in S. Smal-Stockyj, Ukrainisches Lesebuch (Sammlung Goschen), Berlin-
Leipzig, 1927.

376 RUSSIAN BALLADS

‘byliny 5 profess to tell us tales of the Ukraine, which are almost
wholly unknown in that region. At most mere wisps of that tradi¬
tion can be grasped. There is some vague memory of IPja
associated with place-names in the Ukraine. Zurilo or Dzurilo,
hero of a dancing song, is probably to be identified with Curilo
Plenkovic, and both with a family of boyars who took a prominent
part in the affairs of Przemysl, Cholm, Galicia, and Podolia from
the end of the fourteenth century to the beginning of the seven¬
teenth (18). There is a game in dialogue in which mention is made
of a Prince Roman (12). This Roman may be the Galician prince
Roman Mstislavic of the early thirteenth century, who (if not
another) may have been the hero of the ‘bylina 5 of Prince Roman
and Mary the White Swan. But these are slight indications, and
they do not imply any real knowledge of the Kievite cycle in the
country round Kiev.

The fact is that the Ukraine suffered the worst shock of the
Tatar invasions, so destructive as to blot out nearly all memory of
the glorious past. Her folk-literature begins a new epoch with the
ruin of the capital in 1240. Unable to find a new centre of gravity,
the Ukraine did not recover until the Polish kings began to inter¬
vene by reason of their forward policy in the fifteenth and sixteenth
centuries. They brought with them the seeds of a definitely western
culture. Rhyme is a characteristic of most Ukrainian folk-songs,
even those of narrative cast. The lyrical element predominates,
as in Poland, and there are several narratives of an obviously
international type. Ukrainian metrics have some interesting
features, but none that mark them off from the general practice of
western Europe, outside the ‘dump. There are the usual love-songs
(boy and girl, man and wife, hate, betrayal, death and lonely sur¬
vival). Other pieces follow the calendar of ceremonial occasions:
Christmas (koljadki), New Year (scedrivki), May Day (vesnjanki),
Easter (gaivki), Midsummer (kupaPni pisn'i), Harvest (obzinkovi
pism), together with marriages and funerals. A conspicuous fea¬
ture is the number of dance songs (kolomiiki, kozacki, sabadaski,
cabaraski), as in Poland. There are various lengths of line, from
octosyllables to sixteen-syllabled lines, but all this verse is subject
to measure and to rhyme, generally in couplets. Whatever elements
of narrative there may be, they are interrupted by refrains and by
repetitions and the movement is lyrical.

The ‘dump alone stand out from the mass. They are normally

UKRAINIA ' 377

rhymed in couplets, though the rhyme may be imperfect or fail
altogether; but the length of the line is entirely free. They are de¬
voted to narratives, but are not the only medium for narrative verse.
A twenty-stringed lute (the ‘kobza’) is used for them, the instru¬
mentalist filling in the interstices of the lines with improvisations.
An attempt has been made to devise a system of scansion common
to the ‘dumi 5 and ‘byliny’, but there is here even less evidence for
the recurrence of certain stresses and pauses. The two types are not
identical, but, in view of the metrical irregularity also apparent in
Igor's Expedition , it seems natural to believe that the ‘byliny’ and
the 'dumi 5 are independent derivatives from the old prosody of
Kiev.

The themes of the ‘dump are modem and historical. They are
concerned with expeditions against the Turks, the complaints and
escapes of prisoners, and the Cossack against the Poles. The style
and treatment are generally undistinguished, and even plebeian.
One misses the supernatural, which diversifies the action of some
of the best ‘byliny 5 . Aleksei Popovic and the Storm in the Black Sea
(44) is a Jonah legend, marred by an excess of words. The crimes
to which this unpardonable sinner confessed are trivial—they
include irreverence and failure to say his prayers — and the whole
piece is in painful contrast to the vivid imagination that shapes the
song of Sadko . The name of the hero is also that of one of the
‘bogatyrs’; but it means no more than Alec the Priest’s Son, so that
one cannot be certain of the identity. Aljosa Popovic, however,
was, as we have seen, involved in one or two serious pieces of ballad
wrongdoing. The escapes are those of Samuil Koska from Trebi-
zond (45) and of three brothers from Azov (35), the latter being
raised above the commonplace by a dash of tragedy. The best of
the ‘dumi’ is that of Marusja Bogoslavka (46). The plot is very
simple. The lady has herself turned Turk by entering a Turkish
harem (doubtless as a captive), but on Easter Eve, while her lord
is at prayers, she frees five hundred Cossacks from his dungeons,
sending by them a hopeless message to her parents. There is a
naive but effective use of psychology and indubitable pathos in the
ending; Marusja saved others, herself she could not save :

This request alone I make you, pass not by Boguslav’s town.

To my father dear and mother make this news known:

that my father dear grieve not,
alienate not store of treasure, ground or plot,

461s 3 c

378 RUSSIAN BALLADS

that no store of wealth he save,
neither me, Marusja the slave,
child of Boguslav the priest,
evermore seek to release,

for become a Turk I am, I’m become a Mussulman,
for the Turk’s magnificence,
and for my concupiscence.

It is suggested that the case of Marusja Bogoslavka would be that
of Roxolana (Rossa), wife of Suleiman I, who seems to have been
f P ri esfs daughter from Galicia (d. 1558). Something of the old
‘bogatyr’ spirit breathes in Cossack Golota’s Duel with a Tatar (43).

For the most part we are dealing with fragments in regular
metres on quite general themes. It is only possible to identify a
name or an allusion by chance. An adventurous and romantic
scholarship has thus room to work out identifications, as Chodzko
did, stretching back to a remote pagan antiquity, with the dynasty
of Rurik as almost a middle point in the series. Dragomanov,
though more prudent, assigns his first twenty items to ‘the age of
the “druSiny” and princes’. These pieces are almost entirely
dancing-songs, and not at all easy to interpret in an exact fashion.
Some of them mention names of persons or places. In one the
dancers propose to sail down the silent Don and attack Constanti¬
nople ; others speak of sieges of Constantinople, Lwow, Mogila,
Kamenec, Mier, &c., the Greeks and others being specified as
enemies. If these are references to real events, then they take us
back to dates in the remote middle ages. The classical attack on
Constantinople was that of Oleg in the early tenth century. But in
the absence of narrative context or other guarantees of historicity,
there is nothing to compel us to believe these sieges were more than
figments of fancy. Take, for instance, The Duel with the Turkish
King (6). A Russian hero, it is alleged, seized a Turkish (or Czech)
king and would not release him. Turkish sultans appeared at the
head of large armies only in 1620 and 1672, and no Russian gained
then the ‘spolia opima’. There is no way of connecting the allega¬
tions with any Czech king. The ‘spolia opima’ were won by
Svjatopolk Izjaslavic in 1103, when he captured the Polovcy
leader Khan Belduz. The ballad might be emended in this sense,
but there is not really any reason to believe that this is more than
an imagined case. So, too, with a dance which speaks of the arrest
of some one called Ivan while dancing (11).

UKRAINIA 379

It is more easy to believe that some of the songs which relate to
Tatar devastations may contain a trace of the contemporary' horror
roused by Batu. This form of savagery', it is true, continued to be
present in the experience of Ukrainians until the eighteenth cen¬
tury was well advanced; but doubtless the first barbarians set the
model of pitiless ruin:

On the plain a birch tree lies,

on the birch a cuckoo cries,
questioning the birch I ween:

Wherefore, birch-tree, art not green ?

—Green and sweet how could I be ?

Tatars have stood under me.

Cleft by sword, my twigs down came,
as they stirred the brilliant flame,
with their spears they turned the ground,
by my roots the water found,
watering their horses round.

(SmaJ-Stockvj, p. 71.)

There exists, at all events, a Ukrainian prose legend of the fall of
Kiev (15)- The hero was a certain Mihallik, so strong that he could
shoot an arrow from Kiev into the Tatar camp; the Tatars de¬
manded his surrender so as to deprive the city of its defence; and
when the prince of Kiev was likely' to comply', Mihailik unhinged
the Golden Gates and carried them through the Tatar host to
Constantinople. Another famous bowyer was Baida (40), who,
rather than submit to the Grand Turk, shot three arrows which
killed the sultan, sultana, and a princess. The word means ‘idler,
reveller’. The root idea is thus the same as we find in the ‘bylina’
of Vasilii the Drunkard', but it has been proposed, on not very con¬
clusive evidence, to identify this hero with a ‘starost’ of the Cher-
kassians who did much harm to the Turks of the Crimea, the
Ukraine, and Moldavia between 1556 and 1563.

The remaining narratives are such as might be encountered any¬
where in Europe, and must in some cases be regarded as borrowed
goods in the Ukraine. Some indication has already been given of
them in the paragraphs dealing with Poland, since the legends of
Polish Galicia are also Ruthenian. There is a Ukrainian Girl who
went to War (69), but it is somewhat inconsequential. The usual
conclusion has been lost, and the ballad has been contaminated
with The Duel with a Turkish King (6 c), in which a hero neglects

380 RUSSIAN BALLADS

the excellent advice to keep in the middle of the army, exposed
neither in front nor behind. The ballad exists in White Russia also,
and has closest affinity with the Czech form. The Girl ransomed by
her Lover from the Turk (33) is yet another example of the power of
true love. The Robber Boyar (20) is a bandit whose bride has to
wash out bloody cloths, and sees in one of them her brother’s
hand; the story is found also in Czechoslovakia, and originated
probably in Serbia (Karadzic i. 668, Susil 139). From the Balkans
came The Brother purchases his Sister as a Slave (63); and a Father
(or Brother) betrays his Daughter (or Sister) to a Turk (65, 66) is a
Serbian theme borrowed probably by the Galician Ruthenians
from the Czechs (Susil 128). The cases are distinguished by the
names of the actors, since the father and daughter are called
Andriecko and Mariecka, but the brother and sister are Roman
and Olena. The verses are arranged in octosyllabic couplets.

The Ukrainian folk-songs now extant thus belong wholly, or as
far as sure inferences can be made, to the post-Kievite era; some
may be contemporary with the great Tatar raids, but otherwise the
oldest evidence is for the sixteenth century. Brought within the
orbit of western civilization by the Poles, the form of their ballads
is occidental, apart from the ‘dumi’. The themes, save those based
on facts, derive from the west or from the Balkans, and in either
case a determining factor has been the contiguity of Czechs and
Ruthenians in Galicia.

NOTES

Note A

S. B. Hustvedt, A Melodic Index of Child's Ballad Tunes , Berkeley
(Cal.), 1936; see the notice in Modern Language Review , xxxiii, 1938,
pp. 295-7. For my own convenience I have introduced some modifica¬
tions into Mr. Hustvedt’s system, due partly to a wish to simplify it
typographically, since it ought to be printed easily, typed without an
abnormal keyboard (though I allow myself foreign accents), and written
by hand without risk of ambiguity; modifications partly due also to the
examination of tunes which are not composed within the limits usual in
English balladry.

The names of notes may be obtained by differences of type and
capitalization whether in print, on a typewriter (preferably with red and
black or blue ribbon), or in handwriting. The middle octave may be
represented by Roman letters: c d e f g a b. One can then obtain the
names of notes in two octaves below and two above by using large
capitals (or underlined capitals), small capitals, roman, italic, italic
capitals: C —B c— b c —b c — b C — B. Some of the notes of the highest
octave are used by Pelay Briz for his Cansons de la Terra, Barcelona-Paris,
1866-87. Generally speaking the range required falls within two octaves
named c—b c — b.

The whole tune may then be transcribed by stating (i) the key, (ii)
the time, (iii) the unit length of note. A rest of any length is indicated
by V.

The key or mode can be stated simply. As for the time, Hustvedt

recommends writing the two numbers side by side. For the more
complicated Balkan times, involving more than two numbers, a dash or
vertical line would be needed between the halves of the fraction.

The unit length is represented by U4 U8,. &c., 4 being a crochet, 8 a
quaver. Most ballad tunes are best transcribed in these units, but one
may also use on occasion a semibreve (U2) or semiquaver (U16). Each
letter of type will therefore be of the length of the unit chosen; for
instance, under U8, V is a quaver of that pitch. Dots will lengthen the
note by one unit of length. Under U8, ‘c. 5 is a crochet, and ‘c .. .’a semi¬
breve. To write notes of half the unit length, put them in brackets; ‘(c)*
under the same circumstances is a semiquaver. It is not usually desirable
to show too many brackets in a transcription, and the main thing to be
kept in mind is to give as brief and clear a formula as possible. In some
eastern tunes there are very rapid trills, which may be omitted for some
purposes, or may be entered in double brackets if necessary. There are
notes spaced with intervals a little greater or less than ours, which may
be indicated by putting a point above or below the letter. There are also
optional trills, of imprecise shape, which may be represented : by a line
drawn over one or more notes.

Bars must be entered. Vertical lines serve when the bar is certain (]),

382 NOTES

dots when uncertain (:). The latter kind are typical of the oldest ballad
tunes m the east and west. Hustvedt indicates sharps and flats and
naturals by the usual musical signs. These are not on most typewriters
and I find that some confusion may arise through the fact that on staff
notation and with Mr. Hustvedt the accidental is entered before the note
but in usual transcription it comes after. I think accents may serve The
acute (') easily represents a sharp; the grave (') a flat; and the tittle (~) a
natural, since it was originally a superimposed ‘n’.

By these means most simple melodies may be printed or typed within
the dimension of a square inch; an enormous saving on the space occupied
by staff notation.

The tune so transcribed is characterized by a melodic contour. This
contour can be set down in a quasi-algebraical way by counting each
rise and fall in semitones, ignoring identical notes. By using roman and
italic (or underlined) numbers one can get symbols for twenty rising
semitones: i-o i-o, where V is ten and ‘o’ is twenty. The alphabet
gives symbols for twenty-four falling semitones, omitting I and O
as liable to confusion with numbers. Actually , one only needs about
twelve.

These formulas may be used for the whole tune or any part. For pur¬
poses of comparison it is enough to give the curve of the first phrase,
corresponding to the first line of quatrains or, sometimes, the first hemi¬
stich of long divided lines. In a general catalogue Mr. Hustvedt recom-
mends entering in numerical order first, and then in alphabetical order*
so tunes in n . . 12 . . 13 . . &c., then Ai . . Az . . and later

AA . . ., AB . . . There would have to be cross-indexing, however, for it
happens that the first interval is often the most capricious. Singers make
different jumps to reach the same continuing pitch. In this way tunes are
sometimes identical, except for the first interval. I am not quite certain
whether all tunes which have identical melodic contours under Hustvedt’s
transcription are in fact identical. The details can be ascertained by con¬
sulting the full transcription.

Personally, I find it convenient to indicate in transcription by a blank
where each line ends. To make the above clear, I give a transcription and
characterization of God Save the King ( = My Country ’tis of Thee ) as it
would appear on a card-index.

God Save the King 3C12
Church Hymnary 631
1 sharp 44 Ua
gga/f (.g)a/
bbc/b(.a)g/

agf/g../
ddd/d(.c)b/
ccc/c(.b)a/
h(cbzg)/b(.c)dl
(ec) ba/g../

NOTES
Note B

3 %

4 dvk M d l nat l r IS 18 an ex P Iana tj° n ‘obscuri per obscuriora’.
A considerable number of sung ‘byliny’ have been collected, but their
interpretation is still indefinite. Still the ‘byliny’ singers actually cope in

epkminstrTk T* * ^7 SUperficiaI,y similar *>' that of the Spanish
iLTfT r > S ° me , Senera notlons may be P ut forward tentativelv. The
onlv nl y 7 are Sh ° rter than ^ Rabies, or longer. If shorter, ihere is
onty one important accent m the line: the last. If longer, there are two.

' Kors g f’® a complicated account of their metre, with the aid of

P m 7 iZ ir?! S ’ - m 1118 VeHkorusski J a p &”i zapisannija dla Ricarda

o T ' er c?n g ’ T 7 ’ 3nd appIkd if to Prince in his Slozo

o Polku Igoreve, St. Petersburg, 1909. I have not succeeded in under-

standmg the system. Professor S. Konovalov, who has a markedly

^ifm Ca ma ? ner , ° f P ronoun «ng ‘byliny’, accents the 3rd and gth
syllables, or 3rd, 7th, and 1 ith, according to the length of the line. This
regularity is obtained by adding unaccented syllables or omitting them in
certain cases, and there is no doubt that it corresponds in that respect

to the practice of the popular singers. So

V slavnom velikom IMovegrade

becomes in recitation

"V slaavnom velikom Novegrade,

with main accents on the 3rd and 9th. There^are many lines which are
sung to an irregular time, but with two bars, which rather resemble the
two chief stresses of Yugoslav ‘pesme’. The first of these stresses may
not occur, but still the sense of measured utterance is maintained by the
second, as also doubtless by subtle adaptations of tempo. Each line is
treely improvised. Generally speaking, the melodic curve rises to the
first bar; it then oscillates on two or three notes, and descends after the
second mam accent to a point lower than the beginning. Thus Gil herding,
Onezskija Byliny, St. Petersburg, 1873, No. 73, records hearing the minstrel
Rjabnin sing VoVga i Mikula as follows:

4 fiats irregular U8

g.fff.gf/g.f.fi.r/
g.fffedc/d.c.cr
cc/eededccd/c. .bb. .r/

(e..e)edfdd/c..BB.r/

g.f.f.gf/g.fff.

ff / gffgfedc/d.c. cr

cd/(e.e.e)ecddcd/(c,.B)BBBr &c.

Mussgorskii has given a similar transcription, pitched somewhat higher.
The Russian melody is more developed than the simplest Spanish ballad
tunes, and so a fortiori more than the Spanish epics may have been. They
may have had the tonal austerity of Montenegrin heroic songs. But the
Russian practice, shows how music of the most formless sort may suffice
to impose form on irregular words. It is a warning also against’looking
for formal elements (e.g. four-accent verse) where they need not have
existed.

3§4

NOTES

Note C

This legend is complicated in the Balkans by the presence of Akritic
poems of a similar content. They refer to the Kidnapping of a bride who is
identified, on epic authority, as the mother of Digenis Akritas. In these
poems, both ballad and epic, wooing and marriage follow the heroic
pattern of bride-stealing, as they do in the Serbian ‘junacke pesmek
Digenis Akritas steals his own bride from the powerful family of Doukas,
despite their contempt for his lineage. He uses Philopappos, the apelate,
as a go-between, and is also favoured by possession of a.magical ‘tambourak
In some ballads we find lannakis, also an apelate, is the candidate
favoured by the Doukas clan. This story corresponds to the ballad
Liogenneti (Politis, Eklogai apo ta tragoudia , Athens, 1932, 74) or King
Levandis* Girl (S. Kyriakides, 'O Digenis Akritas, Athens, n.d., p. 140;
p. 149 Ianni’s Marriage). The details of these stories are often widely
different, so that the ballad tends to resemble in parts the other set of
kidnappings. One complication has been the transference to Digenis of
an episode which may belong properly to Philopappos.

According to the epos Digenis’ mother was kidnapped by the emir
Mousour when her five brothers were away from home. They hear of the
affair and give chase, and a duel occurs in which the youngest brother,
Mikrokostantinos, overcomes Mousour. After some misunderstandings,
Mousour agrees to take the lady for his lawful wife, and then brings
over his whole tribe to the Christian side. The ballads give a different
account of these things. According to Andronikos’ Son, his mother was
‘enceinte 5 when captured, and it was uncertain whether Digenis was
the son of Andronikos or of the emir, who had also his Moslem wife.
Alternatively they represent the lady as a Camilla, who yields to the
Saracen after a fight, only on condition of his baptism. The kidnapping
becomes that of a hero’s bride; the hero being, apparently, not Digenis,
but lannakis (S. Kyriakides, ibid., p. 8; Politis 75 does not give the
name).

The ballad then runs: lannakis or Digenis is eating and drinking, when
he suddenly realizes that his wife has been kidnapped. (No reason is
given.) He goes to converse with his horses, and picks out the oldest and
most trusty. He races over the plains, and learns from an old man (some¬
times a herd) and from his old mother that the parties are still at table.
He meets his mother in his garden, and she fails to recognize him. His
wife recognizes him at once, but contrives to pass him off as her brother,
and on that pretext offers him a stoup of wine. The hero, lannakis or
Digenis, seizes her, and they ride away. Some of these episodes figure
in Asturian Spanish ballads of Count Dirlos.

These Greek ballads do not begin with a summons to war on the
morning after the wedding, nor do they mention a duration of time. They
may lack their proper beginning; but as things are, we seem bound to
consider these details to be a western feature. But the two conversations,
modelled on those of Odysseus with the swineherd and his father, and the
mention of a garden or vineyard, are often found in Balkan ballads, to an

NOTES 3 § s

t^hLttrr 1 ? SU ? ge f s Greek “fluence, and there are other Greek
Tff" ll 0 fa u S r Iar ^ Thus the Serbian Marko Kraljevic
durine- whirb'fh^f j 1 *- ^ surnrnor!S t0 ' var and the period of nine years
off bv S r '' lS Then the St0 ^ char ^s: she is carried

o tm -I> ^ ® Germans and Marko chases and recovers her. Jankamc

STJ I ™ P ™° nr T! t shows Stojan «* a prisoner, who is saddened bv the
Set free t * reR T? age - He has a PP°mted a nine-year period.

’ r 6 COnverses WIth ^ old mot^r in the vineyard, appears at the
wedding feast, is recognized by the bride who tells his sister, and the
affair is concluded by marrying the sister to the rejected groom. In
Pomorovac Todor we have the return of a prisoner; his wife is callous, and
all ends m blood. The Bulgarian Simon and his Sister (Miladinov 60
opens with the summons to war and conversation with the bride. Simon
has a magic bouquet which fades on her inconstancy or peril. He returns
converses with his father in the vineyard, and the ballad concludes in the
manner of Jankovic Stojan’s Imprisonment .

The Rhodian Greek poem appears to be of the western type. The West
European poems present the same characteristics with only minor changes
and Dobrynja and Aljoia resembles them. The vacillations in Serbia and
Bulgaria would appear, on the face of it, to offer evidence of contamination.
We haveto remember, however, that the story of Gerhardt von Holenbach,
retailed by Caesarius of Heisterbach about 1200, is connected with the
Crusades, which implied contact between the Latins and the Greeks. An
Akntic ballad enjoys, ‘a priori’, a reputation for antiquin’, despite the
appearance of an italianism like ‘kourtesia’ in the extant text. It is there¬
fore possible that the western legend may be a reworking of the Akritic
legend.

We have also to note that the Akritic legend fits the cycle of Gaiferos and

L Escriveta at least as well as does the Waltharius. Gaiferos opens with a
scene of revelry; there is a hasty ride over the Pyrenees, and an Instant
recognition by his bride; they ride away together. We must ignore the
Carolingian setting in either case. If derived from the Waltharius, we
have to admit that these ballads lack one leading characteristic of that epos,
namely, the fact that the hero and heroine are both hostages at the ccinrt
of a heathen king (Attila), when they agree to escape; we note also that the
ballads make little of the fight against odds, which is the great scene of the
epos. If we were to adopt the Greek alternative, these difficulties would
disappear; we should be faced by the apparent location in Aquitaine of the
Gaiferos story (Walter was £ o£ Spain’ or ‘of Aquitaine', and in French
Gautier might easily lead to Gaiffier), and with the fact that the Greek
source seems more remote than the German, which we know to have been
accessible to the Latins since the tenth century. On the other, hand,
Greek ballads travel as far west as Corsica, whence it is but a short step
to Provence, the home of VEscriveta.

Until the Akritic texts are comprehensively known it will not be
possible to determine these relationships, or whether they really exist.
The western cycles may be independent, of the Greek. Details in. the
Serbian and Bulgarian. £ pesme’ can hardly be deemed independent.

' 4615 • 3D

3 86 NOTES

Note D

The bibliography of this subject is enormous. It is contained in the
principal works cited in this note, and there is a special bibliography: De
Beaurepaire-Froment, Bibliograpkie des Chants Populaires Frangais, Paris,
n.d. The modern oral lyric is collected into E. Rolland’s Recueil de
Chansons Populaires, 6 vols., Paris, 1883-7 (with music). There are many
regional collections. The most important are probably: Comte de
Puymaigre, Chants populaires recueillis dans le pays messin, 1865, which I
know only by excerpts, and J. Bujeaud, Chants et Chansons populaires des
provinces de VOuest, 2 vols., Niort, 1866. D. Arbaud’s Chants populaires
de la Provence, 2 vols., Aix, 1862-4, is somewhat meagre, but of special
importance because of the difference of language. M. Haupt’s Franzosische
Volkslieder, Leipzig, 1877, is eclectic, with a tendency to prefer older
songs, and it is in some ways the best general introduction to the subject.
The older pieces, of the fifteenth and sixteenth centuries, are those which
chiefly concern us here. There are two principal published ‘chansonniers’,
viz. A. Gast6’s Chansons normandes du XV e siecle, publiees pour la premiere
fois sur les manuscrits de Bayeux et de Vire, Caen, 1866, and G. Paris et
A. Gevaert, Chansons du XV e siecle, publiees d’apres le manuscrit de la
Bibliotheque Nationale de Paris (SATF), Paris, new ed. 1935, with a
musical appendix. The material offered by these collections is mixed, so
that it is a great help to encounter a discriminating selection like that of
Th. Gerold, Chansons populaires des XV e et XVF siecles, Strasbourg, n.d.,
which has an admirable introduction. One must have at hand also A.
Jeanroy’s Origines de la Poesie lyrique en France, Paris, 1889, 1904, and
other editions.

Somewhat apart from these collections stands G. Doncieux’s Romancero
populaire de la France, Paris, 1904. The first word is Spanish, and the
notion of a corpus of narratives is somewhat contrary to the French genius,
but in the same degree it makes the collection highly valuable for purposes
of comparison with Spanish and German balladry. Doncieux has aimed
at producing a standard text for each of his ballads, constructed out of the
different versions available to him. Similarly, J. Tiersot has noted down
standard tunes, ignoring variants and the presence of different tunes for
given ballads. For comparative purposes we have to go back to Doncieux
and Tiersot’s sources, since significant details may not happen to have met
with their editorial approval.

Note E

A. Duran’s Romancero general (Biblioteca de Autores Espanoles),
2 vols., Madrid, aimed at completeness and contained 1887 pieces. For
many years the scholarly world has awaited the completion of a definitive
Romancero by Don Ramon Men^ndez Pidal, which should contain
about four times the material collected by Duran. The tragedy lately
enacted in Spain may have dispersed the notes covering a generation of
labour. For the present we can only follow Sr Men6ndez Pidal’s doctrine
in the articles and books he has written, which are all of the highest

NOTES 3 g 7

intema tion al interest. The older ballads are to be encountered in F. J.
_ 0 5 n< ^ Hofmann, Primavera y Flor de Romances, as edited by

M; Menendez y Pelayo in his Antologia de Poetas Uncos Castellanos ,
vm-x; followed by the same scholar’s Tratado de los ziejos Romances in
vols. xi-xii. References are to the ballads as numbered in the Primavera
or t0 the volume and page of the Antologia ,

Note F

Flecker s excellent translation will serve to show those who care to make
the comparison some of the difficulties of rendering Spanish ballads into
English. He cuts off a syllable or two from each line; justifiably, because
there are more syllables in Spanish to the same number of words. As
my object is to reproduce the form of the original, I have given the lines
their Spanish length, at the cost of a little padding. Flecker has not been
able to avoid some padding. Then there is the difficult matter of the
assonance. Several keywords suggest the assonance in ee which is used by
Flecker as well as Longfellow: ‘’sea, he, she, cramasv, tree’, See, But
Flecker was unable to carry 7 this assonance through the poem, and, on the
other hand, his rhymes are too perfect, being full rhymes. 1 have
endeavoured to use a more sonorous assonance in o, and, in order to mark
its quality 7 as assonance and not rhyme, I have gone out of my way to
break strings of rhymes when these offered themselves. The'Spanish
poet was under no such necessity. Assonance was, and is, a device so
familiar that no inconvenience is suffered if a series of full rhymes appears
among them. But in English such a series suggests rhyme, and the
occasional assonances produce a wrong effect of imperfection. Assonance
in two syllables, as in many ballads, is more than I dare attempt.

Note G

Danmarks gamle Folkeviser (DGF), by S. Grundtvig and A. OMk
(Copenhagen, 1853-1920), is not easy to obtain in England, and not often
complete. I have on my shelves Abrahamson, Nyerup, and Rahbeck's
Udvalgte Danske Viser ( UDV ) (Copenhagen, 1812-14), which is sufficient
for the sort of general comparisons instituted in this book. S. Grundtvig’s
Danmarks Folkeviser i Udvalg (DFU) (Copenhagen, 1882), and A.
Olrik’s Danske Folkeviser i Udvalg ( Olrik ) (Copenhagen, 1927), are
excellent anthologies, the latter having a valuable preface. R. C. A.
Prior’s Ancient Danish Ballads , London, i860, corresponds to the first
three volumes of DGF, E. T. Kristensen’s Jyske Folkeviser og Toner , 100
gamle jyske Folkeviser, and Gamle Viser i Folkemunde (Copenhagen, 1871,
1889, 1891), give the living oral tradition.

For Sweden there are two sources: A. L Arwidsson’s Svenska Form-
sanger (Stockholm, 1834-42), and E. G. Geijer and A. A. Afzelius’
Svenska Folk-visor , as revised by Bergstrom (Stockholm, 1880), with a
volume of tunes by L. Hoijer. For Norway I have consulted M. B.
Landstad’s Nor she Folkeviser (Christiania, 1853), without having access
to S. Bugge’s Gamle norske Folkeviser (Christiania, 1858). The Icelandic

388 NOTES

ballads are in S. Grundtvig’s fzlensk Fornkvcedi (Copenhagen, 1854-8),
to which was added a second volume in 1859-85. The Faeroese ballads
were gathered by Grundtvig into his manuscript Corpus carminum fceroen-
stum (CCF), from which various extracts have been made. I have used
V. U. Hammershaimb’s Fcewsk Anthologi (Copenhagen, 1886-91).

Note H

There is a handy one-volume edition of the English and Scottish Ballads ,
edited by H. C. Sargent and G. L. Kittredge, Cambridge, Mass., 1904.
A. Quiller-Couch’s Oxford Book of Ballads, Oxford, 1910, used some
:es not in Child, but in general goes to show that Child’s selection is
nitive. The texts are for readers: eclectic. The leading student of the
broadside ballad has been Hyder E. Rollins, who has many valuable
editions to his credit. One, of an anthological nature, is Old English
Ballads (i 553 ~ I ^ 2 5 )> Cambridge, 1920. Sir C. Firth’s essay on ‘Ballads
and Broadsides’ (Essays, Oxford, 1938; reprinted from Shakespeare's
England, 1916) shows how full Shakespeare’s mind was of these catches.
R. L. Greene’s Early English Carols, Oxford, 1935, is a study of an allied
literary form. A useful general account of ballad-poetry from the English
and American standpoints is G. H. Gerould’s Ballad of Tradition, Oxford,
1932. Professor Gerould deals summarily—but sufficiently from the
standpoint of this book, with the new ballads of the negroes, cowboys, and
lumberjacks of America. Child’s appeal for American material proved
fruitless. Olive Dame Campbell and C. J. Sharp’s English Folk Songs from
the Southern Appalachians, New York, 1917, contained eighteen quite new
ballads, and was of the utmost importance from the musical standpoint.
The collection of ballads in Virginia, the Carolinas, Maryland, and even
in Maine and other states has been most vigorous. A reader who dips into
Dorothy Scarborough’s A Song Catcher in Southern Mountains, New
York, 1937, will encounter not only ballads, but also an agreeable descrip¬
tion of the sort of people who still sing them and believe in them. A
copious bibliography accompanies Professor Gerould’s book.

Note I

L. Erk and F. Bohme, Deutscher Liederhort, Leipzig, 1893-5 1 J* Meier,
Deutsche Volkslieder mit ihren Melodien : Balladen (3 fascicules have
appeared), Berlin and Leipzig, 1935-7. Except for the study of the history
of literature these publications render unnecessary the study of Herder,
Arnim and Brentano, Uhland, &c. They also include the Dutch and
Flemish evidence, which can be consulted best in Hoffmann von Fallers-
leben, Niederlandische Volkslieder, Hannover, 1856 (Horce Belgicce, ii).
R. von Liliencron’s Historische Volkslieder der Deutschen, Leipzig,
1865-9, contains a great many historical poems for which there is little or
no evidence of traditional transmission; those relevant to our study have
been gathered into Erk-Bohme. F. Arnold’s Das deutsche Volkslied,
Prenzlau, 1927,4s a really useful anthology and introduction; so is also
J. Sahr’s Das deutsche Volkslied, ed. P. Sartori, Berlin and Leipzig, 1924, in

NOTES 3§9

the Goschen series. A general introduction to the ballad and its influence

deut^BaSXC^^ fr ° m W - ^ ^

iNOTE j

I have not been able to get satisfactory documents for narrative poetry
n Hungary, as our English libraries are not well stocked™

lyrical^ Out'of!! aSiB ® la>S ? Ung ? ri ™ F ^-Son gs , Budapest, is ‘en&elv
Mafn!’ 1 *v f 44 pieces gIVen by H - M6lIer in Das Lied der Volker
Mamz, only three are narratives which serve for comparison with those of
other countnes; the rest are mostly lyrics. There are 37 folk-songs ln
W. Tolnai Ungansches Lesehuch (Sammlung Goschen), Berlin-Leipzig
su^e^^ttf? 1110 ^ 7171 ' 1 ' 181 ' Authorities on international ballads have not
and°so T d h h 3117 theme has originated among the Magyars,

bv T H SchwV C r Se r' f ? r rdying ° n the description ^ven

DD J e6s' R„ h ^n k v Geschwhte der un sanschen Litteratur, Leipzig, 1899,
w -8 ?' 7 o!kspoesie und ihr Einfluss’. My purpose is not so
much to describe Magyar ballads as to assign them their place Tn^nter!

national traditional song; and that is undisputed. [I have later, by the

o '& n / S \ r 0 , T l r * ^f eorge ® uda y> been able to consult the admirable

fnd Mr 'l 35? WHch Dn G - ° rtuta y bas edited,

and Mr. Buday has illustrated with striking woodcuts.]

Note K

Bohdan Zaleski (Waclaw z Oleska), Piesni Polskte i Ruskie Ludu
Gahcyjskiego, Lwow, 1833, divides his collection into ‘women’s songs 1
(476 pp.), and men’s songs’ (38 pp.). The latter are the true narratives.
Ihe division, however, which was taken over from Karadzic, does not
seem relevant to Polish folk-song. To the fuller anthologies of later
editors I have, unfortunately, not been able to resort. R. Walter’s
Dudelsack , Schalmei und Geige : Polniscke Volkslieder , Hamburg (no date)
with an epilogue by Dr. Lucjan Kamieiiski, helps to correct the impres¬
sions one gets from Zaleski, but these may still be false in some important
There are 26 Polish songs included in H. Moller, Dm Lied
der Volker. Specifically Galician songs recur in V. Antonovic and M.
Dragomanov, Istoriceskija Pesni Malorusskago Xaroda, Kiev, 1874.

Note L

A lay song is called a ‘daina 5 , plural ‘dairies 5 in Lithuanian, but ‘dainas*
in Latvian. The word may be connected with the Rumanian ‘doinak It
means no more than ‘song 5 , but it must be distinguished from ‘giesme 5 ‘a
religious song 5 . No account is taken of religious songs in this essay. The
most useful collection of Lithuanian ‘dainos’ is G. H. P. Nesselmann
Litthauische Volkslieder , Berlin, 1853, but I have also consulted A.
Juskevic, Lietiwiskos Ddjnos , Kazan, 1880-2. Rhesa, Dainos (ed. Fr,
Kurschat), Berlin, 1843, X Leskien, Litauisches Lesebuck> Heidelberg,
1919 (17 ‘dainos 5 ), &c. U. Katzenelenbogen’s The Dama } Chicago, 1935,

39 °

NOTES

has an excellent introduction, and really sensitive translations into English
of 98 Lithuanian and 283 Latvian pieces. I have not had access to Kr.
Barons and H. Wissendorff’s great Latmju dainas, Mitau and St. Peters¬
burg, 1894-19x5, but think that the indications given by U. Katzenelen-
bogen and the 176 examples in J. Endzelin’s Lettisches Lesebuch, Heidel¬
berg, 1922, suffice to place the discussion of Latvian folksong on a sound
basis. On a point of particular interest—the survival of pagan deities and
customs in this region—Mr. Janis Karklins was good enough to send me
the valuable Chansons mythologiques lettones (Paris, 1929) by M. Jonval. It
contains an introduction dealing with this mythology, and 1219 poems

in Latvian and French.

Note M

An apology has been tendered in the preface for accepting second-hand
evidence of Finnish and Esthonian balladry. None the less, the references
in this chapter are to poems which have been studied in the original. An
interval of three months occurred between the preface and the final draft
of this chapter, and in consequence I have been able to do some, though
inadequate, original reading. Candour thus requires that the initial
apology still stand, with this modification.

The Finnish ballads have been collected in E. Lonnrot, Kanteletar ,
Helsinki, 1864 (2nd ed.), especially in his third section, ‘Virsi-Lauluja’,
and in the introduction. For the Kalevala I have used W. Kirby’s transla¬
tion reprinted in ‘Everyman’s Library’ and the Finnish edition by Roth-
sten and Forsman issued at Helsinki in 1887. C. J. Billson’s Popular
Poetry of the Finns, London, 1900, is a pleasant introduction, but lament¬
ably short.

The most useful collection of Esthonian songs is J. Hurt s Vana Kannel,
Dorpat, 1886. The first volume has a complete German translation, and
my observations have been restricted to these translated pieces. For the
Kalevipoeg there is W. Kirby’s summary in The Hero of Esthonia, London,
1895, vol. i; and the Esthonian text is available to me in an edition issued
at Kuopio in 1862.

Note N

Initially all vowels are supposed to alliterate with each other. The
alliterations may overlap from one line to another, though in the second a
new alliteration rules. Some notion of the subtleties of this style may be
obtained from noticing the alliterations and echoes in
Hierelevi, /zautelevi, Then she rubbed and then she softened,

muna muuttui neitoseksi, till the egg became a maiden,
mika raeille nimeksi— and the maiden’s name—what was it ?

Sorsatarko, >Suometarko ? Wild Duck’s daughter ? Finland’s daughter ?
Ei ole jSorsatar soria, Nay, she was not Wild Duck s daughter,

iSuometar nimi soria. but was Finland’s pretty daughter.

Kanteletar, ‘Virsi-Lauluja’, 1 , lines 10-15.

Longfellow’s Hiawatha metre reproduces only the more monotonous
aspects of Finnish prosody, though it is the nearest English equivalent.

NOTES 39 i

It is impossible in English, with its more varied sound system, to repro¬
duce the Finnish echoes in all their subtlety.

Note O

Adonz, in Byzantmische Zeitschrift , xxix, 1930, discusses the relation
between Digenis and the Armenian David of Sasoun; Speranskii dis¬
cussed the Russian romance of Devgemi in the Russian Academy’s
Memoirs (Section of Language and Literature) for 1922. H. Gregoire
posits an Arabic Geste de JS/Ielitene , relating to the exploits and death of the
emir and martyr 'Umar al-Nu'man in 863. From this supposed Geste
would descend, on the one hand, the Moslem episodes of the Greek poem
{Geste de VEmir), and on the other the Histoire du roi Omar-al-Nemdn et
ses deux fils merveilleux Scharkdn et Daoidi-J.lakdn (Mardrus’s translation
of the Arabian Nights). The episode should be assigned, according to
Gregoire, to the early Baghdad cycle, not to the late Egyptian cycle of
stories (to which it had been attributed by other scholars). Then there is
the ‘great Arabic romance of chivalry 5 , the Ddt al-Himma wa'l-Battdl , for
which we are referred to an article by Cagnard in the Journal Asiatique
for 1926, and lastly the Turkish heroic tale of Sayyid Battdl. What is
hypothetical in this account of the matter is the so-called Geste de
Melitene. Supposing there were no such work, and that the Moslem
episodes of the Greek poem were merely novelesque developments from
certain facts of history, then it seems far from impossible to regard the
legend in the Arabian Nights —uncertain of date, as it is—as a derivative
from a Greek source, in so far as the two coincide; and so on. for the other
romances. The references are those given by H. Gregoire.

Note P

The Serbian ballads were collected by V. S. Karadzic, Srpske narodne
Pjesme (my edition is that of Belgrade, 1895), and by an anonymous
German in the Erlangen manuscript edited by G. Gezeman, Erlangenski
Rukopis , Carlowitz, 1925, for which I have to thank Professor Yladeta
Popovic, of Belgrade. Milos ©uric’s Srpske narodne Pesme, Belgrade (no
date) is a convenient popular issue. There are other collections, par¬
ticularly of women’s songs, into which I have not thought it necessary to
inquire. That aspect of Serbian folk-song is sufficiently treated by
L. K. Goetz in Volkslied und Volksleben der Kroaten und Serben, Heidel¬
berg, 1936-7, 2 vols. A full-length treatment of the men’s songs is given
by H. M. and N. K. Chadwick in The Growth of Literature, Cambridge,
1936, ii. 2: ‘Yugoslav Oral Poetry’ (pp. 299-456), and in D. Subotic’s
Yugoslav popular Ballads , Cambridge, 1932. P. Popovic summarized
research down to 1932 in Misao , xl, and D. Kostic issued notes on .the
second volume of Karadzic under the title Tumacenja dnige knjige
Srpskih narodnik Pjesama Vuka St. Karadzica, Belgrade, 1937. A. Dozon- '
collected a number of pieces for his Poesies series .

The standard Bulgarian collection is by D. and K. Miladinov, BMgarski
narodni Pjesni , 2nd ed., Sofia, 1891. A. Dozen’s Chansons poptdaires

392 NOTES

bulgares inedites, Paris, 1875, has been of great help since it contains both
translations and an impeccable glossary. A. Dozon’s Contes albanais,
Paris, 1881, contains a few lines of Albanian verse, which show coinci¬
dences of form and theme with the Serbian.

Note Q

N. Iorga, Istoria Literaturii romane§ti, i, Bucharest, 1925, chap, i:
‘Balada populara romaneasca—originarea §i ciclurile ei.’ The standard of
reference is V. Alecsandri, Poezii popular e, recently re-edited by G. Gluglea
(Bucharest, 1938), with improved annotations. It is incomplete as a
picture of Rumanian balladry, and I have drawn also on G. D. Teodorescu,
Poesii populare romane, Bucharest, 1885; G. G. Tocilescu, Materialnri
folkloristice, i: Poesia poporana, Bucharest, 1900; O. Densusianu, Flori
alese din Cintecele poporului, Bucharest, 1920; I. A. Candrea and O.
Densusianu, Din Popor, Bucharest, 1908; and L. Salvini, Canti popolari
romeni, Lanciano, 1931; as well as the anthologies of Rumanian texts
published by M. Gaster, C. Tagliavini, and S. Pu^cariu and I. Breazu.
For several of these books I am indebted to Sir William Craigie.

Note R

P. N. Rybnikov, Pesni, 3 vols., Moscow, 1909-10 (revised edition);
A. GiTferding, Onezskija Byliny, St. Petersburg, 1873; M. Speranskii,
Byliny, 2 vols., Moscow, 1916; N. K. Chadwick, Russian heroic Poetry,
Cambridge, 1932. I have not seen Skaftymov, Poetika i Genezis Bylin,
Saratov, 1924, but have found Speranskii’s notes helpful and also the
careful summary of research in Keltujala, Kurs Istorii Russkoi Literatury,
i, St. Petersburg, 1913. A masterly statement in English is contained in
H. M. and N. K. Chadwick, The Growth of Literature, ii. 1: ‘Russian Oral
Literature’, Cambridge, 1936; and R. Trautmann, Die Volksdichtung der
Grossrussen, i: Das Heldenlied {Die Byline), Heidelberg, 1935, is authorita¬
tive. F. W. Neumann, Geschichte der russischen Ballade, Konigsberg,
1937, deals with the consequences of this folk-literature for the artistic
writers of the nineteenth century. It is only in the latter half of his period
that the ‘byliny’ begin to operate directly on the artistic conscience. In
the earlier years the standard of ballad-making is set by Percy and
Herder’s followers.

Aage and Else, 83, 86, 221, 222, 262.
Abenamar, 28, no, 162, 163, 253.
Abendgang, see Evening Walk.
Absalom, 189.

Accident de Don Inigo, 128.

Adam, 189.

Adrianople, 64, 311.

Affianced to a Turk, see Turk’s Bride
Afonso, 156, 165, 188, 191.

After Death, 127.

Agincourt, 233.

Agnes and the Merman, see Merman
Agnes Bernauer, 243, 249.

Aguilar, 6, 160, 162.

Aiglentine, 136.

Aike, 298.

Ajmer i, 102, 103, 168, 176.

Ai’ol, 6, 26, 176.

Akritas built a Castle, see Digenis.
Alarcos, 50, 145, i 54j I7 6.

Alatar, 163.

Albanina, see Blancanina.

Albayaldos, 163.

Albert (King), 216.

Alda (Aude), 102, 173.

Aldelaga and Olope, 106, 304.
Aldingar, 19, 66, 67, 74, 82, 11S, 182,
195^206, 229, 232, 233, 234.
Aleksei Popovic, 367, 377.

Alexander of Metz (Count in the
Plough), 60, 76, 104, 196, 204, 223,
244, 258, 280, 370.

Alf (Proud Sir), 211.

Alf i Odderskjaer, 101, 202, 217.

Alf of Tonsberg, 216.

Alhama, 50, 127, 164.

Alidakis, 313.

Alimon, 352.

Alimo? (Toma), 351.

Alison Gross, 237.

Aljosa Popovic, 56, 64,83,86,104,130,
354, 357, 362, 364, 367, 368, 377*
see Dobrynja.

Alora, 156.

Amazon, see Warrior-maiden.
Ambrogio and Lietta, 145, 182.

Amis and Armies, 104, 176.

Amourxs (Amouropoulos), 105, 106,
107, 304, 307, 308, 362, 369.
Ancient Mariner, 126.

Andrew (St.), 152.

Andriecko and MarieSka, 380.

Andro and Marica, 342.

Andronikos’ Son, 106, 306, note C.
Androutsos, 315.

Angela, 182, 268,

4615

Angelfyr and Helmer, 211, 217.
Angelica and Medoro, 12c.

Animal Life, 280, 28S.

Animals’ Wedding, 270.

Anna (Annie, Annet), 224, 238,
Anne Cozik, 14S.

Anrus and Death, 77, 277.

Ansels, 95, 167.

Antequera, 161, 165.

Anthony (St.), 152.

Antonine (St.), 152.

Anzolin (Angiolino), 83, 149, 221.
Apraksja, 64, 356, 360, 368, 369.
Arabian Nights, 106.

Aragon (Lady of), 52, 87, 181.
Arbola, 1S2.

Argentine, 136.

Arge$, see Manole.

Aria the Infanticide, 78.

Armada, 127.

Armstrong (Johnnie), 6, 231.

Amaldos, 29, 58, 130, 168, 179, i$o,
189, note F.

Amau, 192.

Arts (bridge, Adana), 7, 87, 309, 315,
316, 326, 330, 350.

Artevelde, 249.

Arthur and Cornwall, 104, 203, 235.

Asanaga’s Palace, 343.

Asanaginica, 340, 341.

Asmund, 218.

Astason, 227.

Athelston, 204.

Atreiden (Die spanische), 125.

Aubrays (Les), 147.

Ausenda, 174.

Ausenia, see Poisoner.

Avdotja of Rjazan, 373.

Avenger’s Sword, see Sword of Ven¬
geance.

Avenging Husband, see Husband’s
Vengeance.

Avenging Prince, no, 184.

Axel Tordsson and Valborg, 58, 206,
223, 224.

Aye d’Avignon, 97, 117.

Azov, 374, 377.

Babeli, see DurslL
Babylon, 240.

Badiu, 352.

Baeza, 160.

Baida, 379.

Balaurul, 351.

Ballade de Florentin Prunier, 136.
Banovic Strahinja, 88, 341, 363. .

INDEX

394

Barbara Allan, 241.

Barbarfn, 163.

Barbiere, 105, 134, 13B, 140.

Barton (Andrew), 231.

Basselin, 140.

Bataille perdue, 128.

Bathsheba, 361, 368.

Beatrice, 135.

Beauty’s Castle, 308.

Beggar from Hungary, 280.

Beichan and Susie Pye, 182, 240, 330.
Belgrade (battle), see Eugene.

Bell (Adam), 229.

Belle se sied, 141, 183.

Bengerd, 6, 65, 213.

Benjamin’s Cup, 59, 108, 370.

Beowulf, 92, 97, 200, 201, 235.

Bernal Frances, 178, 179.

Bernardo del Carpio, 166, 168.

Biche Blanche, see White Beast.

Billee, 140, 184, 205, 218.

Binnorie, 19, 66, 109.

Birds’ Wedding, 135, 270, 288.

Birs, 68, 246.

Bishop of Penanstanc, 145.

Biter bit, 274.

Biterolf, see Detlieb.

Bjarkamal, 65, 198, 211.

Blak og Ravn hin Brune, 222.
Blancanina, 178, 180, 239.

Blanche de Bourbon, 157, 159, 160.
Blancheflour and Jellyflorice, 204, 236;

see Flos and Blancflos.

Bland, 241.

Blaubliimlein, 267.

Blighty, 242.

Bloody Son, 295.

Bliicher, 254, 289.

Bluebeard, 60.

Bohmenschlacht, 197.

Bojcic Alija and Glumac Asanaga, 335.
Bojovnice, see Warrior-maiden.

Bona Viuda, 143.

Bonello, 151.

Borba so Smertju, 129.

Boris Godunov, 65, 373, 374.

Bosmer, 221.

Botany Bay, 241.

Botzaris, 128, 312, 313.

Bo valias, 184.

Boy and Mantle, 104, 203, 235,
Brand, 76, xoi, 238.

Brankovici, 326, 332.

Breitenfels, 253.

Bremberger (Brennenberg), 106, 204,

257.

Bride (False), 266.

Bride avenged by Brothers, 85,99,143.
Bridesmaid into Bride, 87, 182, 316.
Broomfield Hill, 237.

Brother (Cruel, Grim), 267, 280.
Brother and Sister, see Jovo and his
Sister.

Brother and Sister, 148.

Brother betrays Sister, 380.

Brother purchases Sister, 380.

Brothers (five), 289, 290.

Brothers (Nine) and Sister, 225, 372.
Brown Robin, 337.

Bruder Veit, 68.

Brunanburh, 198.

Bruncvik, 85, 258.

Brunswick (Lord of), 139.

Bryan of Brittany, 310.

Brynhild, 98.

Bticar, 94, 184, 191.

Bueso, 51, 85, 177, 202, 257.

Bunyan (Paul), 242.

Burmand, 115, 205, 218.

Burning of Rome, 51.

Busso von Exleben, 68, 249.

Butcher’s Boy, 241.

Butler (Sir John), 233.

Byrge (Birger) and his Brothers, 66,
213.

Cabin Boy, 79, 183.

Cadi’s Daughter, 352.

Calainos, 175, 176.

Captain (Pitiless), 254.

Captain and the Beauty, 136, 149.
Captain’s Daughter, 271.

Captive Girl, 185, 189, 190.

Captured Lady, 277.

Carini (Princess, Baroness, of), 69, 151.
Carl and Rigmor, 217.

Castellan de Coucy, 204, 257.

, Castle in Austria, 54, 71, 75, no, 138,
196, 269, 270, 284.

Catherine (St.), 59, 152, 218, 261,
296.

Catherine II, 375.

Catinella, 152.

Catoise, 147.

Cawline, 204, 235.

Ceivan’s Daughter, see Zlatija.
Chaplain (Slanderous), 260.

Charles XII, 297.

Charon, 59, 60, 77, 108, 129, 318.
Chasse du Bourgrave, 128.

Chaste Servant, 59, 260.

Chevy Chase, 68, 113, 229, 230, 233.
Child is born to the Princess, 183.
Chiliopappos, see Philopappos.
Chisholm Trail, 242.

Chords, see Harp.

Chotim, 282.

Christ, 108, 336.

Christine, 128.

Churls in Flanders, 68, 24s

INDEX

Cid (Poem, Mocedades), 7, 13, 48, 57,
94 ) 96, 106, 120, 121, 122, 124, 128,
166, 168, 169, 171, 172, 186, 189.
Claire Fontaine, 140.

Claros of Montalban, 104, 141, 183,
372 :

Claudine, 80, 184.

Claus Molinaer, 272.

Clement the Mason’s Wife, 277.
Clementine, 3.

Clerk in a Basket, 259.

Clerk’s twa Sons of Oxenford, 69, 139,
238.

Clotildo, 85, 99, 143.

Cock (Grey), 236.

Cola Pesce, 82, 141, 288, 318, see
Diver.

Colvill (Clerk), 45, 83, 142, 143, 221,
236.

Comendadores de Cordoba, 156, 165.
Companions (Two), 265.

Comtesse de Ponthieu, 97, 177.
Constantine (Marriage and Departure),
83.

Constantine and Arete (The Dead
Voyager), 86, 87, 237, 309, 310, 316,
338 .

Constantine Brancoveanu, 351.
Constantine Dragazes, 312.
Constantine the Great, 337.
Constantine the Little, 86.
Constantinople, 64, 312.

Corbea, 351.

Corbies, see Ravens.

Com Ship, 140.

Cossack Nuczaj, 282.

Cossack’s Farewell to his Horse, 283.
Count in the Plough, Count of Rome,
see Alexander of Metz.

Cmojevici, 332.

Cronica Sarracina, 95, 167.

Crowned Snake, 262.

Cuckoo and Turtledove, 351.
Cunegunde, 67.

Curilo PlenkoviS, 254, 361, 368, 369,
376 .

Custozza, 254.

Daemon Lover, see Harris.

Dagmar, 65, 213.

Dance in Ribe, 31, no.

Dancer (Drowned), see Helene.
Dancer (Pardoned), 262.

Daskaloianni, 10, 313.

Daughter at Mother’s Grave, see
Orphans.

Daughter of King Under-Waves, 235,
Daughters avenge Father, 225.

David/ 368.

Dead Boy, 276.

461s

395

Dead Brother’s Return, see Constan¬
tine and Arete.

Dead Man, 273.

Degener and Lussewine, 269.
Delgadina, 183.

Demakes, 312.

Des Chapelles, 147.

Detlieb (Detlev), 101, 364, 367.
Deutschland iiber Alles, 254.

Devil’s Bride, 152.

Devgenii, 362.

Devuska-voin, see Warrior-maiden.
Diamond (Lady), 204.

Diaz (Pedro), 160.

Diaz de Bivar (Ruv), see Cid.

Dibra (Bridge of), 87, 316.

Dietrich (Diderik,Theodoric,Thid'rek),
I 4) 57) 97, 99, too, 103, no, 115,
196, 197, 200, 201, 203, 205, 214,
217, 218, 255, 364..

Digenis Akritas, 32, 58, 59, 63, 77, 105,
106, 129, 214, 303, 304, 305, 306,
307, 308, 330, 362, notes C, 6.
Dinah, 189.

Diocletian (Duklijan), 337.

Didos, 25,86,104,154,176,257, noteC.
Disguised Margrave’s Son, 101.
Dittmarsche.fi, 216.

Diver, 82, 138, 141, 149, 315, see Cola
Pesce.

Dives, 59, 236, 295, 296, 336.

Djuk Stepanovic, 107, 354, 362, 366,

368, 369.

Djurja Danilic, 335.

Dobrynja, 11, 12, 13, 56, 64, 86, 104,
258, 354 , 357 , 360^364, 367, 368,

369, note C.

Doette, 135, 139, 182, 267.

Doidin, 344, 352.

Don and Nepra, 369.

Donciia, 344, 352.

Donna Lombarda, 52, 80, 82, 141,
145, 149, 156, 182, 200, 241, 269,
274, see Poisoner.

Doon de la Roche, 176.

Douglas, 29, 32.

Douglas Tragedy, 3, 53, 77, 101, 109,
115, 238, 241.

Dragon-fight, ioi, 217.

Dragon’s Love, 338.

Draumekvaecti, 226, 227.

Dream (Untold), 372.

Drewicz, 282.

Drummer, 140.

Drums (Three), 51.

Duchess of Bragan^a, 165.

Duel with Turkish King, 378, 379..
Duke of Guimaraes, 165.

Dummy, 339.

Dunal, 368, 369.

3E2

396 INDEX

Durandarte, 102, 122, 173.

Durham Field, 229, 232.

Dursli and Babeli, 265, 288.

Du§an, see Stepan Dusan.

Earl o’ Mar’s Daughter, 237.

Earl o’ Murray, 23 r.

Early rising in the Morning, 286.
Ebbe Skammelson, 224.

Edda, Eddica Minora, 10, 20, 26, 27,
63, 76, 77 , 92 , 93 , 94 ,98, 99 , 101,1 15,
201,202,207,210,211,216,217,222.
Edmund and Benedikt, 223, 225.
Edom o’ Gordon, 231.

Edward, 34, 77, 129, 239, 295.
Eleanor’s Confession, 67, 232.

Elena, 83.

Elfes, 124.

Elf-hillock, 221.

Elf-shot (Elveskud), 54, 82, 83, 124,
125, 126, 128, 142, 143, 148, 149,
178, 220, 221, 273, 284.

Elias (St.) 361.

Elina’s Death, 297.

Elselein, 259.

Elves’ Dance, 83, 221.

Embarcation of the Singing-girl, 140.
Emirissa, 318, 319.

Emund, 102.

Enchanted Daughter, 273.

Engel, 225.

English Marriage, 139, 144.

En passant par la Lorraine, 45.

Epple von Gailingen, 68, 248.

Erceg Stepan, 81, 141, 342.

Erembors, 136.

Erik Emun’s Murder, 65, 195, 214.
Erko the Latin, 334, 335.

Erlkonig, 83, 124, 129, 220, 221, 239.
Ermak, 30, 65, 360.

Ermanaric’s Death, 98, 203, 255.
Ernst, Ernest (Duke, Herzog), 104,
204, 258.

Esbern Snare, 65, 213.

Escrivana (Escriveta), 51, 78, 97, 141,
145, 148, 177, note C.

Esel, 253.

Espinelo, 184.

Estudiants de Tortosa, see Hanged
Scholars.

Estmere, 204, 235.

Etin, 77, 236.

Etmanml, 362.

Eudoxia’s Death, 308.

Eugene, 251, 254, 335.

Eugenoula, 49, 308, 309.

Evening Walk, 204, 258, 259.

Evil Mother, 81.

Evil Wife, 319.

Eydun, 213.

Fadrique, 29, 156, 158, 159, 188, 189.
Fajardo, 6, 30, 161, 163.

Falcon on the Poplar, 283.

Faliena’s Boasts, 337.

Falkenstein (Lord of), 270.

Famine, 148.

Fanchon, 138.

Father and Son, 277.

Father betrays Daughter, 380.

Fedor, 374.

Fehrbellin, 253.

Fernan Gonzalez, 57, 95, 166, 168.
Fiancde, 83.

Fiancee as Poisoner, see Poisoner.
Fiddling Hunchback, 263.

Fierabras, 176.

Fight with Dragon, see Dragon-fight.
Fille du roi Loys (Louis) ,104.
Finnekonster, 86.

Finnsburg, 68, 97, 200, 235.
Fior-d’auliva, 149.

Fir, 270, 277.

Fire-king, 239.

First Love, 86, 274.

Flambeau d’Amour, see Torch of Love.
Flerida, 191.

Florencios, 104, 183.

Floris, 145.

Floris V, 272.

Flos and Blancflos (Floire et Blanca-
flor), 104, 204, 257.

Fldvant, 102.

Flower of Northumberland, 240.
Flower of Serving-men, 240.

Fonte frida, 180, 181.

Foot-page, 223.

Foundling, 226.

Frsendehevn, 2x6.

Francois I, 56, 74 , I39> * 44 > *93-
Frangoise et Pierre, 148.

Frankie and Albert, 241.

Frederick’s Wife, 267, 273.

Fritsche Grad, 248.

Frundsberg, 139, 250.

Fulga, 350.

Gabrielle d’Estrdes, 53, 56, 139, 144.
Gaiferos, 79, 97, 104, 122, 175, 176,
177, 185, note C.

Galanzuca, 183.

Gandia, 188.

Garcia, 185.

Gavan’s Lady, 336.

Gawain’s Marriage, 104, 203, 235.
Gay Goshawk, 141, 183, 238.

Geipa tattur, 102.

Genevieve (St.), 152.

Genoveva, 271.

Gentle Lady, 47, 70, 83,156, 178, 180,
192, 194.

INDEX

George (St.), 59, 218, 261, 281, 306.
Georgie, 34.

Gerineldo, 50, 89, 103, 176.

German Rhine, 254.

Germand Gladensvend, 219.

Gian Lorenzo, 191.

Girl and Cuckoo, 265.

Girl and Dragon, 138, 140.

Girl and Hazel, 270, 280.'

Girl and Saracen, 306.

Girl at her Lover’s Grave, 273.

Girl ransomed by her Lover, 380.

Girl who went to War, see Warrior-
maiden.

GlSb, 357 , 369.

Golea, 351.

Goliath, 189.

Gollheim, 245.

Golota’s Duel, 378.

Gonzalo of Jaen, 6, 7, 30, 161, 162.
Gorel’s Daughter, 217.

Gormond et Isembart, 198.
Gouyquet, 147.

Gramvold, 216.

Granson, 68.

Grass-girl, 280.

Grave (Unquiet), 238.

Gregory (Lord), 34.

Grimhild’s Vengeance, 26, 66, 98, 196,
202, 212, 217.

Grozdanka, 339.

Gruica and the Pasha from Zagorje,
335 .

Grujo Novakovic, 341.

Guarinos, 77, 103, 129, 173, 175*
Guingamp, 147.

Haakon (Haagen Haagenson), 66, 213.
Haevnersvaerdet, see Sword of Ven¬
geance.

Hagbard and Signe, 65, 101, 203, 207,
211, 217, 255.

Hallewijn (Holofemes), 54, 55, 59, 84,
108, 142, 182, 219, 236, 239, 259,
260, 270, 273, 276, 280, 295.
Hamilton (Mary), 231.

Handrias and Rajsenberk, 248, 279.
Handschuh, 125, 185.

Hanged Students, 51, 69, 139, 145,
194, 238.

Hardy (John), 242.

Hare’s Complaint, 270.

Hares (Two), 270.

Harlaw, 68, 231.

Harp (Speaking), 223.

Harp’s Power, 54, 222.

Harris (James), 238, 240, 266.

Hasten, hasten, Mother, 277.
Hasting, 147.

Hateful Deed, 269.

397

Havsfrun, see Mermaid.

Heathen Princess, see Sultan’s Daugh¬
ter.

Heavenly Lime, 261.

Heidenrosiein, 264.

Heldenbuch, 255. 363.

Helen, 59, 108?

Helene, 83, 142, 220.

Helmer Blaa, 224.

Henori (St.), 147.

Henrietta Maria, 139.

Henry (Bishop), 297.

Henry (John), 242.

Henry (King), 81, 203.

Henry the Lion (Brunswick), 58, 60,
85, 99 , 257 , 258.

Hercegovinian Stepan, see Erceg
Stepan.

Hermund illi, 218.

Hildebrand, 68, 97, 107, 197, 201, 203,
tt .?5°, 2 54> 255, 277, 309, 363, 364.

Hilde s Sorrow, 224.

Hilzicka, 280.

Hindrask, 80.

Hogna tattur, 98.

Hoheniinden, 127.

Holenbach (Gerhardt von), note C.
Holger (Olger), see Ogier,

Homecomer, 274.

Horn, 86, 104, 203, 235, 257.

Hoten, 369.

Hromund, 218.

Hugdietrich, 203, 255, 363.

Hugh le Blond, see Aldingar.

Hugh of Lincoln, 67, 232.

Hungry Child, 277.

Hunter from Greece, 98, 255.

Hunting (Young), 240.

Hunting of the Cheviot, see Chevy
Chase.

Husband’s Return, 86, 140, 152, 258,
283, 317.

Husband’s Vengeance, 317, 330.

Hutten, 251.

lancea (lencea) Sabiencea, 81, 352.

lannakis, see loannakis.

Ianni’s Marriage, note C.

Idolisce, 363, 367.

Igor, 65, 92, 108, 355, 356, 359, 377.

I heard a Sickle, 264.

Ileana Cosanzeana, 345, 351.

IMja Smiljanic, 334.

Ilincuta Sandului, 351.

Il’ja, 13, 58, 64, 65, 97, 98, 100, 101,
107, 354 > 35 % 359 * 3&>, 3 $ 3 * 3 %
367,368.

Iliugi, 218.

Umarinen, 298.

Incorruptible, 265.

308 INDEX

Inez de Castro, 165.

Infantes de Lara, 33, 63, 95, 96, 166,
168, 169, 170, 186, 187.

Infanticide, 269, 274, 283, 315.
Infantina, 178.

Ingeborg’s Disguise, 223.

Ingefred Torluf’s Daughter, 223.
Ingrese (Bella), 149.

Inkeri’s Suitors, 296.

Innocence proved, 81.

Inquietudes de Don Simuel, 128.
Interrupted Wedding, 89.

Ioannakis, 105, 303, 307, note C.
Isabel and Earl Patrick, 237.

Isabel and the Elf Knight, 85, 236,

260.

Isabel de Liar, 165, 191.

Isabiaus, see Ysabiaus.

Iskren and Milica, 88, 341, 363.

Ivan Danilovic, 369.

Ivan the Merchant’s Son, 369.

Ivan the Terrible, 65, 364, 365, 372,
373 , 374 -

Ivo Karlovic, 344.

Ivo of Senj, 334, 335.

Jacob, 281.

Jagers (Tenth), 254.

Jaksici, 332, 333.

James and Brown, 232.

Jancu Voda, 351.

Janko of Kotor, 334.

Jankovic Stojan’s imprisonment, see
Stojan.

Jasia, 284.

Jerina of Smederevo, 333.

John and the Bishop, 203, 204, 235.
John Hunyadi, 333, 343, 350.

John (King), 83.

John (St.), 338, 339.

Jomfru i Hindeham, 222.

Jonah, 368, 377.

Jonas, 225.

Jose, 362.

Joseph, 59, 67, 108, 281, 361, 362, 370.
Joseph (St.), 152.

Jovo and his Sister, 86, 338.

Juan of Biscay, 158.

Judas, 62, 67, 197, 230, 233, 236, 260,

261.

Judith, 59, 84, 108.

Jugovici, 7, 109, 326, 328.

Julianesa, 79, 185.

Jumala, 297.

Juris of Senj, 335.

Jurja of Smederevo, 332.

Kadar Kata, 276.

Kaferhochzeit, 270.

Kaica, 333.

Kalevala, 14, 39,40, 91, 297, 298, 299,
300, 366.

Kalevipoeg, 14, 297, 298, 299, 300,
366, 367.

Kalin (Kalka), 65, 360, 367.
Kaniowski, 284.

Karl, 216.

Karlamagnussaga, 102, 218.

Kawi-ali, 363.

Kemp Owyne, 235, 237.

Kerekes Andras, 276.

Kerels, see Churls.

Kerenstein (Lady of), 265, 266.

Keys of Heaven, 241.

Kidnapping, note C.

Kilian (St.), 261.

Kimiskes, 304.

King and Queen of Buda, 342.

King of Buda, 333.

King of Milan, 267.

King’s Children (Two), 54, 84, 258,
259, 270, 276, 280, 284.

King’s Daughter, 268.

Kinmonth Willie, 30.

Kinnamos, 105, 106, 303, 306.

Koico, 339.

Kitchie Boy, 235.

Kitel, see Shroud.

Kitty of Niedergiirig, 280.

Knight (Baffled), 238.

Knight and Shepherd, 270.

Knight and Shepherdess, 148.

Knight and Squire, 269.

Knight from Steiermark, 204.

Knight of Stauffenberg, 83, 221.
Knights (Twa), 81, 238, 315.

Knight’s Ghost, 237.

Knut i Borg, 223.

Koat-ar-Ster, 147.

Kolokotronis, 30, 312.

Kolyvan, 367.

Komm ’runter, mein Gretchen, 279.
Konig in Thule, 124, 129.

Kordyle, 312.

Kosovo, 324, 325, 326, 327, 328, 329.
Kozubaj, 283.

Krali Marko, see Marko.

Ksehja, 374.

Kudrun, 85, 98, 99, 116, 143, 148, 177,
202, 256, 257, 274, 280.

Kunz von Kaufungen, 250.

Kutusov, 375.

Kyr Borias, 318, 319.

Lady in the Wood, 224.

Laily Worm, 237.

Laime, 62, 290, 291.

Lammour, 147.

Lancelot (Lanzarote), 104, 174, 175.
Landres, 102.

399

INDEX

Lannion, 147.

Lass o’ Roch Royal, 240.

Lave and Jon, 54, 224.

Lawrence (St.), 59, 261.

Lazar, 7, 32, 109, 129, 234, 325, 32^
327 , 328, 329.

Lazar and Petkana, 86, 338, 344.
Lazarus, 59.

Leander, 54, 59, 84,108, 142, 258, 25c
273, 280, 283, 284.

Lear, 274.

Leesom Brand, 240.

Leipzig, 253.

Leiva, 139.

Lemminkainen, 298, 299, 300.

Lena, 65, 66, 195, 215.

Lenore, 86, 123, 124, 126, 129, 221
222, 237, 239, 262, 277.

Lepanto, 127, 312, 319.

Levandis, note C.

Liazes, 312, 313.

Liebesprobe, no, 182, 266, 267, 280
283.

Liebestod, 194, 204.

Light Brigade, 127.

Lijon, 298,

Lilia, 276.

Linda, 299.

Lindenschmid, 57, 68, 116, 243, 248.
Liogennete, note C.

Lionel, 235.

Lippold von Homboken, 68, 195, 196,
T- 2 i 5 ‘>

JLjubov mertveca, 129.

Ljubovic, 334.

Ljudmila, 129.

Locmaria, 147.

Logie (Laird o’), 231.

Long Serpent, 4, 28, 102, 203, 218.
Lom (Lord of), 240.

Louis XI, 140; —XVI, 139, 144.
Lover as Confessor, 152.

Lovers (Two), 81.

Lovers’ Flight, 152.

Lover’s Return, 317.

Lover’s Wreath, 340.

Lovmand and Thor, 224.

Loys’ (Louis) Daughter, 138, 141,148,
183, 238, 276.

Lucy (St.), 152.

Ludwigslied, 68, 197, 198.

Llitzen, 253.

Lyylikki, 299.

Macintosh (Willie), 231.

Magda, 330.

Magdalene, 59, 78, 218, 273, 281, 283,
296, 336 .

Magdeburg, 252, 253.

Magicians (Twa), 238.

I Magnus, 221.

I Maid and Palmer, 78, 230.

Maid in Dragon-form, 235.

Maid in the Forest, 277.

», ; Malamka, 339.

I Malamos, 312.

’ Maldon, 198.

I Malfred, 221, 223, 224.

1, | Mamaf, 360, 369.

| Man in the Hay, 268.
j Manetas, 317. '

j Manote (Mano), 71, 87, 277, 309, 316,
\ . 344 , 349 , 350, 352.
j Manuel de Leon, 185.

Mansfeldt, 253.

| Marcisia, 239, 283.

J Margaret (Maid of Norway), 66, 67,
j 213, 216, 232.

J Margaret (Proud Lady), 238.

| Margaret’s Ghost, 129.

| Margravine, 267.

S Marguerite, 142.
j Marguerite Chasles, 147.
j Marguerite Laurent, 149
| Maria de Padilla, 157, 158.
Marianson’s Rings, 81, 107, 141, 143,
145, 189, 223,238,315.33°» 352,364.
Mane Tili, 147.

Marignano, 244, 249.

Marina (Marinka), 357, 368, 374.
Mariner, 182.

Marinero, see Shipman.

Marivonnic, 148.

Marko Kraljevic, 13, 14, 30,49,56, 57,
6 4» 74* 77, 81, 82, 116,119, 215,235,
258, 307, 329 , 33 °> 334 , 332 , 333 s
340, 342, 343, 344, 352, note C.
Marlbrough (Malbrook), 53, 136, 138,
139.

Marquillos, 182.

Marquis of Mantua, see Ogier.
Marriage, (Hapless, Luckless), 140,

143, 221, 273, 317.

Marsin, 102, 173.

Marsk Stig, 25, 56, 206, 215.
Martinhos (Marquitos), 80.

Martino, 152,

Marusja Bogoslavka, 377, 378.
Maruska and Janosek, 274.

Mary (St.), 59, 152, 261, 296.

Mary the White Swan, 374, 377.
Master of Santiago’s Murder, see
Fadrique.

Mathurin (St.), 147.

Matthias Corvinus, 333, 343.
Maumariee, 105, 116, 133, 136, .149,
178, 342.

Maumariee vengde, see Bride avenged.
Matirianos, Si, 238, 315, 364.

Maurice (Child), 109.

INDEX

400

May Collin, 85.

May Song, 50, 88, 105, 134, 179, 191,
320.

Mazeppa, 128, 317.

Meererin, 99, 256, 257.

Melisenda, 176.

Mermaid, 85, 98, 283.

Merman, 83, 220, 262, 277, 280, see
Nix.

Metso'isos, 312, 313.

Michael (St.), 36.

Michael of Batak (St.), 363.

Mihata, 334.

Mihail Danilovic, 369.

Mihail Kazimirov, 369.

Mihailik, 379.

Mihailo Potyk, 25, 88, 342, 363, 369.
Mihu, 350.

Mikula, 331, 362, 363, 365, 366, 367,
370.

Milica, 327, 328.

Milionis, 312.

Milja, 334, 335.

Mill Song, 320.

Miller of Arcos, 187.

Miller's Daughter, 267.

Milos Obilic, 7, 326, 328, 330.
Mimering (Mimecan, Memering), see
Aldingar.

Minstrel’s Son, 263.

Miorita, 346, 348, 349.

Missolonghi, 313.

Mitrii VasiPevic and Domna Alexan-
drovna, 372, 373.

Mizil Crai, 80, 352.

Modus Ottinc, 198.

Molnar Anna, 84, 276.

Monferrina, 85.

Montejicar, 162.

Montesinos, 6, 7, 26, 102, 103, 122,
168, 175, 178.

Moon’s Faithlessness, see Perkuns.
Moon’s Wedding, 338.

Moraima, 185.

Moriana, 79, 156, 165, 182, 185.
Moringer, 60, 70, 85, 86, 104, 107, 126,
r 34> 177, 204, 223, 235, 239, 244,
257, 258, 266, 274, 307, 331, 352,
364, 368.

Morten of Fogelsang, 222.

Moses, 189.

Mo$neagul, see Old Man.

Mother beneath theMould, see Orphan.
Mother resurrected, 84.

Mother’s Malison, 237, 267.
Mother-in-law prisoner to Son-in-law,
283.

Mpoukovalas, 312.

Mudarra, 33, 63, 95, 96,121, 128, 170,
171, 180.

Munroe (Young), 242.

Murderer, 273, 295.

Murderess, 283, 295.

Murray (Outlaw), 231.

Murten, 68, 246.

Musgrave and Lady Barnard, 240, 241.
Music Stepan, 328.

Mutschelbeck, 269.

Muza, 163, 164.

Nafels, 68, 246.

Najera, 157.

Nalezena Sestra, see Sister redis¬
covered.

Nancy, 68, 246.

Nann, 83, 142, 143, 148, 221.

Narva, 297.

Nastasja, 367, 368.

Nastasja Politovskaja, 104, 183, 372.
Nastyna, 284.

Nest’atna Svat’ba, see Marriage.
Nibelungs (Nibelungenlied), 20, 66,
90, 98, 99, 196, 202, 211, 212, 256.
Nice, 193.

Nicholas, (St.), 59, 281, 361, 366.
Niels Ebbeson, 6, 13, 30, 205, 215.
Nilus and Hilde, 224, 225.

Nobleman in the Sack, 280.

Novak, 57, 334.

Ntritzas, 312.

Nucenzie, 149.

Nun, 268.

Nuho Vero, 181.

Nutmegs and Pinks, 288.
Nutmeg-tree, 257.

Oancea, 352.

Occasion manqu^e, 178.

Odvald, 102.

Ogier (Olger, Marquis of Mantua), 14,
26, 27, 69, 100, 102, 103, 115, 122,
168, 178, 205, 215, 218, 330.

Olaf (St.), 4, 28, 29, 102, 203, 218.
Olaf Lily-rose, see Elf-shot.

Olaf Strangeson, 225.

Old Man, 86, 352.

Ole Morske, 29.

Oleaz, 352.

Oleg, see Vol’ga.

Ol’ga, 129.

Olinos, 141, 174, 179, 180, 181.
Oliveros, 175.

Oluvu kvEedi, 102.

Olympus and Kissabos, 314.

Omer and Merjem, 342.

Orange-girl, 140.

Orange-seller, 140.

Orfeo, 236.

Orm Stdrolf’s Sons, 218.

Ormurin langi, see Long Serpent.

40i

INDEX

Orphans, 83, 84, 108, 221, 262, 273,
280, 287, 288, 291.

Ortnit, 100 107, 2 SS , 358, 363.
Utterburn, 68, 229, 231, 232, 233.
Owl and Eagle, 270, 280.

Owl at Sparrow’s Wedding, 287.
Oxenstjema, 253.

Palatine, or the Bloody Marriage, 266.

Palbeli Antal, 83, 277.

Palle, 225,

Palmer, 184.

Pastor que estas, see Gentle Lady.
Pauna§, 351, 352 .

Pavia, 68, 244, 250.

Pavle Zecanin, 340.

Pedro, 83, 182, 221.

Pedro the Cruel, 70, 156, 157, 158,
I 59 > 160.

Peranzules, 184.

Percy, 3 o, 32.

Perkuns (Perkunas, Perun), 62, 290,
291, 298.

Pernette, 104, 135, 138, 141, 183.
Perronnelle, 137, 138, 140, 288.

Peter (St.), 59.

Peter and his Concubine, 224.

Peter of Varadin, 333.

Peter the Great, 374, 375.
Philopappos, 105, 107, 303, 305, 306,
.307-

Pied Piper, 262.

Pierre de Grenoble, 140.

Pikker (Pikne), 297, 298.

Pilgrim to Compostela, 59, 60.
Pilgrims (Forty), 361, 370.

Pine and Palm, 126.

Poisoned Lord, 279.

Poisoned Man’s Will, 81, 145.
Poisoned Marchioness, see Gabrielle.
Poisoner, 80, 81, 269, 274, 276, 277,
283.

Poliokastro, 311.

Poltava, 297, 375.

Pommier Doux, 144.

Pomorovac Todor, see Todor.

Popovic Stojan, 341.

Porcheronne, see Swine-girl.
Porphyrios, 303, 306, 307.
Portocarrero, 189.

Potiphar’s Wife (Zuleikha) 59,67,108.
Praise God, ye pious Christians, 251.
Pretended Death as Matchmaker, 263.
Prevented Remarriage, see Remarriage.
Prijezda, 332, 333-
Prince a-wooing, 101.

Princes of Tver, 65, 373.

Princess, 145.

Princess and King of France’s Son,
183.

Princess rediscovered, 99.

Prior of St. John, 160.

Prisoner, 180.

Prisoners of Nantes, 140.

Prisoner’s Return, 86, 133.

Proteus, 238.

Puntus (Jakko), 297.

Pyramus, 59, 108, 134, 204, 259.

Queen of Hungary, 251.

Rada, 352.

Radoica, 81, 325, 343.

Raimund, 145.

Ramon, 83, 221.

Ramon Berenguer and. the Empress,

67, 182; see Aldingar.

Ramos (Alfonso), 184.

Randal, 53, 81, 109, 145, 239.
Ransom, 265.

Ravanica, 326, 343.

Ravengaard and Memering, 233, see
Aldingar.

Ravens (Three, Twa Corbies), 73, 129,
240.

Ravisher’s Reparation, 283.

Reed, (Percy), 231.

Reedisdale and Wise William, 81, 238.
Regin the Smith, 98.

Regnar and Kragelille, 101, 217.
Relatives (Uncharitable), 295.
Remarriage prevented, 176, 182.
Renaud (Roi, Jean), 45, 83, 138, 142,
182, 211, see Elf-shot.

Renaud Wife-killer, 85, 142, 260. see
Hallewijn.

Return by Night, 86.

Revenge, 127.

Reynald of Montauban, 172, 173, 176,
Rhodanthe kai Dosikles, 362.

Rihold (Ribbolt) and Guldborg, 53,
76, 77, iqx, 202, 216, 222, 244.
Ribolt’s Fight with Ailer, 217.

Richard of the Lee, 235.

Richmode von Adocht, 268.

Rico Franco, 85, 1S2, 260.

Riding down from Bangor, 3.

Riga, 374 -
Rimaardson, 216.

Ring and Veil, 352.

.Rings, see Marianson.

Ritter von StaufFenberg, see Knight.
Robbers (Evil), 279, 280.

Robber’s Bride, 77, 88, 283, 343, 380.
Robert, 141.

Robert the Devil, 148.

Robin and Makine, 178,

Robin Hood, 6, 13, 25, 32, 5.7,76, 306,
229,231,232,234,235,332.

Robyn and Gandeiyn, 233. .

INDEX

402

Roddyngar, see Aldingar.

Roderick (Rodrigo, King), 57, 95, 128,
166, 167.

Roger I, 150.

Roland (Chanson de), 90, 91, 106.
Roland and Godelinde, 257, 271.
Roland to the Dark Tower came, 123,
220.

Roman (Prince), 357, 366, 369, 373,
376.

Roman and Olena, 77, 283, 380.
Roman Grue Grozavanul, 351.
Romance de Don Fadrique, 128.
Romance de Doha Blanca, 128.
Romance mauresque, 128, 170.
Roncesvalles, 94, 102, 172, 218.

Rosa fresca, 51, 180, 194.

Rosalia (St.), 152.

Rose of England, 233.

Rosengarten, 363.

Roses (Three), 268.

Rosmadec, 147.

Rosmer Havsmand, 123, 220.
Rozmelchon, 148.

Rubas, see Shroud.

Riibezal, 262, 280.

Sadko, 130, 223, 337 , 357 , 37 °, 371 -
Sailors (Seafarers), 79.
Saint-Germain-en-Laye, 254.

Saisnes, 176.

Sale of Wife, 343.

Sallemand, 226.

Salmi, 299.

Saluzzo’s Will, 144.

Samaritan Woman (Magdalene), 59,
78, 108, 218, 236, 273, 336.
Samovile (Fountain of), 345.

Samson, 59, 108, 2x8, 361, 362, 363,
364, 367.

Samuil Koska, 377.

Saracen Moor, 79, 145.

Saracens, 148.

Saunders (Clerk), 77, 240.

Saur, 107, 361, 362, 369.

Sawa, 284.

Saxon Maiden, 252, 253.

Sayavedra, 6, 30, 161, 162.

Scelkan Dudent’evic, 373.
Schlangenkochin, see Poisoner.
Schnadahupfl, 245.

Scholars of Ponthieu, see Hanged
Scholars.

Schiittensam, 248.

Scott (Willie), 6.

Sea-fight and Slave, 312.

Sea-suitors, 296.

Sea-swallows, 127.

Sea-wolves, 147.

Secret of the Sea, 130.

Sedan, 243, 254.

Sempach, 68, 126, 239, 244, 245, 246,

247*

Serajevo, 254.

Serapheim (Bishop), 312.

Serrallonga, 57, 193.

Serrana de la Vera, 165.

Servant-girl, 193.

Seven Golden Hills, 288.

Sevilla, 184.

She who feigned Death to guard her
Honour, 141.

She who went to see her Mistress in
Hell, 148.

Sheep saved from Wolf, 138.
Shepherd (Dead), 348.

Shine on us, lovely Sunshine, 264.
Ship Catharine, 79, 183.

Shipman, 53.

Short Straw, 140, 148, 183, 205.
Shroud, 76, 262, 273, 280.

Sickingen, 140, 243, 252.

Signs of the Beloved, 182.

Sigurd (Siegfried), 98, 201.

Silkie (Great), 237.

Silvertooth, 80, 315.

Simeon the Foundling, 337.

Simo the Latin, 335.

Simon and his Sister, note C.

Simon’s Hermitage (St.), 88, 181.
Sinclair (Malcolm), 19, 216.

Sinner, 77, 273.

Sisman, 343.

Sister avenged, 85.

Sister frees Brother, 370.

Sister (Lost, rediscovered, stolen), 99,
274 > 287.

Sister (Unmerciful), 269.

Sister tests Brother, 340.

Sister (Two), 77, 148, 223, 238.
Sisters at Heaven’s Gate, 262.

Sivard Snarensvend, 98, 216.

Skadar (Founding of), 87, 316, 326.
Skull, 185.

Sleepers (Seven, Seven Brothers), 3,
77, 216.

Smith’s Daughter, 262.

Smolensk, 374.

Sokol’nik, 367.

Sold Miller’s Wife, 269.

Soldier’s Return, see Husband’s Re¬
turn.

Solomon, 59, 189, 361, 363, 370.
Solovel, 364, 367.

Some men for sudden Joy do weep,
123.

Song’s Reward, 266.

Sorbs’ Victories, 277.

Sore-wounded, 221.

Sorrow’s Might, 128.

INDEX

Soul, 262.

Spaniard in Oran, 120, 186.

Sparrow’s Wedding, 288.

Spencer (Sir Hugh), 232.

Spens (Sir Patrick), 16, 67, 82, 109,
IX 3> 213, 231.

Springfield Mountain, 241.

Squire of low Degree, 61, 103.
Squirrel and his Wife, 270.

Stagolee, 242.

Starkadsmal, 198, 211.

Starling and Bath-tub, 99,

Stayr, 82, 258, 369, 370.

Stenka Razin, 374.

Stepan Dusan (Dushan), 25, 325, 327,
332 , 343 -

Stephen (St.), 60, 108, 218, 230, 233,
336 .

Steutlinger, 268.

Stig Hvide, 65, 213, 214.

Stojan and his Mother, 81.

Stojan Jankovic, 86, 334, 335, 342,
343, note C.

Stortebeker, 68, 243, 248, 249.
Stralsund, 253.

Strassburg, 243, 254.

Students of Toulouse, see Hanged
Scholars.

Stideli, 85, 99, 202.

Suffolk Miracle, 87, 237,238,240,309.
Suhman, 369.

Suitor (Tricked), 226.

Sultan’s Daughter, 54, 76, 218, 262.
Sun (Count), 182.

Sun and Moon woo a Maiden, 338.
Sunflower Kingdom, 374.

Suometar’s Wooers, 299, note H.
Susanna of Homldd, 276.

Svejdal, 101, 202, 211, 216, 238.

Sven i Rosengard, 239.

Svend Grade, 65, 66, 195, 212, 214.
Svetlana, 129.

Svjatogor, 362, 364, 365, 366, 367.
Swabian Round Table, 270.

Swabian War, 246.

Swallow Song, 88, 135, 320.

Sweet William’s Ghost, 86, 123, 222,
237, 262.

Swine-girl, 53, 74, 141.

Sword of Vengeance, 102, no, 217.
Syropoula, see Zerbopoula.

Syropoulos (Skleropoulos), 106, 304,
307.

Taara, 297.

Tailor and Dwarfs, 148.

Talamanca-la Grua (Caterina), 151.
Tam Lin, 35, 82, 237, 238.

Tamar, 189.

Tannenbaum, see Fir.

403

Tannhauser, 58, 60,70,74,75,127,134,
196, 199, 204, 223, 244, 257, 2s>8,
263.

Tarfe, 164.

Taucher, 125, 142.

Tell, 60.

Tests, 352.

Tete du Comte, 128.

Thidrekssaga, 62, 63, 64, 97, 98, 99,
107, 197, 200, 201, 202, 216, 217,
21S, 358, 362, 364.

Thijsken van der Schiide, 248, 272.
Thomas and fair Annet, 238.
Thomas Rymer, 229, 236, 237.

Thor (Tord af Havsgaard), 26, 27, 93,
202, 210, 212, 216.

Thor and Thure, 223, 224.

Thorn and Olive, 181.

Thou art mine, 263.

Tidemand, 213.

Tierhochzeit, see Animals’ Wedding.

Todorof Pomorje, 86, 330, 342, noteC.

Todtenamt, 267.

Tomillas, 6.

Tommaso di Savoja, 144.

Torch of Love, 84, 138, 142.

Torskild Trundeson, 224.

Tournament (They were seven and
seventy), 99.

Toveiille, 65, 214, 215.

Tour (Francois de la), 147.

Trebizond, 31.

Tristan, 26, 27, 69, 104, 174, 181, 204,
205, 217, 218, 259.

Troilus, 59, 108.

Trucks Daughters, 235, 372.
Tsamados, 98, 107, 277, 304, 309, 363,
Tsar and Maiden, 339.

Tsoulakis, 319.

Tsoulkas, 312.

Tsouvaras, 312.

Tuck (Friar), 6 .

Tudorel, 351.

Tugarin, 362, 368.

Tuoni, 297.

Turkish Wooer, 274.

Turk’s Bride, 81, 88, 274, 283.

*Twixt Arras and Douay, 134.

Tyolet, 104, 174.

Ukko, 297.

Ulinger, see Hallewijn.

Ungersvend, 217.

Unter der Linde, see Liebesprobe.
Unverdorben (Peter), 270.

Urraca, 172.

Valdovinos, 103, 176.

Vanamoinen (Wanemuine), 29S, 300,
366.

INDEX

4°4

Vanegas, 160.

Vasili! Buslaev, 64, 107, 357, 370, 371,
372.

Vasili! Kazimirov, 360, 368.

Vasili! the Drunkard, 369, 379.
Vasilissa, 370.

Veit and Heini, 250.

Velez (Count), 104, 183, 372.

Vellido Dolfos, 153.

Verdant River, 161, 162.

Vergil, 259.

Verinen poika, see Bloody Son, Ed¬
ward.

Verner, 222.

Verrua, 144.

Viderik Verlandson, 99, 200, 201.
Vidra, 35352.

Viipuri (Viborg), 297.

Vil’jam i Margarita, 129.

Villareal, 165.

Villfar, 220.

Villum, 225.

Virguleina, 149.

Vladimir, 11, 24, 56, 62, 64, 65, 100,
107, 118, 345, 355, 356, 357, 358,
360, 367, 369.

Vodas (Black), 277.

Vogelhochzeit, see Birds’ Wedding.
Voichita, 86, 344, 352.

Vol’ga (Volh), 359, 362, 364, 365, 366,
367, 378, note B.

Vonved (Sven), 216.

Vorwirth, see Shroud.

Voyage (Sir Peter’s), 79.

Vujadin, 335.,

Vuk Mandulic, 334.

Vukasin, 326, 329, 330, 352.

Vukosav, 335.

Wager, 82.

Waggoner’s Lad, 241.

Waldere (Waltharius), 68, 97, 100,177,
200, 235.

Wallace, 231.

Wallenstein, 253.

Warrior-maiden, 61, 79, 80, 184, 274,
2 iS, 339 , 343 , 352 , 379 , 380.
Wasserman, see Merman.

Watch on the Rhine, 254.

Waters (Child), 145, 182, 239.

Waters (Two), 54.

Wax Child, 146.

Wedderburn’s Courtship, 238.
Wedding, see Marriage.

Wedding of Bear and Wolf, 288.

Wee, wee Man, 236.

Weissenberg (Lady of), 181, 268, 352.
White Beast, 142, 222.

White-footed Stag, 174.

Widow, 83, 148.

Widsith, 235.

Wife (Murderous), 239.

Wife of Usher’s Well, 34, 238.
Willelmus, 272.

William the Good, 151.

Willie’s fatal Visit, 238.

Winter Rose, 261.

Wise Woman, 269.

Witty Answer, 239.

Wodny Muz, see Merman.

Woeful Janja, 343.

Wolfdietrich, 203, 255, 280.

Wooing, 257.

Ximena, 128.

Ydoine, 135, 141, 183.

Youth and Prudent Woman, see
Nastasja Politovskaja.

Ysabiaus, 135, 268.

Zaide, 121, 164.

Zamora, 96, 153, 166, 168, 172.
Zerbopoula, 36, 317.

Zlatija, 80, 315, 343.

Zuleikha, see Potiphar’s Wife.

2 urilo, see Curilo.

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "European Balladry"

See other formats